Uptake of prevention of mother-to-child-transmission
using Option B+ in northern rural Malawi:
a retrospective cohort study
Alison J Price,1Michael Kayange,2Basia Zaba,1Frank M Chimbwandira,2
Andreas Jahn,2Zengani Chirwa,2Aisha NZ Dasgupta,1Cynthia Katundu,3
Jacqueline L Saul,1Judith R Glynn,1Olivier Koole,1Amelia C Crampin1
1London School of Hygiene
and Tropical Medicine, London,
2Malawi Ministry of Health,
3Karonga Prevention Study,
Dr Alison J Price,
London School of Hygiene and
Tropical Medicine, Keppel
Street, London WC1E 7HT, UK;
Received 26 August 2013
Revised 21 February 2014
Accepted 17 March 2014
Published Online First
8 April 2014
To cite: Price AJ,
Kayange M, Zaba B, et al.
Sex Transm Infect
To identify points of dropout on the pathway from
offering HIV testing to maintenance on antiretroviral
therapy (ART), following the introduction of the Option
B+ policy for pregnant women in Malawi (lifelong ART
for HIV-positive mothers and 6 weeks nevirapine for the
infants), a retrospective cohort study within a
demographic surveillance system in northern Malawi.
Women living in the demographic surveillance system
who initiated antenatal care (ANC) between July 2011
(date of policy change) and January 2013, were eligible
for inclusion. Women who consented were interviewed
at home about their health facility attendance and care
since pregnancy, including antenatal clinic (ANC) visits,
delivery and postpartum care. Women’s reports, patient-
held health records and clinic health records were
manually linked to ascertain service use. Among 395
women, 86% had tested for HIV before the pregnancy,
90% tested or re-tested at the ANC visit, and <1% had
never tested. Among 53 mothers known to be HIV-
positive before attending ANC, 15 (28%) were already
on ART prior to pregnancy. Ten women tested HIV-
positive for the first time during pregnancy. Of the 47
HIV-positive mothers not already on ART, 26/47 (55%)
started treatment during pregnancy. All but five women
who started ART were still on treatment at the time of
study interview. HIV testing was almost universal and
most women who initiated ART were retained in care.
However, nearly half of eligible pregnant women not on
ART at the start of ANC had not taken up the invitation
to initiate (lifelong) ART by the time of delivery, leaving
their infants potentially HIV-exposed.
Combination antiretroviral therapy (ART) is highly
effective for prevention of mother-to-child transmis-
sion (PMTCT) of HIV , yet globally an estimated
390 000 children were infected during pregnancy,
delivery or breastfeeding in 2011, 90% of whom
were in sub-Saharan Africa.1In July 2011, the
Malawi Ministry of Health implemented ‘Option B+’
nationally, a policy to initiate ART for life, regardless
of CD4 count, for all HIV-positive pregnant or breast-
feeding women, with decentralised provider-initiated
HIV testing and counselling, and single tablet
regimen, triple therapy ART (in the form of
tenofovir-lamivudine-efavirenz), delivered at antenatal
clinics (ANC). The baby should receive 6 weeks of
nevirapine, PCR-based HIV screening at 6 weeks,
additional rapid testing at 12 months and 24 months
and ART if HIV-positive. Option B+ replaced a
modified ‘Option A’ regimen of daily zidovudine
during pregnancy and combination lamivudine-
nevirapine at delivery for the mother (or in case of
home delivery, single dose nevirapine to take home)
and 1–4 weeks of daily zidovudine syrup for the
This public health approach to PMTCT aims to
provide a simplified and standardised regimen2in a
country with limited access to CD4 counts,3high
fertility (total fertility rate=5.7), and short birth
intervals (median=3 years) and where breastfeeding
is currently the only feasible option to prevent
infant malnutrition.4However, in an overburdened
health system relying predominantly on patient-held
records (health passport, ART identity card) and
clinic registers, and where women are known to
change facilities for ANC services and delivery,
sharing information between facilities and services
to ensure that women and infants who need
PMTCTare identified, initiated and retained in care,
is a challenge.
