Increased Risk of Depression in Patients with Parkinson’s Disease: A Nationwide Cohort Study
Yi-Ting Hsu, M.D., Chien-Chang Liao, Ph.D., Shih-Ni Chang, M.P.H., Yu-Wan Yang, M.D., Chon-Haw Tsai, M.D., Ta-Liang Chen, M.D., Fung-Chang Sung, Ph.D.
Neuroscience Laboratory, Department of Neurology, China Medical University Hospital, Taichung, Taiwan (YTH, YWY, CHT); School of Medicine, China Medical University, Taichung, Taiwan (YTH, YWY, CHT); Department of Anesthesiology, Taipei Medical University Hospital, Taipei, Taiwan (CCL, TLC); School of Chinese Medicine, China Medical University, Taichung, Taiwan; Management Office for Health Data, China Medical University Hospital, Taichung, Taiwan (CCL, SNC, FCS); Department of Public Health, China Medical University, Taichung, Taiwan (FCS)
CCL has equal contribution with first author; TLC has equal contribution with corresponding author.
Corresponding author / address reprint requests to: Fung-Chang Sung, PhD, MPH
Professor
China Medical University College of Public Health 91 Hsueh Shih Road
Taichung 404, Taiwan Tel: 886-4-2206-2295 Fax: 886-4-2201-9901
E-mail: [email protected]
Word count: 37843457 (including title page, references, legends and structured abstract) Running title: Depression and Parkinson’s disease
Funding: The study was supported by grants from the Executive Yuan National Science Council (NSC 99-2314-B-039-016-MY2), China Medical University Hospital (DMR-99-176), and the Taiwan Department of Health Clinical Trial and Research Center of Excellence (DOH102-TD-B-111-004)
Relevant conflicts of interest/financial disclosures: None.
Ethical approval: Insurance reimbursement claims from the National Health Insurance Research Database are available for public access. To protect personal privacy, the electronic database with patient identifications was decoded and scrambled for further research access. Although informed consent was not required because of this privacy protection, the study was evaluated and approved by the National Health Research Institutes.
Abstract
Objective: The association between Parkinson’s disease and depression remains unclear, particularly in Asian population. The purpose of this study is to investigate the risk of depression in patients with Parkinson’s disease using population-based data.
Methods: Based on the National Health Insurance Research Database of Taiwan, we identified 1698 patients with Parkinson’s disease aged 40 years or older diagnosed in 2000-2003. With frequency matching procedure, we randomly selected 6792 subjects without Parkinson’s disease stratified by sex and age. Both cohorts were followed-up until the end of 2008 or diagnosis of depression. Risk of depression associated with Parkinson’s disease was estimated in the multivariate Cox hazards regressions. Parkinson’s disease patients were more prevalent with diabetes, hypertension, and hyperlipidemia at the baseline.
Results: Compared to cohort without Parkinson’s disease, the hazard ratio (HR) for depression in Parkinson’s disease patients was 4.06 (95% confidence interval [CI]: 3.15-5.23), which increased to 4.267 (95% CI: 3.29-5.513) after adjustment for age, sex, occupation, urbanization, living region, income, and coexisting medical conditionsdiabetes, hypertension, and hyperlipidemia. In the sex-stratification, the HR of depression for men with Parkinson’s disease was 4.42 (95% CI: 2.93-6.678) compared with men without Parkinson’s disease. The HR for the association between Parkinson’s disease and depression in women was 4.221 (95% CI: 3.02-5.883.02-5.87).
of depression, particularly for men. Integrated care for early identification and treatment of depression are crucial for patients with Parkinson’s disease.
INTRODUCTION
Parkinson’s disease (PD) is a neurodegenerative disorder characterized by motor symptoms of tremor, bradykinesia, rigidity, and postural instability. Non-motor symptoms such as psychiatric, cognitive and autonomic symptoms, and depression and sleep disorders are often underestimated, but have major impact on PD patients. The prevalence of depression in patients with PD varies among populations, from 2.7% to more than 90%, possibly due to different methodologies and study populations (1). A comprehensive review of the 26 major studies concerning PD and depression indicates that the mean prevalence of depression is 40% among PD patients (2). Depression is known to have major impact on the quality of life, and economic burden for PD patients, and is a risk factor of impaired cognitive functions (3-5).
The relationship between depression and PD is an area of concern. Three large register studies have concluded that depressed patients are at higher risk of PD (6-8). On the other hand, several cross-sectional studies have reported elevated prevalence of depression in patients with PD (9,10). The previous studies were either restricted to small populations or cross-sectional designs, or lack proper adjustments for the potential confounding factors for depression (9-11). Few have focused on the longitudinal relationship between PD and subsequent risk of depression.
without PD in a nationwide retrospective cohort study using Taiwan’s National Health Insurance Research Database.
