The validity of self-reported likelihood of HIV infection among the general population in rural Malawi

You Know What I Know: Interviewer Knowledge Effects in Subjective Expectation Elicitation

Directly eliciting individuals' subjective beliefs via surveys is increasingly popular in social science research, but doing so via face-to-face surveys has an important downside: the interviewer's knowledge of the topic may spill over onto the respondent's recorded beliefs. Using a randomized experiment that used interviewers to implement an information treatment, we show that reported beliefs are significantly shifted by interviewer knowledge. Trained interviewers primed respondents to use the exact numbers used in the training, nudging them away from higher answers; recorded responses decreased by about 0.3 standard deviations of the initial belief distribution. Furthermore, respondents with stronger prior beliefs were less affected by interviewer knowledge. We suggest corrections for this issue from the perspectives of interviewer recruitment, survey design, and experiment setup.

Distributional impact of the Malawian Essential Health Package

In low- and middle-income countries (LMICs), making the best use of scarce resources is essential to achieving universal health coverage. The design of health benefits packages creates the opportunity to select interventions on the basis of explicit objectives. Distributional cost-effectiveness analysis (DCEA) provides a framework to evaluate interventions based on two objectives: increasing population health and reducing health inequality. We conduct aggregate DCEA of potential health benefits package interventions to demonstrate the feasibility of this approach in LMICs, using the case of the Malawian health benefits package. We use publicly available survey and census data common to LMICs and describe what challenges we encountered and how we addressed them. We estimate that diseases targeted by the health benefits package are most prevalent in the poorest population quintile and least prevalent in the richest quintile. The survey data we use indicate socioeconomic patterns in intervention uptake that diminish the population health gain and inequality reduction from the package. We find that a similar set of interventions would be prioritized when impact on health inequality is incorporated alongside impact on overall population health. However, conclusions about the impact of individual interventions on health inequalities are sensitive to assumptions regarding the health opportunity cost, the utilization of interventions, the distribution of diseases across population groups and the level of aversion to inequality. Our results suggest that efforts to improve access to the Essential Health Package could be targeted to specific interventions to improve the health of the poorest fastest but that identifying these interventions is uncertain. This exploratory work has shown the potential for applying the DCEA framework to inform health benefits package design within the LMIC setting and to provide insight into the equity impact of a health benefits package.

Assessing the validity of and factors that influence accurate self-reporting of HIV status after testing: a population-based study

OBJECTIVES

To assess the validity of self-reported HIV status, and investigate factors that influence accurate reporting of HIV-positive status, in a population tested and informed of their HIV test result.

DESIGN

Prospective cohort study.

METHODS

We compared self-reported HIV status with biomarker-confirmed HIV test status among participants of Karonga Health and Demographic Surveillance Site in rural northern Malawi. We linked information on HIV test results to subsequent self-reported HIV status, and calculated sensitivity, specificity, positive predictive value and negative predictive value for self-reported HIV status (considered as a diagnostic test). We used Poisson regression with robust variance estimators to examine predictors of accurate self-reporting of HIV-positive status.

RESULTS

Among 17 445 adults who tested for HIV, were recorded as having received their HIV test results, and had a subsequent self-reported HIV status between 2007 and 2018: positive predictive value of self-reported HIV status was 98.0% (95% confidence interval: 97.3-98.7); negative predictive value was 98.3 (98.1-98.5); sensitivity was 86.1% (84.5-87.7); and specificity was 99.8% (99.7-99.9). Among true HIV-positive people, those who were younger, interviewed in community settings, and had tested for HIV longer ago were more likely to misreport their HIV-positive status.

CONCLUSION

In this setting, self-report provides good estimates of test-detected HIV prevalence, suggesting that it can be used when HIV test results are not available. Despite frequent HIV testing, younger people and those interviewed in community settings were less likely to accurately report their HIV-positive status. More research on barriers to self-reporting of HIV status is needed in these subgroups.

