2-HSJI Vol 2 (2) 30 Jan 2012

Vol. 2, No.
1, April 2011
Editorial note
51
Volume 2, number 2, December 2011, pages 51-100, ISSN 2087-7021
Health Science Journal of Indonesia

TABLE OF CONTENTS
A. Suwandono S. Idaiani
Editorial note Inter-rater reliability of Health of Nations Outcome Scale (HoNOS) among mental health nurses in Aceh HoNOS has a good inter-rater agreement among mental health nurses in Aceh. Cold working room temperature increased moderate/severe qualitative work stressor risk in Air Traffic Controllers A working room temperature that was too cold increased risk of moderate/ severe overload qualitative work stressor among air traffic controllers. Risk factors of post partum haemorrhage in Indonesia Eclampsia is the strongest for post partum haemorrhage. Signs or symptoms of complications in pregnancy and risk of caesarean section: an Indonesia national study Women with any sign or symptom of complications in pregnancy have an increased risk of c-section. Quality of refill drinking water in Greater Jakarta in 2010 Refill drinking water in Greater Jakarta in general has a good quality of the physical and chemical parameters. Fingernail biting increase the risk of soil transmitted helminth (STH) infection in elementary school children Fingernail biting and no hand-washing before meals and no hand washing with soap after defecation increased the risk of soil transmitted helminth infection. Sensitivity and specificity of immunocytochemical assay for detection of Dengue virus 3 infection in mosquito Immunocytochemical assay has a high sensitivity and high specificity to detect DENV-3 infection on mosquito head squash. Virus culture and real-time RT-PCR in influenza-like illness cases in Indonesia 2007-2008 Virus culture was still essential and considerably efficient to support real-time RT-PCR detection in ILI cases in Indonesia. Several dominant clinical symptoms associated with InfluenzaA in Indonesia In addition to fever and coughing, runny nose, sore throat, and ever had fever dominantly associated with influenza A.
51 53
D. Astuti B. Basuki H. Mulijadi R. P. Jekti E. Suarthana Suparmi B. Basuki
58
66 71
A. A. Kurniatri
77
L. Sofiana S. Sumarni M. Ipa D. Widiastuti B. Yunianto S. R. Umniyati N. Wijayanti I. L. Indalao V. Setiawaty H. A. Pawestri Subangkit Roselinda N. Fitri
81
87
92
96
52
Buang et al.
Health Science Indones
Acknowledgment of Reviewers Volume 2, 2011

Ani Isnawati
Center for Basic Biomedical and Health Technology, Institute for Health Research and Development, Ministry of Health of Indonesia
Joedo Prihartono Retno Gitawati
Faculty of Medicine, Universitas Indonesia, Jakarta Center for Basic Biomedical and Health Technology, Institute for Health Research and Development, Ministry of Health of Indonesia
Bob Tilden
HSJI international advisor
Christine Thayer
Saleha Sungkar
HSJI international advisor, France
Faculty of Medicine, Universitas Indonesia, Jakarta Supratman Sukowati Centre for Technology in Public Health Intervention, National Institute for Health Research and Development, Ministry of Health of Indonesia Suryadi Gunawan Center for Applied Technology and Clinical Epidemiology, National Institute for Health Research and Development, Ministry of Health of Indonesia Tris Eryando Center for Applied Technology and Clinical Epidemiology, National Institute for Health Research and Development, Ministry of Health of Indonesia
Didi Danukusumo
Faculty of Medicine, Universitas Indonesia, Jakarta
Elisabeth Emerson
Former WHO consultant; HSJI international advisor; Minnesota, USA
Eva Suarthana
Gendrowahyuono
Center for Applied Technology and Clinical Epidemiology, National Institute for Health Research and Development, Ministry of Health of Indonesia
Irmansyah
Weda Yuwana
Air Traffic Control of Soekarno-Hatta Airport, Jakarta
Health Science Journal of Indonesia

Editor-in-chief: Trihono; Deputy Editor-in-chief: Agus Suwandono; Editorial board: Asri Adisasmita, Atmarita, Betty Rossiehermiati, Herqutanto, Minarma Siagian, Muchtaruddin Mansyur, Ratna Juwita, Sanjaya, Siti Isfandari, Sudarto Ronoatmodjo. International Editorial Advisory Board: Bastaman Basuki (Universitas Indonesia, Indonesia), Christine Thayer (Health for Development, France), Elisabeth Emerson (Minnesota Department of Health, USA), Hans-Joachim Freisleben (German-Indonesian Medical Association, Germany), Martin Weber (WHO Indonesia, Indonesia), Robert Tilden (Consultant, Indonesia) Layout: Jerico Franciscus Pardosi, Muhammad Kamil; General Affairs: Cahyorini, Endang Sri Widyaningsih, Erwin Mustikowati, Leny Wulandari, Siwi Wresniati. Printed by Badan Penerbit Fakultas Kedokteran Universitas Indonesia. Subscription: The journal is published quarterly and should be subscribed for a full year. Advertisement: Only advertisements of health science or related products will be allowed space in this journal. For all inquiries please contact Health Science Journal of Indonesia Editorial Office at Badan Litbangkes, Gedung Pusat Teknologi Intervensi Kesehatan Masyarakat, Jln. Percetakan Negara No. 29 Jakarta Pusat 10560, Indonesia; Tel (62-21) 42872393. Fax (62-21) 42872392. E-mail: healthsciencejournal@yahoo.com
Vol. 2, No. 2, December 2011
Editorial note
51
Editorial Note
Dissemination of Health Research for Better Health Programs
As stated by the Director of Research Policy & Cooperation, World Health Organization (WHO), Geneva, Dr. Tikki Pangetu et al (2003), knowledge produced by health research, if disseminated widely, is a global public good. Knowledge contributes to the policies, activities, and performance of health systems, and to the improvement of individuals and populations health. Using existing knowledge adapted to local conditions is particularly crucial in achieving the local, country as well as Millennium Development Goals and other world committment of health development goals. To achieve these and other health-related goals, a well-functioning health system must be able to access and utilize research-based knowledge and the products of research. WHO SEARO Regional Director, Dr Samlee Plianbangchang (2005) said that the vision of knowledge management and sharing/dissemination was to attain global health equity through better knowledge management and sharing/dissemination, was extremely important to health development. The contribution of knowledge management and sharing/dissemination to health for all will be enhanced if the application of this concept was succesfully implemented to empower the people at large to be able to plan and take care of their own health effectively. Another real challenge was to ensure accessibility and utilization of the relevant knowledge in health development processes at all levels. The publication of research ndings is considered to be the primary output of the research process. Research can be utilized in two main ways: rst, for developing new tools (drugs, vaccines, devices and other applications) to improve health; and second, for translating, communicating, and promoting the utilization of research to inform health policies, strategies, and practices, particularly within health systems. Research can also be used to educate the population and change public opinions and practices. It is generally agreed that a wide gap exists between current health systems and the needs that health systems should address, one major cause being the inability to synthesize existing research outputs and apply existing knowledgetowards improving interventions and the performance of health systems (WHO, 2004, Knowledge for Better Health). One of the major dream of the National Institute of Health Research and Development (NIHRD) is to be a scientic institution which can drive the national health development based on the research evidences found by NIHRD researchers and its networking partners. The Health Science Journal of Indonesia (HSJI) is one of the efforts of NIHRD in knowledge dissemination, to share the research ndings conducted by health researchers in NIHRD and other academic and private research institutions. Dissemination of health research through HSJI hopefully can be used to improve the quality and efciency of health programs and policy formulation, as well as to improve capability of the health researchers in inuencing the use of evidence based to advocate the health providers for better health programs. The third edition of HSJI consists of 9 articles covering several diciplines of health knowledge such as mental health nurses in Aceh, quality work stressor risk in air trafc controllers, risk factors of post partum haemorrhage in Indonesia, complications in pregnancy and risk of caesarean section in Indonesia, quality of rell drinking water in Jakarta, ngernail biting and the risk of soil transmitted helminth (STH) infection in elementary school children, immunocytochemical assay for detection of Dengue virus 3 infection in mosquito, and virus culture and to real-time RT-PCR in inuenza-like illness cases in Indonesia. Several of them are original research carried out by NIHRD and other health institution researchers while some of them are using data based on the results of National Health Basic Research 2010 or Riskesdas 2010.
52
Suwamdhono
Initially this third and the fourth edition of HSJI were to focus on the special topic of Riskesdas 2007 and Riskesdas 2010 particularly in some topic related to the achievement efforts of health development programs in Indonesia to the targets of Millenium Development Goals 2015 as well as the analysis of speciment related to non-communicable and communicable diseases based on the specimens taken during the Riskesdas 2007, but unfortunately the NIHRD researchers were too busy with the implementation of National Health Facility Research in 2011 (Rifaskes 2011) and other health research such as research for Development of Area with Health Problems (PDBK), so that the theme was changed to a more general health research focus with some analysis of Riskesdas 2010. We are hoping, however, those topics, and some of the Rifaskes 2011 preliminary results as well as the research of PDBK results can be published in the next editions of the HSJI. In the up coming editions, HSJI editors invite all health researchers across Indonesia, particularly those who are coordinated under the National Health Research and Development Network (Jaringan Penelitian dan Pengembangan Kesehatan or Jarlitbangkes) to publish their research ndings in this journal by submitting their scientic articles to HSJI. Thank you very much to all of the national and international editors and reviewers who have worked seriously to publish this HSJI edition. The HSJI also congratulates all of the authors who had their articles published in this third edition and for their successful negotiation with our peer review team members and editors. Our special thanks and best appreciation is forwarded to Professor Bastaman Basuki who has wisely and seriously carried out some necessary writing skill training and assistance to the researchers in NIHRD in developing their articles.
Agus Suwandono Deputy Editor-in-chief Health Science Journal of Indonesia, National Institute of Health Research and Development (NIHRD) Ministry of Health, Republic of Indonesia
Inter-rater reliability of HoNOS
53
Inter-rater reliability of Health of Nations Outcome Scale (HoNOS) among mental health nurses in Aceh
Sri Idaiani
Centre for Applied Health Technology and Clinical Epidemiology, National Institute of Health Research and Development, Indonesia Ministry of Health
Abstrak
Latar belakang: Alat ukur Health of Nations Outcome Scale (HoNOS) dapat digunakan untuk keperluan rutin dan menilai perkembangan status fungsional pasien gangguan jiwa di klinik maupun komunitas. Tujuan penelitian adalah melakukan uji kesepakatan (agreement) bagi perawat jiwa yang akan menggunakan alat ukur ini di Indonesia. Metode: Lima puluh lima orang pasien psikosis yang sedang di rawat di Rumah Sakit Jiwa (RSJ) Provinsi Aceh dinilai secara bergiliran oleh 11 orang perawat jiwa menggunakan alat ukur HoNOS. Penilaian dilakukan pada bulan September 2011. HoNOS terdiri dari 11 pertanyaan. Kesesuaian di antara perawat terhadap masing-masing pertanyaan dinilai menggunakan koesien intra-class classication correlation (ICC). Hasil: Empat puluh dua persen pasien berusia 31-40 tahun, termuda berusia 24 tahun dan tertua berusia 59 tahun, mayoritas laki-laki, dan 38% mengalami psikosis selama 5-10 tahun. Tiga puluh enam persen perawat berusia >40 tahun sebagai penilai, separuhnya perempuan, dan 55% telah bekerja lebih dari 10 tahun. Nilai ICC untuk masing-masing pertanyaan secara umum baik (berkisar antara 0,8-0,9). Kesepakatan yang baik didapatkan di antara perawat Rumah Sakit Jiwa, perawat Puskesmas, maupun gabungan keduanya. Kesimpulan: HoNOS memiliki inter-rater agreement yang baik dan dapat digunakan pada penelitian dengan setting yang sama. Untuk penggunaan di populasi yang lebih besar dan berasal dari daerah yang berbeda disarankan untuk melakukan uji reliabilitas serta validitas dengan jumlah sampel yang lebih besar. (Health Science Indones 2011;2:53-7) Kata kunci: HoNOS, agreement, psikosis
Abstract
Background: The Health of Nations Outcome Scale (HoNOS) instrument could be used for routine purposes and assessing the functional status of the mental health patients in clinical and community settings. The objective of this study was to evaluate the agreement of the scale among mental health nurses who would use this tool in Indonesia. Methods: Fifty ve psychotic patients who were hospitalized at a mental hospital in Aceh were evaluated by 11 mental health nurses using the HoNOS instrument. The agreement between the nurses on each questionnaire item was evaluated using the intra-class correlation (ICC) coefcient. Results: Forty-two percent of the patients were 31-40 years of age, the youngest was 24 and the oldest was 59, most of them were males, and 38% had psychosis for 5-10 years. Thirty-six percent of the nurses aged >40 years as raters, half of them were females, and 55% had worked for >10 years. ICC values were generally good (ranging from 0.8 to 0.9) among the mental health hospital nurses, as well as the community health centers nurses and a combination of both. Conclusion: This instrument showed a good inter-rater agreement and could be used in future research with the same settings. For a wider use in different regions, it is recommended to test the reliability and the validity of the HoNOS in a larger study population. (Health Science Indones 2011;2:53-7) Key words: HoNOS, agreement, psychosis.
Corresponding author: Sri Idaiani E-mail: sriidaiani@yahoo.com
54
Idaiani
Assessment of the effectiveness of a mental health intervention requires a measurement tool to assess the improvement of the outcomes of the mental disorders. Problems in the assessment of these outcomes include the absence of the following: a clear statement of the purpose of the intervention or treatment, clear outcome indicators, standardization of how to conduct such study, and the appropriate time intervals to assess the improvement. These ndings from a study by Barbach and colleagues was cited by Trauer.1 Assessment of the outcome of mental disorders using functional status is better than simply assessing the improvement of clinical symptoms.2 Functional status assessment emphasizes more on improvement of ones ability in terms of functioning on a daily basis. If someone has a late disability, then the most important thing is the ability to function optimally.2 The Health of Nations Outcome Scale (HoNOS) is a measuring tool that can assess the development of functional status, especially for patients with mental disorders, both schizophrenia, psychotic and nonpsychotic disorders, and others.1,3 This tool has been used in many countries, especially in the United Kingdom, Australia, and New Zealand. It has also been used in Canada, Germany, Italy, France and Norway. Assessments of the results have been published in numerous articles and publications.1,3-5 This tool was designed by Wing and colleagues in 1990.3 The HoNOS can be used for assessing the outcome of the mental disorders and other routine assessments. Previous studies show that the validity of the content, the tests reliability, and the inter-rater reliability were good1,3-6 Other measuring tools that have been used in Indonesia are the Positive and Negative Symptoms Scale (PANSS) and the Brief Psychiatric Rating Scale (BPRS), which only measures the degree of symptoms alone; and the Global Assessment Functioning (GAF) to measure the patients functional status.1 However, GAF is widely used in clinical settings by psychiatrists to measure the improvement of the patients functional status.2 GAF was designed as a part of the fth axis of the Diagnostic and Statistical Manual of Mental Disorders (DSM). METHODS This study was conducted in Aceh Province because
Aceh is the only province in Indonesia that has community mental health programs that have been implemented in all districts and cities. In the preparation phase, the HoNOS questionnaires were rst translated from English into Bahasa Indonesia. Afterwards, they were back translated into English by different translators. This step was followed by a group discussion panel of researchers from NIHRD who produced a questionnaire measuring tool with content and translation in Bahasa Indonesia, in a way that it would at least be understandable by the community health center (CHC) nurses. A pilot study was done to evaluate the HoNOS. Ten psychotic patients from six CHCs in the city of Banda Aceh and Aceh Besar district were assessed by six mental health nurses from the CHC. Each nurse assessed 1-2 patients using the HONOS based on the information contained in CHC medical records and interviewed patients and their families. Afterwards, the researchers had three sessions to discuss issues related to language, content, and the feasibility of this measuring tool. The Indonesian version of HoNOS, especially for the Aceh population, was nalized after the group discussion. The inter-rater agreement of the Indonesian version of the HoNOS were evaluated among 11 nurses (six nurses who served at CHC in Banda Aceh and Aceh Besar district and ve nurses who served at the Province of Aceh Mental Hospital). The mental health nurses involved in this study were nurses who run the mental health program at CHC and the selected CHCs have ample mental health patients. Nurses voluntarily joined the study. The ow of the research was described in Figure1. The nurses evaluated 55 psychotic patients who were being treated at the Province of Aceh Mental Hospital. The inclusion criteria includes residing in the city of Banda Aceh or Aceh Besar district and the patient was not expected to be discharged within 2 weeks from study inception. Patients in special wards (i.e. drug user or having physical comorbidity) were also excluded. Of 63 eligible patients, 55 patients were assessed by 11 mental health nurses between September 26 and October 4, 2011. Each rater assessed each patient. Before assessing the patients, at the same time all nurses joined a 3-hour short course about how to use the measuring instrument in the hope they would have the same understanding of the meaning of the questions. The guideline was translated from the original, which
55
HoNOS: English version Group discussion Translation and back translation
HoNOS: Indonesian version
Pilot study: assessment of 10 psychotic patients by 6 nurses at Community Health Centers
Reliability test (inter-rater agreement): Assessment of 55 psychotic patients treated at mental health hospital by 6 CHC nurses and 5 hospital nurses
Figure 1. Flow of research
was developed by Wing and colleagues. The HoNOS consists of four dimensions of behavior, impairment, and social symptoms. The four dimensions were translated into 12 items of questions. Items 1-3 were the domain of behavior; items 4-5 were included in the impairment domain; items 6-8 were in the symptoms domain; and items 9-12 were in the social domain. For each item, the score ranged from 0 to 4, with the following criteria: 0 = no problem; 1 = minor problem that does not interfere with the function; 2 = mild problem 3 = moderate to severe problem; and 4 = severe to very severe problem.. A value of 9 was assigned when the rater did not know or the condition of the patient did not t the score criteria. Total scores and sub-total scores for each domain were calculated. For this analysis, a patients education was divided into three categories (low = no school or not completed primary school; medium = completed primary school or junior high school; high = senior high school or graduate schools). However, for the nurses only two categories (medium = high school; high = graduate schools) were used. Inter rater-reliability analysis was performed by calculating intra-class correlation (ICC) coefcients using a two way mixed model and absolute agreement
type. The average measure coefcients were reported. ICC is a good method to evaluate inter-raters agreement because ICC controls for measurement errors that might occur.7 Analysis was done using SPSS version 15.0. Research permission was obtained from the Ministry of Home Affairs, while the ethics approval was obtained from the Ethics Committee of the National Institute of Health Research and Development (NIHRD), Ministry of Health of the Republic of Indonesia. RESULTS The characteristics of 55 patients are presented in Table 1. Patients were mostly male and more than 80% were between the ages of 24 and 39. Thirty-six percent of the nurses were 40 or more years, half of them were female, and 55% had tenure of more than 10 years (Table 2). All nurses had attended the basic level of training on community mental health nursing (CMHN). Four nurses had an advanced course on CMHN. Item number 2 of the HoNOS was not included in the analysis because there were only three patients who were reported to have a self harm problem. This low number might be explained because of good observation by hospital staff. Item number 3 (i.e. substance use) was
56
Idaiani
also not 1. analyzed patients who used substances, Table Some because demographic characteristics of the including patients alcohol, were excluded from the study. Therefore, the analysis was just for 10 items.
n = 55 22 23 10 49 6 10 20 25 22 18 6 9 % Age 24-30 years 31-39 years 40-59 years Gender Male Female Education Low Medium High Duration of illness 5-10 years 11-15 years 16-20 years 21-41 years 40.0 41.8 18.2 84.5 10.3 18.2 36.4 45.5 37.9 31.0 10.3 15.5
In general, the inter-rater agreement was good (ICC coefcients ranged from very good (0.81+), and good
Table 2. Some demographic and employment characteristics of the nurses (raters)

n Age 27-30 year 31-39 year 40-47 year Gender Male Female Education Medium High CMHN course Never Basic Intermediate Advance Period of employment 5-10 year 11-20 year 21-26 year Work sites Mental hospital Community health center 4 3 4 5 6 2 9 3 3 1 4 5 3 3 5 6 % 36.4 27.3 36.4 45.5 54.5 18.2 71.8 27.3 27.3 9.1 36.4 45.5 27.3 27.3 45.5 54.5
Nurses from the CHC were involved in the pilot assessment of the HoNOS, while the hospital nurses were not. The rst group might know the measurement better than the second group. Therefore, the ICC were evaluated in three groups: the CHC nurses, the hospital nurses, and all nurses.
Table 3. Intra-class correlation coefcients of HoNOS items

