The majority of the HIV-infected women in this study who started ART for life during pregnancy because of low CD4 cell count had developed detectable viral load at 24 months after delivery. This finding was particularly surprising given that most of the women reported adherence and were clinically doing well. Most of the women acknowledged poor adherence and gave explanations for it after being informed about their viral load trend.
Previous studies have found that costs of transport to and from the clinic, health clinic fees, and lost wages due to long waiting times at clinics have been the main barriers to adequate ART adherence [11, 13, 23–25]. Our findings suggest, however, that despite provision of free ART drugs in an ideal setting with continuity among providers as well as “extras” like bus fare, free medical services, and management of opportunistic infections, it was still a challenge for the women to adhere adequately over a two-year period. We discovered that the motivation to adhere to ART dropped after cessation of breastfeeding and saving the child. The drop in motivation was compounded by poverty, the overwhelming demands of everyday life, and the fear of stigma if relatives and neighbors found out they were infected with HIV. Motivation was also affected by the fact that most women felt well and were clinically stable and had been put on ART because they were diagnosed during pregnancy. However, all the women had been informed that their CD4 cell counts were low and that they needed to adhere to ART for life. To our knowledge this is the first study to document a decrease of motivation to adhere to lifelong ART after cessation of breastfeeding.
Some of the other reasons we found for non-adherence have also been reported in previous studies: the stigma of HIV [24, 26], the forced secrecy around the disease and around taking ART, general poverty, the overwhelming demands of everyday life, a general sense of hopelessness, and a sense that one is well and therefore need not follow the demanding drug regimen [11, 13, 15, 27–30]. The gripping detail in which women described their travails, however, breathes life into the challenges to adhering to ART even among those women who need the medication for their own immediate health and survival, many of whom were single caretakers of their infants.
Most women in this study had consistently reported high ART adherence during the follow-up in the Mitra Plus study and some even kept insisting, until we showed them the viral load results. Social desirability may be particularly strong among mothers given the strong expectations they face from health care staff in terms of ART adherence and our study suggests that self-report is not a reliable measure as it may well exaggerate adherence rates. This is very important clinically as poor adherence and virologic failure have serious consequences including increased risk of morbidity and mortality and development of drug resistance which can be costly to manage in resource-limited settings [14–16].
Poverty and HIV-related stigma combined with cultural gender norms and traditions, significantly reduces women’s power to make decisions regarding their own health in this Tanzanian society, and is probably quite generalizable to many other urban sub-Saharan African settings. Women are generally less educated and often lack their own income making them dependent on men for survival [31, 32]. Lack of food and crowded housing conditions are real barriers to ART adherence directly linked to poverty. Public awareness of PMTCT and the relationship to breastfeeding and early weaning makes both adherence to ART and exclusive breastfeeding a major challenge especially among the most vulnerable, least empowered women due to societal stigma associated with HIV causing neighbors and relatives to closely observe and gossip around any signs of HIV-related deviations from normal post-partum behavior. This challenge may be dealt with by introducing structural interventions aimed at alleviation of poverty, food-shortages and the hardships of women’s lives such as educational opportunities, small scale business or job-training support, more involvement of fathers and partners in pregnancy and child-care at the community level. Efforts should also be made to reduce stigma and improve compliance by protecting and promoting the client’s right to privacy and confidentiality both in the community and in healthcare settings.
The major limitation of this study is that it was done at the end of the follow up period when women might have forgotten what really happened in the past 2 years. Case notes kept in the clinic were used to try to remind them of some events that happened as they were recorded at certain time points during their follow up period. The fact that MN worked in the project throughout the study period and was well known by the participants before the interviews could have caused some social desirability bias, where mothers answered with what they assumed would be the right thing to say, rather than what they actually practiced. However, to reduce such bias the interviewers worked to carefully ensure friendly environment to enable open discussions and only after some time told the women about their high viral loads that had raised our concern. Most women then volunteered what appeared to be truthful information regarding lack of adherence. We also underlined the separate roles of the first author of being a researcher rather than their clinician at the time of this study.