Incidence, prevalence, and mortality of localized scleroderma in Quebec, Canada: a population-based study

Summary

Background

Localized scleroderma is an understudied autoimmune disease characterized by fibrosis of the skin and/or subcutaneous tissue. To date, only 6 articles reported on the incidence and/or prevalence estimates globally, with significant design limitations and risk of bias. None of the studies originated from Canada or investigated mortality/geospatial epidemiology. We aimed to study the incidence, prevalence, mortality and spatiotemporal trends of localized scleroderma in Quebec, Canada, stratified by sex and age.

Methods

Quebec populational health administrative databases were used to identify localized scleroderma cases from 1989 to 2019. Crude incidence rate, age-standardized incidence rate, prevalence and mortality analyses were conducted using negative binomial random walk models. Spatial analyses were conducted using a Poisson Besag-York-Mollié regression model.

Findings

There were 6063 incident localized scleroderma cases identified over the total period of the study (mean age 53.0, standard deviation [SD] 20.2 years at diagnosis). The overall age and sex-standardized incidence rate was 3.25/100,000 person-years [95% Confidence Interval (CI) 3.17–3.33]. Among 6063 incident cases, 4510 (74.4%) were female and 1553 (25.6%) were male, yielding a female-to-male ratio of approximately 3:1. In females, we noted an initial increase in age-standardized incidence rate followed by a plateau and a decrease after 2013 (average annual percent change −2.0 [95% CI −3.7 to −0.2]%). In males, a steady decrease in age-standardized incidence rate was observed (average annual percent change −3.3 [95% CI −5.0 to −1.8]%). The highest incidence rate was observed in the 60–79 year-old age group for females and the 80+ group for males. Age-standardized incidence rate varied geographically with hotspots identified in the south of Quebec. The average prevalence was 24.5/100,000 [95% CI 24.3–24.8]. The overall standardized mortality ratio was comparable for females (1.04 [95% CI 0.95–1.14]) and males (1.14 [95% CI 0.98–1.33]) and decreased steadily over time for both sexes (from 1.31 [95% CI 1.06–1.58] in 1996 to 0.81 [95% CI 0.66–0.98] in 2019). Standardized mortality ratio analysis revealed excess death only in females aged 40–59 years.

Interpretation

From 1989 onward, we report an initial increase in the age and sex-standardized incidence rate of localized scleroderma in Quebec followed by a recent decrease after 2013, as well as a generally increasing prevalence from 1996 to 2019. Standardized mortality ratio analysis confirmed the clinical observation that localized scleroderma is a morbid rather than life-threatening disease. We demonstrate an uneven geographic distribution of localized scleroderma incidence in Quebec.

Funding

This project was funded by Canadian Dermatology Foundation, National Scleroderma Foundation and Canadian Institutes of Health Research. Dr. Netchiporouk received FRQS Junior 1 Clinician Scientist Salary Award.

Research in context.

Evidence before this study

Localized scleroderma, also known as morphea, is an autoimmune disease primarily characterized by fibrosis of the skin, but may also involve underlying tissues. The depth and location of connective tissue involvement in localized scleroderma poses considerable challenges, associated with disfigurement and potential disability. Studying the epidemiology of localized scleroderma to understand its burden and identify factors influencing its impact are essential in addressing the challenges posed by this disease.

We searched PubMed for epidemiological studies assessing the incidence, prevalence, and mortality of localized scleroderma patients published before January 2023 without restrictions. The following search term strategy was used: ((“eosinophilic fasciitis” OR “morphea” OR “localized scleroderma” OR “linear scleroderma”) AND (“prevalence” OR “incidence” OR “mortality” OR “epidemiology”).

Only 6 original manuscripts and 1 systematic review were identified. Three studies were conducted in Europe, two in North America, and only 1 study originated from Africa. Estimated incidence rates appeared higher in North America, particularly among adult, female, and Caucasian patients, although significant variations between studies and design limitations warrant further investigation. A single study (1960–1993) reported an increased incidence over time, however newer data is required. Furthermore, no articles studying spatial epidemiology nor mortality were identified.

