Provision and Access

"Antiretroviral therapy has rendered HIV a treatable condition and people on effective therapy can have a life expectancy that approaches that of HIV-uninfected people" (Mastro et al., 2011: 2049). In Uganda, for example, patients receiving combination ART can now expect an almost normal life expectancy (Mills et al., 2011b). Globally, mortality rates from HIV declined by 19% between 2004 and 2009 (Atun and Bataringaya, 2011).

"I cook scones for my children and do not get tired. I do chores, pounding. When the sun rises... I go to the garden or farm. People say, 'You have tilled that garden on your own?'" HIV-positive woman on antiretroviral treatment, Malawi (Mkandawire-Valhmu and Stevens, 2010: 690)

Evidence has repeatedly demonstrated that antiretroviral therapy has been successfully administered in a range of situations and populations with good adherence, good patient retention, and good clinical outcomes in resource-poor settings common to many developing countries; results have been similar to those achieved in resource-rich countries. Increasing provision and access across all populations is critical to continuing that success. Further, accelerating treatment access for adults with young children can reduce the numbers of orphans, and improve pediatric survival and social wellbeing. [See Orphans and Vulnerable Children]

Treatment Must Be Equitably Available to All in Medical Need

Medical need for treatment is based on established guidelines for clinical staging and CD4 counts (see WHO guidelines below). Many countries have a treatment gap between those already tested and known to be HIV-positive and those who are able to access treatment. For example, in Kenya, two million people have HIV but only 300,000 are on treatment (Siringi, 2010). In sub-Saharan Africa, where the majority of those living with HIV reside, there is no health information system that adequately tracks those who are HIV-positive but not yet eligible for treatment according to their countries' national guidelines. Therefore, the vast majority is not retained in care until they are eligible to receive treatment (Rosen and Fox, 2011). "The immediate implementation of an aggressive strategy aimed at rapidly expanding antiretroviral therapy to all those in medical need, based on current medical guidelines, is fully warranted" (NIDA and IAS, 2010: 37).

To date, more women than men have accessed treatment. "Women are often more likely than men to attend health services because of dedicated provision of reproductive and child health clinics" (Braitstein et al., 2008b: 53). Data disaggregated by sex show that adult women are slightly advantaged over adult men in access to antiretroviral therapy in low- and middle-income countries. "Among 109 reporting countries, the estimated antiretroviral therapy coverage was higher among women, estimated at 53%, than among men (40%)" (WHO, 2011c: 15). A study in Thailand found that proportionally more women (53%) were receiving antiretroviral therapy than men, despite a higher proportion of HIV infections among men, in part due to initial prioritization of pregnant women to receive treatment (Le Coeur et al., 2009).


Gender roles and norms affect treatment access. A study in Zimbabwe found that many women felt unable to disclose their serostatus, forcing them to take their ARVs in secret and to do so without support from their partner, affecting adherence. Men saw HIV/AIDS as a threat to their manhood and their lack of participation in HIV services, along with multiple partnerships, left women vulnerable to re-infection. Women also reported their husbands stealing their ARVs (Skovdal et al., 2011b).

"Men will only come to us when they are bedridden and brought to us in a wheelbarrow." Nurse for HIV-positive men on treatment, Zimbabwe (Skovdal et al., 2011b)A study in Nigeria found that men were more likely to have a more rapid progression to death after initiation of HAART, most likely due to initiating treatment at CD4 counts of less than 50 (DeSilva et al., 2009). A study in Uganda found that men who received treatment were much more likely to die than women because of late presentation for treatment (Mills et al., 2011c). A review of the 631,985 patients on ART in South Africa found that men were more likely to be lost to follow-up and more likely to die than women (Klausner et al., 2011). Greater attention needs to be paid to ensuring that HIV-positive men know their serostatus, have access to condoms and understand the need for consistent and correct condom use, and have equitable access to treatment.

