Secondary Logo

Journal Logo

Epidemiology and Social

Towards UNAIDS Fast-Track goals

targeting priority geographic areas for HIV prevention and care in Zimbabwe

Cuadros, Diego F.a,b; Li, Jingjinga; Mukandavire, Zindogac; Musuka, Godfrey N.d; Branscum, Adam J.e; Sartorius, Bennf; Mugurungi, Oweng; Tanser, Frankf,h

Author Information
doi: 10.1097/QAD.0000000000002052

Abstract

Introduction

The Joint United Nations Programme on HIV/AIDS (UNAIDS) under its ambitious 2020 targets of 90-90-90 [90% of all people living with HIV (PLHIV) know their HIV status, 90% of all people with diagnosed HIV infection receive sustained antiretroviral therapy (ART) and 90% of all people receiving ART have viral suppression] [1], which has been escalated to 95-95-95 by 2030, has paved the way for more structured and effective responses to the HIV epidemic, particularly in South-Saharan Africa (SSA). Zimbabwe, a country that had one of the worst HIV epidemics in the world [2,3], is working to achieve the 90-90-90 goals. Supported by rapid and effective mobilization against the epidemic, Zimbabwe has been able to halve its HIV prevalence over the last 15 years [3]. Recently, the Zimbabwe Population-Based HIV Impact Assessment (ZIMPHIA) survey showed that the country is moving rapidly towards the 90-90-90 goals for 2020, with 73% of PLHIV diagnosed, 87% of those diagnosed on ART and 86% of those on ART virally suppressed [4].

Despite notable progress made by Zimbabwe, HIV prevalence in the country remains high, with about 14% of adults aged 15–49 years infected with the virus [5] and more than 33 000 new infections occurring every year [4]. This complex epidemiological context of the HIV epidemic, coupled with flatlining of financial resources from international donor organizations and limited domestic capacity to fund its own response represent a challenge for achieving and maintaining the UNAIDS 2020 goals, and the even more ambitious 95-95-95 UNAIDS targets by 2030. Therefore, enhanced targeting of spending and allocative efficiency of limited resources through novel methodological and strategic approaches would facilitate reaching these UNAIDS Fast-Track targets.

Recent research has consistently demonstrated substantial spatial variation in the HIV epidemic at different geographical scales, including at the province [6], district [7,8], subdistrict and more localized levels [9–18] in several SSA countries (including Zimbabwe). The HIV landscape in Zimbabwe is not homogeneous, with the infection clustering in the southern part of the country, between the provinces of Bulawayo, Matabeleland North and Matabeleland South [19,20]. These provinces also had lower ART coverage than provinces located in areas with low HIV prevalence (e.g. Mashonaland West, Central and Eats and Masvingo) [4]. Studies focused on localized areas in Zimbabwe have also found substantial variation of the HIV epidemic. For example, east Zimbabwe had spatial clustering of HIV infection in urban areas around the small towns of Nyazura and Nyanga, as well as inequalities in accessibility and uptake of HIV services in these areas [21].

Identifying areas where the burden of HIV infection is concentrated will help target prevention and treatment interventions to populations at a higher risk of infection, and knowledge about high-risk areas is prerequisite to implementing successful surveillance programmes and optimally allocating resources [13,22–26]. However, most measures of HIV occurrence are frequently available only for large geographical administrative units such as districts or provinces, thereby obscuring important localized aspects of the HIV transmission process. Given the limited and declining resources devoted to HIV prevention, treatment, care and control as a result of donor fatigue and shifting donor priorities over time, it is crucial that intervention efforts are directed to vulnerable high-risk populations in local geographical areas of greatest HIV burden [27,28]. Microlevel data on the spatial distribution of HIV prevalence could help prevent new HIV infections and increase treatment rates by prioritizing resources targeting localized high burden areas and aligning service delivery modalities to the specific needs of the local communities on these areas.

Given this context, the main objective of this study was to construct high-resolution maps of intranational estimates of HIV measures (prevalence and density of HIV-infected population) in Zimbabwe based on spatial variables. In addition, a detailed spatial analysis of the current stage of the 90-90-90 indicators in Zimbabwe was conducted. This information is aimed to assist the country to strategically target localized geographic areas and populations in order to maximize programme outputs and boosting programme efficiency to achieve the UNAIDS targets in Zimbabwe.

