Association of Paid Sick Leave With Job Retention and Financial Burden Among Working Patients With Colorectal Cancer

Workers who develop serious illnesses, such as colorectal cancer (CRC), can incur economic hardship, regardless of insurance coverage.^1-4 Paid sick leave could reduce the need to take unpaid time off during treatment. However, 40% of US workers have no paid sick leave. Its provision is not mandated under the Affordable Care Act or the Family Medical Leave Act, nor is it part of health insurance coverage.⁵

This study examined the association between access to paid sick leave and job retention and personal financial burden among patients with CRC.

After the relevant institutional review boards granted a waiver of consent, surveys were mailed to and telephone follow-up conducted with adults with stage III CRC reported to the Surveillance, Epidemiology, and End Results cancer registries of Georgia and metropolitan Detroit between August 2011 and March 2013. Patients were contacted 4 months postoperatively and could respond up to 12 months postoperatively; only those employed at diagnosis were analyzed.

The primary outcomes were job retention and personal financial burden. Patient report of causality was determined with the survey item stem, “As a result of cancer or its treatment …” Job retention was defined as employment at both diagnosis and survey completion (mean, 8 months after diagnosis) together with a negative response to the binary stemmed questions “I changed/quit/lost my job.” Personal financial burden was assessed with a previously validated⁴ composite score of six 5-point Likert questions asking how CRC or its treatment affected personal finances. Higher scores denote increased financial burden.

The primary independent variable was access to paid sick leave at the time of diagnosis. Percentages of job retention and mean personal financial burden were calculated among those with and without paid sick leave adjusting for covariates in Table 1. Multiple imputation was used to account for the 12% of missing annual household income data.

We used χ² tests to assess associations and multivariable logistic regression for adjusted analyses. Statistical tests were 2-sided; a P value of <.05 was considered significant. Analyses were conducted with SAS version 9.4 (SAS Institute Inc).

Among 567 employed respondents (68% response rate), 58% were men; 68% were white; 28% had less than a high school education, 35% reported annual household income of less than $50 000, and 56% had access to paid sick leave.

Respondents without vs those with paid leave reported significantly higher personal financial burden: 28% vs 18% borrowed money, respectively, 29% vs 14% had difficulties making credit card payments, 50% vs 35% reduced spending for food and clothing, and 57% vs 47% reduced recreational spending; all P values <.001.

Fifty-five percent retained their jobs. Others were newly disabled (26%), retired (7%), or unemployed (8%) or had found new jobs (4%). Those who retained their jobs were significantly more likely to be men, white, married, without comorbid disease, and were more highly educated and were more likely to have a higher annual income, private health insurance, and access to paid sick leave (Table 1).

After covariate adjustment, 59% (95% CI, 57%-62%) of respondents with paid sick leave retained their jobs vs 33% (95% CI, 31%-34%) without paid sick leave (P < .001). The covariate-adjusted mean personal financial burden score among respondents with paid sick leave was 2.19 (95% CI, 2.11-2.27) vs 2.81 (95% CI, 2.69-2.93) among those without (P < .001). Further adjustment for job retention did not attenuate the association, suggesting that job retention did not explain the reduced personal financial burden among those with paid sick leave (Table 2).

In this population-based survey, 45% of working individuals with stage III CRC did not retain their jobs reportedly due to their cancer diagnosis and treatment. Paid sick leave was associated with a greater likelihood of job retention and reduced personal financial burden. Lack of paid sick leave may be a proxy for low-wage jobs or low socioeconomic status. However, our data show that the association with paid sick leave was robust to education, income, and health insurance.

The analyses were limited to a sample of respondents in 2 states with 1 disease, but the population-based sampling achieved broad demographic representation. In addition, specific reasons for job change or loss were not requested, and causality was self-reported. Other potential contributory factors to job loss may reflect economic trends external to the study or other uncontrolled confounders.

Corresponding Author: Arden M. Morris, MD, MPH, Center for Healthcare Outcomes and Policy, University of Michigan, 1500 E Medical Center Dr, TC2421F, Ann Arbor, MI 48109 (ammsurg@med.umich.edu).

Author Contributions: Dr Morris had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Veenstra, Regenbogen, Hawley, Abrahamse, Morris.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Veenstra, Abrahamse, Banerjee, Morris.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Regenbogen, Abrahamse, Banerjee, Morris.

Obtained funding: Morris.

Study supervision: Hawley, Morris.

Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Veenstra reported receiving institutional funding from Pfizer unrelated to this research letter. No other disclosures were reported.

Funding/Support: Dr Morris and the study are supported by research scholar grant 11-097-01-CPHPS from the American Cancer Society.

Role of the Funder/Support: The American Cancer Society played no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: We thank Ashley Gay, MOA (Department of Surgery, University of Michigan), Kevin Ward, PhD (Department of Epidemiology, Emory University), and Ikuko Kato, PhD (Department of Pathology, Wayne State University), for their assistance in data collection and support. All were supported by research scholar grant 11-097-01-CPHPS from the American Cancer Society.