Social health inequalities during the course of chronic HIV disease in the era of highly active antiretroviral therapy

Introduction High-income countries have seen major shifts in the HIV epidemic since the mid-1990s [1]. The dynamics of the epidemic have changed, with an increasing proportion of persons infected through heterosexual contacts and thus an increasing proportion of women living with HIV/AIDS. Moreover, the epidemic appears to have shifted towards more socially vulnerable populations: HIV is moving into poorer and more deprived communities, and the proportions of infected young adults belonging to ethnic minorities and of disadvantaged young women have increased sharply [1]. At the same time, the wide-scale diffusion of highly active antiretroviral therapy, known as HAART, first introduced in 1996, has led to rapid changes in the natural history of HIV disease. In most cases, HAART has slowed down or even stopped the progression toward AIDS and death that was so relentless and almost continuous before then. HIV infection has, therefore, become a chronic disease in settings where HAART has been widely introduced. Important problems accompany these new therapies, however: their efficacy largely depends on adherence to treatment; toxic effects are frequent and can severely reduce the quality of life and augment cardiovascular risks; the emergence of resistance to these drugs, which is promoted by poor adherence, results in treatment failures; finally, the complexity of these treatments is even greater in case of comorbidity, which is relatively common in these patients. In particular, co-infection with hepatitis C virus is widespread among intravenous drug users. For all of these reasons, the benefits of the most recent advances in HIV infection management may differ among groups of patients independently of their clinical characteristics, and this can lead to inequalities in the consequences of the disease in settings where HAART is widely used. In particular, socioeconomic characteristics may play a major role in these inequalities. The epidemiological study of these social inequalities may be approached from two points of view. The first approach looks at whether social characteristics influence the effects of the chronic disease on morbidity and mortality. The second considers the social consequences of chronic HIV infection: how does it affect the patient's social situation and do these consequences differ according to the patients’ baseline social characteristics? Social inequalities have already been reported in the context of other chronic diseases. We present here a summary review of the data available to document the existence of such inequalities in the particular context of chronic HIV infection, to characterize them and to attempt to understand how they are generated. Furthermore, we present an overview of current knowledge concerning social inequalities in the more global context of chronic disease in general. Social inequalities in the context of chronic disease Several chronic diseases, including various cancers [2–4], diabetes [5], rheumatoid arthritis [6–8], ischaemic heart disease [9] and self-reported limiting long-standing illnesses (LLI) [10,11] have served as a framework for the study of social inequalities. Studies have shown that socioeconomic characteristics significantly affect health status among patients living with a chronic disease. Survival after diagnosis of cancer and ischaemic heart disease has been shown to be associated with socioeconomic factors, including ethnic origin, social class and marital status, with the risk of death highest for minorities, the lowest social classes and the unmarried [2–4,9]. Potential causes have been suggested for these inequalities in survival, including differences in the degree of disease advancement at the time of diagnosis and treatment start, in the type of care and treatment prescribed, in treatment adherence and in associated comorbidity. No study has formally tested these hypotheses, however. Differences in the type of treatment according to ethnic origin have been reported among patients with ischaemic heart disease [12], chronic kidney failure [13] and breast cancer [14], with patients of ethnic minorities more likely than the others to receive suboptimal therapies. Level of social support has also been reported to be associated with survival after a diagnosis of breast cancer: patients with the highest level of social support have the lowest risk of dying [15]. A study conducted in Finland, however, showed that the social gradient in mortality observed in the general population did not appear among the population of diabetic patients. The authors concluded that the explanation probably lay in Finland's highly equitable health system, which provides equal treatment to all patients, regardless of their socioeconomic characteristics [5]. Further data report on the social impact of various chronic diseases and its variations according to socioeconomic characteristics. Several studies have shown that diabetes affects employment rates and resources to only a limited degree [16–18]. Rheumatoid arthritis, by comparison, is associated with high rates of disability and consequently with a decrease in income level [7,19,20]. The risk of disability among patients with rheumatoid arthritis increases with disease severity and duration of illness, reaching approximately 40% by 2 years after diagnosis [19,20]. Nonetheless, after adjustment for disease characteristics, patients with the lowest educational level, the poorest psychological status, the highest age and those with the most physically demanding work have