Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Nov 1.
Published in final edited form as: Pain Med. 2013 Apr 22;14(5):686–691. doi: 10.1111/pme.12089

Depressive Symptoms, Pain, Chronic Medical Morbidity, and Interleukin-6 among Primary Care Patients

Ellen L Poleshuck 1,2, Nancy L Talbot 1, Jan A Moynihan 1, Benjamin P Chapman 1, Kathi L Heffner 1
PMCID: PMC5665657  NIHMSID: NIHMS913509  PMID: 23691936

Abstract

Objective

Pain, chronic medical morbidity, and depression are highly prevalent problems that frequently co-occur in primary care. Elevated levels of inflammation are linked with all three of these conditions and may play an important role in patients’ comorbidities. The current study aimed to examine if the associations among pain, chronic medical morbidity, and the inflammatory marker interleukin (IL)-6 are dependent on depression status in primary care patients.

Setting, Subjects and Outcome Measures

Primary care patients (N = 106) aged 40 and older were assessed for pain (SF-36), chronic medical morbidity (checklist of chronic health conditions), and depressive symptoms (CES-D), and provided a blood sample for the measurement of serum IL-6.

Results

Among patients with elevated depressive symptoms, higher IL-6 levels were associated with both greater pain and greater chronic medical comorbidity. IL-6 was unrelated to pain or chronic medical comorbidity among patients without clinically significant depressive symptoms. In adjusted analyses, chronic medical morbidity did not fully explain the association between IL-6 and pain, and depression severity and pain remained independently associated after adjustment for chronic medical comorbidity.

Conclusions

Depression may increase primary care patients’ vulnerability to pain and inflammation.

Keywords: persistent pain, depression, inflammation, primary care

Introduction

Depressive symptoms are often elevated among primary care patients with pain (1). Importantly, the presence of depressive symptoms can impede effective management of pain and chronic medical morbidity, which often co-occur (2). Pain, chronic medical morbidity, and depression are each associated with alterations in biological function that may exacerbate symptoms in any or all of these domains and, thus, may help explain their co-occurrence. Understanding the biological links among these comorbidities may help inform comprehensive approaches to treatment addressing both physical and mental health issues.

Neuroimmunological evidence suggests that inflammation may serve as a nexus for understanding links between depression and co-occurring pain and chronic medical morbidity. In clinical studies, greater pain, chronic medical morbidity and depressive symptoms are each independently associated with higher levels of inflammatory markers, including proinflammatory cytokines (3, 4). Proinflammatory cytokines are also implicated in the etiology of depressive symptoms (5, 6), pain (3, 4, 7–15), and a host of chronic diseases that increase risk for depression and pain, including cancer, heart disease and diabetes (16, 17). Little is known, however, about how depression, pain, and inflammation are associated concurrently, or the role of chronic medical morbidity in these associations in primary care patients presenting for care. Understanding the interplay among all these frequently co-occurring factors in primary care patients may help providers to prioritize appropriate targets of treatment and reduce long-term health risks in a patient population for whom complex presentations accompanied by treatment-resistance is a significant problem.

In this study, we explored the associations among depressive symptoms, pain, medical morbidity, and the proinflammatory cytokine interleukin (IL)-6 in an at-risk sample: middle-aged and older low-income patients presenting for care at an urban, primary care clinic. We hypothesized that patients with elevated depressive symptoms would have more pain, higher levels of the proinflammatory cytokine interleukin (IL)-6 and greater chronic medical morbidity, and that they would also show stronger associations among their pain, inflammation and chronic morbidity compared to patients without evidence of depression. We also explored whether any observed associations between IL-6 and pain were explained by chronic medical morbidity.

Methods

Participants and Procedures

This study is part of a larger study on links between psychological factors and physical health in middle-aged and older primary care patients (18). Patients aged 40 and older were recruited in person and through flyers at the Family Medicine Center (FMC) of the University of Rochester Medical Center (URMC), and subsequently attended a study appointment at the FMC or the URMC General Clinical Research Center at their convenience (N=107). One patient withdrew from the study in the year following enrollment, leaving 106 available for the current analysis.

