The establishment and scale-up of HIV antiretroviral therapy (ART) programs in sub-Saharan Africa has been among the most successful global public health efforts. In recent years, ART coverage in Eastern and Southern Africa has nearly doubled, from 24% in 2010 to 54% in 2015, with over 10 million people living with HIV (PLWHIV) currently on ART in this region.1 The resulting health benefits have been very significant, including gains in adult life expectancy of more than a decade.2,3 As PLWHIV age on ART, the prevalence of comorbid noncommunicable diseases (NCDs),4 such as diabetes and hypertension, have also increased, with recent studies showing a diabetes prevalence of 13.7% in African adults over 55 years old and hypertension in more than 50% of older adults in Southern Africa.5–7 However, evidence to guide programs and policies for those affected remains limited.8–12 As such, there is a pressing need for greater research at the intersection of these epidemics, especially in low and middle-income countries that face a dual burden of high HIV prevalence and rising cardiometabolic conditions coupled with weak health system infrastructure.13
The role of ART programs and their impact on health system performance in sub-Saharan Africa has been an area of controversy for many years. Some experts have expressed concern that ART programs may exacerbate health inequalities by directing scarce resources to PLWHIV or diverting those resources away from primary health care systems that serve the general adult population to HIV-related programs.14–16 By contrast, others have argued that ART programs may actually serve as a backbone on which to build stronger health systems by increasing the overall infusion of resources to the system and strengthening both human resources for health and systems for laboratory and clinical medicine that are common to ART programs and other types of health care service provision.14,17 Moreover, the marginal costs of adding these basic diagnostic and treatment services for diabetes and hypertension to existing ART programs is likely to be low because clinical infrastructure and personnel are already in place. With few data to support or refute these claims, there remains a lack of clarity about the effect of ART programs on health system performance for conditions other than HIV.5,18–20
Within a large cohort of adults aged >40 years in Agincourt, South Africa, this study aims to describe (1) differences in utilization of health care services for diabetes and hypertension among HIV-positive versus HIV-negative persons and (2) the relationship between ART use and health care service utilization for diabetes and hypertension among HIV-positive persons. We hypothesize that the HIV-infected individuals who use ART have greater access to diagnosis and care for comorbid diabetes and hypertension as compared to individuals not receiving ART, irrespective of their HIV infection status. In essence, we postulate that ART programs may provide a platform to strengthen health system performance for other chronic conditions.
Overview of the HAALSI Cohort
The Health and Aging in Africa: A Longitudinal Study of an INDEPTH Community in South Africa (HAALSI) aims to understand population aging in a rapidly transitioning community of rural Northeast South Africa. The cohort consists of adults aged 40 and older and is nested within a Health and Demographic Surveillance System (HDSS) covering the Agincourt subdistrict adjacent to Southern Mozambique.21 The subdistrict is a rural region comprising 31 villages with high rates of migration. The local population is served by 6 clinics and 2 health centers along with 3 district hospitals that are located within 60 km of the study site.
Enrollment in the HAALSI cohort, with participants randomly sampled from the Agincourt HDSS, began in November 2014 and closed in November 2015.22 Individuals 40 years and older as of July 1, 2014 who had been living in the study site for the 12 months before were eligible for selection. A total of 6281 women and men were selected for the main household survey to achieve a target sample size of 5000 people. Those who participated in earlier studies and met the eligibility criteria were selected with 100% probability, whereas the remainder of the sample was selected randomly from the 2013 HDSS census, stratifying on sex to achieve equal numbers of men and women.23
The baseline survey, on which this analysis is based, included self-reported demographic, health, and economic information, questions on self-reported health care service utilization for HIV, diabetes, and hypertension, as well as anthropometry and laboratory studies. The survey was administered by trained, local fieldworkers who recorded participant responses in a Computer-Assisted Personal Interview System. As part of the survey administration, these trained fieldworkers also collected blood through finger prick and prepared dried bloodspot (DBS) samples from each participant who consented to blood collection. The anthropometry and laboratory data included height, weight, blood pressure, and point-of-care glucose. In addition, the DBS were later tested for HIV antibody and viral load in all consenting participants. The HIV screening and confirmatory enzyme-linked immunosorbent assays used were the Vironostika HIV 1/2 Ag/Ab MicroELISA System (BioMérieux, Marcy-l'Étoile, France) and the Roche Cobas E411 Combi Ag, respectively. The viral load platform was BioMérieux NucliSens with a lower limit of detection of <100 copies by DBS.