Materials and methods

Study area and data sources

The study area was Zimbabwe, a landlocked country in southern Africa. Zimbabwe has the sixth highest HIV prevalence in SSA, with approximately 1.2 million persons aged 15–64 years living with HIV in 2016 [4]. The main source of data for this study was the Zimbabwe Demographic and Health Survey (ZDHS) conducted in 2015 [5]. Individuals were enrolled in the ZDHS via a two-stage sampling procedure to select households. A total of 400 ZDHS sample locations were selected. The global positioning system was used to identify and record the geographical coordinates of each ZDHS sample location (Fig. 1) [29]. The study population was limited to adults aged 15–49 years, resulting in a total of 9054 women and 7082 men included in the study. Anonymous HIV testing was performed with the informed consent of all sampled individuals. Further details related to the ZDHS methodology, study design and data can be found elsewhere [5,30–32].

Fig. 1
Fig. 1:
The overall study area (top left) and Zimbabwe Demographic and Health Survey sample locations.Maps were created using ArcGIS by ESRI version 10.3 (http://www.esri.com) [29].

Population density throughout Zimbabwe was obtained from the WorldPop database [33] (Figure S1). Estimates for additional spatial variables, including the percentage of PLHIV who were diagnosed, the percentage of people diagnosed with HIV who were treated by ART and the percentage on ART who were virally suppressed, were obtained at provincial scale from the 2016 ZIMPHIA report [4]. Lastly, estimated number of new HIV infections and the average number of clients per ART treatment site at the province level were obtained from the report on Smart Investment to End HIV AIDS in Zimbabwe based on Hotspot Analysis [34].

Definitions of variables

To assess the association between socioeconomic and behavioural factors and the prevalence of HIV infection, we selected a variety of demographic, socioeconomic and behavioural attributes from the ZDHS dataset and other sources. Fifteen variables measured for men and 14 for women included wealth index, number of children under five years old in household (women only), male circumcision (men only), union status, number of unions, currently reside with husband/wife/partners, number of wives (men only), number of husband's other wives (women only), age at first sex, employment, condom use, number of sexual partners including spouse, ever been tested for HIV and total number of lifetime sexual partners. Environmental variables included the normalized difference vegetation index (NDVI), distance to primary roads, distance to main cities and population density (Supplementary Figure S1; http://links.lww.com/QAD/B393). The definition and descriptive statistics for each variable are presented in Supplementary Materials; http://links.lww.com/QAD/B393.

Selection of spatial variables and mapping

To map the geographical distribution of HIV prevalence in Zimbabwe as well as the geographic dispersion of HIV-infected men and women using spatial variables, we implemented a method previously developed to describe the spatial structure of HIV prevalence in several SSA countries [35]. Further details related to this methodology can be found in Supplementary Materials; http://links.lww.com/QAD/B393.

Spatial distribution of the 90-90-90 indicators

The spatial distributions of HIV incidence in 2013 and the average number of clients per ART treatment site were estimated using data from the report on Smart Investment to End HIV AIDS in Zimbabwe Based on the Hotspot Analysis [34]. Data at the provincial levels on the percentage of PLHIV who have been diagnosed, the percentage diagnosed with HIV who were receiving ART and the percentage of those receiving ART who were virally suppressed were generated by combining estimates from the 2016 ZIMPHIA report [4] with the geographical densities of HIV-infected men and women obtained in this study to determine the spatial distributions of the number of men and women who were lacking ART coverage and viral suppression in Zimbabwe in 2015.

Further details related to the methodology, study design and data can be found in Supplementary Materials; http://links.lww.com/QAD/B393.

Results

Spatial variable selection

Supplementary Table S2 presents the results of the bivariate logistic regression and spatial correlation analysis (http://links.lww.com/QAD/B393). The variables selected in the final logistic regression models for men and women are summarized in Table 1. For men, the final logistic model included currently in union, condom use, number of lifetime sexual partners and population density. For women, the selected variables were poverty, number of unions, age at first sex below 20 years old, condom use, number of lifetime sexual partners and population density. Maps of the variables selected for women (Figure S2) and men (Figure S3) are included in Supplementary Materials; http://links.lww.com/QAD/B393.

Table 1
Table 1:
Variables in the final logistic regression models for men and women.

HIV prevalence distribution

The spatial distributions of HIV prevalence estimated using the final multivariable logistic regression models for women and men are presented in Figure 2a and d, respectively. These maps illustrate the distribution of the proportion of individuals who are HIV-positive among the entire adult population, with the highest proportion of the HIV-infected population (prevalence) located in the south-western part of the country. The estimated (pixel-level) HIV prevalence distribution for women ranged from 10.6 to 33.62%, with a median of 17.8%. For men, the estimated (pixel-level) HIV prevalence distribution ranged from 6.9 to 21.3%, with a median of 10.7%.