At the study appointment, following written informed consent, participants completed an assessment conducted by B. Chapman or a trained research assistant of demographics, depressive symptoms using the Center for Epidemiologic Studies Depression Scale-Revised (CES-D-R) (19) and self-rated health-related quality of life from the 36-item Medical Outcomes Study Survey Form (SF-36) (19, 20). The SF-36 composite Bodily Pain score was derived from items assessing pain intensity and interference over the past 4 weeks following published guidelines (21); lower scores reflect more severe and limiting pain. Patients also completed a checklist of chronic health-conditions based on Fortin’s primary care based multimorbidity assessment (22, 23). For the checklist, patients reported whether a physician had diagnosed them with any of 25 common chronic medical morbidities spanning respiratory, musculoskeletal, gastro-intestinal, neurological, endocrine, cardiac and vascular systems. These were summed to form a chronic medical morbidity index based on the total number of affected organ systems (23). Following the assessment, patients provided a blood sample via venipuncture performed by a trained phlebotomist or clinical research nurse. The majority of blood samples (approximately 67%) were obtained in the afternoon (a random subsample of patients attended the study session in the morning given availability of research staff). Following venipuncture, blood was kept on ice, centrifuged, and serum stored at −80 °C. Body weight from the medical chart was available for 85 patients; for the remaining 21 patients, body weight was derived by regression-based imputation from a model with R-squares around .20. Patients missing body weight data were demographically similar to those with the data.

IL-6 was assayed using standard enzyme-linked immunosorbent assay protocols and anticytokine antibody pairs (BD Biosciences, San Diego, California). The absorbance of the color reaction was measured at 405 nm using an automated Opsys MR Microplate Reader (Thermo Labsystems, Chantilly, Virginia). The minimum detectable limit for IL-6 is 0.039 pg/mL. The intra-assay variability for both IL-6 is approximately 4% and the inter-assay variability approximately 5%. Patients were compensated $50 for participation in the study procedures. The study was approved by the University of Rochester Internal Review Board.

Data analysis

IL-6 values were log transformed to improve normality

Following descriptive analysis of the sample, demographic and health characteristics of patients with and without elevated depressive symptoms were compared using chi-square or t-tests. Participants were considered to have elevated depressive symptoms if they had a CES-D score ≥ 16, suggesting moderate or greater symptom severity (24). Associations between IL-6 levels, pain reports, and the chronic medical morbidity index in patients with high and low depressive symptoms were examined using Pearson correlations. Logistic and linear regression models were used to further test whether associations between depressive symptomatology and pain, and IL-6 and pain, were independent of chronic medical morbidity. Given well-established associations between age and IL-6, age was selected a priori as a covariate in regression models.

Results

Sample Descriptives

Characteristics of the patients are detailed in Table 1. Using a CESD-R cutoff score of 16, indicating moderate or greater depressive symptom severity, 45.3% (n = 48) of the sample met screening criteria for clinical levels of depression. Patients with and without elevated depressive symptoms did not differ significantly by age, race/ethnicity, education, income or gender distribution (all p’s > .05). Commonly endorsed chronic conditions (that is, at least 10% of patients endorsing the condition) included arthritis (47.1%; n = 50), hypertension (35.8%; n = 38), asthma (19.8%; n = 21), type 2 diabetes (18.9%; n = 20), herniated disk (13.2%; n = 14), osteoporosis (12.3%; n = 13), and persistent migraines (11.3%; n = 12). Self-reported prevalence of persistent migraines was greater for patients with elevated depressive symptoms (18.8%) compared to patients without (5.3%); however, this difference was not statistically significant (χ2(1) = 4.68, p = .06). As seen in Table 1, patients with elevated depressive symptoms had higher body weight than patients without elevated depressive symptoms; therefore, all models were adjusted for body weight in addition to age.

