Abstract
Background:
The epidemiology of non-malignant—as opposed to malignant—skin diseases is frequently overlooked. Contemporary literature suggests there is a high economic burden and risk of health complications of non-malignant skin diseases, which warrants further study. In contrast to the United States, Canada is particularly devoid of research in this area.
Objective:
This study aimed to explore the national trends in the mortality rates for non-malignant skin diseases (ICD L00-L98) in the Canadian context from 2000 to 2023.
Methods:
A population-level ecological analysis was conducted using publicly available mortality data from Statistics Canada’s Vital Statistics—Death Database. Crude and age-adjusted mortality rates were calculated, and a subsequent Joinpoint regression analysis identified significant inflection points in temporal trends. Annual percent change (APC) was also computed to quantify and delineate fluctuations over time.
Results:
Across the aggregate of non-malignant skin conditions, there was a net rise in the mortality rates from 2000 (1.00/100 000) to 2023 (2.50/100 000). An inflection point occurred in 2006; the mortality rates shifted from an initial non-significant decline (APC = −3.84%;
Conclusion:
This study examined the trends in mortality rates of non-malignant dermatological mortality trends in the Canadian context, noting an overall increase in mortality rates over time. These findings underscore the significance of mortality attributable to non-neoplastic skin diseases with regard to clinical care and future policy interventions.
Keywords
Introduction
When compared to skin malignancies, such as melanoma, non-malignant skin conditions frequently get overlooked in contemporary dermatological research. For instance, cellulitis and toxic epidermal necrolysis (TEN) can have severe outcomes, including death, secondary to their systemic complications.1,2 In addition to contributing to comorbidities and death, chronic non-malignant skin diseases have consequential impacts on the societal and institutional level. For instance, non-malignant skin conditions lead to rising healthcare costs because of their recurring and persistent nature. In the United States (US), the financial burden of skin disease was ~35.9 billion USD in 1997, and 75 billion USD by 2013.3,4 In Canada, by 2013, non-malignant skin diseases contributed to ~2.1 billion CAD. 5 The economic impact and inpatient admission trends are underscored in population-based analyses in Ontario, Canada.6,7 Evidence supporting the financial strain from non-malignant skin conditions accentuates the need to further study underlying factors that may be driving these trends.
In addition to economic burden, skin diseases are associated with profound psychological distress and diminishing social function. The mere occurrence of certain skin diseases, such as psoriasis, is positively correlated with depression, social isolation, loneliness, and an overall diminished quality of life.8,9 Ultimately, while the economic and psychological burdens are well-documented in contemporary literature, a gap exists surrounding overall mortality rates.
Mortality and morbidity are crucial markers that allow for the identification of risk factors and quantitative comparisons across groups. 10 In the context of the US, only 2 studies explored the mortality burden of non-neoplastic skin diseases using the Centers for Disease Control and Prevention Wonder dataset.11,12 However, there is a scarcity of research examining the epidemiology of non-cancerous dermatological conditions outside of the US. Further investigation is warranted in Canada, given the paucity of data surrounding the epidemiology of non-malignant skin conditions.
This study attempts to address a major gap in the literature: that is, the scarcity of studies that assess the trends in the annual mortality rates of non-malignant skin conditions in a Canadian context. In this paper, we aim to provide the first comprehensive national assessment of long-term mortality trends in non-malignant dermatological diseases. In line with similar analyses in the US, the authors of this study hypothesized that there would be an overall increase in the mortality rates from 2000 to 2023, with disproportionately greater representation by decubitus ulcers and cellulitis.11,12
Materials and Methods
A population-level ecological analysis was conducted using the publicly available data obtained from the Vital Statistics—Death Database (VSDD) provided by Statistics Canada.
13
The use of publicly available deidentified data ensured that our study complies with patient privacy laws. Institutional ethics approval was not required for this study. The retrieved data represented the mortality counts of all non-malignant skin and subcutaneous diseases (ICD L00-L98) across Canada from 2000 to 2023.
