Table 2.
Number of eligible participants recruited to trial, and those not randomised due to treatment preference.
| Conventional RCTs (n = 42) | |||
|---|---|---|---|
| Author | Number of eligible participants consenting to randomisation arms | Number of eligible patients not randomised because of treatment preference n (%) | Is preference expressed by patients (in addition to parents) |
| Allen 2013 | 1 (10%) (Feasibility) | 7 (70%) | Yes (only patient preference reported) |
| Allmark 2010 | 325 (81%) | Unclear, preference reported qualitatively [45] ‘30 declined’ ‘45 other reasons’ [90] |
n/a neonates |
| Banks 2012 | 76 (50%) (Pilot) | 6 (4%) | No |
| Barratt 2013 | 258 (33%) | Not reported. 9 (26%) of non-responders reported concern with being in either the intervention or control group, but only 37/305 non-responders replied to question. |
No |
| Bauchner 1996 | 648 (total eligible not reported) | Not reported. Parents were asked their preference at enrollment and 551 (85%) of those randomised preferred single-dose therapy over standard therapy. |
n/a children under 6yrs |
| Blickman 2013 | 142 (88%) | 4 (2%) | Unclear (patients aged 4yrs + were asked to complete a standardised study instrument) |
| Byrne-Davis 2010 | 521 (71%) [103] | Not reported, preference reported qualitatively [47] 215 (29%) not randomly assigned; 97 refused, 7 had Down's syndrome, 4 because of toxic effects, 28 other reason, 79 unknown [103] |
No |
| Caldwell 2003 | Not reported (multiple trials) | Not reported, preference reported qualitatively. | Participant age not stated. |
| Carvalho 2013 | Unclear 48 'recruited' [82] 44 (100%) 'randomised' [99] | Not reported, preference reported qualitatively [82] 3 (7%) parents refused allocated interventions post-randomisation in x 2 trial arms [99] |
No |
| Chappuy 2014 | Not reported | Not reported. Some Parents felt that standard treatment was the best arm for their child because it was less risky |
Participant age not stated. |
| Duncan 2004 | 50 different participants randomised. Total eligible not reported. | 8 (between 12 and 16%) | No |
| Eiser 2005 |
1621 (90%) [98] | 181 (10%) declined randomisation (opted for PRED; 165 DEXA; 16) [98] Preference reported qualitatively, 16 (32%) ‘agreed reluctantly to randomisation’ [49]. |
No |
| Forsander 1995 | 38 (93%) | Not reported Immediately after randomisation 3 families in the control arm reported that they would have preferred the family therapeutic care arm. |
No |
| Glogowska 2001 | 159 (69%) [94] | Not reported, preference reported qualitatively [50] Declined trial in total 70 (31%) [94] |
n/a children under 4yrs |
| Harth 1990 | 72 (55%) | 40 (30%) families declined because of ‘concern about side effects of the new drug’ (ketotifen) 60 declined in total. | n/a children under 3yrs |
| Hissink Muller 2011 | Not reported | Not reported. 41% participating parents reported a preference for therapy with methotrexate and etanercept and 6% had hoped against assignment to this group. Primary aversion was highest (25%) in the prednisone group [40]. Declined trial n = 38 (29%) [96]. |
No |
| Hissink Muller 2012 | Not reported | Not reported. 65% participating parents reported a preference for therapy with etanercept. 5 parents and 2 patients participated in the study to access treatment with etanercept, as initial treatment was not possible nor reimbursed in daily practice. |
Yes |
| Johnson 2007 | Not reported | Not reported. Participating families stated: Close monitoring arm - 27% parents and 70% participants were glad to be in that arm. 74% parents and 35% participants sometimes wished they had been assigned the intervention arm. Intervention arm - 53% parents and 21% participants were glad to be in that arm. 25% parents and 47% participants sometimes wished they had been assigned the closely monitored arm. |
Yes |
| Johnson 2009 | Not reported | Not reported. Participating families were blinded to treatment but were asked which treatment arm they would have preferred. 60% parents and 53% participants chose the capsule condition. 8% parents and 21% participants chose the no intervention condition. Very few participants and parents (3%) chose the insulin injection condition. |
Yes |
| Jollye 2009 | Not reported, multiple trials. | Not reported, preference reported qualitatively. | n/a neonates |
| Levi 2000 | Not reported, multiple trials. | Unclear. 3 (13.6%) stated they declined participation because they felt more comfortable with a “tried and true” method. |
No |
| Miner 2007 | 41 (82%) | Unclear. Declined randomised 9 (18%) reasons not reported. After allocation 4 (10%) parents requested that their child receive nebulized fentanyl rather than the assigned IV fentanyl. |
No |
| Payne 2004 | Unclear Calculated as; 322 (69%) of eligible patients. Paper reports recruitment rate of 75% |
59 (50%) ‘Around half of the eligible participants who refused to participate did so because there was a 50% chance of the child being randomised to the inhalational induction arm’. | No |
| (PENTA) Paediatric European Network for Treatment of AIDS 1999 | 197 | 4 (3%) parents stated explicitly that they were concerned with the use of placebo. | No |