Some studies have found that equity in access differs by age group: In Malawi, 10,000 people are on treatment, with proportionately more females accessing treatment than men. However, in the 15 to 19 year age group, more men are proportionately on treatment despite the fact that HIV prevalence in this age group is higher among women. There were more women than men on treatment for ages 30 to 39, yet HIV prevalence in this age group is higher in men as compared to women. Key affected populations, such as sex workers and those who use drugs may face many barriers in accessing treatment.

Cost is another factor in treatment access. Providing treatment at no cost may substantially increase women's access to life-saving therapy. A study of AIDS-related deaths in Addis Ababa, Ethiopia found that following the launch of no-cost antiretroviral therapy in 2005, women died from AIDS at almost the same rate as men. Prior to no-cost antiretroviral therapy, more women than men died of AIDS, possibly due to sex differences in access to resources for financing treatment (Reniers et al., 2009). Treatment provided at no cost can substantially increase both women and men's access to life-saving therapy (Musoko et al., 2011).

"Steps to expand HAART access to cover persons requiring it based on current standards are urgently needed" (Amirkhanian et al., 2011: 775). "The most crucial issue for distribution is how to ensure that equity considerations are appropriately addressed in resource-poor settings when treatment is not available to all who need it" (Padian et al., 2011b: 271).

Ensuring equitable access to treatment is not clear-cut. For example, it is unclear whether access for women is higher than men simply because PMTCT programs facilitate HIV testing and treatment or whether women living with HIV who do not want or are unable to get pregnant still have more access than men to treatment (Eyakuze et al., 2008). Still, "the need for increased and equitable access to AIDS treatment cannot be overstated" (UNAIDS and WHO, 2004 cited in UNAIDS et al., 2004a).

Sex Differences May Also Play a Role in Clinical Outcomes and Must Be Further Explored

"For too long, medical research has turned a blind eye to differences in disease prevalence, progression, and clinical outcomes between women and men. Women continue to be underrepresented in clinical trials... Even in studies in which both men and women participate, systematic analysis of data to identify sex-based differences is lacking" (Heidari et al., 2012b: 2). While in AIDS treatment, access by sex has been disaggregated in many studies, few studies have analyzed sex differences. "In general, studies to date have not shown differences in virologic efficacy of ART by gender" (Collazos et al., 2007; Fardet et al., 2006; Currier et al., 2010 cited in DHHS, 2011: 103), although a number of studies have suggested that gender or sex may influence the frequency, presentation, and severity of selected ARV related adverse events (Clark et al., 2005 cited in DHHS, 2011; Hawkins et al., 2011 cited in Johannessen, 2011). A study found differences in virologic failure by sex, with women having an advantage over men (Firnhaber et al., 2012b). Another study also found that women were more likely to have viral suppression and better ART outcomes (Kipp et al., 2010) and another study found that women were in better clinical condition than men (Nunes et al., 2010). "Although data are limited, there is also evidence that women may metabolize and respond to specific medications, including ARV drugs, differently than men" (Gandhi et al., 2004; Floridia et al., 2008; Ofotokun et al, 2007 cited in DHHS, 2011: 104). There may be different normal ranges for CD4 counts for women as compared to men (Prins et al., 1999 cited in Heidari et al., 2010). In one study in Uganda, clinical presentation and responses of Kaposi Sarcoma differed between men and women (Phipps et al., 2010). In some studies, women experience more adverse drug reactions than men (Hasan et al., 2011).

