Introduction
In 2014, The Joint United Nations Programme on HIV/AIDS (UNAIDS) launched the 95-95-95 targets. The aim was to diagnose 95% of all HIV-positive individuals, provide antiretroviral therapy (ART) for 95% of those diagnosed and achieve viral suppression for 95% of those treated by 2030 [1].
Significant progress has been made in controlling the HIV epidemic in the past 20 years, with substantial declines in incidence [2]. Despite successes in some regions, many countries are still falling behind UNAIDS targets [3] and only 67% of people with HIV (PWH) globally were accessing treatment in 2019 [4]. Approximately 690 000 people died from HIV-related illnesses and 1.7 million people became newly infected with HIV in 2019 [4].
A successful national response to HIV requires, amongst other things; sufficient medical staff, access to testing and medications, antistigma campaigns, community engagement, political will and financial resources. However, improvements in resources for HIV responses in low-income and middle-income countries stalled in 2017, and global HIV funding fell by 7% between 2017 and 2019 [3]. UNAIDS estimates that global HIV funding must increase by 30% in order to respond to the growing HIV epidemic in 2020 [3].
At the same time that countries are being urged to increase funding for national HIV responses, the cost of ART remains a barrier to HIV treatment. Whilst price reductions have been seen in some countries, prices for ART have even been seen to rise in others. Prices for HIV antiretroviral drugs can vary by country, with the lowest prices in low-income and middle-income countries and the highest prices in upper-middle and high-income countries [5]. In the USA in 2012, the minimum cost for ART per person per year (pppy) was $24 970. By 2018, this had risen to $36 080; an increase 3.5 times faster than inflation [6]. By contrast, in 2018, the cost of dolutegravir (DTG) in Georgia was $27 pppy and the cost of efavirenz (EFV) was $39 pppy [5]. Large international donors, such as the Global Fund and the US President's Emergency Plan For AIDS Relief (PEPFAR) are placing emphasis on sustainable financing and encouraging country-led HIV responses [7]. Sustainable financing is a necessary long-term goal in ending the HIV epidemic; however, it places considerable strain on countries that rely heavily on external funding for HIV treatment.
The coronavirus disease 2019 (COVID-19) pandemic has further exacerbated the problem of HIV treatment access. In July 2020, 73 countries reported increased risk of ART stockouts as a result of COVID-19, and HIV-related deaths were predicted to increase by 10% in high-burden settings as a result of drug shortages and lack of capacity in the healthcare system [8,9]. Lockdowns in some countries have led to ART production shortages as it forces factories and supply chains to run at a reduced capacity. The situation has been worsened by ARTs such as lopinavir/ritonavir (LPV/r) and atazanavir/ritonavir (ATV/r) being repurposed in clinical trials for potential effects against COVID-19 [10]. The new challenges presented by COVID-19 have aggravated existing barriers to treatment access and made the need for improvements in treatment access even more acute.
Effective treatment of HIV both reduces HIV morbidity and mortality and also prevents virus transmission; this reduction in virus spread is necessary to achieve UNAIDS targets and to end AIDS [11]. For UNAIDS treatment targets to be realised, it is paramount that ART be made available to all those who need it to combat the HIV epidemic – the best drugs at the best price. We calculated the minimum costs of production for HIV medications and considered whether HIV treatment could be made more accessible through equitable pricing.
Methods
Epidemiological data
Epidemiological data on the number of PWH, HIV prevalence and the proportion of PWH on ART were compiled for all countries with available data, of which there were 164 countries. Data were extracted primarily from the UNAIDS 2020 database; where data were not available in the 2020 database, the 2019 database was used [12]. UNAIDS data were supplemented with data from publications, national surveillance and published reports [13–18]. The number of annual new infections, HIV-related deaths and the mother-to-child transmission (MTCT) rate (the proportion of births to HIV-positive mothers resulting in vertical transmission) were also acquired from the UNAIDS 2020 and 2019 databases, although these data were available for fewer than 164 countries (115 countries for new infections, 151 for HIV-related deaths and 111 for MTCT rate). The number of live births was extracted from the United Nations Populations Division (UNPD) Population Prospects [19]. Countries were classified as having low, medium or high HIV prevalence such that the global number of PWH was equally distributed across the three categories.
