Short communication
Psychological distress and cancer mortality

https://doi.org/10.1016/j.jpsychores.2008.11.002Get rights and content

Abstract

Background

Psychological distress, such as ongoing depression and anxiety-related symptomatology, has been associated with a higher risk of incident cancer and poorer survival, although previous studies have not compared prognostic and etiological effects within the same sample. We examined the association between psychological distress and cancer mortality in a sample comprising participants with and without previous cancer admissions.

Methods

Data were collected from a community-based sample of 15,453 men and women (including 295 people with cancer history) and prospectively linked to a patient-based database of cancer registry and deaths during an average follow-up of 7.0±3.3 years. Psychological distress was assessed using the 12-item version of the General Health Questionnaire (GHQ-12).

Results

There were 425 incident cancer deaths. Psychological distress (GHQ-12 ≥4) was associated with increased cancer mortality in participants with cancer history [age, gender, social status, marital status, body mass index, smoking, alcohol, and physical activity; adjusted hazard ratio (HR)=1.97; 95% confidence interval (95% CI)=1.05–3.71; P=.035], but not in participants without cancer history. Among participants without cancer history, there was, however, an association between distress and lung cancer death (age- and gender-adjusted HR=2.04; 95% CI=1.36–3.06; P=.001), although adjustment for covariates attenuated this association.

Conclusions

Psychological distress was a predictor of cancer mortality, especially in lung cancer. The presence of participants with cancer history in community-based cohorts may overestimate the association between psychological distress and subsequent cancer mortality.

Introduction

Cumulative evidence suggests that genetic, environmental (e.g., pollution, infection, and geophysical factors), lifestyle (e.g., tobacco, alcohol, diet, and physical activity), endocrine, and socioeconomic factors play a role in carcinogenesis [1]. Numerous epidemiological studies have also investigated psychosocial risk factors for cancer [2], [3], [4], [5], although the association between these variables and cancer is incompletely understood. A meta-analysis focusing on only prospective studies recently showed that psychological distress, including depression, anxiety, and poor quality of life, was associated with a 13% increase in the risk of cancer incidence and a 27% increase in the risk of cancer mortality [5], but a large amount of heterogeneity was observed between studies. Psychological distress also predicts poorer prognosis in subjects with prior cancer diagnosis [5], although etiological and prognostic effects have not been previously compared within the same study samples. These types of study are important because the comparison of associations observed in participants with and without cancer history are more precise when the populations are drawn from the same study setting with equivalent methodology.

It is feasible that the presence of participants with cancer history within a community-based sample might mask the true influence of psychological distress on cancer mortality and might contribute to heterogeneity. This type of analysis requires a large sample size for a more detailed comparison. Therefore, the aim of the present study was to investigate whether the presence of participants with cancer history might bias the association between psychological distress and cancer mortality.

Section snippets

Study sample and procedures

The Scottish Health Survey (SHS) is a periodic survey (typically every 3–5 years) that draws a nationally representative sample of the general population living in households. The present analyses combined data from the 1995, 1998, and 2003 SHS, which incorporated 15,453 participants (44.9% male; mean age, 53.0±12.1 years). Participants gave full informed consent to participate in the study, and ethical approval was obtained from the London Research Ethics Council. Survey interviewers visited

Results

There were 295 participants with cancer history at baseline, and their cancer diagnosis was made, on average, 4.9±1.5 years before the baseline assessment. Among these participants, 47.8% were admitted with breast cancer, 21.2% were admitted with bowel cancer, 12.6% were admitted with bladder cancer, and the remainder were admitted with cancer of the trachea/lung (3.8%), prostate (9.2%), stomach (1.7%), and ovaries (3.8%). Participants with cancer history were older and more likely to be

Discussion

In the present study, we observed an association between psychological distress and incident cancer death, although the presence of participants with cancer history appeared to contribute to an inflated effect. When the analysis was limited to participants without cancer history, psychological distress was associated with an increased risk of lung cancer mortality. This association persisted after removing deaths in the first year of follow-up; thus, it is unlikely that this finding can be

Acknowledgments

Drs. Hamer and Molloy were supported by the British Heart Foundation, UK, and Dr. Chida was supported by the Kanae Foundation for the Promotion of Medical Science, Japan. The SHS is funded by the Scottish Executive. We thank Professor James Coyne for his helpful comments on an earlier draft of this manuscript.

References (15)

There are more references available in the full text version of this article.

Cited by (217)

View all citing articles on Scopus

The views expressed in this article are those of the authors and not necessarily of the funding bodies.

View full text