Table 3.
Analysis of Family Intervention for psychosis (FIp) compared to standard care (random-effects model).
| Outcome | Time of data collection | Trials k | Participants, n FIp/control | Estimate | Summary of estimate [95% CI] | Z, p | Favors FIp/control | Heterogeneity | ||
|---|---|---|---|---|---|---|---|---|---|---|
| Q test | I2 (%) | |||||||||
| Service user | (1) Symptoms (BPRS; PANSS, SOPS) | End of treatment | 4 | 129/130 | SMD | −0.26 [− 0.61, 0.09] | 1.5, p < 0.15 | – | Q = 5.4, p = 0.14 | 45 |
| Up to 2 years follow-up | 2 | 81/82 | SMD | −0.85 [−1.05, −0.20] | 2.6, p = 0.01* | Favors FIp | Q = 3.6, p = 0.06 | 72 | ||
| (2) Functioning (C/GAS) | End of treatment | 4 | 87/88 | SMD | 0.74 [0.13, 1.36] | 2.4, p = 0.02* | Favors FIp | Q = 10.0, p = 0.02 | 70 | |
| Up to 2 years follow-up | 1 | 25/24 | SMD | 0.22 [−0.34, 0.79] | 0.8, p = 0.43 | – | n/a | n/a | ||
| (3) Relapse (num. of people hospitalized/relapse in symptoms/transition to psychosis) | End of treatment | 7 | 303/291 | RR1 | 0.58 [0.34, 1.00] | 0.3, p = 0.05* | Favors FIp | Q = 12.13, p = 0.06 | 51 | |
| Up to 5 years follow-up | 3 | 242/232 | RR | 0.98 [0.32, 2.99] | 0.0, p = 0.98 | – | Q = 9.4, p = 0.009 | 79 | ||
| (4)Length of hospitalization throughout treatment /follow-up | End of treatment | 3 | 81/80 | SMD | −0.58 [−1.43, 0.27] | 1.3, p = 0.18 | – | Q = 6.1, p = 0.01 | 84 | |
| Up to 2 years follow-up | 2 | 33/43 | SMD | −0.12 [−0.58, 0.35] | 0.5, p = 0. 62 | – | Q = 0.0, p = 0.99 | 0 | ||
| Carer | (5) n of carers changed from high to low EE (CFI, FMSS) | End of treatment | 2 | 27/22 | OR1 | 16.76 [1.96, 143.44] | 2.6, p = 0.01* | Favors FIp | Q = 0.42, p = 0.52 | 0 |
| (6) EE: criticism (FQ, PRS) | End of treatment | 3 | 94/97 | SMD | −0.84 [−1.15, −0.53] | 5.3, p < 0.001* | Favors FIp | Q = 31.1, p < 0.001 | 94 | |
| Up to 2.5 years follow-up | 1 | 10/11 | SMD | −0.96 [−1.87, −0.05] | 2.1, p = 0.04* | Favors FIp | n/a | n/a | ||
| (7) EE: emotional over involvement (FQ) | End of treatment | 2 | 79/83 | SMD | −0.08 [−2.14,1.97] | 0.1, p = 0.94 | – | Q = 33.5, p < 0.001 | 97 | |
| Up to 2.5 years follow-up | 2 | 66/69 | SMD | −0.45 [−1.94, 1.04] | 0.6, p = 0.56 | - | Q = 8.6, p = 0.004 | 88 | ||
| (8) Communication conflict (FCS, FES, and clinician coding) | End of treatment | 3 | 80/70 | SMD | −0.44 [−0.77, −0.12] | 2.7, p = 0.008* | Favors FIp | Q = 1.4, p = 0.51 | 0 | |
| (9) Caregiver burden (ECI) | End of treatment | 3 | 135/139 | SMD | −0.72 [−0.97, −0.47] | 5.7, p < 0.001* | Favors FIp | Q = 17.4, p < 0.001 | 88 | |
| Up to 2.5 years follow-up | 3 | 122/125 | SMD | −0.31[−1.53, 0.91] | 0.5, p = 0.62 | – | Q = 36.6, p < 0.001 | 95 | ||
BPRS, Brief Psychosis Rating Scale; CCT, Controlled Clinical Trial; (C) GAS, (Children's) General Assessment Scales; CFI, Camberwell Family Interview; ECI, Experience of Caregiving Questionnaire; FES, Family Environment Scale; FCS, Family Conflict Scale; FMSS, Five Minute Speech Sample; FQ, Family Questionnaire; GHQ-28, General Health Questionnaires; K-10, Kessler Psychological Distress Scale; PANNS, Positive and Negative symptoms Scales; PRS, Patient Rejection Scale, SOPS = Scale of Psychosis-Risk Symptoms.