METHODS Data sources
This study used data from reimbursement claims of a universal National Health Insurance program, which was launched in March 1995 by incorporating 13 insurance programs in Taiwan. This insurance program provided health care to 99% of the population (12). The National Health Research Institutes (NHRI) had the responsibility for managing the insurance data. We obtained from NHRI the claims data of 1 million randomly selected subjects from 23 million insured individuals registered from 1996 to 2008, with the scrambled identifications of insured individuals and contracted heath care institutions. The claims data included the registry of medical facilities, details of inpatient orders, ambulatory care, as well as dental services and prescriptions. Individuals’ identifiers in this dataset have been encrypted to protect the privacy and confidentiality of enrollee. Diagnoses were coded using the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM). Ethical approval
This study was conducted in accordance with the Helsinki Declaration. To protect personal privacy, the electronic database was decoded with patient identifications scrambled for further public access for research. According to NHRI regulations, informed consent is not required due to decoded and scrambled patient identification (13,14).
Study sample
(ICD-9 code 332) from 2000 to 2003. In order to ensure the validity of the diagnosis, only new patients with at least one visit with neurologist’s diagnoses of PD after the index date were eligible. For each PD case, four sex and age-matched and depression-free subjects without PD history were randomly selected for comparison as the non-PD cohort. Both cohorts were followed-up until December 31, 2008, or upon the diagnosis of depression (ICD-9 296.2 and 296.3) or censored for loss to follow-up or death. The National Health Insurance Research Database has been verified as a valid resource for population-based research (15).
In Taiwan, general physicians usually adopted the Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV criteria or geriatric depression scale for diagnosing depression in daily medical practice. In the absence of a comprehensive clinical diagnostic interview, most neurologists or psychiatrists have been trained to use DSM-IV for establishing depressive diagnoses according to standardized criteria. Based on the National Statistics of Regional Standard Classification, all subjects were grouped into four geographic areas (North, Central, South and East and off-shore Islands) and three urbanization levels (low, moderate, and high). There are 359 townships and city districts in Taiwan. We calculated the population density (persons/km2) by dividing the population (persons) by the area (km2) for each of these
administrative units according to the statistics from the Ministry of Interior, Taiwan. The first, second, and third tertiles were considered areas of low, moderate and high urbanization, respectively (13,14).
Statistical analysis
Baseline distribution of age, sex, occupation, urbanization, residential area, and income were compared between two cohorts. We used Chi-square tests to examine the categorical variables. Baseline co-morbidities (diabetes, hypertension and hyperlipidemia) between PD and non-PD cohorts were also examined by Chi-square test. Because this is a retrospective cohort study with the calculation of person-years during the follow-up period, we selected Cox proportional hazard model to estimate the risk of depression for PD and non-PD cohorts. We used unadjusted Cox proportional hazard model to calculate the incidence rate ratio of depression associated with PD by sex and age. The multivariate Cox proportional hazards regression analysis with backward selection was performed to evaluate the hazard ratio (HR) and 95% confidence intervals (CIs) of depression in PD patients compared with non-PD subjects after adjusted for age, sex, occupation, urbanization, lowregion, income, coexisting medical conditionsdiabetes, hypertension, and hyperlipidemia. In this study, the proportional hazards assumption was evaluated by inspecting residual and log-log plots.the test of goodness of fit for each model was performed by using global test. The cumulative depression-free rates for both cohorts were estimated using the Kaplan-Meier method, and used the log-rank test to examine the difference. All analyses were performed using the SAS statistical software (version 9.1 for Windows; SAS Institute, Inc., Cary, NC, USA). A two-tailed p <0.05 was considered statistically significant.
RESULTS
Table 1 shows demographic characteristics and co-morbidities between PD and non-PD cohorts. Distributions in age and sex were similar between two cohorts. The PD patients had lower income and employment rates, and tended to live in southern Taiwan. The PD patients were also more prevalent than non-PD cohort with comorbidities of diabetes, hypertension and hyperlipidemia.
Table 2 shows the analysis of the incidence of depression in both cohorts developed in the follow-up period. The incidence in the PD cohort was 3.98 times greater than that in the non-PD cohort (10.3 vs. 2.59 per 1000 person-years). Men of the non-non-PD cohort had the lowest incidence of depression (2.02 per 1000 person-years). The sex-specific PD to non-PD incidence rate ratio of depression was greater in women than in men.