No 1 4 5 6 7 8 9 10 11 12 HoNOS items* Agression Cognitive problem Physical illness and disability Hallucination and delusion Depresssion Other symptoms Relationship Activity Daily Living Residential enviroment Day time activity HONOS total score 1-10 items 1-8 items Intra-class correlation coefcients* CHC nurses MH nurses (n = 6) (n = 5) 0.93 0.97 0.92 0.97 0.85 0.90 0.93 0.90 0.92 0.93 0.92 0.94 0.91 0.94 0.92 0.95 0.91 0.88 0.84 0.61 0.91 0.85 0.92 0.96 Total (n = 11) 0.96 0.94 0.89 0.93 0.92 0.95 0.93 0.91 0.84 0.82 0.94 0.91
*The original items 2 and 3 were excluded because only 3 subjects had experienced self harm, while drug users were excluded from the study.
57
(0.61-0.80). Nurses from CHC were involved in the pilot study, while the hospital nurses were not. This might cause bias as the CHC nurses might have a better understanding of the tool. However, the ICC coefcients between the nurse groups turned out to be comparable. DISCUSSION The reliability of the HoNOS among mental health nurses in this study was reliable. Previous studies in 293 patients in Manchester, 100 patients in Nottingham, and 50 patients in Geelong showed good reliability. Patients involved in those studies also had a diagnosis of mental disorder, which varied and was not limited to psychosis.1-5 With these ndings, the HoNOS could be used in future studies. Nevertheless, this research was done in Aceh because the province has a good community mental health program that was initiated after the tsunami disaster in 2004. In early 2005, the community mental health program was introduced and further developed in all health centers in Aceh province. At this moment mental health nurses could be found not only in mental hospitals, but also scattered in almost all CHCs in Aceh province. This is not the case in other provinces. Moreover, this study only assessed psychotic patients and the nurses who evaluated the subjects were mental health nurses. Therefore, if the HoNOS is going to be used in other places with different settings (e.g. other provinces and not limited to psychotic patients), the reliability of this measurement should be re-evaluated. This study has several limitations. Sample size can affect the internal consistency, reliability and interrater reliability.8 Although this study was done in 55 psychosis patients, the ndings were comparable to much bigger studies in Manchester and Nottingham1,3,5 The validity, content, predictive, and diagnostic criteria of the translated HoNOS were not evaluated,
although they were important aspects in trans- cultural epidemiology, because this study was a small part of a main study assessing the effectiveness of the community mental health program for psychotic patients at CHCs. Complete evaluation of these aspects should be considered in future studies.9 Acknowledgments The author wishes to thank all subjects who willingly participated in this study. The author would also like to express her sincerest gratitude to Prof. Bastaman Basuki, Dr. Eva Suarthana, Dr. Muchtaruddin Mansyur, and Dr. Elisabeth Emerson for technical assistance in preparing this nal draft. REFERENCES
1. Trauer T, Callaly T, Hantz P, et al. Health of the Nation Outcome Scales Results of the Victorian eld trial. Br J Psychiatry. 1999;174:380-8. Csipke E, Wykes T. Global functioning scales. In: Thornicoft GT, M, editor. Mental Health Outcome Measures. 3 ed. London: RC Psych Pub; 2010. Wing J, Beevor A, Curtis J, et al. Health of the Nation Outcome Scales (Ho NOS) research and development. Br J Psychiatry. 1998;172:11-8. Pirkis J, Burgess P, Kirk P, Dodson S, Combs T, Williamson M. A review of the psychometric properties of the Health of the Nation Outcome Scales (HoNOS) family of measures. Health Qual Life Outcomes [serial on the Internet]. 2005;3. Trauer T. The subscale structure of the Health of the Nation Outcome Scales (HoNOS. J Ment Health 1999;8:499-509. Preston N. The Health of the Nation Outcome Scales: validating factorial structure and invariance across two health services. Aust N Z J Psychiatry. 2000;34:512-9. Weir J. Quantifying test-retest reliability using the intra class correlation coefcient and the SEM. J Strength Cond Res. 2005;19(1):231-40. Charter R. Study samples are too small to produce sufciently precise reliability coefcients. J Gen Psychol. 2003;130:117-29. Ommeren V. Validity issues in trans cultural epidemiology. Br J Psychiatry. 2003;182:376-8.
2.
3.
4.
5.
6.
7.
8.
9.
58
Astuti et al.
Cold working room temperature increased moderate/severe qualitative work stressor risk in Air Trafc Controllers
Dewi Astuti,1,2 Bastaman Basuki,1 Herman Mulijadi3
1 2
Department of Community Medicine, Faculty of Medicine, Universitas Indonesia Indonesian Civil Aviation Institute, Ministry of Transportation, Indonesia 2 Aviation and Aerospace Health Institute, Indonesia Abstrak Latar belakang: Pemandu lalu lintas udara (PLLU) kemungkinan lebih besar terkena stresor kerja kualitatif. Tujuan penelitian untuk mengidentikasi beberapa faktor yang berkaitan dengan stresor kerja kualitatif moderat (SBKL) sedang di antara PLLU di Bandar Udara Internasional Soekarno-Hatta. Metode: Studi potong lintang dilakukan pada bulan November 2008 dengan subjek PLLU aktif bekerja minimal 6 bulan. Penelitian menggunakan kuesioner standar survei diagnostik stres dan kuesioner stresor rumah tangga. Kuesioner diisi oleh subjek. Hasil: Subjek berumur 27-55 tahun terdiri dari 122 PLLU dengan SBKL sedang/berat dan serta 13 (9,6%) PLLU dengan SBKL rendah. Model menunjukkan bahwa mereka yang merasa dibandingkan dengan yang tidak merasa suhu ruangan terlalu dingin mempunyai 11-lipat risiko SBKL sedang/berat [rasio odds suaian (ORa) = 10,63: 95% interval kepercayaan (CI) = 1,79-65,59]. Dibandingkan dengan subjek tanpa stresor ketaksaan peran, mereka yang mempunyai stresor ketaksaan peran sedang/berat berisiko 8,2-lipat SBKL sedang/berat (ORa = 8,23: 95% CI = 1,13-59,90). Di samping itu, mereka yang mempunyai stresor tanggung jawab sedang/berat mendapatkan dibandingkan dengan tanpa stesor ini 6,6-kali berisiko SBKL sedang/berat (ORa = 6,64: 95% CI = 1.13-38.85), Selanjutntya mereka yang mempunyai dibandingkan dengan yang tanpa stresor pengembangan karir sedang/berat mempunyai 3,7-kali risiko SBKL sedang/berat (ORa = 3,67: 95% CI = 0.88-15.35; P = 0,075). Kesimpulan: Subjek LLU yang merasa suhu ruangan terlalu dingin, stresor ketaksaan peran, tanggung jawab personal dan pengembangan karir sedang/berat mengalami peningkatan risiko SBKL sedang/berat. (Health
Science Indones 2011;2:58-65)

Kata kunci: suhu dingin, stresor beban kerja kualitatif, pemandu lalu lintas udara Abstract Back ground: Air trafc controllers (ATCs) have a high level of responsibility which may lead to qualitative work load stressor (QLWS). This study identied several risk factors related to moderate qualitative work load stressor among the ATCs. Methods: This cross-sectional study was conducted in November 2008 at Soekarno-Hatta International Airport. Subjects consisted of active ATCs with a minimum of six months total working tenure. The study used standard diagnostic as well as home stressor questionnaire surveys. All questionnaires were lled in by the participants. Results: Subjects were aged 2755 years, consisted of 112 ATCs who had moderate and 13 (9.6%) ATCs who had slight QLWS. Those who felt than did not feel the working room temperature was not too cold had 11fold moderate/severe QLWS [adjusted odds ratio (ORa) = 10.63: 95% condence interval (CI) = 1.79-65.59]. Those who had than did not have moderate/severe role ambiguity stressor had 8.2-fold risk of moderate/ severe QLWS (ORa = 8.23: 95% CI = 1.13-59.90). Those who had than did not have moderate/severe personal responsibility stressor had 6,6-fold risk for moderate/severe QLWS (ORa = 6.64: 95% CI = 1.13-38.85). In terms of the career development stressor, those who had it than did not have it had a 3.7-fold risk for moderate/ severe QLWS (ORa = 3,67: 95% CI = 0.88-15.35; P = 0.075). Conclusion: Those who felt the room temperature was too cold, moderate/severe role ambiguity, personal responsibility, as well as career development stressor were at increased risk for moderate/severe QLWS.
(Health Science Indones 2011;2:58-65)

Key words: working room temperature, qualitative work stressor, air trafc controller
Corresponding author: Dewi Astuti E-mail: dewi.astuti@rocketmail.com
Air trafc controller and qualitative stressor
59
Stressors and stress are main issues affecting the human factor in aviation. Stressors are the source of stresses. The effect of stress could produce psychological and physical reactions such as fear, anxiety, frustration, loss of motivation, decreased attention, slower reaction time, and decreased situational awareness.1,2 Air Trafc Controllers (ATCs) are required to work at optimal levels according to international standards, and there is intolerance for any error or mistake in order to guarantee perfect air trafc operation.3 It was believed the ATCs had excessive workload demands and a high level of responsibility. They also were at risk for experiencing work stress generated by their working environment or home stresses.4,5 A previous report noted that there were 11 near misses at the Soekarno-Hatta International Airport in 2006, more than the standard maximum for near misses of 3 miss per year. The report also found positive correlation between work stressors and performance. Moderatesevere qualitative work load stressors would produce stress that would decrease performance. Hence, the medium-heavy qualitative work load stressors would increase operational error errors or mistakes.[6] This study aimed to identify work environment stressors and other risk factors related to the moderate qualitative workload among the ATCs. METHODS This cross-sectional study was conducted in November 2008. Subjects consisted of active ATCs at SoekarnoHatta International Airport who had a minimum of six months total working tenure. Subjects were given information by researchers and lled in special questionnaires in their ofce or at home. Demographic and behavioral factors questionnaires identied age, gender, marital status, number of children, education, smoking habits and sports. Work characteristics questionnaires included information on job title, working unit, length of employment, experience in problem, stress management training, and second job/additional job. Working environment stressor questionnaires included information on lighting in the room, noise, working chair, conguration of the room, crowded working room, and other complaints. Stressor questionnaires used standard diagnostics for identifying stress.7 Working stressor questionnaires
consisted of role conict, ambiguity, quantitative workload, qualitative workload, career development, and personal responsibility. Work stressors were determined by a diagnostic survey questionnaire which consisted of 6 stressor groups (quantitative workload, role ambiguity, personal responsibility, career development, role conict and quantitative workload stressors). Moreover, each sub-group of stressors consisted of 5 questions and every question had a score from 1 to 7. One was the lowest and 7 were the highest score. The total score for each stressor ranged from 5 through 35. Furthermore, each stressor group was divided into 3 categories (low = 1-10; moderate = 11-23; severe = 24-35). More details are as follows: Qualitative workload stressor was work variability that required technical and intellectual ability above a workers abilities. It consisted of: job standard demands that were too high; assigned tasks that were sometimes too difcult/complicated; tasks that became more complex day to day; organizations expectations that exceeded my abilities and skills; and inadequate training or experience to accomplish my job. Role ambiguity was the workers feeling that he/she does not have enough information to do the job or does not understand the job sufciently to fulll the expectations of the role. It consisted of: my tasks and job description were not well dened; I did not clearly know to whom and who to report; I did not have authority in doing my obligations; I did not clearly understand what my goals are; and I did not understand my job role in the organization Personal responsibility was being responsible for customers safety. It consisted of: I was responsible for the development of other employees; I was responsible to guide and/or help my colleagues with their problems; I acted or made decisions that affected the safety and welfare of others; My responsibility was primarily about taking care of people rather than things; and I was responsible for my colleagues careers/ futures. Career development was potential stress aroused because of work uncertainty, over promotion and low promotion. It consisted of: I didnt have enough opportunities to advance in this organization; if I want to advance my position, I have to nd another job in another unit; my career will suffer if I stay in this organization; I only had minimal opportunity to
60
Astuti et al.
develop and learn new knowledge and skill in this organization; and I feel stuck in my career. Role conict was conict that was `created because of a mismatch between role demands and personal needs. It consisted of: I was doing unnecessary tasks; I was caught between my supervisor and my staff; Formal line of command was not obeyed; I was doing work that was not being done by colleagues and was their responsibility; I received contradictory orders from one or more person(s). Quantitative Workload Stressor related to limited time. It consisted of: I had to bring my work home every noon or weekend to stay on schedule; I spent too much time at unnecessary meetings and wasted my time; I was responsible for all kinds of jobs at the same time and almost uncontrollable; I really had more tasks then could be accomplished in one day; and I felt that I didnt have periodic time to rest. Working room lighting, noise, working chair, working room space and conguration of ones work station were categorized based on the subjects perceptions: The Home Stress Checklist questionnaire included ones role in the family, home physical factors, home tension and privacy.8 Home stressors included household conditions that might increase or create stress. These were categorized into four groups consisting of role in home, physical home stressor, home tension, and privacy. To determine a score for role in the home subjects were asked what is your role in your home: as the main source of family income, as a father/mother, as a husband/wife, as a nancial support to other family members, as a payer, as a gardener, as a home decorator, as a household repairman. Each role identied was given the score of one. The total score was determined by adding roles and categories (low = less than 2; moderate = 3-5; severe =6-7. Physical home stressor was a physical or home environment that could create stress. The question consisted of: neighborhood noise, small house, messy house, leaking and/or damage to part of ones house, dense neighborhood, ooded neighborhood, unsafe neighborhood. Each physical home stressor identied was given a score of one. The total score was determined by adding all items identied. The resulting score placed the subject into one of three categories (low = less than 1; moderate = 2-3; severe =4-5).
Home tension was the responders perception of their current home environment. The results placed the subject into one of three categories (low = small dispute and can be resolved; moderate = several tensions, but it was still tolerable; severe = the tension is very high). Privacy was time for personal matters without interference from others. The question was if you were at home, how often would you have time for yourself relaxing and enjoying an activity (low = always; moderate = seldom; and (severe = rare). Sports habit was physical exercise to maintain responder health and was divided into three categories (light such as walking 2-4 km/hour; moderate such as biking 16-20 km/hour, heavy such as jogging 6-9 km/hour).9 Out of 171 ATCs, 135 (78%) participated this study. For this analysis we excluded 19 subjects who had severe qualitative work stressors. Data analysis used Cox regression 10 using Stata version 9.
RESULTS Table 1 shows that most of the subjects had moderate/ severe QLWS were male, and aged between 27-55 years. In addition, Table 1 shows that subjects who had slight and moderate/severe QLWS were equally distributed with respect to gender, age group, number of children, and sport habit. However, those who were not yet married were less likely had a lower risk of having moderate/severe QLWS. Table 2 shows that subjects with slight and moderate/ severe QLWS were equally distributed in terms of job title, working unit, length of employment (11-23 years), experience with accident control, and stress management training. However, those who had 6-10 years of work experiences were less likely had lower risk to be moderate/severe QLWS than those with 11-30 years. Table 3 shows that subjects who had slight and moderate/severe QLWS were similarly distributed with respect to working room lighting, noise, working chair, working room space, and conguration of the working room.
61
Table 4 shows that subjects with slight and moderate/ severe QLWS were similarly distributed with respect to family role, home tension and privacy. However, those who had moderate than low physical home stressors were at a greater risk for moderate/severe QLWS than those with low physical home stressors. Subjects who felt than did not feel the working room temperature was not too cold had 11-fold moderate/
severe QLWS [adjusted odds ratio (ORa) = 10.63]. Those who had than did not have moderate/severe role ambiguity stressor had 8.2-fold risk of moderate/severe QLWS (ORa = 8.23). Those who had than did not have moderate/severe personal responsibility stressor had 6,6-fold risk for moderate/severe QLWS (ORa = 6.64). In terms of the career development stressor, those who had it than did not have it had a 3.7-fold risk for moderate/severe QLWS (ORa = 3.67; P = 0.075).
Table 1. Some demographic, habits characteristics and risk of moderate qualitative work load stressor
Qualitative work load stressor Slight (n=13) n Gender Male Female Age 21-29 years 30-39 years 40-49 years 50-55 years Marital status Married Not yet married Divorce/widow(-er) Number of children None 1-2 children 3-4 children Sport habit None Mild Moderate/ Heavy Smoking habits Never Ever Current 12 1 1 8 2 2 11 2 0 3 6 4 3 8 2 0 4 6 3 Moderate/severe (n=122) n 113 9 10 59 37 16 117 3 2 15 78 29 31 68 22 1 65 30 27 Crude odds ratio
95% condence interval
1.00 0.96 1.00 0.73 1.85 0.80 1.00 0.14 n/a 1.00 2.60 1.45 1.00 0.82 1.06
Reference 8.21-0.82 Reference 0.08-6.65 0.15-22.53 0.09-10.01 Reference 0.02-0.93
0.967
0.785 0.630 0.863
0.043
Reference 0.58-11.55 0.28- 7.33 Reference 0.20-3.31 0.11-11.64
0.209 0.653
0.784 0.948
1.00 0.30 0.55
Reference 0.08-1.17 0.11-2.64
0.084 0.459
n/a=not applicable
62
Astuti et al.
Table 2. Several work characteristics and risk of moderate qualitative work load stressor
Qualitative work load stressor Slight Moderate/severe (n=13) (n=122) Job title Operator Supervisor Working unit Air control service Aerodrome control/approach Total length of employment 0-5 years 6-10 years 11-15 years 16-30 years Experience controlling control near miss accident Never Near miss Accident Stress management training Ever Never n/a=not applicable 11 2 7 6 3 6 2 2 95 27 57 65 33 16 38 35 1.00 1.56 1.00 1.33 1.00 0.24 1.72 1.59 Reference 0.33-7.48 Reference 0.42-4.19 Reference 0.05-1.10 0.27-19.10 0.25-10.13 0.576 Crude odds ratio 95% condence interval P
0.626
0.066 0.562 0.623
8 5 0 1 12
58 59 5 35 87
1.00 1.63 n/a 1.00 0.21
Reference 0.50-5.26
0.416
Reference 0.02-1.65
0.137
Table 3. Some environment work characteristics and risk of moderate qualitative work load stressor
Qualitative work load stressor Slight Moderate/severe (n=13) (n=122) Working room lighting Bright Dim Noise Normal Noisy Working chair Comfortable Uncomfortable Working room space Not crowded Crowded Work station Comfortable Uncomfortable 9 4 62 60 1.00 2.18 Reference 0.64-7.45 0.215 12 1 99 23 1.00 2.79 Reference 0.34-22.54 0.336 6 7 64 58 1.00 0.78 Reference 0.25-2.25 0.666 12 1 7 6 110 12 64 58 1.00 1.31 1.00 1.06 Reference 0.16-10.96 Reference 0.34-3.33 0.804
Crude odds ratio
95% condence interval
0.924
63
Table 4. Several home stressors and risk of slight and moderate qualitative work load stressor
Qualitative work load stressor Slight Moderate/severe (n=13) (n=122) Family role Low Moderate Severe Physical home stressor Low Moderate Severe Home tension Low Moderate Severe Personal privacy Always Seldom Rare n/a=not applicable 3 9 1 12 1 0 10 3 0 5 7 1 28 75 19 92 28 2 91 30 1 40 63 19 Crude odds ratio 1.00 0.89 2.03 1.00 3.65 n/a 1.00 1.10 n/a 1.00 1.13 2.38 95% condence interval Reference 0.23-3.53 0.20-21.07 Reference 0.45-29.33 P
0.872 0.551
0.223
Reference 0.28-4.26
0.891
Reference 0.33-3.79 0.26-21.77
0.849 0.44
Table 5. The relationship among working room temperature and some of stressors and risk of qualitative work load stressor
Qualitative work load stressor Slight Moderate/severe (n=13) (n=122) Working room temperature Cold Too cold Role ambiguity stressor Low Moderate/severe Personal responsibility stressor Low Moderate/severe Career development stressor Low Moderate/severe 5 8 20 102 1.00 3.67 Reference 0.88-15.35 0.075 2 11 6 116 1.00 6.64 Reference 1.13-38.85 0.036 1 2 48 6 1.00 8.23 Reference 1.13-59.90 0.010 0.037 3 10 9 113 1.00 10.63 Reference 1.79-65.59 0.010 Adjusted odds ratio 95% condence interval
*Adjusted each others for risk factors listed on this Table.
64
Astuti et al.
DISCUSSION This study has some limitations such as limited subjects and a relatively high number of non-respondents among eligible subjects. Most of the non-respondent subjects were on leave or training. Though we explained as clearly as possible the questionnaire, the answers were based on subjects perceptions. The working room temperature in the radar controller room was 1819 C, and in the tower room the temperature was 20C. Prevalence of the low personal responsibility (6.0%) was lower than reported in the Police study,11 and newspaper worker study.12 Since the ATCs worked as a team, they shared the responsibility for passenger safety within the group and with other sectors. Our nal model revealed that subjects with low versus moderate-high personal responsibility stressor had a 6.4 times increased risk for slight qualitative work load stressor. This might be due to ATCs who felt low personal responsibility would feel the load of qualitative work was also low. This result was similar with an earlier study that found a relation between personal responsibilities and work stressor among the newspaper workers.12 Low role ambiguity among the ATCs (63.7%) was higher than that in the Police Brigade study (44.2%),11 and with the newspaper worker study (39.1%).12 Low role ambiguity among the ATCs is caused by duty, reporting system, competency and responsibility of ATCs to obey standard operation procedures. The model also showed subjects with low role ambiguity had a 10.59 (p 0.032) increased risk of the slight qualitative work load stressor compared to the moderate-high role ambiguity stressor. The reason was ATCs with low role ambiguity stressor had better self condence in doing their job with the slight qualitative work load. ATCs with slight qualitative work load (10.4%) had a value lower than that in the Police Brigade study (13.2%),11 or with the newspaper worker study (29.6%).12 Even with responsibility for an increased frequency of airplane ights, ATCs were supported with high technology equipment such as telecommunication, radar etc that would reduce workload difculties. Career development stressor of ATCs (18.5%) was lower than that in the Police study (13.2%),11 or
newspaper worker study(31.3%).12 Since ATCs had specic professional jobs and a slim organizational structure they understood that their career development was limited. Result showed the subjects with low development career stressor had a 2.7 increased risk of slight qualitative work load compared to the mediumhigh career development stressor (P=0,161) Of 135 subjects, 5 (5,1%) were unmarried, 2 divorced and the others (94,9%) were married. Our study shows that compared to married subjects, unmarried subjects had a 78% lower risk for a slight qualitative workload stressor. This might be due to household problems over duty, economical problems in the family that married ATCs might have and which could result an increased qualitative workload stressor. Our model shows that subjects with the perception that the room temperature was uncomfortable had a 22-fold increased risk for moderate qualitative workload. The temperature of the working rooms was relatively low. This cold working room temperature was not for personnel but to preserve equipment. Hence, this condition needs to be managed to control the side effects of cold temperature. For example, warmer jackets could be provided. In conclusion, a too cold room temperature and other moderate/severe stressors increased the risk of moderate qualitative workload stressor for ATCs. Acknowledgments We thank to Chief of Air Trafc Service Management to allow us to conduct this study. Special gratitudes for ATCs who were cooperatively participated in this study. We also thank Dr. Elisabeth Emerson for reviewing the nal draft and her excellent suggestions. REFERENCES
1. Human Factors Digest No. 8. ICAO Circular. Human factors in air trafc control. Montreal. ICAO. 1993 p. 33-4. NIOSH. Stress at work. [Cited 2008 Oktober 22]. Available from http://www.cdc.gov/niosh/topics/stres/ General Directorate of Air Trafc of Indonesia. Air trac services. [Cited 2008 Mei 14]. Available from http://www.dephub.go.id Top 10 Most stressful jobs. [Cited 2008 October 20]. Available from www.cdc.gov/ulcer/myth.htm Costa G. Occupational stress and stress prevention in air trafc control: Literature review. [Cited 2008
2. 3.
4. 5.
65
October 22]. Available from www.ilo.org/public/ english/protection/condtrav/pdf/wc-gc-95.pdf. 6. Sena A. Work environment and stress management in air trafc controllers at Soekarno-Hatta airport [thesis]. Tangerang. Indonesia Aviation College. 2006. 7. Isfandari S. Instrument survey study on stress and stress strain diagnosis. Jakarta: National Institute of Health Research & Development; 1992. 8. Kaplan PS, Stein J. Psychology of adjustment. California: Wadsworth Publishing Company; 1984. 9. Faculty of Medicine of Universitas Indonesia. Field study guidance. Jakarta. The Faculty. 2003. 10. Barros AJD, Hirakata VN. Alternative for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimates the prevalence ratio. BMC Med Research Method. Oct.
11.
12.
13.
14.
2003;3. [Cited 2008 October 20]. Available from http// www.biomedcentral.com/1471-2288/3/21. Kanam R. Qualitative work overload and other risk factors related to hypertension risk among the Police Mobile Brigade (Brimob), Med J Indones 2008; 17:188-196. Setiawan ZY. Work stress and mental health trend of mental emotional in news paper editors in Jakarta [thesis]. Jakarta. Univ Indonesia. 2006. Costa G. Working and health conditions of Italian air trafc controllers. Int J Occupy Safety Ergonomic. 2001; 6:365-82 [Cited 2008 Oktober 20]. Available from http://www.ciop.pl/807.html Widyahening IS, High level of work Stressors increase the risk of mental-emotional disturbances among airline pilots. Med J Indones 2007; 16:117-21.
66
Jekti and Suarthana
Risk factors of post partum haemorrhage in Indonesia