Disparities Between HIV Testing Levels and the Self-Reported HIV-Negative Status of Sexually Active College Students

The methodological issues of using self-reported human immunodeficiency virus (HIV) testing history to determine HIV infection status are understudied. This study aimed to assess the disparities between HIV testing history and self-reported HIV status, with a specific objective of estimating the prevalence of self-reported HIV-negative status in sexually active students who had never had an HIV test (inaccurate self-disclosure). Participants were recruited from four Hong Kong universities; 255 students were included in the analysis, with 16.1% of participants being men who have sex with men (MSM). Overall, 81.6% of participants reported they never had been tested for HIV, with inaccurate self-disclosure by 65.9% of participants. Among participants who used condoms inconsistently, only 20.2% had had HIV tests, with inaccurate self-disclosure by 66.7% of subjects. Among MSM students, only 36.6% had had HIV tests, with inaccurate self-disclosure by 61.0% of participants. Among MSM students who used condoms inconsistently, only 35.1% had had HIV tests, with inaccurate self-disclosure by 62.2% of participants. The findings raise concerns about the use of self-reported HIV status in clinical practice and research. The low prevalence of HIV testing found suggests that interventions to promote HIV testing should be continued.

Why do marital partners of people living with HIV not test for HIV? A qualitative study in Lusaka, Zambia

BACKGROUND

Knowledge of HIV status is crucial for HIV prevention and management in marital relationships. Yet some marital partners of people living with HIV decline HIV testing despite knowing the HIV-positive status of their partners. To date, little research has explored the reasons for this.

METHODS

An exploratory qualitative study was undertaken in Lusaka, Zambia, between March 2010 and September 2011, nested within a larger ethnographic study. In-depth interviews were held with individuals who knew the HIV-positive status of their marital partners but never sought HIV testing (n = 30) and HIV service providers of a public sector clinic (n = 10). A focus group discussion was also conducted with eight (8) lay HIV counsellors. Data was transcribed, coded and managed using ATLAS.ti and analysed using latent content analysis.

RESULTS

The overarching barrier to uptake of HIV testing was study participants' perception of their physical health, reinforced by uptake of herbal remedies and conventional non-HIV medication to mitigate perceived HIV-related symptoms. They indicated willingness to test for HIV if they noticed a decline in physical health and other alternative forms of care became ineffective. Also, some study participants viewed themselves as already infected with HIV on account of the HIV-positive status of their marital partners, with some opting for faith healing to get 'cured'. Other barriers were the perceived psychological burden of living with HIV, modulated by lay belief that knowledge of HIV-positive status led to rapid physical deterioration of health. Perceived inability to sustain uptake of life-long treatment - influenced by a negative attitude towards treatment - further undermined uptake of HIV testing. Self-stigma, which manifested itself through fear of blame and a need to maintain moral credibility in marital relationships, also undermined uptake of HIV testing.

CONCLUSIONS

Improving uptake of HIV testing requires a multi-pronged approach that addresses self-stigma, lay risk perceptions, negative treatment and health beliefs and the perceived psychological burden of living with HIV. Strengthening couple HIV testing services, including addressing conflict and addressing gendered power relationships are also warranted to facilitate joint knowledge, acceptance and management of HIV status in marital relationships.

Comparison of self-reported HIV testing data with medical records data in Houston, TX 2012-2013

PURPOSE

To assess the agreement between self-reported and medical record data on HIV status and dates of first positive and last negative HIV tests.

METHODS

Participants were recruited from patients attending Houston health clinics during 2012-2013. Self-reported data were collected using a questionnaire and compared with medical record data. Agreement of HIV status was assessed using kappa statistics and of HIV test dates using concordance correlation coefficient. The extent of difference between self-reported and medical record test dates was determined.

RESULTS

Agreement between self-reported and medical record data was good on HIV status and date of first positive HIV test, but poor on date of last negative HIV test. About half of participants that self-reported never tested had HIV test results in medical records. Agreement varied by sex, race and/or ethnicity, and medical care facility. For HIV-positive persons, more self-reported first positive HIV test dates preceded medical record dates, with a median difference of 6 months. For HIV-negative persons, more medical record dates of last negative HIV test preceded self-reported dates, with a median difference of 2 months.

CONCLUSIONS

Studies relying on self-reported HIV status other than HIV positive and self-reported date of last negative should consider including information from additional sources to validate the self-reported data.

Cohort Profile: The Malawi Longitudinal Study of Families and Health (MLSFH)

The Malawi Longitudinal Study of Families and Health (MLSFH) is one of very few long-standing, publicly available longitudinal cohort studies in a sub-Saharan African (SSA) context. It provides a rare record of more than a decade of demographic, socioeconomic and health conditions in one of the world's poorest countries. The MLSFH was initially established in 1998 to study social network influences on fertility behaviours and HIV risk perceptions, and over time the focus of the study expanded to include health, sexual behaviours, intergenerational relations and family/household dynamics. The currently available data include MLSFH rounds collected in 1998, 2001, 2004, 2006, 2008, 2010 and 2012 for up to 4000 individuals, providing information about socioeconomic and demographic characteristics, sexual behaviours, marriage, household/family structure, risk perceptions, social networks and social capital, intergenerational relations, HIV/AIDS and other dimensions of health. The MLSFH public use data can be requested on the project website: http://www.malawi.pop.upenn.edu/.

