Key results
In this cohort of individuals presenting to HIV care at two clinics in Nairobi, Kenya, approximately one-third presented to care with advanced HIV, suggesting important opportunities still exist to encourage earlier diagnosis and treatment. We found that delayed diagnosis was more common than delayed linkage to care in explaining presentation to care with advanced HIV. Although 59 % of those presenting with advanced HIV had been previously diagnosed with HIV, almost ¾ of these individuals presented to care within three months of their initial diagnosis. Given the average rate of decline of CD4 T lymphocytes [
28,
29], it is unlikely that many of the individuals who presented to care with advanced HIV within three months of their previous diagnosis would have had non-advanced HIV when they were initially diagnosed.
Overall, there was a strong linear increase in the likelihood of presenting with advanced HIV among age groups older than 30. The proportion of those presenting with advanced HIV also varied by clinic, but this may be expected as larger structural and contextual correlates are likely to vary in different care settings. In this instance, we noted that Baba Dogo lacks the same level of community outreach programs present in Kibera. Community outreach programs in Kibera include home-based HIV testing and counselling (HBTC) by various organizations, including Amref Health Africa, Liverpool VCT, and the Centers for Disease Control and Prevention (CDC). Kibera also benefits from numerous clinics at which HIV can be tested, including a Médicins Sans Frontières (MSF) clinic that opened during the time of recruitment for this study. While HBTC efforts exist in Baba Dogo, they are less prevalent. These community-based efforts may drive earlier testing and linkage to care, and may have led to comparatively fewer clients presenting with advanced HIV in Kibera.
Comparability with other studies
Various definitions of ‘advanced HIV disease’ have been used in the literature, and the term has frequently been used interchangeably with the term ‘late presentation’. Definitions of ‘advanced HIV disease’ from studies in sub-Saharan Africa have included: CD4 count <100 cells/μL or WHO stage 4 [
10]; WHO stage 3 or 4 [
9]; and CD4 < 100 cells/μL [
14]. The wide array of definitions used makes it difficult to compare the proportions of individuals presenting to care with advanced HIV across studies. For instance, in a large, multi-country study by Lahuerta et al., 19 % of those enrolling in care were classified as having advanced HIV [
10], compared to 33 % in our study; however, Lahuerta et al. used a lower CD4 threshold of 100 cells/μL, so more individuals might have been classified as having advanced HIV than if a higher threshold of 200 cells/μL had been used. The recent development of consensus definitions of ‘late presentation’ (CD4 below 350 cells/μL or presenting with an AIDS-defining event), ‘presentation with advanced HIV’, and even ‘presentation for care’ [
1,
17], and their use going forward, will facilitate comparison between studies in the future.
During the course of this study, the clinics transitioned from initiating treatment at CD4 counts of 350 cells/μL or less to the 2013 WHO’s recommendations to initiate treatment at 500 cells/μL or lower [
30]. The benefits of which include improved survival, immune recovery, and a decreased risk of transmission [
30]. With approximately one-third of patients presenting to care with advanced HIV, and over half of the population presenting with a CD4 count lower than 350 cells/μL, the majority of patients at these clinics will not be affected by the change in treatment guidelines, or more recent recommendations to initiate treatment upon diagnosis, regardless of CD4 count. Over the past decade, CD4 count at presentation has not markedly increased in sub-Saharan Africa [
7], and while it is too early to tell whether implementation of the new guidelines will promote earlier presentation to care, our study emphasizes that it is critical to develop and implement strategies that encourage earlier diagnosis. Without this, stated targets of expansion of therapy to those who are eligible and the intended individual- and population-level effects of the new WHO recommendations will not be fully realised.
The importance of earlier diagnosis is further supported by our findings that presentation with advanced HIV was largely due to delayed diagnosis, rather than a delay in seeking care after diagnosis. Prior studies on presentation with advanced HIV in the region did not examine prior diagnoses [
13], considered new diagnoses only [
14], or were based on clinical records [
9,
10], thereby restricting their ability to investigate the pathway to presentation with advanced HIV. In addition to our finding that approximately ¾ of those with advanced HIV who had been previously diagnosed presented to care within three months, the proportion of individuals who had had a previous diagnosis was similar between those with or without advanced HIV, and the median time to first presentation to HIV care did not significantly differ between the two groups. This supports our conclusion that advanced HIV at first presentation was primarily due to delayed diagnosis. This is not to underestimate the importance of promoting timely linkage to care; however, as almost ¼ of those with advanced HIV (who had been previously diagnosed) took longer than three months to present to care, and many individuals who test positive do not link to care [
31,
32].