uptake in a rural population in the first 18 months
(July 2011–January 2013) following policy change,
women who had given birth and cross-referencing
with clinic registers, to established potential points
in the system of care where HIV-infected women
may fail to be identified or are lost to care services,
mother’s health and potential mother-to-child HIV
The Karonga Prevention Study demographic surveil-
lance system (DSS) in northern Malawi covers an
area of 135 km2and a population of 35 000, with
linkage to local clinical services.5Four annual HIV
serosurveys (2007–2011), with 70% participation
and 98% requesting their results, show HIV preva-
lence is 9% in adult women.6Trained local key
informants (n=280) provide notification of all
births (including stillbirths) in the DSS on a monthly
basis, and the date and location of each birth and
the parent’s details are recorded at a home visit
from a study interviewer. Each key informant covers
a ‘cluster’ of about 25 households. Any births
missed by an informant are captured during the
re-census. Five health facilities provide antenatal
Scan to access more
Price AJ, et al. Sex Transm Infect 2014;90:309–314. doi:10.1136/sextrans-2013-051336309
services with integrated ARTcare and four provide routine deliv-
eries (obstetric complications during labour, delivery and the
postnatal period are referred to the district hospital, 70 km north
of the DSS). Pregnant women are recommended to attend at least
four ANC visits, starting before week 17 of pregnancy, and to
deliver in a health facility, using a skilled attendant.4
A retrospective cohort study was conducted among women
resident in the DSS who gave birth (either a live or stillbirth)
after the PMTCT policy changed (1 July 2011). All HIV-positive
and HIV-unknown women were included. A random sample of
HIV-negative women were also included to ascertain whether
they were offered HIV counselling and testing (HCT), to avoid
inadvertent disclosure of HIV status, and to identify any serocon-
verters. Mothers who had moved within the DSS area since the
registration of a birth were sought at the new location.
HCT-trained study interviewers, who did not know the women’s
HIV status, visited women at home to ask about HIV testing and
PMTCTservice uptake along the continuum of care (ANC, deliv-
ery and postpartum services). Written consent was sought for the
interview, for HIV testing if indicated, and for subsequent review
of health records at the clinic. Rapid testing was offered to all
women unless they reported that they were HIV-positive. All par-
ticipants were precounselled and those who consented to rapid
HIV testing were given post-test counselling. All HIV-positive
women not already on ART who were still breastfeeding were
interviewed about their reasons for not starting treatment (if pre-
viously tested positive) and informed about ART services in the
district and given a referral letter.
All women who were HIV-positive (by report or rapid test),
or HIV-unknown (no test or last negative test on database prior
to 2011), were interviewed about breastfeeding and use of HIV
care services. HIV-positive and HIV-unknown mothers who did
not have a PCR test recorded in the child’s health passport were
offered PCR HIV testing of their infant(s), with precounselling.
Mothers of babies potentially at risk of mother to child trans-
mission (MTCT) (unknown HIV status, or HIV-positive and no
ART) were advised to seek care at a government facility.
The patient-held record and self-report of health centre
attendance and service use was checked against clinic registers.
If records were not identified at the facility reported, they were
sought at all other clinics in the DSS. Identification used the
name, age, facility and date of attendance and where available,
HCT, ART and delivery identification numbers (recorded in the
patient and clinic records by health staff). Authorisation to
search the registers was obtained from the district health office
and conducted at the convenience of each facility. Ethics
approval was obtained from the Malawi National Health
Sciences Research Committee and the ethics committee of the
London School of Hygiene and Tropical Medicine.
Data were analysed using Stata V .12 software (Stata Corp,
College Station, Texas, USA). Data from the HIV serosurveys,
self-report, patient-held and clinic records were combined to
assess services received, and points of dropout in PMTCTcare.