Prevalence, correlates, and risk perception of HIV infection among heroin users in Central Taiwan

We investigated the prevalence and correlated factors of human immunodeficiency virus (HIV) among heroin users attending methadone maintenance treatment (MMT) programs in Central Taiwan, and explored the degree of risk perception of HIV infection among the participants. Our study participants were 781 heroin users seeking treatment at the MMT program at Tsaotun Psychiatric Center in Taiwan. The presence of HIV antibodies was identified by microparticle enzyme immunoassay and confirmed by western blot. Multivariate logistic regression was used to identify the independent correlates of HIV infection. The mean age of the sample was 36.1 years [standard deviation (SD) = 7.6]; of the patients, 710 (90.9%) were men. The prevalence of HIV infection among our study population was 20.7%. Multivariate logistic regression analysis revealed that HIV infection was independently associated with the age of the patients of initial heroin use, heroin injection use, nondrug-related criminal convictions, needle-sharing behaviors, and sharing injection paraphernalia. A strong agreement existed between self-reported HIV serostatus and the results of laboratory analyses, with 88.8% of patients reporting their condition correctly. We found a high rate of HIV infection among patients in the MMT program. Factors associated with HIV infection were mostly related to drug-use behaviors. These findings stress the importance of education regarding drug-risk behaviors.

The effect of marriage and HIV risks on condom use acceptability in rural Malawi

A large and increasing proportion of HIV transmissions in sub-Saharan Africa occur within marriage. Condom use within marriage could, therefore, be an important prevention strategy, but there is considerable debate about whether married couples would be willing to use condoms. This paper contributes to this debate by identifying key factors that affect the acceptability of condom use within marriage and actual condom use among men and women in rural Malawi, using three waves of longitudinal data from 2004, 2006 and 2008. Specifically, we focused on the effect of (1) entry into first marriage, (2) respondent's HIV status, HIV perceptions, and risk behaviors, and (3) spouse's HIV characteristics on condom use acceptability within marriage and actual condom use with a spouse or steady partner. Using fixed-effects regression, we found that getting married coincides with a pronounced attitudinal shift regarding the acceptability of condom use within marriage that cannot be explained by differences in fertility status or selection into marriage. In addition, we found that, for women, perceived HIV status of the respondent and spouse generally had greater influence than actual HIV status on the acceptability of condom use within marriage and actual condom use with a spouse or steady partner, even after HIV status is known; while actual HIV status and HIV risk behaviors are generally more important among men. Although condom use within marriage remained low, these findings suggest that attitudes about and use of condoms are susceptible to change and that both marital status and perceptions of risk are important influences on condom use.

Now or never: perceived HIV status and fertility intentions in rural Mozambique

As the HIV epidemic evolves, researchers are devoting increased attention to the infection's effect on various life-course activities, including marriage and reproduction. The impact of HIV on decisions about childbearing is particularly important, given the role that vertical transmission plays in the persistence of the epidemic. Previous studies on HIV and fertility intentions have yielded inconsistent results. This article expands on prior research by taking into account preferred timing of childbearing. Using data from a population-based survey in rural Mozambique, we show that higher perceived risk of HIV is associated with greater likelihood of both wanting to speed up childbearing and wanting to stop having children. The "now or never" approach to childbearing is shown to be consistent with the widely held belief that HIV infection is incompatible with childbearing in the long term.

HIV infection and fertility preferences in rural Malawi

Although HIV-prevalence and fertility rates in sub-Saharan Africa are among the highest in the world, little is known about how HIV infection affects the fertility preferences of men and women in the region. A quasi-experimental design and in-depth interviews conducted in rural Malawi are employed to examine how and through what pathways learning that one is HIV positive alters a person's childbearing desires. Among rural Malawians, particularly men, the desire to have more children decreases after receiving a positive HIV-test result. The motivations underlying this effect are greatly influenced by gender: women fear the physical health consequences of HIV-positive pregnancies and childbearing, whereas men see childbearing as futile because they anticipate their own early death and the deaths of their future children. Considerable ambivalence remains, nevertheless, particularly among women who strategize to live normal lives in spite of their infection, but whose definitions of "normal" vary.

