Doctor, graduate in tropical medicine. He began working with MSF in 2016 on issues related to access to HIV care for men in the Homa Bay district of Kenya under the supervision of Jean-Hervé Bradol and Marc Le Pape. This research will form part of his medical thesis, which will be published in a CRASH book. Then, in 2019, he joined the oncology project in Bamako, Mali, as a palliative care doctor and researcher on the trajectories of breast and cervical cancer patients. He worked with MSF in Kinshasa as a doctor in a ward caring for patients living with HIV at the AIDS stage. Since 2022, he has been pursuing a master's degree in the sociology of health at EHESS which, in conjunction with CRASH, has led him to take an interest in palliative care practices in Malawi and the development of the discipline in a humanitarian context.
Chapter 4 - Discussion
I. LIMITATIONS OF THE STUDY
In order to allow the men to express themselves as freely as possible, rather than using closed-ended questions, we conducted the interviews in a conversational manner; this gave us more detailed knowledge about the population. Repeating the interviews to the point of saturation served the same purpose.
Subjects were recruited as opportunities presented themselves; meetings with informants led, when possible, to harder-to-reach segments of the population. In some cases the informants were people involved with MSF, resulting in selection bias, in that such people might have had an interest in presenting a positive view of themselves or MSF, and thus an inaccurate or partial view of reality. To minimize such bias we forged a number of ties to the population, and the meetings happened in different ways – through MSF, through the translator who was with us throughout the study, and directly, through various day-to-day encounters. Those personal encounters likewise introduced bias, in that they depended on the interviewers’ areas of interest and on their own representations, values, affinities, and ability to create connections. The interviews were done mainly during MSF-organized testing campaigns. Since the testing was done randomly in the villages and during scheduled testing sessions, the men we encountered varied in age, socioprofessional background, religious affiliation, and HIV status. One of the shortcomings of our approach was that the our population was not representative of the population as a whole, and that by recruiting people who had already been tested or who were being tested for the first time, we omitted an entire segment of the population that refused to be tested, and so were unable to study their representations regarding testing. I would address that issue by saying that the interview guide was constructed to retrace the subject’s history regarding testing before it was done and to elicit their reactions when they were offered testing and when they first became aware of HIV. By retracing their life histories, we learned about their past behaviors and practices, about the reasons for test refusal among men of different age groups, about acceptance of and doubts about the testing, and about the adoption of certain behaviors – preventive, in particular. In that way we were able to study a wide variety of representations regarding testing.
However, using anecdotal information about things that happened years ago may have introduced bias, since the veracity of subjects’ reports about past events may be questionable. To try to minimize the effects of this bias, we recruited men who differed according to how long it had been since their first test or diagnosis. Those who had been tested in the late 1990s or early 2000s gave us information about the representations of populations who had lived through the emergence of HIV and testing. That data was compared to the data from men tested more recently. To check the veracity of what they said about past events, the comparison looked for correlations and similarities in the representations, or category effects. This enabled us to see whether and how representations and practices had changed over time, in parallel with care and testing practices and with changes in the society as a whole.
The formal and informal interviews with organizational, governmental, medical, and religious stakeholders gave us information about the representations of the different actors involved in disseminating messages to the population. This also helped confirm or refute the information gathered during interviews with HIV-positive or negative men, and to back up ¬– or question – specific statements. One potential source of bias was my position as a researcher and, at the same time, a doctor (at least I was considered as such in most cases) – and an MSF doctor at that – and it sometimes felt as though I was being equated with the practitioners. That may have led to a situation where the interviewee wanted to “satisfy” the interviewer and give the “right” response. Seeing a researcher in a position of authority might cause the informant (in some cases a patient) to relate little from their individual perspective, and to talk about norms rather than his own personal practices or experience. To avoid that situation, we clearly stated that I would have no role in their care; I explained the aim of the study and how it would unfold, and emphasized the anonymity of the information collected. In addition, there was the question of whether my presence might alter the “normal” behavior of the patients questioned, or that of the doctors or counselors during the testing and treatment sequences I observed. As Howard Becker (64), and later Jean-Pierre Olivier (65), showed, an observer’s presence has only a minor influence on these operations, since the researcher has only a limited impact on the individuals, whose problems go far beyond the framework of the interview or the sequence the researcher is exploring.
