Asian Nursing Research 9 (2015) 318e327

Contents lists available at ScienceDirect

Asian Nursing Research

journal homepage: www.asian-nursingresearch.com

Research Article

Blended Infant MassageeParenting Enhancement Program on

Recovering Substance-Abusing Mothers' Parenting Stress, Self-Esteem,
Depression, Maternal Attachment, and Mother-Infant Interaction
Luz S. Porter, PhD, ARNP, FAANP, FAAN, 1, * Brian O. Porter, MD, PhD, MPH, 2
Virginia McCoy, PhD, 3 Vivian Bango-Sanchez, EdD, ARNP, 4 Bonnie Kissel, EdD, ARNP, 4
Marjorie Williams, DNP, ARNP, FNP-BC, 5 Sachin Nunnewar, MD, MPH 6
1
College of Nursing and Health Sciences, Florida International University, Miami, FL, USA
2
External IMPEP Study Co-Investigator, University of Miami, Miami, FL, USA
3
Department of Health Promotion and Disease Prevention, Robert Stemple School of Public Health, Florida International University, Miami, FL, USA
4
Memorial Regional Hospital, Hollywood, FL, USA
5
AIDS Healthcare Foundation, Miami Beach, FL, USA
6
IMPEP Study Team, Robert Stemple School of Public Health, Florida International University, Miami, FL, USA

a r t i c l e i n f o s u m m a r y

Article history: Purpose: This study aimed to determine whether a blended Infant MassageeParenting Enhancement
Received 25 July 2014 Program (IMPEP) improved maternal psychosocial health outcomes (parenting stress, depressive
Received in revised form symptoms, self-esteem, maternal attachment) and maternal-infant interaction among substance-
7 August 2015
addicted mothers (SAMs) actively engaged in outpatient rehabilitation.
Accepted 16 September 2015
Methods: Designed as a randomized, three-group controlled trial testing two levels of psychoeducational
intervention (IMPEP vs. PEP) and a control group (standard care parenting resources), the study was conducted
Keywords:
in two substance abuse centers in southeast Florida on a convenience sample of 138 recovering SAM-infant
massage
mental health
pairs. IMPEP or PEP classes were held weekly on Weeks 2e5, with data collected at baseline (Week 1), Week
mothers 6, and Week 12 via structured interviews, observation (Observation Checklist on Maternal-Infant Interaction),
parenting education and self-administered questionnaires (Abidin Parenting Stress Index, Beck Depression Inventory, Rosenberg
substance-related disorders Self-Esteem Scale, Muller's Maternal Attachment Inventory), analyzed descriptively and inferentially using
Kruskall-Wallis analysis of variance and post hoc Wilcoxon rank sum and Mann-Whitney U tests.
Results: Both IMPEP and PEP groups had signicantly increased Parenting Stress Index scores (decreased
parenting stress) and decreased Beck Depression Inventory scores (decreased depressive symptoms)
compared to controls at Week 12, whereas there were no clinically meaningful differences among study
groups in Rosenberg Self-Esteem Scale, Muller's Maternal Attachment Inventory, or Observation
Checklist on Maternal-Infant Interaction scores. Only the IMPEP group showed signicant improvements
in both psychological and physical (waist-hip ratio) measures of parenting stress over time.
Conclusions: The ndings suggest that infant massage blended into a structured parenting program has
value-added effects in decreasing parenting stress and maternal depressive symptoms, but not on SAM's
self-esteem, attachment, or maternal-infant interaction.
Copyright 2015, Korean Society of Nursing Science. Published by Elsevier. This is an open access article
under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction
At the height of public health concern in the 1990s, U.S. health
agencies estimated nearly 4.6 million women of childbearing age
* Correspondence to: Luz S. Porter, PhD, ARNP, FAAN, FAANP, College of Nursing
and Health Sciences, Florida International University, 16295 SW 14th Street,
Pembroke Pines, FL 33027, USA.
E-mail addresses: luzporter@comcast.net, porterl@u.edu
regularly used cocaine, with drug-exposed births rising from 300,000
in 1992 to 750,000 in 1995 [1e3]. Given the myriad of negative life
events (e.g., domestic violence, sexually transmitted infections)
experienced by substance-abusing mothers (SAMs) even when in
recovery, SAMs are vulnerable to mental health problems including
parenting stress, depression, low self-esteem, and deviant maternal
attachment [4]. The purpose of this study was to evaluate the effects
of a blended Infant MassageeParenting Enhancement Program
(IMPEP) on mental health outcomes among recovering SAMs.