Added value of this study

Our study provides a detailed examination of the age and sex-stratified incidence, prevalence, mortality and spatiotemporal distribution of localized scleroderma in Quebec, Canada, using populational administrative data from 1996–2019. We demonstrate an initial increase in localized scleroderma incidence from 1996 to 2004, followed by a plateau until 2013 and a decrease until 2019, specifically more pronounced after 2017, possibly related to changes in the Quebec billing system or better disease recognition in preceding years. We show an increasing prevalence over time, similar to other autoimmune skin diseases. We report reassuring mortality data confirming the clinical observation of localized scleroderma being a highly morbid but usually not a life-threatening disease. An increased risk of death limited to 40-59 year-old females was observed and deserves further investigations. We show an uneven geographic distribution of localized scleroderma incidence across Quebec, with clustered hotspots in the south of the province, prompting further research into the risk factors associated with the incidence and severity of this disease.

Implications of all the available evidence

The results of our study set the stage for further in-depth analyses of the epidemiology of localized scleroderma in Quebec, Canada and globally. A better understanding of localized scleroderma's epidemiology is essential to raise awareness of this morbid disease, and stimulate research funding and patient advocacy for improved healthcare resource planning and distribution.

Introduction

Morphea, also referred to as localized scleroderma, is an autoimmune disease characterized by fibrosis of the skin and/or subcutaneous tissue and is thought to be externally triggered (e.g., radiation, chemotherapy, trauma) in genetically predisposed individuals.¹ The diagnosis of localized scleroderma is typically established clinically, however, histopathological and/or imaging evaluation may be needed for diagnosis confirmation and/or extent assessment.²

The epidemiological data on localized scleroderma is scarce. In a recent systematic review (2022), we highlighted that only 6 articles reported incidence and/or prevalence estimates globally with no studies originating from Canada or studying mortality or geospatial epidemiology.³ A better understanding of localized scleroderma epidemiology is essential to raise awareness of this disease, stimulate research funding and patient advocacy for improved healthcare resource planning and distribution.⁴ Using populational healthcare administrative data for the province of Quebec (Canada) spanning 30 years (1989–2019), we aimed to investigate trends in incidence, prevalence, and mortality of localized scleroderma over time with detailed analysis by sex and age. Geographic variability in incidence was studied to evaluate geographic clustering of localized scleroderma.

Methods

Study design

The observational study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement for observational studies.⁵ Institutional ethical approval was obtained (#2019–5296).

Data sources

The province of Quebec is the second largest province in Canada by population counts and the largest by land area.⁶ All Quebec citizens benefit from universal healthcare services. In 2020, ∼8.3 million Quebecers were covered by universal health care, representing ∼97% of the entire Quebec population.⁷ Health information is stored in provincial databases that are linked by a unique patient's identifier. The Fichier d'inscription des personnes assurées (FIPA) records patients' demographics. Physician billing diagnosis codes, based on the International Classification of Diseases 9th and 10th revisions (ICD-9 and ICD-10 codes), are documented in the Régie de l’assurance maladie du Québec. Until 2017, the Régie de l’assurance maladie du Québec exclusively employed ICD-9 codes, transitioning to a dual usage of ICD-9 and ICD-10 codes thereafter. Information related to hospitalizations, including primary and secondary diagnoses (ICD-9 until 2006 and ICD-10 thereafter), is stored in the Maintenance et exploitation des données pour l'étude de la clientèle hospitalière (MED-ÉCHO) database, while mortality data (date and causes-ICD-9 until 2020 and ICD-10 thereafter) are maintained in the provincial vital statistics database. Data concerning the demographic distribution of the Quebec population, broken down by age and sex for all available years (1989–2020), was obtained from the Régie de l’assurance maladie du Québec and MED-ÉCHO databases.

