Universal Testing, Expanded Treatment, and Incidence of HIV Infection in Botswana.

BACKGROUND: The feasibility of reducing the population-level incidence of human immunodeficiency virus (HIV) infection by increasing community coverage of antiretroviral therapy (ART) and male circumcision is unknown. METHODS: We conducted a pair-matched, community-randomized trial in 30 rural or periurban communities in Botswana from 2013 to 2018. Participants in 15 villages in the intervention group received HIV testing and counseling, linkage to care, ART (started at a higher CD4 count than in standard care), and increased access to male circumcision services. The standard-care group also consisted of 15 villages. Universal ART became available in both groups in mid-2016. We enrolled a random sample of participants from approximately 20% of households in each community and measured the incidence of HIV infection through testing performed approximately once per year. The prespecified primary analysis was a permutation test of HIV incidence ratios. Pair-stratified Cox models were used to calculate 95% confidence intervals. RESULTS: Of 12,610 enrollees (81% of eligible household members), 29% were HIV-positive. Of the 8974 HIV-negative persons (4487 per group), 95% were retested for HIV infection over a median of 29 months. A total of 57 participants in the intervention group and 90 participants in the standard-care group acquired HIV infection (annualized HIV incidence, 0.59% and 0.92%, respectively). The unadjusted HIV incidence ratio in the intervention group as compared with the standard-care group was 0.69 (P = 0.09) by permutation test (95% confidence interval [CI], 0.46 to 0.90 by pair-stratified Cox model). An end-of-trial survey in six communities (three per group) showed a significantly greater increase in the percentage of HIV-positive participants with an HIV-1 RNA level of 400 copies per milliliter or less in the intervention group (18 percentage points, from 70% to 88%) than in the standard-care group (8 percentage points, from 75% to 83%) (relative risk, 1.12; 95% CI, 1.09 to 1.16). The percentage of men who underwent circumcision increased by 10 percentage points in the intervention group and 2 percentage points in the standard-care group (relative risk, 1.26; 95% CI, 1.17 to 1.35). CONCLUSIONS: Expanded HIV testing, linkage to care, and ART coverage were associated with increased population viral suppression. (Funded by the President's Emergency Plan for AIDS Relief and others; Ya Tsie ClinicalTrials.gov number, NCT01965470.).

Epidemiological and viral characteristics of undiagnosed HIV infections in Botswana.

BACKGROUND: HIV-1 is endemic in Botswana. The country's primary challenge is identifying people living with HIV who are unaware of their status. We evaluated factors associated with undiagnosed HIV infection using HIV-1 phylogenetic, behavioural, and demographic data. METHODS: As part of the Botswana Combination Prevention Project, 20% of households in 30 villages were tested for HIV and followed from 2013 to 2018. A total of 12,610 participants were enrolled, 3596 tested HIV-positive at enrolment, and 147 participants acquired HIV during the trial. Extensive socio-demographic and behavioural data were collected from participants and next-generation sequences were generated for HIV-positive cases. We compared three groups of participants: (1) those previously known to be HIV-positive at enrolment (n = 2995); (2) those newly diagnosed at enrolment (n = 601) and (3) those who tested HIV-negative at enrolment but tested HIV-positive during follow-up (n = 147). We searched for differences in demographic and behavioural factors between known and newly diagnosed group using logistic regression. We also compared the topology of each group in HIV-1 phylogenies and used a genetic diversity-based algorithm to classify infections as recent (< 1 year) or chronic (≥ 1 year). RESULTS: Being male (aOR = 2.23) and younger than 35 years old (aOR = 8.08) was associated with undiagnosed HIV infection (p < 0.001), as was inconsistent condom use (aOR = 1.76). Women were more likely to have undiagnosed infections if they were married, educated, and tested frequently. For men, being divorced increased their risk. The genetic diversity-based algorithm classified most incident infections as recent (75.0%), but almost none of known infections (2.0%). The estimated proportion of recent infections among new diagnoses was 37.0% (p < 0.001). CONCLUSION: Our results indicate that those with undiagnosed infections are likely to be young men and women who do not use condoms consistently. Among women, several factors were predictive: being married, educated, and testing frequently increased risk. Men at risk were more difficult to delineate. A sizeable proportion of undiagnosed infections were recent based on a genetic diversity-based classifier. In the era of "test and treat all", pre-exposure prophylaxis may be prioritized towards individuals who self-identify or who can be identified using these predictors in order to halt onward transmission in time.