Table 1.

Sample Characteristics for Total Sample and for Patients with Low and High Depressive Symptoms

Total Sample (n = 106) CESD-R < 16 (n = 58) CESD-R ≥ 16 (n = 48) p
Age [mean (range)] 51.8 (40–80) 53.0 (40–80) 50.4 (40–70) .13
Female [n (%)] 83 (78.3) 47 (81.0) 36 (75.0) .49
Race [n (%)]a
 White, non-Hispanic 44 (41.5) 25 (43.1) 19 (39.6) .84
 Black, non Hispanic 57 (53.8) 28 (48.3) 29 (60.4) .32
 Hispanic 3 (2.8) 2 (3.4) 1 (2.1) 1.0
 American Indian or Alaska Native 5 (4.7) 4 (6.9) 1 (2.1) .37
 Other 1 (0.9) 1 (2.1) 0 (0.0) 1.0
Education [n (%)]
 No high school diploma 31 (29.2) 16 (27.6) 15 (31.3) .21
 GED or graduated from high school 21 (22.6) 9 (15.5) 15 (31.3)
 Some college 9 (8.5) 5 (8.6) 4 (8.3)
 Associate’s degree or 2 years of college 24 (22.6) 14 (24.1) 10 (20.8)
 College graduate 11 (10.4) 9 (15.5) 2 (4.2)
 Graduate degree 7 (6.6) 5 (8.6) 2 (4.2)
Household income level [n (%)]
 Less than $20,000 per year 65 (61.3) 31 (53.4) 34 (70.8) .08
 Equal to or greater than $20,000 41 (38.7) 27 (46.6) 14 (29.2)
Body weight (lb) 204.8 (47.6) 196.4 (39.9) 214.9 (54.2) .05
SF-36 Bodily pain score [mean (SD)]b 48.5 (30.7) 57.4 (28.5) 38.0 (30.0) .001
 Pain intensity rating [mean (SD)] 3.4 ( 1.6) 3.8 (1.5)c 2.9 (1.5)d .004
 Pain interference [mean (SD)] 3.5 (1.6) 4.6 (1.5)e 3.3 (1.8)f < .001
CESD-R Depressive symptoms [mean (SD)] 17.0 (13.4) 6.9 (4.3) 29.3 (9.8) < .001
Chronic medical morbidity Index [mean(SD)] 3.9 (2.9) 3.2 (2.2) 4.7 (3.5) .006
IL-6 (pg/mL) [mean(SD)]g 3.84 (2.97) 3.70 (2.82) 4.02 (3.16) .59
Open in a new tab

CESD-R = Center for Epidemiological Studies Depression Scale-Revised; SF-36 = Medical Outcomes Study Survey Form – 36 item

a

Race/ethnicity proportions total more than 100% due to 4 subjects endorsing more than one race/ethnicity.

b

Bodily pain scores represent population-based T scores derived from the pain intensity and pain interference items of the SF-36. Higher scores indicate less impact of pain in the past 4 weeks. Scores on the pain intensity item ranges from 1 (very severe) to 6 (none) and scores on the pain intensity item range from 1 (extremely) to 6 (none).

c

Equates to mild to moderate pain intensity

d

Equates to moderate to severe pain intensity

e

Equates to not-at-all to a-little-bit pain interference

f

Equates to a-little-bit to moderate pain interference

g

IL-6 data were missing for 3 patients. Comparisons between patients with high and low depressive symptoms were analyzed using log transformed IL-6 values.

Depression, Pain, Chronic medical morbidity, and IL-6

Patients with elevated depressive symptoms reported more bodily pain, including more pain intensity and pain interference (higher scores indicate less pain), and had more chronic medical morbidity relative to patients below the screening cutoff for depression (Table 1). IL-6 levels did not differ between patients with and without elevated depressive symptoms.