Another question concerns my own subjectivity and involvement with the men with whom I had established a relationship. That is, whether creating that relationship changed my relationship to the study subject or my involvement with my interlocutors. The issue is knowing at what point emotional involvement, whether significant or modest, negatively impacts interpretation of the results. As the investigator, I vacillated between wishing to be included in order to obtain certain knowledge and wishing to remain objective.
One final source of bias, which is inherent to all qualitative research, has to do with the representativeness of the results detailed in our study. This relates mainly to questions about the population studied. Are the results applicable to the entire male Luo population? To what extent are they applicable to other male populations? Since this was not a statistical study, it cannot be used to study the distribution of identified behaviors and practices.
II. DISCUSSION OF THE MAIN STUDY RESULTS
Despite the fact that MSF has succeeded in placing several thousand individuals on antiretrovirals, no one has yet been able to reduce the very high incidence of HIV in Homa Bay to the national average. Yet reducing that incidence is necessary to achieving the UNAIDS goal of ending the epidemic by 2030.
In Homa Bay, the desire to achieve that goal has led to the targeting of so-called “vulnerable” populations with regard to HIV – i.e., women and children – on the one hand, and the apparent vectors of the disease – i.e., men – on the other. The effort to bring men into the health care system, however, seems to have been difficult, only partially successful, and the subject of debate on how best to raise their awareness and get them to take responsibility in the fight against HIV.
Few studies have looked at the specific characteristics of the male population or their personal experience with the epidemic in the region where this study was done.
We will therefore start by discussing the key distinct attributes of the male population facing HIV, and its reactions to institutional messages on prevention and care-seeking. Then, by comparing those distinct attributes to the data we obtained from observing the care being offered, we will attempt to assess how well the health care system has lived up to the male population’s expectations.
HIV, which is extremely severe in the region, has caused significant societal transformations. Recognition of the disease came late there, and the initial absence of a health care system capable of diagnosing or treating AIDS-related pathologies allowed the disease to become a plague; no solution seemed to emerge. Hence it was the oldest subjects that were most likely to equate an HIV diagnosis with almost immediate death. At the same time, the association between HIV risk and certain practices deemed morally reprehensible made it a stigmatizing disease. Those prejudices were due, in part, to the messages coming from institutions in high-income countries, as well as to messages from the HIV organizations working in sub-Saharan Africa. Though the HIV epidemic began in the early 1980s, political awareness of it in Kenya came later, and it was initially associated with certain high-risk practices, themselves attributed to a stigmatized segment of the population. This occurred in other countries – and in the global North, in particular, where HIV was first blamed on homosexuals alone, and then on the 4Hs: homosexuals, Haitians, hemophiliacs and heroin addicts (66). It gradually became apparent that the infection spared no one, yet the prejudices persisted. In Homa Bay especially, where the majority of residents are Luo, the severity of the epidemic was quickly blamed on traditional practices such as polygamy and levirate marriage. That belief was reinforced by the medical institutions. They considered such practices an “obvious” contributing factor, which they associated with what they believed was more – and thus higher risk – sex. At the same time, Western authorities legitimized the notion of a specific “African sexuality”, which they used early in the epidemic to explain the higher prevalence on the African continent. Here is anthropologist Jean-Pierre Dozon’s analysis of that ideology (67):
“[translation] Whether or not Africa was its cradle, it gave AIDS many springboards for its expansion, thanks to the excessively free and unbridled character of sexual mores (…). In that regard, any excuse can be used to support that assertion; some invoke traditions, attributing to them the anachronistic but performative expression, “sexual promiscuity” (which they seem to confuse with inherently instinctual polygamy); some scrutinize African modernity, finding only permissiveness and frivolity”
No studies have proven that such practices are responsible for the high HIV prevalence in this region, however. Several authors dispute the idea that sexuality differs between developing and developed countries, or that there is a difference that would explain HIV’s more rapid spread in some areas (68). Based on HIV studies done in Africa, some have pointed out the difficulty of proving that sexual behavior alone is responsible for the severity of the epidemic on that continent. They mention the possibility that higher HIV incidence is due to other transmission routes, and in particular medical procedures such as injections and transfusions (69) (70).
Though testing in Kenya began in the 1990s, the testing campaigns appeared ineffective because testing – without the possibility of subsequent treatment – unquestionably led to harsh social exclusion. Access to free antiretroviral drugs in the early 2000s offered some hope. HIV gradually evolved from a disease that was fatal in the relatively short term into a disease that could conceivably be survived, provided one took the drugs. That was a pivotal period – not just for the populations targeted by treatment programs, but for the programs’ sponsors as well.