Fig. 2
Fig. 2:
High resolution maps of HIV indicators in Zimbabwe.High resolution maps of HIV prevalence in Zimbabwe for (a) women and (d) men in 2015; geographic dispersion of HIV-infected (b) women and (e) men in Zimbabwe. High HIV burden areas are illustrated in red for (c) women and (f) men. Maps were created using ArcGIS by ESRI version 10.3 (http://www.esri.com) [29].

The geographic distributions of the estimated number of HIV-infected women and men are presented in Figure 2b and e, respectively. These maps illustrate the distribution of the density of HIV-infected individuals in Zimbabwe. Most of the HIV-infected individuals were located in urban settings (Harare, Norton, Chitungwiza, Ruwa and Bulawayo). There were an estimated 625 319 HIV-infected women and 419 964 HIV-infected men aged 15–49 years in 2015. Figure 2c and f illustrate the location of high HIV burden areas for women and men, respectively, where both HIV prevalence and density of HIV-infected individuals were positively correlated. For women, the high burden areas corresponded to only 1.4% of the total area in Zimbabwe but contained 18.8% of the total number of HIV-infected women. Similarly, for men, these areas corresponded to 2.4% of the total area in Zimbabwe but contained 26.5% of the total number of HIV-infected men.

Gaps in antiretroviral therapy coverage

Figure 3a–d illustrates the provincial-level spatial distribution of several epidemiological and treatment and care measures, including estimated HIV incidence in 2013 (Fig. 3a), ART coverage (Fig. 3b) and viral load suppression among HIV-infected individuals receiving ART (Fig. 3c). Figure 3d presents the average number of clients per ART treatment site at the district level. The geographic distributions of the number of HIV-infected women and men aged 15–49 years who were lacking ART coverage and viral suppression in 2015 are presented in Fig. 3e and f, respectively. We estimated that a total of 353 613 women (56.5%) and 237 804 men (56.6%) aged 15–49 years were lacking standard ART coverage and viral suppression in Zimbabwe in 2015 (Table 2).

Fig. 3
Fig. 3:
Maps of HIV services distribution in Zimbabwe.Province-level maps of (a) new HIV infection in 2015, (b) ART coverage and (c) percentage of viral load suppression among all HIV-infected individuals identified in the population-based ZIMPHIA survey. Map in (d) presents estimates of the average number of clients per ART site per district in 2015. Geographical dispersion of men (e) and women (f) lacking ART coverage and not virally suppressed in 2015. Maps were created using ArcGIS by ESRI version 10.3 (http://www.esri.com) [29].
Table 2
Table 2:
Pixel-level epidemiological indices estimation for each sex nationally in Zimbabwe and in its high HIV burden areas.

Discussion

The study identified geographical variation of HIV prevalence for both men and women in Zimbabwe. The burden of infection is concentrated in the south-western part of the country, between the provinces of Matabeleland North and South, and the city of Bulawayo, which is consistent with previous studies [19,20]. Similar geographical patterns with high HIV prevalence for males were identified in urban areas, particularly in the eastern part of the country (Mutare and Chipinge). High numbers of HIV-infected men and women occurred in northern and eastern Zimbabwe. The burden of HIV-infected individuals appears to be concentrated in urban settings (Harare, Norton, Chitungwiza, Ruwa and Bulawayo) with an HIV-infected population density higher than 1000 HIV-infected women and men per 5 km2. However, the distribution of HIV-infected women was more dispersed than the distribution for men.

Our results suggest that sexual behaviour could be a key driver of the HIV prevalence distribution in Zimbabwe. Behavioural factors such as high number of lifetime sexual partners (> two lifetime sexual partners), condom use and sexual debut (age at first sex below 20 years in girls) showed a distinct geographical distribution that was associated with the spatial prevalence of HIV infection for both women and men. These results are consistent with previous data from Malawi [36], in which risky sexual behaviour described by high number of lifetime sexual partners explained much of the spatial variation in HIV prevalence. In addition, several other studies have identified sexual risk behaviours such as early age of first sex [37], multiple partnerships [38,39] and long-term concurrent partnerships [40,41] to be linked to HIV infection in Zimbabwe. Moreover, condom use has been found to be associated with a high risk of HIV infection, including in Zimbabwe [4,21,42].