| Rovers 2000 | 187 | Not reported. 19 (10%) parents withdrew consent straight after randomisation (15 in ventilation tubes arm and 4 in watchful waiting arm). 10 (5%) children in the watchful waiting arm were treated with ventilation tubes. |
n/a children under 2yrs |
| Sammons 2007 | Unclear 245 'randomised’ [55] 252 (85%) ‘randomised’ [89] |
25 (9%) declining families stated they wanted a specific treatment (IV; 20 or oral; 5) [55] 43 (15%) declined to take part; n = 6 (2%) excluded post randomisation reasons: 4 withdrawn by parents/2 by clinician (no further detail provided) [89]. |
No |
| Sandler 2014 | 78 (87%) | 7 (8%) Three did not want to wear headgear for anchorage, three did not want the Nance button palatal arches, but only one patient did not want to take part because he or she was unhappy at “the thought of temporary anchorage devices”. | No |
| Sartain 2002 | 399 (86%) | 10 (2%) 7 families withdrew from ‘hospital care’ arm because they wanted the ‘hospital at home’ arm |
Yes |
| Schuttelaar 2010 | 160 | 4 (2%) Preferred only dermatologist (n = 2), preferred only nurse practitioner (n = 2). | No |
| Sederberg-olsen 1998 | 429 | 120 (10%) parents insisted that the child had grommet insertion performed at the time of randomisation. | No |
| Shilling 2011 | MASCOT: 63 [97] MENDS: 146 (84%) [88] POP: [still recruiting] TIPIT: 153 (57%) [108] |
Unclear, preference reported qualitatively. MASCOT Assessed for eligibility (n = 898), Not registered (n = 732), Excluded (n = 103) [97]. MMENDS 27 (16%) assessed for eligibility but not randomised: ‘declined 11’ ‘other 16’ [88]. TIPIT 57 (21%) assessed for eligibility but not randomised: ‘refused’ [105]. |
Yes |
| Snowdon 1997 | 185 (79%) [95] | Unclear. ‘majority of parents had a keen preference for ECMO treatment arm’. Preference reported qualitatively [58]. 48 (21%) were registered but not randomised; 14 died, 19 improved and 15 parents refused trial participation [95]. |
n/a neonates |
| Spandorfer 2005 | 73 | 24 (7%) A further 3 parents refused participation after randomisation to oral rehydration therapy before starting treatment. |
n/a children under 3yrs |
| Sureshkumar 2012 | 412 (37%) [85] | 214 (19%) Prefer antibiotics 71/Prefer no antibiotics 143 [85]. Primary paper reports patients excluded because ‘participation refused by parent’ 1935 [92] |
No |
| Tercyak 1998 | 56 | 2 (5%) | Yes (only patient preference reported) |
| Willey 2005 | 31 | Not reported. 19/31 patients completed a preference questionnaire/10 (43%) preference for oral, 2 (9%) for suppositories, 7 (30%) no preference/preference for oral more pronounced among girls 5 (83%). |
Yes |
| Williams 2013 | 94 | Not reported. A significantly larger percentage of parents and patients in the cast group reported that they would not choose the same method of immobilization again at all time points (baseline, days; 1, 3, 7, 21 after injury). |
No |
| Woodgate 2010 | Not reported (multiple trials) | Not reported, preference reported qualitatively. | n/a neonates |
| Woolfall 2013 | MASCOT: 63 [97] MENDS: 146 (84%) [88] POP: [still recruiting] TIPIT: 153 (57%) [108] |
Unclear, preference reported qualitatively. MASCOT Assessed for eligibility (n = 898), Not registered (n = 732), Excluded (n = 103) [97]. MMENDS 27 (16%) assessed for eligibility but not randomised: ‘declined 11’ ‘other 16’ [88]. TIPIT 57 (21%) assessed for eligibility but not randomised; ‘refused’ [105]. |
No |
| Wright 2005 | 108 (46%) | 41 (33%) | No |
| Wynn 2010 | 234 (29%) | 2% unwilling to take placebo. | n/a children under 2yrs |
| Young 2006 | Not reported. | 125 ‘Reluctance toward medication treatment accounted for 44.7% of study refusals and was disproportionately common among ethnic minority families’. |
No |
| RCTs with non-randomised preference arms (n = 10) | |||
| Cunningham 2011 | Not reported. (multiple trials) | Not reported. A small number of patients who were eligible declined the trial as they had a treatment preference. These were patients allocated to both intervention groups, so one treatment option was not preferred to the other. Preference arms added. |
Unclear |
| Gowers 2010 | 170 (68%) | 28 (11%) Not randomised, patient preference. |
Yes |
| Lock 2010 | 268 (26%) | 286 (28%) declined any follow up, authors assumed that all had a patient preference. 461 (45%) opted for preference arms in cohort. |
Only in qualitative sample. Authors did not attempt to differentiate between parent/child preferences in RCT/preference samples. |
| Mattila 2007 | 137 (45%) | 169 (55%) opted for preference arms. | n/a children under 2yrs |
| Paradise 1984 | 91 (49%) | 96 (51%) opted for preference arms. | No |
| Paradise 1990 | 99 (46%) | 114 (54%) opted for preference arms. | No |
| Reddihough 1998 | 34 (49%) | 32 (46%) declined randomisation. | n/a children under 3yrs |
| Rovers 2001 |
187 (48%) | 133 (34%) opted for non-randomised cohort arms. 66 (17%) refused randomisation/follow up via cohort. |
n/a children under 1yrs |
| Van Wijk 2014 | 84 (21%) | 186 (46%) opted for preference arms. | n/a children under 1yrs |
| Weinstein 2013 | 155 (14%) | 228 (21%) opted for preference arms. 297 (27%) declined all follow-up due to preference. 216 (20%) no to randomisation. |
No |