http://dx.doi.org/10.1016/j.anr.2015.09.002
p1976-1317 e2093-7482/Copyright 2015, Korean Society of Nursing Science. Published by Elsevier. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).
 L.S. Porter et al. / Asian Nursing Research 9 (2015) 318e327
Parenting stress (i.e., feeling unduly pressured by infant care and
the parenting role) can negatively impact child development, either
directly or indirectly through parenting behaviors [5,6]. For
recovering SAMs, self-consciousness over a history of incarceration,
mandated rehabilitation programs, and vulnerability to recidivism
may impair the ability to cope with the stresses of parenting related
not only to the need to care for their infants but also to the need to
satisfy intense cravings created by their substance addiction [7].
Depression is also a recognized risk factor for drug and alcohol
addiction [8,9], as more than one-third of females using drugs
having had a major depressive episode [2]. Depression develops
most frequently in people's twenties [10], which corresponds to the
period of greatest fertility. An estimated 10%e20% of women
struggle with major depression before, during, or after childbirth
[11]. Women who are of lower socioeconomic status, less educated,
unemployed, and from ethnic/racial minority populations are more
likely to experience depression [12e14]. Thus, risk factors associ-
ated with both depression and maternal substance abuse
contribute to a vicious cycle for recovering SAMs, with their infants
being at increased risk of developing insecure attachments, nega-
tive affect, and dysregulated attention and arousal [15].
Potentially serving as a counterbalance to these negative
psychologic states, self-esteem is the disposition to experience
oneself as competent to cope with basic life challenges and
worthy of happiness [16]. Self-esteem is inuenced by both
environmental and psychological factors [17,18], with low self-
esteem leading to dysfunctional relationships, neurosis, anxiety,
and depression, acting as a universal denominator in addiction
[19]. Positive self-esteem can serve as a buffer from anxiety, guilt,
shame, criticism and depression, thereby promoting mother-
infant interaction.
Developmental theorists assert the most important task of an
infant's life is to establish basic trust and security in others [20,21].
Attachment involves a reciprocal, interactive system between
mother and infant in which both individuals are active participants,
with infants evoking responses from adults who are sensitive and
committed to observe, understand, and respond appropriately to
cues. Attachments vary in quality from secure to anxious and may
be thwarted by risk factors such as poverty, mental illness, maternal
substance abuse, and lack of social support [20,22]. In the case of
SAMs, mother-infant interactions are at risk due to the mother's
guilt, poor self-image, and lack of parenting condence [23], as well
as the infant's impaired ability to send clear cues and respond
favorably to the mother.
Multidimensional intervention programs have been tested on
both SAMs and depressed mothers to improve parenting skills and
prevent negative sequelae such as child abuse [24,25]. However,
these studies did not incorporate infant massage as a key modality.
Practiced in many countries for centuries, infant massage is
reective of different cultural practices involving physical touch to
communicate warmth and gentleness to the infant [26].
Individual trials have suggested the benets of infant massage
for mother-infant bonding and infant health outcomes, including
protection against infection, weight gain in both preterm and full-
term infants, improvements on the Bayley Infant Scales of Mental
and Motor Development, and increased bone mineral density
[27e30]. However, more recent studies [31] and a Cochrane review
[32] of 34 randomized infant massage trials did not conrm sta-
tistically signicant improvements in infant measures of physical
and mental development. As this metaanalysis was focused on low-
risk groups of parents and infants, its ndings may not generalize to
SAMs and their infants. In turn, another recent metaanalysis
conrmed infant massage led to statistically signicant improve-
ments in daily weight gain in preterm infants and reduced their
length of hospital stay [33].
319
To date, infant massage has been applied primarily in hospital
settings, with most controlled studies focused only on infant health
outcomes. However, infant massage holds the promise of
improving mental health outcomes in high-risk mothers as well, as
studies have suggested infant massage may improve mother-infant
interaction in mothers with postpartum depression [34,35] and
decrease depressive symptoms in mothers of preterm infants [36].
Purpose of study
This study was conducted to evaluate the effects of infant
massage integrated into a structured multidimensional parenting
enhancement program (PEP) on key mental health outcomes
among recovering SAMs. We hypothesized that compared to a PEP
intervention without infant massage or standard of care parenting
education, the IMPEP will decrease parenting stress and depression
in recovering SAMs, while improving self-esteem, maternal
attachment, and mother-infant interaction.
Methods
Study design
This 3-month study utilized a randomized, controlled three-
arm experimental design with repeated measures over time in
maternal health outcomes to compare the postintervention ef-
cacy of each of two levels of psychoeducational intervention
(IMPEP and PEP) versus a standard of care educational control
(Control) in recovering SAMs, who were evaluated on-site at their
outpatient drug treatment facilities (Figure 1). In order to ensure
timely accrual of a sufcient number of participants for the
classroom-based interventions, clusters of 3e6 recovering SAMs
and their infants (SAM-infant dyads) were randomly assigned
within each study site to one of the three study groups in a 1:1:1
ratio. All patients underwent an extensive baseline evaluation at
Week 1, completing several psychosocial and physiological in-
struments to capture levels of parenting stress, depressive symp-
toms, self-esteem, and maternal attachment. This was followed by
a 4-week intervention period, with members of the IMPEP and PEP
groups taking once weekly classes on a consistent (but separate)
day and time at Weeks 2, 3, 4, and 5. Members of the Control group
received no specic group-based intervention and received only
the standard of care parenting education materials offered to all
recovering SAMs (including IMPEP and PEP participants) at their
respective outpatient treatment programs. At Week 6, a post-
intervention evaluation was completed with the same psychosocial
and physiological battery, which was repeated at Week 12. Stra-
tegies to maximize data integrity and minimize between group
cross-contamination in this study have been described previously
in a methodological article [37].
IMPEP group
Infant massage techniques were taught and practiced only
within the IMPEP intervention, which consisted of four weekly 2-
hour classes of 4e6 mother-infant dyads each, held in a dedicated
meeting space at each study site. These psychoeducational group
sessions were taught by certied nurse practitioners and incorpo-
rated both demonstration and supervised practice of infant mas-
sage techniques, interactive group discussions, as well as question-
and-answer periods focused on childcare practices. Mothers
learned simple games and interactive action-based songs to pro-
mote infant stimulation and mother-infant communication during
infant massage sessions. The IMPEP curriculum also included
demonstration and practice sessions on properly assessing infant
320 L.S. Porter et al. / Asian Nursing Research 9 (2015) 318e327
n n
n n
n n
Figure 1. Participant disposition: screening, randomization, intervention, and assessment Note. IMPEP infant massageeparenting enhancement program, PEP parenting
enhancement program.
temperature, pulse, and respirations, as well as infant cardiopul-
monary resuscitation [38]. Group participants had a 10-minute
break after the rst hour of each class and were provided re-
freshments consisting of fresh fruits, orange juice, and cookies. The
last class included a party celebrating the mothers' completion of
the program, and each mother received an appreciation certicate
and a gift basket of childcare items for her baby, tailored to the age
of her infant.
PEP group
Mothers in this study group participated in four weekly PEP
classes of 4e6 mother-infant dyads, each lasting for 1.5 hours and
held in a dedicated meeting space at each study site. The PEP cur-
riculum was identical to that of the IMPEP and followed the same
sequence of activities, other than excluding infant massage teach-
ing and practice. PEP classes were taught by the same certied
nurse practitioners who also taught IMPEP classes at each study
n
participants
n
n
n
site. PEP classes were scheduled on alternate days as IMPEP ses-
sions to reduce cross-contamination between study groups. How-
ever, interaction among SAMs within the same treatment group
was encouraged as a means of social support. As in the IMPEP
group, a celebratory party was held in honor of the mothers upon
completion of the PEP, and each mother received an appreciation
certicate and an age-appropriate gift basket for her baby, as for the
IMPEP group.
Control group
Mothers in the Control group did not participate in either the
IMPEP or PEP classes. However, they were provided with the usual
health care, social services, and parenting education already avail-
able to all mothers (including IMPEP and PEP participants) at their
respective service agencies, and were assessed with the same
psychosocial and physiological battery on an identical timeline as
the IMPEP and PEP groups. An appreciation certicate and an age-
 L.S. Porter et al. / Asian Nursing Research 9 (2015) 318e327 321

