A pilot cluster-controlled trial of interventions to improve detection of depression in primary healthcare in Ethiopia.

BACKGROUND

The low recognition of depression in primary healthcare (PHC) remains a major obstacle to rendering adequate care for people with depression globally. This study aimed to evaluate the feasibility and potential benefit of a contextually developed multicomponent and multilevel intervention to improve the identification of depression in PHC.

METHODS

A pilot, four-arm, parallel-group, cluster, non-randomised controlled trial was conducted in a predominantly rural district in Ethiopia. The active interventions were allocated to three PHC facilities: (1) a core multicomponent intervention focusing on providers-a manualised training package along with system intervention (mobile application, posters, quality improvement and supervision) (Level-I/Arm I), (2) Level-I intervention plus a 4-item screening questionnaire administered by triage nurses (Level-II/Arm II), (3) Level-II intervention plus service user awareness raising (Level-III/Arm III). In the control facility, standard integrated mental healthcare (care by providers trained in the standard WHO mhGAP intervention guide) was available. The outcomes were the identification of depression and the feasibility and acceptability of implementation by PHC clinicians. Quantitative and qualitative data were collected post-intervention. Descriptive analysis and thematic analysis were used to analyse the data.

RESULTS

A total of 21 providers (14 clinicians and 7 triage nurses) and 1659 adult outpatients participated in the study. Overall, 116 outpatients (7.0%) received a diagnosis of depression by PHC clinicians. Detection of depression was significantly better in the active intervention arms combined: 8.3% (n = 115/1380) vs. 0.4% (n = 1/279) in the control arm. Level-II and Level-III intervention arms had significantly higher rates of detection (10.1% Level II, 9.2% Level III) compared with Level I (5.2%); however, there was no significant difference between Level-II and Level-III. The interventions demonstrated very good acceptability, feasibility and appropriateness although screening, which was included in the Level II and Level III intervention arms, had relatively lower acceptability and an overall low positive predictive value.

CONCLUSIONS

The tested interventions hold good promise of enhancing the identification of depression in PHC, with excellent feasibility and acceptability parameters. Although screening has good potential, implementation in routine care requires further optimisation and evaluation.

TRIAL REGISTRATION

Trial ID: PACTR202206723109626. Registration date: 21 June 2022. Retrospectively registered.