Clinical outcomes
HIV infection and death rates were calculated for each country by dividing the number of infections and deaths, respectively, by the number of PWH. Weighted ordinary least squares (OLS) log-linear regression was used to estimate the associations between ART coverage and (a) HIV infection rate, (b) HIV death rate and (c) MTCT rate, weighted by the number of PWH. Model selection was conducted in a bi-directional stepwise manner using the stepAIC function from the MASS package for R version 3.6.3 [20]. A P value threshold of 0.05 was used for significance. In addition to the natural log of ART coverage, covariates considered in model selection were: UNAIDS region, country income level (low, medium, high) and adult HIV prevalence. The resulting regression models were used to estimate the infection, death and MTCT rates in each country assuming (i) current service coverage and (ii) 95--95--95 service coverage (i.e. 90.25% of PWH on ART). If a country's ART coverage was already greater than 90.25%, we assumed that it remained unchanged.
The numbers of HIV infections and HIV-related deaths under current and 95--95--95 service coverage were estimated by multiplying model-predicted infection and death rates by the number of PWH in each country. The number of MTCTs was estimated by multiplying the predicted MTCT rate by the number of live births and the HIV prevalence.
Generic drug pricing
The methods used to estimate the minimum costs of production have been described previously [21–23]. Briefly, we analysed the costs of exports of Active Pharmaceutical Ingredient (API) from India using the online tracking database Panjiva [24], which shows details of all shipments of API with quantities and costs per kilogram. Panjiva was selected as it provides real-time, up-to-date shipping data. We used all available costing data for each drug API found on Panjiva, excluding shipments less than 1 kg in size. We also excluded the lowest and highest 15% of results based on prices per kg so as to exclude outliers. The weighted mean cost of API per kg (weighted by shipment size) was taken. The estimated cost of API per dosage regimen (in grams) was derived from the cost per kilogram of API. A conversion cost of US $0.01 per tablet was used. A multiplier based on API mass was applied to account for the price of excipients, which are additional substances needed to convert API into finished pharmaceutical products. Our estimated API costs presume that production is carried out by a generic provider of APIs, where associated costs of capital investment, overhead and labour are not as high as for production by originator companies. A profit margin of 10% and Indian taxation of 27% on profit was added. We have used the term ‘treatment cost’ in reference solely to the price of ART but of course the cost of holistic treatment for a PWH will also include diagnostics, blood monitoring tests, counselling, nursing and physician time and other support.
We investigated the minimum costs of production for a selection of antiretroviral drugs with available data, including: abacavir (ABC), atazanavir (ATV), dolutegravir (DTG), efavirenz (EFV), emtricitabine (FTC), lamivudine (3TC), tenofovir alafenamide (TAF), tenofovir disoproxil fumarate (TDF) and ritonavir (/r). We estimated the cost of each of these drugs for a 1-year treatment course at WHO standardized ‘defined daily dosing’. The shortlist of all antiretroviral drugs investigated was not based on treatment guidelines but was chosen to make maximum use of available data and give a holistic picture of the minimum costs of production; therefore, we present data for all antiretroviral drugs with sufficient data available to estimate the minimum costs of production.
Cost of global treatment
We estimated the cost to treat all 164 countries in our analysis at current service coverage and 95--95--95 service coverage. The cost to treat each country was calculated by multiplying the estimated generic cost for 1 year of TDF/3TC/DTG combination therapy by ART coverage and the number of PWH. We focussed on TDF/3TC/DTG as it is the current WHO-recommended first-line treatment [25].
Current pharmaceutical sales
Annual antiretroviral sales were compiled from pharmaceutical quarterly financial sales reports. Data were converted to 2018 USD and adjusted for inflation using the World Bank GDP deflator (annual %).
Results
Most new HIV infections and deaths occur in lower prevalence countries
High prevalence countries (Botswana, Eswatini, Lesotho, Namibia, South Africa and Zimbabwe) had higher ART coverage rates compared with low and medium prevalence countries, as shown in Table 1; indeed, Botswana and Namibia have some of the highest ART coverage rates in the world. Medium and low prevalence countries had higher numbers of new infections and HIV-related deaths, with medium prevalence countries having the highest numbers of both infections and deaths. This is consistent with previous results using 2018 UNAIDS data [26]. Epidemiological data by country are shown in the Supplementary Material, https://links.lww.com/QAD/C191.