Mapping of HIV-1C Transmission Networks Reveals Extensive Spread of Viral Lineages Across Villages in Botswana Treatment-as-Prevention Trial.

BACKGROUND: Phylogenetic mapping of HIV-1 lineages circulating across defined geographical locations is promising for better understanding HIV transmission networks to design optimal prevention interventions. METHODS: We obtained near full-length HIV-1 genome sequences from people living with HIV (PLWH), including participants on antiretroviral treatment in the Botswana Combination Prevention Project, conducted in 30 Botswana communities in 2013-2018. Phylogenetic relationships among viral sequences were estimated by maximum likelihood. RESULTS: We obtained 6078 near full-length HIV-1C genome sequences from 6075 PLWH. We identified 984 phylogenetically distinct HIV-1 lineages (molecular HIV clusters) circulating in Botswana by mid-2018, with 2-27 members per cluster. Of these, dyads accounted for 62%, approximately 32% (n = 316) were found in single communities, and 68% (n = 668) were spread across multiple communities. Men in clusters were approximately 3 years older than women (median age 42 years, vs 39 years; P < .0001). In 65% of clusters, men were older than women, while in 35% of clusters women were older than men. The majority of identified viral lineages were spread across multiple communities. CONCLUSIONS: A large number of circulating phylogenetically distinct HIV-1C lineages (molecular HIV clusters) suggests highly diversified HIV transmission networks across Botswana communities by 2018.

Implementation of Universal HIV Testing and Treatment to Reduce HIV Incidence in Botswana: the Ya Tsie Study.

PURPOSE OF REVIEW: Antiretroviral treatment (ART) can dramatically reduce the risk of HIV transmission, but the feasibility of scaling up HIV testing, linkage and treatment to very high population levels, and its impact on population HIV incidence, were unknown. We review key findings from a community-randomized trial in which we evaluated the impact of "universal test and treat" (UTT) on population HIV incidence in Botswana, a resource-constrained country with both high HIV prevalence and high ART coverage before study inception. RECENT FINDINGS: We conducted a community-randomized trial (the "Ya Tsie" trial or Botswana Combination Prevention Project (BCPP)) in 30 villages in Botswana from 2013 to 2018, with the goal of determining whether a combination of prevention interventions-with a focus on universal HIV testing and treatment-would reduce population-level HIV incidence. The intervention included universal HIV testing (home-based and mobile), active linkage to HIV care and treatment with patient tracing for persons not linking, universal ART coverage, rapid ART start (at the first clinic visit), and enhanced male circumcision services. Botswana had very high HIV diagnosis, treatment, and viral suppression levels (approaching the UNAIDS "90-90-90" targets) prior to intervention roll-out. By study end, we were able to exceed the overall 95-95-95 coverage target of 86%: an estimated 88% of all persons living with HIV were on ART and had viral suppression in the Ya Tsie intervention arm. In addition, annual HIV incidence was 30% lower in the intervention arm as compared with the control arm over a 29-month follow-up period. With universal HIV testing and relatively simple linkage activities, it was possible to achieve one of the highest reported population levels of HIV diagnosis, linkage to care, and viral suppression globally and to reduce population HIV incidence by about one-third over a short period of time (< 3 years). We were able to significantly increase population viral suppression and to decrease HIV incidence even in a resource-constrained setting with pre-existing very high testing and treatment coverage. Universal community-based HIV testing and tracing of individuals through the HIV care cascade were key intervention components.

Efficient estimation of human immunodeficiency virus incidence rate using a pooled cross-sectional cohort study design.

Development of methods to accurately estimate human immunodeficiency virus (HIV) incidence rate remains a challenge. Ideally, one would follow a random sample of HIV-negative individuals under a longitudinal study design and identify incident cases as they arise. Such designs can be prohibitively resource intensive and therefore alternative designs may be preferable. We propose such a simple, less resource-intensive study design and develop a weighted log likelihood approach which simultaneously accounts for selection bias and outcome misclassification error. The design is based on a cross-sectional survey which queries individuals' time since last HIV-negative test, validates their test results with formal documentation whenever possible, and tests all persons who do not have documentation of being HIV-positive. To gain efficiency, we update the weighted log likelihood function with potentially misclassified self-reports from individuals who could not produce documentation of a prior HIV-negative test and investigate large sample properties of validated sub-sample only versus pooled sample estimators through extensive Monte Carlo simulations. We illustrate our method by estimating incidence rate for individuals who tested HIV-negative within 1.5 and 5 years prior to Botswana Combination Prevention Project enrolment. This article establishes that accurate estimates of HIV incidence rate can be obtained from individuals' history of testing in a cross-sectional cohort study design by appropriately accounting for selection bias and misclassification error. Moreover, this approach is notably less resource-intensive compared to longitudinal and laboratory-based methods.

