User login
Clinical question: Can a prediction model based on administrative and clinical data identify potentially avoidable 30-day readmissions in medical patients prior to discharge?
Background: An estimated 18% of Medicare beneficiaries are readmitted to the hospital within 30 days of discharge, costing nearly $17 billion per year. Interventions to reduce readmission rates are costly and should be focused on high-risk patients. To date, using models to predict 30-day readmission has been problematic and unreliable.
Study design: Retrospective cohort.
Setting: Academic medical center in Boston.
Synopsis: Using consecutive discharges from all medical services of Brigham and Women’s Hospital occurring over one year, this study derived and internally validated a prediction model for potentially avoidable 30-day readmissions. Of 10,731 discharges, there were 2,399 (22%) 30-day readmissions, and 879 (8.5%) were deemed potentially avoidable. Seven independent predictors for readmission were identified and used to create a predictor score referred to as the HOSPITAL score. Predictors included hemoglobin and sodium levels at discharge, number of hospitalizations in the past year, and four features of the index hospitalization, including type, discharge from an oncology service, presence of procedures, and length of stay. The score was internally validated and found to predict potentially avoidable 30-day readmission in medical patients with fair discriminatory power and good calibration.
This study is unique in that none of the classic comorbidities (e.g. congestive heart failure) were associated with a higher risk of 30-day readmission. Previously unrecognized predictors, including hemoglobin, sodium, and number of procedures performed, were incorporated. This suggests that comorbidities are not as important as illness severity or clinical instability. Hospitalists should await studies that externally validate the HOSPITAL score before incorporating it into practice.
Bottom line: A unique and simple seven-item prediction model identifies potentially avoidable 30-day readmissions but needs to be externally validated before being widely utilized.
Citation: Donze J, Drahomir A, Williams D, Schnipper JL. Potentially avoidable 30-day hospital readmissions in medical patients. JAMA Intern Med. 2013;137(8):632-638.
Clinical question: Can a prediction model based on administrative and clinical data identify potentially avoidable 30-day readmissions in medical patients prior to discharge?
Background: An estimated 18% of Medicare beneficiaries are readmitted to the hospital within 30 days of discharge, costing nearly $17 billion per year. Interventions to reduce readmission rates are costly and should be focused on high-risk patients. To date, using models to predict 30-day readmission has been problematic and unreliable.
Study design: Retrospective cohort.
Setting: Academic medical center in Boston.
Synopsis: Using consecutive discharges from all medical services of Brigham and Women’s Hospital occurring over one year, this study derived and internally validated a prediction model for potentially avoidable 30-day readmissions. Of 10,731 discharges, there were 2,399 (22%) 30-day readmissions, and 879 (8.5%) were deemed potentially avoidable. Seven independent predictors for readmission were identified and used to create a predictor score referred to as the HOSPITAL score. Predictors included hemoglobin and sodium levels at discharge, number of hospitalizations in the past year, and four features of the index hospitalization, including type, discharge from an oncology service, presence of procedures, and length of stay. The score was internally validated and found to predict potentially avoidable 30-day readmission in medical patients with fair discriminatory power and good calibration.
This study is unique in that none of the classic comorbidities (e.g. congestive heart failure) were associated with a higher risk of 30-day readmission. Previously unrecognized predictors, including hemoglobin, sodium, and number of procedures performed, were incorporated. This suggests that comorbidities are not as important as illness severity or clinical instability. Hospitalists should await studies that externally validate the HOSPITAL score before incorporating it into practice.
Bottom line: A unique and simple seven-item prediction model identifies potentially avoidable 30-day readmissions but needs to be externally validated before being widely utilized.
Citation: Donze J, Drahomir A, Williams D, Schnipper JL. Potentially avoidable 30-day hospital readmissions in medical patients. JAMA Intern Med. 2013;137(8):632-638.
Clinical question: Can a prediction model based on administrative and clinical data identify potentially avoidable 30-day readmissions in medical patients prior to discharge?
Background: An estimated 18% of Medicare beneficiaries are readmitted to the hospital within 30 days of discharge, costing nearly $17 billion per year. Interventions to reduce readmission rates are costly and should be focused on high-risk patients. To date, using models to predict 30-day readmission has been problematic and unreliable.
Study design: Retrospective cohort.
Setting: Academic medical center in Boston.
Synopsis: Using consecutive discharges from all medical services of Brigham and Women’s Hospital occurring over one year, this study derived and internally validated a prediction model for potentially avoidable 30-day readmissions. Of 10,731 discharges, there were 2,399 (22%) 30-day readmissions, and 879 (8.5%) were deemed potentially avoidable. Seven independent predictors for readmission were identified and used to create a predictor score referred to as the HOSPITAL score. Predictors included hemoglobin and sodium levels at discharge, number of hospitalizations in the past year, and four features of the index hospitalization, including type, discharge from an oncology service, presence of procedures, and length of stay. The score was internally validated and found to predict potentially avoidable 30-day readmission in medical patients with fair discriminatory power and good calibration.
This study is unique in that none of the classic comorbidities (e.g. congestive heart failure) were associated with a higher risk of 30-day readmission. Previously unrecognized predictors, including hemoglobin, sodium, and number of procedures performed, were incorporated. This suggests that comorbidities are not as important as illness severity or clinical instability. Hospitalists should await studies that externally validate the HOSPITAL score before incorporating it into practice.
Bottom line: A unique and simple seven-item prediction model identifies potentially avoidable 30-day readmissions but needs to be externally validated before being widely utilized.
Citation: Donze J, Drahomir A, Williams D, Schnipper JL. Potentially avoidable 30-day hospital readmissions in medical patients. JAMA Intern Med. 2013;137(8):632-638.