Table 1 -
Summary of HIV epidemiological data, stratified by HIV prevalence.
|
Prevalence |
|
Low (<1.2%) |
Medium (1.2--12.1%) |
High (>12.1%) |
| Number of countries |
128 |
30 |
6 |
| Epidemic size (PWH) |
13 249 458 |
13 997 600 |
10 030 000 |
| Mean prevalence |
0.3% |
3.5% |
19.3% |
| Mean % of PWH on ART |
58.2% |
53.2% |
80.3% |
| Number of new annual infections |
328 119 |
634 100 |
271 900 |
| Number of annual HIV-related deaths |
188 189 |
289 900 |
107 100 |
HIV prevalence category cutoffs were calculated to minimize the difference in the number of PWH between groups. The number of MTCTs is not shown as not all countries had data available for the MTCT rate. MTCT, mother-to-child transmissions; PWH, people with HIV.
Increased antiretroviral therapy coverage will have substantial clinical benefits
Significant covariates from stepwise model selection for weighted log-linear models of the association between ART coverage and HIV death rate were UNAIDS region, country income level and adult prevalence. For models of HIV infection rate, significant covariates were country income level and adult prevalence. For models of MTCT rate, significant covariates were UNAIDS region and adult prevalence. GDP was not significant for any of the three models. Model equations, coefficients, standard errors, P values and data used to fit the models are shown in the supplementary material, https://links.lww.com/QAD/C191. Increased ART coverage was found to be associated with decreased HIV death rates, decreased HIV infection rates and decreased MTCT rates. Fitted log-linear model equations were used to generate country-level estimates; the model-estimated effect of achieving 95--95--95 on the HIV death rate, HIV infection rate and MTCT rate in 164 countries is shown in Fig. 1.
;)
Fig. 1: The model-estimated effect of increased antiretroviral therapy coverage on HIV death rate, infection rate and mother-to-child transmission rate.
We estimated that if all 164 countries in our analysis had 95-95-95 service coverage, we could prevent 66 308 MTCTs, 241 811 HIV-related deaths and 631 398 new HIV infections every year.
Generic treatments can be produced for substantially less than patented drugs
We estimated that TDF/3TC/DTG combination therapy could be produced for $59 pppy. The TDF component could cost $20, 3TC $27 and DTG $12. The estimated costs of other combination therapies were also affordable. We estimated that ABC/3TC/DTG could cost $75 pppy ($37 for the ABC component). In the USA, ABC/3TC/DTG sold for $38 628 pppy in July 2020. The estimated costs of other generic drugs are shown in Table 2.
Table 2 -
Estimated cost of generic drugs.
| Drug |
Cost per kilo of generic API |
Dosage per day (mg) |
Cost of formulated generic drug pppy |
| Abacavir (ABC) |
$258.09 |
300 |
$36.61 |
| Atazanavir (ATV) |
$621.51 |
300 |
$81.46 |
| Dolutegravir (DTG) |
$360.05 |
50 |
$11.62 |
| Efavirenz (EFV) |
$169.22 |
600 |
$47.18 |
| Emtricitabine (FTC) |
$345.11 |
200 |
$32.94 |
| Lamivudine (3TC) |
$182.24 |
300 |
$27.25 |
| Tenofovir alafenamide (TAF) |
$5673.90 |
25 |
$62.52 |
| Tenofovir disoproxil fumarate (TDF) |
$157.15 |
245 |
$20.48 |
| Ritonavir (r) |
$231.04 |
100 |
$13.83 |
The estimated cost per kilo of generic API, combined with the daily dosing, is used to calculate the cost pppy. API, Active Pharmaceutical Ingredient; ppy, person per year.
Current global expenditure could be substantially reduced
We estimated that all 164 countries in our analysis could be treated with TDF/3TC/DTG (at $59 pppy) at 95--95--95 service levels for $2 billion annually. In comparison, global expenditure on HIV pharmaceuticals in 2019 was $28 billion. At $2 billion a year, the 164 countries in our analysis could have been treated at 95--95--95 in 2019 using generic medications for the cost of only 4 weeks of current annual sales. As shown in Table 2, the costs of formulated generic drugs pppy vary. For a more expensive combination, such as EFV/FTC/TDF at $101 pppy, the cost to treat 164 countries at 95--95--95 using generic medications would have been $3.4 billion; this is still only 12% of the global expenditure on HIV pharmaceuticals in 2019.