Case definitions

As previously described in the literature using administrative data,8, 9, 10 the following case definition algorithm was used to identify a case of localized scleroderma: 1) any hospitalization that included localized scleroderma as a primary or non-primary discharge diagnosis ICD-9 (701.0) and/or ICD-10 (L94.0, L94.1); and/or 2) ≥ 2 physician (any specialty) billing claims ICD-9 (701.0) and/or ICD-10 (L94.0, L94.1) at least 2 months apart and within 2 years, and/or 3) ≥ 1 billing claim (ICD-9 or ICD-10) made by a specialist (dermatologist or rheumatologist). To minimize misclassification bias, patients who received systemic sclerosis codes (ICD-9 code 710.1 and ICD-10 code M34) at any time point were excluded.

Statistical analyses

Incidence

A conservative, 7 year-long washout period (1989–1995) was applied to remove prevalent cases, as the mean duration of localized scleroderma varies from 3 to 5 years depending on the subtype.^1,3 Year 2020 was excluded to account for case definition criteria as data from the year 2021 was not available at the time of analysis. A negative binomial random walk model was used to estimate crude incidence rate trends over time stratified by age group (0–19, 20–39, 40–59, 60–79, ≥80) and sex as previously described.¹¹ Annual (1996–2019) age-standardized incidence rates stratified by sex were calculated using negative binomial random walk models with the provincial population distribution in the corresponding year as the denominator (Supplementary Table S1). Negative binomial random walk models were used to calculate the average annual percent change in age-standardized incidence rate as recommended for overdispersed count data.^11,12 Standardized incidence ratios (by age and sex) were computed for each forward sortation area (a 3-digit postal code) from 1996 to 2019. The standardized incidence ratio compares the actual age and sex-standardized incidence rate of localized scleroderma in a specific forward sortation area to what would be expected if the age and sex-standardized incidence rate in that area were the same as that of the general population during the study period. As per Régie de l’assurance maladie du Québec/MED-ÉCHO confidentiality rules, forward sortation areas with a population of fewer than 5 individuals were not included in the analysis. All estimates were age-standardized unless otherwise noted.

Prevalence

For each year from 1996 to 2019, we defined prevalent cases as Quebec residents with a diagnosis of localized scleroderma in the past 7 years. Trends in crude prevalence rates for years 1996–2019 were estimated using flexible negative binomial random walk models for males and females separately, as well as combined.

Mortality

All-cause annual standardized mortality ratios adjusted for age and sex from 1996 to 2019 were estimated using the flexible Bayesian random walk model, with the general Quebec population as the standard population as before.¹¹ Cause specific deaths were identified from the provincial vital statistics database using the primary causes of death (ICD-9 and ICD-10 codes). All estimates were age-standardized unless otherwise noted.

Comorbidities

Localized scleroderma comorbidities were identified using validated ICD-9/10 definitions applied to billing claims and hospitalizations within a 2-year window, as detailed in the Supplementary Methods.

Geospatial mapping

A Poisson Besag-York Mollié regression model with a spatially correlated random effect and smoothing was used to model the standardized incidence ratio per forward sortation area. ArcMap 10.8.2 was used to generate maps showing the geographic distribution of the average standardized incidence ratio over the study period.¹¹

Software packages

The negative binomial generalized linear models, random walk models, and Besag-York Mollié model were fitted with the integrated nested Laplace approximation (INLA) implemented in the R-INLA package. Default non-informative prior distributions in R-INLA were used for all model parameters and hyperparameters. Standardized mortality ratios over the full study period were calculated with the epi. smr function in the epiR R package. Confidence intervals were calculated with Byar's approximation method. Further details are available in the Supplementary Material.

Role of the funding source

The study sponsors were not involved in the design of the study; the collection, analysis, and interpretation of data; the writing of the report; or the decision to submit the paper for publication.

Results

Incidence

In total, 6063 incident cases of localized scleroderma were identified between 1996 and 2019 (Table 1). Most (74.4%) were females. The mean age at diagnosis was 53.0, SD 20.2 (standard deviation, SD) years for the entire population. When stratified by sex, the mean age at diagnosis in females was 54.6, SD 19.8 years compared to 48.4, SD 20.8 years in males. Pediatric patients accounted for 7.7% of incident cases (0–19 age group) whereas 16.7% of patients were between 20 and 39 years old, 32.7% were between 40 and 59 years old, 33.9% were between 60 and 79 years old, and 7.0% were over 80 years old. The most common comorbidities found in this cohort were diabetes (9.8% including type 1 and 2), thyroid disease (7.4%), lupus (2.4%) and rheumatoid arthritis (1.3%) (Supplementary Table S2).

