536086 research-article2015
MSJ0010.1177/1352458514536086Multiple Sclerosis JournalMW Koch, S Patten
MULTIPLE SCLEROSIS MSJ JOURNAL
Research Paper
Depression in multiple sclerosis: A long-term longitudinal study Marcus W Koch, Scott Patten, Sandy Berzins, Simon Zhornitsky, Jamie Greenfield, Winona Wall and Luanne M Metz
Multiple Sclerosis Journal 2015, Vol. 21(1) 76–82 DOI: 10.1177/ 1352458514536086 © The Author(s), 2015. Reprints and permissions: http://www.sagepub.co.uk/ journalsPermissions.nav
Abstract Background: Depression is a common comorbidity in multiple sclerosis (MS), but little is known about its long-term prognosis. Depression in the general population is usually episodic with relatively shortlasting depressive episodes. In this study we investigate the long-term prognosis of depression in MS. Methods: Using data from a large longitudinal observational study and from the Calgary MS clinic database, we investigated changes in Center for Epidemiological Studies Depression Scale (CESD) scores in MS patients over four years of follow-up. We used logistic regression to investigate the association of the factors sex, age, disease duration, Expanded Disability Status Scale (EDSS), depression at baseline, and antidepressant use with depression at each year of follow-up. Results: CESD scores remained largely stable, or decreased slightly over four years of follow-up, whereas EDSS scores steadily increased. Depression at baseline was the strongest predictor of depression at follow-up; the other factors were not or not consistently associated with depression at follow-up. As expected, antidepressant use was associated with a greater risk of depression at follow-up. Starting and stopping antidepressant treatment during follow-up was not associated with the risk of depression at follow-up or with significant change in CESD scores. Conclusion: In contrast to depression in the general population, depression in MS is largely chronic, which suggests a different pathophysiology.
Keywords: Multiple sclerosis, depression Date received: 12 January 2014; revised: 22 April 2014; accepted: 24 April 2014
Introduction Depression is a common co-morbidity in patients with multiple sclerosis (MS).1 The lifetime prevalence of depression in MS patients is estimated around 50%,2,3 which is three times that of the general population. The one-year period prevalence of major depression in MS was found to be around 15% (double that of participants without MS) in an analysis of a Canada-wide health questionnaire study.4 The cause for the much greater prevalence of depression in patients with MS compared to the general population is unclear. Possible causes range from the psychological view that the depression may be a reaction to having a chronic debilitating disease, to the mechanistic view that depression could be an expression of underlying brain damage.
While the prevalence of depression in MS is well documented, much less is known about the prognosis of MS-related depression in the longer term. Depression outside the setting of MS is generally considered an episodic disease, with depressive episodes lasting for a certain time, after which patients recover, either fully or partially. For example, in the most widely used diagnostic systems of the Diagnostic and Statistical Manual of Mental Disorders, fourth (DSM-IV) and fifth editions (DSM-5), major depressive disorder is sub-classified either as “single episode” or “recurrent episode.” The duration of these major depressive episodes varies, but is usually measured in weeks or months rather than years.5,6 If an episode of depression lasts for longer than two years, the diagnosis becomes that of a persistent depressive disorder.7 A pivotal study on the duration of major
Correspondence to: Marcus W. Koch Division of Neurology, Department of Community Health Sciences, University of Calgary, Heritage Medical Research Building, 3330 Hospital Drive NW, Calgary, Alberta, T3A 0B1, Canada. [email protected] Marcus W Koch Simon Zhornitsky Jamie Greenfield Winona Wall Luanne M Metz Department of Clinical Neurosciences, Division of Neurology, University of Calgary, Alberta, Canada Sandy Berzins Marcus W Koch Department of Community Health Sciences, University of Calgary, Alberta, Canada Scott Patten Department of Community Health Sciences/Department of Psychiatry, University of Calgary, Alberta, Canada
76 http://msj.sagepub.com
Downloaded from msj.sagepub.com at UNIVERSITY OF WATERLOO on February 24, 2015