Among patients with elevated depressive symptoms, higher levels of IL-6 were associated with lower SF-36 Bodily Pain scores (reflecting more severe, limiting pain; r(42) = −.39, p = .01; see also Table 2, Model 1) and more chronic medical morbidity (r(42) = .32, p = .03), after controlling for body weight and age. Among patients without elevated depressive symptoms, no associations were observed between IL-6 and pain (r(51) = .14, p = .32), and IL-6 and chronic medical morbidity (r(53) = .10, p = .48) after adjustment for body weight and age.

Table 2.

Mediation Analyses of SF-36 Bodily Pain Scores as a Function of IL-6 and Chronic Medical Morbidity in Patients with Elevated Depressive Symptoms

Predictors B(SE) t F df R2 Adj R2
Model 1
IL-6 −46.40 (17.12) −2.72** 3.43* (3, 42) .20 .14
Model 2
Chronic medical morbidity −4.68 (1.10) −4.927***
IL-6 −25.16 (15.25) −1.65 8.17*** (4, 41) .44 .39
Open in a new tab

Models adjusted for body weight and age. Based on bootstrapping mediational analyses (26), the indirect effect of IL-6 on SF-36 Bodily Pain scores through chronic medical morbidity was not significant (95% CI: −48.73, 2.17).

*

p < .05

**

p < .01

***

p < .001

Next we examined whether chronic medical morbidity explained the association between pain and IL-6 among individuals with depressive symptoms by comparing analyses adjusted for and not adjusted for chronic condition count (25). When regressing pain onto IL-6 and chronic medical morbidity, the relationship between IL-6 and SF-36 Bodily Pain scores was reduced (Table 2, Model 2). To test the significance of the change in coefficients between these two models, recommended bootstrapping procedures were used (26). The method estimates standard errors and bias-corrected and accelerated 95% confidence intervals of the effect of IL-6 on Bodily Pain scores through chronic medical morbidity. These analyses did not support that chronic medical morbidity explains the relationship between IL-6 and SF-36 Bodily Pain scores (95% CI: −48.72, 2.17).

Finally, in a logistic regression model of depressive symptomatology controlling for body weight (χ2(2) = 15.37, p = .002), bodily pain (OR = 0.57, CI (95%) = 0.34 – 0.95) remained associated with elevated depressive symptoms after adjustment for the chronic medical morbidity index (OR = 1.24, CI(95%) = .75 – 2.07).

Discussion

The current study explored the potential role of inflammation in links between primary care patients’ comorbid depression, pain, and medical morbidity, three of the most common concerns in primary care practice. We found that depressive symptoms moderate the relationships between inflammation and pain, and inflammation and medical morbidity. Specifically, among patients with elevated depressive symptoms, IL-6 was associated with greater pain reports and chronic medical morbidity. In contrast, those relationships were not found among patients without clinically significant depressive symptoms. Subsequent analyses suggested that chronic medical morbidity did not fully (or statistically significantly) explain the association between IL-6 and pain in patients with elevated depressive symptoms. These findings suggest that depressive symptoms, whether alone or in combination with chronic disease burden, may increase vulnerability to pain and inflammation in primary care patients. Notably, depression severity and pain severity remained associated after adjustment for chronic medical morbidity, suggesting independent links between depression and pain which is consistent with the extensive literature pointing to reciprocal relationships between depression and pain.

As is common among low-income middle-aged and older adults in medical settings (27), nearly half of all study patients reported elevated depressive symptoms. Although IL-6 levels were slightly higher in primary care patients with clinically significant levels of depressive symptoms compared to those without, this difference did not reach statistical significance in our study. Given that the correlation between patients’ IL-6 levels and CES-D scores (~.20, NS) are consistent with effect sizes reported from meta-analytic studies (.2 – .3) (e.g., (3)), we suspect our null findings here may be due to a small sample size.