Rabea Pangerti Jekti,1 Eva Suarthana2
Centre for Biomedical and Applied Health Technology, National Institute of Health Research and Development, Ministry of Health Republic Indonesia 2 Department of Community Medicine, Faculty of Medicine, Universitas Indonesia
1
Abstrak
Latar belakang: Perdarahan post-partum (PPH) merupakan salah satu trias klasik penyebab kematian ibu. Studi ini mengevaluasi beberapa faktor risiko PPH, khususnya riwayat antenatal, natal, dan post-natal. Metode: Analisis menggunakan sebagian data dari studi potong lintang Riset Kesehatan Dasar (Riskedas) 2010. Subjek yang dipakai unuk analisis ini ialah wanita yang menikah berumur 13-49 tahun dan melahirkan anak terakhir antara 1 Januari 2005 sampai 31 Juli 2010. Perdarahan post-partum berdasarkan konrmasi petugas kesehatan tentang telah terjadinya perdarahan dua atau lebih kain (masing-masing 1,5 m) selama proses persalinan. Hasil: Pada analisis ini terdpat 601 subjek yang mengalami PPH dan 19.583 subjek tidak mengalami PPH. Eklamsia meningkatkan risiko PPH 3,5 kali (95% interval kepercayaan (CI) = 2,534,69), ketuban pecah dini meningkatkan risiko PPH 2,2 kali (95% CI = 1,69-2,83), placenta previa meningkatkan risiko PPH 2,1 kali (95% CI = 1,29-3,31). Dibandingkan kehamilan aterm, wanita dengan kehamilan prematur berisiko PPH 82% lebih tinggi (95% CI = 1,332,49), sedangkan yang dengan kehamilan post-term berisiko PPH 72% lebih tinggi (95% CI = 1,162,57). Dibandingkan wanita dengan paritas 1-2, risiko PPH pada wanita yang berparitas 3-5 dan 6 atau lebih berturut-turut adalah 24% dan 81% lebih tinggi. Kesimpulan: Eklampsia merupakan faktor risiko PPH terkuat. Placenta previa, ketuban pecah dini, kehamilan prematur atau post-term, serta paritas yang tinggi juga meningkatkan risiko PPH. (Health Science Indones 2011;2:66-70) Kata kunci: perdarahan postpartum, eklampsia
Abstract Background: Post-partum haemorrhage (PPH) is one of the classic triad of causes of maternal death. This analysis aimed to evaluate several risk factors of PPH. Methods: This analysis using a cross-sectional Basic Health Research (Riskesdas) 2010 data. For this analysis, the subjects consisted of married women aged 13-49 years, who gave birth of their last child between January 1, 2005 and August 2010, who had a probability of PPH history. The PPH was dened as bleeding more than two wet pieces materials, 1.5 m each, during giving birth. Results: This analysis noted 601 subjects had PPH and 19,583 subjects did not have PPH. Post-partum haemorrhage related to demographic (education level, and economic level), gynecologic (parity) as well as obstetric factors. Those who had than did not have eclampsia had 3.5-fold PPH [95% condence interval (CI) = 2.534.69]. Those who had than did not have premature rupture of the membranes had 2.2-fold PPH (95% CI) = 2.534.69). Those who had than did not have placenta previa had 2.1-fold PPH (95% CI) =1.293.31). In term of uterine rupture, those who had than did not uterine rupture had 65% increase PPH (95% CI) = 1.112.46). Compared to women with 1-2 parity, women with 3-5 and 5 or more parity had an increased PPH risk for 24% and 81% respectively. Conclusion: Eclampsia was the strongest risk factor of PPH. Other risk factors of PPH include premature rupture of the membranes, placenta previa, premature or post-term pregnancies, and high parity. (Health Science
Indones 2011;2:66-70)
Key words: post-partum haemorrhage, eclampsia
Corresponding author: Rabea Pangerti Jekti E-mail: yekti.yekti.24@gmail.com
Eclampsia and post partum haemorrhage
67
In developing countries, maternal mortality rates (MMR) are still higher than 100 women per 100,000 live births.1 The World Health Organizations statistics suggest that 25% of the maternal deaths are due to PPH, accounting for more than 100,000 maternal deaths per year.2 It was estimated that there were 140,000 maternal deaths per year or 1 woman died every 4 minutes.3 The Indonesian Demographic and Health Survey (IDHS) in 2007 indicated that the MMR in Indonesia was the highest in Asia (228 per 100,000 live births). This rate was about 3-6-time higher than MMRs in South East Asian countries and more than 50 times MMR in developed countries. The National MediumTerm Development Plans (RPJMN) in 2005-2009 targeted a reduction of MMR from 390 in 1990 to 228 per 100,000 live births in 2007.4 According to the Ministry of Health Report in 1998, the main causes of maternal mortality (more than 90%) in Indonesia are the triad classic, namely haemorrhage (40%-60%), toxemia gravidarum (20%-30%) and infection (20%-30%). The causes of the classic triad are known as the three late: too late to recognize the danger signs of the pregnancy, too late to refer the mother to a referral center, and too late to get help by the health provider.5 The National Institute of Health Research and Development conducted the Basic Health Research (Riset Kesehatan Nasional or Riskesdas) in 2010. One aspect observed in Riskesdas 2010 was the health of the pregnant women. Therefore, Riskesdas 2010 was expected to provide evidence-based data about post partum haemorrhage (PPH). This analysis aimed to evaluated several risk factors PPH
mean to evaluate the development of several health status of the Indonesian people; changes in health problems; as well as the progress of health development efforts at the national and provincial level in a three-year period. Some indicators that were collected include nutritional status of children (e.g. starvation); maternal and child health status (e.g., lower child mortality and improved maternal health); the prevalence of malaria and tuberculosis (e.g., decrease morbidity); access to drinking water sources; as well as safety and basic sanitation facilities. Data was collected through interviews and measurements. Laboratory tests for diagnosis of malaria and tuberculosis was performed in the eld at the referred public health centers (blood sample for malaria testing and sputum for tuberculosis testing). Riskesdas was approved by the ethical committee of the National Institute of Health Research and Development of the Ministry of Health of the Republic Indonesia. For this analysis, the subjects consisted of married women aged 13-49 years, who gave birth of their last child between January 1, 2005 and August 2010, who had a probability of PPH history. The study consists of 601 subjects with positive PPH history and 19,583 subjects with negative PPH history. In Riskesdas 2010, PPH was dened as bleeding more than two wet pieces materials, 1.5 m each, during giving birth. Independent variables consisted of demographic characteristics (living in urban or rural area, age of the subject in 2010, educational level, employment status, marital status, and economic status). Antenatal, natal and postnatal history characteristics include age at last birth (age of the subject during the last delivery); parity (the number of born children); birth spacing (interval between the last and the previous child), knowledge about pregnancy (danger signs of pregnancy and childbirth), iron tablet consumption during pregnancy, caesarian-section or non-vaginal delivery, eclampsia or eclampsia (characterized by leg swelling, hypertension, and/or seizures), uterine rupture (the incidence of uterine tear during childbirth), the premature rupture of the amniotic sac (amniotic discharge six hours or more before the child was born), placenta previa (the birth canal was blocked by the placenta). The presence of complications during pregnancies and childbirth was conrmed by the health workers. Independent variables were generally dichotomized, except for employment status (i.e., student, housewife; government employee/
METHODS This analysis used a cross-sectional study Basic Health Research (Riskesdas) 2010 data. Riskesdas sampling followed the National Health Survey (SUSENAS) sampling frame. With a larger number of samples than the National Health Survey, Riskesdas data could better describe the health prole of the districts, cities, or provinces in Indonesia. Riskesdas 2010 provides specic information on the health Millenium Development Goals (MDG) according to the commitment of global health efforts at national level. In addition, it is also a
68
Jekti and Suarthana
army/police; labor/ farmer/ shermen; and others), age during the last gestation (i.e., 13-19 years, 20-34, and 35-49 years), parity (i.e.1-2 children, 3-5 children, > 5 children or grand multipara), and term of pregnancy (preterm, aterm, post-term). Analysis was done using logistic regression in Stata 9.0. Independent variables with a statistical signicance of p <0.25 were included in the multiple regression analysis. Backward stepwise selection was applied to obtain the nal model with a statistical signicance level of p <0.05. RESULTS Total number of Riskesdas 2010 samples was 91,711 subjects, of which 59,382 were married. Of 19,506 women who had a probability of having post partum
haemorrhage, 18,905 (96.9%) did not have PPH, while 601 (3.1%) did have PPH. Table 1 reveals that those who had and who did have PPH similarly distributed with respect of residence, marital status, employment status, age birth, and birth spacing. On the sides, those who had higher education, government/army/police employment, lowlowest economic level had higher risk experiencing PPH compared to respective references. Our nal model reveals (Table 2) that PPH related to demographic (education level, and economic level), gynecologic (parity) as well as obstetric eclampsia, premature rupture of the membranes, placenta previa, uterine rupture, and premature or post-term birth] factors.
Table 1. Univariable association between some demographic characteristics and post partum haemorrhage
Post partum haemorrhage No Yes (n=18,905) (n=601) n % n % Residence Urban Rural Age 1319 years 2034 years 3549 years Marital status Married Single parent Education level None PrimaryJunior high school Senior high school above Employment status Housewife, student Government, army, police Labor, farmer, shermen Others Economic level Middleupper Lowlowest Age birth 1319 years 20-34 years 3549 years Birth spacing 24 months 23 months 9,254 9,651 504 13,363 5,038 18,806 99 6,926 11,583 396 9,683 3,768 3,481 1,973 10,179 `8,726 1,331 14,328 3,246 17,361 1,544 48.9 51.1 2.7 70.7 26.6 99.5 0.5 36.6 61.3 2.1 51.2 20.0 18.4 10.4 53.8 46.2 7.0 75.8 17.2 91.8 8.2 307 294 15 412 174 598 3 186 395 20 308 98 127 68 303 298 37 455 109 552 49 51.1 48.9 2.5 68.5 29.0 99.5 0.5 31.0 65.7 3.3 51.3 16.3 21.1 11.3 50.4 49.6 6.2 75.7 18.1 91.8 8.2 P Crude odds ratio 95% Condence interval
1.00 0.92 1.00 1.04 1.16 1.00 0.98 1.00 1.27 1.88 1.00 0.82 1.15 1.08 1.00 1.15 1.00 1.14 1.20 1.00 0.99
Reference 0.781.08 Reference 0.611.75 0.681.98 Reference 0.671.44 Reference 1.061.51 1.173.02 Reference 0.651.03 0.931.42 0.831.42 Reference 0.981.35 1.00 0.81 1.61 0.83 1.76 1.00 0.74-1.34
0.304
0.895 0.586
0.935
0.008 0.009
0.087 0.201 0.556
0.097
0.443 0.328
0.990
Eclampsia and post partum haemorrhage
69
Table 2. Relationship between several gynecologic and obstetric characteristics and post partum haemorrhage
Post partum haemorrhage No Yes (n=18,905) (n=601) n % n % Adjusted odds ratio* 95% Condence interval Reference 2.534.69 P
Eclampsia No 18,506 97.9 540 89.9 1.00 Yes 399 2.1 61 10.1 3.45 Premature rupture of the membrane No 17,872 94.5 510 84.9 1.00 Yes 1,033 5.5 91 15.1 2.18 Placenta previa No 18,705 98.9 576 95.8 1.00 Yes 200 1.1 25 4.2 2.07 Uterine rupture No 18,548 98.1 566 94.2 1.00 Yes 357 1.9 35 5.8 1.65 Parity 12 children 12,649 66.9 364 60.6 1.00 35 children 5,506 29.1 198 32.9 1.25 6 or more children 750 4.0 39 6.5 1.81 Birth term Aterm 17,698 93.6 525 87.3 1.00 Preterm 734 3.9 48 8.0 1.82 Post-term 473 2.5 28 4.7 1.72 *Adjusted each other among risk factors listed on this Table, education level, and economic level
0.001
Reference 1.692.83 Reference 1.293.31 Reference 1.112.46 Reference 1.03-1.48 1.232.47 Reference 1.332.49 1.162.57
0.001
0.002
0.014
0.021 0.002
0.001 0.008
The strongest obstetric factor was eclampsia. Those who had than did not have eclampsia had 3.5-fold experiencing of PPH. Those who had than did not have premature rupture of the membranes had 2.2fold experiencing PPH. Those who had than did not have placenta previa had 2.1-fold experiencing PPH. In term of uterine rupture, those who had than did not uterine rupture had 65% increase of experiencing PPH. Furthermore, more parities, pre-term as well as postterm birth increased PPH risk. DISCUSSION This study shows that pre- or post-term pregnancy, uterine rupture, placenta previa, premature rupture of the membranes, and eclampsia were risk factors of PPH. The operational denition of the study variables, including eclampsia, were not well dened as in hospital-based studies, which could cause selection bias. Second, Riskesdas data were based on recall, which may also cause recall-bias. Nevertheless,
Riskesdas 2010 data is the largest community-based study in Indonesia. It has rich data for evaluating the MDGs, especially in reproductive health. A review by Jouppila stated that most PPH cases are caused by uterine atony, maternal soft-tissue trauma, retained placenta or its parts, and obstetric coagulopathy.6 The factors most signicantly associated with haemorrhage include advanced maternal age, prolonged labor, eclampsia, obesity of mother, multiple pregnancy, a birth weight of more than 4000g, and previous postpartum haemorrhage.6 A twenty-year cohort study (1978-1997) by Kramer and colleagues demonstrated that major independent risk factors for PPH included prior Caesarean section, placenta previa or low lying placenta, marginal umbilical cord insertion in the placenta, transverse lie, labour induction and augmentation, uterine or cervical trauma at delivery, and gestational age < 32 weeks.7 In concordance, in this study we found that multiparity, eclampsia, premature rupture of the membranes and placenta previa were strong risk factors of PPH (OR > 2). We also found that uterine
70
Jekti and Suarthana
rupture, pre-term as well as post-term birth increased the risk of PPH. Eclampsia is a vascular endothelial disorder, which is clinically dened by hypertension and proteinuria, with or without pathologic edema. Eclampsia can range from mild to severe. If uncontrolled, eclampsia could lead to maternal death or cause growth restrictions of the fetus. The 1994 Household Health Survey found the major causes of maternal deaths in Indonesia included haemorrhage, eclampsia, infections, and obstructed labor.8 Our study revealed that eclampsia was the strongest risk factor of PPH. We found risk of PPH increased with increasing level of education. An earlier evaluation of the factors inuencing complication during delivery in Indonesia using the 2007 IDHS dataset shows that surprisingly breaking of water excessive vaginal bleeding during delivery increased with increasing level of education, qualication of health services, frequency of antenatal care attendance. The risk of these complications was higher respondent lived in urban area compared those who lived in rural area. The author argued that health aspects of health and safe delivery are not included in the subjects of formal education in Indonesia. Therefore, increasing education does not have direct correlation with an increase in knowledge about delivery complications. She also hypothesized that respondents choose better health services for delivery when they experience complications. Data on antenatal care visit shows 78.7% of respondents were very active in attending of antenatal care (more than 4 times) and 21.3 % attended antenatal service less than the recommended three antenatal visits. This might suggest that respondents who though that they had any symptom of complications, visited antenatal care more frequently.9 In conclusion, eclampsia was the strongest risk factor of PPH. Other risk factors of PPH included premature
rupture of the membranes, placenta previa, premature or post-term pregnancies, and high parity. Our ndings imply the need of increasing awareness on health and safe delivery. In particular, early identication and prompt treatment for pregnant women with hypertension by health workers during antenatal case visits. REFERENCES 1. Hogan MC, Foreman KJ, Naghavi M, et al. Maternal
mortality for 181 countries, 19802008: a systematic analysis of progress towards Millennium Development Goal. The Lancet. 2010; 375:1609-23. Abouzahr C. Antepartum and postpartum haemorrhage. In: Murray CJ, Lopez AD, eds. Health dimensions of sex and reproduction. Boston, Mass: Harvard University Press; 1998:172-4. American College of Obstetricians and Gynecologists (ACOG). Practice Bulletin #76: Postpartum hemorrhage. Obstet Gynecol. 2006;108:1039-47. Indonesia National Planning and Development Board. A road map to accelerate achievements of the MDGs in Indonesia. Jakarta: The Board; 2010. Ministry of Health of Idonsia. Acceleration effort to decrease maternal mortality rate. Jakarta: The Ministry; 1998. Indonesia. Jouppila P. Postpartum hemorrhage. Curr Opin Obstet Gynecol. 1995;7:446-50. Kramer MS, Dahhou M, Vallerand D, et al. Risk factors for postpartum hemorrhage: can we explain the recent temporal increase? J Obstet Gynaecol Can. 2011;33:810-9. Central Bureau of Statistics (CBS), National Family Planning Coordinating Board, Ministry of Health and Macro International Inc. Indonesia Demographic and Health Survey 1994. Jakarta: CBS; 1995. Reviani N. Factors inuencing complication during delivery in Indonesia 2007 [Thesis]. Adelaide: The Flinders University of South Australia; 2010.
2.
3. 4. 5. 6. 7.
8.
9.
Symptom of cmplications and risk of c-section
71
Signs or symptoms of complications in pregnancy and risk of caesarean section: an Indonesia national study
Suparmi,1 Bastaman Basuki2
1 2
The National Institute of Health Resesarch and Development, Indonesian Ministry of Health Department of Community Medicine, Faculty of Medicine Universitas Indonesia Abstrak Latar belakang: Pada beberapa tahun terakhir kejadian seksio sesarea (c-sesarea) meningkat. Tujuan analisis ialah untuk mengidentikasi beberapa tanda atau gejala yang berbahaya selama kehamilan terhadap c-sesarea. Metode: Analisis ini memakai sebagian data Riset Kesehatan Dasar (Riskesdas) 2010. Sub-sampel dengan metode multistage stratied sampling di seluruh Indonesia di antara wanita yang menikah atau pernah menikah berumur 10-49 tahun yang melahirkan bayi antara 1 January 2005 sampai 31 August 2010. Analisis mempergunakan regresi Cox dengan waktu konstan. Hasil: Prevalensi c-sesarea sebesar 10,8% di antara 20.501 wanita. Rasio prevalensi kota dan desa ialah 2,9. Wanita yang pernah dibandingkan yang tidak pernah mengalami sebarang tanda atau gejala komplikasi kehamilan berisiko lebih besar mengalami c-sesarea. Wanita yang pernah dibandingkan yang tidak pernah mengalami demam tinggi berisiko 2,3-lipat c-sesarea [risiko relatif suaian (RRa) = 2,33; 95% interval kepercayaan (CI) = 1,69-3,34]. Wanita yang pernah mengalami dibandingkan yang tidak pernah mengalami perdarahan per vaginam berisiko 2,1-lipat mengalami c-sesarea (RRa = 2,12; 95% CI = 1.,5-2,58). Risiko yang terkecil (96%) terjadi pada wanita yang pernah mengalami kejang atau pingsan (RRa = 1.96; 95% CI = 1,41-2,73). Kesimpulan: Wanita yang pernah dibandingkan yang tidak pernah sebarang tanda atau gejala komplikasi kehamilan berisiko lebih besar mengalami c-sesarea. (Health Science Indones 2011;2:71-6) Kata kunci: seksio sesarea, persepsi masyarakat, komplikasi, , Indonesia Abstract Background: In the last years, the frequency of cesarean section (c-section) has risen. This study was aimed to identify several signs or symptoms of complications during pregnancy increased the risk of c-section (c-section). Methods: Data were derived from the Basic Health Survey (Riskesdas) 2010. The sub-sample was married or divorced women aged 10-49 years between January 1, 2005 and August 2010 in Indonesia based on multistage stratied sampling methods. Analysis used Cox regression with constant time. Results: The c-section rate was 10.8% among 20,501 women. Urban and rural ratio of c-section rate was 2.9. Women who reported than who did not report any signs or symptoms of complications during their pregnancies had a higher risk of c-section. Women who reported high fever had 2.3-fold for c-section [adjusted relative risk (RRa) = 2.33; 95% condence interval (CI) = 1.69-3.34]. Moreover, those who reported compared to those who did reported bleeding had 2.1-fold increase risk of c-section (RRa = 2.12; 95% CI = 1.75-2.58). The lowest risk (96%) was among those who ever had convulsion/fainted (RRa = 1.96; 95% CI = 1.41-2.73). Conclusion: Women who reported any signs or symptoms of complications during their pregnancies had an increased risk of c-section. (Health Science Indones 2012;2:71-6) Key words: cesarean section, community perception, pregnancy complications, Indonesia
Corresponding author: Suparmi E-mail: suparmi.mi@gmail.com
72
Suparmi and Basuki
In developing countries, the frequency of cesarean section (c-section) has risen in the past few years.1 Prior studies reported an inverse association between c-section rate and maternal and infant mortality at population level in low income countries.1,2 Although c-section is considered relatively safe, it poses a higher risk of some complications than a vaginal delivery, in addition, risks of certain peripartum complications have long been associated with c-section.3 Furthermore, c-section rates above a certain limit have not shown additional benet for the mother or the baby, and some studies have reported that high c-section rates could be linked to negative consequences in maternal and child heath.4 The trend toward increasing c-section suggests that the incidence of those complications might also be on the rise. On the other hand, the impact of c-section trends might be modied by changes in population health or improvements in obstetric care.5 Most prior studies, however, presented clinical complications related to c-section risk. Few reviewed community perceptions of medical complications during pregnancy related to c-section. This study aimed to identify several community perceptions on signs and symptoms that indicate complications in pregnancy related to the risk of c-section. METHODS The data analyzed originated from the Basic Health Survey (Riskesdas) 2010 of Indonesia. Riskesdas 2010 is a cross sectional survey which provided specic information on the health Millennium Development Goals (MDG) according to the commitment of global health efforts at the national and provincial level.6 Sampling was multistage stratied sampling. Specially trained interviewers collected data using the questionnaire. The interviewers consisted of 104 teams. Each team consisted of one supervisor, one eld editor and data entry, and two interviewers. Some indicators collected include nutritional status of children, maternal and child health status, malaria and tuberculosis, access to drinking water sources, as well as the safe and basic sanitation facilities. Data were collected through interviews and measure-
ments. Laboratory tests for diagnosis of malaria and tuberculosis were performed in the eld at the respective referred public health centers. The eligible population was all households in the entire Republic of Indonesia having equal probability of being included. Nationally representative sample of Riskesdas 2010 was 33 provinces with over 441 districts/cities of the total 497 districts/cities in Indoesia. The interview was held in the respondents home. The subjects consisted of 69,300 households, with 251,388 respondents, and 20,591 ever or still married women age 10-59 years. The sub-sample included in the analysis were women meeting all criteria: (1) who gave birth of their last child between January 1, 2005 and August 2010, (2) who did not have ectopic pregnancy, (3) aged 10-49 years. For this analysis, 30 subjects were excluded because of ectopic pregnancy, 50 women because their age were 50-59 years, and 10 subjects because of incomplete data, leaving 20,501 subjects for this analysis. Riskesdas 2010 study was approved by the Ethical Committee of National Institute of Health Research and Development (NIHRD) Ministry of Health of Republic of Indonesia. For this analysis, women were classied as having a c-section if the c-section was either done at a government, private, or maternity hospital and if assisted by medical doctors. Risk factor variables consisted of suspected complications during pregnancy, baby size at birth, antenatal care (ANC), term of delivery, and demographic characteristics (i.e age group, education, working status, place of living and levels of expenditure per capita). Perception on signs and symptoms that indicate pregnancy was threatened based on mothers report consisted of 6 subgroups (none, very painful stomach ache, bleeding, high fever, convulsion/collapse, and others) Baby size was consisted of ve subgroups based on mothers perception of the baby size after birth (average, very small, small, large, and very large). Antenatal care was divided into two subgroups (complete and incomplete). Women were classied as having a complete ANC if they mentioned height and weight measurements, blood pressure measurement, urine sample taken, abdominal examination, and
73
informed of signs of pregnancy complications. Term of delivery was divided into 3 subgroups (aterm, preterm and postterm). Women were classied as aterm delivery if they delivered in 9 months of pregnancy; preterm if they had delivery in 7-8 months of pregnancy; and postterm if they had delivery in 10 months of pregnancy. Maternal age was into 3 subgroups (13-19 years, 2034 years, and 35-49 years). Education was based on the last education obtained by the respondent (none, primary education, secondary education or above). Never and no primary education were grouped together into none, primary and junior high school were grouped together into primary education, high school or above were grouped together. Working status was divided into housewife/student, private/government employee, farmer/sherman/laborer, and others. Place of living was divided into urban and rural. Level of expenditure per capita originally was divided into 5 levels, with quintile 5 as the highest quintile and the lowest or poorest quintile was 1.
To assess the risk of c-section, the data were analyzed by Cox regression with constant time.6 The analysis used STATA 9.0 software. NIHRD gave permission to analyze part of Riskesdas 2010 data. RESULTS The youngest subject who had c-section was 16 years old. Table 1 showed the women who had c-section rate was 10.8% (2,217 among 20,501 women, and urbanrural c-section rate ratio was 2.9 16.1%/5.5%). Most of the subjects had lower education (52.2%), and aged 20-34 years (70.7%). Table 1 showed that the highest occurrence of c-section was found in women with 3549 years of age, secondary education or above, private/government employee, living in urban area, and has the highest level of expenditure per capita subjects. Furthermore Table 1 noted that older, private/government employee or farmer/sherman/laborer, living in urban area, higher parities, term as well as postterm pregnancy, complete ante natal care, and more wealthy
Table 1. Several demographic factors and the risk of c-section