Discussion
HIV drugs are being produced for substantially less than their sale price under patent. We have estimated that TDF/3TC/DTG could be sold for only $59 pppy, whilst still providing for a 10% profit. Despite this, cost of treatment remains a barrier to HIV treatment. At $59 pppy for generic TDF/3TC/DTG, 95--95--95 could be reached in 164 countries for a total cost of $2 billion a year. The extent to which some countries are over-paying for antiretroviral drugs is further dramatically illustrated in the USA, where ABC/3TC/DTG sold for $38 628 pppy in July 2020; over 500 times the estimated cost of generic ABC/3TC/DTG ($75). In 2019, global expenditure on Biktarvy alone (TAF/FTC/BIC, Gilead Sciences, Foster City, CA, USA) was $4.7 billion. Often, the sale prices for patented drugs, including HIV antiretroviral drugs, are not made publicly available; therefore, these estimates provide only a snapshot of the high prices paid for antiretroviral drugs. It is important to remember that these high drug prices result in substantial profit for the global pharmaceutical industry at the expense of thousands of people with HIV who cannot afford to access treatment; annual global pharmaceutical sales for antiretroviral drugs in 2019 were $28 billion, and cumulative antiretroviral drugs sales since 2000 were $303 billion.
We have predicted astounding clinical benefits if the UNAIDS 95--95--95 targets were reached worldwide; 66 308 MTCTs, 241 811 HIV-related deaths and 631 398 new HIV infections could be prevented every year. Whilst the 90--90--90 UNAIDS diagnosis, treatment and viral suppression targets for 2020 are due to be narrowly missed (in 2019 global rates are at 81, 82 and 88%, respectively), the target of having less than 500 000 new infections and less than 500 000 AIDS-related deaths by 2020 is unfortunately very far away. In 2019, there were 690 000 AIDS-related deaths and an estimated 1.7 million people who were newly infected with HIV – which is unchanged since 2018. The number of people who need to start ART to reach 95--95--95 is significant. We have estimated that an additional 9.1 million people would have needed to be on ART in 2020 in order to reach 95--95--95 targets.
In the USA alone, we estimated that an additional 296 410 people need to start ART to reach 95--95--95. There are currently 1 173 900 PWH in the USA. In 2019, the US Department of Health and Human Services (HHS) announced a plan to reduce HIV infections by 90% by 2030, and proposed a $291 million dollar investment in the plan [27]. The USA spent $34.8 billion on combined domestic and global HIV prevention efforts in 2019, $21.5 billion of which was spent on domestic treatment and care [28]. At $38 628 pppy for treatment (ABC/3TC/DTG), over $45 billion would be needed to treat all PWH in the USA. The price of HIV drugs under patent must come down for countries to achieve 95--95--95.
One key reason for these high drug prices are patent laws providing pharmaceutical companies with monopoly power for up to 20 years from filing. In most countries, the patents for TDF and 3TC expired in 2021, which will allow generic competition to bring down prices. In most countries, DTG patents will not expire until 2025--2029. In the USA, antiretroviral drug prices remain high. However, antiretroviral drug prices have reduced greatly in other low-income and middle-income countries [5], making further reductions to minimum cost price more attainable.
It must be noted that drug pricing is not the only barrier to the global HIV response. Key enablers of effective HIV control include education and healthcare access as well as justice systems designed to protect marginalized societal groups. Legal barriers to HIV treatment access include but are not limited to criminalization of sex work and same-sex sexual relationships, a lack of legal recognition for transgender people and punitive policies for drug use [29]. Drug pricing is an issue of health equity but it is one part of a wider fabric of sociopolitical equality problems that must be addressed to improve global HIV ART access. We also note that the amount of by-product from API production can by up to 100× that of the weight of API produced. Furthermore, manufacturing of ARVs does have a significant environmental impact, which we have not assessed in our analysis. The profit maximization priority and poor environmental regulation enforcement in some countries may lead to corners being cut in terms of safe, green and efficient production and biochemical waste disposal.
