Return to work or leaving work? Differences of return to work between breast cancer patients and the general population and determinants of return to work.

Abstract

Purpose: It was examined whether employment among breast cancer survivors was lower than in the general population 4 to 6 years after surgery. We also examined whether disease severity, post-surgical treatment, social, and workplace characteristics have effects on employment as primary outcome, and whether the distance from surgery to observation may determine employment.

Methods: We performed a multicentric observational study with four survey waves. Data were collected based on mailed surveys and patient records. Patients were up to 63 years old at entry with TNM-tumour stages T0 to TIV. Comparisons with the general population were performed by drawing controls from the German Socio-Economic Panel.

Results: N = 372 breast cancer survivors participated in all surveys (= 82.2% of the initial sample). Their rate of occupationally active women was lower than in the general population (OR_patients = 0.59; 95% CI = 0.42-0.84; p < 0.01). Among patients, tumour stage had no effects on employment 12 months after surgery; 4-6 years later, this was the case only among patients with the most unfavourable tumour stage (OR = 0.16; p = 0.01; 95% CI = 0.04-0.58). Antihormone therapy was unrelated with employment (OR = 0.80; p = 0.27; 95% CI = 0.54-1.19); inpatient rehabilitation was negatively associated at 12 months after surgery (OR = 0.47; p = 0.02; 95% CI = 0.25-0.89) and unrelated at the last survey wave (OR = 0.95; p = 0.86; 95% CI = 0.55-1.64). Compared with the lowest level of occupational autonomy, it was unrelated with employment 12 months after surgery (OR = 0.79; p = 0.75; 95% CI = 0.18-4.41), but for the highest level of autonomy, it had significant effects 4 to 6 years later (OR = 4.56; p = 0.04; 95% CI = 1.10-18.81). Effort-reward imbalance as a continuously scaled indicator of pre-surgery occupational distress was significantly associated with return to work 12 months after surgery (OR = 0.13; p < 0.01; 95% CI = 0.06-0.31), but it had no effect at the last survey wave (OR = 0.64; p = 0.31; 95% CI = 0.28-1.50). One year after surgery, education at higher levels had no significant effects on return to work (OR = 1.30; p = 0.57; 95% CI = 0.56-3.00 for the highest level compared with the lowest one), only at the last measurement marked differences by education emerged (OR = 2.23; p = 0.03; 95% CI = 1.08-4.63).

Conclusion: Temporal distance between surgery and survey wave determines whether potentially influencing factors have effects. Disease severity and post-surgical treatment were unrelated to employment. Whether work-related and socio-demographic factors are determining employment depends on the date of measurement.