Delivery Vaginal C-section (n=18,284) (n=2,217) n % n % Residence Rural Urban Age group 13-19 years 20-34 years 35-49 years Education None Primary education Secondary education or above Working status Housewife/student Private/government employee Farmer/sherman/laborer Others Level of expenditure Lowest Second Middle Fourth Highest 9,723 8,561 516 12,968 4,800 2,422 10,003 5,859 9,404 3,206 3,761 1,913 4,889 4,274 3,833 3,165 2,123 94.5 83.9 93.3 89.5 88.1 95.7 93.5 80.6 89.8 80.7 95.3 90.5 96.0 93.5 90.1 85.0 74.2 571 1,646 37 1,529 651 108 700 1,409 1,063 767 185 202 203 296 420 560 738 5.5 16.1 6.7 10.6 11.9 4.3 6.5 19.4 10.2 19.3 4.7 9.6 4.0 6.5 9.9 15.0 25.8 Crude relative risk 95% condence interval
1.00 2.91 1.00 1.58 1.79 1.00 1.53 4.54 1.00 1.90 0.46 0.94 1.00 1.63 2.48 3.77 6.47
Reference 2.64-3.20 Reference 1.14-2.18 1.28-2.49 Reference 1.25-1.88 3.73-5.52 Reference 1.73-2.09 0.39-0.54 0.81-1.09 Reference 1.36-1.94 2.09-2.93 3.21-4.43 5.54-7.56
0.000
0.006 0.001
0.000 0.000
0.000 0.000 0.424
0.000 0.000 0.000 0.000
74
Suparmi and Basuki
were more likely to have c-section deliveries. Table 2 noted that women with higher parity were less likely to have c-section deliveries. On the hand, those who with preterm as well as postterm pregnancies, complete ante natal care, and did not deliver normal baby were more likely to be c-section delivery. Table 3 showed that women who did not have any signs or symptoms of complications during their pregnancies were 93.5% (1,865/19,172). The most common (1.3%) reported signs or symptoms of complication during
their pregnancies was bleeding (276/19,172), and 1.2% women was very severe abdominal pain (256/19,172). The least were fever, 0.6%, and convulsions, 0.7% (113/19,172 and 138/19,172 respectively). In the nal model (Table 3), revealed that compared to women who did not report complication before their labors, all women who reported complications during pregnancy had a higher risk of experiencing c-section. Women who reported high fever had 2.3-fold risk of experiencing c-section. The lowest risk was among women who reported very severe abdominal pain,
Table 2. Several gynecologic factors and the risk of c-section

Delivery Vaginal (n=18,284) n Parity 1-2 3-5 6-15 Pregnancy Aterm Preterm Postterm ANC completeness Incomplete Complete Baby size Average Very small Small Large Very large 11,862 5,612 810 17,224 625 435 16,826 1,458 13,082 112 1,057 3,493 540 % 87.6 91.7 96.2 89.8 77.5 83.2 90.1 79.9 90.5 73.2 85.9 86.6 85.7 n 1,676 509 32 1,948 181 88 1,851 366 1,374 41 173 539 90 C-section (n=2,217) % 12.4 8.3 3.8 10.2 22.5 16.8 9.9 20.1 9.5 26.8 14.1 13.4 14.3 1.00 0.67 0.31 1.00 2.21 1.66 1.00 2.03 1.00 3.49 1.56 1.47 1.59 Reference 0.61-0.74 0.22-0.44 Reference 1.90-2.57 1.34-2.05 Reference 1.81-2.27 Reference 2.43-5.01 1.31-1.85 1.32-1.63 1.26-1.99 Crude relative risk 95% condence interval P
0.000 0.000
0.000 0.000
0.000
0.000 0.000 0.000 0.000
Table 3. Sign or symptoms of complication during pregnancies that indicate threatened pregnancy and risk of c-section
Delivery Vaginal (n=18,284) n % Complication during pregnancy None Very severe abdominal pain Bleeding High fever Convulsion/collapse Others 17,307 221 199 87 111 359 90.3 86.3 72.1 77.0 80.4 65.8 C-section (n=2,217) n % 1,865 35 77 26 27 187 9.7 13.7 27.9 23.0 19.6 34.2 Adjusted relative risk* 95% condence interval P
1.00 1.21 2.12 2.33 1.96 2.57
Reference 0.89-1.67 1.75-2.58 1.69-3.34 1.41-2.73 2.25-2.94
0.221 0.000 0.000 0.000 0.000
*Adjusted for residence, age group, education, working status, parity, antenatal care completeness, level of expenditure, and baby size.
75
which had a 21% increase in experiencing c-section.
recommendation is questionable.9 The nding showed that the percentage of women who did not have any signs or symptoms of complications during their pregnancies was 93.5% out of 19,172. This number was higher than previous report (Indonesia Demography and Health Survey IDHS 2007-2008) which noted that 89% out of 14,043 women reported no complications during their pregnancy.7 Moreover, the ndings revealed the most common complaints was bleeding (1.3%), followed by very severe abdominal pains (1.2%). The least ones were fever, 0.6% and convulsion,0.7%. On the other hand, the IDHS report revealed among those who reported complications, 3% had excessive vaginal bleeding, and 1% had fever. 7 Bleeding, fever, and convulsion were relatively small percentage, but these three complaints increased the risk of c-section by about 2-fold for each (relative risk of 2.12; 2.33; and 1.96 respectively). Therefore, these three complaints are recommended to be considered as early warning potential factors which increased the risk of c-section and have to be detected during ANC visits. In conclusion, women who reported any signs or symptoms of complications during their pregnancies that indicate pregnancy were threatened had an increased risk of c-section. Acknowledgments The authors wish to express their sincerest gratitude to The National Institute of Health Research and Development, Indonesia Ministry of Health, particularly to Dr. Trihono for the use of datasets. REFERENCES
1. Stanton CK, Holtz SA. Levels and trends in cesarean birth in the developing world, Studies Fam Planning. 2006;37:418. Althabe F, Belizn JM. Caesarean section: the paradox. Lancet. 2006; 368 (9546):1472-3. [cited 2011 October 26]. Available at: http://www.thelancet.it/journals/ lancet/article/PIIS0140-6736(06)69616-5/full text World Health Report. The global numbers and costs of additionally needed and unnecessary caesarean sections performed per year: overuse as a barrier to universal coverage. The Organization [cited 2011 October 26]. Available at: http://www.who.int/healthsystems/topics/ nancing/healthreport/30C-sectioncosts.pdf
DISCUSSION Several limitations must be considered in interpreting the ndings. Firstly, this analysis was based on data from a national wide survey (Riskesdas 2010). Secondly, data on complications during pregnancy and baby size were based on community perception of medical complications during pregnancy and baby size women perceptions. Thirdly, the Riskesdas 2010 did not have detailed risk factors data related to c-section, such as, on placenta previa and abruption, breech position, cord prolapsed, failure to progress in labor, repeated caesarean sections, cephalopelvic disproportion, fetal distress, birth defects, and multiple births, and demand to have caesarean section from the subjects as well as from the respective medical doctors. In spite of these limitations, this study used a national wide survey with a large sample consisting of 20,591 ever or still married women. Our analysis noted that the rate of women who had c-section was 10.8%, and the ratio of c-section between urban and rural was 2.9 times. The c-section rate was found to be higher than previous reports based on calculated regional c-section rates in Southeast Asia, which was 5%.1 But this nding was almost similar to the best current estimate of the overall rate of c-section delivery in developing countries, which was 12%, based on a study using nationally representative data from 82 nations with a median reference year of 1996.1 On the other hand, ndings on urban-rural rate ratio was almost similar to the ratio among urban women in the developing world, which was on average, three times as high as those among rural women.8 Indonesia is a developing country with both high rates of maternal mortality and a marked disparity in c-section rate between urban and rural women. The low c-section rate in rural showed that those who are at greatest risk for obstetric complications do not have adequate access to the procedure as stated in a prior study.1 Furthermore, the rate of women who had c-section was within range of the World Health Organization (WHO) recommendation that a nations c-section birth rate should be in the range of 515%.8 However, this
2.
3.
76
Suparmi and Basuki
4.
5. 6.
Belizn JM,Althabe F, Cafferata ML. Health consequences of the increasing caesarean section rates. Epidemiology. 2007;18: 485-6 [cited 2011 October 26]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/17568221 Ministry of Health Republic of Indonesia. Basic Health Survey Report 2010. Jakarta. The Ministry. 2011. Barros A, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an empirical comparison
7.
8. 9.
of models that directly estimate the prevalence ratio. BioMed Central. 2003;3:1-13. Badan Pusat Statistik-Statistics Indonesia (BPS), ORC Macro. Indonesia Demographic and Health Survey 2007-2008. Calverton, Maryland, USA: BPS and ORC Macro. 2008. World Health Organization. Appropriate technology for birth. Lancet. 1985;2:436-7. Chaillet N, Dub, Dugas M, et al. Identifying barriers and facilitators towards implementing guidelines to reduce caesarean section rates in Quebec. Bull World Health Organ. 2007 October; 85:7917.
Quality of drinking water
77
Quality of rell drinking water in Greater Jakarta in 2010

