Read Aloud the Text Content

This audio was created by Woord's Text to Speech service by content creators from all around the world.

Text Content or SSML code:

4.3. Age and gender The average age at the time of diagnosis (46.0 years of age) and the fact that men are younger than women at the time of diagnosis are in line with several other reports [3,5,8,27]. However, in contrast to previous studies, the incidence among women does not display a second peak nor does it exceed the incidence of men at older age [3,8,27–29]. The vast majority of women diagnosed at a late age do not receive therapy, suggesting a milder course of disease. However, it is notable that the incidence of women receiving SC continuously in the three years following diagnosis shows a steady increase as age advances. This might indicate that the burden of severe/chronic sarcoidosis is more substantial in women diagnosed at an older age (Supplementary 4). Distribution of gender among patients with sarcoidosis varies between studies and geographical regions. Two large studies from Asia found a predominance of female cases [5,7] whereas studies from the UK and USA found no difference between genders [26,30]. The present study finds a slightly higher share of male cases (56%), which is in line with a recent study that found 55% male cases based on nationwide data from the Swedish National Patient Registry [8]. On the contrary, several Nordic studies from 1950 to 1982 – before the establishment of the National Patient Registries and when citizens were offered routine chest x-ray screening – found a predominance of female cases [31]. Although these regional and age-associated differences between genders have been observed for decades, the causes are still unclear. Differences in occupational exposure, genetics and hormones have been suggested as possible contributors. However, no conclusion has been reached [24,32,33]. 4.4. Regional differences Regional differences in incidence and prevalence within countries have been reported in previous studies [34,35]. A population-based Swedish study [8] found the highest prevalence of sarcoidosis in less density populated areas, which is in contrast to our findings of the lowest incidence in the Capital Region, which is by far the densest populated region in Denmark. Furthermore, the regional differences in IP is not reflected by differences in regional SC treatment since the share of non-treated cases show little variation between regions (51–56%), (Supplementary 3). A recent population-based Swedish study found vast regional differences in dispensed SC ranging from 30 to 50% of cases however, they did not find any consistency in the regional differences in treatment and prevalence [8,36]. The regional differences in incidence observed in the present study however, corresponds well with Byg et al. [3], who also found the lowest incidence in the eastern part of Denmark (approximately corresponding to the Capital Region and Region Zealand). The explanation for these regional differences, that seem to have persisted for at least four decades, is unclear. Denmark has universal tax-funded healthcare with equal access for all citizens, and the socioeconomic and ethnic diversity is small. Differences in unmeasured factors, such as environmental, occupational and social exposures and genetic clustering between the western part of Denmark and the Capital Region might explain some of the regional differences, however this is yet to be explored. persisted for at least four decades, is unclear. Denmark has universal tax-funded healthcare with equal access for all citizens, and the socioeconomic and ethnic diversity is small. Differences in unmeasured factors, such as environmental, occupational and social exposures and genetic clustering between the western part of Denmark and the Capital Region might explain some of the regional differences, however this is yet to be explored. PET is rare and does not seem to correlate with incidence; the Southern Region only performs PET in 0.3% of cases, as compared to 6.3% in the Capital Region. The impact of diagnostic approach on incidence is supported by Byg et al. [3], who found the highest incidence of sarcoidosis in the region with the most biopsies performed (48%), and in the region with the lowest incidence only 25% of cases had a biopsy. The non-treatment case group tend to have fewer diagnostic procedures performed than patients who are treated within the first three years. The + MTX/AZA treatment group has a particularly high share of diagnostic procedures performed, which might indicate more severe/ complicated disease at the time of diagnosis in this subgroup of patients (Fig. 6). 4.6. Strengths, limitations and generalizability Sarcoidosis is not validated in the DNPR, thus to reduce the risk of misclassification, cases were restricted to having a minimum of two sarcoidosis-related visits. However, we cannot rule out that a small share of individuals with only one visit did actually have sarcoidosis but were discontinued, which would tend to underestimate the incidence/prevalence of sarcoidosis. Furthermore, to address the potential misdiagnosis of cancer and sarcoidosis or sarcoid-like reaction, cases with a concurrent diagnosis of cancer and sarcoidosis were excluded. Cancer diagnoses have a high PPV in the DNPR thus, thus we expect misclassification of cancer to be minimal [18]. We cannot rule out, that cancer and sarcoidosis did occur concurrent in a small share of cases, which would tend to underestimate the incidence and prevalence of sarcoidosis. Regional differences in coding of diagnostic procedures may tend to both over-estimate and under-estimate the share of diagnostic procedures. Furthermore, we cannot rule out that some diagnostic procedures were performed earlier than three months prior to diagnosis, which would tend to underestimate the share of diagnostic procedures. The DRMPS provides information on collected medication, except for medication dispensed at hospitals, which could tend to underestimate the extent of treatment in some patients. On the contrary, we do not know if the collected drugs were ingested, which would lead to an overestimation of SC treatment in other patients. Furthermore, we have no information on the indications for treatment, nor if SC was prescribed but not collected, which could also lead to misclassification within the treatment groups. The population-based study design, within a tax-funded, uniformly organised health care system and virtually complete follow-up on all cases, reduces the risk of selection bias [37]. Importantly, most cases of sarcoidosis are diagnosed in hospital settings thus, we expect our data to be largely representative of the entire Danish sarcoidosis population and generalizable to populations with similar demographics and standards of medical care as those in Northern Europe. 5. Conclusion and future implications We find an increasing incidence and prevalence of sarcoidosis from 2001 to 2015. Our results suggest that the increase in incidence is widely seen among the milder cases (non-treatment group) and, put in a contemporary context, this is likely to be explained by an overall increase in diagnostic procedures. One quarter of cases are diagnosed at the age of 30–39 years. However, the age-associated incidence is markedly higher in men than women aged 30–69 years. Incidence seems to correlate well with the number of diagnostic procedures performed and with disease severity. Regional differences in incidence are observed and further research should focus on investigating risk factors, such as genetics and environmental factors, which potentially contribute to these regional differences.