Among women resident in the DSS who gave birth after 1 July
2011 and before 31 January 2013, 568 were selected to partici-
pate. Of these, 35 (6.2%) had left the DSS, 45 (7.9%) could not
be found and one (0.2%) had died. Of the 487 women found,
483 (99.2%) consented to interview and review of clinic
records. Eighty-eight women had first attended ANC before the
policy change so were excluded, leaving 395 women who
initiated ANC or did not attend ANC during their pregnancy
but delivered after policy change: 53 with a prior HIV-positive
test result and 112 with a recent HIV-negative test result on the
study database and 230 whose HIV status could not be ascer-
tained from study records (either never tested during the sero-
survey or last negative test prior to 2011).
Characteristics of these 395 women are shown in table 1.
The mean age was 25 years and 61% had not completed
primary education. Patient-held records were available for
80% (316/395); ANC register entries were found for 88% of
those who reported at least one ANC attendance in the DSS
(338/386; two never attended and seven attended outside the
DSS); and delivery register entries were found for 80% of
those who reported a hospital birth in the DSS (261/326; 37
delivered at a facility outside the DSS and 32 delivered at
home or with a traditional birth attendant). Most women com-
menced ANC during the second trimester (61%) and attended
at least two ANC visits (94%), with some facility switching
(9%). Only 47 women (12%) were accompanied to ANC by
their husband, and four by other relatives. Almost all women
(92%) reported that they delivered at a government health
facility (inside or outside the DSS) and 85% used a maternity
facility at the same location as their ANC (310/363; data not
shown). A third (36%) of women lived >5 km from the
nearest health facility.
In total, 86% reported or were documented to have ever
tested for HIV before their most recent pregnancy, 90% tested
or retested at ANC or HCT clinic (on the day of an ANC visit),
and <1% never tested (data not shown). Of newly diagnosed
HIV-positive mothers 80% had previously tested and of women
who did not retest during ANC, 9% reported that testing and
counselling was not offered by government health staff.
ART uptake is summarised in figure 1. Of 53 mothers identi-
fied via serosurvey or self-report as HIV-positive prior to ANC,
16 (30%) were already on ART before pregnancy. Ten add-
itional women tested HIV-positive for the first time during preg-
nancy. Among HIV-positive mothers not already on ART, 26/47
(55%) started treatment during pregnancy; almost half (21/47:
45%; table 1) reported they had been referred to a separate
ART clinic and eight of these referrals occurred 6 months or
more after policy implementation. Eight women known to be
HIV-positive status during ANC by refusing to test and by
reporting a negative status during the study interview. However
one of these women was found to have started ART during
pregnancy. Of 21 HIV-positive women who did not start ART,
14 revealed their HIV-positive status during interview and were
asked about ART referral: three reported being offered ART in
ANC, nine reported referral to an ART clinic (of whom only
two attended and one was assessed but refused treatment; data
not shown), and two claimed they had not received advice
regarding ART. All of these women reported that ART was not
offered at delivery, although four were referred, but they all
refused treatment. For 15 of these 21 women there was no
record of their HIV-positive status in the ANC register. Reasons
reported for not starting ART included fear of disclosure (to
husband (8) to others (2)), no trust in ART (10), fear of side
effects (6), transport costs (1), preference for herbal treatment
(4), inadequate referral from health staff (4) and the belief that
ART was not yet needed, due to good health (15). Women
could report more than one reason.
Of 63 HIV-positive women, 42/63 (67%) received ARTwhile
pregnant and during delivery and among the remaining 21, two
received nevirapine during labour and one started ART 6 weeks
after delivery. All but one of the at-risk babies were breastfed; at
least 20% of these babies did not receive nevirapine syrup and
310Price AJ, et al. Sex Transm Infect 2014;90:309–314. doi:10.1136/sextrans-2013-051336
Characteristics of 395 women in prevention of mother-to-child-transmission (PMTCT) study by HIV and ART status*
on ART prior
on ART during
women§ HIV-unknown¶TotalPer cent
Number of ANC visits
Age at delivery (years)
Highest attained education level
First presented at ANC**
Attended ANC with guardian at least once††
No guardian (all visits)
Delivered at a health centre††
Offered ART in ANC‡‡
Referred to ART clinic
Not offered or referred
Distance to health centre¶¶ (km)
1626 21 31319395100
24.3 11 15
Price AJ, et al. Sex Transm Infect 2014;90:309–314. doi:10.1136/sextrans-2013-051336
on ART prior
on ART during
women§HIV-unknown¶ TotalPer cent
Visited Health Centre for PCR***
Infant age 6–8 weeks
Infant age 9+ weeks
Six weeks oral Nevirapine for child***
24 16 NA
*HIV status determined by information held in the demographic surveillance site (DSS) HIV database (annual HIV serosurveys 2007–2011), clinic records and a woman’s self-report.