Arifayu Addiena Kurniatri
Center for Basic Biomedical and Health Technology, Institute for Health Research and Development, Ministry of Health Abstrak Latar belakang: Air yang layak minum kian langka di perkotaan. Air yang dikonsumsi masyarakat harus memenuhi persyaratan kualitas air minum yang aman bagi kesehatan. Banyak masyarakat yang memanfaatkan air minum isi ulang untuk memenuhi kebutuhan air minum sehari-hari. Pada kajian ini disajikan penilaian kualitas air minum isi ulang berdasarkan pemeriksaan parameter sik dan kimia pada tahun 2010. Metode: Desain penelitian adalah cross-sectional. Sampel didapatkan dari masyarakat yang mengajukan permintaan pemeriksaan kualitas air minum di Laboratorium Farmasi di Pusat Biomedis dan Teknologi Dasar Kesehatan, Kementerian Kesehatan selama Januari-Desember 2010. Masing-masing sampel dilakukan pemeriksaan kualitas sik dan kandungan zat kimia berdasarkan PERMENKES Nomor 492/MENKES/PER/ IV/2010. Pemeriksaan secara sik meliputi TDS (jumlah zat padat terlarut), kekeruhan, suhu, dan warna. Pemeriksaan secara kimia meliputi nitrit, besi, kesadahan, klorida, mangan, pH, sulfat, dan senyawa organik KMnO4. Sampel dikatakan tidak memenuhi syarat jika salah satu atau lebih parameter memiliki nilai diluar batas maksimum. Hasil: Di antara 121 sampel terdapat 23,1% (28) sampel tidak memenuhi syarat PERMENKES. Seluruh sampel yang tidak memenuhi syarat karena nilai pH di luar batas yang diperbolehkan dan 1 sampel karena kandungan mangan di atas 0,4 mg/l. Kesimpulan: Air minum isi ulang di Jabotabek sebagian besar memenuhi syarat atau layak untuk diminum. Adapun sampel yang tidak memenuhi syarat karena kadar mangan yang tinggi dan pH yang terlalu rendah dan terlalu tinggi. (Health Science Indones 2011;2:77-80) Kata kunci: kualitas air, sik dan kimia, Jabotabek Abstract Background: Drinking water is increasingly scarce in urban areas. Water consumed by the public must meet the requirements of safe drinking water quality for health. Many people use rell drinking water to meet the needs of drinking water daily. This study presents the quality of rell drinking water assessment based on the examination of physical and chemical parameters in 2010. Methods: The study design is cross-sectional. Samples were obtained from the public who requested the examination of drinking water quality in the Laboratory of Pharmaceutical Technology at the Center for Basic Biomedical and Health, Ministry of Health from January to December 2010. Each sample was assessed for physical and chemical content based on Miniter of Health decree Number 492/MENKES/PER/ IV/2010. Physical examination includes TDS (the amount of dissolved solids), turbidity, temperature, and color. Chemical examination includes nitrites, iron, hardness, chloride, manganese, pH, sulfate, and organic compound KMnO4. The sample is said not to meet the standard quality if one or more parameters have a value beyond the maximum limit. Results: Among 121 samples, about 23.1% (28) samples did not meet the requirement set by the Miniter of Health decree. All samples which did not qualify because the pH value was beyond the permitted limits and one sample because the content of manganese was above 0.4 mg / l. Conclusions: Most rell drinking water in Greater Jakarta meets the quality requirement. Samples that did not meet the requirement because of high levels of manganese. too low or too high pH. (Health Science Indones 2011;2:77-80) Key words: quality of water, physical and chemical properties, Jabodetabek
Corresponding author: Arifayu Addiena Kurniatri E-mail: arifayuaddiena@ymail.com
78
Kurniatri
Water is a very basic requirement for human beings. Current problem is that drinking water is increasingly difcult to obtain, especially in urban areas. Water supply in Indonesia such as piped water has not been evenly distributed in each region so that the community generally use ground water (wells), river water, rain water, water sources (springs) and others. Ground water has become unsafe due to contaminated drinking water material from septic tanks and surface water.1 The sources of water becoming polluted by untreated industrial waste due to the use of excess capacity compared to renewal ability. Currently besides industrial pollution caused by waste, pollution caused by domestic waste has demonstrated a serious effect. In Jakarta for example, as a result of inadequate urban waste water treatment facilities, river bodies have been polluted by domestic sewage, even the bodies of the river designated as a raw material of drinking water has been contaminated as well.2 If contaminated water is consumed by the public, it would cause health problems. Before consumption, water must be processed to eliminate or reduce levels of contaminated material to a level that is safe. Safe drinking water is when it meets the requirements for physical, microbiological, chemical and radioactive materials contained in the mandatory parameters and additional parameters. Mandatory parameters that must be followed by all providers of drinking water include:3 (1) physical parameters including TDS (Total Dissolved Solids), turbidity, and color; chemical parameter: problem of toxic chemical compounds, heavy metals, discoloration and avor; biological parameter determined by the presence of pathogenic bacteria contained in the water. Many people use rell drinking water to fulll their daily needs. With so many business of rell drinking water, more attention is required to guard the quality of drinking water produced. By implementing that way, the consumer is protected as well as the business of rell drinking water itself. In previous studies describing quality of drinking water in Greater Jakarta during the years 2007-2009, it was found that 27.5% of samples did not meet the requirements for drinking water.4 Following the review, this paper examines the quality assessment of physical and chemical parameters specic to rell drinking water in Greater Jakarta in 2010. METHODS
This study uses water quality data checks are performed at the Laboratory of Pharmacy, Center for Basic Biomedical and Health Technology in 2010. The study design is cross-sectional and type of experimental is laboratory research. Study results provide information about physical and chemical quality of rell water so that water, thus determined eligible or ineligible quality of drinking water.4 Samples were water proposed by communities of the Greater Jakarta area to be examined in the laboratory to obtain water quality certication. The number of samples obtained during January-December 2010 was as many as 426 samples. Samples analyzed were only rell drinking water and those with complete inspection data. Hence, the total number of samples analyzed was 148 samples. Assessment of drinking water quality was carried out by examining the physical and chemical quality of each sample. Physical quality of the examined includes Total Dissolved Solids (TDS), turbidity, and color. Chemical quality was checked by measuring the level of nitrite, iron, hardness, chloride, manganese, pH, sulfate, and KMnO4 organic compounds in water samples.3 Determination whether the drinking water met the quality requirement was by comparing each parameter against the maximum limit in accordance with the Minister of Health decree Number 492/MENKES/PER/ IV/2010. The requirements according to the Health Minister are as follows: TDS 500 NTU; turbidity 5 TCU; color of 15 mg / l; nitrite 3 mg / l; iron 0.3 mg / l; hardness 500 mg / l; chloride 250 mg / l; manganese 0.4 mg / l; pH 6.5 to 8.5; sulfate 250 mg / l; KMnO4 10 mg / l. If any parameter exceeds a specied threshold then it is concluded that water samples are not qualied or not good to drink. Analysis was conducted by Stata 9 statistical program. RESULTS Of the 426 samples of water quality inspections in 2010 at the Laboratory of Pharmacy, 121 samples were used in this study. Table 1 shows that the impropriety rell drinking water in Greater Jakarta was caused by two variables; i.e. that manganese is too high and the pH is too low and too high. Test results vary widely in terms of physical and
Quality of drinking water
79
Table 1. Quality of rell drinking water based on physical and chemical parameters Meet the quality requirement
Parameter Unit Mean SD Min. Max. CV (%)
Standard n
Yes % 100 100 100 100 100 100 100 99.2 76.9 100 100 n 0 0 0 0 0 0 0 1 28 0 0
No % 0 0 0 0 0 0 0 0.8 23.1 0 0
Physical TDS Turbidity Color Chemical Nitrite Iron Hardness Chloride Manganese pH Sulfate KMnO4
mg/l NTU TCU mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l
87.32 0.23 4.56 0.01 0.02 37.73 11.74 0.01 6.85 3.77 0.84
50.62 0.28 2.19 0.01 0.02 28.06 4.80 0.05 0.61 4.69 0.78
0 0 0 0 0 0.5 3.0 0 5.2 0 0
290.0 1.7 12.0 0.09 0.1 168.0 34.0 0.5 8.5 21.2 5.0
57.97 121.74 48.03 100.00 100.00 74.37 40.89 500.00 8.91 124.40 91.86
500 5 15 3 0.3 500 250 0.4 6.5-8.5 250 10
121 121 121 121 121 121 121 120 93 121 121
chemical parameters, seen from the CV> 20%, except for pH. DISCUSSION Limitations of this paper is that the data obtained was recorded data from the results of the proposed community water quality in 2010, and information regarding the origin of the sample was not recorded in full. However, samples were obtained during the months of January to December 2010 and the examination was performed by standard procedures. Another limitation was that the data of biological parameters was not obtained because it was done in other laboratories. Therefore, water quality can only be seen from the physical and chemical parameters. Rell drinking water is generally made from water that has been treated with a variety of processes such as by ozonation, UV radiation, adsorption by activated carbon ltration or membran (reversed osmosis).5 Water treatment done in depots of rell drinking water in Greater Jakarta seemed to have been able to produce excellent physical quality of water. It can be seen in Table 1 where all samples meet the requirements of physical parameters.
Drinking water should not only be safe but the physical quality is also a priority.6 Poor physical water quality can indicate what chemicals contained in water. Water turbidity for instance, can be caused by the presence of organic materials and inorganic materials contained in the water like mud and materials derived from waste. In terms of aesthetics, turbidity in the water is associated with the possibility of contamination by sewage.1 There are two parameters of the chemical that cause one sample that does not qualify, i.e. the content of manganese and pH. According to Permenkes 492/ MENKES/PER/IV/2010, the pH value should not be less than 6.5 or more than 8.5. Water with low pH (below 6.5) is acidic and not recommended for consumption because it may indirectly be bad for health. In addition, low pH water is corrosive because it can dissolve metals that are not required by the body. Water with high pH (above 8.5) is alkaline. Water that is alkaline usually contains excess ions that are not needed by the body. The pH value of drinking water depends on the pH of raw water and the processing process. Screening/ ltration stage was instrumental in normalizing the pH value, because at this stage both anions and cations that lead to high / low pH can be ltered.7 High content in water can cause water to become
80
Kurniatri
turbid, brown, and smelling the metal manganese.8 This study found one sample that had excessive levels of manganese, but did not nd a sample that has turbidity and color that exceeds the maximum limit. This may occur because in the sample that does not qualify the parameter of manganese levels was 0.5 mg / l whereas the requirement is 0.4 mg / l. Excess levels of manganese was of very little difference, so it can be said that excess manganese that occurs in this sample did not affect the physical quality of water. Manganese in water can be removed by chlorination followed by ltration.6 In conclusion, rell drinking water quality in Greater Jakarta according to chemical and physical parameters was generally of good quality. Water quality parameters that did not meet drinking water quality standard in rell drinking water were manganese and pH. It is necessary for drinking water depots to normalize the pH and reduce the levels of manganese in their drinking water products. Acknowledgments The author would like to thank the Laboratory of Pharmacy, Center for Basic Biomedical and Health Technology, Ministry of Health and Dra. Ani Isnawati as Head of the Laboratory for her permission to use the data. REFERENCE
1. Widiyanti NLPM, Ristiati NP. Qualitative analysis of coliform bacteria on drinking water rell depot
2.
3. 4. 5.
6.
7.
8.
in Singaraja, Bali. Health Ecology. 2004[cited 2011 November 17 2011];3:64-73. Available from http:// www.pusat3.litbang.depkes.go.id/data/vol%203/ Ni%20 Putu%20_2.pdf. Indonesia. Herlambang AH. Water pollution and its managing strategy. JAI. 2006[citation 17 November 2011];2:1629. Downloaded from http://ejurnal.bppt.go.id/ejurnal/ index.php/JAI/article/view/57/10. Indonesia. Minister of Health decree No. 492/Menkes/Per/IV/2010 on drinking water quality requirements. The Ministry of Health, Republic of Indonesia. Indonesia. Sukmayati A, Isnawati A, Mariana R. Drinking water quality outlook in Jabotabek, Year 2007-2009. Indonesian J Pharmaceutical 2009;3:83-92. Indonesia. Sitorus S.. Drinking Water quality analysis by ozonization process, ultraviolet, and reverse osmosis. Mulawarman J Chemical. 2009[citated 17 November 2011]; 6:30-2. Available from http://isjd.pdii.lipi.go.id/ admin/jurnal/62093032.pdf. Indonesia. WHO Guidelines for drinking-water quality 4th edition. Geneva. WHO, 2011 [citated 2011 November 17]. Available from http://www.who.int/water_ sanitation_health/publications/2011/dwq_guidelines/ en/. Indonesia. Athena, Sukar, Hendro M, Anwar D. Water processing inuence in normalizing acidity (pH) in drinking water rell depot. Health Research Development Media. 2005;15:19-24. Indonesia. A Rahmita, Roq I. Water quality and performance of processing unit in drinking water processing installation [thesis], Bandung. Institut Teknologi Bandung. 2009. [citated 17 2011 November 17]. Available from http:// www.ftsl.itb.ac.id/kk/rekayasa_air_dan_limbah_cair/ wp-content/uploads/2010/11/pi-w2-rahmita-astari15305049.pdf. Indonesia/
Risk factors for helminthiasis
81
Fingernail biting increase the risk of soil transmitted helminth (STH) infection in elementary school children
Liena Soana,1 Sri Sumarni,2 Mara Ipa3
1 2
Faculty of Public Health, Universitas Ahmad Dahlan, Yogyakarta Departement of Parasitology, Faculty of Medicine, Universitas Gadjah Mada 3 Vector Borne Disease Control Research and Development Council, Ciamis, West Java
Abstrak
Latar belakang: Infeksi cacing usus yang ditularkan melalui tanah (Soil Transmitted Helminth-STH) merupakan infeksi tersering dan terbanyak di antara infeksi-infeksi parasit. Kunci pemberantasan kecacingan adalah memperbaiki higiene perorangan dan sanitasi lingkungan. Tujuan penelitian ini adalah untuk Mengidentikasi faktor risiko yang meningkatkan infeksi STH pada anak sekolah dasar (SD). Metode: Penelitian ini menggunakan desain potong lintang dengan sampel purposif. Penelitian ini dilakukan di suatu SD di wilayah kerja Puskesmas di Yogyakarta pada bulan Oktober sampai Desember 2009. Pengumpulan data dilakukan dengan survei tinja dan wawancara. Pemeriksaan tinja menggunakan metode Kato Katz dan wawancara menggunakan kuesioner. Hasil: Di antara 211 subjek, 52 (24,6%) murid mengidap cacingan dan yang terbanyak adalah Trichusis trihiura, sedangkan cacingan Ascaris lumbricoides, sedangkan infeksi campuran Trichuris trichiura dan Hookworm sangat jarang. Risiko tertinggi (2,8 kali lipat) terjadi di antara murid yang mempuyai kebiasaan menggigit kuku jari dibandngkan dengan yang tidak mempunyai kebiasan ini [risiko relatif suaian (RRa) = 2,80; 95% interval kepercayaan (CI) = 1,22-4,04]. Subjek yang tidak mencuci tangan sebelum makan atau tidak mencuci tangan dengan sabun setelah buang air besar mempunyai risiko 2,2 kali terhadap terinfeksi cacingan. Kesimpulan: Kebiasaan menggigit kuku jari, tidak mencuci tangan sebelum makan dan tidak mencuci tangan dengan sabun setelah buang air besar mempertinggi risiko infeksi cacingan. (Health Science Indones 2011;2:81-6) Kata kunci: cacingan, kebiasaan menggigit kuku jari, cuci tangan
Abstract
Background: Intestinal worm infections transmitted through the soil are the most common infection among parasitic infections. The key to worm eradication is to improve personal hygiene and environmental sanitation. This study aimed to identify several risk factors related to occurrence of Soil Transmitted Helminth (STH) infections in elementary school. Methods: A cross sectional study with purposive sampling method was carried out in a primary school children in a area of a Yogyakarta health Center from October to December 2009. Stool was examined by using the Kato Katz method and pupils were interviewed by questionnaires. Results: Two hundred and eleven subjects participated in this study, and 52 subjects (24.6%) had STH infection. The most frequent STH infection was Trichuris trichiura, and the least was mixed infection (Ascaris lumbricoides, Trichuris trichiura and Hookworm). The highest risk (2.8-fold) occurred among those with a habit of ngernail biting compared to those who did not bite ngernails [adjusted relative risk (RRa) = 2.80; 95% condence interval (CI) = 1.22-4.04]. No hand washing before meals as well as no hand washing with soap after passing stool also increased the risk of STH infection by 2.2-fold. Conclusion: Fingernail biting and no hand washing before meals as well as no hand washing with soap after passing stool increased the risk of STH infections. (Health Science Indones 2011;2:81-6) Keywords: soil transmitted helminth, ngernail biting, hand-washing
Corresponding author: Sri Idaiani E-mail: liena_manisku@yahoo.co.id
82
Soana et al.
Soil Transmitted Helminth (STH) infections are the most common among parasitic infections. These intestinal worms, such as roundworms (Ascaris lumbricoides), whip worms (Trichuris trichiura), and hookworms (Necator americanus), usually infect children. They are transmitted through the soil and into the human body and more than one billion people worldwide suffer from worm infestation.1 In Indonesia, the number of STH cases, especially in children, is also quite high, although it is not considered a serious public health problem yet. Even though STH does not cause death, but the impact is very severe, since it can affect the intelligence and mental development of children. This is especially true during the growth and development of the children. Therefore worm infestations still pose a threat to the future of the children, even though it has been largely ignored by the community A report from one district in the province of Yogyakarta found the prevalence of worm infection in 2008 to be 9.8%, while in an area served by another health center, the prevalence was 7.8%.2 Several risk factors suspected to be related to the high incidence of STH are, among others, inadequate sanitation, poor personal hygiene, low level of education and poor socio-economic level, poor knowledge, attitude, and health behavior, and also geographic conditions suitable for breeding of the worms.3 The purpose of this study was to identify the association of personal hygiene and the other factors to STH infections in elementary school children in the area of a primary health center in Yogyakarta. METHODS A cross sectional study with purposive sampling method was carried out in a primary school. The primary school was located in the area of Kulonprogo in Yogyakarta health center. The subjects were rst grade primary school children. The number of primary schools in the Kokap I health center area is 25, consisting of 15 state primary schools (SDN) and 10 private primary schools. The study lasted 3 months from October to December 2009. All the children were asked for fecal samples and then interviewed. The fecal samples were collected in cooperation with the classroom teachers. The teachers instructed the children to collect their stool in a pot that has been
provided. The children were asked to pass stool in the morning before going to school, promptly collected the stool sample and come to the school as soon as possible. The stool collected was examined by Kato-Katz method. A sample the size of a green bean was taken with a stick and placed on waxed paper (which is water impermeable) and ltered through a ne steel mesh (screen ware) approximately 3x4 cm. Stool that had been ltered was taken with a stick and then pressed with perforated cardboard (Kato cardboard) on a coded slide. The stool sample was then covered with cellophane tape (22 x 30 mm) which had been soaked in a solution of green Gliserynmalachiet for 24 hours. The stool preparation was thinned and attened out. Excess uid on the edges of the cellophane tape was drained by placing the preparations upside down on tissue paper.4 The preparations were labeled (name of student, school, and date of execution) and then examined under a compound microscope. Data of personal hygiene and sanitation were obtained through questionnaires lled out by the respondents. There were eight aspects for personal hygiene. These were hand washing before meals, using the toilet, hand washing with soap after passing stool, trimming ngernails, ngernail biting, using sandals/shoes, bathing twice a day with soap, and using clean water for bathing & drinking.5 There were ten aspects for sanitation of home environment clean water, toilet, soap, distance of toilet to water source/well, water storage, types of ooring, yard, wastebasket, drinking water, and kitchen.6 Cox regression analysis with constant time using STATA 9.0 software was done to determine the risk factors for STH infections. A risk factor was considered to be a potential confounder if the univariate test it had a P-value 0f <0.25, and considered as a candidate for multivariate model along with all known risk factors for STH infections. Ninety-ve percent condence intervals were based on the standard error of coefcient estimates. Relative risks (RR) were estimated by the maximum likelihood. This survey was conducted with approval by the Ethics Committee of Medicine and Health Research, Faculty of Medicine, Universitas Gadjah Mada, Yogyakarta. RESULTS A number of 211 subjects participated in this study, and 52 subjects (24.6%) had STH infection. The STH infections were not only single infection (one type of worm
83
only) but also mixed type infections with 2 or 3 types of worms. Table 1 showed that the STH infection with the highest frequency was single infection of Trichuris trichiura and the lowest frequency was mixed infection of Ascaris lumbricoides, Trichuris trichiura, and Hookworm. Table 1. Results of stool examination for Soil Transmitted Helminths infection
Species of worms Ascaris lumbricoides Trichuris trichiura A. lumbricoides & T. trichiura T. trichiura & Hookworm A. lumbricoides, T. trichiura & Hookworm Total Frequency 8 25 15 3 1 52 Percentage (%) 15.4 48.1 28.8 5.8 1.9 100.0
Table 2 showed that subjects with positive and negative STH infections were similarly distributed with respect to trimming of ngernails, using clean water for bathing & drinking, and using soap for bathing. On the other hand, subjects who did not wear sandals or shoes, or did not use the toilet were more likely to have STH infection compared to the reference group. Table 3 revealed four dominant risk factors that increased the risk STH infections. Those with a habit of biting their ngernails had the highest risk (2.8-fold) of STH infections compared to those who did not. No hand washing before meals as well as no hand washing with soap after defecation also increased the risk for STH infection. Those who did not wash their hands before meals had a 2.2-fold increased risk for STH infections. The same increased risk for STH infection (2.2-fold) was found among those who did not wash their hand with soap after defecation.
Table 2. Several personal hygiene characteristics and the risk of Soil Transmitted Helminth infections STH infection Negative Positive (n=149) (n=52) n % n % Trimming ngernails Yes No Using clean water for bathing & drinking Yes No Using soap for bathing Yes No Using toilet Yes No Using sandals/shoes Yes No 92 57 73.6 75.0 33 19 26.4 25.0
Crude Relative Risk
95% Condence Interval Reference 0.54-1.66
1.00 0.95
0.850
142 7 141 8 142 7 139 10
74.7 63.6 74.2 72.7 75.9 50.0 76.8 50.0
48 4 49 3 45 7 42 10
25.3 36.4 25.8 27.3 24.1 50.0 23.2 50.0
1.00 1.44 1.00 1.06 1.00 2.08 1.00 2.15
Reference 0.52-3.99 Reference 0.33-3.39 Reference 0.94-4.60 Reference 1.08-4.29
0.484
0.925
0.072
0.029
84
Soana et al.
Table 3. Fingernails biting and several personal hygiene and risk of Soil Transmitted Helminth infections
STH infection Negative Positive (n=149) (n=52) n n Fingernails biting Yes No Hand washing before meals Yes No Hand washing with soap after defecation Yes No House sanitation Good 140 9 93 56 76.9 47.4 88.6 58.3 42 23.1 10 52.6 12 11.4 40 41.7
Adjusted Relative Risk
95% Condence Interval
1.00 2.80
Reference 1.14-4.55
0.019
1.00 2.23
Reference 1.22-4.04
0.009
98 51 123 26
88.3 56.7 81.5 52.0
13 11.7 39 43.3 28 18.5 24 48.0
1.00 2.16 1.00
Reference 1.21-3.85 Reference
0.009
0.000
DISCUSSION Soil Transmitted Helminth infection is a disease typical of tropical and sub-tropical regions, especially in rural areas, urban slums and densely populated areas. Although all ages can be infected with worms, the highest prevalence can found in children. STH infections in children can be only a single infection (one type of worm only) or they can be mixed infections involving 2 or 3 types of worms. Mixed infections usually indicate recurrent infections in these children. The results showed that poor house sanitation was a risk factor for STH infections. This result was similar to a study in the Karanganyar district.3 The high prevalence of STH was probably due to geographic and environmental sanitation of the area which were favorable for the proliferation of the worms. The poor environmental sanitation surrounding the houses made it possible for the occurrence of continuous reinfection, since Ascaris lumbricoides can be present up to more than 1 year, whereas Trichuris trichiura and Hookworms for 5-10 years. This condition will produce an accumulation of larvae in the body.7 Reinfection of Ascaris lumbricoides and Trichuris trichiura in humans (self reinfection) are possible through food contaminated by infective eggs. These ingested eggs will hatch in the duodenum where the
larvae will stick to the intestinal villi and move to the proximal colon or eventually to the entire colon. Adult worms will burrow into the anterior part of the intestinal mucosa and begin to produce eggs. The eggs will exit the body along with the feces. Outside the body the eggs will mature and ready to reinfect again for the same person or another person (another host). Reinfection of Hookworms in humans is through penetration of the skin. After entering through the skin, the larvae will be taken by venous blood ow to the lungs. The larvae will then enter the alveoli, trachea, and pharynx to be swallowed and passed to the small intestines. The worms then produce eggs which will pass to the outside world along with feces to reinfect the person or another persons.8 Poor personal hygiene was also found to be a risk factor for STH infections in primary school children in Kokap I health center, Kulon Progo, Yogyakarta. This result was similar to a study in Malaysia that showed hygiene as a risk factor for the occurrence of Soil Transmitted Helminths.9 Poor personal hygiene will facilitate the occurrence of STH infections by continuous reinfection. The occurrence of reinfection in children often resulted from direct contact with soil containing infective eggs. In a village reinfection of ascariasis in children under 10 years was found to be lower than in children the age of 10 years and over.
85
This was probably caused by the spread of infection elsewhere outside the village Sentanan.10 Humans are infected with Ascaris lumbricoides and Trichuris trichiura through swallowing mature eggs from contaminated soil, whereas infection by hookworm larvae are through penetration of the lariform in the soil into the skin.7 This was supported by the poor environmental sanitation of the houses state where most of the houses have oors of beaten earth which facilitated the occurrence of STH infections. The transmission of Ascaris lumbricoides, both new infection and reinfection, after treatment was slow. This was probably the result of the dry sandy soil conditions which was unfavorable for the growth of Ascaris eggs.10 The results of this study differed from the results of a study conducted in Mataram, which showed that there was no signicant correlation between personal hygiene and the prevalence of worm infections, and substantial infection of A. lumbrcoides and hookworms in elementary school students. Since poor personal hygiene is one of the factors that play a role in the STH infections, poor people with unsanitary behavior have a greater possibility to be infected by all types of worms.11 This proved that the occurrence of STH infections is determined by man himself. One aspect of personal hygiene which has a greater inuence on STH infections was hand washing before meals. This nding was similar to other studies.12 Hand washing before meals using soap and water played an important role in the prevention of STH infections, because washing with soap can mechanically remove dirt along with parasites from the hands.13 In children, frequent infections directly via hands contaminated by soil containing infective eggs, the children most frequently diseased worms because usually their ngers inserted into the mouth or eating rice without washing hands, but the occasional person stomach also contained adult worms, worms are commonly encountered roundworms, hookworms, thread worms, tapeworms and pinworms. This is consistent with the theory that humans infected with Ascaris lumbricoides and Trichuris trichiura by swallowing infective eggs contaminate food, drinks and cutlery.14 In this study ngernails biting was found to increase risk factor for STH infection. This result was similar to a study in Southern Nigeria, licking/biting of ngers
were signicantly higher risk for A. lumbricoides and T. trichiura infection.15 Fingernails biting behavior is bad behavior and have a signicant association of STH infection.5 Fingernail which dirty and infected the soil is containing infective eggs are medium in STH infection transmission. Fingernails are usually to be the place of transmission of worm eggs from soil into the body. Fingernails always be cut in two days and short to avoid the transmission of worms from hand to mouth.14 STH transmission them through dirty hands, ngernails tucked worm eggs are likely to be swallowed when eating, this is compounded if they are not accustomed to hand washing before meal using soap.16 In conclusion, ngernails biting and no hand washing before meals, as well as no hand washing with soap after passing stool increased the risk of STH infection. Acknowledgments The authors wish to thank all subjects who willingly participated in this study. The authors would also like to express their sincerest gratitude for Dwi Ciptorini and Dr. M. Juffrie for their technical assistance. REFERENCES
1. Nakita. Helminthiasis can lower the intelligent of child. [Internet]. 2009. [cited 2010 March 20]. Available from: http://mylovbaby.blogspot.com/2009/01/kecacinganturunkan-kecerdasan-anak.html. Indonesian. Health Center of Kokap I. Prole of Kokap I health center, Kulon Progo. Yogyakarta; 2008. Indonesian. Suhartono. Factors of associated with the incidence and intensity of Helminthiasis in elementary school children in Karanganyar, Central Java in 1995. Media Medika Indonesia. 1998; 33: 3-6. Indonesian. Ministry of Health of Indonesia. Minister of Health act on helminthiasis control guidance. [cited 2009 August l3]. Available from http://www.depkes.go.id/downloads/ Kepmenkes/ KecacingandanFilariasis/. Indonesian. Gazali, Muhammad. Relationship personal hygiene children in elementary school and environmental health conditions with helminthiasis in Air Periukan, Seluma, Bengkulu. [thesis]. Yogyakarta: Universitas Gadjah Mada. 2008. Indonesian. Gunawan, Rudy. Interpretation on healthy house in healthy house plan, Indah S (ed). Yogyakarta: Kanisius. 2009. Indonesian.
2. 3.
4.
5.
6.
86
Soana et al.
7.
Garcia LC, Bruckner DA. Diagnostic of medical parasitology. Padmasutra L, editor. Jakarta: EGC medical book; 1996. (Indonesian}. 8. Sandjaja B. Medical parasitology, medical parasitology. 2nd ed. Pedo Herri editor. Jakarta. Prestasi Pustaka. 2007. Indonesian. 9. Hidayah NI, Teoh ST, Hillman E. Socio-environmental predictors of soil transmitted helminthiasis in a rural community in Malaysia. Southeast Asian J Trop Med Public Health [Internet].1997. 28:811-4. Available from: http://www.tm.mahidol.ac.th/seameo/publication.htm. 10. Noerhajati. Impact of clean water providing, toilet of families and health environment toward the prevalence of Ascaris lumbricoides in Kasongan, Medika. 17:72331. Indonesian. 11. Brown HW. Human intestinal nematodes in the basic clinical paracitology. Pribadi W editor. Jakarta: Gramedia; 1983. Indonesian.
12. Hidayat T. Environmental health, personal hygiene and intencity of worm disease with nutritional status in primary school in Mataram [thesis]. Yogyakarta: Universitas Gadjah Mada; 2002. Indonesian. 13. Agoes D. Behavior of hands washing before meals and helminthiasis in elementary school children in the West Sumatra. [Internet]. 2008. [cited 2011 Juli 27]. Available from: http://www.promosikesehatan. com/?act=article&id=423. Indonesian. 14. Gandahusada S, Illahude DH, Pribadi W. Helminthology in medical parasitology. Jakarta: Balai Penerbit Fakultas Kedokteran Universitas Indonesia; 2006. Indonesian. 15. Nmour J.C, Onojafe J.O, Omu B.A. Anthropogenic Indices of Soil Transmitted Helminthiasis among Children in Delta State, Southern Nigeria. Iranian J Publ Health. 2009;38:31-8. 16. Onggowaluyo JS. Medical Parasitology I (Helminthology). Jakarta: EGC. 2002. Indonesian.
Immunocytochemical assay and Dengue virus 3
87
Sensitivity and specicity of immunocytochemical assay for detection of Dengue virus 3 infection in mosquito
Dyah Widiastuti,1 Bambang Yunianto,1 Sitti Rahmah Umniyati,2 Nastiti Wijayanti3
Banjarnegara Vector Control Research Unit, National Institute of Health Research and Development Department of Parasitology, Faculty of Medicine, Universitas Gadjah Mada 3 Faculty of Biology, Universitas Gadjah Mada
1 2
Abstrak Latar belakang: Survei virologi pada nyamuk vektor dapat digunakan sebagai Sistem Kewaspadaan Dini untuk mencegah penularan Demam dengue di suatu daerah. Pemeriksaan laboratoris untuk deteksi virus Dengue pada nyamuk seperti isolasi virus, Polymerase Chain Reaction (PCR) dan Direct Fluorescent-Antibody (DFA) memerlukan keahlian yang tinggi, peralatan yang mahal dan waktu yang lama. Suatu metode berdasarkan imunositokimia menggunakan antibody monoclonal DSSE10 memiliki beberapa kelebihan. Tujuan penelitian ini untuk mengevaluasi sensititas dan spesitas pemeriksaan imunositokimia dibandingkan metode Reverse Transcription-Polymerase Chain Reaction (RT-PCR) untuk mendeteksi infeksi Virus Dengue 3. Metode: Penelitian eksperimental dilakukan di laboratorium Parasitologi Fakultas Kedokteran Universitas Gajah Mada (UGM) pada bulan Mei 2009-Oktober 2010. Sebanyak 22 Ae. aegypti yang diinfeksi virus Dengue 3 digunakan sebagai kelompok infeksius dan 35 nyamuk yang tidak diinfeksi sebagai kelompok non infeksius. Pemeriksaan imunositokimia Streptavidin Biotin Peroxidase Complex (SBPC) menggunakan antibodi monoklonal DSSE10 dilakukan pada sediaan head squash Ae .aegypti untuk mendeteksi antigen virus Dengue 3. Pemeriksaan RT-PCR sebagai baku emas diaplikasikan pada toraks nyamuk. Hasil: Nilai Kappa menunjukkan kesepakatan yang baik antara dua orang pemeriksa (0,63). Imunositokimia mendeteksi antigen virus Dengue-3 dengan sensitivitas yang sama dengan RT-PCR (sensitivitas 100%). Namun spesisitas IC lebih rendah dibanding RT-PCR (spesisitas 91%) karena beberapa hasil positif palsu muncul pada pemeriksaan ini. Kesimpulan: Metode IC memiliki nilai sensitivitas dan spesisitas yang tinggi dibandingkan dengan metode RT-PCR. Metode IC ini dapat digunakan untuk surveilans virus Dengue pada nyamuk vektor. (Health Science Indones 2011;2:87-91) Kata kunci: imunositokimia, DSSE10, head squash, dengue Abstract Background: Virological surveillance provides an early warning sign for the risk of transmission in an area. Laboratory tests for dengue virus infection on mosquitoes include isolation of the virus, Polymerase Chain Reaction (PCR) and Direct Fluorescent-Antibody (DFA) requires a high level of technical skill, expensive equipment, and time-consuming. A method based on immunocytochemical (IC) using monoclonal antibody DSSE10 has several advantages. This study aimed to evaluate sensitivity and specicity IC assay compared with Reverse Transcription-Polymerase Chain Reaction (RT-PCR) as gold standard to detect Dengue Virus (DENV)-3 infections in mosquito Aedes aegypti. Methods: An experimental study was conducted in laboratory of Medical Parasitology, Faculty of Medicine, Universitas Gadjah Mada (UGM) in May 2009 until October 2010. A total of 22 articially-infected adult Ae. aegypti mosquitoes of DENV 3 were used as infectious samples and 35 non-infected adult Ae. aegypti mosquitoes were used as normal ones. The IC Streptavidin Biotin Peroxidase Complex (SBPC) assay using monoclonal antibody DSSE10 was applied in mosquito head squash to detect Dengue virus antigen. RT-PCR as a gold standard was applied in mosquito thorax. Results: The kappa value showed a good agreement between two observers (kappa value 0.63). IC could detect dengue virus antigen as sensitive as RT-PCR (sensitivity 100%). But IC was less specic than RT-PCR (specicity 91%) because some false positive results were found in this method. Conclusion: The IC method has a high sensitivity and high specicity compared with RT-PCR. This IC method may be useful for virological surveillance of dengue infected Aedes mosquitoes. (Health Science Indones 2011;2:87-91) Keywords: immunocytochemical, DSSE10, head squash, dengue
Corresponding author: Dyah Widiastuti E-mail: umi.azki@gmail.com
88
Widiastuti
Dengue is a prominent disease in tropic and subtropic areas. The major disease burden is found in Southeast Asia and the Western Pacic.1 There are four antigenically related but distinct serotypes of dengue virus, designated DEN-1, DEN-2, DEN-3, and DEN4.2 DEN-3 has been recognized as the predominant serotype in many recent epidemic occurrences of DHF in Indonesia. Dengue viruses are transmitted to humans by the bite of infective female mosquitoes of the genus Aedes. Aedes (Ae.) aegypti is considered the main vector because this species is closely associated with human habitation, but in some regions other Aedes species, such as Ae. albopictus3 and Ae. polynesiensis4, are also involved. Virological surveillance provides an early warning sign for the risk of Dengue virus transmission in an area.5 Laboratory tests for detection mosquitoes infected with dengue viruses include isolation of the virus and demonstration of a specic viral antigen or RNA. Isolation of the virus is the most denitive approach, but the techniques involved require a relatively high level of technical skill, equipment, and are timeconsuming.6 Detection of nucleic acid is an alternative method to detect infected mosquitoes. Polymerase Chain Reaction (PCR) is one technique available for the laboratory diagnosis of dengue infection. This molecular technique is rapid, highly sensitive and specic.6 However PCR requires a relatively expensive equipment, such as a thermalcycler. Direct detection of dengue antigen, such as the Direct Fluorescent-Antibody (DFA) test is labor-intensive and requires uorescent microscope and cryo-freezer.6 Therefore, a method based on immunocytochemistry (IC) involving enzyme conjugates such as peroxidase and phosphatase in conjunction with either polyclonal or monoclonal antibodies has been developed to detect dengue antigen. This IC method is a common laboratory technique that uses antibodies that target specic peptides or protein antigens in the cell via specic epitopes. This method has 94.3% sensitivity, 90% specicity for whole blood samples.7 Therefore, IC is useful in detecting dengue virus infection. Monoclonal antibody against DENV-3 was produced by the Dengue Team of Universitas Gadjah Mada
(UGM).8 In this study, the newly developed MAbs DSSE10, which recognize NS1 of dengue virus serotype 3 (DEN-3), respectively, were used in staining mosquito head squash. This study aimed to evaluate sensitivity and specicity IC assay compared with RT-PCR to detect Dengue Virus-3 infection in mosquito Aedes aegypti. METHODS An experimental study was conducted in laboratory of Medical Parasitology, Faculty of Medicine, UGM from May 2009 until October 2010. A total of 22 articiallyinfected adult Ae. aegypti mosquitoes of DENV 3 were used as infectious samples and 35 non-infected adult Ae. aegypti mosquitoes were used as normal ones. Dengue virus type 3 (H-87) in C6/36 cell lines was obtained from Naval Medical Research Unit 2 (NAMRU-2), Jakarta. Eggs of laboratory colony Ae. aegypti on dry lter paper, were reared to adults in the laboratory of Medical Parasitology, Faculty of Medicine, UGM. Three days old adults female Ae. aegypti were collected by manual aspirators for use in the experiments. Three-day old adult females Ae. aegypti were experimentally infected with DENV-3 using a sterile parenteral inoculation technique.9 Mosquitoes were immobilized over wet ice for 5-10 minutes before being injected with virus suspension in the membrane area of the intrathoracic. Inoculation procedures took place under a dissecting microscope using a calibrated capillary needle and syringe plunger. Infected mosquitoes were held in small cylindrical cages covered with mosquito netting, and they were incubated at 271C and a relative humidity of 886% and maintained on 10% sucrose for 7 days. They were collected at 5, 6, and 7 days after inoculation and separated into caput and thorax. RT-PCR was applied in mosquito thorax, while caputs were kept at -70C for IC test. Negative controls comprised uninfected Ae. aegypti from non-endemic area of DHF and Anopheles mosquitoes from Salatiga district, Central Java province. The Anopheles mosquitoes were used as negative control tissue because they are not the vector of dengue virus.
89
RNA extraction was done in accordance with the protocol of High Pure Viral Isolation Kit (Roche, Germany). Dengue viral RNA in mosquitoes were detected by reverse transcriptase polymerase chain reaction (RT-PCR) using DENV-3 -specic primers.10 The RT-PCR product was analyzed by agarose gelelectrophoresis on a 1.5% agarose gel (invitrogen) containing ethidium bromide (0.5 g/ml). For the DNA size marker, 100 bp DNA ladder (0.1 mg/ml) was used. Electrophoresis was set at 100 volts/cm2 and was run for 30-45 minutes. The expected size of 538 bp was identied as being of DENV-3 respectively. Caputs of Ae. aegypti from infectious and non-infectious group were put on the object glasses then pressed under cover glass with the eraser part of a pencil. Each object glass can be lled with 10-15 caputs. The cover glass was removed, and the object glass was put into a bottle lled with alcohol 70%. The preparation was dried in room temperature for around 30 minutes. Afterwards, it was xed with cold acetone (-200C) in freezer for 3-5 minutes, then dried in laminary ow.11 Preparation was xed with cold methanol (-200C) for 3-5 minutes and washed with PBS. To eliminate the endogenous peroxidase activity, the preparation was soaked in peroxidase blocking solution (1 part of hydrogen peroxide 30% + 9 part of absolute methanol) at room temperature for 10 minutes. Preparation was incubated in prediluted blocking solution for 10 minutes in room temperature (250C).11 One hundred mL primary antibody (DSSE10 1:10 monoclonal antibody) was added to the preparation (adjusted until all part was soaked) and incubated on damp tray at room temperature (250C) for 60 minutes or overnight in the refrigerator. Then it was washed twice with (fresh) PBS for 2 minutes; 100mL biotinylated universal secondary antibody was added, and the preparation was incubated at room temperature (250C) for 10 minutes, and then washed twice with fresh PBS for 2 minutes.11 The preparation was incubated with ready to use streptavidin-peroxidase-complex reagent for 10 minutes and then washed twice with PBS for 2 minutes; incubated in 100 mL peroxidase substrate solution (DAB) for 2-10 minutes (the thicker the preparation, the longer the incubation time), and then washed with tap water; then 100 mL Mayer hematoxyllin (counter stain) was added, incubated for 1-3 minutes, and then washed with tap water. The preparation was then soaked
in alcohol, washed, and then soaked in xylol. Drops of mounting media were added on the preparation and covered with cover glass. When it was already dry, the preparation was ready to be evaluated under light microscope with magnication of 40x, 100x, 400x, and 1000x. The microscopic examinations involved two observers. The rst observer was an experienced technician of Medical Parasitology laboratory, Faculty of Medicine UGM. The second observer was the researcher. If the preparation showed a brown color, it meant that the preparation contained DEN viral antigen. Meanwhile if the preparation showed blue or pale color (as in the negative control) the preparation did not contain DEN viral antigen. Laboratory mosquitoes infected with DENV-3 were dissected into 2 parts (head and thorax) on day 5, 6, and 7 days post-inoculation. Each mosquito thorax was amplied by RT- PCR to detect dengue virus. All of RT-PCR product of uninfected mosquito showed negative result, meaning there were no DENV inside these mosquitoes. These groups were used as noninfectious group. Sensitivity and specicity were measured based on Hermann formula. Validity and reliability was determined based on kappa value by Landis and Koch.12 RESULTS All head squash preparation of infectious groups showed positive result based on IC (Figure 1). DENV-3 infection in Ae. aegypti was shown as discrete brownish granules between the whole brain tissues. Most of head squash preparation of non-infectious group showed blue and pale brain tissues, meaning there is no DENV-3 antigen inside these mosquitoes. Figure 1 showed IC staining positive result on A (5 days), B (6 days), and C (7 days) post-inoculation and negative result on non-infected mosquito (D). For reliability, an agreement has achieved for kappa between two observers. Inter observer agreement result shown in Table 1. The rst and the second observers were in agreement for detecting Dengue viral antigen on head squash preparation by immunocytochemistry using mABs DSSE10 (Kappa value was 0.63).
90
Widiastuti
Figure 1. All head squash preparations of infectious groups
DISCUSSION Table 1. Inter-observer agreement of immunocytochemical assay

