PERSPECTIVE

camps listed by the United Nations camp-management cluster

reportedly have no water or sanitation agency, and most are far
from reaching the established
guidelines for sanitation in humanitarian emergencies.3 The living conditions of most of Haitis
poor, whether theyre living in
camps or communities, are equal
ly miserable in terms of the risk
of diarrheal disease.
The reported numbers of cases
and deaths, though shocking, represent only a fraction of the epidemics true toll. We have seen
scores of patients die at the gates
of the hospital or within minutes
after admission. Through our network of community health workers, we have learned of hundreds
of patients who died at home or
en route to the hospital. In the
first 48 hours, the case fatality
rate at our facilities was as high
as 10%. Though it dropped to less
than 2% in the ensuing days as
the health system was reinforced locally and patients began to present earlier in the

Responding to Cholera in Post-Earthquake Haiti

course of disease, mortality will

most likely climb as the disease
spreads and Haitis fragile health
system falters.
This most recent crisis in Haiti
has reinforced certain lessons
regarding the provision of services to the poor. Complementary prevention and care should
be the primary focus of the relief effort. Vaccination must be
considered as an adjunct for controlling the epidemic, and antibiotics should be used in the
treatment of all hospitalized patients. These endeavors should
proceed in concert with muchneeded improvements to sanitation and accessibility of potable
water. More generally, reliable
partnerships are essential, especially if local partners are dependable and have practical experience and complementary assets.
Long-term reinforcement of the
public-sector health system is a
wise investment, permitting provision of a basic minimum set of
services that can be built upon in
times of crisis. And community

health workers who can be rapidly mobilized as educators, distributors of supplies, and first
responders are a reliable backbone of health care. In Haiti,
such workers can bring the timesensitive lifesaving therapy of
oral rehydration right to the patients door.
Disclosure forms provided by the authors are available with the full text of this
article at NEJM.org.
From the Department of Global Health and
Social Medicine, Harvard Medical School;
the Division of Global Health Equity,
Brigham and Womens Hospital; and Partners in Health all in Boston.
This article (10.1056/NEJMp1012997) was
published on December 9, 2010, at NEJM
.org.
1. Sullivan CA, Meigh JR, Giacomello AM.
The Water Poverty Index: development and
application at the community scale. Nat Resour Forum 2003;27:189-99.
2. Ministre de la Sant Publique et de la
Population, Haiti. Enqute mortalit, morbidit et utilisation des services (EMMUSIV): Haiti, 2005-2006. (http://new.paho.org/
hai/index.php?option=com_docman&task=
doc_download&gid=25&Itemid=.)
3. 101112 WASH Cluster situation report.
November 12, 2010. (http://haiti.humanitarian
response.info/Default.aspx?tabid=83.)
Copyright 2010 Massachusetts Medical Society.

Antibiotics for Both Moderate and Severe Cholera

Eric J. Nelson, M.D., Ph.D., Danielle S. Nelson, M.D., M.P.H., Mohammed A. Salam, M.B., B.S., and David A. Sack, M.D.
Related article, p. 33

he 2010 Haitian cholera outbreak has pressed local and

international experts into rapid
action against a disease that is
new to many health care providers in Haiti. The World Health
Organization (WHO) has timetested management protocols for
emerging cholera outbreaks. These
protocols have been used by the
Haitian government to fight an
epidemic that is merely one of
several recent tragedies in Haiti.
The use of these protocols has

allowed for a high standard of

care in this complex and evolving medical landscape. But whereas the current WHO choleratreatment protocol (www.who.int/
mediacentre/factsheets/fs107/en/
index.html) recommends anti
biotics for only severe cases, the
approach of the International
Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B),
recommends antibiotics for both
severe and moderate cases.
Several antibiotics are effec-

tive in the treatment of cholera,

including doxycycline, ciprofloxacin, and azithromycin, assuming
that the cholera strain is sensitive. Currently, the epidemic strain
in Haiti is susceptible to tetracycline (a proxy for doxycycline) and
azithromycin but is resistant to
nalidixic acid, sulfisoxazole, and
trimethoprimsulfamethoxazole.
The WHO advocates giving antibiotics to patients with cholera
only when their illness is judged
to be severe. This recommen-

n engl j med 364;1 nejm.org january 6, 2011

The New England Journal of Medicine

Downloaded from nejm.org on May 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.

