PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of hhspaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Occup Environ Med. Author manuscript; available in PMC 2017 August 1.
Published in final edited form as:
PMCID: PMC4978136
NIHMSID: NIHMS791823

Response to Letter to the Editor

Sadie H. Conway, Ph.D., Post-Doctoral Research Fellow,* Lisa A. Pompeii, Ph.D., Associate Professor, Robert E. Roberts, Ph.D., Professor, Jack L. Follis, Ph.D., Assistant Professor, and David Gimeno, Ph.D., Associate Professor

To the editor

The authors thank Dr. Kawada for his interest in our study and appreciate the continued opportunity to discuss the manuscript “Dose-Response Relation between Work Hours and Cardiovascular Disease Risk: Findings from the Panel Study of Income Dynamics”.

Dr. Kawada indicated that specific risk factors for cardiovascular disease (CVD) that are also associated with long work hours (LWH) were not included in our analysis of the relationship between LWH and CVD. This was acknowledged in the manuscript as a study limitation resulting from the lack of availability of those variables in the Panel Study of Income Dynamics (PSID), which was the source of the data. The PSID did not capture variables related to sleep patterns, shift work, or workplace personal relationships. Although the PSID provided information on participants’ diagnosable mental health conditions, this information lacked occupational context, and it could not be determined if job-related factors influenced the mental health of participants. There also was no basis for determining the relationship between sleep patterns and mental health, as the PSID did not have sleep variables.

Sleep disruption due to long work hours and contributing to the development of CVD would be an intermediate variable; similarly, poor mental health resulting from LWH and increasing the risk of poor sleep quality would be on the pathway between the exposure and outcome. As intermediaries, adjusting for either would have introduced over-adjustment bias into our measures of association.1 Although we appreciate Dr. Kawada’s suggestion of adjustment variables, we believe our model is appropriately adjusted. While we acknowledge that residual confounding may exist in our measures of association, we do not believe the magnitude of residual confounding would nullify the observed results.

Further, it should be noted that the identification of a causal relationship between sleep duration and CVD was not a goal of this study. It also was not a goal of this study to evaluate the influence of mental health on CVD.

In summary, our analysis of panel survey data drawn from a representative sample of U.S. households allowed us to identify a dose-response relationship between increasing work hour durations and the risk of CVD, thereby contributing to the growing body of longitudinal studies demonstrating an association between working long hours and adverse health outcomes.

Acknowledgments

Acknowledgement of Funding

Funding for this project was received from the National Institute for Occupational Safety and Health (Grant No. 5T42OH008421).

Footnotes

Conflicts of Interest:

None declared.

Contributor Information

Sadie H. Conway, University of Texas Health Science Center, School of Public Health, Houston, Texas, USA.

Lisa A. Pompeii, University of Texas Health Science Center, School of Public Health, Houston, Texas, USA.

Robert E. Roberts, University of Texas Health Science Center, School of Public Health, San Antonio, Texas, USA.

Jack L. Follis, University of St. Thomas, Houston, Texas, USA.

David Gimeno, University of Texas Health Science Center, School of Public Health, San Antonio, Texas, USA.

Reference

1. Schisterman EF, Cole SR, Platt RW. Overadjustment bias and unnecessary adjustment in epidemiologic studies. Epidemiology. 2009;20:488–495. [PMC free article] [PubMed]