Reactions to Voluntary Counseling

and Testing in Rural Malawi

Rebecca Thornton
Agatha Bula
Kondwani Chavula
Simona Bignami-Van Assche
Susan Cotts Watkins

PWP-CCPR-2008-002

January 2008

California Center for Population Research

On-Line Working Paper Series
Reactions to Voluntary Counseling and Testing in rural Malawi

Rebecca Thornton, Agatha Bula, Kondwani Chavula, Simona Bignami-Van Assche, Susan Cotts
Watkins

Kennedy School of Government, Harvard University, Cambridge, MA 02138

Rebecca Thornton
PhD Candidate

University of North Carolina Project, Lilongwe, Malawi

Agatha Bula
Senior Research Nurse

Malawi Diffusion and Ideational Change Project, Zomba, Malawi

Kondwani Chavula
VCT Project Manager

Harvard Initiative for Global Health, Harvard University, Cambridge, MA 02138

Simona Bignami-Van Assche
Research Associate

Population Studies Center, University of Pennsylvania, Philadelphia, PA 19104

Susan Cotts Watkins
Full Professor

Correspondence to: S Watkins, [email protected]

Word count: 2,415 words

Abstract

As part of a study that tested individuals for sexually transmitted infections in rural Malawi, we
examine the reactions to the provision of voluntary counseling and testing (VCT). This paper
presents response rates as well as summarizing qualitative data on community comments. Our
primary substantive conclusion is that despite mixed community opinions about the value of
VCT, there was unexpectedly high participation in VCT. This has important implications for
current methods of assessing the acceptability of VCT based on research that poses hypothetical
questions rather than measuring actual behavior.

2
Introduction

In Malawi, as elsewhere in high HIV-prevalence countries of sub-Saharan Africa, residents of

rural areas have little opportunity to be tested for HIV and to learn their results. Voluntary
counseling and testing (VCT) for HIV has been available in freestanding facilities in Malawi for
about five years, but most VCT centers are located in major urban areas. In conjunction with the
expected availability of antiretrovirals through programs such as the Global Fund, the Malawi
Ministry of Health and Population plans the rapid expansion of VCT facilities as soon as
technically feasible. To permit evaluating the impact of these programs, it is crucial to
distinguish between what people say about the acceptability of VCT and what they do when it is
provided.

There have been relatively few studies on reactions to the availability of VCT, especially in the
rural areas of highly AIDS-affected areas of sub-Saharan Africa. Several studies examine the
determinants of HIV testing and returning for results, the majority of which either take place in
urban testing centers or ask individuals to discuss hypothetically why they would or would not
get tested. The findings are mixed and often indicate respondents’ fears of testing and worries
about learning they are infected [1-3].

Data from the nationally-representative 2000 Malawi Demographic and Health Survey (MDHS)
show that 8 percent of people in rural areas report having been tested for HIV, compared to 18
percent of people in urban areas [4]. In rural areas there seems to be a higher demand for testing
than in urban areas (74 percent of people in rural areas who had never been tested would have
liked to, versus 66 percent of people in urban areas), and a lower knowledge of the availability of
testing facilities (66 percent of respondents in rural areas said that they knew a source for a HIV
test, compared to 88 percent in urban areas). Another study that also asked hypothetical
questions came to more pessimistic conclusions. Based on survey questions and semi-structured
is more pessimistic. In 2003, approximately 200 rural and urban Malawians were asked about the
acceptability of HIV testing and counseling and whether they would want to learn their results.
[5]. This study found that respondents were knowledgeable about HIV/AIDS and most were
aware of the existence of testing facilities. They concluded, however, that testing was not yet
acceptable to the general public: most respondents said that they might be tested at some point in
their lives, but they were not yet ready.

We examine the reactions to VCT in the rural communities of our research project. Despite
mixed verbal feedback, we find an overall acceptance of VCT. This has important implications
on the conclusions drawn from research that asks for opinions rather than measuring actual
behavior.

Methods

This study occurred in the context of a larger project, the Malawi Diffusion and Ideational
Change Project (MDICP). A core activity of the MDICP is an ongoing panel survey. The project
was approved by ethical review boards at the University of Pennsylvania and the University of
Malawi. The main sample consists of ever-married women and their current husbands. In 2004,
3
this sample was augmented by a randomly chosen sample of adolescents (married and
unmarried) aged 15-24 years. The MDICP is conducted in the rural areas of three districts, each
taken to represent one of the three regions of the country, North, Center and South. However, a
comparison of the characteristics of the MDICP sample with those of the rural population
surveyed in the MDHS indicates that the MDICP sample is representative of the national rural
population (more details on sampling and fieldwork procedures, as well as the survey data, are
available for free download from the project’s website: http://malawi.pop.upenn.edu).

