Why hiv prevalent in africa

Approximately one-third A similarly large proportion of people living with HIV is distributed throughout the larger number of grid cells that have more moderate spatial concentrations of people living with HIV: This increase was due to a corresponding increase in population, as prevalence in sub-Saharan Africa as a whole declined over this same period, from 5. The increase in people living with HIV was larger in locations with high spatial concentrations of people with HIV compared to those with fewer people living with HIV: in , the total number of people with HIV in grid cells in which there are estimated to be fewer than people with HIV was nearly identical 6.

However, the number of people living with HIV in grid cells in which there are estimated to be more than 1, people increased by This study provides a comprehensive quantification of subnational trends in HIV prevalence and the number of people living with HIV in sub-Saharan Africa. These estimates highlight substantial differences between and within countries in levels and trends in HIV prevalence and the spatial concentration of people living with HIV.

For discussion of the advantages of this analysis compared to earlier analyses, important limitations of the present analysis and potential future directions, see Supplementary Discussion. Subnational estimates of HIV prevalence can be used to more efficiently target resources and interventions.

Estimates of the prevalence of HIV and the number of people living with HIV at local levels provide important information about the number of people who are potentially in need of diagnosis and treatment services.

Additionally, in the absence of local information on HIV incidence, information about HIV prevalence can be used to target primary prevention strategies: modelling studies that compare geographically targeted to non-geographically targeted prevention strategies have found that geographically targeted strategies are more efficient in preventing new HIV infections under the same budgetary constraints 11 , Our analysis highlights several challenges to bringing HIV infection under control in Africa.

Growing population size coupled with continued high incidence 1 , 4 of new HIV infections and increased life expectancy among people living with HIV 31 , 32 , 33 , 34 has led to an increase in the number of people living with HIV in sub-Saharan Africa since Despite this increase, spending on HIV in sub-Saharan Africa has declined in recent years, largely as a result of a reduction in development assistance for health 9.

Our estimates also highlight the diversity of the HIV epidemic: although a large number of people living with HIV are concentrated in a few select areas Fig. The most effective treatment and prevention strategies probably differ between areas in which many people live with HIV and those with a smaller number of people living with HIV, and economies of scale may be harder to realize in the latter case.

Nonetheless, it is essential to ensure that people living with HIV have access to appropriate health services regardless of their location. The results of this analysis describe a multifaceted picture of patterns of changing HIV prevalence across sub-Saharan Africa, with many areas experiencing increases over the same period in which other areas experienced declines.

Changes in HIV prevalence are the outcome of a complex interaction between incidence, mortality and migration patterns. Globally, the large-scale expansion of ART coverage has reduced mortality among people living with HIV, offsetting declines in incidence and resulting in an overall increase in HIV prevalence since 1 , 4 , At the region and country levels, trends in mortality and incidence have varied, which has resulted in differing trends in the prevalence of HIV 1 , 4 , Exploration of this dynamic at a subnational level is warranted, although it is complicated by the relative lack of directly observed empirical data on HIV incidence and mortality in sub-Saharan Africa Nonetheless, existing evidence indicates that subnational increases in prevalence should not be interpreted as inherently alarming without additional consideration of incidence and mortality trends.

Despite progress in recent decades, HIV continues to impose a substantial health burden on countries in sub-Saharan Africa. The estimates from this analysis highlight the degree to which the effect of this epidemic varies, even within countries.

These local data provide a new tool for policymakers, programme implementers and researchers to use to assess local needs, efficiently target interventions and ultimately work towards bringing HIV infection under control in Africa. No statistical methods were used to predetermine sample size.

The experiments were not randomized and the investigators were not blinded to allocation during experiments and outcome assessment. The period of — and the age group of 15—49 years were selected to optimize the contemporaneousness of the estimates and to maximize data availability—there were relatively few large-scale seroprevalence surveys conducted before , and most seroprevalence surveys focus on adults, in which 15—49 years was the most commonly reported age range.

The methodology used here is similar to that used for previous analyses of mortality in children under 5 years of age 41 , child growth failure 42 and education 43 in Africa. Extended Data Figure 5 provides an overview of the analytic process.

Each step is described below and additional details are available in the Supplementary Information , including a discussion of the limitations of this approach.

We compiled a dataset of 29, data points from seroprevalence surveys in 41 countries and 9, data points from sentinel surveillance of antenatal care clinics ANC data in 46 countries. Data from seroprevalence surveys were originally in one of three forms: survey microdata that is, individual-level survey responses , survey reports or published literature Supplementary Table 2.

For surveys with available microdata, we extracted variables related to age, HIV blood test result, location and survey weights. After subsetting the data to ages 15—49 years and excluding rows with missing information on any of these variables, we collapsed the data by calculating the weighted HIV prevalence at the finest spatial resolution available.

