30 Day Readmissions in Patients with Deep Vein Thrombosis in United States

Introduction: Deep vein thrombosis (DVT) is an important cause of the morbidity and mortality in the United States (US). National estimates of 30-day readmissions in DVT patients in the US are unknown. The objective of our study was to estimate readmission rates and identify causes, predictors and cost of readmissions in DVT patients.

Methods: We used National Readmission Dataset (NRD - the year 2013), Healthcare Cost and Utilization Project (HCUP), Agency for Healthcare Research and Quality which represents one of the largest random sample of discharges from all hospitals, excluding rehabilitation and long-term acute care hospitals. NRD is designed to generate national estimates of readmission analysis.NRD contains approximately unweighted 14 million discharges and weighted 36 million discharges for the year 2013. Discharge weights were utilized to generate the national estimates. The patients with DVT were identified by primary discharge diagnosis with ICD9-CM code 451-453. All cause unplanned 30-day readmission rates were calculated for patients admitted between January and November 2013 by excluding elective readmissions. Deyo's modification of Charlson comorbidity index was used to define the severity of co-morbid conditions. Using SAS version 9.3, survey procedures were implemented to adjust for stratified cluster design of NRD with DOMAIN, STRATA, CLUSTER and WEIGHT statement. A p-value of less than 0.05 was considered significant. The independent predictors of unplanned 30-day readmissions were identified by logistic regression. The cost of readmission was calculated by multiplying total charges with the cost to charge ratio provided by HCUP.

Results: The NRD contained 60,556 unique DVT patients with 126,362 admissions (weighted N = 288,553) in 2013. After excluding elective readmissions, all cause30-day readmission rate was 13.8%. The top causes of unplanned readmissions were phlebitis (15.9%), septicemia (6.6%), pulmonary heart disease (4.2%), skin and subcutaneous tissue infections (4.0%), gastrointestinal hemorrhage (3.8%), complication of implant or graft (2.9%), nonhypertensive congestive heart failure (2.7%), pneumonia (2.7%), acute, unspecified renal failure (2.4%), and urinary tract infections (2.4%). The multivariate predictors for higher 30 day unplanned readmissions were Charlson comorbidity index (OR 1.13, p<0.0001), large bedside hospitals (OR 1.19, p=0.0008), metropolitan teaching hospitals (OR 1.08, p<0.0001), weekend admissions (OR 1.08, p=0.03), Medicaid payer (OR 1.25, p<0.0001), discharge against medical advice (OR 2.7, p<0.0001), discharge to facility (OR 1.54, p<0.0001), discharge to home health care (OR 1.38, p<0.0001), any bleeding complications (OR 1.13, p=0.017), congestive heart failure (OR 1.44, p<0.0001), chronic pulmonary disease (OR 1.28, p<0.0001), cancer (OR 1.60, p<0.0001), major operative procedures (OR 1.36, p<0.0001), mechanical ventilation (OR 1.79, p=0.016).The multivariate predictors for lower 30 day unplanned readmissions were higher age (OR 0.99, p<0.0001), non-metropolitan hospitals (OR 0.81, p<0.0001), elective admission (OR 0.73, p<0.0001), self-pay, no charge or other pay (OR 0.82, p=0.002), private payer including HMO (health maintenance organization) (OR 0.75, p<0.0001). The estimated total cost of unplanned 30-day readmissions in DVT patients was $ 0.88 billion for 2013.

Conclusions: The unplanned 30-day readmission rates and the cost are high in DVT patients in the US. Phlebitis is the most common cause of unplanned 30-day readmission, which is potentially avoidable. Further research is needed to identify preventable readmissions, strategies to cut down the readmissions and eventually reduce the cost of readmissions in patients admitted with DVT.