With the arrival of antiretrovirals, populations and caregivers were faced with the problem of managing a chronic disease. How to provide information, prevention, diagnosis, and follow-up care to an entire population, especially in a rural area where the health system was less sophisticated than in Western countries, and where the HIV prevalence was far higher?
Once free drugs became available, poor treatment adherence became less acceptable. Then came the attempts to explain. It was at that point that assumptions blaming Luo culture for the lack of adherence to treatment campaigns first appeared (29). For example, campaign sponsors and doctors thought the population was using traditional medicine, rather than HIV testing and antiretrovirals. Yet it was shown that poor participation in HIV testing was primarily an institutional problem, and that testing was not being offered to enough of the patients consulting at hospitals. At the same time, the preconceived notion that the Luo population was at higher risk due to their continued engagement in traditional practices persisted, and was used to explain the higher HIV prevalence in the region.
Recently, the lower demand for testing and poor treatment adherence among men was being blamed on male “arrogance”, which supposedly led them to deny their vulnerability to HIV. In addition, the high HIV incidence in the region was being attributed to their “irresponsible” sexual behavior. And according to some members of the medical community, men’s “more unbridled and aggressive” male sexuality explained why they were less likely to be tested, out of fear of revealing high risk sexual behaviors (7).
A. PREVENTION MESSAGES AND BEHAVIORS
Several institutions are involved in disseminating prevention messages; they convey a variety of values and recommendations.
First, the communities with traditional values seem to have lost, at least in part, their role in educating the young about sex. Some have pointed out that the role of grandmothers in giving sex-related advice has disappeared. In addition, the value placed on formal education and abandonment of tradition in favor of a Christian education have led to the questioning of Chira, a belief that dictated rules of behavior considered by some a means of controlling sexuality.
The sex education dispensed by educational or religious institutions strongly advocates abstinence before marriage, and often treats condom use as secondary. We know that teachers have trouble tackling or adapting sex education; we know that adolescents reject the messages transmitted by educational institutions. One APHRC (African Population and Health Research Center) study on adolescents (71), published in a journal as we were doing our study, showed that they consider such messages moralistic and unrealistic. And as our interviews revealed, some adolescents found such messages hypocritical. There was a major incongruity between young men’s competitive sexuality (a source of social recognition) and the rigidity of the prevention messages, which failed to consider their motivations for engaging in sex.
When condom-related messages were conveyed, they generally emphasized men’s exclusive responsible for condom use. Often, in fact, only boys are instructed on condom use. As anthropologist Laurent Vidal showed (72), in an effort to promote virginity, girls were given no instruction on their future sexuality, and that lack of instruction rendered them incapable of negotiating condom use. Other studies have revealed that there is very little teaching on the existence and use of condoms (45) (73). One might wonder about shared male-female responsibility for the lack of condom use. Indeed, women are often unable to negotiate this prevention method due to an unequal power relationship between men and women. In addition, women seem to be ignorant or make erroneous assumptions about condoms. Men themselves describe having trouble negotiating the use of these prevention methods with their partners or spouses.
Information from the medical community does have an impact on perceived risk and prevention. Men seem to have the most respect for information and prevention advice from the medical community. But that medical advice is sometimes misunderstood, misused, or expressed poorly, causing some men to ignore their sense of vulnerability. Reasoning based on false or inadequate medical advice on prevention allows those men to ignore the advice. One study (74) analyzed the perception of risk prior to seroconversion. It found four types of risk perception or risk rationalization; one of them was called “protective reasoning”, in which someone constructs a more or less valid argument to justify ignoring their sense of vulnerability.
We have shown that access to HIV information dispensed by medical, educational, religious, and community-based organizations did not necessarily translate into awareness of the HIV risk – which we defined as either a change in behavior for the purpose of self-protection, or as an awareness of the proximity of HIV and a potential threat. While exposure to HIV-related mortality or morbidity was an important source of initial information for the men we met, it was above all the primary means of HIV awareness. In some cases, seeing the effects of the disease led to a complete rejection of sex, in particular among men who had no access to detailed information about prevention methods and whose awareness of the disease came abruptly. Some men modified their sexual behavior by having fewer partners, but did not necessarily use a condom. Lastly, some began using barrier protection, which they weighed against other concerns, such as their desire to have children or the potential difficulty posed by the negative image of condom use in couples.