†One woman started antiretroviral treatment (ART) while pregnant but before antenatal clinic (ANC).
‡Does not include two women who received a single dose of nevirapine at delivery and one who started ART 6 weeks after delivery.
§Random sample of HIV-negative women in DSS who attended ANC after 1 July 2011.
¶HIV-unknown if no DSS test or clinic record and woman reported no HIV counselling and testing (HCT) during ANC or delivery. Includes two women who did not attend ANC or deliver at a health facility.
**Excludes two women who did not attend ANC.
††Delivered at a health facility inside or outside the DSS.
‡‡Eight women with a positive test on the DSS database concealed their HIV-positive status during ANC and the study interview. Six reported a negative test result in ANC and two reported HCT refusal.
§§An ART start date was found for one of these women.
¶¶Calculated as distance (km) to nearest health centre. HCT, ANC, maternity and ART services available at each centre.
***Information on PCR testing and baby nevirapine was not available for six women who reported a negative status during the study.
Price AJ, et al. Sex Transm Infect 2014;90:309–314. doi:10.1136/sextrans-2013-051336
at least 80% of mothers did not return for a PCR test within
6 weeks of delivery (table 1).
Overall, 88% of women (38/43) who had started ART,
including all of those already on ART prior to pregnancy, were
identified from self-report and clinic records as on treatment at
the time of interview (mean 4.3 months (SD 2.3) postdelivery).
Of those women who started ART during ANC, 81% (21/26)
were retained in care at 6 months.
While these findings should be interpreted with caution due to
the small numbers, in the first 18 months of policy change HIV
testing was almost universal and all women who had initiated
ART prior to pregnancy, and most who had started during ANC,
were retained in care. At first ANC attendance 30% of
HIV-positive pregnant women were already on ARTand this pro-
portion is likely to increase as Option B+ is rolled out. All the
children of those women received appropriate treatment at deliv-
ery. However 43% of HIV-positive pregnant women not already
on ART did not start ART during pregnancy or delivery, and at
least 20% of HIVexposed babies did not receive nevirapine.
The high level of HIV testing uptake in ANC is similar to
that observed in a recent meta-analysis of opt-out testing strat-
egies in sub-Saharan Africa.7In our population failure to test
was largely due to refusal rather than programmatic issues such
as staff shortage or unavailability of test kits, as observed else-
where.8In other populations enhanced HIV testing uptake has
been observed among women who attend ANC with their part-
ners.9The effect of male participation on PMTCT uptake in
sub-Saharan Africa is not well understood and may vary.10 11In
our study, HIV testing uptake was high, and the number who
attended with a husband was too small to assess any differences.
The majority of women, including those who did not start
ART, attended ANC on three or more occasions, consistent with
national surveillance data,4thereby providing several opportun-
ities for ART initiation. However, failure to record positive HIV
results in the ANC register—as observed for 71% of those who
did not start ART—may contribute to missed opportunities for
counselling and initiation of PMTCTat subsequent ANC visits.