a higher risk of disability than the others [7,8,20]. Studies conducted in the general population have also shown that self-reported LLI significantly affects employment. In the Netherlands, reporting a LLI was associated with increased job mobility for men and women and an increased risk of stopping work among men [21]. In the UK, two studies have shown that an economic downturn has the most adverse effect on employment rates among persons reporting a LLI, especially among the lowest social classes [22,23]. In Sweden, a study has found that the adverse social consequences of recession, including economic inactivity, unemployment and financial difficulties, are reported more frequently by workers with LLI and that, in this group, the risk of economic inactivity is higher among manual than non-manual workers [23,24]. Finally, chronic disease, and particularly rheumatoid arthritis, has been associated with a decrease in the size of patients’ social networks [8, 20,25]. Social inequalities in the course of chronic HIV infection Impact of social characteristics on health status and its determinants in the context of HIV infection Many prospective surveys aimed primarily at studying the prognosis of patients with HIV have focused secondarily on the influence of socioeconomic factors on the clinical and/or biological progression of the disease. Some of these studies have reported gender, ethnic group, transmission category, income level, occupational status or educational level to be associated with disease course [26–32], but these associations have not been reported consistently. The extreme diversity of studies in terms of methodology, population samples and cofactors considered suggests that a more systematic approach would be useful in studying the influence of social situation on HIV disease. The effect of socioeconomic factors may be explained by processes that were not considered when these various studies were planned. Several types of difference associated with sociodemographic characteristics may explain the social inequalities reported in these studies: Differences in access to health care The rapid evolution of antiretroviral treatments in recent years has made obvious the differences in health-care access among people with HIV. For example, in the United States and in Italy, the use of HAART has revealed differences in patient survival according to sociodemographic characteristics, differences that were undetectable in the absence of effective treatment [30,31]. In France, national AIDS case surveillance data showed that more than three quarters of the patients who developed full-blown AIDS during the first quarter of 2001 had never received any antiretroviral treatment: 52% had not been aware that they were HIV seropositive, 24% knew they were HIV infected but were not receiving treatment, and 24% were being treated [33]. Women and non-citizens were over-represented in the first two groups, a finding which suggests that access to HIV diagnosis and antiretroviral treatments was unequal even though both were available free of charge [34]. Differences in access to health care may occur at different steps of the process: the time until entering medical care, the type of health-care setting, the pattern of medical care use, the access to antiretroviral treatments, and the type of antiretroviral treatment. Time until entering medical care. Data from the United States and Europe have shown that delayed presentation for medical care for HIV infection is associated with ethnicity (belonging to an ethnic minority) and HIV transmission through heterosexual contact [32, 35,36]. Time to entering care includes the period between HIV infection and testing and the period between awareness of infection and presentation for primary medical care. The determinants of the duration for these two periods may be different. Health-care setting. Several studies have shown that the type of care and treatment depend on physicians’ experience with HIV infection [37,38]. In the United States, patients’ socioeconomic status [39] and whether they acquired HIV through drug injection [40] have been shown to determine these features. The physician's level of experience and access to antiretroviral treatments also depend on whether patients are followed in urban or rural areas [41]. However, for patients followed in the same type of health-care setting, sociodemographic characteristics do not affect their disease course, treatment adherence or type of treatment [42]. Pattern of medical care use. In the United States, ethnic origin, gender, HIV transmission category and socioeconomic status have been shown to be associated with patients’ pattern of care for their HIV infection [43]. Two major patterns can be distinguished: white non-drug user males tend to use outpatient clinics or private physicians, whereas women, minorities and drug users tend to use emergency rooms. Likewise, in France in 2001, use of a private physician and type of hospital have been shown to be associated with patients’ sociodemographic characteristics [44]. Access to antiretroviral treatments. Several years after the introduction of zidovudine therapy in 1987, a study reported that access to this new antiretroviral agent in the United States had been delayed among patients who were minorities, women and drug users [45]. Since 1996, many studies in the United States and Europe have reported that, for patients at the same disease stage, access to HAART depends largely on sociodemographic characteristics, including age, gender, ethnic origin, HIV transmission category, educational level, occupational status and socioeconomic status [30,38,43,46–52]. For drug users, access also depends on continued illicit drug use and enrolment in a drug treatment program [38,53–55] even though ease of access to medical care enhances HAART use among active drug users [56]. Guidelines for antiretroviral treatments define patients with a CD4 cell count < 500 × 106 cells/l as eligible for therapy. Nonetheless, in 1999 in Baltimore, as many as 32% of HIV-infected drug users had received no antiretroviral treatment 3.5 years after their CD4 cell count had fallen below this level [57]. This heterogeneity may be explained by differences in access to health care and by physicians’ reluctance to initiate complex long-term treatment in patients whose compliance is expected to be poor [54]. Type of antiretroviral treatment. Several studies in the United States and Europe have shown that, in the group of patients taking HAART, the probability of receiving a treatment that includes a protease inhibitor differs according to gender, ethnic origin, HIV transmission category (especially history of drug abuse), history of incarceration and treatment with psychiatric medications [51–53,57–59]. Differences in adherence to treatments HAART is a complex combination of long-term treatments and as such presents significant difficulties in adherence. Perceived treatment-related constraints and side-effects may depend on the way disease is managed socially, and in particular on whether the disease is kept secret from family and friends [60]. Strict adherence to HAART is crucial to maintain a low viral load and to prevent the development of drug-resistant virus [61]. Demographic and psychosocial factors are associated with the degree of adherence [61–65]. First, socioeconomic difficulties, especially those related to emotional stress and alcohol or drug abuse, negatively affect treatment adherence [64,66]. Patients in precarious situations may express their difficulty in disease management by poor adherence [60]. Nonetheless, poor adherence is also encountered among those in stable, comfortable socioeconomic situations [67]. Youth is associated with poor adherence [62], as is continued drug use [55]. Adherence in patients with a past history of drug use, however, is as good as in those with no drug use history [68]. Adherence to HAART also depends on dynamic factors such as duration of the disease, the complexity of the regimen, medication side-effects, interactions with the health-care system, and negative life events [61–65,69,70]. Differences in comorbidity HIV and hepatitis C virus co-infection, which is particularly widespread among drug users, may accelerate the course of both diseases [27,71–74]. Mortality studies in Western Europe and in the United States have shown that end-stage liver disease as a complication of hepatitis C virus infection has become the leading cause of death among HIV-infected people since 1998 [75–77]. The high risk of other comorbidity among these patients (i.e., psychiatric diseases [40,78], pulmonary infections [79–81], tuberculosis or other sexually transmitted diseases) may affect the HIV prognosis, either by direct interaction with it or because of the even greater complexity of the care needed, or both [40]. A recent study in the United States has reported that depression is associated with clinical and biological progression of HIV disease [82]; it does not, however, provide information on the mechanisms by which depression adversely influences the disease course. Does depression act directly on the neuroendocrine and immunological systems? Or does this action come from secondary changes in behaviour (e.g., drug abuse) or adherence to treatments or health-care use, or some combination of all or several of these? Comorbidity also affects non-HIV-related mortality. In Canada, for example, among patients infected with HIV through drug injection and prospectively followed from 1996, only 34% of the deaths have been related to HIV [83]. Differences in living conditions, life style and health behaviours Living conditions and life style have consequences that may directly affect the course of HIV infection. A recent study has reported that cocaine exerts a specific effect that enhances HIV replication [84]. Aspects of life style are also markers of risk for other diseases or injuries, such as tuberculosis, suicides, fatal overdoses or accidents This is especially true for intravenous drug use [85–87], alcohol and/or psychoactive substance abuse [86], early unemployment [88], poor socioeconomic conditions [89], homelessness [87,89,90], incarceration [91,92], social and affective isolation, recent immigration or emotional stress. Such factors may also influence behaviours with their own effects on health, including diet, exercise and smoking. These behaviours may further enhance some HAART side-effects, particularly metabolic disorders associated with lipodystrophy [93]. Moreover, in the context of chronic disease, systematic preventive care (e.g., dental or gynaecological follow-up) independent of the disease itself is essential. Living conditions and life style may have an impact on all of these health behaviours. Differences in psychosocial characteristics Only a few studies have focused on the influence of social support on the course of HIV disease, although it has been shown to affect survival in other chronic diseases [15]. A recent study in California (USA) reported that social support improved the survival of parents with HIV [28]. Social support has also been associated with the way of coping with HIV disease [94] and with quality of life [95]. Social impact of chronic HIV infection Impact on occupational status Studies in the