Table 1:

Characteristics of districts participating in intervention and control groups

	Intervention	Control
Number of clusters	7	7
Number of districts	43	39
Number of managers	86	77
District health officers	43	38
District tuberculosis supervisors	43	39
Sex of managers
Male	78(91%)	71 (92%)
Female	8 (9%)	6 (8%)
Regions
Southwest	13	12
East	12	11
East-central	18	16
Number of districts per randomisation cluster	5(5–6)	5(5–6)
Number of adults in active HIV care at the district level	5182 (2340–8346)	3456 (1949–8260)
Number of adults in active HIV care at the two largest clinics in each district	2099 (1270–3304)	1897 (1181–3378)
HIV prevalence: proportion of adults in HIV care among the total adult population in each district	4·7% (1·8–6·2)	2·3% (1·7–5·0)
IPT uptake*: proportion of adults in HIV care at the two largest clinics who had received IPT in the quarter immediately preceding the measurement period	1·8% (0·4–5·2)	2·2% (0·6–5·3)
Active tuberculosis prevalence†: proportion of adults with tuberculosis disease diagnosis among adults in active HIV care at the district level in the quarter immediately preceding the measurement period	0·4% (0·3–0·6)	0·3% (0·2–0·4)

Data are n, n (%), or median (IQR). Randomisation was conducted within pairs of clusters matched on region, number of adults in HIV care, presence of large urban centres, and presence of a community that had participated from 2013 to 2017 in the SEARCH universal HIV test-and-treat trial. IPT=isoniazid preventive therapy.

^*Missing data on four clinics.

^†Missing data on two districts.