appropriate gift basket for her baby were given to each mother in Table 1 Sample Demographics by Study Group.
the Control group upon study completion, as for the IMPEP and PEP
Variables IMPEP (n 62) PEP (n 37) Control (n 39)
groups.
n % n % n %
Setting and sample Ethnicity
Asian/Pacic Islander 3 4.8 0 0 0 0
The study was conducted from September 1, 2003 to August 31, Black 17 27.4 15 40.5 19 48.7
Hispanic 9 14.5 5 13.5 9 23.1
2008 at two community addiction treatment centers in Miami- Native American 2 3.2 1 2.7 0 0
Dade and Broward Counties, Florida. Both centers offered resi- White 30 48.4 15 40.5 11 28.2
dential addiction treatment services in an outpatient setting. Other 1 1.6 1 2.7 0 0
Assuming an attrition rate of 25.0%, a Type I error or a of .05, and a Marital status
Single 51 82.3 27 73.0 29 74.4
medium effect size (.25), a sample size of 52 participants per arm
Married 5 8.1 4 10.8 4 10.3
was originally predicted to provide 80.0% power to detect a within- Divorced/widowed 6 9.7 6 16.2 6 15.4
subjects change of this magnitude following the IMPEP interven- Education level
tion [39]. However, the nal completed sample of 121 mother- 0e6 yr (elementary) 3 4.8 0 0 2 5.1
infant dyads at Week 12 did not meet this goal. 7e12 yr (secondary) 48 77.4 30 83.3 30 76.9
Technical/vocational 10 16.1 5 13.9 5 12.8
A total of 170 recovering SAMs were screened for the study, with College 1 1.6 1 2.8 2 5.1
the most common reason for screen failure being lost to follow-up. Annual income
Of the initial 138 SAMs who consented and enrolled in the study, 17  $10,000 51 89.5 27 81.8 37 100.0
mothers failed to complete the study, resulting in an overall $11,000e$20,000 2 3.5 1 3.0 0 0
$21,000e$30,000 2 3.5 1 3.0 0 0
retention rate of 87.7% through Week 12. As depicted in Figure 1,
$31,000e$40,000 1 1.8 2 6.1 0 0
the nal study sample consisted of 51 mother-infant dyads in the  $41,000 1 1.8 2 6.0 0 0
IMPEP group (82.3% retention), 35 in the PEP group (94.6% reten- Living arrangements
tion), and 35 in the Control group (89.7% retention) who, in addi- With parents 3 4.9 1 2.7 3 7.7
tion to the baseline assessment, had at least one other Relatives/friends 10 16.4 7 18.9 9 23.1
Alone with baby 36 59.0 22 59.5 16 41.0
postintervention assessment at Week 6 or Week 12. With spouse/partner 2 3.3 3 8.1 5 12.8
The convenience sample was composed of 138 SAMs (of 170 Other 10 16.4 4 10.8 6 15.4
screened) and their infants aged 4e52 weeks at the time of study Children
enrollment, recruited from two community addiction treatment 1 16 26.2 5 13.5 17 43.6
2 16 26.2 9 24.3 9 23.1
centers in southeast Florida, serving a predominantly low-income
3 9 14.8 8 21.6 6 15.4
client-base. All SAMs registered at these sites were considered for 4 13 21.3 7 18.9 4 10.3
study participation, regardless of drugs of abuse (including 5 7 11.4 6 21.6 3 7.7
alcohol), age, race/ethnicity, marital status, religion, education, or Miscarriages or abortions
income level, although participation was entirely voluntary. Inclu- None 29 46.8 17 45.9 20 51.3
1e2 19 30.6 15 40.5 15 38.5
sion criteria required mothers to have an infant 12 months of age or
3e4 12 19.4 5 13.5 4 10.3
younger, speak and comprehend either English, Spanish, or Creole, 5 2 3.2 0 0 0 0
independently perform activities of daily living, maintain attention Gestation
and follow directions, and willingly provide informed consent in Full-term 44 84.6 32 91.4 37 94.9
Premature 8 15.4 3 8.6 2 5.1
writing. For mothers with more than one infant, only the mother's
Method of delivery
youngest infant was enrolled in the study and comprised the Natural childbirth 27 51.9 22 62.9 26 66.7
mother-infant dyad. The ethnically diverse nature of south Florida Caesarian section 25 48.1 13 37.1 13 33.3
was reected in the ethnic distribution of the study sample Gender of infant
(Table 1). Male 26 41.9 17 45.9 20 51.3
Female 36 58.1 20 54.1 19 48.7
Use of street drugs
Ethical considerations No 7 11.5 7 18.9 6 15.4
Yes 54 88.5 30 81.1 33 84.6
Participant recruitment started upon approval of the research Drink alcoholic beverages
No 44 71.0 27 73.0 26 66.7
protocol by the Florida International University Institutional Re-
Yes 18 29.0 10 27.0 13 33.3
view Board and the chief executive ofcers of all collaborating
agencies. Study participation was entirely voluntary, and although Note. IMPEP infant massage-parenting enhancement program; PEP parenting
enhancement program.
participants were encouraged to attend all intervention classes and
complete all follow-up assessments, mothers and their infants
were free to leave the study at any time, incurring no penalty or range: 36e180), a sphygmomanometer for systolic and diastolic
limitation in services from their respective treatment centers. This blood pressure measurements (measured in a seated position
study was conducted following the principles of ethical clinical preferentially in the right arm following 5 minutes of rest, per
research described in the Declaration of Helsinki. Centers for Disease Control and Prevention guidelines) with higher
readings reecting greater stress levels [41], and waist-to-hip ratio
Measurements/instruments [42] as an indirect marker of fat deposition related to cortisol stress
hormone levels with higher ratios reecting greater stress levels
Outcome variables were measured using both objective and [43]. Depressive symptoms were measured using the Beck
subjective assessment tools that are well established and validated Depression Inventory (BDI) [44,45] with higher scores indicating
in clinical research literature. The psychosocial and physiological greater depressive symptoms (Cronbach a .80e.92; 21 items, 4-
dimensions of parenting stress were quantied using the Parenting point scale, and range: 0e63). Self-esteem was measured using
Stress Index/Short Form (PSI) [40] with lower scores indicating the Rosenberg Self-Esteem Scale (RSE) [44,46] with higher scores
greater stress levels (Cronbach a .70e.71; 36 items, 5-point scale, indicating greater self-esteem levels (Cronbach a .87; 10 items, 4-
322 L.S. Porter et al. / Asian Nursing Research 9 (2015) 318e327
point scale, range: 10e40). Maternal attachment was measured
using the Maternal Attachment Inventory (MAI) [47], with higher
scores indicating greater attachment (Cronbach a .76; 26 items,
4-point scale, range: 26e104) and the Observation Checklist on
Mother-Infant Interaction (OMII) [48]. Consistent with the reli-
ability indices reported in the literature, the self-reported ques-
tionnaire instruments demonstrated high reliability within the
current sample. Cronbach a for PSI was .95; it was .90 for BDI, .92 for
RSE, and .95 for MAI.
Data were collected with the OMII as previously described [48].
Maternal and infant interactive behaviors were coded separately by
two trained raters from sessions videotaped through a one-way
mirror, using a coding system that scored both frequency and
duration of interaction during the last 3 minutes of a 5-minute free
play period and the rst 2 minutes of a mother-infant reunion
period following a planned 2-minute separation. Verbal and
educational interaction, visual monitoring, and structuring of toys
by the mothers were coded, as well as warmth and negative affect
toward their infants. Infants' positive affect and anger were coded
during the free play phase, while secure/approach and insecure/
avoidant measures were coded during the reunion phase. A scoring
methodology was applied in which points were awarded for infant
behaviors reecting secure attachment and subtracted for avoidant,
resistant, or insecure behaviors, with higher scores reecting more
positive interaction. Interrater reliability for the OMII (r .95) was
established prior to the start of study by utilizing a panel of 5 in-
dependent, blinded raters to evaluate 10 mother-infant dyads,
which demonstrated 80%e100% agreement (at least 4 of 5 raters)
on all OMII items across all 10 mother-infant dyads.
Nursing intervention
The intervention that was tested in this study is a blended
IMPEP, based upon Porter's PEP [49], which was originally devel-
oped as a systematically planned, cognitive-behavioral, client-
centered, health-focused educational parenting program for high-
risk mothers and has been validated in multiple studies across
cultures [50,51]. Built upon the belief that mothers are the primary
caregivers and health resource for the family, the PEP is delivered
by qualied nurses, with knowledge reinforcement through
demonstration and return demonstration, covering several
parenting topics, including growth and development, basic infant
health assessments, child safety, infant cardiopulmonary resusci-
tation [38], self-care, and relaxation techniques for the mother.
The IMPEP is an enhanced version of the PEP in which infant
massage teaching has been incorporated, based on previously
described massage techniques [26]. These infant massage tech-
niques apply slow, gentle, and rhythmic strokes on the skin sur-
face, varying from light rubbing to light kneading and gentle
squeezing, with just enough pressure to stimulate blood circula-
tion, while conveying love and condence from parent to child. A
detailed description of the infant massage protocol has been
published previously [52]. In brief, each IMPEP class is allotted 2
hours, allowing for demonstration and return-demonstration of
the infant massage technique (30 minutes), a 10-minute break for
snacks and back stretch, a focused discussion, and an open ques-
tion and answer session. Mothers will be taught infant massage
techniques sequentially in parts (legs and feet, abdomen, arms and
hands, face, back) and directed to practice infant massage at home
for at least 10 minutes each morning and evening. It is anticipated
that by Week 3 of the study period, each mother will have learned
all infant massage components. Mothers will also be taught games
and songs that can be used during infant massage to further
promote communication that stimulates the infant's developing
senses. All teaching materials were translated into Spanish and
Creole to reect the large Hispanic and Haitian communities of
South Florida.
Development and evaluation of the IMPEP was guided by Ban-
dura's Social Learning Theory [53], which postulates that a person's
decision to change or avoid a behavior is determined by that per-
son's expectations and incentives, as the individual must have
knowledge regarding causal relationships and believe that she is
capable of affecting her own behavior. She must acquire new skills
through behavioral rehearsal and believe that she can use these
skills to effectively avoid or change adverse behavior, with suf-
cient incentives and positive reinforcement to change or avoid the
behavior. In turn, the IMPEP provides a means of facilitating
recovering SAMs to make cognitive-behavioral changes through
learning and applying new parenting knowledge and skills, with a
special focus on infant stimulation via massage taught in an inter-
active group setting. Mothers engaged in infant massage are ex-
pected to gain condence in applying specic massage strokes, be
encouraged to listen to their baby's body signals, and respond aptly,
thereby becoming empowered to help their infant relieve
discomfort. The IMPEP also provides social support among mother
peers in a small group setting, while enhancing SAMs' parenting
skills and nurturing relaxed, pleasurable interactions between
mother and infant.
Data collection/procedure
Data were collected via self-administered questionnaires,
guided interviews, body measurements, and blood pressure as-
sessments, by certied nurse practitioners who were different from
those who taught IMPEP/PEP classes and were blinded to group
assignment. Demographic data were collected on ethnicity, age,
educational level, annual income (as a measure of socioeconomic
status), and history of drug abuse, pregnancy, and childbirth.
Maternal outcome data on parenting stress, depressive symptoms,
self-esteem and maternal attachment were collected at a pre-
intervention baseline assessment, a postintervention assessment
at Week 6, and follow-up at Week 12. Mother-infant interaction
data were collected at baseline, Week 6, and Week 12. Gift cards
worth US$10 were provided to each mother, regardless of treat-
ment group, at the end of each assessment visit, for the purpose of
enhancing participant retention.
Data analysis
The sample distribution of individual outcome variables was
tested for normality using the Kolmogorov-Smirnov and Shapiro-
Wilk tests, which indicated that none of the outcome variables
met the assumption of normality. Therefore, nonparametric
Kruskal-Wallis analysis of variance (ANOVA) with post hoc Wil-
coxon rank sum and Mann-Whitney U tests were used to compare
ranked differences within and between treatment groups, respec-
tively, along with adjusted z scores for two-group pairwise com-
parisons. All tests were conducted using SPSS version 13.0 (SPSS
Inc., Chicago, IL, USA), and results were considered signicant at a
two-sided signicance level of p < .05. Increases in Type I error
resulting from repeated measurements of signicance were
partially controlled by the use of repeated measures ANOVA to test
a series of predetermined comparative hypotheses regarding po-
tential differences between the treatment groups. Mother-infant
dyads could be removed from the study for failure to attend 2 or
more classes, although participants were not replaced, and data
were analyzed on an intention-to-treat basis.
Extreme data values resulting from processing or measurement
errors reecting more than 10.0% of missing data in a specic tool
were excluded from inferential tests, as supported by literature
 L.S. Porter et al. / Asian Nursing Research 9 (2015) 318e327 323

