Evaluating the july effect on sickle cell crisis hospitalizations: A national inpatient sample analysis from 2016–2022

Abstract

Abstract Background: Sickle cell disease is a condition seen widely across the nation with acute sickle cell crisis being one of the primary reasons for hospital admissions. Caring for admitted patients during a sickle cell crisis can be particularly challenging, especially with the influx of new interns starting their training nationwide in teaching hospitals every July. We conducted this study using a large, nationally representative database to better understand how care may vary in different stages of clinical practice. Methods: We performed a retrospective analysis of inpatient admissions from the National Inpatient Sample (NIS) using data from 2016 to 2022 to compare patient outcomes in July, when new medical interns begin, versus February. February was selected as a comparison month to represent the midpoint of the academic year. We included patients aged 18 years or older who were admitted to teaching hospitals with a primary diagnosis of sickle cell crisis, identified using ICD-10-CM codes. National estimates were calculated using survey weights in Stata. The primary outcome was in-hospital mortality. Secondary outcomes included length of stay (LOS), total hospital charges, and inpatient complications. Multivariable logistic regression was used to evaluate mortality, while linear regression was used to assess LOS and hospital charges. All models were adjusted for age, sex, race, primary payer, hospital characteristics, and comorbidities using the Charlson Comorbidity Index (CCI). Statistical significance was defined as p < 0.05. Results: From 2016 to 2022, 38,614 patients were admitted in July with sickle cell crisis (mean age 33.7 years; 44.0% male, 55.9% female), compared to 36,984 in February (mean age 33.9 years; 45.0% male, 54.9% female). The in-hospital mortality rate was 0.4% in July and 0.2% in February. A non-significant trend toward higher mortality was observed in July compared to February (adjusted odds ratio [aOR] 0.667, p = 0.076). Length of stay (LOS) decreased from 5.6 days in July to 5.4 days in February. Mean total hospital charges were higher in July ($59,910) compared to February ($53,850), reflecting an unadjusted difference of $6,060. After adjustment, the average hospital charges in February were significantly lower by $8,011 (p < 0.001). Rates of acute chest syndrome were identical in February and July (0.7%, p = 0.8), as were rates of splenic sequestration (0.00004%, p = 0.9). No cerebral infarcts were reported in either month (p = 1.0). Pulmonary embolism occurred in 0.9% of February and 1.0% of July admissions (p = 0.08), while deep vein thrombosis occurred in 0.3% and 0.5%, respectively (p = 0.08). Hematuria rates were equal at 0.6% in both months (p = 0.99). Conclusion: July often raises concerns about potential care disruptions associated with the arrival of new medical interns; however, our analysis offers reassuring findings. Among patients hospitalized with sickle cell crisis at teaching hospitals, we observed no significant differences in in-hospital mortality or complication rates between July and February. While hospital charges were higher in July and length of stay was modestly shorter in February, overall care quality appeared consistent and uncompromised throughout the academic year.