Voluntary Medical Male Circumcision and Perceived Sexual Functioning, Satisfaction, and Risk Behavior: A Qualitative Study in Botswana.

This article examines perceptions of sexual functioning, satisfaction, and risk-taking related to voluntary medical male circumcision (VMMC) in Botswana. Twenty-seven focus group discussions were conducted in four purposively selected communities with community leaders, men, and women. Discussions were analyzed using an inductive content analytic approach. Perceptions of VMMC's impact on sexual functioning and satisfaction varied. Increased satisfaction was attributed to improved penile health and increased ejaculatory latency time, whereas decreased satisfaction was attributed to erectile dysfunction and increased vaginal irritation during sex. Most participants thought sexual disinhibition occurred after circumcision; nevertheless, some women said they used male circumcision status as a marker of HIV status, thereby influencing sexual decision-making and partner selection. Messaging should emphasize that VMMC does not afford complete HIV protection. Optimizing VMMC's impact requires increasing uptake while minimizing behavioral disinhibition, with a balance between potential messaging of improved sexual functioning and satisfaction and the potential impact on sexual disinhibition.

Methodological Challenges When Studying Distance to Care as an Exposure in Health Research.

Distance to care is a common exposure and proposed instrumental variable in health research, but it is vulnerable to violations of fundamental identifiability conditions for causal inference. We used data collected from the Botswana Birth Outcomes Surveillance study between 2014 and 2016 to outline 4 challenges and potential biases when using distance to care as an exposure and as a proposed instrument: selection bias, unmeasured confounding, lack of sufficiently well-defined interventions, and measurement error. We describe how these issues can arise, and we propose sensitivity analyses for estimating the degree of bias.

Implementation of Instrumental Variable Bounds for Data Missing Not at Random.

Instrumental variables are routinely used to recover a consistent estimator of an exposure causal effect in the presence of unmeasured confounding. Instrumental variable approaches to account for nonignorable missing data also exist but are less familiar to epidemiologists. Like instrumental variables for exposure causal effects, instrumental variables for missing data rely on exclusion restriction and instrumental variable relevance assumptions. Yet these two conditions alone are insufficient for point identification. For estimation, researchers have invoked a third assumption, typically involving fairly restrictive parametric constraints. Inferences can be sensitive to these parametric assumptions, which are typically not empirically testable. The purpose of our article is to discuss another approach for leveraging a valid instrumental variable. Although the approach is insufficient for nonparametric identification, it can nonetheless provide informative inferences about the presence, direction, and magnitude of selection bias, without invoking a third untestable parametric assumption. An important contribution of this article is an Excel spreadsheet tool that can be used to obtain empirical evidence of selection bias and calculate bounds and corresponding Bayesian 95% credible intervals for a nonidentifiable population proportion. For illustrative purposes, we used the spreadsheet tool to analyze HIV prevalence data collected by the 2007 Zambia Demographic and Health Survey (DHS).

A general instrumental variable framework for regression analysis with outcome missing not at random.

The instrumental variable (IV) design is a well-known approach for unbiased evaluation of causal effects in the presence of unobserved confounding. In this article, we study the IV approach to account for selection bias in regression analysis with outcome missing not at random. In such a setting, a valid IV is a variable which (i) predicts the nonresponse process, and (ii) is independent of the outcome in the underlying population. We show that under the additional assumption (iii) that the IV is independent of the magnitude of selection bias due to nonresponse, the population regression in view is nonparametrically identified. For point estimation under (i)-(iii), we propose a simple complete-case analysis which modifies the regression of primary interest by carefully incorporating the IV to account for selection bias. The approach is developed for the identity, log and logit link functions. For inferences about the marginal mean of a binary outcome assuming (i) and (ii) only, we describe novel and approximately sharp bounds which unlike Robins-Manski bounds, are smooth in model parameters, therefore allowing for a straightforward approach to account for uncertainty due to sampling variability. These bounds provide a more honest account of uncertainty and allows one to assess the extent to which a violation of the key identifying condition (iii) might affect inferences. For illustration, the methods are used to account for selection bias induced by HIV testing nonparticipation in the evaluation of HIV prevalence in the Zambian Demographic and Health Surveys.