suggesting that 5.0%e10.0% of missing data is the maximum extent Table 2 Descriptive Statistics by Study Group for Measures of Maternal Stress,
allowable for the proposed research [54]. This approach balances Depression, Self-esteem, and Mother-infant Interaction.

the potential loss of power due to a smaller sample size resulting Variable Baseline Week 6 Week 12
from listwise deletions versus a decrease in variance and the
Study Group n M SD n M SD n M SD
introduction of bias associated with mean substitution of missing
values. Parenting Stress Indexa
IMPEP 35 146.66 20.91 41 151.73 21.93 40 158.00 32.70
PEP 28 153.71 20.38 22 159.82 18.73 29 164.86 16.74
Results Control 25 149.25 26.72 24 147.58 25.00 30 147.67 23.50
b
Systolic BP
The results reported in this manuscript are limited to maternal IMPEP 43 119.00 12.42 44 116.75 13.96 43 117.53 13.20
mental and physiological health outcomes. Infant health outcomes PEP 31 121.03 18.60 24 117.46 14.49 21 118.90 11.33
Control 26 113.19 17.20 26 124.12 14.08 27 120.70 16.27
(e.g., growth and development, safety, immunization rates) will be Diastolic BPc
reported separately. IMPEP 43 75.51 12.62 44 76.55 11.25 43 79.28 11.80
PEP 31 81.94 9.81 24 78.25 10.66 21 77.67 10.67
Demographics Control 26 77.46 12.78 26 81.31 13.17 27 78.37 16.33
d
Waist-to-Hip Ratio
IMPEP 62 0.87 0.08 59 0.83 0.08 51 0.85 0.07
Mothers ranged in age from 18 to 42 years (IMPEP mean: 28.4 PEP 37 0.88 0.07 35 0.87 0.06 35 0.85 0.15
years; PEP mean: 29.2 years; Control mean: 25.8 years), and infants Control 39 0.87 0.06 35 0.87 0.07 35 0.87 0.06
were aged 1e4 months at the time of enrollment. Table 1 depicts Beck Depression Inventorye
the demographic distribution of the participants by study groups IMPEP 62 10.84 8.63 59 6.14 8.20 51 3.16 5.31
PEP 37 9.92 9.08 35 6.63 6.17 35 4.91 5.07
and overall sample. Some imbalances were seen between groups in Control 39 7.85 7.33 34 8.50 7.89 35 9.74 9.71
ethnic distribution and the proportion of mothers with only one Rosenberg Self-Esteem f

child, whereas other traits were generally comparable across IMPEP 62 31.08 6.85 59 30.76 7.04 51 30.31 6.91
groups. The SAMs in this study were more likely to report street PEP 37 30.35 8.44 35 31.66 7.90 35 30.37 7.56
Control 39 30.18 7.16 35 27.43 4.73 35 26.23 7.34
drug use (e.g., marijuana, crack, cocaine) than alcohol use, although
Maternal Attachment Inventoryg
specic usage patterns were not quantied in this study. IMPEP 61 3.92 0.25 56 3.88 0.27 50 3.89 0.27
PEP 35 3.91 0.26 34 3.80 0.43 35 3.87 0.27
Parenting stress Control 37 3.75 0.44 30 3.51 0.64 35 3.54 0.71
Observation Checklist on Mother-Infant Interactionh
IMPEP 60 103.35 11.69 59 110.08 11.44 51 119.41 11.50
Consistent with the hypothesis that the IMPEP intervention will PEP 37 103.81 13.63 35 111.80 9.26 35 116.77 9.37
preferentially decrease parenting stress levels, we evaluated Control 39 97.21 10.08 35 106.17 19.68 35 112.74 9.95
whether at Week 12, the IMPEP group demonstrated a greater in-
Note. BP blood pressure; IMPEP infant massageeparenting enhancement
crease in PSI score (reecting greater reductions in parenting program; PEP parenting enhancement program.
stress), a greater decrease in SBP and DBP readings, and a greater a
Range of 35e100 highly stressed, 101e150 moderately stressed,
decrease in WHR compared to the PEP and Control groups (Table 2). 151e175 usually not stressed.
b
Results of the corresponding Kruskal-Wallis ANOVA tests of com- Below 120 normal, 120e139 prehypertension, 140e159 stage 1 hyper-
tension, 160 or more stage 2 hypertension.
parison are summarized in Table 3. Collectively, these results c
Below 80 normal, 80e90 prehypertension, 90e99 stage 1 hypertension,
demonstrate that while there were no statistically signicant dif- 100 or more stage 2 hypertension.
ferences between groups in PSI scores at baseline, only scores for d
For women ideal 0.7, standard 0.8, at-risk 0.9, high risk > 0.9 [42].
e
the IMPEP group increased signicantly over time from baseline to 0e9 no to minimal depression, 10e16 minor depression, 17e29 average
depression, 30e63 major/severe depression.
Week 12 (adjusted z 4.189, p < .001). Moreover, post hoc pairwise f
Less than 15 low self-esteem, 15e25 within normal range, > 25 high self-
comparisons to the Control group at Week 12 showed that PSI esteem.
scores at Week 12 for both the IMPEP (adjusted z 3.409, p .001) g
Mean response of 26 items; range of 1 (lowest score)e4 (highest score).
and PEP (adjusted z 2.816, p .005) groups were signicantly h
Below 60 minimally displayed, 60e90 moderately displayed,
higher than those for the control group, indicating less parenting 91e120 maximally displayed.

stress for both intervention groups at postintervention follow-up.