Observer 2 (+) (-) (+) 15 0 16 (-) 10 32 41 Total 25 32 57
Observer 1
Total Kappa value = 0.63
Immunocytochemistry is a powerful method for the identication of proteins or antigen in cells and tissues. The monoclonal antibody used in this study was secreted by DSSE10 clone belonging to IgG class and IgG1 subclass. The monoclonal antibody was secreted by a single hybrid (DSSE10) which was generated from the third fusion recognized as DENV complex specic epitope and showed no cross-reactivity to Chikungunya and Japanese Enchephalitis antigens based on Western blotting analyses. The mAb DSSE10 reacts to non-structural protein (NS1).11 The detection of DENV-3 antigen in mosquitoes by IC assay using monoclonal antibody DSSE10 has high sensitivity (100%), which was more sensitive than the detection in human thick blood smears. Detection of Dengue virus in human thick blood smears by IC using monoclonal antibody DSSE10 gives 94.3% sensitivity and 90% specicity.7 This difference in sensitivity value may be caused by the different preparation between head squash and blood smears. Head squash preparation was dried by wet-xed smear, meanwhile blood smear preparation was dried by air drying. Air dried preparation often exhibits relatively weak immunoassaying. This is probably because the dried cells exhibit an overall lower antigen density.13 The IC stains are indispensable for problem solving in detection of infectious agent. In order to evaluate them appropriately, it is critical to be aware of a true positive stain and a false positive stain. A true positive stain shows chromogen deposition in cells or structures that truly contain the antigen of interest. In contrast, a false positive stain is one where the chromogen is localized to cells or structures that in reality lack the antigen of interest.14
Table 2 showed that IC has good sensitivity (100%) and good specicity (91%). Furthermore, analysis showed probability that the samples infected with DEN virus was 88% if diagnostic test showed positive result at IC assay. In addition, the probability of samples noninfected with DENV-3 was 100% if diagnostic test result showed negative at IC assay. Table 2. Sensitivity and specicity of immunocytochemical assay
RT-PCR (+) (+) Immunocytochemistry Total (-) 22 0 22 (-) 3 32 35 Total 25 32 57
Sensitivity = 100% Specicity Positive predictive value = 88% predictive value = 100%
= 91% Negative
91
The main cause of non-specic background staining is endogenous peroxidase activity which is found in many tissues and can be detected by reacting xed tissue sections with DAB substrate13. To eliminate endogenous peroxidase activity in this study was by the pretreatment of the tissue section with hydrogen peroxide prior to incubation of primary antibody. However Immunocytochemical stains in this study showed false positive in 3 samples of non-infected mosquitoes. The IC assay has several advantages over the PCR method. First, IC process is easier than PCR. Besides, IC assay does not require specic equipment such as thermalcycler, so this assay could be done in every laboratory and more cost-effective. Moreover, head squash preparation of IC assay could be stored for along times. So, with these advantages, IC has a chance to replace other method of dengue virus detection. Detection of Dengue virus in mosquitoes would give valuable information as early warning system to prevent Dengue Fever transmission. Chow noted that by detecting of Dengue virus infection in the mosquito vectors before its introduction into the human population it was possible to predict an outbreak six weeks in advance of the occurrence of the rst human case in Singapore.15 When the number of dengue infection in mosquitoes increases the Local Health Ofce should make an effective vector control program to prevent the virus introduction to the human populations. Therefore, it needs further study to evaluate the application of immunocytochemical assay for detection Dengue virus infection in mosquito head squash in the eld level to develop an effective early warning system for Dengue Fever prevention. In conclusion, immunocytochemical assay could be used in detection DENV-3 infection on mosquito head squash. This method has a high sensitivity, high specicity and good inter observer agreement.
2. 3. 4. 5. 6.
7.
8.
9. 10.
11.
12. 13. 14.
REFERENCES
1. Gubler DJ. Epidemic Dengue/Dengue hemorrhagic fever as a public health, social and economic problem in the 21st century. Trends Microbiol. 2002;10:100-3.
15.
Halstead SB. Pathogenesis of dengue: challenges to molecular biology. Science 1988;239:476-81. Fontenille D, Toto JC. Aedes (Stegomyia) albopictus (Skuse), a potential new Dengue vector in Southern Cameroon. Emerg Infect Dis. 2001;7:66-74. Rosen L, Rozeboom LE, Sweet BH, Sabin AB. The Transmission of Dengue by Aedes polynesiensis marks. Am J Trop Med Hyg. 1954;3:878-82. Samuel PP, Tyagi BK. Diagnostic methods for detection & isolation of Dengue viruses from vector mosquitoes. Indian J Med Res. 2006;123:615-28. Lanciotti RS, Calisher CH, Gubler DJ, et al. Rapid detection and typing of Dengue viruses from clinical samples using reverse transcriptase chain reaction. J Clin Microbiol. 1992; 30:545-51. Mulyaningrum U. Evaluation of immunocytochemical assay to detect Dengue virus infection on thick and thin human blood smear [thesis]. Yogyakarta. Univ Gadjah Mada; 2010. Indonesian. Sutaryo, Umniyati SR, Wahyono D. Production of monoclonal antibodies against Dengue virus 3 for DHF patients and vector. Report of RUT-3 Year I. Yogyakarta. Faculty of Medicine, Univ Gadjah Mada. 2005. Indonesian. Rosen L, Gubler DJ. The use of mosquitoes to detect and propagate Dengue viruses. Am J Trop Med Hyg. 1974;23:1153-60. Yong YK, Thayan R, Chong HT, Sekaran SD. Rapid detection and serotyping of Dengue virus by multiplex RT-PCR and real-time SYBR green RT-PCR. Singapore Med J.2007. 48:662-68. Umniyati SR. Immunocytochemistry using monoclonal antibody DSSC7 for pathogenesis of Dengue virus infection and transovarial transmission and virologic surveilance of Dengue vector [dissertation]. Yogyakarta. Univ Gadjah Mada; 2009. Indonesian. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977; 33:159-74. Farmilo AJ, Stead RH. Immunochemical staining methods. 3rd ed. Boenisch T, editor. Carpinteria (Ca): DAKO Corporation; 2001. Miller RT. True positive vs. false positive staining. The focus immunohistochemistry. 2001. p. 1-2 [cited 2011 July 11]. Available from: http://www.propath.com/ index2.php?option=com_content&do_pdf=1&id=148 Chow VT, Chan YC, Yong R, et al. Monitoring of dengue viruses in eld-caught Aedes aegypti and Aedes albopictus mosquitoes by a type-specic polymerase chain reaction and cycle sequencing. Am J Trop Med Hyg. 1998;5:578-860.
92
Indalao et al.
Virus culture and real-time RT-PCR in identifying inuenza viruses from inuenzalike illness cases in Indonesia 2007-2008
Irene L. Indalao, Vivi Setiawaty, Hana A. Pawestri, Subangkit
Centre for Biomedical and Basic Technology of Health, NIHRD-MOH, Jl. Percetakan Negara 23, Jakarta 10560 Abstrak
Latar belakang: Adanya perbedaan hasil antara kultur virus dengan real-time polymerase chain reaction (RTPCR) yang digunakan dalam surveilans inuenza-like illness (ILI) menunjukkan perlunya mengevaluasi hasil kultur virus yang didapatkan dengan hasil RT-PCR sebagai pembanding.Tujuan penelitian ini adalah untuk mengevaluasi apakah kultur virus masih dapat diandalkan untuk studi surveilans ILI. Metode: Usap hidung dan usap tenggorok didapatkan dari 20 sentinel ILI di Indonesia selama tanun 20072008. Identikasi kultur virus dilakukan dengan menggunakan metode hemaglutinasi dan hemaglutinasi inhibisi. RT-PCR menggunakan primer yang bersifat spesik untuk inuensa A (A/H1N1, A/H3N2 and A/ H5N1) dan inuensa B. Primer disediakan oleh Center for Disease Control and Prevention, USA. Hasil positif kultur virus dibandingkan dengan hasil RT-PCR berdasarkan persentase kesamaan hasil. Hasil: Sebanyak 112 spesimen dari 4277 spesimen kasus ILI didapatkan hasil positif inuenza dengan metode kultur. Kesamaan hasil positif inuenza kultur virus dibandingkan dengan real-time RT-PCR adalah 69.6%. Pada penelitian ini juga ditemukan bahwa 30,4 % (n=112) hasil real-time RT-PCR yang ditemukan positif inuenza tidak dapat dideteksi oleh metode kultur. Kesimpulan: Metode kultur masih relevan untuk surveilans ILI meskipun hasil positif Inuenza dari kultur virus lebih sedikit dari pada hasil positif Inuenza yang terdeteksi dengan metode PCR. (Health Science Indones 2011;2:92-5) Kata kunci: inuenza-Like Illness, RT-PCR, inuenza, kultur.
Abstract
Introduction: From the inuenza-like illness (ILI) surveillance in Indonesia, we learned that there was disagreement between virus culture and reverse trancriptase polymerase chain reaction (RT-PCR). This implies the need to evaluate whether virus culture is still a relevant method to be used in ILI surveillance. Methods: The ILI specimens obtained from 20 ILI sentinels in Indonesia in 2007-2008. Real-time RTPCR using primers were specic for inuenza A (A/H1N1, A/H3N2 and A/H5N1) and Inuenza B. The sequence of these primers was provided by the CDC, Atlanta. Virus culture identication was conducted with hemagglutination and hemagglutination inhibition methods. We evaluated the percentage of concordance between positive culture results vs its RT-PCR results. Results: A number of 112 inuenza positive in culture method from 4277 ILI specimens were compared with real-time RT-PCR result. There was 69.6% of virus culture result was in concordant with real-time RTPCR result. We also found that 30.4% of positive result using real-time RT-PCR were not detectable by virus culture. Conclusion: Virus culture was still essential and considerably efcient to support real-time RT-PCR detection in ILI cases in Indonesia although the positive Inuenza results by virus culture less than RT-PCR. (Health Science Indones 2011;2:92-5) Key words: inuenza-like illness, RT-PCR, inuenza, culture
Corresponding author: Vivi Setiawaty E-mail: vilitbang@yahoo.com
Virus culture and real-time RT-PCR
93
Inuenza-Like Illness (ILI) is a disease that shows inuenza symptoms, such as cough, runny or stuffy nose, sore throat, headache or body aches, but not an inuenza disease.1 The surveillance of ILI cases is important because it detects inuenza virus in patient. Since most people suffered with inuenza symptoms is not diagnosed or even seeks treatment, ILI surveillance has to be maintained with reliable detection method, so that it will screen not only inuenza cases, but also, other disease with similar symptoms.2,3 It could also detect whether there was antigenic drift or shift.4 The National Institute of Health Research and Development, Ministry of Health (NIHRD), Indonesia, has served as the national referral laboratory in Indonesia for inuenza. Since 1999, NIHRD has conducted ILI surveillance as one of its main research activity. Reverse trancriptase (RT-PCR) is a method for molecular detection recommended by the United States- Centers for Disease Control and Prevention (CDC) to identify inuenza virus infection in ILI cases.5 The principle of this method is similar to Polymerase Chain Reaction (PCR), which is to amplify the inuenza virus RNA genome using a pair of oligonucleotide primers in order to generate copies of a certain DNA sequence, however, in RT-PCR, the RNA strand has to be reverse-transcribed into cDNA (complement DNA) rst using the reverse transcriptase enzyme. RT-PCR was considered more sensitive than by culture and ELISA.6-8 RT-PCR provides a specic and sensitive method for detection of inuenza viruses A and B and discriminates between virus subtypes;8-10 a considerable tool for inuenza surveillance. Although, in the other hand, this technique requires a high level of skill and complex laboratory infrastructure, takes several hours to perform and is considered not suitable for lower level of expertise.9 The virus culture is considered a sensitive and useful technique for diagnosis of inuenza virus.3 Specimens usually used in this detection methods are nasal, throat, and rectal swabs.11 Inuenza virus which may be contained within these specimens, when the specimens added to the cell line, will adhere directly to the cell and will infect other cells. After the designated time, the isolation from the cell line and the identication of the virus could be done. This isolated virus could be identied by serological method as type A or B with hemaglutination inhibition test. Although detection of
inuenza virus using virus culture requires time more than PCR up to 2 weeks, it was believed to be useful as alternative detection method which is necessary to avoid false negative results.7 Cell culture also has been suggested to be performed to obtain early and late in the season inuenza virus isolates which in turn is important to make sure that suitable vaccine strains will be available for the following year.12 In this study, our institute used two different methods, namely Real Time RT-PCR and Virus Culture. There were several researches that also aimed to compare result between different inuenza detection methods in Europe and America, but most of them were focused only to see which one had the highest sensitivity or specicity.5,6 They also used different detection object, such as Inuenza A Virus and Respiratory Syncytial Virus.5,6 Meanwhile, this research was directed to evaluate whether virus culture method was still relevant to be applied in ILI surveillance. The objective of this result was to evaluate the percentage of concordance between RT-PCR and virus culture. This study was not aimed to determine which was the best detection method between those two methods, but to evaluate whether virus culture was still reliable as an alternative test for RT-PCR. METHODS Nasal and throat swabs were collected throughout Indonesia from 20 ILI sentinels who operated as primary health services in Indonesia. Viral culture is done by using 112 positive results from 4277 specimens obtained during 2007-2008. Each specimen was screened by RTPCR prior to virus culture. Virus culture was done in biosafety cabinet class (BSC) IIA within BSL2 Inuenza Laboratory, Center for Biomedical and Pharmaceutical Research and Development, NIHRD, Jakarta. The molecular method used in this study was Quantitative Reverse Transcriptase Polymerase Chain Reaction (qRT-PCR) utilizing Thermal Cycler IQ5 (Biorad, USA). QiAmp RNA viral mini Kit (Qiagen, Germany) was used to extract the specimens according to manufacturers instruction. Five sets of primers and probes which are syntesized by Invitrogen and Sigma (USA) were treated to the specimens to detect Inuenza A, B, A/H1N1, A/H3N2. The reagent was one step qRTPCR Superscript III with Platinum Taq Polymerase
94
Indalao et al.
(Invitrogen, USA). The amplication procedure were as follows: 5 l extracted RNA of each specimen was treated by 500 C within 30 minutes (reverse transcriptase activation), then 950 C for 2 minutes (taq inhibitor activation). Forty ve PCR cycles were applied in this procedure. The denaturation step was done at 950 C for 5 seconds, subsequently, and the annealing step was performed on 550 C in 30 seconds. These specimens were cultured to identify the type and subtype of inuenza virus. The cell line used in this culture was Madin Darby Canine Kidney (MDCK) cell (NAMRU II). Specimen was transported using Hanks solution (GIBCO/Invitrogen, USA) with 0.2% penicillin (Invitrogen, USA), 0.2% streptomycin (Invitrogen, USA) and 0.4% phenol red (Sigma, USA) in it. Cell culture was inoculated and passaged in two series to allow propagation of the virus. The identication of virus culture was held using Hemagglutination and Hemagglutination Inhibition assays based on World Health Organization guideline in inuenza virus culture and characterization.10 The reference antigen and antiserum A/H1N1, A/H3N3, B/ Malaysia, B/Shanghai were supplied by WHO along with positive control, negative control, and Receptor Destroying Enzyme (RDE) Denka Seiken Co., Ltd. in inuenza virus culture and characterization package. We collected data of virus culture which reported positive results during ILI surveillance 2007-2008. Based on these results, we compared them with their RT-PCR result. The concordance percentage was meant to give information how many positive result obtained from virus culture was actually have the same results with RT-PCR. The percentage was obtained by dividing the number of positive inuenza in virus with number of positive result in cultureRT-PCR.12 RESULTS The comparative analysis between culture virus and RT-PCR of 112 positive inuenza by RT-PCR were 70 (62.5%) Inuenza B Viruses (IBV), seven out of 112 (6,1%) were Inuenza A Virus (IAV) by both methods. The discordance results could be found that four out of 112 (3.6%) was identied as IAV by RT-PCR but negative by culture, 27.7 % was positive IBV by RTPCR but negative by culture. Furthermore, from 112 specimens positive result based on RT-PCR, there were only 78 positive inuenza
specimens according to culture. The concordant percentage of virus culture to RT-PCR was relatively high (69.6%) and also for inuenza A/H1N1 (7 of 10), A/H3N2 (1 of 2) and B (70 0f 101), whereas the percentage numbers were not less than 50 %. Table 1 showed that there were negative PCR results identied as positive IAV (A/H1N1 and H3N2) or IBV by culture method. In total, there was 30.4% of negative result by virus culture which could be identied by RTPCR as positive inuenza.
Table 1. The result difference between RT-PCR and virus culture Culture A/H1N1 A/H3N2 Inuenza B virus Negative TOTAL RT-PCR A/H3N2 Inuenza B virus 0 0 1 0 0 70 1 31 2 101
A/H1N1 7 0 0 3 10
DISCUSSION There was a concordance between the virus culture and RT-PCR result as there were matched positive results between virus culture and RT-PCR. The signicance of the concordance between these results could be quantied to see the efciency of virus culture against RT-PCR. Therefore, the percentage number of type and subtype of inuenza virus from RT-PCR against virus culture was calculated by dividing the number of positive inuenza in virus culture with number of positive result in RT-PCR. We understand that RT-PCR method is an expensive method compare to culture. However the results from this study showed that virus culture was still considered as a reliable detection method since the percentage numbers of the consented results, mostly, were around 60%. This would mean most of the RT-PCRs results were also could be identied with virus culture although the virus culture is a time consuming method. Negative results by virus culture shown on table 1 can be detected by RT-PCR. There were several reasons to explain about these phenomena i.e. the quality of the specimen either due to sampling method, lack of the rapid transportation since Indonesia is an archipelago
Virus culture and real-time RT-PCR
95
country, or the nature of the specimen itself. These factors might cause low concentration of viral RNA or RNA degradation and lead to negative result in RTPCR.5,7,13 Nevertheless, this event could be evidence for the capability of virus culture in propagating the viral load so that they could be detected by virus culture method. In conclusion, virus culture was still essential and considerably efcient to support real-time RT-PCR detection in ILI cases in Indonesia although the positive Inuenza results by virus culture less than RT-PCR. Acknowledgments This research was a part of the ILI and Severe Acute Respiratory Infection Study conducted by The National Insitute of Health Research and Development (NIHRD), Ministry of Health of Indonesia in collaboration with United States of Centers for Disease Control and Prevention (US-CDC)-Atlanta. Data were entirely the property of NIHRD, Ministry of Health of Indonesia. We thank to Agustiningsih, Triyani, Oerip Pancawati, Sumarno, Sri Susilowati, Ratih Renindya Putri, and Sinta Purnamawati from Laboratory of Virology, NIHRD and ILI-SARI study team for their support and technical experties in this research. REFERENCES
1. World Health Organization. WHO manual on animal inuenza diagnosis and surveillance. http:// www.wpro. who .int/NR/rdonlyres/EFD2B9A72 2 6 5 - 4 A D 0 - B C 9 8 - 9 7 9 3 7 B 4 FA 8 3 C / 0 / manualonanimalaidiagnosisandsurveillance.pdf 2002. Herrmann B, Larsson C, Wirgart ZB. Simultaneous detection and typing of Inuenza Viruses A and B by a nested reverse transcription-PCR: Comparison to virus isolation and antigen detection by immunouorescence and optical immunoassay (FLU OIA). J Clin Microbiol. 2001;39:134-8. Shaffer LE, Rowe SA, Reed DE. Early detection of Inuenza Like Illness: Developing a multi-variate
4.
5. 6. 7. 8.
9.
10.
11.
12.
2.
13.
14.
3.
approach. Advance in Disease Surveillance. 2007; 2:67. Frisbie B, Tang YW, Grifn M, et al. Surveillance of childhood inuenza virus infection: What is the best diagnostic method to use for archival samples?. J Clin Microbiol. 2004;42:1181-4. US CDC. Inuenza symptoms.www.cdc.gov.gov/u/ symptomps.htm. 2007. Compans RW, Herrler G. Mucosal immunology. 3rd Edition. Boston: Elsevier Academic Press; 2005. Leland, DS, Ginocchio CC. Role of cell culture for virus detection in the age of technology. Clin Microbiol. Rev. 2007; 20:49-78. Steininger C, Kundi M, Aberle SW, et al. Effectiveness of reverse transcription-PCR, virus isolation, enzymelinked immunosorbent assay for diagnosis of Inuenza A virus infection in different age groups. J Clin Microbiol. 2002:40:2051-6. Chan KH, Maldeis N, Pope W, et al. Evaluation of the directigen Flu A_B test for rapid diagnosis of inuenza virus type A and B infections. J. Clin. Microbiol. 2002;40:1675-80. Van Elden LJR, Nijhuis M, Schipper P, et al. Simultaneous detection of inuenza viruses A and B using real time quantitative PCR. J Clin Microbiol. 2001;39:196-200. Ruest A, Michaud S, Deslandes S, et al. frost eh. comparison of the directigen Flu A_B test, the quickvue inuenza test, and clinical case denition to viral culture and reverse transcription-PCR for rapid diagnosis of inuenza virus infection. J Clin Microbiol. 2003;41:3487-93. Magnard CM, Valette M, Aymard M, et al. Comparison of two nested PCR, cell culture, and antigen detection for the diagnosis of upper respiratory tract infections due to inuenza viruses. J. Med. Virol. 1999;59:21520. Krafft AE, Russel K, Hawksworth AW, et al. Evaluation of PCR testing of ethanol-xed nasal swab specimens as an augmented surveillance strategy for inuenza virus and adenovirus identication. J Clin Microbiol. 2005;43:1768-75. Carman WF, Wallace LA, Walker J, et al. Rapid virological surveillance of community inuenza infection in general practice. BMJ. 2002;321:736-7.
96
Roselinda and Fitri et al.
Several dominant clinical symptoms associated with Inuenza A in Indonesia