Between April and August 2004, MDICP respondents were offered the opportunity to be tested
for HIV and other sexually transmitted infections (STIs) [6]. A total of 2,816 respondents were
tested for HIV or at least one STI. The results show that HIV prevalence is about 7 percent for
the entire sample (adults and adolescents of either sex in all three sites), varying from 4.5 percent
in the Northern region to 8.2 percent in the Southern region (Table 1).

-----------------------
Table 1 about here
-----------------------

Consent was asked separately for the HIV and STI tests; for never-married adolescents, consent
from a parent or guardian was also asked. Respondents who agreed to be tested were counseled
by a trained nurse. The nurse collected HIV samples using oral swabs (ORASURE).1 For STIs,
men were asked to provide urine samples (for gonorrhea and Chlamydia), while women were
asked to provide vaginal swabs (for Chlamydia, gonorrhea and trichomonas). The specimens
were analyzed at the University of North Carolina laboratory in the capital, Lilongwe. Testing
was not anonymous in order to permit linking the HIV and STI results with the survey responses.
However, to preserve the respondent’s confidentiality, the specimens were given a special
identification number, such that only the MDICP Biomarker Coordinator could link the test and
survey results. This identification number was recorded on a Polaroid picture of the respondent
taken by the nurse, and given to the respondent as identification required to be presented in order
to receive results. When results were available, typically about five to seven weeks after testing
occurred, nurses trained in VCT counseling returned to the villages in order to communicate the
results to those who wished them.

Results were made available throughout each study area in small portable tents, which provided
privacy. There were several VCT (tent) sites in each surveyed area, so that all respondents’
homes were within five kilometers from at least one site. A few days before the results were to
be available, the VCT team (consisting of local Form-4 graduates) visited all respondents in their
homes to communicate the location of the tent and the specific week the team would be in their
village. Precautions were taken to ensure that respondents received their correct results, and that

1
The accuracy of saliva for detection of HIV antibodies has been shown to be comparable to serum-based tests [7-
12]. For epidemiological purposes in particular, use of saliva seems to offer several advantages over serum. An
important motivation for using saliva in population-based surveys is the assumption that a non-invasive method
might contribute significantly in reducing selection bias due to non-consent. Studies that have used saliva for
detection of HIV antibodies have generally achieved higher consent rates, but data are still lacking to make a sound
evaluation of the ways in which saliva and serum compete with regard to acceptability [13].
4
confidentiality and privacy were maintained. Nurses at each VCT site carefully checked each
person’s Polaroid picture; no results were provided to respondents who could not present their
identification. Post-test counseling and results were then provided. Respondents who tested
positive for a STI were given the appropriate treatment at no cost; respondents who tested
positive for HIV were encouraged to go to the nearest district hospital for a second test and for
more information about availability of treatment with antiretrovirals. All respondents also
received free multi-vitamins after receiving results. The counseling procedure lasted
approximately 45 minutes for each respondent.

To assess the reactions to VCT, we adapted ethnography to our purposes, since our previous
experience suggested that survey respondents are reluctant to criticize our project. We relied on
local ethnographers, identified as members of the MDICP field team by their project T-shirts,
who kept journals in which they wrote what they heard people say about VCT. In one site, the
ethnographers directly asked people for their opinions. In a second site, the ethnographers simply
listened to what was being said publicly, for example a conversation overheard at a local market.
In both sites, however, names were not recorded, nor did the ethnographers know whether the
participants in a conversation were members of the MDIPC sample, and thus whether they had
been tested for HIV or other STIs.

Results

The majority of the 2004 MDICP respondents who were present for the survey agreed to be
tested for HIV and/or STIs (Table 2). The refusal rates are overall quite low, and similar (or even
lower) than those recorded by the 2004 Kenya DHS (11.9 for women and 11.2 for men).

-----------------------
Table 2 about here
-----------------------

The 2001 MDICP survey showed that respondents vastly overestimate the transmission
probabilities of HIV: approximately two-thirds believed that unprotected sex with an infected
person was certain to lead to one’s own infection, with virtually all of the others saying that
infection was very likely. This suggests that many respondents expected to find that they were
HIV+, and were probably very worried about receiving their results. Nonetheless, the majority of
the respondents who were tested returned (Table 3).