Mobility and migration are common factors linked to the spread of HIV in countries in southern Africa [41,43]. Distance to main roads has been identified as a factor associated with the risk of HIV infection linked to mobile individuals [35,44,45]. However, we did not find a statistically significant association between distance to main roads and the prevalence of HIV infection for either men or women in Zimbabwe. According to our results, areas adjacent to main roads had lower poverty levels than areas distant to main roads. A similar pattern was observed for sexual debut in women, in which female populations living distant to main roads had a higher percentage of women engaging in sexual contact at early ages. Women living in high areas of poverty are a vulnerable population and may engage in risky sexual behaviours such as commercial sex or establish a partnership with older men [46–48]. Therefore, along with ART access and HIV testing, the female population living in these communities could be targeted for combination interventions, including preexposure prophylaxis (PrEP) and behavioural interventions such as condom distribution, prevention counselling, family planning and mass media programmes. Moreover, these poorer, hard-to-reach communities could benefit from the implementation and delivery of mobile units for HIV services, which have been demonstrated to be effective in delivering and enhancing HIV services uptake, particularly in isolated areas [49,50].

For men, we identified geographical hotspots (high numbers of HIV-infected men in areas with high HIV prevalence) in the southern and eastern parts of the country, close to the borders with Botswana, Mozambique and South Africa. These high burden areas for men are linked to commercial transport corridors, increased diamond mining activities and mobility between the neighbouring countries of Botswana, Mozambique and South Africa, among other factors that could be fuelling the burden of HIV infections in these areas [51]. Therefore, other combination interventions apart from HIV testing and counselling and ART, such as PrEP, voluntary medical male circumcision (VMMC) and condom distribution could be provided to the male population living in these areas. The benefits of focusing on this high HIV burden area for men may also extend into lower-prevalence locations, particularly for female partners, given the extensive migration that occurs in these areas.

According to the UNAIDS Global Report in 2015 [2], Zimbabwe is among 13 countries in SSA that has registered a dramatic decline of new HIV infections of at least 50% among adults (15–49 years) between 2001 and 2011. Nevertheless, Zimbabwe still faces a high HIV burden. Despite its push towards UNAIDS goals, more effort is needed to completely achieve the goals and diminish the current HIV epidemic in Zimbabwe. Here, we presented a novel methodological approach to investigate the drivers of the spatial variation of the epidemic and assessed the potential geographical gaps in HIV services and care delivery that could hinder Zimbabwe in meeting UNAIDS targets.

Zimbabwe remains committed to achieving UNAIDS targets and adopted 2013 ART guidelines in order to increase and improve access to services [2]. As a result, service delivery in terms of coverage with ART and prevention of mother-to-child transmission have increased by up to five times between 2004 and 2013 [34]. Despite the substantial effort invested by the government of Zimbabwe and its partners to reach UNAIDS goals, several important gaps could prevent it from successfully attaining and maintaining the desired targets by 2030. Although there are small geographical differences in the geographical distribution of HIV prevention, treatment and care services, the delivery of these services has not been completely aligned in concordance with the spatial heterogeneity of the epidemic. Our analysis indicated that more than 230 000 men and 350 000 women are lacking treatment and care in Zimbabwe. A significant fraction of these individuals (22%) were located in the high burden areas (HIV hotspots) identified, which occupy only about 2% of the total area of Zimbabwe. In fact, ART coverage is slightly lower in the provinces with high HIV prevalence (Matabeleland North and South).

The cities of Harare and Bulawayo, with about 113 000 and 33 000 HIV-infected individuals lacking ART coverage and viral suppression, respectively, might be particularly affected by lacking sufficiently focused HIV services coverage to meet demand. Despite the fact that these provinces received two of the highest budgets allocated for HIV treatment in Zimbabwe in 2012 (Fig. S4C) [34], these regions also have the highest estimated average number of clients per ART site, which could be interpreted as low coverage levels in these provinces. This observation together with the high HIV prevalence and high population density of PLHIV in Harare and Bulawayo, suggest the need to intensify the delivery of HIV services and care among other programme interventions to these areas. Furthermore, HIV spending is the lowest in the areas with the highest HIV prevalence in Zimbabwe, which could reflect failures in the HIV service coverage for the population living in these areas and could compromise the achievement of the UNAIDS targets. However, it is important to note that the number of clients per ART site was taken to be the average number of clients receiving service per site, which was calculated using the number of sites per district and the number of PLHIV within each district (data provided in the Hotspot Analysis report). There are factors that could affect this estimate, such as ART clients who travel and receive services from other health centres beyond their district boundaries. Using the total number of sites and the exact number of people on ART could provide more accurate estimates of HIV services coverage.