Ideally, this was at the level of the GPS coordinates that represent the location of a survey cluster, but in instances for which GPS data were not available, the smallest areal unit termed a polygon possible was used instead, typically representing an administrative subdivision.

For surveys for which microdata were unavailable but for which estimates with some subnational resolution were provided in a report or published literature, we extracted these estimates along with information about the sample size and location. Where possible, these data were matched to a specific set of GPS coordinates, and otherwise were matched to a polygon, which most-often represented an administrative subdivision.

In some instances, estimates extracted from reports or published literature were for age groups other than 15—49 years 34 sources representing 1.

In these instances, we used a cross-walking model—that is, an approach for linking disparate data sources in this case data sources reporting for different age groups —that leveraged existing microdata and linear regression to translate the prevalence in the reported age range to the standard 15—49 age range Supplementary Information, section 2.

In both instances, we extracted information on HIV prevalence and sample size by site and year. Sites were geolocated to specific GPS coordinates where possible and otherwise to a polygon that represents an administrative subdivision. In instances in which data were matched to a polygon rather than specific GPS coordinates, we resampled these data to mimic point data.

Specifically, for each observation, we randomly sampled 10, candidate locations within the associated polygon with a probability proportional to the population and then used k -means clustering to generate a reduced set of locations based on the centroid of each k -means cluster. Each of these resulting pseudo-points was assigned the HIV prevalence observed for the polygon as a whole, and the sample size was set to the observed sample size for the polygon as a whole multiplied by the fraction of candidate locations that belonged to that k -means cluster.

Weighting by sample size, This analysis included five pre-existing covariates: 1 travel time to the nearest settlement of more than 50, inhabitants; 2 total population; 3 night-time lights; 4 urbanicity; and 5 malaria incidence Supplementary Table 5. These eight covariates were constructed based on survey data collected and analysed analogously to the HIV data described above , and using geostatistical models similar to those described in the next section Supplementary Table 6 and Supplementary Figs.

In addition, calendar year was used as a covariate. An ensemble covariate modelling approach was implemented to capture possible nonlinear effects and complex interactions among these covariates For each modelling region Extended Data Fig. Each sub-model was fitted using fivefold cross-validation to avoid overfitting, and the out-of-sample predictions from across the five folds were compiled into a single set of predictions that were used to fit the geostatistical model described below.

In addition, each sub-model was also fitted to the full dataset to generate a complete set of in-sample predictions that were subsequently used when generating predictions from the geostatistical model Supplementary Figs.

We modelled HIV prevalence using a spatially and temporally explicit generalized linear mixed effects model:. We modelled the number of HIV-positive individuals Y i , t among a sample N i , t in location i and year t as a binomial variable. HIV prevalence as measured by sentinel surveillance of antenatal care clinics is known to be biased as a measure of HIV prevalence in the general adult population, because it only covers pregnant women who attend ANC, compared to all adult men and women 46 , Sensitivity analyses were carried out to assess sensitivity to hyper-prior specification and are described in detail in the Supplementary Information, section 4.

This model was fitted in R-INLA 48 using the stochastic partial differential equation 49 approach to approximate the continuous spatial and spatio-temporal Gaussian random fields U i and Z i , t , respectively.

Owing to computational constraints, and to allow for regional differences in the relationship between the covariates and HIV prevalence, as well as differences in the temporal and spatial autocorrelation in HIV prevalence, separate models were fitted for each of the four regions Extended Data Fig.

From each fitted model, we generated 1, draws from the approximated joint posterior distribution of all model parameters and used these to construct 1, draws of p i , t , setting I ANC to 0. Fivefold cross-validation was used to assess model performance and to compare among a number of alternative models that use covariates, ANC data and polygon data in a variety of ways Supplementary Figs.

To take advantage of the more structured modelling approach and additional national-level data used by GBD , we performed post hoc calibration of our estimates to the corresponding national-level GBD estimates 1.

For each country and year in our analysis, we defined a raking factor equal to the ratio of the GBD estimate for this country and year to the population-weighted posterior mean HIV prevalence in all grid cells within this country and year Supplementary Fig.

These raking factors were then used to scale each draw of HIV prevalence for each grid cell within that GBD geography and year. Grid cells that crossed international borders within modelling regions were fractionally allocated to multiple countries in proportion to the covered area during this process. This process was carried out for each of the 1, posterior draws after calibration to GBD with final point estimates derived from the mean of these draws and uncertainty intervals from the 2.