Table 1.

Characteristics of Quebec patients with new diagnosis of localized scleroderma between 1996 and 2019, by age and sex.

Age (years)	Male		Female		Both sexes
	n	%	n	%	n	%
0–19	176	2.9	293	4.8	469	7.7
20–39	335	5.5	675	11.1	1010	16.7
40–59	504	8.3	1479	24.4	1983	32.7
60–79	461	7.6	1718	28.3	2179	35.9
80+	77	1.3	345	5.7	422	7.0
All ages	1553	25.6	4510	74.4	6063	100.0

For both sexes combined, the incidence rate over the study period was 3.25/100,000 [95% CI 3.17–3.33/100,000] person-years (Fig. 1). As expected, mean incidence rate was higher [4.80/100,000, 95% CI 4.66–4.94/100,000] in females compared to males [1.68/100,000, 95% CI 1.60–1.76/100,000]. Among females, the incidence rate followed an inverted U shaped curve with an incidence rate of 3.13/100,000 [95% CI 2.31–4.14/100,000] person-years in 1996, reaching a peak of 5.54/100,000 [95% CI 4.79–6.36/100,000] person-years in 2004, followed by a plateau until 2013 and a decrease to 1.75/100,000 [95% CI 1.18–2.55/100,000] person-years in 2019. The average annual percent change for the incidence rate in females was −2.0 [95% CI −3.7 to −0.2]% over the study period. A steady decrease in incidence rate was observed for males over the study period, with an incidence rate of 1.78/100,000 [95% CI 1.38–2.24/100,000] person-years in 1996, and 0.79/100,000 [95% CI 0.54–1.09/100,000] person-years in 2019. Similarly to females, a more abrupt decrease was observed in 2013. The average annual percent change for the incidence rate in males was −3.3 [95% CI −5.0 to −1.8]% over the study period.

Fig. 1.

Age standardized incidence rate (ASIR) over time (1996–2019) for females, males, and both. The grey areas represent the 95% confidence intervals for the annual point estimates from the model.

For both sexes combined, the highest crude incidence rates were observed in patients aged 60–79 years (6.83/100,000 [95% CI 6.54–7.12/100,000] person-years) and the lowest were in the 0-19 years-old age group (1.11/100,000 [95% CI 1.01–1.22/100,000] person-years). For females, the highest crude incidence rate was in the 60–79 years age group (10.20/100,000 [95% CI 9.72–10.70/100,000] person-years), whereas for males it was in the 80+ years age group (3.26/100,000 [95% CI 2.57–4.07/100,000] person-years). In female patients, a more pronounced decrease in the crude incidence rate over time was observed in the 40–59, 60–79, and 80+ age groups compared to those aged <40 years (Fig. 2A). A similar trend was seen in males aged 40–59 years, whereas males in all other age groups had a steady decrease in the crude incidence rate across the years of study (Fig. 2B).

Fig. 2.

Crude incidence rate over time (1996–2019) per age group (0–19, 20–39, 40–59, 60–79, 80+) for females (A) and males (B). The grey areas represent the 95% confidence intervals for the annual point estimates from the model.

Prevalence

The average prevalence of localized scleroderma over the study period for both sexes combined was 24.5/100,000 [95% CI 24.3–24.8/100,000] persons. An increase in this prevalence was observed across the years of study with an average annual percent change of 2.6 [95% CI 2.3–3.0]% (Fig. 3).

Fig. 3.

Prevalence rates over time (1996–2019) for females, males, and both. The peak prevalence for females was 47.1/100,000 persons with a 3.7 [95% Confidence Interval, CI 3.4–4.2]% average annual percent increase. The peak prevalence for males was 13.6/100,000 persons with a 0.4 [95% CI −0.04 to 0.9]% average annual percent increase. The grey areas represent the 95% CI for the annual point estimates from the model. The points around the curve represent the true observed values for each year.