Several study limitations could have affected our results. Given the multiple logistical difficulties in conducting the ZDHS, some of the variables included in our study could have been affected by inherent biases in the data, such as variability in response rates to HIV testing and undersampling of mobile individuals and key core groups at risk [52,53]. For instance, some high-risk subpopulations, such as female sex workers, injection drug users, men who have sex with men and mobile individuals, could have been missed by the ZDHS. It is not clear though whether these high-risk subpopulations were undersampled or if it affected our findings. Epidemics among these high-risk subpopulations could lead to onward transmission of the HIV infection among spouses and clients of female sex workers. Moreover, our results could be affected by patterns of migration, information not provided by ZDHS but that could improve the accuracy of spatial estimation. Like the ZDHS, ZIMPHIA was a population-based nationally representative survey, and some of the variables extracted from this sample could have been affected by inherent biases in the data, such as variability in response rates to HIV testing and self-reporting HIV testing and ART treatment. Moreover, data were aggregated to the province level, reducing the spatial resolution. In addition, data from ZIMPHIA were reported for the entire adult population and were not stratified by sex. Also, following the recent WHO ART guidelines recommendations, we assumed that all adults living with HIV regardless of WHO clinical stage and at any CD4+ cell count were eligible for ART [54], and the calculation of the numbers of HIV-infected individuals as well as individuals lacking ART coverage were based in pixel-level estimates using the map surfaces developed. An additional potential bias is the global positioning system displacement process of the DHS sampling data points, used to preserve their confidentiality [55]. This process could have an impact on the precision of HIV estimations, particularly affecting the location of areas with high HIV burden by several kilometres.

Conclusion

We conducted a detailed spatial analysis of the geographical structure of the HIV epidemic in Zimbabwe to include geographical prioritization as a key component for HIV control. Behavioural and socioeconomic factors such as lifetime number of sexual partners, age at first sex and poverty were associated with the spatial distribution of HIV prevalence for both men and women in Zimbabwe. As Zimbabwe rapidly approaches the UNAIDS 90-90-90 targets, saturation capacity for HIV infected individuals receiving ART in some areas underscores the need for precise geographic targeting of interventions. Also, social and behavioural programmes coupled with more robust prevention interventions such as PrEP in areas with high poverty that overlapped with a high fraction of vulnerable women, as well as interventions targeting the male population such as VMMC in areas with intense economic activity and migration such as mining activities are needed in Zimbabwe. Special attention should therefore be given to developing and implementing tailored prevention programmes that take into consideration the degree of exposure risk, especially in areas identified as HIV hotspots and potential hotspots. In the HIV response of a country, a spatial model and high-resolution maps of HIV indices and continuum of care provide valuable information in support of a complex decision process that may also include, among several other elements, budgetary constraints along with political priorities beyond health outcomes. However, the type of geospatial analysis we conducted in this study could be a complementary component of the decision-making process via continual engagement with stakeholders and other important decision-makers. The model results presented in this study provide information aimed at increasing intervention coverage in those areas with the greatest need, in addition to increasing the functionality and capacity of the existing service delivery centres in a country such as Zimbabwe.

Acknowledgements

The authors thank Measure Demographic and Health Surveys (Measure DHS) for releasing these national surveys in the service of science, and the United States Agency for International Development and other donors supporting these initiatives.

D.F.C. contributed the study and its design, conducted the statistical and spatial modelling analyses, and wrote the first draft of the article. J.L., Z.M., G.N.M. and A.B. contributed to study conception and design, conduct of the statistical modelling analyses, interpretation of the results and writing of the manuscript. B.S., F.T. and O.M. contributed to study conception and design, interpretation of the results and writing of the manuscript

Conflicts of interest

F.T. was supported by the South African Medical Research Council (SA MRC) Flagship grant (MRC-RFA-UFSP-01–2013/UKZN HIVEPI) and a UK Academy of Medical Sciences Newton Advanced Fellowship (NA150161). For the remaining authors, none were declared.

