The influence of lifestyle factors on the long-term health-related quality of life in breast cancer survivors and in comparison to a control cohort

Abstract

Background: In 2020, breast cancer (BC) was the most diagnosed cancer globally as well as in Germany. Due to the demographic ageing and the improvement of cancer therapy, an increase in long-term breast cancer survivors (BCS) was observed. 90% of the BCSs reported late and long-term sequelae caused by the disease and its treatment, which reduced their health-related quality of life (HRQoL). However, adjustable factors, such as lifestyle factors (LSF), on the long-term HRQoL were not conclusively studied yet, and often these were not compared to a control group. Findings on this topic could help to develop interventions to support and improve the quality of life (QoL) of long-term BCSs. Aim: This thesis aimed to investigate which LSF determine the HRQoL in long-term BCSs at approximately 5 and ≥10 years post-diagnosis, and to compare them to cancer-free controls 10 years post-recruitment. Methods: This thesis included women who participated in the MARIE study, a German population-based cohort study composed of 3813 breast cancer women aged 50-74 years at diagnosis (2002-2005), and 7341 age-matched breast cancer-free women. Information on social-economic status, lifestyle factors, and medical history and conditions were collected at baseline (BL) in a face-to-face interview. HRQoL was estimated with the EORTC QLQC30 questionnaire. To analyse the influence of the LSFs, such as physical activity, smoking behaviour, and alcohol consumption, two linear regression models were created. Model 1 was a generalized linear mixed model with an interaction between time-point assessment and LSFs using the participant number as a random effect variable. This model investigated the influence of the LSFs on the long-term HRQoL of cases at 5 and ≥10 yeas post-diagnosis to check whether LSFs had a different effect on the HRQoL at different time-point. Model 2 was a linear regression with an interaction between case-control status and LSFs, which was used to investigate LSFs as determinants of HRQoL of cases to controls ≥10 years postrecruitment. This model was used to look at whether the LSFs had a different influence on the HRQoL of the cases from the controls. Results: LFSs influenced differently the HRQoL of cases at FU1 and FU2, and controls at FU2. Transport PA (walking and cycling) influenced the most subscale of the QoL of cases, although it lost significance at FU2, while alcohol consumption had the least impact on the HRQoL. At FU2, all LSFs influenced similarly many subscales of QoL of cases and controls. Just LSFs related to smoking behaviour influenced differently the HRQoL and had an influence just in the control group. Conclusion: LSFs determined the HRQoL of cases at FU1 but lost importance at FU2. Just smoking behaviour influenced differently some subscale of the HRQoL of cases and controls at FU2, but in general, LSFs still determined the HRQoL. Researchers and clinicians should be aware of these determinants of QoL in order to develop interventions to support the needs of long-term BCSs.