-----------------------
Table 3 about here
-----------------------

Comments in the ethnographic journals about the acceptability of VCT were quite mixed, and
varied by the circumstances in which the comments were heard. When respondents made
positive comments, they often used the concepts of VCT promotion in Malawi and that the
MDICP nurses had used in counseling. For example, people said that it is good to know one’s
body status so that one can plan for the future; that those who are told they are HIV positive can
5
follow the instructions that will lead them to a longer life; and that those who are negative can
change their behavior to avoid subsequent infection. In contrast, when an ethnographer
happened to be passing by someone talking with others about the VCT, the comments were less
formulaic and appeared to be more genuine. The comment below was overheard by an
ethnographer when she was in a local trading center:

“As I was waiting at Ulongwe Trading Center for the minibus to come and pick me up, I decided to have a
snack....The people who were around playing bawo at the veranda of this shop started talking about the
MDICP minibus. One of the men said >People have started getting their results=. Another asked, 'Do they
come to your house and give you the results?' Another said >You go to their tents=. Then one said that >This
is better because when you have the results that you are HIV positive, you can just go to a nearby place and
have beer and reduce the worry.= The other said >Haa, HIV is HIV. Even if you drink, once you get it that=s
all=. Then one man said, >The young guys who have their results are happy.=”

The negative comments were primarily related to the fact that testing and counseling were
unwanted, unnecessary, or possibly malicious (e.g. Satanism, the project was pumping blood and
selling it for the project’s benefit). A second criticism was that the project was discriminatory.
Only some villages were included in the sample, and within the sampled villages only some
households were included in the survey, yet other respondents in the same household or village
also wanted to know their HIV status. (This comment can therefore be interpreted positively, as
it indicates that individuals were eager to be tested and to know their results). A third, and
particularly interesting, critique raised the issue of inequalities in access to treatment. Although
respondents who were HIV-positive were counseled to eat a nutritious diet and to go to the
district hospital for further testing and treatment, there were complaints that people were too
poor either to afford nutritious foods (“the good food which needs much money”, in the word of
one respondent) or to pay for transport to the district hospital. This view was sometimes
associated with the complaint that the MDICP was not helping people in the villages because it
was not providing antiretrovirals to those infected with HIV. An ethnographer wrote that a man
“was then telling me that we [the MDICP] are doing nothing in the villages. We are doing the
VCT but we are not helping them. Had it been that we have medicine to give them if they have
HIV it would have been helpful.” 2

It is of methodological interest that the comments varied according to the circumstances in which
they were heard. If the ethnographer asked a direct question, the responses typically presented
only one extreme view: either that the MDICP is doing good and that it should come more often,
or that it is bad or useless and thus it should stop bothering the villagers.3 When the
ethnographers were simply listening, and especially on the few occasions when they were not
recognized as part of the MDICP team (for example, if they were not wearing the MDICP t-shirt
because they had washed it), villagers were overheard debating the VCT with their friends and
neighbors, making arguments for and against it.
2
Our research design does not allow us to study whether offering ARV treatment increases return rates. See [14] for
a review.
3
We think the negative comments may appear disproportionately in the ethnographic journals for several reasons.
Perhaps they were found them more worthy of recording; or perhaps those with objections voiced them openly and
strongly, whereas respondents who wanted to be tested and to know their results might have kept quiet. Lastly, it is
worth noting that many negative comments were overheard by those who were not part of the tested sample.

6
Discussion and conclusions

Our most important methodological conclusion is that actually offering VCT rather than posing
hypothetical questions is likely to be the only valid way to assess its acceptability. We also
conclude that when researchers attempt to assess reactions to an innovation, particularly one
promoted by the government or other outsiders (such as VCT in rural areas in Malawi), questions
posed directly by a member of a research project are likely to produce favorable responses. The
type of ethnography we used to assess reactions probably more accurately reflects real
ambivalence in rural communities in Malawi.

Our overall substantive conclusion is that while there is considerable fear of testing and anxiety
about learning one’s HIV status, these can be overcome, at least in studies similar to ours.
When respondents were offered testing in their homes and results were given within their
communities, most respondents agreed to be tested and of those, the majority voluntarily came
for their results. Because survey respondents overestimated the transmission probabilities of
HIV, many were probably delighted to learn that they were HIV negative, and thus may be
emboldened to seek re-testing in the future.