While Option B+ aims to deliver ART within ANC services,
45% of eligible women reported that they were referred to sep-
arate ART services. Almost half of these referrals occurred more
than 6 months after policy change, which suggests challenges
associated with provision of the integrated service in the initial
stages of the policy. Higher levels of attrition have been
observed elsewhere when ART services are provided in separate
locations or on different days from ANC, particularly among
those living far from the clinic.12–15Higher early attrition has
also been observed when women initiate ART to prevent
MTCT compared with those who initiate for their own health
the recommended testing and treatment.1Two women received nevirapine at delivery (from residual clinic stock).2Mean duration of antiretroviral
treatment post-delivery was 4.3 months. ANC, Antenatal clinic; ART, antiretroviral therapy; PMTCT, prevention of mother-to-child-transmission.
HIV testing and ART uptake and usage among women in antenatal clinic. Boxes with a thick black border highlight the numbers not following
Price AJ, et al. Sex Transm Infect 2014;90:309–314. doi:10.1136/sextrans-2013-051336313
and among those who initiate at the time of a HIV-positive diag- Download full-text
nosis,16which suggests greater need for education (maternal
health benefits) and support services to address issues of stigma
In our rural setting nearly 80% of women who initiated ART
during the ANC period were retained on treatment 6 months
after starting. Lower retention (70%) has been observed in
urban and larger health facilities in Malawi.16Retention on
ART depends on good referral systems and linkage between
facilities. While dropout was low within the time frame of this
study, potential barriers to continuation in care may include
transportation costs, distance to clinic and human resource con-
straints resulting in excessive waiting times and scheduling
Strengths of the study include regular reporting of births by
village informants19and continuous demographic surveillance
that provide comprehensive birth data for the population. The
recent HIV serosurveys (2007–2011) with high levels of female
participation and linkage to ART clinic data means that it was
possible to identify, with minimal misclassification, almost all
HIV-positive and HIV-unknown women in this population who
initiated ANC since policy change, and who were known to be or
Limitations of the study include the small number of HIV-positive
women and lack of data on HIV-positive women who experi-
enced an early termination (<7 months) of a pregnancy.
Option B+ is considered a cost-effective strategy for ensuring
universal access to ART for PMTCT in Malawi20and has led to
substantial increases in women initiating ART during pregnancy.3
However, our data suggest that more support is required to facili-
tate integration of PMTCT services in ANC and to address the
reluctance of some healthy women to test or reveal their
HIV-positive status in ANC and/or to initiate ART during preg-
nancy. Studies are needed to explore these barriers and to evaluate
the impact of different healthcare delivery strategies, including
patient education and support services for women and their fam-
ilies, on uptake and retention. If refusal remains high, alternative
interventions will also need to be offered to prevent MTCT.
▸ Implementation of the Option B+ policy may be
compromised by reluctance of some women to reveal their
HIV-positive status in ANC and/or to initiate antiretroviral
therapy during pregnancy.
▸ More support is required to facilitate integration of
prevention of mother-to-child-transmission services in ANC.
▸ Barriers to uptake of Option B+ need to be explored and
addressed with targeted health education and support
▸ If refusals remain high, alternative interventions may need to
Correction notice Last name of the fourth author has been corrected since
published Online First.
Handling editor Jackie A Cassell
Acknowledgements The authors thank Lupakisyo Chomo and Maureen Thindwa
for the data collection. The authors also thank the women who consented to
participate in the study.
Contributors BZ conceived the study. BZ, JRG, ANZD, AJP and ACC wrote the
study protocol and contributed to the study design. AJP supervised the data
collection. JLS programmed the data entry. AJP cleaned and analysed the data and
drafted the manuscript. All listed authors read, revised and approved the final
Funding This work was supported by a Wellcome Trust Award (096249/Z/11/A).
Competing interests None.
Patient consent Obtained.
Ethics approval Obtained from the Malawi National Health Sciences Research
Committee and the ethics committee of the London School of Hygiene and Tropical
Provenance and peer review Not commissioned; externally peer reviewed.
Open Access This is an Open Access article distributed in accordance with the
terms of the Creative Commons Attribution (CC BY 3.0) license, which permits
others to distribute, remix, adapt and build upon this work, for commercial use,
provided the original work is properly cited. See: http://creativecommons.org/licenses/
1UNAIDS. Together we will end AIDS. 2012. Joint United Nations Programme on
HIV/AIDS (UNAIDS). 2012.
