Table 3.
Missing data interventions and reported evaluations into their effectiveness (registered Clinical Trial Unit (CTU) survey Questions 6 and 7)
| Missing data intervention | Respondents who have used the intervention (% of all survey respondents, n = 33) | Used routinely (% of all survey respondents, n = 33) | Used occasionally (% of all survey respondents, n = 33) | No response for frequency of use (% of respondents who used intervention) | Before/after evaluation (informal) | Nested RCT evaluation (formal) | Total evaluations |
|---|---|---|---|---|---|---|---|
| Newslettersa | 30 (91%) | 23 (70%) | 5 (15%) | 2 (7%) | 2 | 2 | 4 |
| A timeline of participant visits for sites | 24 (73%) | 19 (58%) | 4 (12%) | 1 (4%) | 0 | 0 | 0 |
| Inclusion of prepaid envelope (questionnaires)b | 28 (85%)c | 19 (58%) | 6 (18%) | 1 (4%) | 0 | 1 | 1 |
| Telephone reminders | 28 (85%) | 18 (55%) | 9 (27%) | 1 (4%) | 2 | 1 | 3 |
| Data collection scheduled with routine care | 25 (76%) | 18 (55%) | 6 (18%) | 1 (4%) | 1 | 1 | 2 |
| Site initiation training on missing data | 19 (58%) | 18 (55%) | 0 (0%) | 1 (5%) | 0 | 0 | 0 |
| Investigator meetings face to face | 22 (67%) | 17 (52%) | 5 (15%) | 0 (0%) | 0 | 1 | 1 |
| Routines site visits by CTU staff | 23 (70%) | 15 (45%) | 8 (24%) | 0 (0%) | 1 | 0 | 1 |
| Targeted recruitment of sites/GPs | 21 (64%) | 15 (45%) | 6 (18%) | 0 (0%) | 0 | 2 | 2 |
| Flexibility in appointment times | 21 (64%) | 15 (45%) | 4 (12%) | 2 (10%) | 1 | 1 | 2 |
| Communication of trial results | 20 (61%) | 15 (45%) | 3 (9%) | 2 (10%) | 1 | 0 | 1 |
| Investigator teleconferences | 22 (67%)d | 15 (45%) | 8 (24%) | 0 (0%) | 0 | 1 | 1 |
| Questionnaires completed in clinicb | 22 (67%) | 15 (45%) | 5 (15%) | 2 (9%) | 1 | 1 | 2 |
| Minimising frequency of questionnairesb | 21 (64%) | 15 (45%) | 3 (9%) | 3 (14%) | 0 | 0 | 0 |
| Short questionnaireb | 24 (73%) | 14 (42%) | 9 (27%) | 1 (4%) | 0 | 1 | 1 |
| Collecting multiple contact details | 22 (67%) | 13 (39%) | 8 (24%) | 1 (5%) | 2 | 1 | 3 |
| Email reminders | 21 (64%) | 13 (39%) | 6 (18%) | 2 (10%) | 4 | 0 | 4 |
| Postal reminders | 23 (70%) | 12 (36%) | 10 (30%) | 1 (4%) | 2 | 2 | 4 |
| Total design method for Questionnairesb | 15 (46%) | 12 (36%) | 3 (9%) | 0 (0%) | 1 | 1 | 2 |
| Re-imbursement of participant expenses | 24 (73%)e | 11 (33%) | 12 (36%) | 0 (0%) | 1 | 1 | 2 |
| Triggered site training on missing data | 23 (70%) | 11 (33%) | 12 (36%) | 0 (0%) | 0 | 0 | 0 |
| Use of routinely collected data | 29 (88%) | 10 (30%) | 17 (52%) | 2 (7%) | 2 | 1 | 3 |
| Contact GPs for missing data/trace patients | 27 (82%) | 10 (30%) | 14 (42%) | 3 (11%) | 0 | 1 | 1 |
| Patient diaries | 26 (79%) | 10 (30%) | 13 (39%) | 3 (12%) | 1 | 0 | 1 |
| Enhanced cover letter (questionnaires)b | 14 (43%) | 9 (27%) | 5 (15%) | 0 (0%) | 1 | 2 | 3 |
| Staggered per patient payments to sites | 20 (61%) | 8 (24%) | 12 (36% | 0 (0%) | 0 | 0 | 0 |
| Patient data entry | 18 (55%) | 8 (24%) | 8 (24%) | 2 (11%) | 0 | 1 | 1 |
| Trial identity cards | 14 (43%) | 8 (24%) | 5 (15%) | 1 (7%) | 0 | 0 | 0 |
| Telephone questionnaires | 20 (61%) | 7 (21%) | 11 (33%) | 2 (10%) | 2 | 1 | 3 |
| Trial website | 18 (55%) | 7 (21%) | 10 (30%) | 1 (6%) | 0 | 0 | 0 |
| Taking contact details for a friend/family | 13 (40%) | 7 (21%) | 6 (18%) | 0 (0%) | 0 | 1 | 1 |
| Long but clear questionnaireb | 10 (31%) | 7 (21%) | 2 (6%) | 1 (10%) | 0 | 0 | 0 |