As mentioned, the cross-sectional data limits us to describing associations only, questions of causality cannot be evaluated, and treatment recommendations must be considered tentative. Nevertheless, clinical epidemiologic work relies on first establishing associations, which we demonstrated. Further, the patients were primarily low-income, middle-aged and older adults from a single primary care practice, and more than half were Black or African American. Whether the associations observed here generalize to non-urban primary care clinics or pain clinics remains to be evaluated; however, the examination of a patient population often under-represented in research and at high risk for the concerns studied is also strength. Pain was measured by a two-item scale, rather than a more comprehensive assessment, and may not have reflected chronic pain due to the 4-week time frame assessed. We measured IL-6 using a single blood draw. Multiple assessments of IL-6 throughout the day on multiple days would allow for analysis of correlated change among IL-6, depressive symptoms and pain, which would allow for examination of causality and improved reliability. Yet our work suggests that IL-6 levels are relatively stable even over periods as long as 34 weeks (28). We were unable to standardize the time in which the blood draws occurred. Given that we found significant associations despite the introduction of random error related to time of administration, we feel confident that the identified associations are real and not due to error. Patients’ medication use (including use of anti-inflammatories) was unavailable. Also, we had to rely on body weight to serve as a proxy for body mass index, which is known to relate to inflammatory cytokine levels. Both can contribute to inflammatory cytokine levels. Study participants represented regular patients at an attentive clinic, so it is reasonable to believe their chronic conditions were being treated appropriately. We thus expect the chronic condition count to reflect medication load to some degree. Finally, we relied on patients’ reports of whether a doctor had ever diagnosed them with chronic medical conditions. Data suggest this is a reasonably accurate technique, and medical records may sometimes be inaccurate (29). Nonetheless, future work using medical records and clinical laboratory data is needed.

Despite limitations, these findings suggest important clinical implications. Understanding these links helps lay the foundation for comprehensive assessment and cohesive, personalized treatment of primary care patients with pain, chronic medical morbidities, and depression. Primary care patients with comorbid pain, chronic medical morbidity and depression may have increased likelihood of elevated inflammation and associated health concerns. Thus, it is particularly critical to assess and treat depression in patients with pain and co-occurring chronic medical morbidities. Indeed, recent evidence suggests that depression treatment may reduce systematic inflammation (30). It is also possible that by reducing inflammation, depression treatment might serve as a pathway to reducing pain associated with chronic medical morbidities. Alternatively, targeting pain in patients with depression and greater chronic medical burden may reduce associated inflammation that can exacerbate depressive symptoms. Intervention trials that target comorbid depression and pain are needed to test effects of such targeted treatment on the management of chronic disease and inflammation.

Conclusions

The observed cross-sectional associations in our study suggest that future work should determine whether pain, chronic medical morbidities, and depression together synergistically increase inflammation, and if treatment of depression can improve inflammation and pain. Understanding these pathways would facilitate navigation of the development of comprehensive yet focused treatments, thus reducing suffering and decreasing risk for life-threatening inflammation-related conditions.

Acknowledgments

This study was funded in part by NIMH K23MH079347, K24MH072712, K08AG031328, R24AG031089 and R21AG023956.

We gratefully acknowledge the valuable contributions of Kelly Bellenger, BS, Linda Chaudron, MD, Catherine Cerulli, JD, PhD, and Stephanie Gamble, PhD.

Footnotes

The authors do not have financial agreements to disclose.

Conflicts of Interest

None of the authors have any conflicting or competing interests to report.

