Emergency department use by persons with MS: A population-based descriptive study with a focus on infection-related visits

We described emergency department (ED) visits (all visits and infection-related) by persons with multiple sclerosis (MS) in British Columbia, Canada (1 April 2012 to 31 December 2017). We identified 15,350 MS cases using health administrative data; 73.4% were women, averaging 51.4 years at study entry. Over 4.9 years of follow-up (mean), 56.0% of MS cases visited an ED (mean = 0.6 visits/person/year; total = 37,072 visits). A diagnosis was documented for 25,698 (69.3%) ED visits, and 18.4% (4725/25,698) were infection-related. Inpatient admissions were reported for 20.4% (5238/25,698) of all and 29.2% (1380/4725) of infection-related ED visits. Findings suggest that the ED plays a substantial role in MS healthcare and infection management.


Introduction
Emergency department (ED) presentations represent an important aspect of health care utilization in the general population. 1,2 Compared to the general population, persons with multiple sclerosis (MS) frequently access the healthcare system, with infections contributing to this higher healthcare use. For example, persons with MS had 41% more infection-related physician claims (adjusted rate ratio = 1.41; 95% confidence interval: 1.36-1.47) versus a sex-, age-, and region-matched non-MS population. 3 However, relatively little is known about ED utilization by persons with MS. [4][5][6] Here, we described overall and infectionrelated ED use in an MS population.

Methods
We performed a descriptive, population-based study in British Columbia (BC), Canada, using linked health administrative data (via Population Data BC 7 ), including Medical Service Plan Billing Information 8 (providing physician claims); the Discharge Abstract Database 9 (providing hospital admissions/discharges); PharmaNet 10 (for prescriptions filled at outpatient/ community pharmacies); Census Geodata (providing socioeconomic status estimates); Registration and Premium Billing files 11 (providing BC residency status via the mandatory healthcare plan registration days); Vital Statistics 12 (capturing death dates); and the National Ambulatory Care Reporting System dataset 13 (providing ED-related visit dates and Diagnosis Shortlist codes, 14  All MS cases were identified with a validated algorithm. 15 Study entry date was the later of the first MS/demyelinating disease-related International Classification of Diseases (ICD-9/10) code, or first disease-modifying drug (DMD) prescription filled, or 1 April 2012 (start of the ED date). All included persons were ⩾18 years old and BC residents for ⩾1 year pre-study entry; follow-up ended at the earliest of death, emigration, or 31 December 2017. Comorbidities were measured using a modified Charlson Comorbidity Index during the 1-year pre-study entry. [16][17][18] MS cases ever filling a DMD prescription during follow-up were described.

Results
We identified 15,350 MS cases (11,270 (73 17,18 The proportion of persons with MS scoring 1 or more on the Comorbidity Index is consistent with prior work 16 conducted in similar cohorts. The most common comorbid conditions (present at study entry) which were identified using the Index were "chronic pulmonary disease" (present in 1102/15,350; 7.2% of the study cohort), "diabetes mellitus without chronic complications" (1011; 6.6%), and "cerebrovascular disease" (629; 4.1%). d Captured as prescriptions filled; some people were exposed to >1 DMD during follow-up; hence, the sum of cases filling first-and second-generation DMD prescriptions exceeded the sum of cases ever filling a DMD prescription; first-generation DMDs included beta-interferon and glatiramer acetate, and second-generation DMDs included natalizumab, fingolimod, dimethyl fumarate, teriflunomide, alemtuzumab, daclizumab, and ocrelizumab. Pre-study entry (1 January 1996 to 31 March 2012) 19.5% (2991/15,350) of MS cases had ever filled a DMD prescription. e The denominator used to estimate the following proportions was the total number of ED visits captured during follow-up (n = 37,072); some people had more than one ED visit during follow-up; hence, the sums of ED visits exceed the total number of persons with at least one ED visit.

Socioeconomic status
There were no clear patterns across the socioeconomic quintiles at study entry for the MS cases: ever/ never filling a DMD prescription during follow-up, ever/never being hospitalized subsequent to an ED visit, or for the five most common ED diagnoses (data not shown).

Conclusion
Over 50% of persons with MS had more than one ED consultation during our nearly 6-year observation period; one-quarter visited an ED three or more times.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Canadian Institutes of Health Research (CIHR) Project and Foundation grant (PJT-156363 and FDN-159934, PI: Tremlett).

Supplemental material
Supplemental material for this article is available online.