Mediation analysis of decisional balance, sun avoidance and sunscreen use in the precontemplation and preparation stages for sun protection

Abstract

Objective: Mediation analyses of sun protection were conducted testing structural equation models using longitudinal data with three waves. An effect was said to be mediated if the standardised path between processes of change, decisional balance and sun protection outcomes was significant.

Design: Longitudinal models of sun protection using data from individuals in the precontemplation (N = 964) and preparation (N = 463) stages who participated of an expert system intervention.

Main outcome measures: Nine processes of change for sun protection, decisional balance constructs of sun protection (pros and cons), sun avoidance behaviour and sunscreen use.

Results: With the exception of two processes in the preparation stage, processes of change predicted the pros (r = .126–.614), and the pros predicted the outcomes (r = .181–.272). Three models with the cons as mediator in the preparation stage, and none in the precontemplation stage, showed a mediated relationship between processes and outcomes.

Conclusion: In general, mediation analyses found both the process of change-to-pros and pros-to-behaviour paths significant for both precontemplation and preparation stages, and for both sun avoidance and sunscreen use outcomes. Findings provide support for the importance of assessing the role of underlying risk cognitions in improving sun protection adherence.

Keywords:

• sun protection
• decisional balance
• mediation
• transtheoretical model of behaviour change

Acknowledgement

The authors acknowledge the support received by the Department of Oncological Sciences-Icahn School of Medicine at Mount Sinai.

Disclosure statement

The authors declare there are no conflicts of interest.

Notes

1. Data from the contemplation stage were available, but it wasn’t included in the mediation analysis given that the sample was small for longitudinal assessments.

Additional information

Funding

Funding for this research was provided by the National Cancer Institute [grant number #P01CA050087]; the National Institute of Aging [grant number #R01AG024490]; and the National Cancer Institute [grant number #R25CA081137].