Psychological interventions for postnatal depression: cluster randomised trial and economic evaluation. The PoNDER trial.

OBJECTIVES

To investigate outcomes for postnatal women attributed to special training for health visitors (HVs) in systematically identifying postnatal depression and delivering psychologically informed interventions, and to establish the cost-effectiveness of the intervention.

DESIGN

A pragmatic randomised cluster trial with clusters allocated to experimental HV training arms or control, with an 18-month follow-up.

SETTING

GP practices in the former Trent Regional Health Authority.

PARTICIPANTS

Women registered with participating GP practices who became 36 weeks pregnant during the recruitment phase of the trial, had a live baby and were on a collaborating HV's caseload for 4 months postnatally.

INTERVENTION

HV training in the assessment of postnatal women, combined with either cognitive behavioural approach (CBA) or person-centred approach (PCA) sessions for eligible women, plus the option of a selective serotonin reuptake inhibitor if indicated.

MAIN OUTCOME MEASURES

The primary outcome was the proportion of at-risk women with a 6-month Edinburgh Postnatal Depression Scale (EPDS) score > or = 12. The primary comparison was between at-risk women in the combined clusters randomised to HV training and women in practices randomised to provide HV usual care. The secondary comparison was to determine any differences between the proportions of women with a 6-month EPDS score > or = 12 in the CBA and PCA groups.

RESULTS

HVs in 101 clusters in 29 primary care trusts collaborated in the study. From 7649 eligible women 4084 (53.4%) consented to take part: 17.3% (595/3449) of women who returned a 6-week questionnaire had a 6-week EPDS score > or = 12 and were at-risk women; 70.3% (418/595) of at-risk women had a 6-month EPDS score available. In total, 45.6% (67/147) of control group (CG) at-risk women had a 6-month EPDS score > or = 12 versus 33.9% (93/271) of intervention group (IG) women (p = 0.036). A total of 32.9% (46/140) of at-risk women in the CBA group versus 35.1% (46/131) in the PCA group had a 6-month EPDS score > or = 12 (p = 0.74). The CG mean 6-month EPDS score for at-risk women was 11.3 (SD 5.8) versus 9.2 (SD 5.4) for the IG (p = 0.002) and this remained statistically significant after adjusting for 6-week variables (p = 0.001). In total, 16.4% (150/914) of all women in the CG had a 6-month EPDS score > or = 12 compared with 11.7% (205/1745) in the IG (p = 0.003). The CG mean 6-month EPDS score for all women was 6.4 (SD 5.2) compared with 5.5 (SD 4.7) for the IG (p < 0.001). The economic analysis results showed a consistent pattern of psychological approaches being cost-effective at funding levels used by the National Institute for Health and Clinical Excellence.

CONCLUSIONS

HV training was effective compared with HV usual care in reducing the proportion of at-risk women with a 6-month EPDS score > or = 12, with a wide confidence interval for the estimated intervention effect, suggesting that the true treatment effect may be small. The effect remained for 1 year. The economic evaluation demonstrated that the HV intervention was highly likely to be cost-effective compared with the control. There was no difference in outcomes between the CBA and the PCA groups.