The overall unadjusted HR of depression for PD cohort compared with non-PD cohort was 4.06 (95% CI: 3.15-5.23, df=1, chi-square=117.4, p<0.0001) in the Model 1 (Table 3, model 1). After adjustments for age, sex, occupation, urbanization, and income, Mmodel 2 shows an adjusted HR of 4.02 (95% CI: 3.12-5.19, df=1, chi-square=115.2, p<0.0001). The HR increased to 4.27 (95% CI: 3.29-5.53, df=1, chi-square=120.6, p<0.0001) after further adjusting for coexisting medical conditions in Model 3.
The Kaplan-Meier analysis showed that depression-free rate was 6% lower in the PD cohort than in non-PD cohort, and was significant during the follow-up period (p <0.0001; log-rank test) (Figure 1).
DISCUSSION
This nationwide retrospective cohort study shows that PD patients are at an elevated risk of depression and the risk remains significant after adjustments for sociodemographic factors and related coexisting diseases. This association seems similar in women and men.
Lower socioeconomic status has been considered as a risk factor for chronic disease associated with worse health conditions, compared to higher socio-economic status (16). Consistent with previous studies, we found that people with lower socio-economic status are at higher risk of PD than those with higher socio-economic status. From the public health perspective, PD, a chronic progressive and degenerative neurologic condition, imposes significant economic burden on both patients and public services.
Metabolic syndrome is characterized by insulin resistance or overt diabetes, central obesity (large waist circumference), high blood pressure, and dyslipidemia (17). Cross-sectional studies consistently report positive associations between depression and metabolic syndrome (18,19). Diabetes mellitus, hypertension, and hyperlipidemia are significant factors associated with PD, which may confound the estimated risk for depression (Table 1). However, the whole magnitude of the observed association between PD and risk of depression is also sustained in a multivariate model. Another interesting issue is the causal relationship between metabolic syndrome and PD. A large prospective study among participants in two large cohorts: the Nurses’ Health Study and the Health Professionals Follow-up Study, also
suggests that high total cholesterol is associated with an increased risk of PD (20). In contrast, another prospective cohort study suggests that PD risk is not significantly related to the history of hypertension, hypercholesterolemia, or diabetes but may modestly decline with increasing blood cholesterol levels (21). There still exist some arguments and needs further investigation.
Previous researchers have found the risk of depression in patients with PD is inconsistently related to the sex (8,11). Studies failed to find gender difference in depression for PD patients because of the limitation of small sample sizes (22,23). The multi-center Priamo study assessing non-motor symptoms and their impact on the quality of life for PD patients found a higher prevalence of depression in women (4). Recently, a population-based UK study has concluded that increased relative risk of depression in PD patients is most pronounced in women and in individuals aged 40-69 years (11). However, our study showed a similar risk of depression in women and men with PD. Female preponderance in depression rates are related to genetic, hormonal, psychological, and socio-cultural factors (23,24). However, our interpretation of the relationship between gender and depression in patients with PD may be limited. Because some patients may have treated depression, in which case they may not have a sufficiently high score on a depression scale. Thus these depressive patients with treatment may be included in the non-depressed group despite having, perhaps, a recurrent depressive disorder.
Studies have investigated extensively the onset age of PD and the disease duration in association with depression for PD patients (25). In the present study, the age-specific of both absolute risk and relative risk of depression for PD patients decreased with age and becomes more pronounced in those aged 40-49 years. And the Kaplan-Meier survival curve explicitly shows that the risk of depression increases with prolonged PD duration. PD may develop over several years before its clinical diagnosis and the diagnosis dates used in the current analysis are dates of the first dates reported by the physician. However, depression almost inevitably emerged with disease progression and might dominate the clinical picture of advanced PD which contributed to severe disability and impaired quality of life. We should pay more attention on the recognition and quantity of the important non-motor symptom, depression, and give available therapy.
