There is limited data published on resource utilization in patients with Crohn's disease. We aimed to find out influence of patient and hospital characteristics on length of stay and cost of care.Methods:
The study cohort was derived from the Nationwide Inpatient sample (NIS) of 2012 to 2013, a subset of the Healthcare Cost and Utilization Project (HCUP) sponsored by the Agency for Healthcare Research and Quality (AHRQ). We identify Crohn's Disease using ICD 9 diagnosis code (555.xx) in principle diagnosis filed. we excluded patients with missing information on age, sex and mortality. Patients with age ≤18 years were excluded as well. Co-morbidities were identified by “cm_” variables in HCUP database. Comorbid conditions were defined by Charlson's Comorbidity Index (CCI). Primary outcomes of study were length of stay (LOS) and cost of care (Cost). We ran survey regression model to find out predictors of length of stay and cost of care.Results:
Twenty five thousand eight hundred fifty-eight (weighted number: 114,285) patients were admitted with Crohn's disease nationwide. Median age of patients was (40, 29–54) years. 55.79% were females and 75.80% were white. Patients with older age (LOS, 95% Confidence Interval, P value) (+0.01 d, +0.01 to +0.01, P < 0.001), higher Charlson score (CCI-1: +0.72 d, +0.36 to +1.09, P < 0.001), (CCI-2: +1.60 d, +1.04 to +2.17, P < 0.001) compared to CCI-0, obesity (+0.48 d, +0.17 to +0.78, P = 0.001), heart failure (+1.36 d, +0.55 to +2.18, P = 0.001), peripheral vascular disease (+1.00 d, +0.35 to +1.65, P = 0.002), emergent admission (+1.06 d, +0.85 to +1.27, P < 0.001) compared to elective admission were associated with longer hospital stay. Hospital characteristics such as hospital teaching status (+0.94 d, +0.78 to +1.09, P <0.001), larger bed size hospitals (Medium: +0.51 d, +0.31 to +0.71, P <0.001), (Large: +0.95 d, +0.76 to +1.15, P < 0.001) compared to small size. Patients with private insurance (−0.32, −0.49 to −0.15, P < 0.001) compared to Medicare/Medicaid had shorter length of stay. As far as cost is concern, patient with higher Charlson Score (Cost, 95% confidence interval, P-value) (CCI-1, +8091 USD, +3952 to +12,230, P < 0.001), (CCI-2, +16,522 USD, +8628 to +24,415, P < 0.001), heart failure (+14,526 USD, +5510 to +23,543, P = 0.002), peripheral vascular disease (+10,292 USD, +2498 to +18,085, P = 0.009), emergent admission (+15,413 USD, +12,854 to +17,973, P < 0.001) compared to elective were associated with higher cost of care. Hospital characteristics such as higher bed size of hospital (Medium: +6088 USD, +4039 to +8137, P < 0.001), (Large: +11,030 USD, +8775 to +13,285, P < 0.001) compared to small bed size and hospital teaching status (+9262 USD, +7290 to +11,233, P < 0.001) compared to non-teaching. Non influential co morbidities were hypertension, diabetes, chronic lung disease, arthritis, liver disease, alcohol abuse.Conclusions:
We identify multiple patient and hospital level characteristics influencing length of stay and cost of care of Crohn's disease patients. Additional intervention would require to decrease disparity in resource utilization amongst different kind of hospitals. Patient with higher resource utilization can be identified and can be intervened to reduce utilization of resources.