Increasing follow-up questionnaire response rates in a randomized controlled trial of telehealth for depression: three embedded controlled studies

Table 3 Effect of photo intervention on questionnaire completion, number of questionnaire reminders, and questionnaire completion rate

	Control	Photo group	Unadjusted odds ratio, b, hazard ratio	95 % confidence interval; P	Adjusted odds ratio, b, hazard ratio^a	95 % confidence interval; P
Primary analysis^b
Completed questionnaire	118/125 (94)	118/126 (94)	0.88	0.31 to 2.49; 0.80	0.84	0.26 to 2.66; 0.76
Sensitivity analysis
Only those actually sent the questionnaire: Number of reminders = 0	116/123 (94)	114/122 (93)	0.86	0.30 to 2.45; 0.78	0.84	0.26 to 2.65; 0.76
Secondary analysis^c
Number of reminders = 0	68/125 (54)	64/126 (51)
Number of reminders = 1	25/125 (20)	27/126 (21)
Number of reminders = 2	15/125 (12)	15/126 (12)
Number of reminders = 3	9/125 (7)	6/126 (5)
Number of reminders = 4	4/125 (3)	5/126 (4)
Number of reminders = 5	4/125 (3)	9/126 (7)	0.17	−0.19 to 0.52; 0.36	0.22	−0.13 to 0.56; 0.22
Sensitivity analysis^c
Only those actually sent the questionnaire: number of reminders = 0	66/123 (54)	60/122 (49)
Number of reminders = 1	25/123 (20)	27/122 (22)
Number of reminders = 2	15/123 (12)	15/122 (12)
Number of reminders = 3	9/123 (7)	6/122 (5)
Number of reminders = 4	4/123 (3)	5/122 (4)
Number of reminders = 5	4/123 (3)	9/122 (7)	0.19	−0.17 to 0.55; 0.31	0.23	−0.12 to 0.57; 0.20
Secondary analysis^d
Time to complete questionnaire, median days	7.0	9.0	0.93	0.72 to 1.20; 0.57	0.85	0.66 to 1.11; 0.25
Sensitivity analysis^d
Only those actually sent the questionnaire: time to complete questionnaire, median days	7.5	10.0	0.91	0.70 to 1.18; 0.49	0.86	0.66 to 1.12; 0.27

Data are reported as number/total (percent), unless otherwise indicated. All primary analyses are intention-to-treat analyses. A sensitivity analysis was carried out on only those participants who were actually sent the questionnaire, as opposed to those who opted to complete the Patient Health Questionnaire (PHQ-9) over the telephone from the outset
^aAnalyses adjusted by participant sex, baseline age and Patient Health Questionnaire (PHQ-9) score, main trial allocation arm, questionnaire method sent (paper or online)
^bLogistic regression analysis of pre-call group allocation on whether the 8-month questionnaire was completed or not (odds ratio)
^cLinear regression of pre-call group allocation on total number of reminders (linear regression coefficient, b).
^dCox regression of pre-call group allocation on time to complete the questionnaire in days since the questionnaire was sent to the participant (hazard ratio). Zero values are not permitted in the time-to-event analysis, and so 0.5 days were allocated for participants who completed the questionnaire the day it was sent to them, while 91 days were allocated to the 15 participants who did not complete the questionnaire. For median: control n = 118 (sensitivity n = 116); photo n = 118 (sensitivity n = 114)

ISSN: 1745-6215