7
References

1 Cartoux M, Meda N, Van de Perre P, Newell ML, de Vincenzi I, Dabis F. Acceptability of

Voluntary HIV Testing by Pregnant Women in Developing Countries: An International
Survey. AIDS 1998; 12: 2489-93.
2 Duffy TA, Wolfe CD, Varden C, Kennedy J, Chrystie IL. Women's Knowledge and
Attitudes, and the Acceptability of Voluntary Antenatal HIV Testing. Br J Obstet Gynaecol
1998; 105: 849-54.
3 Nuwaha F, Kabatesi D, Muganwa M, Whalen CC. Factors Influencing Acceptability of
Voluntary Counselling and Testing for HIV in Bushenyi District of Uganda. East Afr Med J
2002; 79: 626-32.
4 National Statistical Office [Malawi], ORC Macro. Malawi Demographic and Health Survey
2000. Zomba, Malawi and Calverton, Maryland, USA: National Statistical Office and ORC
Macro, 2001.
5 Yoder PS, Matinga P. Voluntary Counselling and Testing (VCT) for HIV in Malawi: Public
Perspectives and Recent VCT Experiences. Calverton, Maryland, USA: ORC Macro, 2004.
6 Bignami-Van Assche S, et al. Protocol for biomarker testing in the 2004 Malawi Diffusion
and Ideational Change Project. SNP Working Paper No.6, University of Pennsylvania,
Philadelphia, 2004.
7 Frangois-Girard C, Thortensson R, Luton P, et al. Multi-center European evaluation of HIV
testing on serum and saliva samples. Transfus Clin Biol 1996; 3:89-98.
8 Tamashiro H, Constantine NT. Serological diagnosis of HIV infection using oral fluid
samples. Bull World Health Organ 1994; 72:135-143.
9 Bruckova M, Stankova M. Detection of HIV virus antibodies in saliva. Epidemiol Mikrobiol
Imunol 1995; 44:127-129.
10 Vall-Mayans M, Casabona J, Rabella N, De Miniac D. Testing of saliva and serum for HIV
in high-risk populations. Ad hoc group for the comparative saliva and serum study. Eur J
Clin Microbiol Infect Dis 1995; 14:710-713.
11 Holm-Hansen C, Constantine NT. Saliva and HIV testing [letter]. Lancet 1993; 341:382-383.
12 Luo NP, Kasolo F, Ngwenia B, du Pont HL, Zumla A. Use of saliva as an alternative to
serum for HIV screening in Africa. South Afr Med J 1995, 85:156-157.
13 Filkenes K, Ndhlovu Z, Kasumba K, Mubanga Musonda R, Sichone M. Studying dynamics
of the HIV epidemic: population-based data compared with sentinel surveillance in Zambia.
AIDS 1998; 12:1227-1234.
14 Glick P. Scaling up HIV Voluntary Counseling and Testing in Africa: What Can Evaluation
Studies Tell Us About Potential Prevention Impacts. SAGA Working Paper, 2004.

8
Table 1: Prevalence of STIs and HIV in the 2004 MDICP, by survey site and sex (respondents
aged 15 years and above)

TOTAL MEN WOMEN

% N % N % N
Chlamydia
South 0.5 974 0.2 447 0.8 527
Center 0.3 771 0.0 371 0.5 400
North 0.1 896 0.2 435 0.0 456
Total 0.3 2,641 0.2 1,253 0.4 1,383
Gonorrhea
South 5.6 977 0.5 447 10.0 530
Center 2.3 772 0.0 371 4.5 401
North 1.1 896 0.5 435 1.8 456
Total 3.1 2,645 0.3 1,253 5.7 1,387
Trichomonas
South 2.7 513 — — 2.7 513
Center 4.1 418 — — 4.1 418
North 0.6 489 — — 0.6 489
Total 2.3 1,458 — — 2.3 1,458
HIV
South 8.2 1007 7.2 445 9.1 562
Center 7.5 848 6.7 390 8.3 458
North 4.5 961 3.4 440 5.4 516
Total 6.7 2,816 5.7 1,275 7.6 1,536

9
Table 2: Coverage of STI and HIV testing (proportion of respondents who were found and who
agreed to be tested), by survey site, sex and sample

TOTAL
Adults Adolescents
STI HIV N STI HIV N
South 91.4 91.8 759 88.1 89.8 352
Center 87.6 89.2 631 85.4 88.7 295
North 90.4 91.7 687 91.1 96.0 348
Total 89.9 90.9 2077 88.3 91.6 879
MEN
Adults Adolescents
STI HIV N STI HIV N
South 92.3 91.7 312 92.0 91.4 174
Center 90.6 90.4 254 91.2 90.7 159
North 89.0 89.3 299 96.1 97.8 178
Total 90.6 90.5 865 93.2 93.4 511
WOMEN
Adults Adolescents
STI HIV N STI HIV N
South 90.8 92.0 447 84.3 86.2 178
Center 86.1 88.8 375 78.7 86.2 136
North 91.5 93.6 388 85.5 93.9 165
Total 89.6 91.4 1210 83.1 89.6 479

10
Table 3: Proportion of respondents tested for HIV or at least one STI who returned for their
results, by survey site, sex and sample

TOTAL
Adults Adolescents
STI HIV STI HIV
South 74.3 74.2 58.2 58.7
Center 81.1 73.9 80.0 77.8
North 61.9 61.4 50.8 50.4
Total 72.2 69.9 59.8 59.5
MEN
Adults Adolescents
STI HIV STI HIV
South 74.7 74.2 68.1 68.6
Center 80.4 72.3 83.4 81.6
North 61.7 61.8 52.0 52.3
Total 71.9 69.5 67.0 66.6
WOMEN
Adults Adolescents
STI HIV STI HIV
South 73.9 74.2 48.2 48.8
Center 81.8 75.5 76.6 73.9
North 62.0 60.9 49.6 48.4
Total 72.4 70.3 52.5 52.4

11