Similar patterns were observed for physiological stress mea-
established by the World Health Organization [42]. Nonetheless, a
sures (Table 2). A signicant decrease in SBP was seen for the IMPEP
signicant decrease in WHR occurred in the IMPEP group from
group from baseline to Week 6 (adjusted z 1.976, p .048),
baseline to Week 6 (adjusted z 3.114, p .002), although it is un-
whereas the Control group experienced a signicant increase in
clear whether this trend was maintained over time, as WHR began to
SBP from both baseline to Week 6 (adjusted z 2.771, p .006) and
increase again at Week 12 (still remaining lower in magnitude than
baseline to Week 12 (adjusted z 2.690, p .007). For DBP, the only
at baseline). The association of the IMPEP, but not the PEP or Control
signicant decrease over time was seen in the PEP group from
groups, with decreases in WHR was supported by the results of the
baseline to Week 6 (adjusted z 2.190, p .029). Of note, however,
Kruskal-Wallis ANOVA, which was signicant for within group dif-
changes in both mean SBP and DBP over time for the IMPEP and PEP
ferences over time only in the IMPEP group (Table 3).
groups were relatively small in magnitude (2e4 mmHg), and the
clinical relevance of such changes is unclear. However, as reected
in the signicant Kruskal-Wallis ANOVA result for SBP in the Con- Depression
trol group over time (Table 3), the clinical relevance of the change in
SBP from baseline to Week 12 of more than 7 mmHg is more Consistent with the hypothesis that the IMPEP will preferen-
apparent, which was not observed in the IMPEP or PEP groups in tially decrease maternal depression levels (i.e., depressive symp-
this randomized study. toms), we evaluated whether the IMPEP group demonstrated a
As depicted in Table 2, the mean WHR values for all three study greater decrease in scores on BDI at Week 12 compared to the PEP
groups were uniformly high at baseline, Week 6, and Week 12, and Control groups. Summary statistics of BDI scores among study
approaching the at-risk range for coronary artery disease ( 0.9), as groups across assessment periods are depicted in Table 2, with the
324 L.S. Porter et al. / Asian Nursing Research 9 (2015) 318e327