PERSPE C T I V E

Antibiotics for both Moderate and Severe Cholera

dation is interpreted to mean that

only patients who present with
severe dehydration (10% dehydration) should be given antibiotics. By contrast, the ICDDR,B
recommends antibiotics for patients with cholera who have severe dehydration as well as for
those with some dehydration
(5 to 10%) who continue to pass
large volumes of diarrheal stool
during their treatment. These
recommendations apply only to
patients who have symptoms
typical of cholera that is, less
than 24 hours of acute watery
diarrhea with dehydration and
usually vomiting. It is crucial in
triage to rapidly assess dehydration, rule out alternative causes
of diarrhea that are common in
areas with poor sanitation and
coexisting infections, and rehydrate aggressively according to
the WHO protocols.
With effective antibiotic therapy, the purging rate is lessened by
about 50%, the illness is shortened by about 50%, and the duration of excretion of Vibrio cholerae
in the stool is shortened to 1 or
2 days. Without effective antibiotic therapy, patients continue to
excrete V. cholerae for 5 or more
days and shed for a longer period
at home.1-3 If antibiotics are used,
patients recover more quickly and
require less rehydration fluid.
Nursing care is lessened, and patients are able to leave the treatment center earlier, as demonstrated in a study that showed
dramatic resolution of diarrhea at
24 hours with azithromycin.1 This
approach maximizes the effectiveness of limited resources
while optimizing patient care.
Regarding transmission, ricewater stools contain 1011 to 1012
V. cholerae organisms per liter. An
infectious dose is 105 to 108 or-

ganisms. These numbers might

explain why 50% of household
contacts of a patient who is the
index case in Bangladesh develop
diarrhea about 2 days after the
index case occurs.4 Although
some of these household contacts may have been infected from
the same source as the index patient, many others are likely to be
true secondary cases. Direct data
are not available to determine
whether household contacts are
protected when the index case is
treated with antibiotics. However,
given the liter volumes of diarrhea, antibiotics will decrease
contamination in the household.
We do not, however, recommend antibiotic prophylaxis for
household contacts because of the
programmatic difficulty in restricting the use of such prophylaxis only to those persons in the
immediate family who are at
highest risk5 and because doing
so would almost certainly drive
antibiotic resistance. Since families of patients with cholera are
at high risk for cholera themselves, they need targeted education about safe water and sanitation, appropriate home use of
oral rehydration solution, and information about the availability
of treatment facilities in case illness does occur.
Some may argue that emphasizing the importance of anti
biotic therapy may lead to the
misguided belief that this is the
most important component in
the overall management of patients with cholera. With careful
training in instituting appropriate and aggressive rehydration
followed by effective antibiotic
therapy, this misunderstanding
need not occur.
A practical reason for hesitancy
regarding administering antibiot-

ics to patients with cholera relates to the severe vomiting that

usually accompanies infection.
Vomiting generally stops within
a few hours after patients are
rehydrated; thus, the administration of the antibiotic should be
delayed until the patient is able
to take food and drink without
vomiting. Doxycycline can be associated with nausea and should
be taken with food and plenty of
fluids.
In summary, the use of antibiotics is an urgent issue for all
stakeholders, because effective
antibiotic therapy shortens the
duration of illness and reduces
the shedding of thousands of
infectious doses. Our goal is to
promote more effective care for
large numbers of patients with
cholera while maximizing limited
resources to keep patients who
are discharged early from dying,
reduce the number of repeat hospital admissions, and limit athome shedding of V. cholerae. To
achieve these aims, we believe
that patients with moderate and
severe cholera should be treated
with antibiotics especially in
Haiti, and especially now.
Disclosure forms provided by the authors
are available with the full text of this article at NEJM.org.
From the Lucile Packard Childrens Hospital, Stanford University, Palo Alto, CA
(E.J.N.); the Santa Clara Valley Medical Center, Milpitas, CA (D.S.N.); the International
Centre for Diarrhoeal Disease Research,
Bangladesh, Dhaka, Bangladesh (M.A.S.);
and the Johns Hopkins Bloomberg School
of Public Health, Baltimore, MD (D.A.S.).
This article (10.1056/NEJMp1013771) was
published on December 9, 2010, at NEJM
.org.
1. Saha D, Karim MM, Khan WA, Ahmed S,
Salam MA, Bennish ML. Single-dose azithromycin for the treatment of cholera in adults.
N Engl J Med 2006;354:2452-62.
2. Lindenbaum J, Greenough WB, Islam
MR. Antibiotic therapy of cholera. Bull World
Health Organ 1967;36:871-83.

n engl j med 364;1 nejm.org january 6, 2011

The New England Journal of Medicine

Downloaded from nejm.org on May 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.