The current study supports the close association between PD and subsequent depression, with the risk of depression in PD patients is at least three-fold greater than that in non-PD individuals. The reason of having higher depression risk in PD is poorly understood and the relationship is complex. There have been some debates about depression and PD. Depression is either a reactive process to the chronic, disabling symptoms of PD or is associated with brain pathological changes of PD. These two hypotheses are not mutually exclusive. PD patients and their families clearly must make adjustments to manage this chronic disease because of the physical disability (26). Medication such as levodopa have both
mood-improving and mood-lowering effects (26). In patients with more advanced disease, depression may be associated with “off-drug” fluctuations (27). However, it does not fully explain the high prevalence of depression. Sharing similar biological mechanism in PD and depression has thus been discussed. Two hypotheses about the patho-physiology between PD and depression have also been suggested. Mayeux et al proposed the serotonergic hypotheses and another dopaminergic hypotheses that considers degeneration of the mesolimbic and mesocortical structures of the dopaminergic system as the cause of depression (28-31). Some studies revealed that depression occurs preceding the diagnosis of PD and proposed that depression and PD share a common biological factor (6-8). The cause of high risk in PD patients who are later diagnosed with depression may be of multi-factorial etiology. We believe that the development of depression is related to the neuroendocrine effects of the stress associated with living with PD (32). When people suffered from PD, they may develop depressive symptoms because of loss of job, limited physical function, unable participating family activity and poor quality of life (33). This is different from patients who have depressive disorders before the onset of PD. Another explanation for the association between PD and depression is that patients with PD have received more medical attention than non-PD people in the clinical settings. Because early detection of depressive symptoms by clinical workers in patients with PD may occur, diagnostic bias may explain partly the observed increased risk (11).
The present study has some limitations. First, this study used medical claims to identify diseases, the underestimation of comorbidities such as hypertension, diabetes, hyperlipidemia, depression, and PD should be considered, because some people with mild condition may not visit medical services. Second, diagnostic bias may also exist because more medical attention may be given to depressive symptoms in subjects with PD than those without. Third, the ICD-9 296.2 and 296.3 codes do not classify depression severity and there is no accurate differentiation of depression severity in PD patients by the current ICD-9 code. However, the large population sample of longitudinal design is the strength of the present study, which aims to evaluate the relationship between depression and PD. Correcting for general risk factors like age, sex, socioeconomic status, occupation, and metabolic syndrome for depression highlights the importance of PD as a risk factor. In addition, although we included PD patients required at least 3 medical visits with primary diagnosis, it remains possible that PD was diagnosed in some patients with drug-induced parkinsonism caused by neuroleptics, which are known to contribute to the metabolic syndrome. Finally, this study is limited to retrospective reimbursement claims without individual detailed biochemical data.
In conclusion, there is a positive association between PD and subsequent risk for developing depression. This stresses the importance of being aware of depression in PD patients, especially among the younger and men. From a public health perspective, it is important to understand the strong association of this complication to enable clinicians and
Acknowledgement
This study is based (in part) on data from the National Health Insurance Research Database provided by the National Health Research Institutes. The interpretation and conclusions contained herein do not represent those of the National Health Research Institutes. This study is supported in part by Executive Yuan National Science Council (NSC 99-2314-B-039-016-MY2), China Medical University Hospital (DMR-99-176), and the Taiwan Department of Health Clinical Trial and Research Center of Excellence (DOH102-TD-B-111-004). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the paper.
Authors' Roles
Manuscript preparation: YTH, CCL. Data analysis: CCL, SNC. Interpretation of data: YTH, CCL, SNC, YWY, CHT, TLC, FCS. Revising manuscript: YWY, CHT, TLC, FCS. Final approval of submission: YTH, CCL, SNC, YWY, CHT, TLC, FCS.
REFERENCES
1. Reijnders JS, Ehrt U, Weber WE, et al: A systematic review of prevalence studies of depression in Parkinson's disease. Mov Disord 2008; 23:183-189