References

1. Joint United Nations Programme on HIV/AIDS. UNAIDS 2016–2021 strategy: on the fast-track to end AIDS (Joint United Nations Programme on HIV/AIDS). 2015. http://www.unaids.org/sites/default/files/media_asset/20151027_UNAIDS_PCB37_15_18_EN_rev1.pdf [Accessed 10 August 2017].
2. Global AIDS response progress report 2015. UNAIDS Zimbabwe Country Report. 2015. http://www.unaids.org/sites/default/files/country/documents/ZWE_narrative_report_2015.pdf) [Accessed 15 July 2017].
3. Fraser N. Zimbabwe: analysis of HIV epidemic, response and modes of transmission. Zimbabwe National Aids Council; 2011. https://www.k4health.org/toolkits/zimbabwe-hiv-prevention/zimbabwe-analysis-hiv-epidemic-response-and-mode-transmission [Accessed 15 July 2017].
4. Ministry of Health and Child Care (MOHCC), Zimbabwe. Zimbabwe population-based HIV impact assessment (ZIMPHIA) 2015-16: first report. Harare: MOHCC; 2017. https://phia.icap.columbia.edu/wp-content/uploads/2017/11/ZIMPHIA_First_Report_FINAL.pdf [Accessed 25 November 2017].
5. Zimbabwe National Statistics Agency, ICF International. Zimbabwe Demographic and Health Survey 2015: final report. Zimbabwe National Statistics Agency (ZIMSTAT) and ICF International. Rockville, Maryland; 2016. https://dhsprogram.com/pubs/pdf/FR322/FR322.pdf [Accessed 17 March 2017].
6. Shaikh N, Abdullah F, Lombard CJ, Smit L, Bradshaw D, Makubalo L. Masking through averages-intraprovincial heterogeneity in HIV prevalence within the Western Cape. South Afr Med J 2006; 96:538–543.
7. Kleinschmidt I, Pettifor A, Morris N, MacPhail C, Rees H. Geographic distribution of human immunodeficiency virus in South Africa. Am J Trop Med Hygiene 2007; 77:1163–1169.
8. Coburn BJ, Okano JT, Blower S. Current drivers and geographic patterns of HIV in Lesotho: implications for treatment and prevention in Sub-Saharan Africa. BMC Med 2013; 11:224.
9. Wand H, Whitaker C, Ramjee G. Geoadditive models to assess spatial variation of HIV infections among women in Local communities of Durban, South Africa. Int J Health Geogr 2011; 10:28.
10. Wand H, Ramjee G. Targeting the hotspots: investigating spatial and demographic variations in HIV infection in small communities in South Africa. J Int AIDS Soc 2010; 13:41.
11. Tanser F, Bärnighausen T, Newell M. Identification of localized clusters of high HIV incidence in a widely disseminated rural South African epidemic: a case for targeted intervention strategies. 18th Conference on Retroviruses and Opportunistic Infections (CROI); 27 February–3 March 2011, Boston, Massachusetts.
12. Cuadros DF, Awad SF, Abu-Raddad LJ. Mapping HIV clustering: a strategy for identifying populations at high risk ofHIV infection in sub-Saharan Africa. Int J Health Geogr 2013; 12:28.
13. Fichtenberg CM, Ellen JM. Moving from core groups to risk spaces. Sex Trans Dis 2003; 30:825–826.
14. Hallett T, Anderson S-J, Asante CA, Bartlett N, Bendaud V, Bhatt S, et al. Evaluation of geospatial methods to generate subnational HIV prevalence estimates for local level planning. AIDS 2016; 30:1467–1474.
15. Aral SO, Torrone E, Bernstein K. Geographical targeting to improve progression through the sexually transmitted infection/HIV treatment continua in different populations. Curr Opin HIV AIDS 2015; 10:477–482.
16. Tanser F, Vandormael A, Cuadros D, Phillips AN, de Oliveira T, Tomita A, et al. Effect of population viral load on prospective HIV incidence in a hyperendemic rural African community. Sci Transl Med 2017; 9:pii: eaam8012.
17. Tanser F, Bärnighausen T, Dobra A, Sartorius B. Identifying ‘corridors of HIV transmission’ in a severely affected rural South African population: a case for a shift toward targeted prevention strategies. Int J Epidemiol 2018; 47:537–549.
18. Cuadros DF, Sartorius B, Hall C, Akullian A, Bärnighausen T, Tanser F. Capturing the spatial variability of HIV epidemics in South Africa and Tanzania using routine healthcare facility data. Int J Health Geogr 2018; 17:27.