For those participants with a positive HIV antibody test, further testing for exposure to either emtricitabine (FTC) or lamivudine (3TC) was performed through dried blood spot (DBS). Study samples were analyzed at the Pharmacokinetic Laboratory at the University of Cape Town in South Africa. A semiquantitative liquid chromatography tandem mass spectrometry (LC/MS/MS) assay with a lower limit of detection of 0.02 μg/mL was validated for the determination of 3TC and FTC from DBS. The method consisted of a protein precipitation, followed by high-performance liquid chromatography with tandem mass spectrometry (MS/MS) detection using gradient elution. An AB Sciex API 4000 mass spectrometer at unit resolution in the multiple reaction monitoring mode was used to monitor the transition of the protonated precursor ions at m/z 248.0 and 230.2 to the product ions at m/z 129.9 and 112.0 for FTC and 3TC, respectively. Electrospray ionization was used for ion production.24,25 Samples that fell above the lower limit of detection for either of the antiretroviral drugs tested were classified as positive for ART use. The time from ingestion to reaching this threshold has been estimated at 1.5 days for 3TC.24 Ethical approval for HAALSI was obtained from the University of the Witwatersrand, the Harvard T.H. Chan School of Public Health, and the Mpumalanga Provincial Research and Ethics Committee.
Defining HIV Infection, ART, and Viral Suppression
In this study, HIV infection status was defined using the result of the DBS antibody testing, rather than self-reported status. Those participants who tested positive for exposure to either FTC or 3TC were classified as ART users, irrespective of their self-reported HIV or ART status. Thus, ART exposure in this study is defined as a binary variable indicating that either a drug or drugs (FTC or 3TC or both drugs) were present in the DBS sample or no drug was present in the sample. All first- and second-line antiretroviral regimens in South Africa use one of these 2 drugs as part of a 3 drug combination. Finally, HIV viral load was measured using DBS viral load testing, allowing for a biological assessment of viral suppression. We define viral suppression based on the DBS results at <100 copies/mL, the lower limit of detection for the assay. We also provide a comparative assessment of the rate of viral suppression when it is defined as <400 copies/mL, the upper limit for routine viral load monitoring per South African guidelines and as <1000 copies/mL, the threshold for defining treatment failure per South African guidelines. We performed all supplementary analyses at each of these viral load thresholds.
Utilization of Health Care Services for Hypertension and Diabetes
Measured height and weight were used to define body mass index (BMI). To define hypertension, 3 blood pressure measurements were taken 2 minutes apart for each respondent, using a standard manual blood pressure cuff. The first measurement was dropped; the second and third measurements were averaged.26 Hypertension was defined as a mean systolic blood pressure (SPB) ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg or self-reported use of medication for hypertension prescribed by a doctor, nurse, or health care worker. Diabetes was defined as a fasting plasma glucose ≥7.0 mmol/L or random plasma glucose ≥11.1 mmol/L measured at the time of interview or self-reported use of medication for diabetes mellitus prescribed by a doctor, nurse, or health care worker.27,28 Utilization of care for hypertension and diabetes was assessed using the following variables as reported by each participant: access to diagnosis (ever measurement of blood pressure and ever measurement of blood sugar by a health care provider), receiving advice on lifestyle modification (advised by health care provider to change diet or exercise), awareness of diagnosis (told about diagnosis of hypertension or told about diagnosis of diabetes by health care provider), and access to treatment (self-reported use of medication for blood pressure or self-reported use of medication for blood sugar). The latter 2 outcomes were only assessed among those who met criteria for a diagnosis of the respective condition.