Perception of vulnerability to HIV infection among older people in Nairobi, Kenya: a need for intervention

It is evident that sexual activity tends to decrease with age. Nonetheless, it is still prevalent enough to be considered a risk factor for the spread of HIV among older people. This paper uses quantitative data for 2053 individuals to examine HIV risk perception and correlates of perceived risk among older people aged 50 years and older living in Nairobi slums. It emerged that a majority of older people did not consider themselves at risk of infection. Of those who felt at risk, a greater proportion sensed only a small chance of contracting HIV. Women cited 'no sexual activity' while men mentioned 'having only one and/or a faithful sexual partner' as the primary reasons for perceiving minimal risk of HIV infection. There were no differences by sex in the basis for perceiving moderate-to-great risk of infection. Religion is a key factor in risk perception with Muslims perceiving higher levels of risk and, conversely, devotees irrespective of faith perceiving lower levels of risk. Older people willing to be tested for HIV had a decreased likelihood of perceived risk compared with those unwilling to be tested. This paper recommends evaluation of older people's perception of risk in order to better inform interventions aimed at minimizing their vulnerability to HIV infection.

Gender variation in self-reported likelihood of HIV infection in comparison with HIV test results in rural and urban Nigeria

Background

Behaviour change which is highly influenced by risk perception is a major challenge that HIV prevention efforts need to confront. In this study, we examined the validity of self-reported likelihood of HIV infection among rural and urban reproductive age group Nigerians.

Methods

This is a cross-sectional study of a nationally representative sample of Nigerians. We investigated the concordance between self-reported likelihood of HIV and actual results of HIV test. Multivariate logistic regression analysis was used to assess whether selected respondents' characteristics affect the validity of self-reports.

Results

The HIV prevalence in the urban population was 3.8% (3.1% among males and 4.6% among females) and 3.5% in the rural areas (3.4% among males and 3.7% among females). Almost all the respondents who claimed they have high chances of being infected with HIV actually tested negative (91.6% in urban and 97.9% in rural areas). In contrast, only 8.5% in urban areas and 2.1% in rural areas, of those who claimed high chances of been HIV infected were actually HIV positive. About 2.9% and 4.3% from urban and rural areas respectively tested positive although they claimed very low chances of HIV infection. Age, gender, education and residence are factors associated with validity of respondents' self-perceived risk of HIV infection.

Conclusion

Self-perceived HIV risk is poorly sensitive and moderately specific in the prediction of HIV status. There are differences in the validity of self-perceived risk of HIV across rural and urban populations.

Uncertainty and Fertility in a Generalized AIDS Epidemic

Sociologists widely acknowledge that uncertainty matters for decision making, but they rarely measure it directly. In this article, we demonstrate the importance of theorizing about, measuring, and analyzing uncertainty as experienced by individuals. We adapt a novel probabilistic solicitation technique to measure personal uncertainty about HIV status in a high HIV prevalence area of southern Malawi. Using data from 2,000 young adults (ages 15 to 25 years), we demonstrate that uncertainty about HIV status is widespread and that it expands as young adults assess their proximate and distant futures. In conceptualizing HIV status as something more than sero-status itself, we gain insight into how what individuals know they don't know influences their lives. Young people who are uncertain about their HIV status express desires to accelerate their childbearing relative to their counterparts who are certain they are uninfected. Our approach and findings show that personal uncertainty is a measurable and meaningful phenomenon that can illuminate much about individuals' aspirations and behaviors.

Provider-initiated HIV testing and counseling: increased uptake in two public community health centers in South Africa and implications for scale-up

Background

International guidance recommends the scale up of routinely recommended, offered, and delivered health care provider-initiated HIV testing and counseling (PITC) to increase the proportion of persons who know their HIV status. We compared HIV test uptake under PITC to provider-referral to voluntary counseling and testing (VCT referral) in two primary health centers in South Africa.

Methods

Prior to introducing PITC, clinical providers were instructed to refer systematically selected study participants to VCT. After PITC and HIV rapid test training, providers were asked to recommend, offer and provide HIV testing to study participants during the clinical consultation. Participants were interviewed before and after their consultation to assess their HIV testing experiences.