PERSPECTIVE
3. Rahaman MM, Majid MA, Alam A, Islam
MR. Effects of doxycycline in actively purging
cholera patients: a double-blind clinical
trial. Antimicrob Agents Chemother 1976;
10:610-2.

Antibiotics for both Moderate and Severe Cholera

4. Weil AA, Khan AI, Chowdhury F, et al.
Clinical outcomes in household contacts of
patients with cholera in Bangladesh. Clin Infect Dis 2009;49:1473-9.
5. Khan MU. Efficacy of short course antibi-

otic prophylaxis in controlling cholera in contacts during epidemic. J Trop Med Hyg 1982;
85:27-9.
Copyright 2010 Massachusetts Medical Society.

Rethinking Safety-Net Access for the Uninsured

Mark A. Hall, J.D.

ow that health insurance

reform has begun, safety-net
programs throughout the United
States are struggling to adapt
their missions to suit the postreform composition of the uninsured population. Most such programs are organized at the local
level, with funding largely premised on their serving low-income
uninsured residents. Examples
include well-structured comprehensive care programs in some
major cities, more than 1000
limited-service free clinics, and
dozens of volunteer physicianreferral programs.
When the Affordable Care Act
(ACA) is fully implemented, 8% of
the U.S. population is projected to
remain uninsured. Other than
undocumented immigrants, however, most such people will be
eligible for Medicaid or highly
subsidized private insurance and
will be subject to tax penalties if
they dont obtain coverage. So
beginning in 2014, most people
who are currently served by access programs for the uninsured
will have insurance, be eligible
for insurance, or be undocumented immigrants.
Some people will remain uninsured because their income is
too high for a subsidy but low
enough to make insurance unaffordable (costing more than 8%
of their household income). But
subsidies will be available to people with family incomes up to

400% of the federal poverty level,

which currently calculates to
$88,200 for a family of four
well above the countrys median
household income of about
$50,000.
Access programs for the uninsured usually serve people with
household incomes below about
twice the federal poverty level.
They may therefore be hard
pressed to adapt their missions
to the new uninsured population in ways that will maintain
their fragile support from funders
and volunteers. Since safety-net
systems are already on life support,1 any major shock may
threaten their very existence.
Therefore, access programs must
consider carefully how best to
refocus and justify their function and mission.
First, health care reforms
chickens should not be counted
until theyve hatched. During
the 3 years before full implementation begins, constitutional
challenges and conservative politicians threaten to upend the
ACA.2 Safety-net programs must
remain intact at least until reform takes effect and just in
case it never does. Second, even
after reform, the newly insured
will face barriers to access arising from provider shortages,
transportation difficulties, and
language differences all of
which safety-net organizations
can help to overcome.

Third, the future uninsured

population will probably deserve
more safety-net support than one
might imagine. Some people will
be uninsured temporarily when
their economic circumstances
change. New workers may earn
enough to lose their subsidy for
individual insurance but remain
ineligible for group insurance during the 3-month probationary
period that employers may impose. People without good jobs
whose income increases just
enough to nudge them over
138% of the poverty level will be
disqualified from Medicaid and
be required to purchase subsidized private insurance. It may
be difficult to make this publicto-private transition smoothly.
Medicaid enrollment can start
instantaneously, sometimes even
retroactively, but private coverage
typically begins on the first day
of the month after all forms
have been completed and the
initial check has cleared.
If the experience in Massachusetts is any guide, these wrinkles will probably cause shortterm coverage gaps for many
people (see table). Coverage discontinuity will also occur within
households, when different family members qualify for coverage
from different sources, depending on their citizenship and employment status. Safety-net programs can therefore serve a
critical function in maintaining

n engl j med 364;1 nejm.org january 6, 2011

The New England Journal of Medicine

Downloaded from nejm.org on May 27, 2013. For personal use only. No other uses without permission.
Copyright 2011 Massachusetts Medical Society. All rights reserved.