As expected, prevalence of localized scleroderma over the study period was higher in females [36.1/100,000, 95% CI 35.72–36.5] compared to males [12.8/100,000, 95% CI 12.6–13.0]. Over the study period, females had an increasing trend in prevalence (average annual percent change 3.7 [95% CI 3.4–4.2]%) with a peak of 47.1/100,000 [95% CI 44.7–49.1] in 2017, whereas males had a relatively stable prevalence (average annual percent change 0.4 [95% CI −0.04 to 0.9]%) over the study period (Fig. 3).

The highest prevalence for both sexes combined was similarly observed in the 60–79 years old group (49.4/100,000 [95% CI 48.7–50.2/100,000] persons). The same age group (60–79 years) had the highest prevalence for females (75.9/100,000 [95% CI 75.6–77.2/100,000] persons), and males (20.0/100,000 [95% CI 19.3–20.7/100,000] persons).

Mortality

A total of 680 deaths were reported among study individuals (11.0%) over the study period, mainly in females (75.0% of deaths). The mean age at death was 77.6, SD 12.1 years (73.0, SD 12.7 for females and 79.1, SD 11.5 for males). The most common causes of death in localized scleroderma patients included, respectively, diseases of the circulatory system (39.7%), tumors (35.3%), psychological and behavioral disturbances (7.4%), diseases of the nervous system (5.9%), diseases of the respiratory system (5.9%), infectious and parasitic diseases (2.9%), endocrine/nutritional/metabolic diseases (1.5%) and diseases of the gastrointestinal system (1.5%) (Fig. 4).

Fig. 4.

Number of deaths per cause of death among study individuals.

Age and sex-standardized mortality ratio estimates suggest that overall mortality risk in localized scleroderma patients (males and females combined) is compatible with that of the general population (mortality ratio 1.07 [95% CI 0.99–1.15], p = 0.104). For males, the mortality risk ranges from 2% lower to 33% higher (mortality ratio 1.14 [95% CI 0.98–1.33], p = 0.094), indicating a possible clinically significant elevation. For females, the risk ranges from 2% lower to 15% higher (mortality ratio 1.07, 95% CI 0.98–1.15), suggesting a clinically moderate upper bound.

Flexible Bayesian random walk models indicate a declining mortality risk over time. For males and females combined, the mortality ratio decreased from 1.31 [95% CI 1.06–1.58] in 1996 to 0.81 [95% CI 0.66–0.98] in 2019 (Fig. 5). Similar trends over time were observed for both sexes with the highest mortality ratio of 1.34 [95% CI 1.07–1.65] in females and 1.29 [95% CI 0.86–1.83] in males in 1996 (Fig. 5) and the lowest mortality ratio of 0.78 [95% CI 0.62–0.96] in females and 0.97 [95% CI 0.67–1.35] in males in 2019 (Fig. 5). These findings suggest that relative mortality risk has approached or fallen below that of the general population in recent years.

Fig. 5.

Trends in standardized mortality ratio (SMR), standardized by age and sex, over time (1996–2019) for females and males. The grey areas represent the 95% confidence intervals. The Y axis is a logarithmic scale.

Causes of death before and after 2006 were non-differential (more granular analysis was not feasible due to the sample size).

When looking by age group, an excess death was noted compared to the general population for 40–59-year-old females only (mortality ratio over the study period of 1.49 [95% CI 1.05–2.06], p = 0.026) (Fig. 6). Unfortunately, due to sample size and mandatory data rounding, further in-depth analysis of causes of death was not performed.

Fig. 6.

Standardized mortality ratio (SMR) over the study period per age group and sex. The lines for each point represent the 95% confidence interval. The Y axis is a logarithmic scale.

Geographic distribution of localized scleroderma in Quebec

A total of 401 forward sortation areas in Quebec were included. Geospatial analysis revealed an uneven geographic distribution of age and sex-standardized incidence ratios over Quebec with clustered hotspots in the south part of the province (Fig. 7). The average standardized incidence ratio with spatial modeling spanned between 0.22 and 4.80. The top 10 and lower 10 forward sortation areas based on incidence rate are shown in Supplementary Table S3.

