Research reportDepression and vulnerability to incident physical illness across 10 years
Introduction
The role of physical illness in predicting depression is well-established across a wide spectrum of illnesses. For example, severe, potentially life-threatening illnesses, including heart disease (Holahan et al., 1995), stroke (Robinson, 2003), and cancer (Massie, 2004), are associated with depression. In addition, patients dealing with chronic pain-inducing illnesses, such as back pain (Currie and Wang, 2004), arthritis (Keefe et al., 2002), and migraine (Breslau et al., 2003), are vulnerable to depression. However, while considerable research exists on the role of physical illness in initiating depressive reactions, the reverse process—the role of depression in the onset of physical illness—is much less studied (Farmer et al., 2008, Patten et al., 2008). Yet, this topic is central to the quality of life of depressed individuals and, more broadly, to medical care and to reducing the illness burden on society (Evans et al., 2005). The purpose of the present study was to investigate the prospective relationship between depression and a wide spectrum of physical illnesses across a 10-year period.
Accumulating evidence suggests that depression may increase morbidity and mortality from physical illness (Stover et al., 2003). Evidence for the role of depression in predicting physical illness can be divided into two lines of inquiry. Most research has examined the role of depression in illness course, recovery, and relapse among individuals with existing specific physical illnesses. More recently, however, the investigative focus has broadened to include the role depression plays in predicting the onset of new illnesses among previously healthy individuals. Almost all of this more recent work has focused on depression as a predictor of specific physical illnesses. However, interest is now emerging in the potential predictive role of depression in incident physical illness more generally.
Although depression has been found to predict the course of many physical illnesses, the strongest evidence pertains to heart-related illnesses and, secondarily, to chronic pain. For example, among patients with coronary artery disease followed for from 1 to 20 years, both major depression (Carney et al., 1988) and depressive symptoms (Whooley et al., 2008) are positively associated with adverse cardiac events, including myocardial infarction and cardiac-related mortality. Moreover, considerable evidence indicates that major depression (Frasure-Smith et al., 1993) and depressive symptoms (Ahern et al., 1990, Frasure-Smith et al., 1995, Welin et al., 2000), including low levels of depressive symptoms (Bush et al., 2001) are associated with an increased risk of mortality after myocardial infarction.
Further, both major depression (Connerney et al., 2001) and depressive symptoms (Burg et al., 2003, Contrada et al., 2008, Jenkins et al., 1996) are positively associated with poorer recovery, including less symptom relief and a higher likelihood of cardiac-related rehospitalization in patients recovering from heart, primarily coronary artery bypass, surgery. Major depression is also associated with more intense and longer lasting pain and greater and longer-term disability among chronic pain patients, including patients with diffuse pain at multiple sites (Ericsson et al., 2002), chronic (primarily low back) pain (Bair et al., 2003), and rheumatoid arthritis, (Dickens et al., 2002, Lowe et al., 2004).
More recent work suggests that depression is a predictor of specific physical illnesses in initially healthy individuals. Again, evidence is especially clear with heart-related illnesses and, secondarily, chronic pain. Two independent reviews found that the relative risk for the onset of coronary heart disease was 1.64 for depressed compared to non-depressed individuals followed for an average of 13 years (Rugulies, 2002, Wulsin and Singal, 2003). For example, studies that have followed large population samples for from 4 to 10 years have found that both major depression (Surtees et al., 2008) and depressive symptoms (Ferketich et al., 2000, Marzari et al., 2005) predict incident coronary heart disease in individuals initially free of heart conditions, controlling for common risk factors. There is also evidence that major depression is associated with a marginally increased risk for incident hypertension (Meyer et al., 2004).
Depressed patients are also more likely to report unexplained pain than non-depressed patients. For example, population-based studies have reported links between major depression and incident chronic back pain (Currie and Wang, 2005, Larson et al., 2004) and between depressive symptoms and incident arthritis (Seavey et al., 2003). Further, both major depression (Eaton et al., 1996) and depressive symptoms (Engum, 2007, Golden et al., 2008) have been linked to incident type 2 diabetes in large samples followed for from 3 to 10 years. There is also some evidence from a large sample followed for more than10 years that depressive symptoms may predict incident ischemic stroke (Arbelaez et al., 2007).
Very recently, research on the role of depression as a risk factor for incident physical illness has expanded from a focus on specific physical illnesses to encompass physical illness more generally. For instance, after an extensive literature review, Evans et al. (2005) concluded that the evidence suggests a potential for depression as a risk factor for incident physical illness across a spectrum of illnesses. However, the authors also noted inconsistencies in the literature on depression and incident physical illness and called for additional prospective studies.