Similar to Kigozi et al. in their Ugandan study [
9], we found that older age was associated with advanced HIV at presentation. This may be due to simply having lived long enough for the disease to progress to an advanced stage, or age-associated differences in HIV awareness, knowledge or stigma that may affect testing and other care-seeking behaviours [
33]. Reducing the barriers to and encouraging earlier diagnosis among older adults is particularly important because of the smaller gains made in CD4 response and increased risk of mortality compared to younger age groups once ART is initiated [
33-
35]. A study from South Africa found no association with age and presentation with advanced HIV [
15]; however, in the South African study, age was dichotomized at 40 years, which may have underestimated the variation in risk according to age [
36].
Although not statistically significant, relatively more men than women presented to care with advanced HIV. This is in contrast to other reports [
9,
10,
14], which found strong associations between male gender and presentation with advanced disease. This may have been due to the exclusion of pregnant women from this study. In studies on CD4 at presentation to care, those with a focus on prevention of mother-to-child transmission (PMTCT) reported a higher mean CD4 count at presentation (395 cells/μL) [
7] than non-PMTCT-focussed studies, and those enrolling in PMTCT services have been found to have a lower likelihood of presenting with advanced HIV disease than others [
10]. By excluding pregnant women in this study, the difference between the genders in the risk of presenting with advanced HIV may have been attenuated.
Other factors of interest, such as educational level [
9,
12], hazardous drinking [
9,
12,
15], and travel time to clinic [
9], have been found to be associated with presentation to care with advanced HIV in previous studies but were not in our cohort. There are several possible explanations for this beyond the different populations under study. First, the use of varying definitions of advanced HIV may underlie the differing results: factors that are associated with advanced HIV when a lower CD4 threshold is used may not be similarly associated with advanced HIV when the consensus definition is applied. Other possible explanations include high levels of missing data in some previous studies, which may have impacted findings; investigation of a large number of variables, increasing the likelihood of chance findings; and a larger sample size in some of the previous studies, increasing the power to detect effects.
Strengths and limitations of the study
One of this study’s principal strengths is our collection of information on the date of participants’ first HIV diagnosis, helping us to illuminate the pathway to presentation to care with advanced HIV. A second major strength of this study is the completeness of the data: CD4 data was available for 97 % of the cohort, and information on additional variables, such as alcohol or illicit drug use, was available for all participants. Additional strengths include a high participation rate, which minimized the possibility of non-participation bias. This, combined with the inclusion of two sites in the study, improves this study’s generalizability.
We used a consensus definition of advanced HIV, part of which is based upon presenting with an AIDS-defining event regardless of CD4 count. WHO stage 4 was used as a proxy for AIDS-defining events. Although most AIDS-defining conditions are included in WHO stage 4, WHO staging data was only available for 85 % of the cohort, which may have led to the misclassification of some participants as non-advanced HIV. Furthermore, CD4 was measured at only one point in time. Given laboratory variability in CD4 measurements, and the possibility that other factors may temporarily influence CD4 counts [
1], it would have been preferable to have a confirmatory CD4 count.
Other limitations include the self-reported nature of data on the occurrence and timing of prior HIV diagnoses. There is a lack of data from Kenya on the validity of self-reported HIV testing data; however, studies from other parts of sub-Saharan Africa suggest that HIV-positive individuals may underreport past testing [
37,
38]. While the reasons for underreporting a previous diagnosis are not well understood, one possibility is that individuals fear being turned away from the clinic if they indicate they are aware of their status. At the study sites, care and treatment guidelines were developed to protect clients from being denied care. It is standard clinical practice to test all patients who come to the clinics for HIV care, regardless of whether they have been tested or diagnosed with HIV before. These guidelines may encourage more honest reporting on previous diagnoses than might otherwise occur. To improve the quality of self-reported data, participants completed the questionnaire after being assured of confidentiality. In addition, the questionnaire was administered by an experienced HIV research nurse in a private room. Finally, data from this section of the questionnaire were cross-checked with a later section to assess consistency. A high degree of consistency was found, and any discrepancies were investigated further and resolved. Despite the limitations inherent in self-reported data, the study is unlikely to suffer from recall bias. There is no strong reason to believe that the advanced HIV versus non-advanced HIV groups would differentially report events; however, data on the dates of previous HIV diagnoses may be less valid than if we had had access to clinical records.