14
The study population included all available mortality counts based on the Statistics Canada dataset; thus, no formal sample size calculations were necessary. Missing data were not reported, and sensitivity analyses were not applicable because of reliance on modeled aggregates. Crude mortality rates were calculated by dividing the mortality counts by the year population data from the “
Joinpoint statistical software (developed by the Statistical Research and Applications Branch of the National Cancer Institute, USA), was used for analysis of the collected data.17-19 A Joinpoint regression—which uses a piecewise regression modeling fluctuations over time—was employed to identify deviations in the temporal mortality rate trends. 20 The standard error mean was calculated to generate a Joinpoint regression analysis. Annual percent change (APC) was calculated to identify significant inflection points.
Results
In the Canadian mortality trends for all skin and subcutaneous diseases (ICD-10 L00-L98), an overall increase was noted from 1.00/100 000 in 2000 to 2.50/100 000 in 2023 (Figure S2). The crude mortality trends for all skin and subcutaneous diseases transitioned from a slight decline between 2000 and 2006 (APC = −1.38%; 95% CI −8.59 to 2.14;
As seen in Figure S3, this trend was reflected in both the age-adjusted and crude rates in 2 sub-categories: bullous disorders (L10-L13) and other disorders of the skin (L80-L98), the latter of which included decubitus ulcers. A Joinpoint regression of the crude rate trends and the age-adjusted rate trends demonstrated an inflection point in 2008 for these 2 sub-groups. There were discrepancies in the skin infections when examining the crude rate and age-adjusted data: With regards to bullous disorders, while the initial decline was non-significant for both crude rates (APC = −3.57%; 95% CI −22.7 to 2.78;
As seen in Figures S3 and S4, the V-shaped trend observed with bullous disorders was not found for annual mortality rates of infections of the skin and subcutaneous tissue (L00-L08). There were no Joinpoints computed for the crude mortality data (APC = 6.72%; 95% CI 6.01-7.74;
As depicted in Figure S6, the age-adjusted mortality rates were also delineated by gender. There was 1 inflection point in 2006 for the mortality rates in males, and 2 inflection points—in 2009 and 2015—for the mortality rates in females. With regard to the male population, there was an initial decline until 2006 that was not statistically significant, followed by a sharp increase from 2006 to 2023 (APC = 4.46%; 95% CI 3.79-5.89;
Discussion
This analysis investigated the temporal trends of mortality rates pertaining to non-malignant skin diseases (ICD L00-L98) over the last 2 decades in Canada. There was initially an overall decline in the aggregated mortality rates for all non-malignant skin diseases until 2006, followed by an increase until 2023. Furthermore, 2008 served as a significant inflection point for bullous disorders and “other skin diseases,” from an overall decline to a subsequent increase. In comparison, a notable transition occurred in 2021 for skin infections, where the increase in the annual mortality rates became significantly sharper. Within the category of all non-malignant skin conditions, bullous disorders, “other” skin diseases, and skin infections were strongly associated with increased annual mortality in Canada. Crucially, the “other” skin disease category includes numerous relevant skin conditions—such as pressure ulcers and pyoderma gangrenosum—with the potential to contribute to mortality. In addition, the results suggest mortality rates did not vary substantially by gender. However, the age-adjusted mortality rate trend for males was stable until 2006, and then took a steady upturn until 2023. On the other hand, for females, the increase was nearly twice as sharp overall when compared to males and only occurred from 2009 to 2015. Following this, for females, from 2015 to 2023, there were no significant fluctuations in the mortality rates.