We calculated descriptive demographic and health characteristics stratified into 3 groups: HIV-negative participants, HIV-positive participants who were not currently using ART (HIV+/no ART), and HIV-positive participants who were current ART users based on FTC/3TC exposure testing (HIV+/ART users). Individual participants could only be assigned to one of these groups. Means were compared using 1-way analysis of variance and proportions using χ2 tests. We then calculated the proportion of participants in the HIV+/ART user group who self-reported utilization of each of the care metrics outlined above and compared this to the proportion of people in the HIV-negative and HIV+/no ART groups who self-reported utilization of these same services.
Next, to test our central hypothesis, we used multivariable logistic regressions to assess the association between ART use and each of the following outcomes in terms of diagnosis and advice on lifestyle modification from a health care provider across the cohort: (1) ever receipt of a blood pressure measurement, (2) ever receipt of a blood sugar measurement, (3) ever receipt of advice to change diet, and (4) ever receipt of advice to exercise. We then use logistic regression to assess the association between ART use and each of the following outcomes related to awareness of diagnosis and treatment of diabetes or hypertension among those who met criteria for a diagnosis of either condition: (1) awareness of diabetes diagnosis among diabetics, (2) awareness of hypertension diagnosis among hypertensives, (3) treatment for diabetes among diabetics, and (4) treatment for hypertension among hypertensives. All models were adjusted for age, sex, BMI, household wealth index, and educational attainment. The household wealth index is the quintile ranking of scores derived from a principal components analysis of ownership of household items, livestock, and vehicles.
As a sensitivity analysis, we also performed this multivariate regression to examine whether the relationship was preserved in self-reported ART and for the ART users who were also virally suppressed. As described previously, we examined these relationships using 3 different definitions of viral suppression (<100, <400, and <1000 copies).
HIV Infection, ART, and Viral Suppression in Agincourt, South Africa
Figure 1 depicts the HIV-positive population in the HAALSI study stratified by ART use and viral suppression. Among 4560 participants who underwent DBS testing for HIV antibody, 1048 (23%) were found to be HIV positive. As discussed in detail elsewhere, this is a high HIV prevalence given the older age of this cohort.22 Moreover, 662 (64%) of these HIV-positive participants were also positive for exposure to FTC or 3TC, although only 450 or 68% of the ART users as defined by DBS exposure testing also self-reported ART use. Finally, among the ART users, 479 (72%) were virally suppressed where viral suppression was defined by a viral load of <100 copies, the lower limit of assay detection in this study. When using the alternative viral load threshold of <400 copies, 549 (83%) were suppressed, whereas 594 (90%) were suppressed at a threshold of <1000 copies.
Diabetes and Hypertension in the HIV-positive Population
The demographic and health data, displayed in Table 1, show that age, BMI, and mean systolic blood pressure are all lower among the HIV-positive population, both in those on and off ART, as compared to the HIV-negative population (P < 0.001). There is a lower proportion of HIV-positive participants who have comorbid hypertension (HIV negative: 63.7%, HIV+/no ART: 43.5%, HIV+/ART users: 38.7%) and diabetes (HIV negative: 12.0%, HIV+/no ART: 6.3%, HIV+/ART users: 7.8%) when compared with the HIV-negative population, although the prevalence of these conditions remains high across all 3 groups.
