Seven-question tool aims to prevent readmissions

Seven data points may help predict which discharged patients are at risk for readmission in the next 30 days – critical information now that Medicare penalizes hospitals with excess readmissions.

The HOSPITAL score assesses factors that independently contribute to readmission: hemoglobin at discharge; discharge from an oncology service; sodium level at discharge; procedures employed during index admission; type of index admission; number of admissions in the last 12 months; and length of stay.

The tool scores risk on a 0-13 scale. Patients who score 0-4 points are considered at low risk for a readmission, those with scores of 5-6 are at intermediate risk, and those with scores of 7 or more are considered high risk, according to a study published March 25 in JAMA Internal Medicine (2013 [doi:10.1001/jamainternmed.2013.3023]).

"The use of this simple score before discharge may help target transitional care for patients who might benefit the most and consequently reduce the rate of avoidable readmission," wrote the researchers from Brigham and Women’s Hospital in Boston and the Bern (Switzerland) University Hospital.

The tool does not include functional status, health literacy, degree of social support, or previous medication adherence – all of which are predictors of potentially avoidable readmission, the researchers wrote. They said that they chose not to include these data points because they are infrequently measured and difficult to obtain.

"The study’s goal was to derive a model that could be easily and widely used," they wrote.

To develop the tool, the researchers performed a retrospective cohort study of consecutive adult patient discharges from all of the medical services at the Brigham and Women’s Hospital between July 1, 2009, and June 30, 2010. They compared patients with a potentially avoidable readmission to those who were not readmitted at 30 days. They excluded unavoidable readmissions, such as planned readmissions for any reason, and readmissions for newly developed, unrelated conditions.

Two-thirds of the admissions were randomized into a derivation set (6,141) and one-third was included in a validation set (3,071). The validation set showed good calibration. The Hosmer-Lemeshow goodness-of-fit statistics were P = .28 and P = .15 in the derivation and validation sets, respectively. Nonsignificant P values on the test indicate a "good fit."

The discriminatory power of the HOSPITAL score was considered fair, with a cross-validated C statistic of 0.69 in the derivation set and 0.71 in the validation set.

For instance, in the derivation set, 18.7% of admissions were observed to be at high risk for readmission, compared with 18.3% estimated by the HOSPITAL score. