BACKGROUND: In sub-Saharan Africa, more than 20% of PLHIV present with advanced HIV disease (AHD). Our objective was to project the clinical outcomes, costs, and cost-effectiveness of the WHO-recommended AHD package of care in Malawi.
METHODS: Using the validated CEPAC-I model, we simulated a cohort of PLHIV aged =18y initiating ART with measured CD4 count using published data and assessed 7 strategies applied to people identified with AHD (CD4<200/µL and/or WHO 3/4 disease): 1) No specific AHD care (sputum Xpert for people with TB symptoms), and then sequential addition of: 2) urine LAM; 3) co-trimoxazole (CTX); 4) cryptococcal antigen (CrAg) and CTX; 5) LAM and CTX; 6) LAM, CTX, and CrAg; 7) LAM, CTX, CrAg, and isoniazid preventive therapy (IPT). The cohort had mean age 37y, 51% were female, mean CD4 362/µL (15% with CD4 <200/µL and 4% with WHO3/4 disease and CD4 =200/µL). Among people with CD4 <200/µL, prevalence was: 18-37% TB disease; 20-39% latent TB; 5% asymptomatic cryptococcal disease; 87% of people with TB disease had symptoms. Test costs were $13 (Xpert), $5 (LAM), and $3 (serum CrAg); medication costs were $0.83/month (CTX), $4/month (Fluconazole), $1/month (IPT). Model outcomes included life expectancy, costs, and incremental cost-effectiveness ratio (ICER, $/year-of-life saved [YLS]); we considered ICERs <$640/YLS (Malawi’s annual per capita GDP) cost-effective.
RESULTS: All AHD strategies would improve clinical outcomes and increase costs (Table). The full AHD package, LAM+CTX+CrAg+IPT, would result in the greatest life expectancy (21.61 life-years) and be cost-effective (ICER, $250/YLS). All other strategies would be less efficient than the full AHD package at the cost-effectiveness threshold. Results are most sensitive to TB and cryptococcemia prevalence.
CONCLUSIONS: Using published data, the full AHD package at ART initiation would be cost-effective in Malawi compared with only some elements of the package, when CD4 count is measured.

Table. Model-projected outcomes comparing the clinical outcomes, costs, and cost-effectiveness of different strategies for AHD care for PLHIV initiating ART in Malawi.
Strategy	1y survival (%)	Undiscounted life expectancy (y)	Discounted life expectancy (y)*	Lifetime costs ($)*	ICER ($/YLS)*
No AHD care (ART+Xpert)	93.89	20.89	13.62	1,354	--
+LAM	94.20	21.13	13.74	1,371	140
+CTX	94.31	21.21	13.81	1,412	dom
+CrAg+CTX	94.34	21.21	13.81	1,413	dom
+LAM+CTX	94.64	21.45	13.93	1,430	dom
+LAM+CTX+CrAg	94.66	21.46	13.94	1,431	dom
+LAM+CTX+CrAg+IPT	94.88	21.61	14.03	1,444	250
*Discounted at 3%/year. Dominated (dom): the ICER of this strategy compared to the next more costly strategy is higher and therefore not preferred. Abbreviations: AHD, advanced HIV disease; PLHIV, people living with HIV; ICER, incremental cost-effectiveness ratio; YLS, years-of-life saved; LAM, lipoarabinomannan; CTX, co-trimoxazole; CrAg, cryptococcal antigen; IPT, isoniazid preventive therapy.