MW Koch, S Patten et al. depressive episodes showed that approximately 40% of patients recover from a major depressive episode within three months, 60% within six months, and 80% within 12 months.8,9 Most research on the duration of depression in MS comes from studies using depression scales, and it is important to realize that a high score on a depression scale is not identical to having a major depressive episode as diagnosed by a psychiatrist. Such studies can therefore give an impression of depressive symptoms at the time of the measurement, but do not fully compare to the entity of major depressive episode. Depressive symptoms in MS, however, appear to have a long duration: A study with 10-year follow-up of MS patients with depressive symptoms showed that depression scores between baseline and 10-year follow-up were not significantly different, and that two-thirds of patients depressed at baseline were depressed at 10 years.10 This suggests that depressive symptoms in patients with MS may be substantially different from those experienced by patients with idiopathic depression, with possibly important consequences for their prognosis and response to treatment. In this study, we use data from a large patient cohort study to investigate the duration of depressive symptoms in a large cohort of patients with MS during follow-up of up to four years. Methods Study participants Data for this study were collected in a large longitudinal study that enrolled patients from the Calgary MS clinic, and followed them up with yearly questionnaires for up to four years. The Calgary MS clinic is the main care center for patients with MS in the southern part of the province of Alberta in Canada. Data collection Questionnaire data for this study were collected from June 2002 to January 2010. Center for Epidemiological Studies Depression Scale (CESD)11 scores and information on antidepressant use were self-reported in the questionnaires. CESD scores were categorized using the standard definition of 16 points11–13 or higher to define clinically significant levels of depression. Expanded Disability Status Scale (EDSS)14 scores and disease course evaluated by an MS neurologist within six months of the questionnaire date, as well as information on sex, age and disease duration, were obtained from the Calgary MS clinic database. Follow-up data were assigned to full years of
follow-up if the questionnaire date fell within three months of a full year of follow-up (e.g. a questionnaire completed 1.1 years after baseline was assigned to one-year follow-up). Antidepressant use between baseline and follow-up was coded as “always” for patients using an antidepressant both at baseline and follow-up, “never” for patients using no antidepressant at either baseline or follow-up, “started” for patients starting antidepressant use between baseline and follow-up, and “stopped” for patients stopping antidepressant use between baseline and follow-up. All patients gave their written informed consent to the use of their data. The study was approved by the University of Calgary Research Ethics Board. Statistical analysis We examined the overall distribution of CESD and EDSS scores at baseline and follow-up with violin plots. We investigated the change in CESD and EDSS scores between baseline and follow-up with the paired-samples t-test (for CESD) and the pairedsamples Mann-Whitney test (for EDSS). We compared the sex distribution between depressed and nondepressed patients at baseline and follow-up with the chi square test. The correlation of baseline and follow-up CESD scores was investigated with Pearson’s correlation coefficient. We created binary logistic regression models with depression at follow-up as the dependent variable and sex, age, disease duration, disease course, antidepressant use, depression at baseline and EDSS scores as independent predictor variables. We explored the goodness-of-fit of all regression models with the Le Cessie and van Houwelingen goodness-of-fit test.15 Statistical significance was taken to be at the two-sided 0.05 level. All statistical analyses and graphs were produced with the R statistical software package for Windows Version 3.0.1.16 Results Patient characteristics at baseline and follow-up are given in Table 1. We included 1376 patients at baseline, 984 at one year of follow-up and 967, 457, and 258 at two, three, and four years of follow-up. Between 22% and 29% of patients were classified as depressed during the study. EDSS scores were missing in 52 (4%) patients at baseline, 295 (29%) patients at one year, and in 349 (36%), 168 (37%), and 131 (50%) patients at two, three, and four years of follow-up. CESD scores slightly decreased over time. This decrease reached statistical significance between baseline and one year and between baseline and four
http://msj.sagepub.com 77
Downloaded from msj.sagepub.com at UNIVERSITY OF WATERLOO on February 24, 2015
Multiple Sclerosis Journal 21(1) Table 1. Patient characteristics at baseline and follow-up.
n Depressed (n, %) CESD (mean, SD) EDSS (mean, SD) AD users (n, %) AD use during follow-up (n, %): Never Always Started Stopped Female (n, %) Age (mean, SD) Disease duration (mean, SD) Disease course: RRMS (n, %) SPMS (n, %) PPMS (n, %) Unknown (n, %)
Baseline
1 Year
pa
2 Years
pa
3 Years
pa
4 Years
pa
1376 403 (29) 12.1 (10.9) 3.4 (2.2) 324 (24) —
984 256 (27) 11.5 (10.6) 3.5 (2.1) 238 (24) —
— — 0.04b
MSJ0010.1177/1352458514536086Multiple Sclerosis JournalMW Koch, S Patten
MULTIPLE SCLEROSIS MSJ JOURNAL
Research Paper
Depression in multiple sclerosis: A long-term longitudinal study Marcus W Koch, Scott Patten, Sandy Berzins, Simon Zhornitsky, Jamie Greenfield, Winona Wall and Luanne M Metz
Multiple Sclerosis Journal 2015, Vol. 21(1) 76–82 DOI: 10.1177/ 1352458514536086 © The Author(s), 2015. Reprints and permissions: http://www.sagepub.co.uk/ journalsPermissions.nav
Abstract Background: Depression is a common comorbidity in multiple sclerosis (MS), but little is known about its long-term prognosis. Depression in the general population is usually episodic with relatively shortlasting depressive episodes. In this study we investigate the long-term prognosis of depression in MS. Methods: Using data from a large longitudinal observational study and from the Calgary MS clinic database, we investigated changes in Center for Epidemiological Studies Depression Scale (CESD) scores in MS patients over four years of follow-up. We used logistic regression to investigate the association of the factors sex, age, disease duration, Expanded Disability Status Scale (EDSS), depression at baseline, and antidepressant use with depression at each year of follow-up. Results: CESD scores remained largely stable, or decreased slightly over four years of follow-up, whereas EDSS scores steadily increased. Depression at baseline was the strongest predictor of depression at follow-up; the other factors were not or not consistently associated with depression at follow-up. As expected, antidepressant use was associated with a greater risk of depression at follow-up. Starting and stopping antidepressant treatment during follow-up was not associated with the risk of depression at follow-up or with significant change in CESD scores. Conclusion: In contrast to depression in the general population, depression in MS is largely chronic, which suggests a different pathophysiology.
Keywords: Multiple sclerosis, depression Date received: 12 January 2014; revised: 22 April 2014; accepted: 24 April 2014
Introduction Depression is a common co-morbidity in patients with multiple sclerosis (MS).1 The lifetime prevalence of depression in MS patients is estimated around 50%,2,3 which is three times that of the general population. The one-year period prevalence of major depression in MS was found to be around 15% (double that of participants without MS) in an analysis of a Canada-wide health questionnaire study.4 The cause for the much greater prevalence of depression in patients with MS compared to the general population is unclear. Possible causes range from the psychological view that the depression may be a reaction to having a chronic debilitating disease, to the mechanistic view that depression could be an expression of underlying brain damage.
While the prevalence of depression in MS is well documented, much less is known about the prognosis of MS-related depression in the longer term. Depression outside the setting of MS is generally considered an episodic disease, with depressive episodes lasting for a certain time, after which patients recover, either fully or partially. For example, in the most widely used diagnostic systems of the Diagnostic and Statistical Manual of Mental Disorders, fourth (DSM-IV) and fifth editions (DSM-5), major depressive disorder is sub-classified either as “single episode” or “recurrent episode.” The duration of these major depressive episodes varies, but is usually measured in weeks or months rather than years.5,6 If an episode of depression lasts for longer than two years, the diagnosis becomes that of a persistent depressive disorder.7 A pivotal study on the duration of major
Correspondence to: Marcus W. Koch Division of Neurology, Department of Community Health Sciences, University of Calgary, Heritage Medical Research Building, 3330 Hospital Drive NW, Calgary, Alberta, T3A 0B1, Canada. [email protected] Marcus W Koch Simon Zhornitsky Jamie Greenfield Winona Wall Luanne M Metz Department of Clinical Neurosciences, Division of Neurology, University of Calgary, Alberta, Canada Sandy Berzins Marcus W Koch Department of Community Health Sciences, University of Calgary, Alberta, Canada Scott Patten Department of Community Health Sciences/Department of Psychiatry, University of Calgary, Alberta, Canada
76 http://msj.sagepub.com
Downloaded from msj.sagepub.com at UNIVERSITY OF WATERLOO on February 24, 2015