Point-of-Care Cepheid Xpert HIV-1 Viral Load Test in Rural African Communities Is Feasible and Reliable.

Routine monitoring of HIV-1 RNA or viral load (VL) in patients on antiretroviral therapy (ART) is important, but there are multiple impediments to VL testing in resource-constrained settings. An accurate point-of-care (POC) HIV-1 VL test could alleviate many of these challenges. We compared the performance of the Cepheid Xpert HIV-1 VL assay against the laboratory-based Abbott m2000sp/m2000rt assay (Abbott assay). ART-naive individuals participating in the Botswana Combination Prevention Project in 20 communities provided EDTA-blood specimens during household surveys. Both the POC Xpert HIV-1 VL and Abbott assays were performed on specimens sampled from 277 individuals. We found a high correlation between the Xpert HIV-1 VL and Abbott assay results (r2 = 0.92; P < 0.001). The overall mean difference in the HIV-1 RNA values obtained by Xpert HIV-1 VL assay and Abbott assay was 0.34 log10 copies/ml (95% confidence interval [CI], 0.26 to 0.40 log10 copies/ml) (P < 0.001). Using a clinically relevant level of 1,000 copies/ml as a threshold, agreement was 90.6% (95% CI, 87.9 to 93.1%), with a sensitivity of 98.6% (95% CI, 97.2 to 100%). The two methods agreed on their detectability of HIV-1 RNA (>40 copies/ml) at 97.1% (95% CI, 95.5 to 98.7%), with a sensitivity of 99.6% (95% CI, 97.2 to 100%). The POC Cepheid Xpert HIV-1 VL assay showed high agreement and accuracy with a laboratory-based method of HIV-1 RNA testing. The POC Xpert HIV-1 VL assay tended to overestimate HIV-1 VL, although the difference was below a clinically relevant threshold of 0.5 log10 copies/ml. The POC Cepheid Xpert HIV-1 VL assay is a promising tool for monitoring patients on ART in southern Africa.

Triggering the decision to undergo medical male circumcision: a qualitative study of adult men in Botswana.

In 2007, the World Health Organization endorsed voluntary medical male circumcision (VMMC) as part of comprehensive HIV-prevention strategies. A major challenge facing VMMC programs in sub-Saharan Africa remains demand creation; there is urgent need for data on key elements needed to trigger the decision among eligible men to seek VMMC. Using qualitative methods, we sought to better understand the circumcision decision-making process in Botswana related to VMMC. From July to November 2013, we conducted 27 focus group discussions in four purposively selected communities in Botswana with men (stratified by circumcision status and age), women (stratified by age) and community leaders. All discussions were facilitated by a trained same-sex interviewer, audio recorded, transcribed and translated to English, and analyzed for key themes using an inductive content analytic approach. Improved hygiene was frequently cited as a major benefit of circumcision and many participants believed that cleanliness was directly responsible for the protective effect of VMMC on HIV infection. While protection against HIV was frequently noted as a benefit of VMMC, the data indicate that increased sexual pleasure and perceived attractiveness, not fear of HIV infection, was an underlying reason why men sought VMMC. Data from this qualitative study suggest that more immediate benefits of VMMC, such as improved hygiene and sexual pleasure, play a larger role in the circumcision decision compared with protection from potential HIV infection. These findings have immediate implications for targeted demand creation and mobilization activities for increasing uptake of VMMC among adult men in Botswana.

A composite likelihood approach for estimating HIV prevalence in the presence of spatial variation.

Since 1990, the World Health Organization has recommended HIV surveillance among pregnant women as an essential surveillance activity for countries with generalized HIV epidemics. Despite the widespread availability and potential usefulness of antenatal HIV surveillance, analyses of such data present important challenges. Within an individual clinic, the HIV status of its attendees may be correlated because of similarities in HIV risk among women close in age. Between-clinic correlation may also arise as women often seek antenatal care at clinics located close to their home, and individuals living in nearby communities may share important characteristics or behaviours related to susceptibility. A general estimating equation-based approach for spatially-correlated, binary data such as that antenatal HIV surveillance based on a pairwise composite likelihood has been described. We present an extended version of this model that can accommodate penalized spline estimators and apply it to antenatal HIV surveillance data collected in 2011 in Botswana to estimate the effects of proximity to the 'hotspot' of the country's HIV epidemic and age on HIV prevalence. Finally, we compare the results with a logistic regression analysis, which ignores potential correlation of responses.