| Only collecting the primary outcome for patients with missing data | 17 (52%) | 6 (18%) | 10 (30%) | 1 (6%) | 2 | 1 | 3 |
| Gift | 18 (55%) | 6 (18%) | 10 (30%) | 2 (11%) | 1 | 1 | 2 |
| ONS flagging | 16 (49%)e | 6 (18%) | 7 (21%) | 2 (13%) | 0 | 0 | 0 |
| Flexibility in appointment locations | 14 (43%) | 6 (18%) | 6 (18%) | 2(14%) | 1 | 0 | 1 |
| Money/gift voucher given on completion of a milestone | 17 (52%) | 5 (15%) | 12 (36%) | 0 (0%) | 1 | 0 | 1 |
| Contacting patients between visits | 13 (40%) | 4 (12%) | 9 (27%) | 0 (0%) | 0 | 0 | 0 |
| Christmas and birthday cards | 13 (40%) | 4 (12%) | 9 (27%) | 0 (0%) | 0 | 1 | 1 |
| Freephone number for updating contact | 7 (22%) | 4 (12%) | 3 (9%) | 0 (0%) | 0 | 0 | 0 |
| Follow-up through patient notes only | 19 (58%)e | 3 (9%) | 15 (45%) | 0 (0%) | 0 | 0 | 0 |
| Transport to and from appointments | 8 (25%) | 3 (9%) | 4 (12%) | 1 (13%) | 0 | 0 | 0 |
| SMS text reminders | 16 (49%)e | 2 (6%) | 13 (39%) | 0 (0%) | 3 | 1 | 4 |
| Trial certificate | 8 (25%) | 2 (6%) | 6 (18%) | 0 (0%) | 0 | 0 | 0 |
| Medical questions first in questionnaireb | 2 (7%) | 2 (6%) | 0 (0%) | 0 (0%) | 0 | 0 | 0 |
| Generic questions first in questionnaireb | 3 (10%) | 2 (6%) | 0 (0%) | 1 (33%) | 0 | 0 | 0 |
| Questionnaires sent less than 3 weeks after a visitb | 2 (7%) | 2 (6%) | 0 (0%) | 0 (0%) | 0 | 0 | 0 |
| Other (questionnaires) b | 2 (7%) | 2 (6%) | 0 (0%) | 0 (0%) | 0 | 0 | 0 |
| Money/gift voucher given regardless | 8 (25%) | 1 (3%) | 7 (21%) | 0 (0%) | 1 | 2 | 3 |
| Case management | 3 (10%) | 1 (3%) | 2 (6%) | 0 (0%) | 0 | 0 | 0 |
| Other | 2 (7%) | 1 (3%) | 1 (3%) | 0 (0%) | 0 | 0 | 0 |
| Personal touch (questionnaires)b | 9 (28%) | 1 (3%) | 8 (24%) | 0 (0%) | 0 | 0 | 0 |
| Questions about health issue first in questionnaireb | 1 (4%) | 1 (3%) | 0 (0%) | 0 (0%) | 1 | 0 | 1 |
| Questionnaires sent before clinic visitb | 5 (16%) | 1 (3%) | 4 (12%) | 0 (0%) | 0 | 1 | 1 |
| Prize draw limited to trial participants | 6 (19%) | 0 (0%) | 6 (18%) | 0 (0%) | 0 | 0 | 0 |
| Social media | 5 (16%)e | 0 (0%) | 3 (9%) | 1 (20%) | 0 | 0 | 0 |
| Priority or recorded post (questionnaires)b | 4 (13%) | 0 (0%) | 3 (9%) | 1 (25%) | 0 | 0 | 0 |
| Crèche service | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 | 0 | 0 |
| Behavioural motivation | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 | 0 | 0 |
| Charity donation | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 | 0 | 0 |
| National lottery ticket or similar public draw | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 | 0 | 0 |
| Total | 35 | 31 | 66 | ||||
| Number of strategies that have been evaluated (% of all strategies, n = 59) | 23 (39%) | 26 (44%) | 36 (61%) | ||||
aOne case where the newsletters were used with patients only, nine cases where newsletters were used with both patients and research sites, and 20 cases where they were only used with research sites. bStrategies used to enhance questionnaire response rates from Question 7 of the registered CTU survey cTwo respondent stated that they would not use this intervention again. dOne person reported both occasional and routine use. eOne respondent stated that they would not use this intervention again
CTU Clinical Trial Unit, GP general practitioner, ONS Office for National Statistics, RCT randomised controlled trial, SMS short message service