In a study of 4,531 women from numerous treatment sites in Sub-Saharan Africa, one-third experienced a pregnancy within four years of ART initiation (Myer et al., 2010). Yet, "the design and operation of most HIV treatment services do not explicitly acknowledge the likelihood or occurrence of pregnancy" (Myer et al., 2010: 8). In most settings, the desire for a child for women living with HIV is not discussed in HIV treatment either (Sedio, 2008b). And women who test HIV-positive in pregnancy may be lost to follow up HIV care postpartum (Kaplan et al., 2008 cited in Geng et al., 2010a). Yet most of the world's women living with HIV are of reproductive age and will need either contraception, discussions on how best to safely become pregnant and/or have a safe motherhood and reduce the risk of vertical transmission. [SeeMeeting the Sexual and Reproductive Health Needs of Women Living With HIV and Safe Motherhood and Prevention of Vertical Transmission ]

"The need to better understand the potential role of sex differences in HIV disease progression and treatment response is being increasingly recognized by the research community as an understudied area of inquiry" (Heidari et al., 2010: 4), with women under-represented in clinical trials (Monteforte et al., 2010). But at a minimum, "research data should be disaggregated by sex to ensure opportunities for gender-based analysis..." (Heidari et al., 2010: 70). More research is needed for HIV treatment for women (Abdool Karim et al., 2011d). In addition, women may be more affected by lipodystrophy (Tsuchiya et al., 2012; Han et al., 2011; Elliott et al., 2011) and may be more adversely affected by the common side effects of ARVs that result in anemia. [See also Adherence and Support]

Increased Access Must Also Include Respect for Human Rights

Expanding comprehensive medical services for women living with HIV and providing multiple entry points for careincluding antenatal, family planning and other sexual and reproductive health care services and psychosocial supportwill be essential to increase womens access to optimal ARV treatment. The benefits of treatment access go beyond improvements in health status and can include increasing employment and income for people living with HIV. [See Promoting Women’s Employment, Income and Livelihood Opportunities] However, fear of stigma and discrimination associated with HIV/AIDS may deter women living with HIV from seeking treatment as women living with HIV are at increased risk for being blamed as the source of infection and face more severe consequences of stigma (Hong et al., 2004; Maman et al., 2001a).

Regardless of who has better access to treatment, human rights must be respected. Requiring people living with HIV to disclose their serostatus to sexual partners and/or community members in order to receive treatment, care or support is a human rights violation. Similarly, coercing women to accept contraception in order to access treatment violates women's rights to make their own fertility choices. [See Meeting the Sexual and Reproductive Health Needs of Women Living With HIV] Further, while "treatment buddies" can be supportive, requiring a treatment buddy or medical companion to access ARV therapy may place undue burdens on women and their children: a study of 1,453 patients in Uganda (71% female) on the impact of requiring patients to disclose their HIV status and have a "treatment buddy" or "medical companion" to access ARV therapy found that of the women, 41% chose a child as their medical companion versus 14% of the men. Women feared disclosing their serostatus to their husbands. Only 31% of married women chose their husband as their medical companion, compared with 66% of married men who chose their wife. The majority (75%) of women were either widowed, separated or divorced; whereas 66% of the men were married. One woman noted that a six-year-old child was tasked with reminding her to take her drugs. Most men stated that wives were best for reminding them about ART, as "she is the one that takes care of you" (Foster et al., 2010b: 39).

The Cost of Treatment and Service Delivery Varies Widely

"I think that this medicine [ART] is as precious as my life" Woman on ART in Cambodia, (Elliott et al., 2011: 689)Treatment costs vary widely between sites but per-patient costs drop rapidly as care is scaled up. Calculated costs include medication and other costs associated with providing treatment and follow-up. Costs for providing treatment and care for a newly initiated adult on ARVs ranged from US$61 in one site to US$3,301 in another (Menzies et al., 2011). Even within the same country -- Kenya -- studies found a wide range of costs per patient per year, ranging from US$77 in one site to US$1,160 in another site (Harding et al., 2010). Others have estimated the cost at US$430.64 per person per year (AIDS2031 Consortium, 2010 cited in IOM, 2011: 179). Newly initiated ART patients entail costs of 15 to 20% more than those for established patients (Walker et al., 2011). The use of generic drugs can save significant sums of money, however. Annual surveys from 2005 to 2008 of ARVS purchased in 16 countries by PEPFAR implementing and procurement partners found that availability of generic ARVs was associated with increased ARV procurement and cost savings of more than US$323 million over the four year period, with generics accounting for almost 90% of the 22 million ARV packs purchased with PEPFAR funds in 2008 (Holmes et al., 2010b). Adequate funding for treatment is critical. Reductions in HIV funding threaten to interrupt treatment provision. Interrupted or episodic treatment increases individuals' risk for disease progression and death (SMART, 2006; Kaufmann et al., 2011).