Accounting for selection bias in association studies with complex survey data.

Obtaining representative information from hidden and hard-to-reach populations is fundamental to describe the epidemiology of many sexually transmitted diseases, including HIV. Unfortunately, simple random sampling is impractical in these settings, as no registry of names exists from which to sample the population at random. However, complex sampling designs can be used, as members of these populations tend to congregate at known locations, which can be enumerated and sampled at random. For example, female sex workers may be found at brothels and street corners, whereas injection drug users often come together at shooting galleries. Despite the logistical appeal, complex sampling schemes lead to unequal probabilities of selection, and failure to account for this differential selection can result in biased estimates of population averages and relative risks. However, standard techniques to account for selection can lead to substantial losses in efficiency. Consequently, researchers implement a variety of strategies in an effort to balance validity and efficiency. Some researchers fully or partially account for the survey design, whereas others do nothing and treat the sample as a realization of the population of interest. We use directed acyclic graphs to show how certain survey sampling designs, combined with subject-matter considerations unique to individual exposure-outcome associations, can induce selection bias. Finally, we present a novel yet simple maximum likelihood approach for analyzing complex survey data; this approach optimizes statistical efficiency at no cost to validity. We use simulated data to illustrate this method and compare it with other analytic techniques.

How does sex trafficking increase the risk of HIV Infection? An observational study from Southern India.

Studies have documented the substantial risk of human immunodeficiency virus (HIV) infection endured by sex-trafficked women, but it remains unclear how exposure to trafficking puts its victims at risk. We assessed whether the association between sex trafficking and HIV could be explained by self-reported forced prostitution or young age at entry into prostitution using cross-sectional data collected from 1,814 adult female sex workers in Karnataka, India, between August 2005 and August 2006. Marginal structural logistic regression was used to estimate adjusted odds ratios for HIV infection. Overall, 372 (21%) women met 1 or both criteria used to define sex trafficking: 278 (16%) began sex work before age 18 years, and 107 (5%) reported being forcibly prostituted. Thirteen (0.7%) met both criteria. Forcibly prostituted women were more likely to be HIV-infected than were women who joined the industry voluntarily, independent of age at entering prostitution (odds ratio = 2.30, 95% confidence interval: 1.08, 4.90). Conversely, after adjustment for forced prostitution and other confounders, no association between age at entry into prostitution and HIV was observed. The association between forced prostitution and HIV infection became stronger in the presence of sexual violence (odds ratio = 11.13, 95% confidence interval: 2.41, 51.40). These findings indicate that forced prostitution coupled with sexual violence probably explains the association between sex trafficking and HIV.

Microbiology and risk factors for central line-associated bloodstream infections among pediatric oncology outpatients: a single institution experience of 41 cases.

BACKGROUND: Risk factors for central line-associated bloodstream infections (CLABSI) among children with cancer in the outpatient setting remain poorly defined, and the microbiology may differ from hospital-onset CLABSI. MATERIALS AND METHODS: We conducted a matched case-control study of oncology patients followed at the Dana Farber/Children's Hospital Cancer Center. Cases (N=41) were patients with CLABSI as per National Healthcare Safety Network criteria who had not been hospitalized in the preceding 48 hours. For each case we randomly selected 2 oncology outpatients with a central venous catheter and a clinic visit within 30 days of the case subject's CLABSI. Multivariate conditional logistic regression models were used to identify independent risk factors for CLABSI. We compared the microbiology to that of 54 hospital-onset CLABSI occurring at our institution during the study period. RESULTS: Independent predictors of community-onset CLABSI included neutropenia in the prior week (odds ratio 17.46; 95% confidence interval, 4.71-64.67) and tunneled externalized catheter (vs. implantable port; odds ratio 10.30; 95% confidence interval, 2.42-43.95). Nonenteric gram-negative bacteria were more frequently isolated from CLABSI occurring among outpatients. DISCUSSION: Pediatric oncology outpatients with recent neutropenia or tunneled externalized catheters are at increased risk of CLABSI. The microbiology of community-onset CLABSI differs from hospital-onset CLABSI.