Table 1 also describes self-reported utilization of care for hypertension and diabetes in ART users compared with non-ART users in the HAALSI cohort. The percentage of participants who received a blood pressure or blood sugar measurement or counseling from a health care provider regarding exercise was greater among ART users as compared to non-ART users (HIV negative: 70.0%, 48.7% and 5.5%, HIV+/no ART: 69.2%, 42.1% and 1.3%, HIV+/ART users: 71.5%, 49.0% and 7.8%), as was awareness of diagnosis and treatment for hypertension (HIV negative: 65.2% and 56.5%, HIV+/no ART: 59.6% and 49.1%, HIV+/ART users: 67.6% and 57.8%) and diabetes (HIV negative: 55.9% and 53.9%, HIV+/no ART: 47.8% and 43.5%, HIV+/ART users: 56.9% and 54.9%). These differences were only statistically significant for blood sugar measurement and counseling to exercise.
ART and Utilization of Health Care Services for Diabetes and Hypertension
Multivariable logistic regression models of diagnosis and preventive counseling for diabetes and hypertension (Table 2) showed greater odds of ever receiving a blood pressure measurement [adjusted odds ratio (aOR) 1.27, 95% confidence interval (CI): 1.04 to 1.55], blood sugar measurement (aOR 1.26, 95% CI: 1.05 to 1.51), or counseling regarding exercise (aOR 1.57, 95% CI: 1.11 to 2.22) among HIV+/ART users as compared to those who were HIV negative, after adjusting for age, sex, BMI, education, and household wealth quintile. However, the odds of receiving preventive counseling from a health care provider regarding diet (aOR: 1.24, 95% CI: 0.92 to 1.67) among the HIV+/ART users was not statistically significant.
Regression analyses (Table 3) also demonstrated greater odds of being aware of a diagnosis of hypertension among those hypertensive participants who were HIV+/ART users (aOR 1.52, 95% CI: 1.12 to 2.05) as compared to those who were HIV negative. The HIV+/ART group also had greater odds of receiving treatment for their hypertension (aOR 1.63, 95% CI: 1.21 to 2.19), again compared with those who were HIV negative. Relationships of similar magnitude were seen between ART use and awareness of diabetes diagnosis or treatment of diabetes, but these were not statistically significant because of lower prevalence of diabetes diagnosis and treatment. Although not shown in Tables 2 and 3, all these relationships were equally strong and significant among the subgroup of self-reported ART users and among those who were ART users and virally suppressed (see Supplemental Digital Content, http://links.lww.com/QAI/B29). There was no qualitative difference in the results at alternative thresholds of viral suppression.
This study offers several key findings regarding the relationship between HIV infection, ART use, and utilization of health services for hypertension and diabetes, 2 major cardiometabolic diseases of increasing importance in this region. First, in a community with very high HIV, hypertension, and diabetes prevalence, ART utilization is associated with utilization of needed hypertension and diabetes care, including greater diagnostic testing and preventive counseling regarding lifestyle modification for diabetes and hypertension as well as treatment among those affected by these comorbid chronic diseases. This rural South African cohort was well suited to examine these relationships given the high prevalence of HIV and the similarly high rates of both hypertension and diabetes in this group of older adults.
However, because of the cross-sectional nature of this study, it is unknown whether ART use actually precedes the diagnosis and treatment of hypertension and diabetes or vice versa. The mechanism for the association between these services is unclear but may include better access to or utilization of health care services generally, increased exposure to health information resulting in better health literacy, more accurate self-reporting of service utilization, more assertive health-seeking behavior, or some combination of these factors. One important factor underlying this finding may be that those on ART are required to present to a clinic each month to obtain medications and thus likely come into contact with the health system much more frequently than HIV-negative or HIV-positive people who are not receiving ART. The frequency of this contact may offer more opportunities to be diagnosed, educated, or referred to chronic disease clinical staff for care of other health conditions. The data presented here suggest a potential positive spillover effect from ART utilization to care for other chronic conditions. If true, this may suggest a more central role for ART programs in the delivery of care for cardiometabolic diseases among HIV-positive populations, in particular screening, diagnosis, and preventive care for the increasingly large population initiating ART in this context.