How to integrate HIV care into accessible, nonstigmatizing services into national health systems remains an ongoing challenge. "It is still common in Africa to see newly constructed, well-staffed HIV clinics side by side with crumbling primary health care facilities, with little integration and few linkages between services" (Pfeiffer et al., 2010: 2). In addition, concerns have been raised that HIV/AIDS efforts may shift scarce resources away from other urgent health priorities, rather than create a synergistic effect. But it is important to keep in mind that "The rapid expansion of funding for HIV/AIDS programming provides a unique opportunity to improve all primary health care services... [showing] that rapid ART scale up and system-wide strengthening must go hand in hand" (Pfeiffer et al., 2010: 8). Integration of HIV care services into primary care systems can simultaneously strengthen both HIV services and the broader system in which these services are embedded. For example, a recent analysis of 1,538,612 adults, including 60,303 deaths from 41 surveys in 27 countries found that between 2004 and 2008, all-cause mortality declined more in countries which were the focus of PEPFAR efforts (Bendavid et al., 2012), though "ongoing research is needed to identify best-practice service delivery models" (Filler et al., 2011: e1). [See also Structuring Health Services to Meet Women’s Needs]

The Optimal Timing of Initiating Treatment is Still Being Evaluated

Current WHO guidelines, issued in 2010, state that antiretroviral therapy should be initiated when CD4 counts are below 350 cells/cubic mm. However, the optimal time to initiate treatment is still a subject of considerable debate. "The effect of the timing of the initiation of antiretroviral therapy on clinical and microbiologic outcomes has been controversial in evaluations of the benefit of therapy and of the associated short-and long-term complications and costs" (Cohen et al., 2011a: 494).

Some countries have found lack of lab facilities to measure CD4 counts a major barrier and WHO guidelines need to be adapted to country contexts. When HAART is initiated at lower CD4 counts and with higher viral load, patients develop more AIDS-defining illnesses (De Beaudrap et al., 2010), thus HAART-scale up can decrease the incidence of AIDS-defining illnesses. However, as people living with HIV on ART live longer, there may be an increase in AIDS-associated cancers and other malignancies (Sasco et al., 2010). Also, if antiretroviral therapy fails, patients also develop more AIDS-defining illnesses.

Guidelines from the United States call for initiation at CD4 counts of up to 500, with initiation of treatment with CD4 counts over 500 optional. A study from resource rich settings found that "Initiation of cART at a threshold CD4 count of 500 increases AIDS-free survival. However, mortality did not vary substantially with the use of CD4 thresholds between 300 and 500" (HIV-CAUSAL Collaboration, 2011: 509). Based on data from 309,435 patients from 48 countries, including low- and middle-income countries, median CD4 count at the start of ART increased in most countries, although CD4 counts at initiation for low- and middle-income countries remained under 200 (Mugglin et al., 2012a).