HIV protection was not contemplated in isolation; it was fitted into life plans with varying degrees of ease, depending on the pressures the men faced. Their ability to adapt to the risk depended on how well they combined protective measures with personal desires and prevention with social pressure. For example, social pressure pushed the younger men toward early sexuality, which was associated – especially by those responsible for educating them – with a higher risk of HIV. Those responsible for transmitting educational messages prioritized abstinence-based messages over messages about condom use. Condom use turned out to be complicated for the younger men, because their partners sometimes feared it indicated risk-taking or even known, but undisclosed, seropositivity.
Many of the men we met believed that marriage – which confers another type of social recognition – offered adequate protection from HIV. But since marriage is only possible with enough income (which is difficult in the current context), access to that form of “protection” is limited. Indeed, men can only marry if they can meet their responsibilities by obtaining gainful employment. If that is not possible, they do not consider, or they delay, marriage. Financial difficulties caused the younger men to prefer non-marital relationships.
In some rare cases, they adopted alternative behaviors, attempting to keep their reputations intact while avoiding HIV infection. That involved resisting social pressure and rejecting the behaviors it encouraged – regarding sex, in particular. They based their reputation on alternative behaviors like valuing virginity, to take one example. The solution for men who found those alternatives impossible was social withdrawal, so that they could avoid the pressures around them and protect themselves from HIV.
B. FROM AWARENESS TO TESTING
The second objective of treatment campaigns was to promote testing, in order to get more people on antiretrovirals and reduce the incidence of HIV. Those campaigns encountered certain obstacles; as we have seen, launching a testing campaign was no guarantee that the population would participate. Several factors seemed to condition testing campaign participation in the male population, which was being targeted by actions like door-to-door and night-time testing when this study was being done in 2017. The testing strategy seemed to be conditioned, in part, on a view of masculinity based on reputation and resilience, on the importance of providing economically for the family, and on the social importance of marriage and fatherhood.
First, we have seen that when men first got information on HIV, depending on how they got it, they tended to blame the epidemic on different segments of the population. That categorization of risk seems to have influenced their perception of personal risk. Indeed, some individuals felt that belonging to a certain category protected them or made them vulnerable. The Christians provided the most striking example; they felt protected by their religion or its practices. They often contrasted themselves with traditionalists, who themselves began to feel that their practices – sexual practices, in particular – put them at higher risk for HIV.
Other characteristics like age, marital status, and occupation were used to divide up the population according to HIV risk. For example, Boda-Boda drivers are an occupational category thought to spread the disease, based on the fact that they have cash and/or can pressure their customers.
These contrasting attitudes with regard to HIV risk make prevention measures less effective. Those who feel protected from HIV (like the most religious, for example) slow the testing process, while those in the “opposite” categories are more concerned with such prevention measures, and will adopt them more or less efficaciously.
Second, marriage is considered for a number of reasons, but especially for social recognition – in particular, via fatherhood – or for its economic benefits. We did, however, observe some questioning of marriage due to HIV, either because marriage would mean having to be tested, which some were afraid of, or because the burden of being HIV positive, combined with the burden of marriage, would be financially or physically insupportable (30). Some of the men felt incapable of fulfilling their assigned role as the mainstay of the family (some, who were very young, because their father had died) if they were HIV-positive. Some refused, or postponed, testing. For men who were already married, testing depended on the perceived risk, with some considering their marriage a risk. Some, swayed by institutional messages, considered their polygamous unions high risk and willingly went to get tested. Those who considered their marriage protective were less likely to consider a test necessary. Finally, some preferred to be tested after marriage because being tested prior to marriage might make marriage impossible.
The relationship between fatherhood and HIV follows the same logic. People who are HIV-positive often think having children is impossible, due to the risk of transmission and the economic burden having children represents. While they may temporarily question marriage due to the need to be tested beforehand, this is less true with having children. Our interlocutors said having children was the most important source of social recognition; there is a lot of pressure to do so, causing some men who had firmly rejected marriage to reconsider, so that they could become fathers. Some men believed that finding out that they were HIV-positive would prevent them, physically and psychologically, from having children, since they equated being HIV-positive to having AIDS (fatal in the short term) and it would be too much of a financial or psychological burden.
Third, the fear of stigmatization – due to the importance of reputation – led to test refusal. Indeed, some men felt that their loved ones would see their going to a testing facility as evidence of sexual risk-taking and possible infection (63). In other cases, though rare here, men felt obliged to get tested to preserve their reputation, and because their loved ones might associate refusal with seropositivity. It is easy to understand how the desire to preserve family ties – often intimately related to reputation, particularly in traditional families – might cause test refusal. All the more so when a man has yet to develop symptoms and is afraid to disclose his status to loved ones. Indeed, disclosure sometimes leads to fierce rejection. Some of the men who were HIV-positive and dependent on loved ones saw those ties broken after disclosing their status.
