Objective
Participation bias may be a problem in couple-based psychosocial studies. Therefore, it is important to investigate the characteristics associated with participation. The aim of this study was to analyze whether participation in a longitudinal psychosocial questionnaire study among couples affected by breast cancer was associated with socioeconomic, breast cancer-specific, and other health-related characteristics of the patients and partners.

Methods
The analyzes are based on 2254 couples who were invited to participate in a nationwide survey on psychosocial adjustment among couples dealing with breast cancer. Participating couples (N = 792) were compared with non-participating couples (N = 1462) with regard to socioeconomic and health-related characteristics obtained from nationwide clinical and administrative registers.

Results
Associations were seen between various socioeconomic variables and couple participation. The patient characteristics older age (OR = 0.15 [95 % CI = 0.07–0.55]), low education (OR = 1.95 [95 % CI = 1.46–2.68]), disability pension (OR = 0.59 [95 % CI = 0.39–0.55]), or non-western ethnicity (OR = 0.36 [95 % CI = 0.15–0.82]) reduced couple participation. The partner characteristics older age (OR = 0.23 [95 % CI = 0.15–0.43]), low education (OR = 1.67 [95 % CI = 1.25–2.22]), receiving disability pension (OR = 0.46 [95 % CI = 0.25–0.82]), non-western ethnicity (OR = 0.17 [95 % CI = 0.06–0.49]), or high morbidity (OR = 0.76 [95 % CI = 0.60–0.96]) also reduced couple participation. Furthermore, couples with low income (OR = 1.49 [95 % CI = 1.16–1.95]) had reduced participation. No associations were found between couple participation and breast cancer-related variables.

Conclusions
Socioeconomic characteristics of patients and partners, and morbidity of partners may influence participation in couple-based psychosocial breast cancer research. Breast cancer-related characteristics do not seem to influence participation.