Analysing N-of-1 observational data in health psychology and behavioural medicine: a 10-step SPSS tutorial for beginners

Suzanne McDonald; Rute Vieira; Derek W Johnston

doi:10.1080/21642850.2019.1711096

Analysing N-of-1 observational data in health psychology and behavioural medicine: a 10-step SPSS tutorial for beginners

Health Psychol Behav Med. 2020 Jan 22;8(1):32-54. doi: 10.1080/21642850.2019.1711096.

Authors

Suzanne McDonald¹, Rute Vieira², Derek W Johnston³

Affiliations

¹ Centre for Clinical Research, The University of Queensland, Brisbane, Australia.
² Institute of Applied Health Sciences, University of Aberdeen, Aberdeen, United Kingdom.
³ School of Psychology, University of Aberdeen, Aberdeen, United Kingdom.

Abstract

Background: N-of-1 observational studies can be used to describe natural intra-individual changes in health-related behaviours or symptoms over time, to test behavioural theories and to develop highly personalised health interventions. To date, N-of-1 observational methods have been under-used in health psychology and behavioural medicine. One reason for this may be the perceived complexity of statistical analysis of N-of-1 data. Objective: This tutorial paper describes a 10-step procedure for the analysis of N-of-1 observational data using dynamic regression modelling in SPSS for researchers, students and clinicians who are new to this area. The 10-step procedure is illustrated using real data from an N-of-1 observational study exploring the relationship between pain and physical activity. Conclusion: The availability of a user-friendly and robust statistical technique for the analysis of N-of-1 data using SPSS may foster increased awareness, knowledge and skills and establish N-of-1 designs as a useful methodological tool in health psychology and behavioural medicine.

Keywords: N-of-1; dynamic regression modelling; observational; single-case study; statistical analysis.

Grants and funding

Funding from the Menzies Centre for Australian Studies, King’s College London, University of London (Bicentennial Fellowship Award) awarded to SM to support the conduct of the study described in this tutorial article is gratefully acknowledged.