Roselinda, Nyoman Fitri
Center for Biomedical and Basic Technology of Health, National Institute of Health Research and Development Ministry of Health Republic of Indonesia
Abstrak
Latar belakang: Pada tahap awal infeksi, inuenza A yang dapat menimbulkan pandemi, sangat sulit dibedakan dengan inuenza-like illness (ILI) yang lain. Oleh karena itu gejala klinik sangat penting untuk mendiagnosis secara dini inuenza A. Tujuan penelitian ini untuk mengidentikasi gejala klinik dominan yang berkaitan dengan inuenza A di Indonesia. Metode: Penelitian potong lintang. dilakukan di 20 puskesmas sentinel yang dipilih secara purposif di 19 propinsi di Indonesia tahun 2009. Data dan spesimen dikumpulkan oleh petugas paramedik atau medik puskesmas dari subjek rawat jalan dengan gejala ILI (batuk dan demam). Pemeriksaan Spesimen dilakukan di Pusat Rujukan Inuenza Nasional di Jakarta. Penentuan Inuenza A dengan real time RT-PCR. Hasil: Sebanyak 1802 subjek berdata lengkap untuk analisis inuenza A dari 2728 subjek dengan gejala ILI, dan 23,1% (416 subjek) didiagnosis positif inuenza A. Pada model terakhir terungkap bahwa subjek dengan pilek dibandingkan dengan yang tidak pilek berisiko 3,6 kali lipat Inuenza A [risiko relatif suaian (RRa) = 3,59; 95% interval kepercayaan (CI) = 1,34-9,63]. Subjek dengan nyeri tenggorok dibandingkan dengan yang tanpa nyeri tenggorok berisiko 54% lebih besar menderita Inuenza A (RRa = 1,54; 95% CI = 0,95-2,58; P = 0.082). Seangkan, subjek dengan riwayat demam dibandingkan tanpa riwayat pernah demam dalam dua hari terakhir berisiko 42% lebih besar menderita Inuenza A (RRa = 1,42; 95% CI = 0,97-2.,7; P = 0,069). Kesimpulan: Selain demam dan batuk, keluhan pilek dan nyeri tenggorok, serta riwayat pernah demam dalam dua hari terakhir merupakan faktor risiko dominan yang berhubungan dengan Inuenza A. (Health Science
Indones 2011;2:96-100)
Kata kunci: Inuenza A, cough, muscle pain, runny nose, sore throat
Abstract
Background: Inuenza A has a potential to become a pandemic, in the early stages is difcult to differentiate inuenza to inuenza-like illnesses. Therefore, the dominant clinical symptoms are the important keys to predict inuenza A infection in patients with inuenza-like illnesses (ILI). The aim of this study is to identify additional dominant symptoms associated to inuenza A in Indonesia. Methods: The eligible subjects of this study were outpatient who had ILI symptom, i.e. who had fever (38o or more) and coughing in purposive selected 20 Health Centers in 19 provinces of Indonesia during year 2009. Paramedics and medical staff identied the ILI cases and collected specimens. Laboratory tests for RT-PCR were performed at the National Inuenza Center in Jakarta. Results: Of 2728 specimens, 1802 had complete data for this analysis, and 23.1% (416 subjects) diagnosed positive inuenza A. Those who had than did not have runny nose symptom had 3.6-fold risk of inuenza A [adjusted relative risk (RRa) = 3.59; 95% condence interval (CI) = 1.34-9.63). In term of sore throat, those who had than did not have it had 54% more risk of inuenza A (RRa = 1.54; 95% CI = 0.95-2.58; P = 0.082). Furthermore, those who ever had than did not have fever for the last two days had 42% more risk of inuenza A (RRa = 1.42; 95% CI = 0.97-2.07; 0,069). Conclusion: In addition to fever and coughing, runny nose, sore throat, and ever had fever are dominantly associated with inuenza A. (Health Science Indones 2011;2:96-100) Key words: inuenza A, cough, muscle pain, runny nose, sore throat
Corresponding author: Roselinda E-mail: roselinda@litbang.depkes.go.id
Vol. 2, No. 2, December2011
Clinical Symptoms Associated with Inuenza A
97
Inuenza-like illness (ILI) is a medical diagnosis of possible inuenza or other illness with a set of common symptoms and therefore a signicant source of morbidity and mortality worldwide. Inuenza is an infectious disease caused by Orthomyxoviridae viruses that affects birds and mammals. The initial symptoms of inuenza are similar to ILI, such as fever or history of fever accompanied by cough, runny nose, sore throat, muscle pain and or dyspnea.1,2 There are three types of inuenza based on the main virus that caused, they are A, B, and C. Inuenza type A dan B have similar early symptoms that can cause epidemics and have high case fatality rates as well. Additionally, inuenza type A is more threatening because it can cause pandemics.3 Inuenza type A viruses are divided into subtypes (strains) based on two proteins on the surface of the virus. These proteins are called hemagglutinin (HA) and neuraminidase (NA). Currently, there are 16 different HA subtypes and 9 different NA subtypes. New subtypes of inuenza viruses may occure through processes called antigenic drift and antigenic shift. In 2005, inuenza virus subtype H5N1 pandemic (bird u) occured in Indonesia. Another pandemic occured in 2009 caused by a new subtype of H1N1 (swine u). The symptoms were similar with common seasonal inuenza. There are little known about factors related to inuenza A.3,4 Since inuenza A has similar symptoms with ILI, further examination in laboratory is needed to reveal the type of inuenza and the sub-type if the result is inuenza A. Inuenza A viruses can cause diseases with common symptoms that mostly will be neglected by people. Doctors seldom tell patients to conrm the result to the laboratory tests such as rapid test, RT-PCR and Culture test as the gold standard for inuenza[2]. It has been known that laboratory tests are very expensive, even for the rapid test that has low specivicity and sensitivity for inuenza A. Additionally, RT-PCR and Culture test are taking much time. Furthermore, it will be too late to realize that patients have inuenza A viral infections. Since inuenza A viral infection can cause severe inuenza-like illness among exposed people, virological and epidemiological surveillance are needed to understand the impact of inuenza A virus among people with ILI symptoms in Indonesia. Therefore, we need clinical symptomps to give us high predictions about inuenza A in patients. This study aimed to identify additional dominant symptoms associated to inuenza A in Indonesia.
METHODS A cross sectional study was conducted on cases that included into the ILI surveillance in 19 provinces in Indonesia in 2009. For each provinces appointed one health center except Papua which had two health centers that were chosen purposively. The criteria to select the health centers included: (1) high prevalence of upper respiratory track infection; (2) human resources availability for ILI surveillance; (3) cooperative and willingness to participate; (4) well documented reporting system; (5) and close to airport for easiness shipping to the regional laboratory or the nasional referral laboratory. Procedures: Subjects with symptoms of fever (38 C or more) and cough who presented at primary health centers (outpatients) were enrolled once a week. Informed consents were obtained from subjects. Trained paramedics collected nasal and throat swabs from the patients. They also obtained demographic data and clinical symptoms from patients. All of the specimens were shipped to the regional laboratory in four provinces or to the national referral laboratory in Jakarta. A conrmed case was ened by a positive result of a realtime reverse-transcriptase polymerase chain reaction (rRT-PCR) test. Before shipping, swab of specimens were placed into sterile Hanks Balanced Salt Solution (HBSS) as viral transport media (VTM). Specimens were refrigerated (4C) and shipped weekly with a strick condition. All of the specimens were tested for inuenza by realtime reverse-transcription polymerase chain reaction (rRT-PCR) assay. The QIAamp Viral RNA kit (QIAGEN, Valencia, CA) was used to extract viral RNA according to the manufacture manual. Positive specimens were inoculated into cell culture using Madin Darby Canine Kidney (MDCK). All of the virological and epidemiological data were collected and analized at the Virology Laboratory in Center for Biomedical and Basic Technology of Health. For this analysis age were categorized into ve groups (6-12, 1317, 18-34, 35-49, and 50-82 years old). We also collected the other symptoms which occured among the subjects: history of fever, runny nose, sore throat, muscle pain and dyspnea. These symptoms were obtained by observation and or asked by special trained paramedic or medical doctor during the subjects visited the health centers. Statistical analysis: Of the people enrolled (n=2728), only 1802 samples can be included in this study. We
98
Roselinda and Fitri et al.
excluded uncomplete data and patients under 6 years old because we could not acquire information from them about muscle pain. We performed data analyzed using Cox regression using Stata released 9. RESULTS Of 2728 specimens, 1802 had complete data for this analysis, and 23.1% (416 subjects) diagnosed positive inuenza A. Table 1 shows that in general, those who had and did not have inuenza A was similarly distributed with respect to dyspnea. Similar condition is shown between subjects
age 18-34 years and 6-12 years old. Male subjects and those who had muscle pain than sore throat more likely had increase risk to be inuenza A. Compared to age 6-12 years old, subjects of age 13-17 and 35 or more had more likely increase risk to be inuenza A. Our nal model (Table 2) reveals runny nose, sore throat, and ever had fever are dominantly associated with inuenza A. Those who had and did not have runny nose symptom had 3.6-fold to be Inuenza A. In term of sore throat, those who had than did not have it is 54% more risk to be inuenza A. Furthermore, those who ever had fever than did not have fever had 42% more risk having inuenza A.
Table 1. Several demographic, clinical symptoms and risk of Inuenza A Inuenza A Negative Positive (n=1386) (n=416) Gender Female Male Age group 6-12 13-17 18-34 35-49 50-82 Muscle pain No Yes Dyspnea No Yes Sore throat No Yes 725 661 643 141 322 176 104 455 931 30 56 374 1,012 197 219 210 70 94 34 8 113 303 400 16 97 319 Relative risk 1.00 1.16 1.00 1.34 0.91 0.65 0.29 1.00 1.23 1.00 0.96 1.00 1.16 95% condence interval Reference 0.96-1.41 Reference 1.02-1.76 0.71-1.17 0.45-0.94 0.14-0.58 Reference 0.99-1.53 Reference 0.58-1.58 Reference 0.93-1.46 P
0.120
0.031 0.490 0.023 0.001
0.056
0.876
0.191
Tabel 2. Relatioship between clinical symptoms and risk of Inuenza A Inuenza A Negative Positive (n=1386) (n=416) Runny nose No Yes Sore throat No Yes Ever had fever for the last 2 days No Yes 190 1,196 374 1,012 410 258 27 389 97 319 45 47 P Adjusted relative risk* 95% condence interval 1.00 3.59 1.00 1.54 1.00 1.42 Reference 1.34-9.63 Reference 0.95-2.58
0.011
0.082
Reference 0.97-2.07
0.069
*Adjusted each other among risk factors listed n this Table
Vol. 2, No. 2, December2011
Clinical Symptoms Associated with Inuenza A
99
DISCUSSION In interpreting our nding there are some limitation that must be considered. First, we did not include fever and cough as examined symptoms because those symptoms had been included in CDC protocol for inuenza. Therefore, every subject must have fever and cough. Second, we only studied ve symptoms. It is a very small number if we compare to the number of symptoms that have been known. Third, for this study we did not analyze for combination of symptoms. Symptoms of inuenza can start quite suddenly one to two days after infection.5 These symptoms may include fever, cough, body aches, especially muscle and throat, and dyspnea.1,2 It is difcult to differentiate inuenza to inuenza-like illnesses in the early stages of these infections.5 One study suggested that during local outbreaks of inuenza, the prevalence will be over 70%.6 Rapid laboratory tests can be used to detect inuenza A, especially during the inuenza season. However, it is not frequently experienced because the test itself is usually expensive.6 According to the CDC, rapid diagnostic tests have a sensitivity of 7075% and specicity of 9095% when compared with viral culture.7,8 Unlike viral culture or rapid test, RT-PCR test has more sensitivity in detection and ability to identify which inuenza A subtypes. Therefore, CDC recommends RT-PCR for inuenza for surveillance purposes.9 Our study reveals that among individuals with inuenza A, the most frequent reported symptoms were runny nose (n=389), sore throat (n=319), and muscle pain (n=302). In contrary, Monto et al. in a study stated different result. They said that the most common symptoms for inuenza feverishness, cough, myalgia, and weakness.10 Another study imply that malaise, fever, and cough are the most common symptoms for inuenza.11 In addition, Eccles and Cao et al. also stated fever and cough as the best predictors.5,8 However, in our study we could not use fever and cough as variables because they are included in ILI denition. Furthermore, similar distribution is shown between subjects age 6-12 years and 18-34 years old (P = 0.490). Male subjects had more likely increase risk to have inuenza A. Dyspnea are distributed similarly between those who had and did not have inuenza
A. Male subjects and those who had muscle pain than sore throat more likely had increase risk to be inuenza A. Compared to age 6-12 years old, subjects of age 13-17 and 35 or more had more likely increase risk to be inuenza A. Moreover, Govaert et al. in a study analyzed combination of symptoms and revealed that combination of fever, cough, and acute onset had a high predictive value of inuenza.11 In conclusion, close observation into ILI surveillance in 2009 offered some dominant symptoms (runny nose, sore throat, and ever had fever) that can be used in order to predict inuenza A in patients. Acknowledgments The authors wish to thank all subjects who participated in this study, ILI surveillance team (laboratory, data manajemen and data collection), and CDC Atlanta for funding this ILI surveillance. Lastly to Vivi Setiawaty for her reviewing the nal draft of this manuscript and her suggestion.
REFERENCES
1. Center for Biomedical and Pharmacy and Directorate of Communicable Disease and Environment. Guidance on inuenza-like ilness (ILI) epidemiology and virology surveilance in health center and hospital. Jakarta. The center. 2006. http://www.cdc.gov/u/index.htm. Indonesia. Kasper DL, Fauci AS, Longo DL, et al. Harrisons Principles of internal medicine. 16th edition. New York: McGraw-Hill; 2005. Kamps BS, Hoffmann C, Preiser W. Inuenza report 2006. Paris: Flying Publisher; 2006. Eccles R. Understanding the symptoms of the common cold and inuenza. Lancet Infect Dis. 2005;5:718-25. Rothberg MB, He S, Rose DN. Management of inuenza symptoms in healthy adults. J Gen Intern Med. 2003;18:808-15. Inuenza (Flu) Antiviral Drugs and Related Information [cited 2011 November 18]. Available from http://www. fda.gov/drugs/drugsafety/informationbydrugclass/ ucm100228.htm#AntiviralMedications Cao B, Li XW, Mao Y, Wang J, et al. Clinical features of the initial cases of 2009 pandemic inuenza A (H1N1) virus infection in China. N Engl J Med. 2009; 361: 2507-17. Centers for Disease Control and Prevention. Interim guidance for inuenza surveillance: prioritizing RT-
2.
3. 4. 5.
6.
7.
8.
100 Roselinda and Fitri et al.
9.
PCR testing in laboratories [cited 2011 November 10]. Available from http://www.cdc.gov/h1n1u/screening. htm Monto AS, Gravenstein S, Elliott M, et al. Clinical signs and symptoms predicting inuenza infection. Arch Intern Med. 2000;160:3243-7.
10. Govaert ThME, Dinant GJ, Aretz K, et al. The predictive value of inuenza symptomatology in elderly people. Fam Prac. 1998;15:16-22. 11. Govaert ThME, Dinant GJ, Aretz K, et al. The predictive value of inuenza symptomatology in elderly people. Fam Prac. 1998;15:16-22.