There are numerous clinical implications of the aforementioned findings. Firstly, the high mortality rates of the “other” skin disease are supported by the hypothesis of this study; namely, that the increase in mortality was likely in part driven in part by decubitus ulcers, paralleling data from the US.11,12 One interesting trend was that mortality rates for urticaria and erythema (L50-L53), under which Stevens-Johnson syndrome (SJS) and TEN are listed, were low or negligible in most years. This is of interest as severe drug reactions are among the most life-threatening dermatological conditions. Similarly, low or negligible mortality rates were observed for the dermatitis and eczema category (L20-L30), under which Drug reaction with eosinophilia and systemic symptoms (DRESS) syndrome falls. These findings underscore a greater need to direct greater clinical attention toward the over-represented non-malignant skin conditions, including bullous diseases and skin infections, as opposed to those that may be conventionally construed to acutely cause mortality, including SJS and TEN. Specifically, bullous disorders (including pemphigus vulgaris), cutaneous infections (including cellulitis), and “other” diseases (including decubitus ulcers) comprise the clusters of conditions that have shown the greatest mortality increases. While the relative acuity, diagnostic workup, and management of these conditions vary greatly, clinicians would be prudent to consider that these conditions can be potentially life-threatening. Involving appropriate consultant physicians early in the clinical course, such as dermatologists, rheumatologists, or infectious disease specialists, will be important for improving outcomes.
Changes in the policy and healthcare-levels may have also exacerbated the burden of non-malignant skin diseases. While the ecological design of this study precludes causal inferences from being drawn, we postulate on these findings in the context of several trends. For example, in 2004, the Canadian government introduced the
Antimicrobial resistance may have contributed to the rising mortality trends, particularly given the observed inflection point in 2006. Prior literature supports evolving antimicrobial resistance patterns in Canada in recent years. 22 Although there was an overall decline in the proportion of total Methicillin-resistant Staphylococcus aureus (MRSA) cases in Canada from 2007 to 2016, community-associated MRSA strain types—a leading contributor of recurring skin and soft tissue infections—have increased significantly. 23 This finding may be consistent with the upward rise in mortality rates for aggregated skin diseases and skin infections. In 2009, a coordinated national Pan-Canadian approach to antimicrobial stewardship was outlined, emphasizing coordinated action, education, and surveillance. 22 However, stewardship efforts continued to vary considerably across provinces. 22 Furthermore, increased rates of antibiotic resistance, particularly for Gram-negative and positive organisms, were seen with the SARS-CoV-2 (COVID-19) pandemic.24,25 A greater susceptibility toward infectious skin diseases may ultimately have been due to increased antibiotic resistance and the dysregulated immune responses that are typical of preexisting inflammatory conditions—such as atopic dermatitis and psoriasis. 26 Although we cannot definitively draw causal linkages, future research should investigate the possible association between antimicrobial resistance and cutaneous mortality trends in Canada.
The rising mortality trends in Canada may also be attributed to increasing overall disease prevalence and an aging population predisposing individuals to severe systemic involvement. Between 1990 and 2017, the prevalence of psoriasis and decubitus ulcers in Canada increased considerably (17% and 7%, respectively), which was concurrent with an 18% increase in the overall death rate of all skin diseases, including malignant skin conditions. 27 Although most cases of skin diseases may not directly result in mortality, severe instances may be associated with a higher risk of death due to systemic complications. For example, severe psoriasis can be associated with a higher risk of cardiovascular diseases, and decubitus ulcers can precipitate sepsis.28,29 Moreover, comorbidities that result from an aging population in Canada may contribute to heterogeneity despite standardization, which complicates uniform comparisons. 30 One notable age-related factor is frailty, a key predictor of poorer health outcomes such as hospitalization and mortality. 31 Therefore, solely relying on age-adjusted mortality rate trends may be limiting, as it may obscure the effects of other comorbidities associated with a disproportionately larger aged population in Canada. Future studies should take a more holistic approach toward assessing the trends in skin disease-induced mortality, including factors such as frailty, prevalence of diseases, and other comorbidities.