19. Cuadros DF, Branscum AJ, Mukandavire Z. Temporal stability of HIV prevalence in high-burden areas regardless of declines in national HIV prevalence in Malawi and Zimbabwe. AIDS 2018; 32:1381–1383.
20. Cuadros DF, Abu-Raddad LJ. Spatial variability in HIV prevalence declines in several countries in sub-Saharan Africa. Health Place 2014; 28:45–49.
21. Schaefer R, Gregson S, Takaruza A, Rhead R, Masoka T, Schur N, et al. Spatial patterns of HIV prevalence and service use in East Zimbabwe: implications for future targeting of interventions. J Int AIDS Soc 2017; 20:21409.
22. Wilson D, Halperin DT. Know your epidemic, know your response: a useful approach, if we get it right. Lancet 2008; 372:423–426.
23. Buse K, Dickinson C, Sidibé M. HIV: know your epidemic, act on its politics. J R Soc Med 2008; 101:572–573.
24. Gerberry DJ, Wagner BG, Garcia-Lerma JG, Heneine W, Blower S. Using geospatial modeling to optimize the rollout of antiretroviral-based preexposure HIV interventions in Sub-Saharan Africa. Nat Commun 2014; 5:5454.
25. Barankanira E, Molinari N, Niyongabo T, Laurent C. Spatial analysis of HIV infection and associated individual characteristics in Burundi: indications for effective prevention. BMC Public Health 2016; 16:118.
26. Meyer-Rath G, McGillen JB, Cuadros DF, Hallett TB, Bhatt S, Wabiri N, et al. Targeting the right interventions to the right people and places: the role of geospatial analysis in HIV programme planning. AIDS 2018; 32:957–963.
27. UNAIDS. Practical guidelines for intensifying HIV prevention: towards universal access. 2007. http://data.unaids.org/pub/manual/2007/20070306_prevention_guidelines_towards_universal_access_en.pdf [Accessed 5 September 2017].
28. Controlling the epidemic: delivering on the promise of an AIDS-free generation, PEPFAR 3.0. 2014. https://www.pepfar.gov/documents/organization/234744.pdf [Accessed 5 September 2017].
29. ESRI. ArcGIS 10.x. Redlands, California: ESRI; 2004.
30. Tanzania Commission for AIDS ZAC, National Bureau of Statistics, Office of Chief Government Statistician, ICF international. Tanzania HIV/AIDS and malaria indicator survey 2011-12. Calverton, MD: ICF International; 2013. https://dhsprogram.com/pubs/pdf/AIS11/AIS11.pdf [Accessed 17 March 2017].
31. Tanzania Commission for AIDS ZAC, National Bureau of Statistics, Office of Chief Goverment Statistician, Macro International Inc. Tanzania HIV/AIDS and malaria indicator survey 2007-08. Calverton, MD: Macro International Inc.; 2008. https://dhsprogram.com/pubs/pdf/AIS6/AIS6_05_14_09.pdf [Accessed 17 March 2017].
32. Tanzania Commission for AIDS NBoS, ORC Macro. Tanzania HIV/AIDS indicator survey 2003-04. Calverton, MD: ORC Macro; 2005. https://dhsprogram.com/pubs/pdf/AIS1/AIS1.pdf [Accessed 17 March 2017].
33. Linard C, Gilbert M, Snow RW, Noor AM, Tatem AJ. Population distribution, settlement patterns and accessibility across Africa in 2010. PloS one 2012; 7:e31743.
34. United Nations World Food Programme, Zimbabwe. Smart investment to end HIV AIDS in ZIMBABWE based on Hotspot Analysis. 2015. http://procurement-notices.undp.org/view_file.cfm?doc_id=114244 [Accessed 15 July 2017].
35. Cuadros DF, Li J, Branscum AJ, Akullian A, Jia P, Mziray EN, et al. Mapping the spatial variability of HIV infection in Sub-Saharan Africa: effective information for localized HIV prevention and control. Sci Rep 2017; 7:9093.
36. Palk L, Blower S. Geographic variation in sexual behavior can explain geospatial heterogeneity in the severity of the HIV epidemic in Malawi. BMC Med 2018; 16:22.
37. Pettifor AE, Van Der Straten A, Dunbar MS, Shiboski SC, Padian NS. Early age of first sex: a risk factor for HIV infection among women in Zimbabwe. AIDS 2004; 18:1435–1442.
38. Gregson S, Garnett GP, Nyamukapa CA, Hallett TB, Lewis JJ, Mason PR, et al. HIV decline associated with behavior change in eastern Zimbabwe. Science 2006; 311:664–666.