2. Cummings JL: Depression and Parkinson's disease: a review. Am J Psychiatry 1992; 149:443-454
3. Findley L, Aujla M, Bain PG, et al: Direct economic impact of Parkinson's disease: a research survey in the United Kingdom. Mov Disord 2003; 18:1139-1145
4. Barone P, Antonini A, Colosimo C, et al: The PRIAMO study: A multicenter assessment of nonmotor symptoms and their impact on quality of life in Parkinson's disease. Mov Disord 2009; 24:1641-1649
5. Marder K, Tang MX, Cote L, et al: The frequency and associated risk factors for dementia in patients with Parkinson's disease. Arch Neurol 1995; 52:695-701
6. Nilsson FM, Kessing LV, Bolwig TG: Increased risk of developing Parkinson's disease for patients with major affective disorder: a register study. Acta Psychiatr Scand 2001; 104:380-386
7. Schuurman AG, van den Akker M, Ensinck KT, et al: Increased risk of Parkinson's disease after depression: a retrospective cohort study. Neurology 2002; 58:1501-1504 8. Leentjens AF, Van den Akker M, Metsemakers JF, et al: Higher incidence of depression
preceding the onset of Parkinson's disease: a register study. Mov Disord 2003; 18:414-418
9. Aarsland D, Bronnick K, Alves G, et al: The spectrum of neuropsychiatric symptoms in patients with early untreated Parkinson's disease. J Neurol Neurosurg Psychiatry 2009; 80:928-930
10. Kulisevsky J, Pagonabarraga J, Pascual-Sedano B, et al: Prevalence and correlates of neuropsychiatric symptoms in Parkinson's disease without dementia. Mov Disord 2008; 23:1889-1896
11. Becker C, Brobert GP, Johansson S, et al: Risk of incident depression in patients with Parkinson disease in the UK. Eur J Neurol 2011; 18:448-453
12. Cheng TM: Taiwan's new national health insurance program: genesis and experience so far. Health Aff 2003; 22:61-76
13. Liao CC, Shen WW, Chang CC, et al: Surgical adverse outcomes in patients with schizophrenia: a population-based study. Ann Surg 2013; 257:433-438
14. Liao CC, Chiu WT, Yeh CC, et al: Risk and outcomes for traumatic brain injury in patients with mental disorders. J Neurol Neurosurg Psychiatry 2012; 83:1186-1192
15. Cheng CL, Kao YH, Lin SJ, et al: Validation of the National Health Insurance Research Database with ischemic stroke cases in Taiwan. Pharmacoepidemiol Drug Saf 2011; 20:236-242
16. Mackenbach JP, Stirbu I, Roskam AJ, et al: Socioeconomic inequalities in health in 22 European countries. N Engl J Med 2008; 358:2468-2481
health inequalities. J Gen Intern Med 1995; 10:577-586
18. Koponen H, Jokelainen J, Keinanen-Kiukaanniemi S, et al: Metabolic syndrome predisposes to depressive symptoms: a population-based 7-year follow-up study. J Clin Psychiatry 2008; 69:178-182
19. Skilton MR, Moulin P, Terra JL, et al: Associations between anxiety, depression, and the metabolic syndrome. Biol Psychiatry 2007; 62:1251-1257
20. Hu G, Antikainen R, Jousilahti P, et al: Total cholesterol and the risk of Parkinson disease. Neurology 2008; 70:1972-1979
21. Simon KC, Chen H, Schwarzschild M, et al: Hypertension, hypercholesterolemia, diabetes, and risk of Parkinson disease. Neurology 2007; 69:1688-1695
22. Tandberg E, Larsen JP, Aarsland D, et al: The occurrence of depression in Parkinson's disease. A community-based study. Arch Neurol 1996; 53:175-179
23. Schrag A, Jahanshahi M, Quinn NP: What contributes to depression in Parkinson's disease? Psychol Med 2001; 31:65-73
24. Boorsma M, Joling K, Dussel M, et al: The incidence of depression and its risk factors in Dutch nursing homes and residential care homes. Am J Geriatr Psychiatry 2012; 20:932-942
25. Aarsland D, Påhlhagen S, Ballard CG, et al: Depression in Parkinson disease: epidemiology, mechanisms and management. Nat Rev Neurol 2011; 8:35-47
depression in Parkinson's disease. Biol Psychiatry 2003; 54:363-375
27. Richard IH, Frank S, McDermott MP, et al: The ups and downs of Parkinson disease: a prospective study of mood and anxiety fluctuations. Cogn Behav Neurol 2004; 17:201-207
28. Mayeux R: The "serotonin hypothesis" for depression in Parkinson's disease. Adv Neurol 1990; 53:163-166
29. Fibiger HC: The neurobiological substrates of depression in Parkinson's disease: a hypothesis. Can J Neurol Sci 1984; 11:105-107
30. Frisina PG, Haroutunian V, Libow LS: The neuropathological basis for depression in Parkinson's disease. Parkinsonism Relat Disord 2009; 15:144-148
31. Remy P, Doder M, Lees A, et al: Depression in Parkinson's disease: loss of dopamine and noradrenaline innervation in the limbic system. Brain 2005; 128:1314-1322
32. Willis GL: Parkinson's disease as a neuroendocrine disorder of circadian function: dopamine-melatonin imbalance and the visual system in the genesis and progression of the degenerative process. Rev Neurosci 2008; 19:245-316
33. Rod NH, Bordelon Y, Thompson A, Marcotte E, Ritz B: Major life events and development of major depression in Parkinson's disease patients. Eur J Neurol 2013; 20:663-670
Figure legends
FIGURE 1. The depression-free proportions estimated for the people with and without Parkinson’s diseases using the Kaplan-Meier method (log rank, df=1, chi-square=18.17, p <0.0001)