Our results found that awareness of a hypertension diagnosis is also greater among those hypertensive participants in the HIV+/ART group. ART users who were also hypertensive were also more likely to self-report receiving treatment for their hypertension. However, similar relationships did not exist for awareness of a diabetes diagnosis or treatment of diabetes, although the small number of people with diabetes in the sample may limit the power to draw definitive conclusions regarding this relationship.
In addition, we found a high prevalence of both diabetes and hypertension in this rural South African community specifically among the population over age 40. Although these conditions have not received nearly the same level of attention as HIV, several recent studies in similar settings in Southern Africa have shown comparable prevalence estimates for both diabetes and hypertension in aging African adults.5–7 This lends further evidence that the burden of these 2 cardiovascular risk factors is very important in middle-income countries and that failing to address these chronic diseases could have detrimental effects on the population-level gains in life expectancy and quality of life that have been achieved through ART roll-out.2,29 However, we also show that HIV-positive participants in this rural South African cohort (HIV+/no ART and HIV+/ART) have a lower prevalence of diabetes and hypertension as compared to the HIV-negative population. The lower prevalence of these conditions is likely multifactorial as PLWHIV are on average younger and have a lower BMI than the HIV-negative participants. In particular, their lower BMI may be in part caused by their HIV disease and would be expected to reduce their risk of diabetes and hypertension, although they remain at heightened risk of other opportunistic infections.4,30
However, the fact that the prevalence of diabetes and hypertension remains high in the setting of relatively younger age and lower BMI among PLWHIV may also indicate the potential for an increasing burden of these conditions in the future with the shift toward early ART initiation in HIV disease.31 In particular, early ART initiation may be expected to promote body weight preservation by averting HIV-associated wasting among infected populations, in contrast to older guidelines that recommended ART initiation at a later stage of immunosuppression. Perhaps paradoxically, this could result in even greater rates of hypertension, diabetes, and other noncommunicable diseases among the HIV-positive population in the future. As shown in this study, greater counseling on healthy lifestyle habits for HIV-positive people on ART may lead to improvements in diet and increased physical activity, thereby mitigating the effects of earlier ART initiation on NCDs.
Although these findings provide important insight regarding the design and delivery of care for diabetes and hypertension among PLWHIV, they may also provide lessons that are relevant to care delivery for these conditions in the HIV-negative population. When considering possible overlap between the health systems infrastructure underlying HIV care programs and the health systems structures needed to support growing epidemics of diabetes and hypertension,32,33 these findings suggest that ART programs may serve as a powerful platform for broader population health improvements through increased programmatic integration.34 One excellent example of the potential for this integration has been the Academic Model Providing Access to Healthcare program which provides care to >150,000 adults and children living with HIV/AIDS throughout Western Kenya.35 This collaboration has been able to provide important insights about cardiometabolic diseases in PLWHIV, including high rates of cardiovascular risk, low levels of knowledge, and perceived risk of cardiovascular disease in this population.35 Another example of this may be the Ideal Clinics Initiative in South Africa, a reform effort which in part involves integrating some aspects of care for HIV with care for common chronic conditions such as diabetes and hypertension.36 Finally, the expansion of universal health coverage will potentially be advanced by the integration of services for common key conditions such as HIV, diabetes, and hypertension.
This study has several limitations. First, data on utilization of health services for NCDs were self-reported, including the diagnosis and care received for both diabetes and hypertension. Despite this limitation, in the case of diagnosis and preventive counseling for these NCDs, the use of self-reported data can be illuminating because it reflects not only the health services participants have received but also their understanding of both those diagnostic tests and advice. In the case of counseling on lifestyle modification, this is particularly relevant given that this type of advice can seemingly only be used in a meaningful way if the recipient understands and retains the information provided. However, the limitation of self-reported data includes misreporting and the direction of any resulting bias is difficult to predict. A second limitation of this study is the lack of temporal data regarding the date and duration of HIV infection, the time when ART was initiated, and the timing during which services were received for diabetes and hypertension. Without this information, it is difficult to assess the directionality of the relationship between ART program participation and receipt of care for NCDs. Finally, there are various definitions of viral suppression that could be acceptable in the South African context; however, the supplementary analyses for this population showed a preserved association between virally suppressed ART users and the outcomes of interest across 4 commonly used thresholds.