39. Bassett MT, McFarland WC, Ray S, Mbizvo MT, Machekano R, van de Wijgert JH, et al. Risk factors for HIV infection at enrollment in an urban male factory cohort in Harare, Zimbabwe. J Acquir Immune Defic Syndr 1996; 13:287–293.
40. Goodreau SM, Cassels S, Kasprzyk D, Montaño DE, Greek A, Morris M. Concurrent partnerships, acute infection and HIV epidemic dynamics among young adults in Zimbabwe. AIDS Behav 2012; 16:312–322.
41. Cassels S, Manhart L, Jenness SM, Morris M. Short-term mobility and increased partnership concurrency among men in Zimbabwe. PLoS One 2013; 8:e66342.
42. Gregson S, Nyamukapa CA, Garnett GP, Mason PR, Zhuwau T, Caraël M, et al. Sexual mixing patterns and sex-differentials in teenage exposure to HIV infection in rural Zimbabwe. Lancet 2002; 359:1896–1903.
43. Coffee MP, Garnett GP, Mlilo M, Voeten HA, Chandiwana S, Gregson S. Patterns of movement and risk of HIV infection in rural Zimbabwe. J Infect Dis 2005; 191 (Suppl 1):S159–S167.
44. Tatem AJ, Hemelaar J, Gray RR, Salemi M. Spatial accessibility and the spread of HIV-1 subtypes and recombinants. AIDS 2012; 26:2351–2360.
45. Tanser F, LeSueur D, Solarsh G, Wilkinson D. HIV heterogeneity and proximity of homestead to roads in rural South Africa: an exploration using a geographical information system. Trop Med Int Health 2000; 5:40–46.
46. Lopman B, Lewis J, Nyamukapa C, Mushati P, Chandiwana S, Gregson S. HIV incidence and poverty in Manicaland, Zimbabwe: is HIV becoming a disease of the poor?. AIDS (London, England) 2007; 21 (Suppl 7):S57–S66.
47. Bassett MT, Mhloyi M. Women and AIDS in Zimbabwe: the making of an epidemic. Int J Health Serv 1991; 21:143–156.
48. Schaefer R, Gregson S, Eaton JW, Mugurungi O, Rhead R, Takaruza A, et al. Age-disparate relationships and HIV incidence in adolescent girls and young women: evidence from Zimbabwe. AIDS (London, England) 2017; 31:1461.
49. Suthar AB, Ford N, Bachanas PJ, Wong VJ, Rajan JS, Saltzman AK, et al. Towards universal voluntary HIV testing and counselling: a systematic review and meta-analysis of community-based approaches. PLoS Med 2013; 10:e1001496.
50. Govindasamy D, Ford N, Kranzer K. Risk factors, barriers and facilitators for linkage to antiretroviral therapy care: a systematic review. AIDS 2012; 26:2059–2067.
51. Chimonyo G, Mungure S, Scott P. The social, economical, and environmental implications of diamond mining in Chiadzwa. Center for Research & Development. Open Society Initiative for Southern Africa; 2017. https://www.africaportal.org/publications/the-social-economic-and-environmental-implications-of-diamond-mining-in-chiadzwa/ [Accessed 8 December 2017].
52. Marston M, Harriss K, Slaymaker E. Nonresponse bias in estimates of HIV prevalence due to the mobility of absentees in national population-based surveys: a study of nine national surveys. Sex Transm Infect 2008; 84 (Suppl 1):i71–i77.
53. Mishra V, Barrere B, Hong R, Khan S. Evaluation of bias in HIV seroprevalence estimates from national household surveys. Sex Transm Infect 2008; 84 (Suppl 1):i63–i70.
54. World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection. 2016. 2nd ed. http://apps.who.int/iris/bitstream/handle/10665/208825/9789241549684_eng.pdf;jsessionid=5304E2BF3AE63CF9881301C3DA5E7502?sequence=1 [Accessed 12 August 2018].
55. Burget C, Colston J, Roy T, Zachary B. Geographic displacement procedure and georeferenced data release policy for the demographic and health surveys. DHS spatial analysis reports No. 7. Calverton, Maryland: ICF International; 2013. https://dhsprogram.com/pubs/pdf/SAR7/SAR7.pdf [Accessed 8 December 2017].
Keywords:

antiretroviral therapy coverage; HIV mapping; optimal resource allocation; spatial analysis; UNAIDS 90-90-90 goals; viral suppression

Supplemental Digital Content

Copyright © 2019 Wolters Kluwer Health, Inc.