RCTs evaluating the effects of family based behaviour modification programmes
| Reference | Participants | Interventions, duration | Results | Comments |
|---|---|---|---|---|
| I=intervention; C=controls. | ||||
| Family based programmes with parents as agents of change | ||||
| Epstein58 (USA, 1985) | Obese girls (aged 5–8 years) Mean age: not given 100% female | I: Diet and exercise information plus information on parent management techniques and social learning principles (n=8) C: Diet and exercise information alone (n=11) Follow up: 12 months | Mean (SD) percentage overweight: Baseline: I=41.9 (13.6), C=39.2 (17.1) 12 months: I=15.6 (15.2), C=28 (16.7), (p<0.05) Mean (SD) BMI: Baseline: I=22.8 (2.6), C=22.7 (3) 12 months: I=19.1 (2.8), C=21.4 (3.3) (p<0.05) | Random allocation: Method not described Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Israel39 (USA, 1985) | Overweight children (aged 8–12 years) Mean age: 11 years, 4 months % female: not given | I1: Weight reduction only (WRO); multicomponent behavioural weight reduction programme (n=12) I2: Parent training (PT); as WRO, but preceded by short course for parents in general child management skills (n=12) C: Waiting list control (n=9) Follow up: One year (I1: n=11; I2: n=9) | Mean % overweight: Week 1: I1=53.15, I2=45.88, C=56.02 Week 9: I1=41.49, I2=38.71, C=55.09 Change in % overweight at 9 weeks lower in I1 than I2 group (p<0.025), and lower in I2 than C group (p<0.01) One year: I1=45.53, I2=40.40 Change in % overweight at 1 year increased in I1 group compared with I2 (p<0.001) | Random allocation: Stratified blocks based on child percent overweight and age Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Mellin40 (USA, 1987) | Obese adolescents (aged 12–18 years) Mean age: 15.6 years 79% female | I: 14 × 90 minute sessions using the materials of the SHAPEDOWN programme (encouraging adolescents to make small sustainable changes in diet, exercise, lifestyle and attitudes) plus two parent sessions (n=37) C: No intervention (n=29) Follow up: 15 months from start of intervention | Mean weight change (kg): 3 months: I=–3.11, C=+0.13 6 months: I=–1.40, C=–1.05 15 months: I=–3.88, C=+1.27 Intervention group displayed overall mean weight loss of 5.15 kg compared with control group Programme participation was also associated with a post-treatment and 1 year follow up reduction in relative weight | Random allocation: Method not described Blinding: Children: No Providers: No Outcome assessors: Unclear |
| Israel59 (USA, 1994) | Obese children (aged 8–13 years) Mean age: 10 years 11 months % female: not given | Parents and children met separately for 8 × 90 minutes sessions followed by 9 biweekly sessions for a total of 26 weeks. Treatment consisted of discussions and homework assignments I1: Standard treatment condition (n=18) I2: Enhanced child involvement (n=16) Follow up: 1 and 3 years (I1: n=11; I2 n=9) | Mean percentage overweight: Week 1: I1=45.94, I2=48.10 Week 26: I1=33.43, I2=32.55 1 year: I1=45.15, I2=42.32 3 years: I1=52.30, I2=43.29 Mean percentage over triceps norm: Week 1: I1=131.65, I2=118.43 Week 26: I1=101.3, I2=82.99 1 year: I1=129.83, I2=132.68 No significant between group differences | Random allocation: Method not described Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Epstein41 (USA, 1994) | Obese children (aged 8–12 years) and their parents Mean age: 10.2 years 74% female | I: Parents and children targeted and reinforced for mastery of diet, exercise, weight loss and parenting skills (n=17) C: participants taught behaviour change strategies and provided non-contingent reinforcement at a pace yoked to the intervention group (n=22) Intervention given over 26 weekly meetings and 6 monthly meetings Follow up: 2 years | Mean percentage overweight: Baseline: I= 60.6, C=58.8 6 months: I= 30.5, C=38.8 (p<0.05) 12 months: I=34.1, C=42.1 (p<0.05) 24 months: I=45.2, C=48.2 (p<0.3) | Random allocation: Method not described Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Golan42 (Israel, 1998) | Obese children (aged 6–11 years) Mean age: 9.2 years 62% female | I: Behavioural modification targeted at parents as agents of change, 14 sessions (n=30) C: Children as agents of change. 30 sessions (n=30) Hour long support and educational sessions were conducted by a clinical dietician Follow up: 6 and 12 months. | Percentage overweight: I: Baseline: 39.6 1 year follow up: 24.9 (p<0.001) C: Baseline: 39.1 1 year follow up: 31.0 (p<0.01) Reduction over 1 year was significantly greater in I group than C group (p<0.03) | Random allocation: Method not described Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Family based behaviour modification programmes | ||||