Results

HIV test uptake increased under PITC (OR 2.85, 95% CI 1.71, 4.76), and more patients felt providers answered their questions on HIV (104/141 [74%] versus 73/118 [62%] for VCT referral; p 0.04). After three months, only 4/106 (3.8%) HIV-positive patients had registered for onsite HIV treatment. Providers found PITC useful, but tested very few patients (range 0–15).

Conclusion

PITC increased the uptake of HIV testing compared with referral to onsite VCT, and patients reported a positive response to PITC. However, providing universal PITC will require strong leadership to train and motivate providers, and interventions to link HIV-positive persons to HIV treatment centers.

Disclosure of HIV status between spouses in rural Malawi

Disclosure of HIV status after HIV voluntary counseling and testing has important implications for the spread of the HIV epidemic and the health of individuals who are HIV positive. Here, we use individual and couples level data for currently married respondents from an ongoing longitudinal study in rural Malawi to (1) examine the extent of HIV status disclosure by HIV serostatus; (2) identify reasons for not sharing one's HIV status with a spouse; and (3) evaluate the reliability of self-reports of HIV status disclosure. We find that disclosure of HIV status is relatively common among rural Malawians, where most have shared their status with a spouse, and many disclose to others in the community. However, there are significant differences in disclosure patterns by HIV status and gender. Factors associated with non-disclosure are also gendered, where women who perceive greater HIV/AIDS stigma and HIV positive are less likely to disclose HIV status to a spouse, and men who are worried about HIV infection from extramarital partners are less likely to disclose their HIV status to a spouse. Finally, we test the reliability of self-reported HIV status disclosure and find that self-reports of HIV-positive men are of questionable reliability.

Assessing the importance of gender roles in couples' home-based sexual health services in Malawi

To more effectively address individuals' and couples' sexual and reproductive health needs, innovative service delivery strategies are being explored. These strategies are logistically and ethically complicated, considering prevailing gender inequalities in many contexts. We conducted an exploratory study to assess the acceptability of couples' home-based sexual health services in Malawi. We collected qualitative data from six focus group discussions and 10 husband-wife indepth interviews to gain a more thorough understanding of how gender norms influence acceptability of couples' sexual health services. Findings reveal that women are expected to defer to their husbands and may avoid conflict through covert contraceptive use and non-disclosure of HIV status. Many men felt that accessing sexual health services is stigmatizing, causing some to avoid services or to rely on informal information sources. Gender norms and attitudes toward existing services differentially impact men and women in this setting, influencing the perceived benefits of couples' sexual health services.

HIV risk among currently married couples in rural Malawi: what do spouses know about each other?

In countries with generalized HIV/AIDS epidemics, married couples have a shared risk of acquiring HIV/AIDS. Yet very little research has adopted a couple-level perspective to investigate perceived risk of HIV infection. In this paper, we used population-based data from 768 married monogamous couples in the 2004 Malawi Diffusion and Ideational Change Project (MDICP) to compare respondents' perceptions about their spouses' HIV status to their spouses' actual HIV status. Using chi-squared and Kappa coefficient statistics, we evaluated how accurately respondents assess their spouse's HIV status, and compared the assessment of their spouse's HIV status with their assessment of their own serostatus. We found that individuals tend to overestimate their spouse's as well as their own risk of having HIV. Husbands were generally more accurate in assessing their own risk of HIV infection than that of their wives, but wives were more accurate in assessing the HIV status of their spouses. In our multivariate logistic regression results, we found that marital infidelity is the most important correlate of overestimating individual and spousal HIV risk.

The Malawi Religion Project: Data collection and selected analyses

Scholars have recently become increasingly interested in the role religion plays in the responses to the HIV/AIDS epidemic in sub-Saharan Africa. Here, we present the Malawi Religion Project (MRP), which provides data to examine the relationship between religion and HIV/AIDS through surveys and in-depth interviews with denominational leaders, congregational leaders, and congregation members in three districts of rural Malawi. In the paper, we outline existing perspectives on the religion-HIV/AIDS link, describe the MRP's design, implementation, and subsequent data; provide initial evidence for a series of general research hypotheses; and describe how these data can be used both to extend explorations of these relationships further and as a model for gathering similar data in other contexts. In particular we highlight the unique possibilities this project provides for analyses that link MRP data to the Malawi Diffusion and Ideational Change Project. These linked data produce a multi-level data set covering individuals, congregations and their communities, allowing empirical research on religion, HIV/AIDS risk, related behaviors, attitudes, and norms.

