A population-based projection of psoriatic arthritis in Germany until 2050: analysis of national statutory health insurance data of 65 million German population

Psoriatic arthritis (PsA) is an inflammatory disorder that most often occurs after skin psoriasis and typically involves the peripheral joints, entheses, or axis, which can potentially lead to chronic pain and impaired functioning [1‐3]. For patients, there is a risk of permanent damage, deformity, and disability, which can all impact health-related quality of life and disability-adjusted life years and are each associated with increased social healthcare expenses [3, 4]. However, PsA can be diagnosed and treated early, thus delaying the progression of the disease, if awareness is raised among those with risk factors for musculoskeletal involvement and if they are persuaded to seek early help from a rheumatologist [5].

PsA is a chronic inflammatory disease that occurs more rarely than other non-communicable diseases, and the global epidemiological burden varies by location, with more data available in high-income countries [6, 7]. Scotti et al. in a systematic review, reported that the prevalence of PsA in the general population was 0.13%, with an incidence of 83 per 100,000 person-years [7]. In Germany, Maximilian et al. estimated the prevalence of PsA, using the German health insurance database, to be 0.29% in the adult population in 2010, with the highest prevalence of 0.50% among those in their 50 s [8]. In Norway, Hoff et al. estimated the prevalence in the adult population to be between 0.67% and 0.70%, with an incidence of 41.3 per 100,000 person-years during 2000–2008 [9]. In the USA, Karmacharya et al. estimated the prevalence of PsA in the general population at 0.18%, with a relatively low incidence of 8.5 per 100,000 population in 2000–2017 [10].

The global population-based prevalence of PsA is still unclear and not well described [11]. Only a few countries, such as Denmark, Canada, and the USA, have the capability to organize and publish their PsA population-based surveillance data using a national medical registration system [10, 12, 13]. The global rheumatology literature also lacks data to describe the developments of PsA over certain time periods [11]. Given the limits of existing medical sources, health scientists are encouraged to use a projection model to estimate or forecast future situations and possible trends and developments of PsA based on historical data [14‐16]. Existing research offers examples of successful projections and estimations of other non-communicable diseases using the illness–death model [14‐16], but there are so far no studies in the global rheumatology literature that have applied such mathematical model to project and estimate the future burden of PsA. At the national level, projection estimations could help public health stakeholders to better prepare pharmaceutical interventions and medical resource allocation in order to improve the quality of life of patients with PsA [17].

In this analysis, we used population-based data of 65 million from the German Institute for Medical Documentation and Information (DIMDI) between January 2009 and December 2012 [18]. The specific aims were (i) to conduct a population-based prevalence projection and provide long-term future estimations of PsA patients in Germany until 2050, using the illness–death model and based on historical data; and (ii) using assumptions and references from the literatures, to compare estimations of PsA patient numbers between five different possible scenarios that reflect potential trends and developments of the disease.

This is the first study to use the illness–death model to project PsA case numbers in Germany until 2050, accounting for epidemiological parameters based on the National Statutory Health Insurance data of 65 million German population. The novelty of this study is that it not only fills a gap in the rheumatology research in Germany, but also provides PsA reference data for public health stakeholders to aid in decisions on medical resource allocation and pharmaceutical prevention in order to flatten the epidemiological curve and to improve the quality of life of patients with PsA, particularly among the senior female population (Table 3).

The conservative estimate of overall PsA prevalence in Germany in 2019 was 0.31%, which is similar to the prevalence in Denmark (0.22% in 1997–2011) [11]; and significantly higher than the prevalence in the USA (0.18% in 2015) [10], Czech Republic (0.049% in 2002–2003) [40], Canada (0.17% in 2015) [13], and in the most recent global systematic review (0.13% in 2007–2015) [7]; and significantly lower than in Norway (0.67–0.70% in 2000–2008) [9], Spain (0.58% in 2016–2017) [41], and Italy (0.42% in 2004) [42]. Figure 3 shows the substantial variations in the global prevalence of PsA.

We found that women have a higher prevalence of PsA, which is consistent with the findings from the Norwegian, Canadian, and Danish studies but inconsistent with the US study [9‐11, 13]. Senior women have a higher prevalence than younger women in the scenarios with increasing incidence (Fig. 1), which is also consistent with the Danish study’s finding of PsA being predominant among older women if there is rising incidence [11]. This may be because screening tools are helpful in detecting PsA patients [43, 44], or because older female patients are more likely to seek medical therapy for arthritic symptoms [11]. Another interesting finding is that, in scenario 4 with 5% increasing incidence, even though the number of cases among women (688 thousand) is higher than among men (565 thousand) in 2050, the rate of increase for men between 2019 and 2050 is greater, at 353%, than it is for women, at 341% (Table 2).