Fig. 7.

A. Geographic distribution of standardized incidence ratios in Quebec (1996–2019). B. Clustered hot spots in Quebec (1996–2019).

Discussion

This is the first study investigating the epidemiology of localized scleroderma in Canada and to this level of detail globally. Using Quebec populational data, we studied localized scleroderma incidence, prevalence and mortality by age and sex from 1996 to 2019 as well as assessed temporal and spatial trends.

Over the study period, the incidence rate of localized scleroderma was 3.25/100,000 [95% CI 3.17–3.33/100,000] person-years. While this is slightly higher than previously published estimates from the USA, Sweden, Denmark, Botswana, and the UK, (0.16–2.7/100,000 people-years), these studies were conducted prior to the year 2015.³ For comparison, in our study, the incidence rate in 2015 was 2.75/100,000 [95% CI 2.39–3.18/100,000]. In a recent study conducted by our group, the incidence rate for systemic sclerosis for the same study period was slightly higher at 4.14/100,000 [95% CI 4.05–4.24/100,000] person-years with a 4:1 female predominance.¹¹ Similar to other autoimmune diseases including systemic sclerosis, localized scleroderma incidence in our cohort was found to be higher in females compared to males in a ∼3:1 ratio, which is consistent with localized scleroderma literature.^1,13,14 Cutaneous autoimmune diseases are generally female predominant and this predilection may be explained by several factors including genetics, epigenetics, hormonal, anatomical, and sociocultural differences between males and females, as well as men and women.¹⁴

To our knowledge, this is an initial study to investigate incidence of localized scleroderma separately by age groups. Similar to systemic sclerosis, the highest crude mean incidence rate in localized scleroderma was for ages 60–79 in females (10.20/100,000 [95% CI 9.72–10.70/100,000]) person-years) and over 80 in males (3.26/100,000 [95% CI 2.57–4.07/100,000]) person-years).¹¹ The mean age of our population was 53.0 years, which is higher than the mean age of other cohorts (45.3 years),¹⁵ possibly due to the population-based nature of our research. In our recent systematic review, we did not identify other population-based studies that included individuals of all ages and provided the mean age for their population.¹⁵ As population-based studies are inherently different from registry-based cohorts, it is difficult to compare our results to those previously reported. Some factors that may explain the older mean age of our population include inclusion of all localized scleroderma cases regardless of severity, subtype or treating physician specialty (whereas registry data may be biased towards more severe cases), and shifting demographics of localized scleroderma (more immunotherapy-induced cases in the last years). However, whether our findings are specific to the Quebec population requires further research.

Indeed, aging is an independent risk factor for the development of immune dysregulation. Aging of the immune system or immunosenescence is characterized by increased genotoxic stress signals (e.g., double stranded DNA breaks leading to mutations and neoantigens formation) and inflammaging (e.g., increased levels of proinflammatory cytokines, pro-fibrotic/angiogenic growth factors, extracellular matrix proteins).^16,17 While this hasn't been studied in localized scleroderma, Gniadecki et al. recently demonstrated genomic instability, somatic mutations and “senescence” signature in the skin samples of early systemic sclerosis patients.¹⁸ Interestingly, external factors (e.g., pollution, radiation, chemo/immunotherapy) may induce immunosenescence and inflammaging which could act as triggers in localized scleroderma and systemic sclerosis.¹⁹ Thereby, age specific epidemiology and immunosenescence in localized scleroderma should be explored in future studies.

