Prognostic factors for return to work in breast cancer survivors.

Abstract

Background: Breast cancer is the most common type of cancer in women around the world. Large numbers of people diagnosed with breast cancer are working at the time of diagnosis. Accumulating evidence suggests that breast cancer survivors participate less often in paid work compared to others. Return to work among breast cancer survivors is multifactorial. It is currently unknown which factors are associated with return to work in breast cancer survivors. Therefore, it is important to systematically review and synthesize the literature on the association between sociodemographic, breast cancer-related, other health-related, personal, and work-related factors and return to work in this group of people.

Objectives: The objective is to systematically review and synthesize the literature on the association between sociodemographic, breast cancer-related, other health-related, personal, and work-related factors and return to work in the 24 months following breast cancer diagnosis among breast cancer survivors having paid work at the time of diagnosis.

Search methods: The search strategy included electronic searches in OVID/MEDLINE, Embase.com, EBSCOhost/CINAHL with Full Text, EBSCOhost/PsycINFO, Clarivate Analytics/Web of Science Core Collection and Wiley/Cochrane Library from inception up to 20 January 2023, as well as handsearching references of relevant reviews, included studies, and Google Scholar.

Selection criteria: The following inclusion criteria were applied: - The type of study is a prospective cohort study, retrospective cohort study with time lag between assessment of prognostic factor and outcome, or prognosis study based on a randomized controlled trial. - The study sample included people diagnosed with breast cancer, having paid work at the time of their breast cancer diagnosis. - At least one variable as specified in our variable framework was studied. - Return to work (yes/no), or time to return to work was assessed somewhere between one and 24 months of follow-up. - The article type is an original research article (commentaries, reviews, and editorials were excluded). - Full text of the article is available. - The article was published in a peer-reviewed journal.

Data collection and analysis: Study characteristics and estimates of unadjusted and adjusted associations between one of the variables from the pre-defined variable framework and return to work were extracted. Risk of bias was assessed using the Quality in Prognosis Studies (QUIPS) tool. When at least four adjusted or four unadjusted measures of association (e.g. Odds Ratio (OR)) were available and more or less comparable in terms of how the measures of association were included in the analysis of the original study, a meta-analysis was conducted.

Main results: The systematic searches yielded 14,799 records with 2 identified via other sources. The systematic searches yielded 8486 references after duplicates were removed. We assessed 280 full-text articles for eligibility and excluded 249, including one article that was classified as 'awaiting classification' as it required professional translation. This left 31 articles based on 19 cohorts that fulfilled our inclusion criteria. Seven of the 19 studies could be included in one or more meta-analyses with a total of 2473 participants. All but one study were conducted in either Europe or the USA. The return to work rate ranged from 56% to 88%. From our prespecified variable framework, altogether 35 variables were studied in one or more included studies as prognostic factors. From these, we could combine five factors in the meta-analyses. - We found low-quality evidence that higher age is associated with lower odds of return to work in an adjusted analysis (pooled adjusted OR 0.96, 95% confidence interval (CI) 0.94 to 0.98; 4 studies, 1333 participants). - We found low-quality evidence that lower level of education is associated with lower odds of return to work in an unadjusted analysis (pooled unadjusted OR 0.40, 95% CI 0.29 to 0.55; 4 studies, 1680 participants), but not in an adjusted analysis (pooled adjusted OR 0.60, 95% CI 0.33 to 1.08; 4 studies, 1147 participants). - We found low-quality evidence that not having a partner is not associated with return to work in an unadjusted analysis (pooled unadjusted measures of association: 0.91 95% CI 0.67 to 1.23; 4 studies, 1680 participants). - We found low-quality evidence that receiving chemotherapy was associated with lower odds of return to work in an unadjusted analysis (pooled unadjusted measures of association: 0.48, 95% CI 0.31 to 0.73; 5 studies, 1766 participants). - We found low-quality evidence that receiving radiotherapy is not associated with return to work, respectively (pooled unadjusted measures of association: 1.03, 95% CI 0.64 to 1.17; 4 studies, 1680 participants). Due to the low number of included studies that measured the outcome, time to return to work, it was not possible to pool data of these studies.

Authors' conclusions: We found that higher age and receiving chemotherapy may be associated with lower odds of returning to work in breast cancer survivors (low-quality evidence; for chemotherapy, only pooled unadjusted results were available). Results regarding educational level are inconclusive. We furthermore found that there was no statistically significant adjusted association between having a partner and receiving radiotherapy (low-quality evidence; only unadjusted results were available). Further research is warranted to identify those breast cancer survivors who are at higher risk of not returning to work, so that they can receive timely support.