In summary, our analysis shows that the burden of noncommunicable diseases is substantial among the HIV-positive population in rural South Africa. Our findings suggest that ART increases access to diagnosis and preventive counseling for diabetes and hypertension, as well as awareness of diagnosis and treatment of hypertension. The positive spillover effects from ART utilization to preventive care for other chronic conditions could provide a powerful vehicle for broader population health improvements through increased programmatic integration.
The authors thank Shahira Ahmed, ScD, for her support in the preparation of this manuscript.
1. World Health Organization. Global Update on HIV Treatment 2013. Geneva, Switzerland: World Health Organization; 2013.
2. Bor J, Herbst AJ, Newell ML, et al. Increases in adult life expectancy in rural South Africa: valuing the scale-up of HIV treatment. Science. 2013;339:961–965.
3. Hontelez JA, de Vlas SJ, Baltussen R, et al. The impact of antiretroviral treatment on the age composition of the HIV epidemic in sub-Saharan Africa. AIDS. 2012;26(suppl 1):S19–S30.
4. Feigl AB, Bloom DE, Danaei G, et al. The effect of HIV and the modifying effect of anti-retroviral therapy (ART) on body mass index (BMI) and blood pressure levels in rural South Africa. PLoS One. 2016;11:e0158264.
5. Werfalli M, Engel ME, Musekiwa A, et al. The prevalence of type 2 diabetes among older people in Africa: a systematic review. Lancet Diabetes Endocrinol. 2015;4:72–84.
6. Lloyd-Sherlock P, Beard J, Minicuci N, et al. Hypertension among older adults in low- and middle-income countries: prevalence, awareness and control. Int J Epidemiol. 2014;43:116–128.
7. Kaze AD, Schutte AE, Erqou S, et al. Prevalence of hypertension in older people in Africa: a systematic review and meta-analysis. J Hypertens. 2017;35:1345–1352.
8. Bloomfield GS, Khazanie P, Morris A, et al. HIV and noncommunicable cardiovascular and pulmonary diseases in low- and middle-income countries in the ART era: what we know and best directions for future research. J Acquir Immune Defic Syndr. 2014;67(suppl 1):S40–S53.
9. Barnighausen T, Welz T, Hosegood V, et al. Hiding in the shadows of the HIV epidemic: obesity and hypertension in a rural population with very high HIV prevalence in South Africa. J Hum Hypertens. 2008;22:236–239.
10. Ameh S, Gomez-Olive FX, Kahn K, et al. Predictors of health care use by adults 50 years and over in a rural South African setting. Glob Health Action. 2014;7:24771.
11. Clark SJ, Gomez-Olive FX, Houle B, et al. Cardiometabolic disease risk and HIV status in rural South Africa: establishing a baseline. BMC Public Health. 2015;15:135.
12. Gomez-Olive FX, Thorogood M, Clark B, et al. Self-reported health and health care use in an ageing population in the Agincourt sub-district of rural South Africa. Glob Health Action. 2013;6:19305.
13. The Lancet Diabetes E. HIV and NCDs: the need to build stronger health systems
. Lancet Diabetes Endocrinol. 2016;4:549–550.
14. Yu D, Souteyrand Y, Banda MA, et al. Investment in HIV/AIDS programs: does it help strengthen health systems
in developing countries? Glob Health. 2008;4:8.
15. England R. The dangers of disease specific programmes for developing countries. BMJ. 2007;335:565.
16. Shiffman J. Has donor prioritization of HIV/AIDS displaced aid for other health issues? Health Policy Plan. 2008;23:95–100.
17. McCoy D, Chopra M, Loewenson R, et al. Expanding access to antiretroviral therapy in sub-saharan Africa: avoiding the pitfalls and dangers, capitalizing on the opportunities. Am J Public Health. 2005;95:18–22.