2Malawi Ministry of Health. Clinical Management of HIV in Children and Adults.
3 Chimbwandira F, Mhango E, Makombe S, et al. Impact of an innovative approach
to prevent mother-to-child transmission of HIV—Malawi, July 2011-September
2012. Morb Mortal Wkly Rep 2013;62:148–51.
4 National Statistical Office (NSO) and ICF Macro editor. Malawi Demographic and
Health Survey. 2010th edn. Zomba, Malawi, and Calverton, Maryland, USA: NSO
and ICF Macro, 2011.
5Crampin AC, Dube A, Mboma S, et al. Profile: the Karonga health and
demographic surveillance system. Int J Epidemiol 2012;41:676–85.
6Floyd S, Molesworth A, Dube A, et al. Underestimation of HIV prevalence in surveys
when some people already know their status, and ways to reduce the bias. AIDS
7Wettstein C, Mugglin C, Egger M, et al. Missed opportunities to prevent
mother-to-child-transmission: Systematic review and meta-analysis. AIDS
8 Doherty T, Chopra M, Nsibande D, et al. Improving the coverage of the PMTCT
programme through a participatory quality improvement intervention in South
Africa. BMC Public Health 2009;9:406.
9 Peltzer K, Mosala T, Dana P, et al. Follow-up survey of women who have
undergone a prevention of mother-to-child transmission program in a resource-poor
setting in South Africa. J Assoc Nurses AIDS Care 2008;19:450–60.
10Brusamento S, Ghanotakis E, Tudor Car L, et al. Male involvement for increasing
the effectiveness of prevention of mother-to-child HIV transmission (PMTCT)
programmes. Cochrane Database Syst Rev 2012;10:CD009468.
11Auvinen J, Kylmä J, Suominen T. Male involvement and prevention of
mother-to-child transmission of HIV in Sub-Saharan Africa: an integrative review.
Curr HIV Res 2013;11:169–77.
12Stinson K, Jennings K, Myer L. Integration of antiretroviral therapy services into
antenatal care increases treatment initiation during pregnancy: a cohort study. PLoS
13Mandala J, Torpey K, Kasonde P, et al. Prevention of mother-to-child transmission
of HIV in Zambia: implementing efficacious ARV regimens in primary health centers.
BMC Public Health 2009;9:314.
14Hussain A, Moodley D, Naidoo S, et al. Pregnant women’s access to PMTCT and
ART services in South Africa and implications for universal antiretroviral treatment.
PLoS One 2011;6:e27907.
15Chinkonde JR, Sundby J, Martinson F. The prevention of mother-to-child HIV
transmission programme in Lilongwe, Malawi: why do so many women drop out.
Reprod Health Matters 2009;17:143–51.
16Tenthani L, Haas AD, Tweya H, et al. Retention in care under universal antiretroviral
therapy for HIV-infected pregnant and breastfeeding women (‘Option B+’) in
Malawi. AIDS 2014;28:589–98.
17Chi BH, Adler MR, Bolu O, et al. Progress, challenges, and new opportunities for
the prevention of mother-to-child transmission of HIV under the US president’s
emergency plan for AIDS relief. J Acquired Immune Defic Syndr 2012;60(Suppl. 3):
18Govindasamy D, Ford N, Kranzer K. Risk factors, barriers and facilitators for linkage
to antiretroviral therapy care: a systematic review. AIDS 2012;26:2059–67.
19Jahn A, Crampin AC, Glynn JR, et al. Evaluation of a village-informant driven
demographic surveillance system in Karonga, Northern Malawi. Demogr Res
20Fasawe O, Avila C, Shaffer N, et al. Cost-effectiveness analysis of option B+ for HIV
prevention and treatment of mothers and children in Malawi. PLoS One 2013;8:
314Price AJ, et al. Sex Transm Infect 2014;90:309–314. doi:10.1136/sextrans-2013-051336