Adherence to manualised peer support for people with mental illness: the UPSIDES fidelity scale

Ramona HILTENSPERGER1, Marianne FARKAS2, Yasuhiro KOTERA3, Richard MPANGO4, Rebecca NIXDORF5, Bernd PUSCHNER1

1Department of Psychiatry and Psychotherapy II of Ulm University , Ulm, Germany
2Center for Psychiatric Rehabilitation, Boston University, Boston, United States
3School of Health Sciences, Institute of Mental Health, University of Nottingham, Nottingham, United Kingdom
4Butabika National Referral Hospital, Kampala, Uganda
5Department of Psychiatry, University Medical Center Hamburg-Eppendorf, Hamburg, Germany

Introduction

Fidelity describes the extent to which an intervention is delivered as intended. Assessing adherence to core ingredients of peer support is vital for successful implementation and intervention delivery. Modifications to the implementation of peer support are needed when scaling up to different social, cultural and resource settings. There is a need to develop psychometrically sound scales which measure fidelity to established core principles of peer support. This presentation describes stages of development of the UPSIDES fidelity scale, and reports its psychometric properties.

Method

After construction, scale refinement involved site-level expert consultation, resulting in a service user and a peer support worker version. Both versions were translated into six languages and evaluated at six study sites across five countries (Germany, Uganda, Tanzania, India, Israel). The scale was piloted with a total of 315 participants (257 service users and 58 peer support workers). Psychometric evaluation included analysis of internal consistency, construct validity, and criterion validity.

Results

The final versions the scale consist of 28 items (service users) and 21 items (peer support workers) resp. Both are divided into two subscales (implementation and active ingredients). Item statistics showed a skewed distribution of fidelity values but no restriction of range. The scale showed good levels of internal consistency for both the service user and peer support worker version. The confirmatory factor analyses suggest acceptable fits of the proposed factor structures for both versions. Correlations with relevant external criteria were moderate to high.

Discussion

The UPSIDES fidelity scale is a feasible, reliable, and valid tool to assess fidelity to manualised peer support. The scale will advance process and outcome research on peer support, by supporting a multi-informant perspective on fidelity, enabling exploration of similarities and differences in  fidelity ratings between providers and recipients of peer support.

Funding: EU H2020, GA 779263.