Our observation that most AIDS-related deaths and new infections occur in low and medium prevalence countries is consistent with previous observations from 2019 UNAIDS data [26]. Several strategies, such as system decentralization, task shifting and adherence clubs have helped to improve ART coverage and retention rates in high-prevalence countries [30,31]. In low-prevalence countries, an absence of dialogue about HIV testing and lower levels of awareness in the community and healthcare professionals can lead to increased stigmatization of HIV compared with high-prevalence settings [32,33].
It must be noted that our study examines a hypothetical scenario where all PWH are treated with the same combination therapy (TDF/3TC/DTG). It is true that this is not always realistic; many patients will be offered first-line and second-line treatments, or will be treated according to drug availability at the time of diagnosis. However, for our analysis of costs, assuming use of other generic combination therapies will produce a consistent result: generic drugs can treat PWH for a fraction of the patented cost. As shown in Table 2, the estimated generic costs of all drugs in our study were very low.
Reaching the UNAIDS 95--95--95 targets could prevent substantial numbers of MTCTs, new infections, and HIV-related deaths each year. The annual cost of 95--95--95 treatment coverage for the 164 countries in this study is equivalent to only 4 weeks of global sales at current prices. Significant savings could be made by switching to quality-assured generics. Generic drug access is paramount to reduce HIV infections and deaths. ART must be made accessible at cost price to all who need it.
Acknowledgements
Funding: this study was funded by the International Treatment Preparedness Coalition (ITPC), grant number ITPC_MV_001.
Author contributions: K.H. sourced shipping records, calculated generic costs and conducted all analyses. K.H. wrote the manuscript and produced all figures and tables. J.L. conducted the literature search for country-level epidemiological data and provided comment on the manuscript. A.H. conceived of the study hypotheses, informed on appropriate methodology and provided comment on the manuscript.
Conflicts of interest
There are no conflicts of interest.
References
1. UNAIDS. Understanding Fast-Track: accelerating action to end the AIDS epidemic by 2030. 2015. unaids.org/sites/default/files/media_asset/201506_JC2743_Understanding_FastTrack_en.pdf. [Accessed 10 January 2021]
2. Girum T, Wasie A, Worku A.
Trend of HIV/AIDS for the last 26 years and predicting achievement of the 90-90-90 HIV prevention targets by 2020 in Ethiopia: a time series analysis.
BMC Infect Dis 2018; 18:320.
3. UNAIDS. Global AIDS update 2020. Seizing the moment: tackling entrenched inequalities to end epidemics. 2020.
4. UNAIDS. Global HIV and AIDS statistics - 2020 fact sheet. Available at:
https://www.unaids.org/en/resources/fact-sheet. [Accessed 10 January 2021]
5. Sim J, Hill A.
Is pricing of dolutegravir equitable? A comparative analysis of price and country income level in 52 countries.
J Virus Erad 2018; 4:230–237.
6. McCann NC, Horn TH, Hyle EP, Walensky RP.
HIV antiretroviral therapy costs in the United States, 2012–2018.
JAMA Intern Med 2020; 180:601–603.
7. Vassall A, Remme M, Watts C, Hallett T, Siapka M, Vickerman P, et al.
Financing essential HIV services: a new economic agenda.
PLoS Med 2013; 10:e1001567.
8. Hogan AB, Jewell BL, Sherrard-Smith E, Vesga JF, Watson OJ, Whittaker C, et al.
Potential impact of the COVID-19 pandemic on HIV, tuberculosis, and malaria in low-income and middle-income countries: a modelling study.
Lancet Glob Heal 2020; 8:e1132–e1141.
9. World Health Organization. WHO: access to HIV medicines severely impacted by COVID-19 as AIDS response stalls. 2020. Available at:
https://www.who.int/news-room/detail/06-07-2020-who-access-to-hiv-medicines-severely-impacted-by-covid-19-as-aids-response-stalls. [Accessed 10 January 2021]
10. Ford N, Vitoria M, Rangaraj A, Norris SL, Calmy A, Doherty M.
Systematic review of the efficacy and safety of antiretroviral drugs against SARS, MERS or COVID-19: initial assessment.
J Int AIDS Soc 2020; 23:e25489.
11. Granich RM, Gilks CF, Dye C, De Cock KM, Williams BG.
Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model.
Lancet 2009; 373:48–57.
