Linkage to HIV, TB and non-communicable disease care from a mobile testing unit in Cape Town, South Africa

Abstract

Background: HIV counseling and testing may serve as an entry point for non-communicable disease screening.

Objectives: To determine the yield of newly-diagnosed HIV, tuberculosis (TB) symptoms, diabetes and hypertension, and to assess CD4 count testing, linkage to care as well as correlates of linkage and barriers to care from a mobile testing unit.

Methods: A mobile unit provided screening for HIV, TB symptoms, diabetes and hypertension in Cape Town, South Africa between March 2010 and September 2011. The yield of newly-diagnosed cases of these conditions was measured and clients were followed-up between January and November 2011 to assess linkage. Linkage to care was defined as accessing care within one, three or six months post-HIV diagnosis (dependent on CD4 count) and one month post-diagnosis for other conditions. Clinical and socio-demographic correlates of linkage to care were evaluated using Poisson regression and barriers to care were determined.

Results: Of 9,806 clients screened, the yield of new diagnoses was: HIV (5.5%), TB suspects (10.1%), diabetes (0.8%) and hypertension (58.1%). Linkage to care for HIV-infected clients, TB suspects, diabetics and hypertensives was: 51.3%, 56.7%, 74.1% and 50.0%. Only disclosure of HIV-positive status to family members or partners (RR=2.6, 95% CI: 1.04-6.3, p=0.04) was independently associated with linkage to HIV care. The main barrier to care reported by all groups was lack of time to access a clinic.

Conclusion: Screening for HIV, TB symptoms and hypertension at mobile units in South Africa has a high yield but inadequate linkage. After-hours and weekend clinics may overcome a major barrier to accessing care.

Figure 1. Schematic of the study inclusion and exclusion pathway.

Of the 9,806 clients screened at the mobile unit between March 2010 and September 2011, 276 were eligible for follow-up in the HIV-infected cohort and 488 were eligible for follow-up in the TB and other chronic diseases cohort. 1A: ^=of the 541 newly-diagnosed HIV-infected clients, 512 (95%) had a POC CD4 count done, of whom 502 (98%) received their CD4 count result. *= The yield of new diagnoses indicates the proportion of clients screened and who were newly-diagnosed with HIV, 1 or more TB symptoms, diabetes and hypertension, in that specific hierarchical order. Individuals have not been double-counted. 1B: ^^= There was no significant difference in sex between those that were eligible and those ineligible to participate (p=0.13). $= 21 TB suspects were also newly-diagnosed with HIV. Thus the total number of clients that met the study’s eligibility criteria is less than the cumulative number of clients in the HIV-infected cohort (n=294) and TB and other chronic diseases cohort (n=504), as it excludes these 21 TB suspects from the TB and other chronic diseases cohort who are accounted for in the HIV-infected cohort. A large proportion of clients did not meet the study’s eligibility criteria due to several reasons (i.e. were HIV-negative or known HIV-positive patients, were known TB suspects/diabetics/hypertensives, were diagnosed outside the pre-specified study enrolment period, left without obtaining a referral letter or consent, had no locator details, were < 18 years);. $$= each individual is represented once in one of the three groups (i.e. TB symptoms, diabetes, hypertension); #=individuals that did not receive the full allotted time to link to care and self-reported they never linked to care or attempted but failed to link to care were excluded from the analysis.