Preferences of patients with depression for exercises: a discrete choice experiment

BMC Public Health. 2025 Mar 29;25(1):1192. doi: 10.1186/s12889-025-22404-0.

ABSTRACT

BACKGROUND: Exercise brings great benefits to patients with depression and is recommended as adjunct treatment for depressive disorder. The low compliance to exercise makes the effective management of depression a serious challenge. Preferences for exercise are important for promoting intrinsic motivation, enjoyment, and behavior maintenance. This study quantified the exercise preferences of patients with depression by a discrete choice experiment (DCE), which provided references for the development of intervention programs, so as to improve the physical activity level of patients with depression and promote the effective management of depressive symptoms.

METHODS: The six attributes (Professional guidance, Interestingness, Type, Special venue, Frequency, and Safety) and their levels were developed through a literature review, in-depth interviews, and focus group discussions. The software Ngene 1.2 was used to generate 36 choice sets, which were divided into 4 blocks. A survey instrument was developed adhering to the design-principles of DCE. A mixed logit model was used to estimate the preferences of patients with depression. The choice probability of exercise scenarios were predicted.

RESULTS: The preferences of 323 depressed patients were analyzed. High level of interestingness was the most important attribute level (coefficient = 0.84, P < 0.01), followed by high safety (coefficient = 0.79, P < 0.01). There is a gender-based and daily steps-based difference in preference. The optimal exercise strategy should be fun and safe, exercise 3-4 times a week, venue without restrictions, professional guidance, and a combination of teams and individuals.

CONCLUSION: Based on the findings, patients with depression in this target population prefer exercises with attributes such as intrerestingness, safety, while considering gender-based and daily steps-based deference in preferences.

PMID:40158084 | DOI:10.1186/s12889-025-22404-0