References

  • 1.Bair MJ, Robinson RL, Katon W, Kroenke K. Depression and pain comorbidity - A literature review. Archives of Internal Medicine. 2003;163(20):2433–45. doi: 10.1001/archinte.163.20.2433. [DOI] [PubMed] [Google Scholar]
  • 2.Detweiler-Bedell JB, Friedman MA, Leventhal H, Miller IW, Leventhal EA. Integrating co-morbid depression and chronic physical disease management: identifying and resolving failures in self-regulation. Clin Psychol Rev. 2008 Dec;28(8):1426–46. doi: 10.1016/j.cpr.2008.09.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Howren MB, Lamkin DM, Suls J. Associations of depression with C-reactive protein, IL-1, and IL-6: a meta-analysis. Psychosom Med. 2009 Feb;71(2):171–86. doi: 10.1097/PSY.0b013e3181907c1b. [DOI] [PubMed] [Google Scholar]
  • 4.Dowlati Y, Herrmann N, Swardfager W, Liu H, Sham L, Reim EK, et al. A meta-analysis of cytokines in major depression. Biol Psychiatry. 2010 Mar 1;67(5):446–57. doi: 10.1016/j.biopsych.2009.09.033. [DOI] [PubMed] [Google Scholar]
  • 5.Miller AH, Maletic V, Raison CL. Inflammation and its discontents: the role of cytokines in the pathophysiology of major depression. Biol Psychiatry. 2009 May 1;65(9):732–41. doi: 10.1016/j.biopsych.2008.11.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Dantzer R, O’Connor JC, Freund GG, Johnson RW, Kelley KW. From inflammation to sickness and depression: when the immune system subjugates the brain. Nat Rev Neurosci. 2008 Jan;9(1):46–56. doi: 10.1038/nrn2297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Solovieva S, Leino-Arjas P, Saarela J, Luoma K, Raininko R, Riihimaki H. Possible association of interleukin 1 gene locus polymorphisms with low back pain. Pain. 2004 May;109(1–2):8–19. doi: 10.1016/j.pain.2003.10.020. [DOI] [PubMed] [Google Scholar]
  • 8.Marchand F, Perretti M, McMahon SB. Role of the immune system in chronic pain. Nat Rev Neurosci. 2005 Jul;6(7):521–32. doi: 10.1038/nrn1700. [DOI] [PubMed] [Google Scholar]
  • 9.Hori TOT, Hosoi M, Aou S. Pain modulatory actions of cytokines and prostaglandin E-2 in the brain. Neuroimmunomodulation. 1998;840:269–81. doi: 10.1111/j.1749-6632.1998.tb09567.x. [DOI] [PubMed] [Google Scholar]
  • 10.DeLeo JA, Yezierski RP. The role of neuroinflammation and neuroimmune activation in persistent pain. Pain. 2001 Feb 1;90(1–2):1–6. doi: 10.1016/s0304-3959(00)00490-5. [DOI] [PubMed] [Google Scholar]
  • 11.Watkins LR, Maier SF. The pain of being sick: implications of immune-to-brain communication for understanding pain. Annu Rev Psychol. 2000;51:29–57. doi: 10.1146/annurev.psych.51.1.29. [DOI] [PubMed] [Google Scholar]
  • 12.Watkins LR, Maier SF. Beyond neurons: evidence that immune and glial cells contribute to pathological pain states. Physiol Rev. 2002 Oct;82(4):981–1011. doi: 10.1152/physrev.00011.2002. [DOI] [PubMed] [Google Scholar]
  • 13.Starkweather A, Witek-Janusek L, Mathews HL. Neural-immune interactions: implications for pain management in patients with low-back pain and sciatica. Biol Res Nurs. 2005 Jan;6(3):196–206. doi: 10.1177/1099800404272221. [DOI] [PubMed] [Google Scholar]
  • 14.Sommer C, Kress M. Recent findings on how proinflammatory cytokines cause pain: peripheral mechanisms in inflammatory and neuropathic hyperalgesia. Neurosci Lett. 2004 May 6;361(1–3):184–7. doi: 10.1016/j.neulet.2003.12.007. [DOI] [PubMed] [Google Scholar]
  • 15.De Jongh RFVK, Meert TF, Booij LHDJ, De Deyne CS, Heylen RJ. The role of interleukin-6 in nociception and pain. Anesthesia and Analgesia. 2003;96:1096–103. doi: 10.1213/01.ANE.0000055362.56604.78. [DOI] [PubMed] [Google Scholar]