12. UNAIDS. UNAIDS data 2019. 2019 . Available at:
https://www.unaids.org/en/resources/documents/2019/2019-UNAIDS-data.[Accessed 10 January 2021]
13. European Centre for Disease Prevention and Control. Continuum of HIV care. Monitoring Implementation of the Dublin Declaration on Partnership to Fight HIV/AIDS in Europe and Central Asia: 2017 Progress Report. 2017.
14. Nosyk B, Montaner JSG, Colley G, Lima VD, Chan K, Heath K, et al. STOP HIV/AIDS Study Group.
The cascade of HIV care in British Colombia, Canada, 1996–2011: a population-based retrospective cohort study.
Lancet Infect Dis 2013; 14:40–49.
15. Lo Y-R, Kato M, Phanuphak N, Fujita M, Duc DB, Sopheap S, et al.
Challenges and potential barriers to the uptake of antiretroviral-based prevention in Asia and the Pacific region.
Sex Health 2014; 11:126–136.
16. Pokrovskaya A, Popova A, Ladnaya N, Yurin O.
The cascade of HIV care in Russia, 2011–2013.
J Int AIDS Soc 2014; 17: (4 suppl 3): 19506.
17. Public Health England. HIV in the United Kingdom: towards zero HIV transmissions by 2030. 2019 report. 2019. Available at:
https://www.gov.uk/government/news/hiv-in-the-uk-towards-zero-hiv-transmissions-by-2030. [Accessed 10 January 2021]
18. Kato M, Long NH, Duong BD, Nhan do T, Nguyen TT, Hai NH, et al.
Enhancing the benefits of antiretroviral therapy in Vietnam: towards ending AIDS.
Curr HIV/AIDS Rep 2014; 11:487–495.
19. United Nations. United Nations World Population Prospects 2019. 2019. Available at:
https://population.un.org/wpp/. [Accessed 10 January 2021]
20. R Core Team. R: a language and environment for statistical computing. Vienna, Austria, 2018. Available at:
https://www.r-project.org/. [Accessed 18 November 2020]
21. Hill A, Wang J, Levi J, Heath K, Fortunak J.
Minimum costs to manufacture new treatments for COVID-19.
J virus Erad 2020; 6:61–69.
22. Hill AM, Barber MJ, Gotham D.
Estimated costs of production and potential prices for the WHO Essential Medicines List.
BMJ Glob Heal 2018; 3:e000571.
23. Gotham D, Barber MJ, Hill AM.
Estimation of cost-based prices for injectable medicines in the WHO Essential Medicines List.
BMJ Open 2019; 9:e027780.
24. Panjiva: S&P Global Market Intelligence.
25. World Health Organization. Updated recommendations on first-line and second-line antiretroviral regimens and postexposure prophylaxis and recommendations on early infant diagnosis of HIV. 2018.
26. Kempton J, Hill A, Levi JA, Heath K, Pozniak A.
Most new HIV infections, vertical transmissions and AIDS-related deaths occur in lower-prevalence countries.
J Virus Erad 2019; 5:92–101.
27. Giroir BP.
The time is now to end the HIV epidemic.
Am J Public Health 2020; 110:22–24.
28. Kaiser Family Foundation. US federal funding for HIV/AIDS: trends over time. 2019. Available at:
https://www.kff.org/hivaids/fact-sheet/u-s-federal-funding-for-hivaids-trends-over-time/. [Accessed 10 August 2020]
29. World Health Organization. Consolidated Guidelines on HIV Prevention, Diagnosis, Treatment and Care for Key Populations – 2016 Update. 2016.
30. Haberer JE, Sabin L, Amico KR, Orrell C, Galárraga O, Tsai AC, et al.
Improving antiretroviral therapy adherence in resource-limited settings at scale: a discussion of interventions and recommendations.
J Int AIDS Soc 2017; 20:21371.
31. Suthar AB, Rutherford GW, Horvath T, Doherty MC, Negussie EK.
Improving antiretroviral therapy scale-up and effectiveness through service integration and decentralization.
AIDS 2014; 28: (Suppl 2): S175–S185.
32. Du H, Chi P, Li X.
High HIV prevalence predicts less HIV stigma: a cross-national investigation.
AIDS Care 2018; 30:714–721.
33. Zukoski AP, Thorburn S.
Experiences of stigma and discrimination among adults living with HIV in a low HIV-prevalence context: a qualitative analysis.
AIDS Patient Care STDS 2009; 23:267–276.