Table 3 Nonparametric Inferential Statistics by Study Group for Measures of Maternal while BDI scores among the three groups did not statistically differ
Stress, Depression, Self-esteem, and Mother-infant Interaction. at baseline, by Week 12 scores in the IMPEP and PEP groups were
Sources Independent Kruskal-Wallis Degrees of p signicantly lower than that in the Control group, with the IMPEP
variables ANOVA freedom group further demonstrating a signicantly lower level of depres-
Parenting Stress Index
sive symptoms than the PEP group at Week 12 (adjusted z 2.588;
Between groups Baseline 5.46 2, 137 .065 p .010), despite similar baseline values. Thus, these ndings
Week 6 5.53 2, 125 .063 indicate that while both the IMPEP and PEP interventions were
Week 12 12.97 2, 118 .002 associated with decreasing BDI scores over time, the magnitude of
Within groups IMPEP 17.54 2, 169 < .001
change was greater in the IMPEP group, which showed signicant
PEP 2.36 2, 103 .307
Control 0.18 2, 108 .912 reductions not only from baseline to Week 12, but also from
Systolic BP baseline to Week 6 (adjusted z 3.852; p < .001) and Week 6 to
Between groups Baseline 5.61 2, 134 .061 Week 12 (adjusted z 2.082; p .037). However, given the overall
Week 6 5.69 2, 129 .058 low BDI scores among all groups at all time points, the clinical
Week 12 2.29 2, 120 .318
signicance of these changes is unclear.
Within groups IMPEP 4.02 2, 172 .134
PEP 1.76 2, 106 .414
Control 10.07 2, 105 .007 Self-esteem
Diastolic BP
Between groups Baseline 5.24 2, 135 .073
Consistent with the hypothesis that the IMPEP will preferen-
Week 6 3.94 2, 129 .140
Week 12 0.51 2, 119 .777 tially increase maternal self-esteem levels, we evaluated whether
Within groups IMPEP 2.34 2, 172 .311 the IMPEP group demonstrated a greater increase in scores on the
PEP 5.19 2, 107 .075 RSE at Week 12 compared to the PEP and Control groups. Overall,
Control 2.21 2, 104 .332 mean RSE scores were uniformly high at baseline among all three
Waist-to-hip ratio
study groups (Table 2) and were reective of high self-esteem
Between groups Baseline 1.18 2, 138 .556
Week 6 13.92 2, 129 .001 across all time points, according to previously described strata
Week 12 2.34 2, 121 .311 [46]. Taken into consideration with the Kruskal-Wallis ANOVA re-
Within groups IMPEP 10.16 2, 172 .006 sults shown in Table 3, neither the IMPEP nor the PEP interventions
PEP 1.41 2, 107 .493
was associated with increased self-esteem scores over time. How-
Control 0.48 2, 109 .789
Beck Depression Inventory ever, as RSE scores in the Control group trended downward over
Between groups Baseline 3.48 2, 138 .175 time (albeit nonsignicantly), scores in the Control group were
Week 6 3.17 2, 128 .205 signicantly lower at both Week 6 and Week 12 compared to those
Week 12 16.31 2, 121 < .001 of the IMPEP and the PEP groups. Given the similarity in scores
Within groups IMPEP 36.01 2, 172 < .001
between the IMPEP and PEP groups at all time points, results from
PEP 8.65 2, 107 .013
Control 0.10 2, 108 .949 this study did not support the hypothesis that the IMPEP leads to
Rosenberg Self-Esteem increased self-esteem in SAMs. However, we cannot rule out that
Between groups Baseline 0.86 2, 138 .650 the IMPEP and PEP interventions contributed to maintenance of
Week 6 10.68 2, 129 .005
self-esteem over time, which was not apparent in the Control
Week 12 7.52 2, 121 .023
Within groups IMPEP 0.54 2, 172 .764
group. Given the overall high self-esteem scores among all three
PEP 0.85 2, 107 .653 groups at all time points, the clinical signicance of these results is
Control 5.75 2, 109 .057 unclear.
Maternal Attachment Inventory
Between groups Baseline 6.20 2, 133 .045
Maternal attachment and mother-infant interaction
Week 6 5.53 2, 120 .063
Week 12 3.82 2, 120 .148
Within groups IMPEP 1.29 2, 167 .526 To assess whether the IMPEP improved attachment of SAMs to
PEP 0.74 2, 104 .691 their infants, we evaluated whether SAM-infant dyads in the IMPEP
Control 1.01 2, 102 .603
group had greater increases in scores on either the MAI or the OMII
Observation Checklist on Mother-Infant Interaction
Between groups Baseline 7.46 2, 136 .024
compared to the PEP and Control groups from baseline to Week 12.
Week 6 2.30 2, 129 .317 Summary statistics for these two instruments are shown in Table 2
Week 12 7.00 2, 121 .030 (with MAI score shown as mean item response, rather than as total
Within groups IMPEP 39.20 2, 170 < .001 score), with the comparative Kruskal-Wallis ANOVA in Table 3. MAI
PEP 18.83 2, 107 < .001
scores were similar in both the IMPEP and PEP groups over time,
Control 31.40 2, 109 < .001
whereas scores decreased in the Control group after baseline, albeit
Note. ANOVA analysis of variance; BP blood pressure; IMPEP infant
not signicantly. Post hoc Mann-Whitney U tests indicated that
massageeparenting enhancement program; PEP parenting enhancement
program.
IMPEP mothers had signicantly greater attachment scores at
baseline compared to Control group mothers did (adjusted
z 2.332; p .020). In addition, pairwise comparisons demon-
results of the Kruskal-Wallis ANOVA for between and within group strated that PEP mothers also had signicantly higher attachment
differences over time shown in Table 3. Mean BDI scores in all 3 scores than Control group mothers did at Week 6 (adjusted
groups across all time points were reective of minimal to minor z 2.279; p .023) and Week 12 (adjusted z 2.589; p .010).
depression, according to previously described strata [45]. None- However, given the higher baseline MAI scores in the PEP and
theless, changes in BDI score over time revealed several notable IMPEP groups versus the Control group at baseline, postbaseline
patterns. differences cannot be clearly attributed to either the PEP or IMPEP
Both the IMPEP (adjusted z 5.872; p < .001) and PEP (adjusted interventions. Similarly, as OMII scores increased signicantly over
z 3.025; p .002) groups had statistically signicant decreases in time within each of the three treatment groups from baseline to
BDI scores from baseline to Week 12, whereas scores in the Control Week 12, these changes cannot be attributed to a specic inter-
group increased over time, albeit not signicantly. In addition, vention. Moreover, although OMII scores were higher in the IMPEP
 L.S. Porter et al. / Asian Nursing Research 9 (2015) 318e327
and PEP groups compared to the control group at Week 12, this
same pattern was also seen at baseline. Thus, the higher post-
baseline mother-infant interaction scores could not be attributed to
the IMPEP or PEP interventions.
Discussion
Although the duration of this study was limited, with the nal
follow-up visit occurring 7 weeks after completion of the IMPEP/
PEP classes, the ndings of this study suggest a blended IMPEP
holds promise for value-added improvements in parenting stress
and depressive symptoms, compared with a structured parenting
program alone. Given the pervasiveness of the negative psycho-
social effects of substance addiction on recovering SAMs, the IMPEP
represents a potential low-input, high-impact intervention to
improve maternal health outcomes and facilitate effective
parenting.
The birth of an in utero drug-exposed infant is both a crisis and
an opportunity for a recovering SAM. As a crisis, the mother may
feel an overwhelming sense of guilt about her baby's condition, but
as an opportunity, the mother may perceive the baby's birth as a
powerful motivation to assume a parenting role through active
recovery. Studies have demonstrated high levels of stress in SAMs
[5], which is a signicant impediment to effective parenting. In this
study parenting stress was dened in a more holistic perspective
than the commonly held psychosocial view, which is measured
only with a behavioral tool. In turn, ndings related to both psy-
chological and physiological measures of parenting stress either
improved or did not worsen in the IMPEP group at Week 12 follow
up, compared to the PEP and Control groups. Specically, only the
IMPEP group demonstrated signicant improvements in PSI score
and WHR over time, whereas systolic blood pressure increased in
the Control group, but not in the IMPEP group, from baseline to
Week 12.
WHR data across all three study groups indicated baseline
values were in the midpoint of the high range (0.85e0.90),
reecting risk of coronary artery disease [55]. Thus, even the rela-
tively small yet statistically signicant decreases in WHR, which
were seen in only the IMPEP group, are of potential clinical sig-
nicance. Moreover, although mean changes in both SBP and DBP
for the IMPEP and PEP groups over time were relatively small
(2e4 mmHg), the signicant Kruskal-Wallis ANOVA for SBP in the
Control group and increase in SBP from baseline to Week 12 of
more than 7 mmHg, which was not observed in the IMPEP or PEP
groups, suggests a benecial effect of the IMPEP (and PEP) inter-
vention on this physiological manifestation of stress.