United States [96,97], Europe [44,98,99] and Australia [100,101] have reported high unemployment rates, ranging from 45 to 65%, among people with HIV in the era of HAART. The impact of HIV infection on employment status is major. In Australia, 35–50% of subjects who had been HIV infected for several years reported that they had lost their job since the HIV diagnosis [100,101]. In 1999 in France, 75% of the AIDES association clients who were unemployed had lost their job since diagnosis [98]. Occupational status is associated with the severity of infection, with the sickest patients least likely to still be working. Nonetheless, characteristics other than health status, including socioeconomic and job characteristics, may also have an impact on occupational status during the course of the disease, as reported for other chronic diseases (see above). Early in the HIV epidemic, Yelin et al. [102] reported that, at comparable disease stage, patients with the highest risk of losing their jobs were those with physically demanding work, those who had low control over the pace and scheduling of their work activities and those belonging to ethnic minorities. The level of work security (type of contract) may also be determinant: according to Virtanen et al. [103], health-related work selection may be stronger among non-permanent than permanent employees. As the disease becomes chronic, those with HIV must attempt to normalize their lives, a task that includes the need to (re)enter the workforce [100,101,104]. Returning to work is likely to be more difficult for those with still other disadvantages for job-seeking, e.g., low qualification levels or little (or long-ago) work experience. Moreover, people who receive disability benefits and who are qualified only for insecure or casual jobs may be less likely to consider returning to work, since (re)entering the workforce might jeopardize their benefits and thus their financial situation [105]. Further considerations include their employers’ attitudes and tolerance as well as workplace towards those with HIV Impact on resources and In Australia in and among a of persons with high educational levels and living with reported below the and reported that their had after the support was the major of income for difficulties were independently associated with employment status, health status and Moreover, reported they had since the HIV and reported that their current was for their health In two of France, of the in 2001 for HIV infection care reported below the [44]. Impact on social In France in 2001, of a of who had been HIV infected for a time of years [44]. This may their situation before the HIV diagnosis or from that followed HIV HIV infection also affects especially play a major role for HIV-infected than of the of persons in France reported that they had in different of their social life during the years [44]. The of and of the disease and the of family may between the different of patients. For example, may be stronger among women or minorities than among or effect on HIV infection can be considered to the patient's socioeconomic in various that it is a risk for and to this a dynamic process: the interactions between the patient's the changes by the of and the health-care [60]. HIV disease, other chronic diseases, causes in the social and of those their with time and with other people The health-care can in different social where the disease is from family and friends as a of by access to health-care and or by social contacts with support groups, Finally, may be increased by the treatment, which can be and have toxic impact on social also probably depends on how disease and health care have been in and Data reported in this review between health status, clinical characteristics, and the living conditions and life style of people with chronic HIV disease. These may have been early in the when the natural history of HIV infection social The diffusion of effective since and the of HIV infection into a chronic disease have revealed the existence of social inequalities during the course of HIV infection. Such inequalities have also been reported for other chronic diseases, including rheumatoid arthritis and ischaemic heart disease. is the absence of reported social inequalities among patients with probably related to disease the of a treatment, the limited clinical and the low level of Nonetheless, the mechanisms in social inequalities largely the that social characteristics affect health status and access to health care among people living with a chronic disease appears to be to the impact of these social inequalities. are several HIV infection in countries provides a particularly framework for further study of the health inequalities in chronic disease. First, the population is particularly and in its social characteristics. most of the subjects are young when they become and their social status is still this time and for many interactions between disease course and social This is not the case for other chronic diseases such as rheumatoid arthritis or ischaemic heart disease, which are likely to occur in persons more in in many health care for HIV infection is which appears to only limited financial to health-care This review has however, the of the available data on the the studies in the of social characteristics in surveys has been and most studies are to a specific of the population living with HIV/AIDS. Further studies conducted with among persons living with in countries where HAART has been on a would to the study of the mechanisms in social inequalities in the context of chronic diseases.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.