Lastly, some men – even though aware of HIV – considered it like other diseases such as malaria, for example, and did not think it worth getting tested until symptoms appeared. Others failed to attribute the symptoms to possible HIV infection, and put off being tested. The difference between disease and seropositivity was unclear to some. As we saw, since a positive test was often equated with the disease and the disease was equated with symptoms and death, they felt they might as well wait for the symptoms to get tested and be treated as late as possible. Studies on similar contexts found this to be true as well (15) (23). Yet there is a huge difference between being diagnosed HIV-positive at the AIDS stage and at an earlier stage, particularly in terms of treatment efficacy, reducing opportunistic infections, and lowering the transmission risk. It is understandable that someone without a clear understanding of these differences and the benefits of early treatment, given that the treatment is not curative, would find it difficult to ask for a test that might on one hand endanger their reputation or marriage (75), and on the other, if positive, force them take prevention and treatment measures that might be an additional economic burden (76). The institutions that offer testing operate under a logic that is at odds with the representations and logic guiding the behavior of populations (15). For the former, testing – if positive – makes it possible to establish a treatment regimen to avoid the onset of the illness. For the latter, being tested amounts to being HIV-positive, or in any case considering oneself as such, and being HIV-positive amounts to having the disease, and thus the symptoms, with all their physical and social consequences.
In contrast, although the older generations associate HIV with high mortality, the younger generations more readily normalize HIV and testing. Indeed, seeing HIV-positive patients remain stable on treatment, and the widespread prevalence of the infection and information about the epidemic seems to foster that normalization. Normalization and the existence of a treatment can in some cases favor testing, because the younger generations do not associate testing with early death. (28) (77).
C. FROM TESTING TO ENTRY INTO THE HEALTH CARE SYSTEM
Whether a man tests negative or positive, he is asked to enter the health care system. Men who are HIV-negative are asked to get tested every three months, due to the ubiquitous nature of HIV. Those who are HIV-positive are asked to come in for treatment monitoring and follow-up visits. Both categories are asked to follow the medical advice given at the visit. Whether the person normalized HIV or associated it with rapid, certain death, the message delivered at the testing visit seems to have had a strong impact.
Most of the medical recommendations concern the sexual behavior of both HIV-negative and HIV-positive individuals. Due to the universal nature of HIV – which can be spread via blood, sex, or medical procedures – men often begin to question their own practices and those of their loved ones after being tested. All sex becomes potentially risky, even for HIV-negative couples, who are asked to stop having extra-marital – or even marital – sex, or to have it less often and use protection, despite their desire to have children. While the recommendations vary from one doctor to another, all are restrictive. As our study showed, it was rarely possible to follow them.
For some men, the medical recommendations were hard to reconcile with work, which was necessary to their survival, their family’s well-being, and their social status. Free antiretroviral drugs, first made available by MSF, did not prevent care as a whole from being hard for these men to afford. Now that treatments are available, HIV has become a chronic disease with a new set of problems requiring adaptation. First, according to the medical community, patients should not be on the drugs unless they are sure they can take them with food. Next, patients have to find a way to enrich their diet, in order to improve their health. Those recommendations were hard for most patients to follow, resulting in some degree of noncompliance. They would have had to increase their income-generating activities in order to obey the hygiene and diet rules. But instead, to complicate things, they were told to limit their physical activity, which was for most the only way to earn income, something that is up to the men in many families. According to the medical recommendations, an enriched diet, together with the medications, helps patients work better. That would increase their income, helping them afford a richer diet. And so on. The question is, where to start? For most of the men, following these recommendations to the letter left them very little leeway, since it was difficult for them to afford a better diet, and even more difficult to increase their income. That’s where the biggest economic difficulties came in, resulting, in particular, in the aforementioned change in male-female relationships. In addition, despite health care facilities being distributed throughout the country during the decentralization, the cost of transportation may render them inaccessible. Several men assured us that they wanted to stay away from the health care facilities closest to them to avoid stigmatization.
