Attempted, participation was ultimately based on self-selection, which may affect the
Attempted, participation was ultimately based on self-selection, which may affect the representativeness of the study population. Although the estimated total number of HAE patients differs largely between the countries involved, equal numbers were attempted to recruit in each country, so as to receive the target sample size of 150 patients, which is consistent with other burden of illness studies in rare diseases [26,27] Attempts were made to ensure generalizable country specific sampling by ways of systematic or random selection. Yet, it cannot be excluded that the willingness of a patient to participate in a survey, increases with disease severity. We also performed an analysis of a potential selection bias by route of recruitment, either via patient organization or clinical center. Such analysis demonstrated no significant difference in the proportion of patients recruited via each route of recruitment. Also, the clinical characteristics of the recent attack (pain severity, swelling severity, time totreatment in each country, as well as duration of swelling in German and Spanish participants) showed no significant differences. Outcome measures for direct resource utilization (percentage of patients with medication to treat attack at home, percentage of treated attacks, visits for treatment of recent attack, PubMed ID:https://www.ncbi.nlm.nih.gov/pubmed/28914615 medication used to treat anxiety or depression) were also not significantly different between patients by recruitment route and country. Likewise, indirect resource utilization as measured by impact items (impact on work and school productivity as well as impact on daily activites, each with recent attack or between attacks) and report on being prevented from career or educational advancement were not significantly different between patients by recruitment route and country. Despite different routes of recruitment, the data obtained seem to be consistent. The survey relied on self-reported, retrospective data, which carries the potential for recall bias, although HAE patients often keep symptom diaries. Data collected 3-MAMedChemExpress 3-Methyladenine reflect the treatment options available at the time of the study (in 2011), which varied between countries. As this study was observational with small numbers representing different treatments, treatment-specific subgroup analyses were not feasible. For the most part, country comparisons were not feasible given different treatment availability, cultural attitudes toward health service use, and health care system configuration; all country comparisons made, including those regarding attack frequency and medication use, may reflect such differences. In estimating absenteeism over a one year period, it was assumed that the severity of the last attack and absenteeism during the last attack were representative of the average attack for the patient, although the severity of attacks is known to be unpredictable. InAyg en-P s et al. Orphanet Journal of Rare Diseases 2014, 9:99 http://www.ojrd.com/content/9/1/Page 8 ofaddition, absenteeism data were only available for 60 (n = 72) of patients either working or attending PubMed ID:https://www.ncbi.nlm.nih.gov/pubmed/26795252 school [56 (n = 40) of whom reported absenteeism], although attack severity and frequency did not differ for these patients compared to patients for whom such data was unavailable. Also, this study did not assess lost leisure time for patients and therefore may underestimate patient-level burden. Nevertheless, our findings, particularly those related to absenteeism and long-term career/educational impact, a.