There is considerable research on the epidemiology of malignant skin disease, notably focusing on the overall mortality rates globally and within specific regions, often combining results with those of skin cancers.32-35 One such study was conducted in the Canadian context, which assessed mortality rates and metrics of morbidity for diseases such as melanoma, keratinocyte carcinoma, cellulitis, and decubitus ulcers. 27 Furthermore, a small denomination of studies have examined the mortality rate trends of non-malignant skin conditions in the US.11,12,36 Nonetheless, there are no studies to our knowledge that have fully encompassed all the “non-malignant” skin conditions in the Canadian context, as was done in this study. Thus, we attempted to address this gap in the literature by delineating the trends in mortality rates of non-malignant skin conditions in Canada.
This study has several areas of limitations where future studies can expand. First, the data obtained from Statistics Canada may considerably under-represent or misattribute the role of dermatological conditions in contributing to mortality. For instance, in a patient with concurrent SJS and sepsis, it may be challenging to attribute the cause of mortality to either condition. Because of attribution bias, the cause of death from non-malignant dermatological conditions may be misattributed to other etiologies. Second, this study does not capture the degree of severity of skin diseases, which may be contributing to all-cause mortality. Our study design also captures changes in broad categories of non-malignant dermatological conditions; as such, ascribing mortality rate changes to particular conditions within a category is challenging.
There are also several limitations that stem from the use of the ICD-10 categorization scheme. The ICD-10 frequently groups conditions that may have drastically different mortality risks, which can make assessing category data in aggregate misleading. This is especially the case with the heterogeneous category of “Other Skin Diseases,” which includes L81 (disorders of pigmentation) and L82 (seborrheic keratosis) with L88 (pyoderma gangrenosum) and L89 (decubitus ulcer and pressure area). The overall increase in the “other skin conditions” category that was witnessed in this study was more likely attributed to the latter 2 diseases than the former, thus diluting the interpretation. As well, the use of categorization schemes allows for misclassification bias. 37
Lastly, a noted limitation was with Canada’s aging population, which introduces heterogeneity due to comorbidities and frailty that are not fully captured by age-adjusted mortality rates. Statistics Canada does not provide sufficient granularity to note the association of frailty with the mortality rates within Canada. However, a key strength of Statistics Canada lies in its generalizability: it spans all provinces and territories through the Vital Statistics Council. Future research should strive to fill this gap in the literature by focusing on the impact of frailty and disease severity on skin disease mortality within Canada.
Conclusion
The present study provides a national assessment of the temporal trends of non-malignant skin conditions within Canada. Using Statistics Canada data, this study demonstrated a significant and sharp increase in the aggregate mortality rates from 2000 to 2023. Furthermore, there were notable inflection points for subcategories. Relevant policy-level impacts of the findings are discussed. Additional demographic parameters and their potential correlation with mortality trends should be investigated in future studies. As well, further efforts should explore the cause of death reporting standards to better delineate any discrepancies. Finally, more controlled designs should be implemented to pinpoint the underlying cause of disease, in addition to highlighting any biases.
Supplemental Material
sj-docx-1-cms-10.1177_12034754261431833 – Supplemental material for Mortality Trends of Non-Malignant Skin Diseases in Canada: A Statistics Canada Analysis, 2000 to 2023
Supplemental material, sj-docx-1-cms-10.1177_12034754261431833 for Mortality Trends of Non-Malignant Skin Diseases in Canada: A Statistics Canada Analysis, 2000 to 2023 by Nirav Saini, Ivy Mushamiri, Jason Kreutz, Raed Alhusayen and Fatemeh Jafarian in Journal of Cutaneous Medicine and Surgery
Footnotes
Author Contributions
Nirav Saini was involved with conceptualization (lead), data curation (lead), formal analysis (lead), investigation (lead), methodology (lead), project administration (lead), visualization (lead), writing—original draft (lead), writing—review & editing (lead). Dr Ivy Mushamiri was involved in formal analysis (supporting) and methodology (supporting). Jason Kreutz was involved with writing—original draft (supporting). Dr Raed Alhusayen was involved with supervision (equal) and writing—review and editing (equal). Dr Fatemeh Jafarian was involved with conceptualization (lead), supervision (lead), writing—original draft (lead), writing—review and editing (lead).
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Supplemental Material
Supplemental material for this article is available online.