Children (0–19 years old) represented ∼ 7.7% of all incident localized scleroderma cases with a mean incidence rate (both sexes combined) of 1.11/100,000 [95% CI 1.01–1.22/100,000] person-years. So far, the incidence of pediatric localized scleroderma was estimated in a single UK/Ireland study conducted from 2005 to 2007, suggesting an incidence rate of 0.34/100,000 children-year.²⁰ Case identification relied on physicians reporting localized scleroderma cases through surveys and cases with insufficient information on the report card were excluded (98 of 185 cases), leading to probable underestimation of the true incidence.²⁰

Studying the temporal trends in incidence over the study period, we documented an initial increase in incidence rate in Quebec from 1996 to 2004, followed by a plateau until 2013 and a decrease from 2013 to 2019, specifically more pronounced after 2017 (average annual percent change of −2.4 [95% CI −3.7 to −1.2]% over the study period). Only one study (82 localized scleroderma incident cases diagnosed over a 33 year period) assessed longitudinal trends in localized scleroderma incidence over time and documented an average increase in incidence of 3.6% per year from 1960 to 1993.²¹ While the cause of this increase is not known, a general increase in the incidence of autoimmune diseases is well documented and is likely multifactorial accounting for improved diagnosis and awareness (particularly of earlier stages and milder phenotypes), population aging and extrinsic factors contributing to autoimmune response elicitation.^11,22 The cause of the apparent drop in localized scleroderma incidence rate in Quebec after 2013 and particularly after 2017 is unclear; however, it may be a consequence of better disease recognition in the preceding years and possibly due to changes in the Quebec billing system. Specifically, physician billing procedures gradually switched from handwritten paper slips to an electronic system (switch completed in 2017). While no changes occurred to diagnostic codes or their requirements, a significant drop in the global reporting of diagnostic codes was noted by the Régie de l’assurance maladie du Québec following this transition.²³ Of interest, similar incidence trends overtime were reported in other autoimmune skin diseases, notably psoriasis, where incidence decreased by 20% between 1990 and 2019 worldwide, and most abruptly after 2014 (average annual percent change 1990–2019 of −0.9% [95% CI −1.0 to −0.9]%).^24,25

This is the third study studying the prevalence of localized scleroderma. The average prevalence for both sexes combined was 24.5/100,000 [95% CI 24.3–24.8/100,000] persons with an increasing trend overtime (mean average annual percent change of 2.6 [95% CI 2.3–3.0]%) for females in particular. This is similar to the trends observed in other autoimmune skin diseases such as systemic sclerosis.^11,26 As expected, higher prevalence was seen in females compared to males (∼3:1 ratio) and in the 60–79 year-old age groups. In children, the mean prevalence was 9.5/100,000 [95% CI 9.2–9.8/100,000] children. While our estimates are lower than the study of pediatric localized scleroderma prevalence in the US (2010–2014, 32–36/100,000 children),²⁷ several methodological differences are worth noting. The Beukelman et al. study used Truven MarketScan database (employed-based health insurance, non-populational) and defined prevalent localized scleroderma as any single occurrence ICD-9 code 7010 by any physician.

Given the lack of existing mortality data in localized scleroderma, our study provides initial and reassuring evidence that all-cause mortality in localized scleroderma is overall similar to that of the general population (mortality ratio 1.07 [95% CI 0.99–1.15] for both sexes combined). This mortality ratio is lower than that found during the same study period for systemic sclerosis in Quebec (3.31 [95% CI 3.18–3.45]).¹¹ While an initial ∼30% excess all-cause mortality (mortality ratio 1.31 [95% CI 1.06–1.58]) was noted for both sexes in 1996, it normalized in 2005 and remained comparable to the general population thereafter (mortality ratio of 0.78 [95% CI 0.62–0.96] in females and 0.97 [95% CI 0.67–1.35] in males in 2019). This temporal trend was seen for both ages and sex groups except for females aged 40–59 years where a ∼50% excess mortality (mortality ratio 1.49 [95% CI 1.05–2.06]) was noted throughout the study period. Overall these results confirm the clinical suspicion that for most patients, localized scleroderma is a morbid condition with associated functional and quality of life impairment, but it is not life-threatening.²⁶ The cause of excess deaths prior to 2006 is not clear, but may be attributed to disease recognition and management overtime. Of note, prior to 2000, the most commonly employed treatment options for severe localized scleroderma were penicillamine, diphenylhydantoin, potassium p-aminobenzoate, hydroxychloroquine and cyclosporine, options that are currently seldom used.²⁸ The increased risk of death among the 40-59 year-old female subgroup is intriguing and deserves further investigation, particularly with adjustment for possible confounders such as comorbidities (e.g., cancer) and their treatments. Overall the leading causes of death in our localized scleroderma cohort were concordant with the top causes of mortality in the Quebec general population, namely diseases of the circulatory system (39.7%) and tumors (35.3%), and remained stable over time (before and after 2006).²⁹