MW Koch, S Patten et al. depressive episodes showed that approximately 40% of patients recover from a major depressive episode within three months, 60% within six months, and 80% within 12 months.8,9 Most research on the duration of depression in MS comes from studies using depression scales, and it is important to realize that a high score on a depression scale is not identical to having a major depressive episode as diagnosed by a psychiatrist. Such studies can therefore give an impression of depressive symptoms at the time of the measurement, but do not fully compare to the entity of major depressive episode. Depressive symptoms in MS, however, appear to have a long duration: A study with 10-year follow-up of MS patients with depressive symptoms showed that depression scores between baseline and 10-year follow-up were not significantly different, and that two-thirds of patients depressed at baseline were depressed at 10 years.10 This suggests that depressive symptoms in patients with MS may be substantially different from those experienced by patients with idiopathic depression, with possibly important consequences for their prognosis and response to treatment. In this study, we use data from a large patient cohort study to investigate the duration of depressive symptoms in a large cohort of patients with MS during follow-up of up to four years. Methods Study participants Data for this study were collected in a large longitudinal study that enrolled patients from the Calgary MS clinic, and followed them up with yearly questionnaires for up to four years. The Calgary MS clinic is the main care center for patients with MS in the southern part of the province of Alberta in Canada. Data collection Questionnaire data for this study were collected from June 2002 to January 2010. Center for Epidemiological Studies Depression Scale (CESD)11 scores and information on antidepressant use were self-reported in the questionnaires. CESD scores were categorized using the standard definition of 16 points11–13 or higher to define clinically significant levels of depression. Expanded Disability Status Scale (EDSS)14 scores and disease course evaluated by an MS neurologist within six months of the questionnaire date, as well as information on sex, age and disease duration, were obtained from the Calgary MS clinic database. Follow-up data were assigned to full years of
follow-up if the questionnaire date fell within three months of a full year of follow-up (e.g. a questionnaire completed 1.1 years after baseline was assigned to one-year follow-up). Antidepressant use between baseline and follow-up was coded as “always” for patients using an antidepressant both at baseline and follow-up, “never” for patients using no antidepressant at either baseline or follow-up, “started” for patients starting antidepressant use between baseline and follow-up, and “stopped” for patients stopping antidepressant use between baseline and follow-up. All patients gave their written informed consent to the use of their data. The study was approved by the University of Calgary Research Ethics Board. Statistical analysis We examined the overall distribution of CESD and EDSS scores at baseline and follow-up with violin plots. We investigated the change in CESD and EDSS scores between baseline and follow-up with the paired-samples t-test (for CESD) and the pairedsamples Mann-Whitney test (for EDSS). We compared the sex distribution between depressed and nondepressed patients at baseline and follow-up with the chi square test. The correlation of baseline and follow-up CESD scores was investigated with Pearson’s correlation coefficient. We created binary logistic regression models with depression at follow-up as the dependent variable and sex, age, disease duration, disease course, antidepressant use, depression at baseline and EDSS scores as independent predictor variables. We explored the goodness-of-fit of all regression models with the Le Cessie and van Houwelingen goodness-of-fit test.15 Statistical significance was taken to be at the two-sided 0.05 level. All statistical analyses and graphs were produced with the R statistical software package for Windows Version 3.0.1.16 Results Patient characteristics at baseline and follow-up are given in Table 1. We included 1376 patients at baseline, 984 at one year of follow-up and 967, 457, and 258 at two, three, and four years of follow-up. Between 22% and 29% of patients were classified as depressed during the study. EDSS scores were missing in 52 (4%) patients at baseline, 295 (29%) patients at one year, and in 349 (36%), 168 (37%), and 131 (50%) patients at two, three, and four years of follow-up. CESD scores slightly decreased over time. This decrease reached statistical significance between baseline and one year and between baseline and four
http://msj.sagepub.com 77
Downloaded from msj.sagepub.com at UNIVERSITY OF WATERLOO on February 24, 2015
Multiple Sclerosis Journal 21(1) Table 1. Patient characteristics at baseline and follow-up.
n Depressed (n, %) CESD (mean, SD) EDSS (mean, SD) AD users (n, %) AD use during follow-up (n, %): Never Always Started Stopped Female (n, %) Age (mean, SD) Disease duration (mean, SD) Disease course: RRMS (n, %) SPMS (n, %) PPMS (n, %) Unknown (n, %)
Baseline
1 Year
pa
2 Years
pa
3 Years
pa
4 Years
pa
1376 403 (29) 12.1 (10.9) 3.4 (2.2) 324 (24) —
984 256 (27) 11.5 (10.6) 3.5 (2.1) 238 (24) —
— — 0.04b