Once they became aware of HIV, both HIV-negative and HIV-positive men tended to change their behavior in response to the medical recommendations. This was seen frequently, whether the medical opinions took the form of a simple piece of medical advice or guilt-inducing recommendations implying a value judgement on past practices. Other studies have shown the negative impact of moral judgments (31). Moreover, the medical recommendations often seemed similar from one visit to the other, and so not very personalized, and were aimed especially at avoiding HIV transmission, even if that meant giving up other aspects of life – something they rarely seemed to consider.
In addition, there seemed to be some reappraisal of social ties. This often happened at the time of diagnosis, due in particular to exclusionary behaviors toward HIV-positive men. It may have been a question of self-exclusion, because the medical recommendations were incompatible with continuing the old way of life (drinking alcohol, for example). It may have been a question of being excluded from the family circle, because seropositivity is associated with imminent death or reputation loss. Some men developed strategies for incorporating the disease into their social relationships, while others found affirmation in their ability to face the disease alone. Still others did not have the resources to redefine their social relationships or their masculinity, and suffered as a result.
III. GENERAL CONSIDERATIONS
There are not enough teams working in the field during community-based actions; those that are there lack recognition (84) and are ill-equipped to deal with the expectations and the problems posed by seropositivity and serodiscordance (85)(86). Indeed, our observation of the testing sequence, in parallel with the interviews, showed that there was some difficulty tailoring the information given at each visit to each individual. Some of the HIV-related information was delivered in the exact same way from one visit to the next; there seemed to be no attempt to evaluate the individual’s characteristics or history, though these have an impact on risk-taking and on the patient’s familiarity with testing and treatment. In contrast, other information was delivered at the speaker’s discretion, and it was then that their own representations of the men and their health – and certain overly-simplistic assumptions – became evident. Male communication problems were often advanced to explain noncompliance by men, despite the latter’s reported desire to share, under certain conditions, their experiences and difficulties with HIV. One study (87) showed a connection between the quality of the testing and adherence to subsequent treatment. Noncompliance appeared to be related to weakness in the caregiver-patient relationship, difficulty accessing health care facilities, the fear of being stigmatized for going to those facilities, the inefficiency of the existing health care system, and the lack of help for patients. We observed that the messages delivered by some institutions – governmental institutions, in particular – were viewed with suspicion because those institutions were associated with corruption, most notably as a result of years of political instability and the marginalization of Luo society (88).
The limited use of care by the male population seems, at least in part, to be the result of a mismatch between the care offered and dispensed by people with certain representations (of what a man is and does) and the transformations occurring in the society and population in question (90). As we have seen, the medical recommendations – for those who wished to follow them – do not always fit into the lives of the men, who have other concerns pre-dating, or resulting from, their HIV infection.
They are asked to cut back on their sexuality (a measure that in our study appeared most frequently respected by the men who were considered highly sexual), their desire to have children, and their work. Studies have shown that some patients ignore or deny these recommendations in order to normalize their sex lives, for example (77). Moreover, in our study it appeared that some men, wanting to follow the recommendations, sometimes found themselves struggling to negotiate between medical advice and the need to remain anchored in their reality and their problems – economic problems, in particular. Although some could successfully adopt both systems simultaneously, for others they seemed irreconcilable. These men associated both systems with survival: survival through access to economic resources, which were also essential for getting their drugs and honoring their follow-up visits; social survival through work, sex, and fatherhood, and the respect that comes from those things; and finally, survival through adhering to their treatment regimen and medical recommendations that were supposed to protect them from a deadly infection.
More general studies have questioned the ethics of some HIV treatment interventions, particularly in sub-Saharan Africa (94). The increase in people being tested does not mean unconditional acceptance of the test – as the smaller number of people entering the health system after diagnosis attests. The acceptability of testing campaigns is questionable, because the procedures for obtaining consent and informing patients are sometimes weak. At home visits in Homa Bay, confidentiality is not respected and men are under a lot of pressure to agree to be tested in front of their family and, ultimately, disclose their HIV status. This would seem to make it hard for these men to refuse to be tested – especially since the population has a positive image of MSF, which might lead more people to agree to the test. Spousal testing has undeniable advantages in terms of subsequent care for members of the couple, women in particular, as it makes disclosure to their male partners less risky. Joint intervention has been shown to improve treatment adherence by both members of the couple (34) (95) (96) (97). New testing and care strategies such as decentralization and home-based testing obviously represent progress in epidemiological terms. In some contexts, there are new initiatives that are partially improving the testing rate among men – community-based actions, in particular – thus demonstrating an ability to work in unexpected domains (98). However, as in other contexts (99), this does not yet seem sufficient for reaching equal numbers of men and women in Kenya.