2-HSJI Vol 2 (2) 30 Jan 2012

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

2-HSJI Vol 2 (2) 30 Jan 2012

Transféré par

Droits d'auteur :

Formats disponibles

Vol. 2, No.

Volume 2, number 2, December 2011, pages 51-100, ISSN 2087-7021

Health Science Journal of Indonesia

D. Astuti B. Basuki H. Mulijadi R. P. Jekti E. Suarthana Suparmi B. Basuki

Health Science Indones

Acknowledgment of Reviewers Volume 2, 2011

Joedo Prihartono Retno Gitawati

HSJI international advisor

HSJI international advisor, France

Faculty of Medicine, Universitas Indonesia, Jakarta

Former WHO consultant; HSJI international advisor; Minnesota, USA

Faculty of Medicine, Universitas Indonesia, Jakarta

Faculty of Medicine, Universitas Indonesia, Jakarta

Air Traffic Control of Soekarno-Hatta Airport, Jakarta

Health Science Journal of Indonesia

Vol. 2, No. 2, December 2011

Health Science Indones

Vol. 2, No. 2, December 2011

Inter-rater reliability of HoNOS

Corresponding author: Sri Idaiani E-mail: sriidaiani@yahoo.com

Health Science Indones

Vol. 2, No. 2, December 2011

Inter-rater reliability of HoNOS

HoNOS: English version Group discussion Translation and back translation

HoNOS: Indonesian version

Pilot study: assessment of 10 psychotic patients by 6 nurses at Community Health Centers

Figure 1. Flow of research

Health Science Indones

Table 2. Some demographic and employment characteristics of the nurses (raters)

Table 3. Intra-class correlation coefcients of HoNOS items

Vol. 2, No. 2, December 2011

Inter-rater reliability of HoNOS

Health Science Indones

Science Indones 2011;2:58-65)

(Health Science Indones 2011;2:58-65)

Vol. 2, No. 2, December 2011

Air trafc controller and qualitative stressor

Health Science Indones

Vol. 2, No. 2, December 2011

Air trafc controller and qualitative stressor

95% condence interval

Reference 8.21-0.82 Reference 0.08-6.65 0.15-22.53 0.09-10.01 Reference 0.02-0.93

0.785 0.630 0.863

Reference 0.58-11.55 0.28- 7.33 Reference 0.20-3.31 0.11-11.64

1.00 0.30 0.55

Reference 0.08-1.17 0.11-2.64

Health Science Indones

0.066 0.562 0.623

1.00 1.63 n/a 1.00 0.21

Crude odds ratio

95% condence interval

Vol. 2, No. 2, December 2011

Air trafc controller and qualitative stressor

Reference 0.33-3.79 0.26-21.77

*Adjusted each others for risk factors listed on this Table.

Health Science Indones

Vol. 2, No. 2, December 2011

Air trafc controller and qualitative stressor

Jekti and Suarthana

Health Science Indones

Risk factors of post partum haemorrhage in Indonesia

Corresponding author: Rabea Pangerti Jekti E-mail: yekti.yekti.24@gmail.com