  • 16.Barton BE. Interleukin-6 and new strategies for the treatment of cancer, hyperproliferative diseases and paraneoplastic syndromes. Expert Opin Ther Targets. 2005 Aug;9(4):737–52. doi: 10.1517/14728222.9.4.737. [DOI] [PubMed] [Google Scholar]
  • 17.Smolen JS, Maini RN. Interleukin-6: a new therapeutic target. Arthritis Res Ther. 2006;8( Suppl 2):S5. doi: 10.1186/ar1969. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Chapman BP, Khan A, Harper M, Stockman D, Fiscella K, Walton J, et al. Gender, race/ethnicity, personality, and interleukin-6 in urban primary care patients. Brain Behav Immun. 2009 Jul;23(5):636–42. doi: 10.1016/j.bbi.2008.12.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Eaton WWSC, Ybarra M, Muntaner C, Tien A. In: Center for Epidemiologic Studies Depression Scale: Review and revision (CESD and CESD-R) MM, editor. Princeton, NJ: Lawrence Erlbaum Associates; 2004. [Google Scholar]
  • 20.Ware JE, Sherbourne CD. The MOS 36-Item Short-Form Health Survey (SF-36): I. Conceptual framework and item selection. Medical Care. 1992;30:473–83. [PubMed] [Google Scholar]
  • 21.McHorney CA, Ware JE, Raczek AE. The MOS 36-item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Medical Care. 1993;31:247–63. doi: 10.1097/00005650-199303000-00006. [DOI] [PubMed] [Google Scholar]
  • 22.Brim OG, Ryff CD, Kessler RC, editors. How healthy are we?: a national study of well-being at midlife. Chicago, IL: The University of Chicago Press; 2004. [Google Scholar]
  • 23.Fortin M, Bravo G, Hudon C, Vanasse A, Lapointe L. Prevalence of multimorbidity among adults seen in family practice. Ann Fam Med. 2005 May-Jun;3(3):223–8. doi: 10.1370/afm.272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Schulberg HC, Saul M, McClelland M, Ganguli M, Christy W, Frank R. Assessing Depression in Primary Medical and Psychiatric Practices. Archives of General Psychiatry. 1985;42(12):1164–70. doi: 10.1001/archpsyc.1985.01790350038008. [DOI] [PubMed] [Google Scholar]
  • 25.Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986 Dec;51(6):1173–82. doi: 10.1037//0022-3514.51.6.1173. [DOI] [PubMed] [Google Scholar]
  • 26.Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods. 2008 Aug;40(3):879–91. doi: 10.3758/brm.40.3.879. [DOI] [PubMed] [Google Scholar]
  • 27.Lyness JM, Caine ED, King DA, Cox C, Yoediono Z. Psychiatric disorders in older primary care patients. J Gen Intern Med. 1999 Apr;14(4):249–54. doi: 10.1046/j.1525-1497.1999.00326.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Chapman BP, van Wijngaarden E, Seplaki CL, Talbot N, Duberstein P, Moynihan J. Openness and conscientiousness predict 34-week patterns of Interleukin-6 in older persons. Brain Behav Immun. 2011 May;25(4):667–73. doi: 10.1016/j.bbi.2011.01.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kriegsman DM, Penninx BW, van Eijk JT, Boeke AJ, Deeg DJ. Self-reports and general practitioner information on the presence of chronic diseases in community dwelling elderly. A study on the accuracy of patients’ self-reports and on determinants of inaccuracy. J Clin Epidemiol. 1996 Dec;49(12):1407–17. doi: 10.1016/s0895-4356(96)00274-0. [DOI] [PubMed] [Google Scholar]
  • 30.Thornton LM, Andersen BL, Schuler TA, Carson WE., 3rd A psychological intervention reduces inflammatory markers by alleviating depressive symptoms: secondary analysis of a randomized controlled trial. Psychosom Med. 2009 Sep;71(7):715–24. doi: 10.1097/PSY.0b013e3181b0545c. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES