Introduction PharmaNet is a drug information system established in late 1995 for administrative purposes and to improve patient safety in the Canadian province of British Columbia (BC). PharmaNet contains almost 30 years of data on all prescription drugs and some medical supplies dispensed at outpatient pharmacies in BC. Methods Developed and maintained by the BC Ministry of Health, PharmaNet supports drug dispensing and helps monitor medication use. It also facilitates claims related to PharmaCare, a public programme assisting BC residents in paying for prescription medications, medical supplies, and devices. Outpatient pharmacists are required to submit information to PharmaNet for every prescription dispensed regardless of the payer. Results Data are considered to be complete and valid from January 1st, 1996. As of December 2025, PharmaNet contained over 1.8 billion records and more than 75 million new records are added annually. Each record represents a dispensation and includes over 80 variables capturing patient demographics (e.g., age and sex), drug dispensing (e.g., drug identification number, fill date, and days supply), payment information (e.g., professional fee and dispensing fee), and information pertaining to the prescriber and pharmacy that issued the prescription. Each person is identified by their Personal Health Number, a unique lifelong identifier assigned to each BC resident for accessing healthcare. This also permits linkage with other health datasets, including physician services and hospital records. PharmaNet data can be requested for research purposes through a standard data access process and, once approved, deidentified data are released to a secure remote research environment. Conclusions PharmaNet is a powerful resource for secondary health research. In addition to supporting studies of drug utilisation, it can when linked to other BC administrative databases enable a broad range of health research, including policy evaluation and long-term real-world drug effectiveness and safety. The data have also been used in multi-jurisdictional research through distributed analyses.

While the multiple sclerosis (MS) prodrome may be prolonged, healthcare use by sex >5 years pre-onset remains underexplored. We examined healthcare use up to 29 years pre-MS onset, stratified by sex. Using administrative data from Ontario, Canada (1991-2020), we compared annual physician visit rates by diagnostic chapter between MS and matched non-MS cohorts, stratified by sex. Quasi Poisson models estimated rate ratios (RRs) with 95% confidence intervals (CI). Included were 35,018 MS and 136,007 matched non-MS individuals (mean onset age ∼43 years; standard deviation ∼14 years; ∼69% female in both cohorts). Sex differences were observed for nervous system-related visits; males exhibited higher RRs from 10 years pre-onset ('year -10'), peaking in year -1 for both sexes (RRs: males=28.60;95%CI:24.10-34.00, females=13.60;95%CI:12.56-14.70). This was followed by mental disorders, being higher for males from year -4, peaking in year -1 (RRs: males=2.98;95%CI:2.73-3.25; females=2.09;95%CI:1.99-2.21), then ill-defined signs/symptoms and injury-related visits from year -3 to -1 (e.g., ill-defined signs/symptoms, RRs range: males=1.85-4.26 and females=1.67-3.27). Additionally, males had higher RRs in the 1-2 years pre-onset for respiratory, genitourinary, musculoskeletal, circulatory, digestive, and infection-related visits. Sex differences in healthcare use were evident up to 10 years pre-MS onset across multiple diagnostic chapters. Findings suggest that prodromal MS patterns differ by sex, potentially reflecting differences in symptom presentation, health-seeking behaviour, or diagnostic recognition before MS onset.

BACKGROUND:Few studies assessed drug dispensations for periods >5 years pre-multiple sclerosis (MS) onset or examined the time pre-MS symptom onset. OBJECTIVE:The objective of this study was to examine prescription drug dispensations up to 15 years pre-MS symptom onset. METHODS:Matched cohort study linking prescription, clinical and administrative data, 1996-2018. RESULTS:Among 1243/6212 MS/non-MS persons, those with MS had higher dispensation rate ratios (RRs) pre-onset for systemic antibacterials up to 15 years (RR (range) = 1.15-1.30; q-values (range) = 0.11-0.31), airway drugs up to 8 years (RR (range) = 1.49-1.71; q-values (range) = 0.26-0.35), sex hormones (females only) up to 7 years (RR (range) = 1.19-1.46; q-values (range) = 0.26-0.39) and psychoanaleptics up to 6 years (RR (range) = 1.36-1.59; q-values (range) = 0.09-0.31), but lower beta-blockers (years 13-14 pre-onset, RRs = 0.10; q-values (range) = <0.00001-0.31). After multiple-comparison adjustment, beta-blockers remained significant. CONCLUSION:Drug dispensations differed up to 15 years pre-MS onset and may reflect early disease.

RATIONAL & OBJECTIVE:People with glomerulonephritis (GN) may be at an increased risk of cancer, but existing studies have not accurately clarified the cancer risk in patients with GN. A better understanding of these risks may inform cancer screening strategies and identify potentially modifiable risk factors. This study identified risk factors for cancer among Canadian patients with GN. STUDY DESIGN:Retrospective observational cohort study. SETTING & PARTICIPANTS:Adults with diagnosed GN (n = 4,039) identified using a centralized pathology registry in British Columbia, Canada, between 2000 and 2020. EXPOSURE:Known cancer risk factors, including age, sex, ethnicity, smoking, alcohol abuse, obesity, diabetes, dyslipidemia, hypertension, and cardiovascular disease as well as GN-related potential risk factors, including GN disease type, estimated glomerular filtration rate (eGFR), and level of proteinuria. OUTCOME:All-cause cancer, excluding non-melanoma skin cancer. ANALYTICAL APPROACH:Standardized incidence ratios (SIRs) were calculated using an age- and sex-matched general population. The time to the first cancer event was modeled using a cause-specific hazards model, with death considered as a competing event. RESULTS:The mean age of the cohort was 51 years, and 52% of the participants were male. During a median of 7.8 years of follow-up, 384 patients (9.5%) developed de novo cancer. The 20-year cancer risk was 23%, with an incidence rate 30% higher than the general population (SIR, 1.3 [95% CI, 1.2-1.4]). The risk was most pronounced in patients younger than 40, almost 3-fold higher than in the general population (SIR, 2.9 [95% CI, 1.6-4.6]). Significant increases in cancer incidence were observed for lymphoma (SIR, 3.5), kidney (SIR, 2.6), colorectal (SIR, 2.4), and lung cancers (SIR, 1.5). Elevated risk was observed both before and after the onset of end-stage kidney disease. Age, male sex, baseline eGFR, and GN disease type were independently associated with cancer risk. LIMITATIONS:The lack of immunosuppression data. CONCLUSIONS:Patients with GN have a substantially increased risk of cancer compared with the general population, particularly younger patients who are typically excluded from current screening programs. These findings suggest the need to raise awareness of the cancer risk among people with GN and may inform the further development of tailored cancer screening and prevention strategies, especially among younger adults with GN. PLAIN-LANGUAGE SUMMARY:People with glomerulonephritis (GN) are thought to be at an increased risk of cancer, but previous studies had limitations that limited the accuracy of the cancer risk estimates in this population. This study of 4,039 patients with GN found the incidence of cancer to be 30% higher than in the general population. Among patients under 40 years of age, the risk was nearly 3-fold. The types of cancers at increased risk were colorectal, lung, kidney, and lymphoma. These findings highlight the need to raise awareness of cancer risk in patients with GN and may inform further development of targeted screening and prevention strategies, particularly among younger adults.

BACKGROUND:Increased healthcare use precedes classical multiple sclerosis (MS) symptom onset. Limited evidence exists on sex and age variation. We assessed physician visit patterns pre-MS onset by sex and age. METHODS:Using data from British Columbia, Canada, we compared annual physician visit rates (overall, by reason and specialty) in the 15 years before the neurologist-determined MS symptom onset date (index) and a matched non-MS cohort, stratified by sex and age (<30, 30-49, ≥50). RESULTS:We included 2038 MS and 10 182 non-MS persons (74% female). Mean age (years) at index was 37.6 (females) and 38.7 (males). Compared with matched non-MS persons, females with MS showed earlier and more consistent elevations in physician visits (years -14 to -1), while males had sporadic elevations (years -5, -3 and -1). Females also had longer periods of elevated rate ratios (RRs) for ill-defined signs/symptoms (years -15 to -1), mental disorders (years -14 to -1 except year -7) and musculoskeletal conditions (years -6 to -1). Females exhibited sustained elevated visits by specialty, including general practice (all years; RR ≥1.1), psychiatry (years -12 to -1 except -8 to -6; RRs ≥1.6) and ophthalmology (years -9 to -1 except -2; RRs ≥1.4). Compared with matched non-MS counterparts, those aged 30-49 years had sustained higher RRs for psychiatry visits (years -12 to -1 except -8 and -6; RRs ≥1.9) and ophthalmology (years -9 to -1; RRs ≥1.4). Other age groups had fewer elevated RRs preindex. Across comparisons, RRs were of similar magnitude across sex and age groups. CONCLUSIONS:Sex-specific and age-specific differences in physician visits extended up to 15 years pre-MS onset, suggesting a durable prodromal signature, most evident in females and those aged 30-49 years.

BACKGROUND:Emergency department (ED) visits represent a substantial component of healthcare use. Although increased utilization has been reported before multiple sclerosis (MS) recognition, ED use and its clinical context remain poorly characterized. METHODS:In this population-based matched cohort study, we used linked administrative health data from British Columbia, Canada. Persons with multiple sclerosis (PwMS) were identified using a validated case definition and matched to individuals without MS by age, sex and postal code. ED encounters in the 5 years preceding the index date (first demyelinating disease claim) were examined. Annual ED visit rates were compared using adjusted negative binomial models; discharge diagnoses were assessed using adjusted logistic regression. RESULTS:The cohort included 2383 PwMS and 11,896 matched individuals (mean age = 44.1 years; 69.2% female). ED visit rates were higher among PwMS across all five pre-index years, rising from an adjusted rate ratio (aRR) of 1.38 (95% confidence interval (CI) = 1.07-1.77) at 5 years pre-index to 3.40 (95% CI = 3.07-3.76) in the year immediately pre-index. Excess visits were observed for gastrointestinal, genitourinary, musculoskeletal, respiratory, mental health, and ill-defined conditions. PwMS had higher odds of a non-specific discharge diagnosis (adjusted odds ratio (aOR) = 1.36; 95% CI = 1.22-1.53). CONCLUSIONS:ED utilization was consistently elevated in the 5 years before MS recognition (index date), predominantly involving non-neurologic and non-specific presentations.

BACKGROUND:Little is known about the prodrome in pediatric-onset multiple sclerosis (POMS), especially potential age and sex differences. OBJECTIVE:To compare annual hospitalization and physician visit rates overall, and for physician visits, examine sex- and age-group (<12, 12-15, 16 to <18 years) differences, up to 18 years pre-first demyelinating event (pre-index) in 451 POMS and 1420 matched non-multiple sclerosis (MS) individuals. METHOD:Using Ontario administrative data (1991-2020), we estimated rate ratios (RRs) by visit-related diagnosis, sex, and age using over-dispersed-Poisson models. RESULTS:The POMS cohort showed elevated rates by year 1 pre-index ("year -1"), for ill-defined, other-health-contact, mental-, and respiratory-related hospitalizations (RRs ⩾ 3.6). Elevated physician visits began at year -14 for respiratory (RRs ⩾ 1.3), year -13 for endocrine (RRs ⩾ 2.4) and injury-related (RRs ⩾ 1.2), year -12 for ill-defined (RRs ⩾ 1.3), and year -11 for nervous system (RRs ⩾ 4.0) and mental-related (RRs ⩾ 2.0). Nervous system visits showed the largest sex-gap; males with (vs without) MS had elevated RRs from year -12 and females from year -6; both peaked the year pre-index (males: RR = 47.2; 95% confidence interval (CI): 17.3-128.3; females: RR = 17.0; 95% CI: 8.9-32.3). Among 12- to 15-year-olds, elevated rates were sustained for respiratory-related (from year -12, except year -5), injury-related, and ill-defined (from year -13); (all RR ⩾ 1.3), while findings were more sporadic in the other age groups. CONCLUSION:Findings suggest prolonged, organ-specific healthcare use pre-POMS, with some sex- and age-related differences indicating pre-onset disease heterogeneity.

Motivation Compositional data comprise vectors that describe the constituent parts of a whole. Data arising from various -omics platforms such as 16S and RNA-sequencing are compositional in nature. However, correlations between features on raw counts have no meaningful interpretation. Metrics of proportionality were formulated to address this problem. However, there is an inherent bias that arises when calculating these metrics empirically on count-based measures due to variability in read depths. Results We quantify the bias introduced by empirically calculating proportionality-based association metrics in count data. Additionally, we propose a means of estimating these metrics within a logit-normal multinomial model in pursuit of more accurate estimates. The model-based estimates are shown to outperform empirical estimates in simulated data, and are additionally applied to a mouse embryonic stem-cell single-cell sequencing dataset as well as a pediatric-onset multiple sclerosis metagenomic dataset. Availability and Implementation An R package is available at https://CRAN.R-project.org/package=countprop . Supplementary information Supplementary data are available at Bioinformatics online.

BACKGROUND:Evidence suggests a prodromal phase in multiple sclerosis (MS) identifiable via healthcare use, including psychiatric symptoms. The association between psychiatric morbidity in this phase and future outcomes remains unclear. OBJECTIVES:We investigated the association between psychiatric morbidity in the 5-years pre-MS onset and subsequent disability (EDSS) scores. METHODS:We identified MS patients who visited an MS clinic in British Columbia, Canada (1991-2018) and linked their clinical and population-based health administrative data. Psychiatric morbidity was identified using physician/hospital visits in the 5-years pre-MS onset. Multivariable generalized linear models examined the association between psychiatric morbidity and subsequent EDSS scores. We explored effect modification by sex, age, and MS course and investigated if high psychiatric-related physician visits (>median) or hospitalizations (measures of "psychiatric morbidity burden") were associated with EDSS scores. RESULTS:Among 2212 MS patients, 481 (21.7%) had psychiatric morbidity in the 5-years pre-MS onset. Follow-up averaged 5.2 (SD: 4.9) years (first-to-last EDSS assessment). Psychiatric morbidity pre-MS onset was associated with higher post-diagnosis EDSS scores (covariate-adjusted[adj) β = 0.17; 95% confidence interval (CI): 0.03-0.30). Associations were more pronounced in males (adjβ = 0.43; 95% CI: 0.04-0.83), <30 years (adjβ = 0.44; 95% CI: 0.15-0.73), relapsing-onset-MS (adjβ = 0.22; 95% CI: 0.08-0.37) and high psychiatric-related physician visit burden (adjβ = 0.23; 95% CI: 0.05-0.41), or hospitalizations (adjβ = 0.48; 95% CI: 0.002-0.96). CONCLUSIONS:Psychiatric morbidity before MS recognition was associated with increased future disability, particularly in males, younger individuals, relapsing-onset-MS, and high pre-MS onset psychiatric morbidity burden.

BACKGROUND:Early recognition of multiple sclerosis (MS) remains a pivotal challenge. Little is understood about the trajectories of health care use before recognition of adult-onset MS, and the relationship of the trajectories with subsequent disability. METHODS:We accessed linked clinical and population-based health administrative data in British Columbia, Canada (1991-2020). Using group-based multi-trajectory models, we described the joint trajectories of physician visits, hospitalizations, and prescription classes filled in the 10 years preceding MS recognition - either the first recorded demyelinating event (administrative index date, n = 6349) or MS symptom onset (clinical index date, n = 725) at age ≥ 18 years. Across the identified trajectories, we compared demographics and encounters potentially related to MS pre-index-date. In the clinical cohort, we assessed the relationship between the trajectories and disability scores using a linear mixed-effects model, adjusting for confounders. RESULTS:Before the administrative index date, we identified 3 trajectories: low (57 % of cohort), moderate (36 %), and high (6 %) health care use. Before the clinical index date, we also identified low (51 %), moderate (40 %), and high (9 %) trajectories. In both cohorts, individuals in the moderate and high (versus low) trajectories were more likely to be female and older, and have earlier neurologist and ophthalmologist visits, and nervous system diagnoses pre-index-date. The moderate (versus low) trajectory was associated with higher subsequent disability (adjusted beta coefficient=0.31;95 %CI:0.09-0.53). CONCLUSIONS:Earlier detection of MS by general practitioners and prompt MS drug treatment may be possible for patients accessing health care frequently, potentially mitigating disability progression.

Evidence of increased healthcare use occurring before paediatric-onset multiple sclerosis presentation suggests a prodromal phase. However, little is known of its duration or features, and few studies have accessed a clinical cohort to examine the period before symptom onset. We compared annual rates of healthcare use before paediatric multiple sclerosis onset in clinical and administrative cohorts versus matched non-multiple sclerosis cohorts. We identified persons with paediatric-onset multiple sclerosis from the Swedish Multiple Sclerosis registry and population-based administrative data using a validated algorithm requiring ≥3 hospital or outpatient multiple sclerosis diagnostic codes recorded on separate dates. The index date was multiple sclerosis symptom onset, as recorded in the multiple sclerosis registry by a neurologist (clinical cohort) or the earliest demyelinating disease-related International Classification of Diseases code (administrative cohort). Individuals with age at index <18 years were matched with up to five individuals from the general population on sex, birth year, county of residence at the index date and residency time. Healthcare use was measured as hospital/outpatient diagnoses (International Classification of Diseases chapters) and prescription drug classes (Anatomical Therapeutic Chemical classification system, 2nd level). Yearly rates of hospital and outpatient visits (up to 17 years pre-index) and prescription fills (up to 14 years pre-index) were compared using Quasi-Poisson regression. The clinical/administrative cohorts included 233/206 paediatric-onset multiple sclerosis and 1151/1011 matched individuals, with a mean age at the index of 16 years (standard deviation: 2) in all four groups. In both cohorts, individuals with paediatric-onset multiple sclerosis exhibited elevated healthcare use predominantly 1–10 years pre-index, including, for example, higher prescriptions filled for corticosteroids for dermatological use (rate ratio range: 2.61–3.91) and outpatient visits for ill-defined signs/symptoms (rate ratio range: 2.17–8.64) and unassigned ICD codes (rate ratio range: 2.20–4.17). In the year pre-index, individuals with paediatric-onset multiple sclerosis in both cohorts exhibited higher rates of outpatient visits for neoplasms, nervous system disorders, sense organ conditions, ill-defined signs/symptoms and ‘other health system contact’ (rate ratio range: 2.05–18.00). In the same year, the clinical cohort also had higher rates of prescription fills for ‘other gynecologicals’ (4.08, 95% confidence interval: 1.04–16.09), and the administrative cohort had higher rates for prescriptions filled across eight drug classes (rate ratio range: 1.56–6.49). Healthcare use was higher primarily in the 1–10 years before paediatric-onset multiple sclerosis versus a matched cohort, suggestive of a prodromal phase. During this period, the paediatric-onset multiple sclerosis cohort was more often identified as having ill-defined signs/symptoms, neoplasms and skin-related issues.

Importance Health care use increases before multiple sclerosis (MS) onset. However, most studies have focused on the 5 to 10 years preceding the first demyelinating disease code from administrative data. Few studies have examined patterns before clinically determined MS symptom onset from clinical records. Objective To examine health care use 25 years before MS symptom onset in a clinical cohort from British Columbia, Canada. Design, Setting, and Participants This matched cohort study accessed data prospectively collected from January 1991 to September 2018. All data were released mid-2024 for analysis. The study was conducted in British Columbia using publicly funded universal health insurance data. Patients with MS were identified from MS clinic records and matched with up to 5 individuals randomly selected without replacement from the general population by sex, birth year, socioeconomic status, and postal code of residency. Main Outcomes and Measures Linked clinical and administrative data were used to compare physician visit rates 25 years before MS onset using adjusted negative binomial models and 15 years before MS onset by International Classification of Diseases, Ninth Revision (ICD-9) chapter and physician specialty. Results A total of 2038 patients with MS (mean [SD] age at symptom onset, 37.9 [10.9] years; 1508 female [74.0%]) and 10 182 matched individuals were included. All-cause physician visit rate ratios (RRs) for patients with MS were consistently elevated from 14 years before onset (adjusted RR [ARR], 1.19; 95% CI, 1.07-1.33), peaking the year before MS onset (ARR, 1.28; 95% CI, 1.21-1.35). The RRs for ill-defined symptoms and signs were consistently elevated 15 years before onset, exceeding 1.15 and peaking at 1.37 (95% CI, 1.19-1.56) the year before MS onset. Mental health-related RRs from 14 years before onset were significant (excluding years 7, 5, and 4), with RRs in the 3 years before MS onset ranging from 1.30 (95% CI, 1.05-1.58) to 1.38 (95% CI, 1.12-1.68). Sensory, musculoskeletal, and nervous system RRs were elevated 8, 5, and 4 years before onset, respectively, with, for example, a peak of 2.42 (95% CI, 1.90-3.07) for nervous system concerns the year before MS onset. By physician specialty, general practice visit RRs were significantly elevated in each of the 15 years before MS onset, reaching 1.23 (95% CI, 1.17-1.30) in the year before onset. Psychiatry visit RRs were elevated 12 years before onset (2.59; 95% CI, 1.23-5.47). Neurology and ophthalmology RRs were significantly higher up to 8 to 9 years before onset, peaking the year before MS onset at 5.46 (95% CI, 4.30-6.93) for neurology and 1.64 (95% CI, 1.30-2.08) for ophthalmology. Conclusions and Relevance In this matched cohort study of people with and without MS, health care use was higher among patients with MS 14 to 15 years before MS symptom onset, suggesting that MS may have started earlier than previously thought. Mental health and psychiatric issues along with ill-defined signs and symptoms might be among the earliest features of the prodromal period preceding nervous system-related and neurologic visits by 7 to 11 years.

BACKGROUND:Evidence suggests signs and symptoms may emerge years before the clinical onset of multiple sclerosis (MS). We investigated secondary healthcare use and medication dispensations before MS onset. METHODS:Using linked administrative data, we analyzed a neurologist-diagnosed MS clinical cohort and an algorithm-defined administrative cohort in Sweden (2001-2019). People with MS (PwMS) were matched to up to five non-MS comparators by sex, birth year, residency location, and residency duration before the index date (clinical: symptom onset; administrative: first MS/demyelinating diagnosis code). Annual outpatient visits and hospitalizations by diagnosis codes across 19 years pre-index, and dispensed medications by anatomical and therapeutic classifications across 14 years pre-index were compared. RESULTS:The clinical cohort included 7604 PwMS and 37,974 matched comparators (median age at symptom onset = 35.5; 68.5% females). In the 5 years pre-index, outpatient visits were 11%-73% higher among PwMS for disturbances of sense organs, nervous, musculoskeletal, digestive, and genitourinary systems, mental health/behavior, and ill-defined symptoms/signs, with visits related to sense organs elevated up to 6 years. Hospitalizations were often elevated in the year pre-index. Dispensations of nervous system-related, musculoskeletal, blood-related, metabolic, sensory, respiratory, and dermatological agents were elevated by 6%-22% in the 5 years pre-index, with elevations in nervous system-related and musculoskeletal agents extending up to 6-9 years pre-index. Findings were similar in the administrative cohort with greater magnitudes and longer durations pre-index. CONCLUSIONS:We observed increased hospital, outpatient, and prescription utilization for multiple body systems 6-9 years pre-MS onset. These patterns provide a more comprehensive picture of the MS prodrome.

BackgroundResearch and surveillance using routinely collected health data rely on algorithms or definitions to ascertain disease cases or health measures. Whenever algorithm validation studies are not possible due to the unavailability of a reference standard, algorithm feasibility studies can be used to create and assess algorithms for use in more than one population or jurisdiction. Publication of the methods used to conduct feasibility studies is critical for reproducibility and transparency. Existing guidelines applicable to feasibility studies include the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) and REporting of studies Conducted using Observational Routinely collected health Data (RECORD) guidelines. These guidelines may benefit from additional elements that capture aspects particular to multi-jurisdiction algorithm feasibility studies and ensure their reproducibility. The aim of this paper is to identify the minimum elements for reporting feasibility studies to ensure reproducibility and transparency. MethodsA subcommittee of four individuals with expertise in routinely collected health data, multi-jurisdiction health research, and algorithm development and implementation was formed from Health Data Research Network (HDRN) Canada's Algorithms and Harmonized Data Working Group (AHD-WG). The subcommittee reviewed items within the STROBE and RECORD guidelines and evaluated these items against published feasibility studies. Items to ensure transparent reporting of feasibility studies not contained within STROBE or RECORD guidelines were identified through consensus by subcommittee members using the Nominal Group Technique. The AHD-WG reviewed and approved these additional recommended elements. ResultsEleven new recommended elements were identified: one element for the title and abstract, one for the introduction, five for the methods, and four for the results sections. Recommended elements primarily addressed reporting jurisdictional data variabilities, data harmonization methods, and algorithm implementation techniques. SignificanceImplementation of these recommended elements, alongside the RECORD guidelines, is intended to encourage consistent publication of methods that support reproducibility, as well as increase comparability of algorithms and their use in national and international studies.

OBJECTIVE:Phenotype hospital, physician, and emergency department (ED) visits by diagnoses and specialty up to 29 years pre-multiple sclerosis (MS) onset versus a matched population without MS. METHODS:We identified people with MS (PwMS) using population-based administrative data from Ontario, Canada (1991-2020). The first MS/demyelinating diagnostic code defined MS onset (the index date). Annual rates of healthcare use (hospital, physician, ED) by primary diagnosis (chapter-level) and physician specialty pre-index were compared between PwMS and up to 5 matched population comparators using overdispersed-Poisson regression. RESULTS:Up to 35,018 PwMS and 136,007 population comparators were included. Consistently elevated yearly physician visit rate ratios (RRs) were observed 28 years pre-index for: mental-health (RR > 1.29) and ill-defined signs/symptoms (RR > 1.15), 24 years for: nervous (RR > 1.47), musculoskeletal (RR > 1.21), injury, and respiratory-related issues (RR > 1.07), and 22 years for digestive-system (RR > 1.18). The magnitude increased as the index date approached, peaking the year pre-index for physician, hospital, and ED visit RRs for: nervous-system (range: 12.06-17.13); ill-defined signs/symptoms (range: 3.51-5.45), mental-health (range: 2.13-2.70), musculoskeletal (range: 1.84-2.96), injury (range: 1.58-2.27), digestive-system (range: 1.49-1.78) and respiratory-system (range: 1.37-2.06). By specialty, yearly visit RRs for primary care were > 1.08 for 28 years pre-index, internal medicine exceeded 1.19 for 25 years, and psychiatry and neurology > 1.52 for 24 years pre-index. INTERPRETATION:Higher healthcare use was evident for over two decades before the first demyelinating event. Mental-related, ill-defined signs/symptoms and primary care visits were consistently elevated the longest (28 years pre-index), followed by nervous-system, musculoskeletal, injury, respiratory-related, and digestive-system (22-24 years pre-index). Health-related phenotypical differences appear early in the MS disease process.

BACKGROUND:We investigated the association between multiple sclerosis (MS) and fractures, dislocations/sprains/strains, and burns preceding MS recognition. METHODS:We conducted a cohort study using clinical and population-based health administrative data in British Columbia, Canada (1991-2020). We compared the risk of a fracture, dislocation/sprain/strain, and burn in the six years preceding an MS cases' first demyelinating claim (administrative cohort=9197) or MS symptom onset (clinical cohort=1446) to that of matched general population controls using modified Poisson regression. As sensitivity analyses, we used high-dimensional propensity scores (hdPS) to address residual confounding and targeted maximum likelihood estimation (TMLE) for mis-specification. RESULTS:In the six years before the first demyelinating claim (administrative cohort), the risk of a fracture (adjusted relative risks [adjRR]=1.28;95 %CI:1.20-1.36), dislocation/sprain/strain (adjRR=1.20;95 %CI:1.15-1.23), and burn (adjRR=1.40;95 %CI:1.22-1.62) was higher among MS cases. After hdPS adjustment and TMLE, the adjusted relative risks decreased slightly: fracture (hdPS=1.20; TMLE=1.20), dislocation/sprain/strain (hdPS=1.15; TMLE=1.15), and burn (hdPS=1.25; TMLE=1.26). Pre-MS symptom onset (clinical cohort), the associations were weaker but in the same direction. CONCLUSION:Fractures, dislocations/sprains/strains, and burns were more common among people with MS before its classical recognition, suggesting that MS could be detected earlier.

Chronically elevated type I interferons (-β and -α) can induce atherosclerosis and autoimmunity but whether this link translates into adverse events in interferon-β users with multiple sclerosis is unknown. We therefore aimed to determine whether long-term interferon-β exposure increases the risk of cardiovascular and autoimmune disease in a Canadian population-based cohort with linked hospital/physician visits and filled prescriptions. People with multiple sclerosis were included from the most recent of (i) first diagnostic code or disease-modifying therapy or (ii) prescription data availability (1/JAN/1996), and followed until the earliest of outcome, emigration, death or study end (31/DEC/2017). Associations were tested using stratified Cox regressions with time-dependent covariates. The cohort included 19 360 people with multiple sclerosis followed for a median duration of 11.2 years (Q1-Q3: 5.1-18.7), of whom 3138 (16.2%) ever used an interferon-β. Longer interferon-β therapy was associated with a higher incidence of cardiovascular disease (per 5-year longer treatment: hazard ratio = 1.18; 95% confidence interval: 1.02, 1.37; P = 0.026) but not with autoimmunity (hazard ratio = 0.74; 95% confidence interval: 0.49, 1.11; P = 0.139). This new safety signal should encourage clinicians to optimize cardiovascular prevention in people with multiple sclerosis and may be considered when discussing treatment options in interferon-β users who are at high risk or with established cardiovascular disease.

Abstract Objective Elevated healthcare use before multiple sclerosis (MS) onset suggests earlier opportunity to identify MS. Yet their timing and sociodemographic effects are unclear. We examined rates of healthcare use (and by age/sex) for >two decades pre‐MS onset. Methods We identified people with MS (PwMS) using administrative data from Canada (Ontario) and Sweden (1991–2020) (“administrative” cohort), and the Swedish MS Registry (“clinical” cohort). The first MS/demyelinating diagnostic code (administrative) or symptom onset (clinical) defined MS onset. We compared annual rates of healthcare use (hospital, physician, and emergency‐room [ED]) pre‐onset between PwMS and up to five matched population controls using negative binomial regression, and by age/sex. Results The administrative cohort = 35,018/136,007 PwMS/controls (Ontario), and 10,269/51,297 (Sweden). Rates of healthcare use were higher for PwMS than controls up to 28 (of 29) years (Ontario) and up to 15 (of 19) years (Sweden) pre‐onset. Annual healthcare use rose steadily as onset approached, particularly escalating 7 years pre‐onset in Ontario (e.g., hospital visit rate ratios [RRs] exceeded 1.30), and 6 years in Sweden (physician visit RRs > 1.10). RRs peaked the year pre‐onset (ED visits [Ontario] = 3.04; 95% CI: 2.94–3.13, physician visits [Sweden] = 2.51; 95% CI: 2.44–2.59). In the year pre‐onset, RRs were disproportionately higher for males (ED RRs [Ontario] = 3.30; 95% CI: 3.13–3.48 vs. females = 2.90; 95% CI: 2.79–3.02), and dropped steadily by age (physician visit RRs [Sweden] = 2.61/2.27/1.97/1.72 for 50/40/30/20‐year‐olds). The smaller clinical cohort (7604/37,974 PwMS/controls) exhibited similar patterns, albeit more modest, with RRs elevated up to 5 years pre‐onset (physician visit RR [year‐5] = 1.08; 95% CI: 1.02–1.14; RR [year‐1] = 1.39;1.33–1.46). Interpretation Higher healthcare use was evident decades before MS onset, escalating 6–7 years pre‐onset, peaking the year before, being disproportionately higher for males and older PwMS.