Finally, we document an uneven geographic distribution of localized scleroderma incidence across Quebec, with clustered hotspots in the south reaching 4 times higher incidence rates than Quebec's average. While this is an initial study exploring the geospatial epidemiology of localized scleroderma on a jurisdiction level, similar findings were recently documented in systemic sclerosis and other autoimmune diseases.^11,22 Anecdotal reports of localized scleroderma occurring in married couples³⁰ as well as geographic clusters of localized scleroderma and systemic sclerosis cases near London (UK) airports,³¹ and Rome (Italy) boroughs were described.³² Further research is necessary to identify potential external risk factors associated with the increased incidence of localized scleroderma in these geographic areas.

Our study should be interpreted within its characteristics (populational study using administrative datasets). As we relied on ICD-9/10 codes for case identification, the risk of misclassification and/or missed cases is possible. To mitigate this concern, we adapted the previously validated case definition algorithm from other autoimmune diseases such as systemic sclerosis (including Quebec) which has been shown to have a high sensitivity and specificity.¹¹ The variability in physician billing/coding practices may have influenced trends over time, which was not possible to account for. Additional clinically important details such as localized scleroderma disease subtype, clinical characteristics, or individuals’ race/socioeconomic status are also lacking in the dataset available. As it is a populational study, assessing individual risk factors was not feasible. Another possible limitation of the study design and analysis included model misspecification bias, which we aimed to decrease by using comprehensive population-based data and accounting for demographic and regional variations in localized scleroderma.

In conclusion, in this study of localized scleroderma epidemiology in Quebec, females were noted to have an initial increase in localized scleroderma incidence from 1996 to 2004, followed by a plateau and a decrease after 2013, while males were shown to have a steady decrease in incidence over the study period. Prevalence increased steadily from 1996 to 2019. Mortality ratio analysis revealed that excess death occurred only in females aged 40–59 years, which needs further investigations. Localized scleroderma incidence varied geographically with hotspots in the South of the province, highlighting the importance of additional research on environmental risk factors.

Contributors

SG contributed to conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing–original draft, writing–review & editing. AM contributed to conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing–original draft, writing–review & editing. KA contributed to data curation, writing–review & editing. MK contributed to data curation, formal analysis, investigation, methodology, review & editing. LK contributed to review & editing. MP contributed to methodology, review & editing. CM contributed to methodology, review & editing. GC contributed to review & editing. ER contributed to conceptualization, methodology, writing–review & editing. MO contributed to conceptualization, validation, writing–review & editing. JS contributed to review & editing. JB contributed to review & editing. EN contributed to conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing–original draft, writing–review & editing, and the final decision to submit the manuscript.

Data sharing statement

The data used in this study originates from the © Government of Quebec (2022). All data access and usage were in accordance with © Government of Quebec guidelines, and all necessary precautions were taken to ensure the confidentiality and privacy of the information. Data obtained is used exclusively for the purpose specified in the authorization. Any other use requires additional permission. Hence, if a data sharing request is submitted to the study authors (EN), it will be subject to the © Government of Quebec data sharing permission request.

Editorial disclaimer

The Lancet Group takes a neutral position with respect to territorial claims in published maps and institutional affiliations.

Declaration of interests

Dr. Catherine C. McCuaig discloses honoraria from Pfizer, Sanofi, and Journées de Dermato Péd CHUSJ. She received travel support from Pierre Fabre and served on advisory boards for Bausch and Johnson & Johnson. She was also a past co-chair of the Canadian Lymphedema Framework. No other conflicts of interest relevant to this work were reported.