The IMPEP was also effective in reducing depressive symptoms,
as measured by scores on the BDI, compared to the Control group
through Week 12. Although BDI scores decreased over time in both
the IMPEP and PEP groups, the magnitude of change in the IMPEP
group was slightly greater than in the PEP group, suggesting a
possible value-added effect of infant massage in reducing depres-
sive symptoms among SAMs. Given the clinical signicance of
maternal depression in impacting effective parenting, it is impor-
tant to detect women at risk for developing depressive symptoms
in community-based addiction treatment settings [56,57]. The
present study, however, detected only minimal depressive symp-
toms among recovering SAMs, despite socioeconomic risk factors
(e.g., household income at or below poverty level in over 90% of
mothers). Thus, our ndings do not corroborate those of previous
studies demonstrating SAMs' propensity for depression, particu-
larly when compounded by poor socioeconomic status [58]. It is
possible that the availability of social and rehabilitative services for
recovering SAMs through both public and private agencies
325
facilitated problem-solving and coping strategies in these mothers
to counter depressive inuences [23].
With regard to self-esteem, recovering SAMs in this study
demonstrated high scores on the RSE [16] at all time points. The
high self-esteem scores across all study groups at baseline may
have made it difcult to demonstrate signicant improvements in
self-esteem over time within this sample. Several studies have re-
ported the association of self-esteem with a myriad of demographic
and psychosocial risk factors (57). Although infant massage did not
appear to confer a value-added effect on self-esteem in this study, it
cannot be ruled-out that a structured parenting enhancement
program (either the PEP or IMPEP) may have contributed to the
maintenance of self-esteem in the face of challenges associated
with parenting. Regardless, given the importance of self-esteem to
healthy living and parental adjustment, continued support for
recovering SAMs in the acquisition of effective parenting skills re-
mains pivotal to their recovery process.
Effective parenting by SAMs requires the development of
adaptive parent-infant interaction and adequate social supports,
although the need for secrecy may alienate SAMs from mainstream
culture. In turn, disruptions in early mother-infant interaction
predispose children to depression and relationship problems in
adulthood [11]. The ndings of this study do not support earlier
studies, which reported that massage therapy on infants resulted in
enhanced mother-infant bonding and communication [28,35].
Scores on the MAI were largely stable in both the IMPEP and PEP
groups throughout the study, whereas scores in the Control group
trended downward, albeit not signicantly. Thus, improvements
over time in maternal attachment were not seen, while the
observed increases in mother-infant interaction could not be
attributed to the structured parenting skills program or the infant
massage component.
It is possible that living in residential rehabilitation centers may
have nurtured maternal attachment and spontaneously promoted
mother-infant interaction, as residents are encouraged to develop
informal support systems and attend agency-based parenting
classes, underscoring the importance of integrative social and
rehabilitative services. In addition, the duration of intervention in
this study may have been too limited to effect signicant change in
SAM-infant interaction, as the study only analyzed data up to Week
12, given the high attrition of mother-infant dyads by Week 24 and
Week 48. Participating infants were aged 4e7 months by the time
of this nal analysis, which is younger than the age range for OMII
assessments using the structured reunion scenario, as previously
reported in the literature [48]. Thus, our ndings on maternal
attachment and mother-infant interaction should be interpreted
with caution.
Conclusion
The results of this study demonstrated the value-added effects
of infant massage blended into a systematic parenting skills pro-
gram on reducing parenting stress and maternal depressive
symptoms; although a similar impact on self-esteem, maternal
attachment, and mother-infant interaction was not demonstrated.
Despite the infant massage component of the IMPEP not demon-
strating consistent value-added effects across all the psychosocial
dimensions examined in this study, the potential impact of the
IMPEP is high with respect to parent-child health promotion, as
studies implementing this intervention hold great promise for
expanding the knowledge base of health enhancement among at-
risk populations. The present ndings related to depression and
the psychosocial dimension of parenting stress corroborate previ-
ous studies with respect to the positive health impact of touch,
reinforcing the use of the IMPEP as both a health promotion
326 L.S. Porter et al. / Asian Nursing Research 9 (2015) 318e327
intervention and therapeutic tool by nurse educators, maternal-
child health practitioners, and substance abuse service providers.
With respect to further longitudinal study of SAMs to assess the
long-term impact of the IMPEP, it may be unrealistic to expect
consistent follow-up beyond 3 months, given the attrition rates
observed in this study past Week 12. However, the IMPEP may be
tested further in other vulnerable populations to validate its effects
within different cultural contexts. A cross-sectional study is a
feasible design to determine whether differences in outcomes
among the IMPEP, PEP, and Control groups persist across different
age groups of mothers and children. A mixed research design
merits consideration for future studies to gain greater insights into
the myriad problems confronting recovering SAMs. The primary
limitation of this study was participant attrition which decreased
the nal sample size and limited the study's power. An imbalance in
dropout rates resulted in the IMPEP group being notably larger than
both the PEP and Control groups, potentially limiting the ability of
the randomized, controlled design to fully account for systematic
biases or confounding variables.
Conicts of Interest
We declare no conicts of interest that may bias the study
ndings.
Acknowledgments
We wish to thank all of the mothers and their infants who
participated in the study and the two drug rehabilitation centers
that served as our study sites. We also wish to thank Henriek Meier
(Administrative Assistant) for data processing, Paul Scibelli
(Consultant) for data management, and MINIMAX Consulting for
statistical analysis. This study was supported by the Maternal and
Child Health Bureau (Title V, Social Security Act), Health Resources
and Services Administration, US Department of Health and Human
Services, Grant R40 MC 01063.
References
1. Substance Abuse and Mental Health Services Administration. 2012 national sur-
vey on drug use and health report. SAMHSA newsletter on substance abuse and
mental health. Rockville (MD): Department of Health and Human Services; 2013.
2. Substance Abuse and Mental Health Services Administration. Results from the
2008 National Survey on Drug Use and Health: National Findings. Rockville
(MD): Department of Health and Human Services; 2008.
3. D'Apolito K, Hepworth J. Prominence of withdrawal symptoms in polydrug-
exposed infants. J Perinat Neonatal Nurs. 2001;14(4):46e60.
4. Choi S, Ryan J. Co-occurring problems for substance abusing mothers in child
welfare matching services to improve family reunication. Child Youth Serv
Rev. 2007;29(11):1395e410.
5. Lutz K, Burnson C, Hane A, Samuelson A, Poehlmann J. Parenting stress, social
support, and mother-child interactions in families of multiple and singleton
toddlers born preterm. Fam Relat. 2012;61(4):642e56.
http://dx.doi.org/10.1111/j.1741-3729.2012.00726.x
6. Nair P, Schuler M, Black M, Kettinger L, Harrington D. Cumulative environ-
mental risk in substance abusing women: early intervention, parenting stress,
child abuse potential and child development. Child Abuse Negl. 2003;27(9):
997e1017. http://dx.doi.org/10.1016/S01452134(03)00169-8
7. Field T. Postpartum depression effects on early interactions, parenting and
safety practices: a review. Infant Behav Dev. 2010;33(1):1e6.
http://dx.doi.org/10.1016/j.infbeh.2009.10.005
8. Luthar S, Sexton C. Maternal drug abuse versus maternal depression vulnera-
bility and resilience among school-age and adolescent offspring. Dev Psycho-
pathol. 2007;19(1):205e25. http://dx.doi.org/10.1017/S0954579407070113
9. Riley A, Coiro M, Broitman M, Colantuoni E, Hurley K, Bandeen-Roche K. Mental
health of children of low-income depressed mothers: inuences of parenting,
family environment, and raters. Psychiatr Serv. 2009;60(3):329e36.
http://dx.doi.org/10.1176/ps.2009.60.3.329