| Brownell43 (USA, 1983) | Obese adolescents (aged 12–16 years) and mothers Mean age: not given 79% female | Programme of behaviour modification, nutrition education, exercise instruction and social support I1: Mothers and children met concurrently in separate groups (n (baseline, 16 weeks, 1 year)=14, 13, 12) I2: Children and mothers attended all sessions in the same group (n (baseline, 16 weeks, 1year)=15, 13, 12) I3: Children met in groups, mother did not take part in formal treatment programme (n (baseline, 16 weeks, 1 year)=13, 13, 13) Follow up: One year | Change in % overweight: 16 weeks: I1=–17.1, I2=–7.0, I3=–6.8 1 year: I1=–20.5, I2=–5.5, I3=–6.0 Significant reduction in % overweight for I1 at 16 weeks (p<0.01) and at 1 year (p<0.05) compared with I2 and I3 Mean change in weight (kg): Significant reduction in mean weight (kg) for I1 at 16 weeks (p<0.04) and at 1 year (p<0.01) compared with I2 and I3 | Random allocation: Method not described Blinding: Children: No Providers: No Outcome assessors: Unclear |
| Kirschenbaum44 (USA, 1984) | Overweight children (aged 9–13 years) and their parents Mean age: I1=10.4, I2=11.2, C=10.5 77% female | I1: Parent plus child condition. Parents and children attended all sessions together. Emphasis was placed on the importance of parents and children working together (n=13) I2: Child only condition. Only children attended group sessions (n=9) C: Waiting list control condition (n=8) Follow up: 3 and 12 months | Weight reduction index: Parents and children in groups I1 and I2 lost significantly more weight than those in group C at 9 weeks (p<0.01), 3 months (p<0.01), and at 1 year, although I1 and I2 did not differ significantly from each other at any follow up. Children in group C significantly gained weight at 3 months (p<0.05). Similar results were found for percentage overweight | Random allocation: Stratified by gender, age and initial percentage overweight Blinding: Children: Unclear Parents: Unclear Outcome assessors: Unclear |
| Senediak45 (Australia, 1985) | Overweight children (aged 6–13 years) and their parents Mean age: 10.3 years % female: not given | I1: Rapid behavioural intervention (n=12) I2: Gradual behavioural intervention (n=12) C1: Non-specific control (n=11) C2: Waiting list control (n=10) Follow up: 26 weeks. I1 (n=8), I2 (n=10), C1 (n=7) | Mean percentage overweight: Week 1: I1=34.63, I2=34.93, C1=41.68, C2=37.64 Week 4: I1=29.37, I2=30.70, C1=40.32, C2=39.95 Week 15: I1=20.99, I2=17.84, C1=36.72, C2=no further contact Week 26: I1=19.94, I2=16.64, C1=30.80, C2=no further contact | Random allocation: Method not described Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Graves60 (USA, 1988) | Obese children (aged 6–12 years) and their parents Mean age: 9.3 years % female: not given | Three different treatment protocols for an 8 week weight loss programme I1: Problem solving group (n=not given) I2: Behavioural group (n=not given) I3: Instruction only group (n=not given) Follow up: 3 and 6 months | Children in I1 and I2 groups significantly reduced their body weights, percentages overweight, and BMIs significantly from pre- to post–treatment (p<0.05), whereas children in the I3 group did not. These differences were maintained at 3 and 6 month follow up. The I1 group demonstrated significantly greater reductions in percentage overweight and BMI from post treatment to 3 month follow up (p<0.05) than I2 and I3 groups | Random allocation: Method not described Blinding: Children: Unclear Providers: No Outcome assessors: Unclear |
| Wadden61 (USA, 1990) | Overweight girls (aged 12–16 years) and mothers Mean age: 13.8 years 100% female | All children attended 16 weekly 1 hour treatment sessions following the “weight reduction and pride” (WRAP) programme I1: Child alone (n=19) I2: Mother and child together (n=14) I3: Mother and child separately (n=14) Follow up: 6 months (n=31) | Mean BMI for all participants decreased from 35.2 at baseline to 33.9 at 16 weeks (p<0.001). There were no differential changes among treatment conditions Mean BMI for available participants at 6 month follow up was 35.4, which did not significantly vary from baseline | Random allocation: Stratified on the basis of BMI Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Flodmark46 (Sweden, 1993) | Obese children (aged 10–11 years) and families Mean age: not given 52% female | I1: Family therapy as adjunct to conventional treatment (dietary education by a dietitian, regular visits to a paediatrician, encouraged to exercise) family therapy involved whole family 6 sessions over 12 months (n=24) I2: Conventional treatment (as above) (n=19) C: No intervention (n=50) Follow up: 12 months | BMI, mean (SD): Baseline: I1=24.7 (0.36), I2=25.5 (0.53), C=25.1 (0.35) End of treatment (14–18 months): I1=25.0 (0.53), I2=26.1 (0.72), C=not given 12 month follow up: I1=25.8 (0.73), I2=27.1 (0.88), C=27.9 (0.61) Significantly smaller increase in I1 than in C (p=0.02) | Random allocation: Method not described Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Duffy47 (Australia, 1993) | Overweight children (aged 7–13 years) and at least one parent Mean age: 9.9 years 79% female | Both groups attended 8 weekly sessions of 90 min duration. Nutritional education was based on Epstein’s “traffic light system” I1: Behaviour therapy plus attention placebo control (n=13) I2: Behaviour therapy plus cognitive self management (n=14) Follow up: 3 and 6 months I1: n (3 months, 6months)=10, 8); I2: n (3 months, 6months)=11, 9) | Mean (SD) percentage overweight: I1: Pre-treatment: 51.53 (26.92); post- treatment: 42.43 (25.45) 3 months: 42.84 (24.90) 6 months: 37.09 (21.71) I2: Pre-treatment: 45.48 (17.52) Post-treatment: 37.70 (18.51) 3 months: 38.49 (18.86) 6 months: 37.02 (24.58) Reductions from baseline significant in both groups, but between group differences not significant | Random allocation: Stratified by age group (7–10 years and 10–13 years) Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Braet48 (Belgium, 1997) | Obese children (aged 7–16 years) and their parents Mean age: not given 63% female | Two randomised behaviour therapy groups including seven 90 min and seven family follow up sessions: I1: Individual therapy (n=48) I2: Group therapy (n=45) Follow up: 12 months | Percentage weight loss from baseline (%): I1: 3 months=5.72; 6 months=8.34; 12 months=9.84; all significant (p<0.001) I2: 3 months=3.31; 6 months=8.44; 12 months=13.08; all significant (p<0.001) | Random allocation: Method not described Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Epstein49 (USA, 2000) | Families with at least one child more than 20% overweight Mean age: 10.3 years 52% female | I1: Problem solving taught to parent and child plus standard family based treatment targeting and reinforcing eating and exercise behaviour change (n=17*) I2: Problem solving taught to child plus standard family based treatment (n=18*) I3: Standard family based treatment (n=17*) Follow up: 6, 12 and 24 months *10 drop outs unaccounted for | BMI Z score, mean (SD): Baseline: I1=2.8 (0.9), I2=2.6 (0.9), I3=2.7 (0.8) 6 months: I=1.5 (0.9), I2=1.2 (0.8), I3=1.2 (0.8) 12 months: I1=1.7 (1.0), I2=1.3 (0.9), I3=1.4 (0.9) 24 months: I1=2.3 (1.1), I2=1.7 (0.9), I3=1.6 (1.0) I3 group had larger decrease in mean BMI Z score over 0–24 months (p<0.02) | Random allocation: Families stratified by gender and degree of child and parent obesity Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |
| Goldfield50 (USA, 2001) | Families with obese children (aged 8–12 years) Mean age: I1=9.8 years, I2=10.3 years 71% female | I1: Mixed treatment, whereby participants received a mixture of individualised plus group treatment (n=12) I2: group treatment that did not involve individual therapy (n=12) Follow up: 6 and 12 months | Mean (SD) change in percentage overweight: Baseline–6 months: –9.97 (8.7) Baseline–12 months: –8.04 (10.3) Mean (SD) change in Z-BMI: Baseline–6 months: –0.59 (0.49) Baseline–12 months: –0.64 (0.63) Data for separate groups not given A significant reduction in percentage overweight and Z–BMI was found for both types of intervention over time (p<0.001), though there were no significant differences between interventions. I2 was found to be significantly more cost effective, due to the greater expense of I1 | Random allocation: Method not described Blinding: Children: Unclear Providers: No Outcome assessors: Unclear |
| Programmes with no parental involvement | ||||
| Warschburger51 (Germany, 2001) | Obese children and adolescents (aged 9–19 years) Mean age: I=13.8, C=13.1 % female: not given | Inpatient rehabilitation programme I1. Group which participated in “obesity training” (a three part programme which included a cognitive-behavioural training program, a calorie reduced diet and an exercise program; n=121) I2. Group which undertook the same diet and exercise programmes but received muscle relaxation training instead of the psychological intervention component (n=76) Follow up: 6 and 12 months | Change in mean percentage overweight: 6 weeks: I1=–15.47, I2=–14.03 Both groups significantly reduced their percentage overweight over the course of one year compared with baseline. Differences between the groups were not significant (p values not reported) | Random allocation: Method not described Blinding: Children: Unclear Providers: Unclear Outcome assessors: Unclear |