18. Atun R, Gale EA. The challenge of diabetes in sub-Saharan Africa. Lancet Diabetes Endocrinol. 2015;3:675–677.
19. Levitt NS. Diabetes in Africa: epidemiology, management and healthcare challenges. Heart. 2008;94:1376–1382.
20. Joint United Nations Programme on HIV/AIDS (UNAIDS). Chronic Care of HIV and Noncommunicable Diseases. Joint United Nations Programme on HIV/AIDS (UNAIDS); 2011.
21. Kahn K, Collinson MA, Gomez-Olive FX, et al. Profile: Agincourt health and socio-demographic surveillance system. Int J Epidemiol. 2012;41:988–1001.
22. Rosenberg MS, Gomez-Olive FX, Rohr JK, et al. Sexual behaviors and HIV status: a population-based study among older adults in rural South Africa. J Acquir Immune Defic Syndr. 2017;74:e9–e17.
23. Gaziano TA, Abrahams-Gessel S, Gomez-Olive FX, et al. Cardiometabolic risk in a population of older adults with multiple co-morbidities in rural South Africa: the HAALSI (Health and aging in Africa: longitudinal studies of INDEPTH communities) study. BMC Public Health. 2017;17:206.
24. Kim AA, Mukui I, Young PW, et al. Undisclosed HIV infection and antiretroviral therapy use in the Kenya AIDS indicator survey 2012: relevance to national targets for HIV diagnosis and treatment. AIDS. 2016;30:2685–2695.
25. Koal T, Burhenne H, Romling R, et al. Quantification of antiretroviral drugs in dried blood spot samples by means of liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom. 2005;19:2995–3001.
26. United States Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey: 1999-2000 Data Documentation, Codebook, and Frequencies. Washington, DC: United States Centers for Disease Control and Prevention; 2015.
27. Society for Endocrinology Metabolism and Diabetes of South Africa. The 2012 SEMDSA Guideline for the Management of Type 2 Diabetes Mellitus: Summary. South Africa: Society of Endocrinology, Metabolism and Diabetes in South Africa; 2012.
28. World Health Organization. Definition and Diagnosis of Diabetes and Intermediate Hyperglycemia. Geneva, Switzerland: World Health Organization; 2006.
29. Bor J, Tanser F, Newell ML, et al. In a study of a population cohort in South Africa, HIV patients on antiretrovirals had nearly full recovery of employment. Health Aff (Millwood). 2012;31:1459–1469.
30. Geldsetzer P, Feigl AB, Tanser F, et al. Population-level decline in BMI and systolic blood pressure following mass HIV treatment: evidence from rural KwaZulu-Natal. Obesity (Silver Spring). 2017;25:200–206.
31. Insight Start Study Group; Lundgren JD, Babiker AG, et al. Initiation of antiretroviral therapy in early asymptomatic HIV infection. N Engl J Med. 2015;373:795–807.
32. World Health Organization. Universal Health Coverage: Questions and Answers. 2015. Available at: http://http://www.who.int
/healthsystems/topics/financing/uhc_qa/en/. Accessed December 11, 2015.
33. Reich MR, Harris J, Ikegami N, et al. Moving towards universal health coverage: lessons from 11 country studies. Lancet. 2016;387:811–816.
34. Tollman SM, Kahn K, Sartorius B, et al. Implications of mortality transition for primary health care in rural South Africa: a population-based surveillance study. Lancet. 2008;372:893–901.
35. Temu TM, Kirui N, Wanjalla C, et al. Cardiovascular health knowledge and preventive practices in people living with HIV in Kenya. BMC Infect Dis. 2015;15:421.
36. Mahomed OH, Asmall S, Freeman M. An integrated chronic disease management model: a diagonal approach to health system strengthening in South Africa. J Health Care Poor Underserved. 2014;25:1723–1729.