"Our ongoing scientific and moral challenge will be to continue to narrow the gulf between north and south and to ensure that we do not accept the establishment of two standards of care: one for richer countries and the other for the poorer..." (Granich et al., 2010: 4-5). "... The benefits of long-term ART for those with CD4 counts above 500 are unknown" (Matthews et al., 2010: 1979). However, early data from the US found that initiating treatment at CD4 counts above 500 was correlated with a 2.6 fold increase in viral suppression (Geng et al., 2012a). "There remains substantial uncertainty in terms of the benefit/risk balance for initiation of ART at CD4 cell count over 350 in resource-limited countries" (PEPFAR, 2011a: 9). [See also Antenatal Care - Treatment]

All guidelines agree that patient readiness is key in deciding when to initiate treatment (NIDA and IAS, 2010). Treatment for HIV begins prior to antiretroviral therapy with access to routine monitoring of HIV infection, including diagnosis of opportunistic infections, and routine testing of CD4 counts (WHO, 2009j; WHO, 2006b). Currently, NIH is conducting a clinical trial (Strategic Timing of Antiretroviral Treatment - START) to assess when is the optimal time to initiate antiretroviral therapy, with results expected in 2015 (www.clinicaltrials.gov; Dieffenbach and Fauci, 2011).

Universal Access to Treatment Can Be Achieved

Universal access to antiretroviral therapy in low-income countries can be achieved. A number of countries from Brazil to Ethiopia to Thailand are achieving progress in increasing the numbers of those initiating ARV treatment at recommended CD4 thresholds (Marcellin et al., 2009; ART-LINC, 2008; Kloos et al., 2007; WHO, 2007c). Studies in Tanzania and Kenya have shown that treatment, including the availability of fixed-dose combination antiretroviral therapy, can also be effectively used for children in resource-poor settings with good outcomes in CD4 counts and nutritional status (Ble et al., 2007; Nyandiko et al., 2006; Calmy et al., 2006).

Progress is being made with treatment access. For example, in some settings, patients who were initiated into treatment in more recent years are initiated at higher CD4 counts than in the earlier years of scale-up, and the mortality rate of those initiated has been lowered in more recent years than at the start of scale-up (Mulissa et al., 2010). A retrospective study in rural Malawi from 2000 to 2007 that reviewed 18,473 deaths found a highly significant downward trend in death rates of 37%. Death rates decreased over time as the percentage of people living with HIV could access ART (Mwagomba et al., 2010).

"Yes, I have been told that I have the assurance of receiving drugs for another three years; but what will happen after that? We are told these drugs are very expensive, so wont we die just like chickens?" HIV-positive person currently on treatment, Uganda (Russell and Seeley, 2010: 378)

However, caution is warranted and prevention efforts must continue. Since the scientific evidence now shows improved patient outcomes if patients are initiated at CD4 counts of over 350 and WHO has changed treatment guidance to CD4 350 instead of 200 to reflect this scientific evidence, patients loads are expected to increase (Konde-Lule et al., 2011). Anticipating now how best to deliver ART to "the ever growing cohort" is needed. Better delivery includes enhancing simplicity, efficacy and cost effectiveness of delivery; designing simpler and cheaper ART protocols that are less toxic and easier for health services and patients; piloting innovative service delivery outside of health services, with community or expert patient-based models; reducing drug stock outs; and understanding how to address the gendered dimensions of treatment access and adherence (Zachariah et al., 2011a).

In addition, for both men and women, better systems are needed to reduce the attrition by those who test HIV-positive in accessing ART once they reach national eligibility guidelines (Zachariah et al., 2011b). Tracking those who are alive and on therapy; dead; stopped treatment; transferred to another facility; and those who were lost to follow up is vital to gauge program success. A study that assessed outcomes for patients lost to follow up and who were then tracked in 17 studies from sub-Saharan Africa found that combined mortality was 46% (Brinkhof et al., 2009). Linking the data system with drug forecasting and procurement is needed. Uninterrupted ART drug supplies are essential. "ART regiments need to be as non-toxic and simple to take as possible and they must be offered free-of-charge to the patient" (Harries et al., 2010a: 72).

Identifying what works in treatment overall is beyond the scope of this resource. Some of the basic guidelines are included here and apply equally to everyone. [For more on clinical guidelines for treatment, see WHO guidelines:
http://www.who.int/hiv/pub/arv/adult2010/en/index.html (WHO, 2010g)]