Three recent empirical studies offer further support for a potential link between depression and incident physical illness more generally. Farmer et al. (2008) retrospectively examined the concordance between lifetime histories of recurrent depression and several physical health diagnoses among over 1500 individuals with recurrent depression and almost 900 controls in the United Kingdom. The authors found higher rates of reported physical illnesses, including myocardial infarction, liver disease, gastric ulcer, osteoarthritis, and rheumatoid arthritis, among depressed individuals compared to psychiatrically healthy controls.
In addition, McCusker et al. (2007) prospectively followed over 200 older medical patients for 1 year. The investigators found that, controlling for prior illness, major depression prospectively predicted poorer subsequent physical health status on the Medical Outcome Study Short Form Health Survey (SF-36). Further, Patten et al. (2008) prospectively followed over 15,000 adults in the Canadian National Population Health Survey for up to 8 years. Controlling for prior illness, the authors reported higher rates of several chronic physical illnesses, including heart disease, hypertension, asthma, chronic bronchitis/emphysema, arthritis/rheumatism, back problems, and migraine, among individuals with major depression compared to those without major depression.
The present study extends previous research on depression and physical illness by examining the role of depression as a vulnerability factor for incident physical illness in a prospective design controlling for prior medical conditions. Based on the few available studies on depression and physical illness more generally (Farmer et al., 2008, McCusker et al., 2007, Patten et al., 2008), it was hypothesized that, controlling for prior medical conditions, depression would prospectively predict the likelihood of experiencing new medical illnesses across a 10-year follow-up period.
The present study followed 388 clinically depressed patients and 404 matched community controls across 10 years to examine the prospective link between depression and physical illness. The data are part of a longitudinal project examining the long-term course of unipolar depression. Earlier research on the 10-year follow-up of these samples has examined life context factors in depression remission and relapse (Cronkite et al., 1998, Holahan et al., 2000, Moos et al., 1998) and the role of avoidance coping in predicting drinking behavior (Holahan et al., 2003, Holahan et al., 2004). No previous research with this database has focused on depression and vulnerability to physical illness.
Section snippets
Sample selection and characteristics
Two samples of adults (age 18 or older) were selected: a sample of depressed patients who were entering treatment for unipolar depressive disorders and a sample of community controls who were matched with the patients in terms of area of residence and marital status (for more information on these samples, see Cronkite et al., 1998, Moos et al., 1998). Sociodemographic and health behavior data were indexed at baseline and medical conditions were indexed at baseline and at 1, 4, and 10 years
Tests of potential of control variables
In developing the logistic regression models, we examined six potential control variables (sex, age, tobacco smoking, drinking problems, physical activity, and financial resources) among participants included in the prospective analyses (N = 792). Table 1 shows group comparisons on the potential control variables at baseline for depressed patients and community controls. Depressed patients and community controls did not differ significantly on sex or age. However, at baseline, the depressed
Discussion
The present study followed 388 clinically depressed patients and 404 matched community controls across 10 years to examine the prospective link between depression and physical illness. After accounting for the effects of prior physical illness, as well as of key demographic and health behavior factors, membership in the depressed group was significantly linked to physical illness during the follow-up period. Previous research on the co-morbidity between depression and physical illness has been
Role of funding source
This work was supported by the Marchionne Foundation and Department of Veterans Affairs Health Services Research and Development Service research funds; the Marchionne Foundation and the Department of Veterans Affairs had no further role in study design; collection, analysis, and interpretation of data; writing of the report; or decision to submit the paper for publication.
Conflict of interest
No authors have any actual or potential conflict of interest that could inappropriately influence, or be perceived to influence, this work.
Acknowledgment
We thank John Finney, Ph.D., for advice on the statistical analyses.
References (70)
- et al.
Biobehavioral variables and mortality or cardiac arrest in the Cardiac Arrhythmia Pilot Study (CAPS)
Am. J. Cardiol.
(1990) - et al.
Psychological status and depression in patients with liver cirrhosis
Dig. Liver Dis.
(2005) - et al.
Psychiatric disorders and stages of smoking
Biol. Psychiatry
(2004) - et al.
Even minimal symptoms of depression increase mortality risk after acute myocardial infarction
Am. J. Cardiol.
(2001) - et al.
Relation between depression after coronary artery bypass surgery and 12-month outcome: a prospective study
Lancet
(2001) - et al.
Chronic back pain and major depression in the general Canadian population
Pain
(2004) The role of depression and anxiety in onset of diabetes in a large population-based study
J. Psychosom. Res.
(2007)- et al.
Mood disorders in the medically ill: scientific review and recommendations
Biol. Psychiatry
(2005) - et al.
Prospective association of anxiety, depressive, and addictive disorders with high utilization of primary, specialty and emergency medical care
Soc. Sci. Med.
(2004) - et al.
The association between depression and chronic kidney disease and mortality among patients hospitalized with congestive heart failure
Am. J. Kidney Dis.
(2004)