Department of Emergency Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
Division of Hematology/Oncology, Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
J Natl Med Assoc. 2018 Apr;110(2):149-156. doi: 10.1016/j.jnma.2017.10.006. Epub 2017 Nov 16.
Analyses of administrative and large data sources in Sickle Cell Disease (SCD) can answer questions not suitable for prospective study but have been hampered by lack of validated methods to adjust for individual comorbidities and lack of baseline utilization data over time. We sought to develop a database to characterize inpatient SCD care across New York State and generate a re-weighted sickle-cell specific Charlson Comorbidity index (S-CCI) for use in future large data SCD research. We identified 18,541 individual SCD patients admitted to New York State hospitals between 2005 and 2013 from the SPARCS database. We present data from both a randomly selected derivation cohort, used to develop the S-CCI and a validation cohort, The S-CCI resulted in small improvements in model fit and discrimination while using fewer covariates, allowing a more parsimonious model. Despite being the most common comorbidity, chronic pulmonary disease was not predictive of mortality. Mortality per hospitalization was 0.61%. Many patients (32%) were admitted only once during the nine year period. However, the majority was admitted more frequently with over 15% of patients being admitted more than once per year.
对镰状细胞病（SCD）的行政和大型数据源进行分析可以回答不适合前瞻性研究的问题，但由于缺乏经过验证的方法来调整个体合并症，以及缺乏随时间推移的基线利用数据，这些分析受到了阻碍。我们试图开发一个数据库来描述纽约州的住院 SCD 护理情况，并为未来的大型 SCD 数据研究生成加权的镰状细胞特异性 Charlson 合并症指数（S-CCI）。我们从 SPARCS 数据库中确定了 2005 年至 2013 年期间在纽约州医院住院的 18541 名个体 SCD 患者。我们从一个随机选择的推导队列中展示了数据，该队列用于开发 S-CCI 和验证队列，S-CCI 导致模型拟合和区分度的微小改善，同时使用更少的协变量，允许更简洁的模型。尽管慢性肺部疾病是最常见的合并症，但它不能预测死亡率。每次住院的死亡率为 0.61%。许多患者（32%）在九年期间仅住院一次。然而，大多数患者更频繁地住院，超过 15%的患者每年住院超过一次。