Additionally, estimates of the number of people living with HIV for each grid cell were derived by multiplying estimated prevalence in each grid cell by the corresponding population estimate from WorldPop 25 , 26 , which was also calibrated to match GBD 50 Supplementary Information, section 4. As with calibration, grid cells that crossed borders were fractionally allocated to multiple areas when calculating aggregated prevalence estimates and estimates of people living with HIV.

This analysis is subject to several limitations further discussed in the Supplementary Information, section 5. Most importantly, the accuracy of our estimates is dependent on the quantity and quality of the underlying data.

We have constructed a large database of geolocated HIV prevalence data for the purposes of this analysis. Nonetheless, important gaps in data coverage, both spatial and temporal, remain Extended Data Figs.

Data quality is also likely to be variable and may be problematic for some data sources or locations. For HIV seroprevalence surveys, potential non-response bias is a particular concern 51 and the quality of the underlying data that are used to generate the covariate surfaces may also be suboptimal in some situations—for example, if cultural context influences the interpretation of a survey question or the response to potentially sensitive questions regarding sexual behaviour The information on locations that is associated with the data used in this analysis is also subject to some error and uncertainty.

The numbers are daunting. Adult HIV prevalence is 1. Five million adults and children became newly infected with HIV in , 3. Three million people died from AIDS-related causes in , and 2. Worldwide, AIDS is the fourth leading cause of death.

Life expectancy at birth has plummeted in many African countries, wiping out the gains made since independence. These statistics disguise an important part of the story, however. This belt consists of about 16 countries 3 and stretches from Djibouti and Ethiopia down the east side of the continent through South Africa. According to UNAIDS, all the worst affected countries with prevalence rates over 20 percent are contiguous to one another in the lower part of the continent.

Botswana, Lesotho, Swaziland, and Zimbabwe have prevalence rates above 30 percent. Adult prevalence in Uganda is estimated to be around 5 percent. Yet, the HIV epidemic is not being treated like a crisis.

We were alarmed by the complacency toward the rate of new infections at all levels and the absence of an emergency response, especially for young people. This is no time for business as usual from South Africa or its partners, including the U. The epidemic is exacerbated by its concentration in year-olds, those of reproductive and working age who are the backbone of South Africa.

Without aggressive action to reduce the rate of new infections in young people, HIV will continue to take a tremendous toll on the country for years and generations to come. Collective action is needed to push beyond the complacency and internal barriers to implement policies and interventions that directly target HIV prevention and treatment for young people.

Of the estimated 7. HIV prevalence in other key populations—female sex workers, men who have sex with men, transgender women, and people who inject drugs—remains unacceptably high, in some cases double the national prevalence rate of approximately 19 percent. After the early years of denial, the South African government now finances close to 80 percent of the HIV response, an unparalleled commitment in sub-Saharan Africa, and provides more than 4 million people with life-prolonging anti-retroviral treatment ART.

New infections in young men and women remain alarmingly high nearly 87 percent of the total and viral suppression rates, a key to preventing those living with the virus from passing it on, are under 50 percent for those years old.

With approximately 45 percent of the population under the age of 25 , the sheer numbers of those becoming infected and overall prevalence of HIV will stay alarmingly high without a massive decline in the new HIV infection rate. The central question is how to interrupt HIV transmission in young adults, and where and whom to target. The reasons are both biological and social, including high rates of teenage pregnancy, an epidemic of gender-based and interpersonal violence; lack of quality education; and widespread poverty and unemployment.

High rates of sexually transmitted infections STIs increase the risk of HIV acquisition, and mental health issues can lead to risky behaviors. We need to close the tap. Addressing the range of social, economic, and health issues that put AGYW at risk is one approach. However, tracking the layering of those services has proven to be a challenge, as has widespread scale up.

Another important approach is to reach young men. Prevalence among year-old women is three times higher than in men their age. The challenge is reaching the men. Several countries with a high burden of HIV infection are also progressing along the path to elimination. By the end of , Main navigation Home Health topics All topics ».

Communicable Diseases. Ebola Virus Disease. Traditional Medicine. Tuberculosis TB. Countries All Countries ». News from countries. WHO delivers 6. How a commercial driver is contributing to prevent the spread of Covid in Kumasi in the Ashanti Region of Ghana. WHO emergency teams learning and supporting in Malawi. Accountability » Financial reports General Programme of Work. The Transformation Agenda ».

Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection In addition, one in three people living with HIV presents to care with advanced disease, low CD4 count and at high risk of serious illness and death.

Key fact 1. Featured video. Featured Asset Download the full infographic. Disease burden. There were approximately In , about 1. Most often these tests provide same-day test results, which are essential for same-day diagnosis and early treatment and care.

There is no cure for HIV infection. However, effective antiretroviral ARV drugs can control the virus and help prevent transmission so that people with HIV and those at substantial risk, can enjoy healthy, long and productive lives.