Prevalence and control of hypertension in a high HIV-prevalence setting, insights from a population based study in Botswana.

In a population-based representative sample of adults residing in 22 communities in Botswana, a southern African country with high HIV prevalence, 1 in 4 individuals had high blood pressure. High blood pressure was less prevalent in adults with HIV than without HIV. Sixty percent of persons with high blood pressure had not previously been diagnosed. Among individuals with a prior diagnosis of high blood pressure who reported being prescribed anti-hypertension medications, almost half had elevated blood pressure, irrespective of HIV-status. One-third of adults in this setting (mainly men) declined free non-invasive blood pressure assessments in their households. In conclusion, our study highlights alarmingly high hypertension rates in the community, with low levels of awareness and control, emphasizing the urgent need for community level BP screening and active management to reach recommended targets.

Health Impact and Cost-Effectiveness of HIV Testing, Linkage, and Early Antiretroviral Treatment in the Botswana Combination Prevention Project.

BACKGROUND: The Botswana Combination Prevention Project tested the impact of combination prevention (CP) on HIV incidence in a community-randomized trial. Each trial arm had ∼55,000 people, 26% HIV prevalence, and 72% baseline ART coverage. Results showed intensive testing and linkage campaigns, expanded antiretroviral treatment (ART), and voluntary male medical circumcision referrals increased coverage and decreased incidence over ∼29 months of follow-up. We projected lifetime clinical impact and cost-effectiveness of CP in this population. SETTING: Rural and periurban communities in Botswana. METHODS: We used the Cost-Effectiveness of Preventing AIDS Complications model to estimate lifetime health impact and cost of (1) earlier ART initiation and (2) averting an HIV infection, which we applied to incremental ART initiations and averted infections calculated from trial data. We determined the incremental cost-effectiveness ratio [US$/quality-adjusted life-years (QALY)] for CP vs. standard of care. RESULTS: In CP, 1418 additional people with HIV initiated ART and an additional 304 infections were averted. For each additional person started on ART, life expectancy increased 0.90 QALYs and care costs increased by $869. For each infection averted, life expectancy increased 2.43 QALYs with $9200 in care costs saved. With CP, an additional $1.7 million were spent on prevention and $1.2 million on earlier treatment. These costs were mostly offset by decreased care costs from averted infections, resulting in an incremental cost-effectiveness ratio of $79 per QALY. CONCLUSIONS: Enhanced HIV testing, linkage, and early ART initiation improve life expectancy, reduce transmission, and can be cost-effective or cost-saving in settings like Botswana.

Adjustment for missing data in complex surveys using doubly robust estimation: application to commercial sexual contact among Indian men.

BACKGROUND: The Demographic and Health Survey program routinely collects nationally representative information on HIV-related risk behaviors in many countries, using face-to-face interviews and a complex sampling scheme. If respondents skip questions about behaviors perceived as socially undesirable, such interviews may introduce bias. We sought to implement a doubly robust estimator to correct for dependent missing data in this context. METHODS: We applied 3 methods of adjustment for nonresponse on self-reported commercial sexual contact data from the 2005-2006 India Demographic Health Survey to estimate the prevalence of sexual contact between sexually active men and female sex workers. These methods were inverse-probability weighted regression, outcome regression, and doubly robust estimation-a recently-described approach that is more robust to model misspecification. RESULTS: Compared with an unadjusted prevalence of 0.9% for commercial sexual contact prevalence (95% confidence interval = 0.8%-1.0%), adjustment for nonresponse using doubly robust estimation yielded a prevalence of 1.1% (1.0%-1.2%). We found similar estimates with adjustment by outcome regression and inverse-probability weighting. Marital status was strongly associated with item nonresponse, and correction for nonresponse led to a nearly 80% increase in the prevalence of commercial sexual contact among unmarried men (from 6.9% to 12.1%-12.4%). CONCLUSIONS: Failure to correct for nonresponse produced a bias in self-reported commercial sexual contact. To facilitate the application of these methods (including the doubly robust estimator) to complex survey data settings, we provide analytical variance estimators and the corresponding SAS and MATLAB code. These variance estimators remain valid regardless of whether the modeling assumptions are correct.