10. Marcus S, Flynn H, Blow F, Barry K. Depressive symptoms among pregnant
women screened in obstetric settings. J Womens Health. 2003;12(4):373e80.
http://dx.doi.org/10.1089/154099903765448880
11. McLearn K, Minkovitz C, Strobino D, Marks E, Hou W. Maternal depressive
symptoms at 2 to 4 months post partum and early parenting practices. Arch
Pediatr Adolesc Med. 2006;160(3):279e84.
http://dx.doi.org/10.1001/archpedi.160.3.279
12. Burt V, Stein K. Epidemiology of depression throughout the female cycle. J Clin
Psychiatry (Suppl). 2002;63(7):9e15.
13. Hirth J, Berenson A. Racial/ethnic differences in depressive symptoms among
young women: role of intimate partner violence, trauma, and posttraumatic
stress disorder. J Womens Health. 2012;21(9):966e74.
http://dx.doi.org/10.1089/jwh.2011
14. Kessler R, Berglund P, Demler O, Jin R, Koretz D, Merikangas K, et al. The
epidemiology of major depressive disorder: results from the National Comor-
bidity Survey Replication (NCS-R). JAMA. 2003;289(23):3095e105.
http://dx.doi.org/10.1001/jama.289.23.3095
15. Evenson R, Simon R. Clarifying the relationship between parenthood and
depression. J Health Soc Behav. 2005;46(4):341e58.
http://dx.doi.org/10.1177/002214650504600403
16. Rosenberg M. Society and the adolescent self-image. Revised ed. Middletown
(OC): Wesleyan University Press; 1989.
17. Ansary N, McMahon T, Luthar S. Socioeconomic context and emotional-
behavioral achievement links: concurrent and prospective associations
among low- and high-income youth. J Res Adolesc. 2013;22(1):14e30.
http://dx.doi.org/10.1111/j.15327795.2011.00747
18. Putnick D, Bornstein M, Hendricks C, Painter K, Suwalsky J. Parenting stress,
perceived parenting behaviors, and adolescent self-concept in European
American families. J Fam Psychol. 2010;22(5):752e62.
http://dx.doi.org/10.1037/a0013177
19. Reasoner R. Review for self-esteem research. National Association for Self-
Esteem; 2010.
20. Hamilton C. Continuity and discontinuity of attachment from infancy through
adolescence. Child Dev. 2000;71(3):690e4.
http://dx.doi.org/10.1111/1467-8624.00177
21. Bowlby J. A secure base: parent-child attachment and healthy human devel-
opment. New York (NY): Basic Books; 1988.
22. Perry B. Bonding and attachment in maltreated children: consequences of
emotional neglect in childhood. Parent Caregiver Educ Ser. 2001;1(3).
23. Suchman N, Pajulo M, Decoste C, Mayes L. Parenting interventions for drug
dependent mothers and their young children: the case for an attachment-
based approach. Fam Relations. 2006;55(2):211e26.
http://dx.doi.org/10.1111/j.1741-3729.2006.00371.x
24. Chasnoff I. Drugs and alcohol in pregnancy and the affected children. In:
Lessenger JE, Roper GF, editors. Drug courts: new approach to treatment and
rehabilitation. New York (NY): Springer; 2002. p. 95e111.
25. Jung V, Short R, Letourneau N, Andrews D. Interventions with depressed
mothers and their infants: modifying interactive behaviours. J Affect Disord.
2007;98(3):199e205. http://dx.doi.org/10.1016/j.jad.2006.07.014
26. McClure V. Infant massage: handbook for loving parents. New York (NY):
Bantam Books; 2003.
27. Abdallah B, Badr L, Hawwari M. The efcacy of massage on short and long
terms outcomes in preterm infants. Infant Behav Dev. 2013;36(4):662e9.
http://dx.doi.org/10.1016/j.infbeh.2013.06.009
28. Grol A, Polat S. The effects of baby massage on attachment between mother
and their infants. Asian Nurs Res. 2012;6(1):35e41.
http://dx.doi.org/10.1016/j.anr2012.02.006
29. Reese S, Storm L. Infant massage. New York (NY): Barnes and Noble; 2008.
30. Garmy P. Infant Massageda review. Nordic J Nurs Res Clin Stud. 2007;27(2):
30e4.
31. Seyyedrasooli A, Valizadeh L, Hosseini M, Jafarabadi M, Mohammadzad M.
Effect of vimala massage on physiological jaundice in infants: a randomized
controlled trial. J Caring Sci. 2014;3(3):165e73.
http://dx.doi.org/10.5681/jcs.2014.018
32. Bennett C, Underdown A, Barlow J. Massage for promoting mental and physical
health in typically developing infants under the age of six months. Syst Rev.
2013;(4):5038. http://dx.doi.org/10.1002/14651858.CD005038.pub3
33. Wang L, He J, Zhang X. The efcacy of massage on preterm infants: a meta
analysis. Am J Perinatol. 2013;30(9):731e8.
http://dx.doi.org/10.1055/s-00321332801
34. Poobalan A, Aucott L, Ross L, Smith W, Helms P, Williams J. Effects of postnatal
depression on mother-infant interaction and child development: systematic
review. Br J Psychiatry. 2007;191:378e86.
http://dx.doi.org/10.1192/bjp.bp.106.032789
35. Onozawa K, Glover V, Adams D, Modi N, Kumar R. Infant massage improves
mother infant interaction for mothers with postnatal depression. J Affect Disord.
2001;163(1e3):201e7. http://dx.doi.org/10.1016/S0165-0327(00)00198-1
36. Holditch-Davis D, White-Traut R, Levy J, O'Shea M, Geraldo V, David R.
Maternally administered interventions for preterm infants in the NICU: effects
on maternal psychological distress and mothereinfant relationship. Infant
Behav Dev. 2014;37(4):695e710.
http://dx.doi.org/10.1016/j.infbeh.2014.08.005
37. Porter B, Porter L, McCoy V, Nunnewar S, Lima M, Pryce C. Methodological
challenges in intervention studies. Nurse Res. 2009;63(2):43e63.
http://dx.doi.org/10.7748/nr2009.01.16.2.43.c6761
38. Berg M, Schexnayder S, Chameides L, Terry M, Donoghue A, Hickey R, et al.
American Heart Association guidelines for cardiopulmonary resuscitation and
emergency cardiovascular care. Circulation. 2010;122:S862e75.
http://dx.doi.org/10.1161/CIRCULATIONAHA.110.971085
39. Colton T. Statistics in medicine. Boston (MS): Little, Brown; 1974.
 L.S. Porter et al. / Asian Nursing Research 9 (2015) 318e327
40. Abidin R. Parenting stress index: mannual. 3rd ed. Odessa (TX): Psychological
Assessment Resources; 1996.
41. Joint National Committee 7. Seventh report of the joint national committee on
prevention, detection, evaluation, and treatment of high blood pressure (JNC 7)
[Internet]. Bethesda (MD): National Heart, Blood, and Lung Institute; 2003
[Cited 2014 Jan 6]. Available from: http://www.nhlbi.nih.gov
42. World Health Organization. Waist circumference and waist-hip ratio: report of
a WHO expert consultation. Geneva (Switzerland): WHO Press; 2008.
43. Kubzansky L, Kawachi I, Sparrow D. Socioeconomic status, hostility, and risk
factor clustering in the normative aging study: help from the concept of allo-
static load? Ann Behav Med. 1999;21(4):330e8.
http://dx.doi.org/10.1007/BF02895966
44. Fischer J, Corcoran K. Measures for clinical practice: a sourcebook. 4th ed. New
York (NY): Oxford University Press; 2007.
45. Hicks B. How to interpret the Beck Depression Inventory. 2009 [cited 2013 Nov
12]. Available from: http://www.clivir.com
46. Sinclair S, Blais M, Gansler D, Sandberg E, Bistis K, LoCicero A. Psychometric
properties of the Rosenberg Self-Esteem Scale: overall and across demographic
groups living within the United States. Eval Health Prof. 2010;33(1):56e80.
http://dx.doi.org/10.1177/0163278709356187
47. Muller M. Questionnaire to measure mother-to-infant attachment. J Nurs Meas.
1994;2(2):129e41.
48. Ukeje I, Bendersky M, Lewis M. Mother-infant interaction at 12 months in
prenatally cocaine-exposed children. Am J Drug Alcohol Abuse. 2001;27(2):
203e24.
327
49. Porter L. West Virginia model program on parenting enhancement with high-
risk adolescents. Morgantown (WV): WVU-DHHS, Division of Nursing; 1985.
Grant no. RO1 NUOO810.
50. Porter L, Youssef M, Shaaban I, Ibrahim W. Parenting enhancement among
Egyptian mothers in a tertiary care setting. Pediatr Nurs. 1992;18(40):
329e336,386. http://dx.doi.org/10.4103/1319-1683.68785
51. Porter L, Knicely B. Study of the effects of a parenting enhancement program on
parents' self-esteem, their perception of the infant, and infant health [un-
published master's thesis]. Morgantown (WV): West Virginia University; 1985.
52. Porter L, Porter B. Blended infant massage-parenting enhancement program for
recovering substance abusing mothers. Pediatr Nurs. 2004;30(5):363e372,401.
53. Bandura A. Social foundations of thought and action: a social cognitive model.
Englewood Cliffs (NJ): Prentice-Hall; 1981.
54. Meyers L, Gamst G, Guarino A. Applied multivariate research: design and
interpretation. Thousand Oaks (CA): Sage; 2006.
55. Noble R. Waist-to-hip ratio versus BMI as predictors of cardiac risk in obese
adult women. West J Med. 2001;74(4):240e1.
56. O'Higgins M, St. James I, Glover V. Postnatal depression and mother and infant
outcomes after infant massage. J Affect Disord. 2008;109(1-2):189e92.
http://dx.doi.org/10.1016/j.jad.2007.10.027
57. Rosen L, Breuner C. Primary care from infancy to adolescence. Pediatr Clin
North Am. 2007;54(6):837e58. http://dx.doi.org/10.1016/j.pcl.2007.09.007
58. Suchman E, Luthar S. Maternal addiction, child maladjustment and socio-
demographic risks: implications for parenting behaviors. Addiction. 2000;
95(9):1417e28. http://dx.doi.org/10.1046/j.1360-0443.2000.959141711.x