Prescriber and patient-oriented behavioural interventions to improve use of malaria rapid diagnostic tests in Tanzania: facility-based cluster randomised trial

Table 4 Effect of interventions on recommended antimalarial prescribing among patients with a non-severe, non-malarial illness

Evaluation category	Arm	Number of clusters	Number of patients	Prevalence number (%)	Crude RD ^a (95% CI)	Adjusted RD ^b (95% CI)	P -value
Overall^c(all ages)	Control	12	8942	749 (8%)	0	0
	HW	12	10118	250 (2%)	0.06 (0.04, 0.09)	0.04 (0.01, 0.06)	0.008
	HWP	12	10163	184 (2%)	0.07 (0.04, 0.09)	0.04 (0.01, 0.06)	0.005
<5 years	Control	12	3139	392 (12%)	0	0
	HW	12	3682	153 (4%)	0.09 (0.04, 0.14)	0.06 (0.006, 0.12)	0.03
	HWP	12	3406	95 (3%)	0.09 (0.03, 0.14)	0.05 (0.005, 0.09)	0.03
≥5 years	Control	12	5803	357 (6%)	0	0
	HW	12	6436	97 (2%)	0.05 (0.03, 0.07)	0.02 (0.006, 0.04)	0.008
	HWP	12	6757	89 (1%)	0.05 (0.03, 0.07)	0.03 (0.01, 0.05)	0.002
By Evaluation period
Standard training - period 1 (all arms)	Control	12	656	48 (7%)	0	0
	HW	9	449	3 (1%)	-	-	-
	HWP	11	494	38 (8%)	0.001 (−0.09, 0.09)	−0.01 (−0.06, 0.03)	0.54
Interactive training - period 2 (HW and HWP arms)	Control	12	3320	236 (7%)	0	0
	HW	12	3791	135 (4%)	0.05 (0.02, 0.08)	0.02 (−0.001, 0.05)	0.06
	HWP	12	3802	81 (2%)	0.05 (0.02, 0.09)	0.03 (0.003, 0.05)	0.03
Feedback SMS – period 3 (HW and HWP arms)	Control	12	2392	215 (9%)	0	0
	HW	12	2829	58 (2%)	0.06 (0.04, 0.09)	0.02 (−0.001, 0.05)	0.06
	HWP	12	2891	24 (1%)	0.07 (0.04, 0.10)	0.02 (−0.004, 0.04)	0.09
Feedback + proverb SMS - period 4 (HW and HWP arms)	Control	12	1297	106 (8%)	0	0
	HW	12	1540	18 (1%)	0.07 (0.05, 0.09)	0.03 (0.007, 0.06)	0.01
	HWP	12	1607	7 (0.4%)	0.07 (0.05, 0.09)	0.03 (0.01, 0.06)	0.009

^aAdjusted for stratification, effect estimate is risk difference = control– intervention; control is standard RDT training. ^bAdjusted for facility (stratum, stock-out of ACT, provision of materials, % fever consultations treated with antimalarial prior to the study, % non-febrile consultations treated with antimalarial prior to the study), prescriber (age, education, time at facility) and patient (age) characteristics.
- Insufficient number of clusters or sample size per cluster to conduct a robust analysis. ^cDefined as the period of evaluation from the end of the standard RDT training until the end of the trial. The between-cluster coefficient of variation was estimated as k = 0.02 for comparison of both intervention arms with the control. ACT, artemisinin-based combination therapies; CI, confidence interval; HW, health worker; HWP, health worker plus patient-oriented; RD, risk difference; RDT, rapid diagnostic test; SMS, mobile-phone text message.