In the recent rheumatology literature, there have been few studies describing future projections of the development of PsA prevalence over time, and the current data offer a good reference for decisions regarding PsA patient management. In the conservative scenario (scenario 1), the prevalence of PsA between 2020 and 2050 fluctuates and exceeds 1% only slightly for both genders. This is an encouraging message that shows that maintaining a low disease prevalence until 2050 would represent a possible success in disease prevention efforts. However, in the scenarios with increasing incidence, the peak prevalence for women at the age of 60 may increase markedly, from 1% in 2019 to 3.5% in 2050. In Germany in 2050, an estimated 27% (20 million) of the population will be over 67 years old [22], and the literature indicates that PsA patients are at increased risk for coronary calcification and therefore cardiovascular events and death [45]; the economic burden and mortality from PsA are therefore likely to rise.

From a clinical perspective, a major factor influencing PsA patient management and increased incidence may be the shortage of rheumatologists in Germany. The American College of Rheumatology has estimated that the US demand for patient care services in rheumatology will exceed supply by 4.1 thousand clinical full-time equivalents by 2030 [46], and Germany may face a similar situation, including long waiting lists for specialists that result in delayed treatment and increased incidence of PsA.

Another factor influencing the presented scenarios may be the coding of PsA diagnoses, with most such diagnoses in Germany currently being coded by general practitioners (GP). The coding of a diagnosis (ICD-10) depends firstly on knowledge of the specific diagnosis, which is most commonly held by specialists, and secondly on the availability of therapies: if a GP cannot prescribe a specific PsA medication, the diagnosis is often simply not provided. If patients cannot receive appropriate diagnoses and care from GPs and are not referred to specialists early, this may increase the ultimate incidence. The recent broadening of awareness and knowledge about the disease and its therapeutic options, such as through educational programs or television commercials, could also explain the increase in coded incidence. Such clinical realities could lead to scenarios 3 and 4 and their consequences. It is thus clear that our data could persuade healthcare stakeholders to ensure that more rheumatologists are trained in order to alleviate the clinical problems experienced by PsA patients. Furthermore, the evidence from rheumatology research shows that the endogenous factors of overweight and vitamin D deficiency are likely to increase the incidence of PsA, wherein these possible drivers are increasing, especially in the Western world [47, 48].

The World Health Organization has started advocating for psoriasis patient management [44], but, according to the rheumatology literature, predicting the development of PsA for those already suffering from psoriasis is difficult. Nevertheless, preventive strategies can focus on identifying predictors (e.g., skin and musculoskeletal inflammation, certain comorbidities, and moderate or severe stress) and on treating psoriasis symptoms with individualized therapy strategies early by referring to the specialists [44, 49], and it would thus be possible to slow or even prevent the transition of psoriasis to PsA [49]. With such early interventions to reduce the incidence and to flatten the prevalence curve, like scenarios 2 and 5 in the graph, we might even achieve a decreasing trend in cases from 2035 to 2050, reducing the burden of PsA on the healthcare system (Fig. 2).

The main strengths of this study include its national and population-based data, its large sample size, and the use of the illness–death model in a predictive mathematical estimation of future PsA prevalence and case numbers until 2050. This prospective projection research provides advantages in understanding the trends and developments of PsA over US and Canadian surveys that have examined the prevalence of PsA retrospectively [10, 13]. Our data from 65 million insured data yields a representative and generalizable analysis. However, this study also has limitations. First, the historical data used for the predictions only included patients with public health insurance, thus excluding those with private health insurance or no insurance. However, health insurance is required by law in Germany, so the numbers of those uninsured is very small, and only around 10% are privately insured [20]; thus, although this is a limitation, it is not as profound as it might appear. Second, given that our analysis relies on aggregated insured data, inherent limitations exist due to the absence of detailed documentation regarding the specific diagnostic criteria used for determining PsA [50]. Furthermore, the extrapolated calculation until 2050 relied on insurance claims diagnoses (with possibility of misdiagnosis or overlapping diagnoses) instead of clinically confirmed diagnoses and validations, which may introduce potential bias into the results. Third, we applied the values of the Danish MRR of PsA as an epidemiological parameter into our projection model [12], and as the MRR was used as a pooled value for both sexes, this may represent a disadvantage in the model. Fourth, the five scenarios were developed on the basis of assumptions for the development of disease characteristics, along with the existing literature. Nevertheless, this is the best knowledge available for predicting future situations for PsA. Fifth, our conservative projections are premised on the idea that changes in incidence and mortality will remain constant annually, in accordance with historical trends in disease development. Consequently, our model and calculations would struggle to predict the effects of (i) the abrupt emergence of COVID-19 and (ii) specific time-bound treatment campaigns by pharmaceutical companies or political changes in health policy.

In conclusion, this study fills a gap in the research on the epidemiological burden and future development of PsA until 2050. We observed that the conservative estimate of overall PsA prevalence in Germany in 2019 was 0.31% and that, until 2050, senior women are projected to have a higher prevalence of 3.5% and to be more vulnerable to PsA. Our models aim to convince public health stakeholders to understand the shortcomings and weaknesses of current PsA patient management and to appreciate that, in the long term, it will be necessary to implement preventive strategies to identify predictors and treat psoriasis symptoms early in order to delay or even prevent the transition of psoriasis to PsA.