Admission characteristics, diagnoses and outcomes of HIV-infected patients registered in an ambulatory HIV-care programme in western Kenya.

OBJECTIVE To determine admissions diagnosis and outcomes of HIV-infected patients attending AMPATH ambulatory HIV-care clinics. DESIGN Prospective cohort study. SETTING Academic Model for Prevention and Treatment of HIV/ AIDS (AMPATH) ambulatory HIV-care clinic in western Kenya. RESULTS Between January 2005 and December 2006, 495 HIV-infected patients enrolled in AMPATH were admitted. Median age at admission was 38 years (range: 19-74), 62% females, 375 (76%) initiated cART a median 56 days (range: 1-1288) before admission. Majority (53%) had pre-admission CD4 counts <100 cells/ml and 23% had counts >200 cells/ml. Common admissions diagnoses were: tuberculosis (27%); pneumonia (15%); meningitis (11%); diarrhoea (11%); malaria (6%); severe anaemia (4%); and toxoplasmosis (3%). Deaths occurred in 147 (30%) patients who enrolled at AMPATH a median 44 days (range: 1-711) before admission and died a median 41 days (range: 1-713) after initiating cART. Tuberculosis (27%) and meningitis (14%) were the most common diagnoses in the deceased. Median admission duration was six days (range: 1-30) for deceased patients and eight days (range: 1-44) for survivors (P=0.0024). Deceased patients enrolled in AMPATH or initiated cART more recently, had lower CD4 counts and were more frequently lost to follow-up than survivors (P<0.05 for each comparison). Initiation of cART before admission and clinic appointment adherence were independent predictors of survival. CONCLUSION Although high mortality rate is seen in HIV-infected in-patients, those initiating cART before admission were more likely to survive.