Health Outcomes in Vaso Occlusive Crisis of Sickle Cell Disease - Retrospective Indicators

Abstract

4699 Background Retrospective data analysis is relatively inexpensive & allows evaluation of rare conditions such as Vaso Occlusive Crisis (VOC) of Sickle Cell Disease (SCD) with ease, as compared to conducting actual prospective studies. A particularly useful application of a retrospective data analysis is to act as a pilot study that can be completed in anticipation of a prospective study. The analysis can help to identify feasibility issues & factors which affect study outcomes that need to be adjusted for in prospective studies. Objective To identify health outcome measures employed in retrospective database analysis of individuals suffering from VOC of SCD, from reviewed literature. Method Potentially relevant articles were searched electronically using terms related to VOC & pain crisis in SCD, across PubMed, COCHRANE & CINAHL databases. Dissertations, guidelines to management of VOC, abstracts, conference proceedings & case reports were excluded for lack of sufficient information. Reference lists of prior literature reviews, as well as reference lists of studies included in this review, served to identify additional articles. However, review articles were not included for further analyses. Studies analyzing retrospective databases such as chart reviews, hospital records, electronic medical records (EMR), claims data, health care cost & utilization project database (HCUP) were included for further analyses. Studies qualifying for full review were searched for relevant clinical, humanistic, economic outcomes & clinical endpoints, which have previously been defined elsewhere. Result 1,917 potentially relevant study abstracts were reviewed. Of which, 13 studies qualified for further analyses. 8 clinical outcomes, 6 humanistic outcomes, 3 economic outcomes & 1 clinical endpoint were found. Codes used to identify patients with VOC of SCD in claims databases were (282.42), (282.60), (282.62), (282.64) & (282.69). Four studies used HCUP database, 3 studies used EMRs, & 2 studies each used Medicaid claims data, chart reviews & hospital records for retrospective assessment of a VOC event. Majority of the studies captured length of stay (LOS) (8 studies) & readmission rates (5 studies) as clinical outcomes measure while Wong-Baker Faces scale in children (2 studies) & Numeric Rating scale (2 studies) in adults, were most common measures recorded. Type of hospital admission, type of SCD admission, severity of pain, age, hemoglobin count & polymorphonuclear leukocyte count, associated significantly with length of stay in patients with VOC of SCD (Table 1). Disease severity, severity of pain, use of steroids, age, type of visits & having co-morbid history of asthma associated significantly with 14- & 30-day readmission rates. Number of pain episodes, intensity of pain, LOS, presence of Human Platelet Alloantigen polymorph (HPA-3), amount of analgesic administered, cost of healthcare & 14- & 30-day readmission rates were found to be significant predictors of severity of a VOC (Table 2). Conclusion Research evidence available, while scarce, is applicable, in population of all ages with VOC of SCD. Findings suggest using LOS & readmission rates for assessment of a VOC event in a retrospective database as well as prospective studies aimed to assess treatment efficacies. Care should be taken during design of pilot & retrospective studies to ensure collection & adjustment for variables known to associate significantly with LOS & readmission rate before reaching a clinical conclusion. Further research is necessary to support the use of suggested retrospective indicators of a VOC event. Disclosures: Shafer: Pfizer: Employment. Sivamurthy:Pfizer: Employment. Kollmer:Pfizer: Employment, Equity Ownership. Vendetti:Pfizer: Employment.