Measuring adherence to antiretroviral treatment in resource-poor settings: The feasibility of collecting routine data for key indicators

Table 4 Correlation between facility indicators

Indicators:	2	3	4	5	6	7	8	9	10	11	12	13
1	0.88753	-0.4332	0.19847	0.60229	0.24169	0.08673	0.78536	0.46213	-0.27041	0.02056	0.07059	-0.00039
	<.0001	0.0001	0.0971	<.0001	0.0438	0.5946	<.0001	<.0001	0.0207	0.8629	0.5825	0.9976
	39	74	71	55	70	40	72	71	73	73	63	63
2	1	-0.84969	0.33361	0.59568	0.50325	0.48017	0.77531	0.96964	-0.73886	0.48363	0.29674	0.24588
		<.0001	0.0379	<.0001	0.0013	0.0097	<.0001	<.0001	<.0001	0.0018	0.0745	0.1424
	39	39	39	39	38	28	39	39	39	39	37	37
3		1	-0.4246	-0.75195	-0.49922	-0.05321	-0.4183	-0.7567	0.89989	-0.21967	-0.1382	-0.059
			0.0002	<.0001	<.0001	0.7443	0.0003	<.0001	<.0001	0.06	0.2801	0.646
		75	71	55	70	40	72	71	73	74	63	63
4			1	0.74158	0.61924	-0.08272	0.41409	0.28721	-0.41749	0.19	0.14293	0.09498
				<.0001	<.0001	0.6071	0.0003	0.0159	0.0003	0.1125	0.2718	0.4665
			73	56	72	41	71	70	72	71	61	61
5				1	0.86842	0.16057	0.6507	0.61115	-0.71072	0.27723	0.37337	0.28431
					<.0001	0.38	<.0001	<.0001	<.0001	0.0404	0.009	0.0502
				56	55	32	56	56	56	55	48	48
6					1	-0.02434	0.32879	0.48872	-0.48896	0.26292	0.4468	0.30016
						0.88	0.0055	<.0001	<.0001	0.0279	0.0003	0.0188
					72	41	70	69	71	70	61	61
7						1	0.16434	0.36403	-0.0598	0.08467	0.02428	0.01863
							0.3174	0.0247	0.714	0.6083	0.885	0.9116
						42	39	38	40	39	38	38
8							1	0.42663	-0.56325	0.37349	-0.08788	-0.06737
								0.0002	<.0001	0.0012	0.497	0.6029
							73	72	73	72	62	62
9								1	-0.74195	0.4772	0.18678	0.16058
									<.0001	<.0001	0.146	0.2125
								72	72	71	62	62
10									1	-0.40181	-0.08735	-0.06228
										0.0004	0.496	0.6277
									74	73	63	63
11										1	0.26338	0.25041
											0.0386	0.0496
										74	62	62
12											1	0.92186
												<.0001
											66	66

Key: In each cell
Pearson Correlation Coefficients: Bold correlations = p < 0.05
Prob > |r| under H0: Rho = 0
Number of Observations
Indicators
1. % days covered by dispensed medicine if still in treatment
2. % patients with 95% days covered by dispensed medicine if still in treatment
3. % patients with gap of 30 days or more in dispensed medicine if still in treatment
4. % patients who attended on or before their appointment
5. % patients who attended within 3 days of their appointment
6. % patients who attended within 30 days of their appointment
7. From records: % Self reported good adherence
8. % days covered by dispensed medicine
9. % patients with 95% days covered by dispensed medicine
10. % patients with gap of 30 days or more in dispensed medicine
11. % patients remaining in treatment
12. From exit interview: % full adherence
13. From exit interview: average % adherence

ISSN: 1472-6963