Mental comorbidity and multiple sclerosis: validating administrative data to support population-based surveillance

Ruth Ann Marrie 0 1 John D Fisk Bo Nancy Yu 0 Stella Leung 0 Lawrence Elliott 0 Patricia Caetano 0 Sharon Warren Charity Evans Christina Wolfson Lawrence W Svenson Helen Tremlett James F Blanchard 0 Scott B Patten for the CIHR Team in the Epidemiology Impact of Comorbidity on Multiple Sclerosis 0 Department of Community Health Sciences, University of Manitoba , Winnipeg , Canada 1 Department of Internal Medicine, University of Manitoba , Winnipeg , Canada Background: While mental comorbidity is considered common in multiple sclerosis (MS), its impact is poorly defined; methods are needed to support studies of mental comorbidity. We validated and applied administrative case definitions for any mental comorbidities in MS. Methods: Using administrative health data we identified persons with MS and a matched general population cohort. Administrative case definitions for any mental comorbidity, any mood disorder, depression, anxiety, bipolar disorder and schizophrenia were developed and validated against medical records using a a kappa statistic (k). Using these definitions we estimated the prevalence of these comorbidities in the study populations. Results: Compared to medical records, administrative definitions showed moderate agreement for any mental comorbidity, mood disorders and depression (all k 0.49), fair agreement for anxiety (k = 0.23) and bipolar disorder (k = 0.30), and near perfect agreement for schizophrenia (k = 1.0). The age-standardized prevalence of all mental comorbidities was higher in the MS than in the general populations: depression (31.7% vs. 20.5%), anxiety (35.6% vs. 29.6%), and bipolar disorder (5.83% vs. 3.45%), except for schizophrenia (0.93% vs. 0.93%). Conclusions: Administrative data are a valid means of surveillance of mental comorbidity in MS. The prevalence of mental comorbidities, except schizophrenia, is increased in MS compared to the general population. - Background Although depression and anxiety are considered common in MS [1,2], population-based prevalence estimates for these conditions are rare. Even fewer prevalence estimates exist for bipolar disorder and schizophrenia in the MS population, and they vary widely [3,4]. The paucity of population-based studies of mental comorbidity may reflect the challenges of conducting such studies. However, such studies are needed given the impact of mental comorbidity in MS, including lower quality of life and reduced adherence to treatment [5,6]; and to minimize the biases from using clinic-based samples. Studies of mental comorbidity could potentially use one of several data sources including medical records review, self-report, interviews, or administrative data. Administrative data are population-based in publicly funded health systems such as Canada and are cost-effective and accessible [7]. Such data are useful for assessing the burden of disease at the population level, including health services use and costs [8]. Mental comorbidities can be assessed in clinical samples using structured diagnostic interviews such as the Composite International Diagnostic Interview (CIDI) although these are time consuming and depend heavily on recall of past episodes [9]. Administrative data have the advantage that they are recorded during an episode and need not be recalled later. Administrative data, however, are collected for health system management and are often inadequately validated [7,10]. Indeed, few published case definitions for mental comorbidity have been validated, and efforts to develop and validate case definitions for depression have identified poor concordance with the CIDI Short Form [11], and difficulties distinguishing depression from anxiety [12]. We aimed to validate administrative case definitions for several mental comorbidities in MS, and to describe their prevalence among persons with MS versus a matched cohort from the general population. We hypothesized that the prevalence of depression, anxiety, bipolar disorder and schizophrenia would be higher in the MS population than in the general population. Methods Administrative data We conducted this study in Manitoba, Canada, using anonymized administrative data obtained from Manitoba Health (MH) which provides health care services for more than 98% of Manitoba residents [13]. Since 1984, MH has maintained computerized records of health services claims, which can be linked using a unique personal health identification number (PHIN) identifying the person who received the service. Physician claims include the PHIN, service date, and three-digit International Classification of Disease (ICD)-9-CM code for one physician-assigned diagnosis. Hospitalization records include the PHIN, admission and discharge dates, and up to 16 discharge diagnoses. Before 2004, diagnoses were listed using five-digit ICD-9-CM codes and since 2004 they have been listed using ICD-10CA codes. Since 1996, the Drug Programs Information Network captures outpatient prescription drug dispensations including date, drug name, and drug identification number for Manitoba residents, regardless of payer. The population registry is updated when an individual migrates into or out of Manitoba, or dies. Study populations and validation cohort Using data from 1984 to 2006, we identified all Manitobans with MS using a previously validated administrative case definition [14]. We identified up to 5 controls from the general population for each MS case, matched on sex, year of birth and region of residence (postal code), and excluding anyone with an ICD9/10-code for any demyelinating disease as previously described [14]. As described in detail previously, the medical records of 430 persons with MS were reviewed by a trained abstractor using a standardized data collection form [14,15]. Using each participants PHIN, these clinical data were linked with the administrative databases. Administrative case definitions We aimed to develop case definitions for depression, anxiety, bipolar disorder and schizophrenia using established approaches [16]. Developing case definitions for mental comorbidity raised challenges. Although hospital claims provide 5-digit ICD codes, physician claims in Manitoba have only three digits, reducing the specificity of coding. For example, at the 3-digit level, the same code (296) describes bipolar I disorder, most recent episodic manic (296.4) and major depressive disorder recurrent episode (296.3). Therefore, we initially created an omnibus definition for mental comorbidity to capture persons with any of the mental comorbidities of interest, followed by an any mood or anxiety disorders definition which captured depressive disorders, anxiety disorders and bipolar disorder. Finally, we developed case definitions for individual mental comorbidities. We identified ICD-9/10 codes for the individual comorbidities and the combination definitions (Additional file 1: Table S1). While incorporating prescription claims might improve specificity of the case (...truncated)