Research spotlight: Using RWD sources to describe the patient burden of a poorly understood rare disease
Liz Garry, Ph.D., M.P.H.
Senior Principal Scientist, Aetion
This month, Aetion and Boehringer Ingelheim published two co-authored papers which demonstrate how real-world data (RWD) can help describe the burden of disease in patients with a rare and poorly understood condition—in this case, generalized pustular psoriasis (GPP).
The studies, published in the Journal of Psoriasis and Psoriatic Arthritis (JPPA), were led by Jeffrey Crowley, MD, from the Bakersfield Dermatology and Skin Cancer Medical Group, and Jeffrey M. Sobell, MD, of SkinCare Physicians, along with other experts on psoriatic conditions.
While the two studies employed the same design, they were independent studies that used separate U.S. commercial RWD sources: IBM® MarketScan® and Optum® Clinformatics® Data Mart, respectively. All analyses were implemented via the Aetion Evidence Platform®.
Intent of the research
The objective of both studies was to use a real-world setting to describe patients with GPP, a rare form of psoriasis characterized by relapsing or remittent flares of pus-filled blisters, rather than the “plaques,” areas of thickened skin, that are common with other forms of psoriasis. Since the burden of disease among this population has not been well captured previously, we compared the demographics, comorbidities, medication use, health care resource utilization, and costs among patients with GPP to patients with plaque psoriasis, which has a burden of disease that is generally well understood and defined. Lastly, we compared our findings to a random sample of patients without any psoriasis to describe how our study cohort compares to the general population.
Process for the analyses
Our first task was to select a cohort of patients with GPP using the data that we had available. The transition to International Classification of Diseases, 10th revision (ICD-10) in the U.S. in October 2015 newly assigned GPP its own diagnosis code (L40.1). Since GPP had previously only been identifiable via a composite group with just one code for “other psoriasis,” we began our study period when the unique code first became available, and ended at the latest date we had available in each of the data sources: September 30, 2018, and March 31, 2019, respectively.
In order to be included in the GPP and plaque psoriasis cohorts, we required patients to have at least one inpatient claim or two outpatient claims with an ICD-10 code indicating GPP or plaque psoriasis—L40.1 or L40.0, respectively—separated by 30 days but within 365 days of one another. This is in line with standard practice for identifying patients in claims databases.
The general cohort required no history of psoriasis at the time of match, and patients were matched 4:1 according to date, age, and sex. All three cohorts were required to have at least one year of continuous enrollment with both medical and pharmacy coverage prior to the index date, which we considered the baseline period.
We analyzed the data using descriptive statistics, including the mean and standard deviation for normally distributed continuous variables, median and quartiles for skewed continuous variables, and frequencies and percentages for categorical variables.
Use of the Aetion Evidence Platform allowed us to more easily replicate the same study design in the MarketScan and Optum data sources. Conducting this study in two separate data sources allowed us to explore a larger sample size and areas with consistent trends. Understanding the particularities of each data source also helped us consider important differences, such as the distribution of age. For example, Optum data includes more patients aged 65 or older with Medicare Advantage, who may have a greater burden of disease and more comorbidities. We also see variance of formularies, as the MarketScan data includes multiple insurance providers while Optum includes only one.
Findings from Aetion’s analyses
In both RWD sources, when compared to patients with plaque psoriasis and patients without any psoriasis in the matched cohort, patients with GPP were more likely to have claims indicating comorbidities during the one-year baseline period, as well as greater medication use (see Figure 2 from each paper below), health care resource utilization, and costs.
MarketScan Data (Sobell et al.):
Optum Data (Crowley et al.):
Implications of the research
This study is one of the first to use RWD to characterize patients with GPP. By describing this poorly understood population using two large, nationally representative data sources, we can begin to identify trends that may be obscured in smaller clinical and case studies. Further, using patients with plaque psoriasis as a benchmark for comparison, which is far more common and well-documented, provided us with a clear benchmark to compare this target population to.
In addition, GPP has no specifically approved treatment in the U.S., highlighting an unmet need among this patient population. Increased understanding of the differences between patients with GPP and those with plaque psoriasis should encourage the development of specific treatments and management practices for patients with GPP to help alleviate their burden of illness. For example, more common antibiotic use among patients with GPP, who were also hospitalized more frequently and for longer durations, suggests that patients with GPP may be more susceptible to severe infections and other comorbid complications. Increased use of opioid pain medication and a greater prevalence of anxiety and depression suggests that patients with GPP not only experience a greater physical burden of pain, but also an emotional burden, which may require a more multidisciplinary approach than has been used for patients with plaque psoriasis. Additional RWD analyses that further dig into these areas of research could lead to increased treatment options and a reduced burden of disease.
The Evidence Digest
Subscribe to our newsletter for insights and updates on research and applications for real-world evidence.