The impact of HIV status and perceived status on fertility desires in rural Malawi

Surprisingly little is known about how or if knowledge of one's HIV status influences the desire to have children in sub-Saharan Africa. This paper takes advantage of a unique situation in rural Malawi where no one knew their HIV status prior to testing being introduced as part of an ongoing panel study. Using a sample of men and women (n = 1,380) who were interviewed in 2001 and 2006, we examine how HIV positive and negative test results impact respondents' desires to continue childbearing. We then consider how prior perceptions of one's status influence this effect. Respondents who received a positive test result reduced their childbearing desires. When self-assessed likelihood of infection was considered, however, only those who were surprised by their test result altered their fertility preferences, a finding which held true for both positive and negative results. The implications of the results and potential applications to other HIV/AIDS research are discussed.

Overestimating HIV infection: The construction and accuracy of subjective probabilities of HIV infection in rural Malawi

In the absence of HIV testing, how do rural Malawians assess their HIV status? In this paper, we use a unique dataset that includes respondents' HIV status as well as their subjective likelihood of HIV infection. These data show that many rural Malawians overestimate their likelihood of current HIV infection. The discrepancy between actual and perceived status raises an important question: Why are so many wrong? We begin by identifying determinants of self-assessed HIV status, and then compare these assessments with HIV biomarker results. Finally, we ask what characteristics of individuals are associated with errors in self-assessments.

The Demand for, and Impact of, Learning HIV Status

This paper evaluates an experiment in which individuals in rural Malawi were randomly assigned monetary incentives to learn their HIV results after being tested. Distance to the HIV results centers was also randomly assigned. Without any incentive, 34 percent of the participants learned their HIV results. However, even the smallest incentive doubled that share. Using the randomly assigned incentives and distance from results centers as instruments for the knowledge of HIV status, sexually active HIV-positive individuals who learned their results are three times more likely to purchase condoms two months later than sexually active HIV-positive individuals who did not learn their results; however, HIV-positive individuals who learned their results purchase only two additional condoms than those who did not. There is no significant effect of learning HIV-negative status on the purchase of condoms.

The Demand for, and Impact of, Learning HIV Status

This paper evaluates an experiment in which individuals in rural Malawi were randomly assigned monetary incentives to learn their HIV results after being tested. Distance to the HIV results centers was also randomly assigned. Without any incentive, 34 percent of the participants learned their HIV results. However, even the smallest incentive doubled that share. Using the randomly assigned incentives and distance from results centers as instruments for the knowledge of HIV status, sexually active HIV-positive individuals who learned their results are three times more likely to purchase condoms two months later than sexually active HIV-positive individuals who did not learn their results; however, HIV-positive individuals who learned their results purchase only two additional condoms than those who did not. There is no significant effect of learning HIV-negative status on the purchase of condoms.

Ethical and public health considerations in HIV counseling and testing: policy implications

HIV counseling and testing is broadly considered a critical component of HIV transmission-prevention and treatment efforts. Given the severity of the AIDS pandemic in sub-Saharan Africa, the potential societal benefit of testing is invoked to call for its massive expansion and to justify a shift from voluntary to routine testing. Surprisingly little evidence has demonstrated, however, that such a shift will result in the intended benefits to communities, particularly that of reducing the horizontal transmission of HIV. This analysis addresses and critiques the assumptions underlying a serostatus-based approach to behavior change and discusses the ethical consequences of transferring control of the decision to be tested from the individual to the provider. It concludes with a discussion of the implications for HIV counseling and testing policies and proposes alternatives to routine testing that have the potential to be effective while preserving the right to know one's HIV status.

Social networks and HIV/AIDs risk perceptions

Understanding the determinants of individuals' perceptions of their risk of becoming infected with HIV and their perceptions of acceptable strategies of prevention is an essential step toward curtailing the spread of this disease. We focus in this article on learning and decision-making about AIDS in the context of high uncertainty about the disease and appropriate behavioral responses. We argue that social interactions are important for both. Using longitudinal survey data from rural Kenya and Malawi, we test this hypothesis. We investigate whether social interactions--and especially the extent to which social network partners perceive themselves to be at risk--exert causal influences on respondents' risk perceptions and on one approach to prevention, spousal communication about the threat of AIDS to the couple and their children. The study explicitly allows for the possibility that important characteristics, such as unobserved preferences or community characteristics, determine not only the outcomes of interest but also the size and composition of networks. The most important empirical result is that social networks have significant and substantial effects on risk perceptions and the adoption of new behaviors even after we control for unobserved factors.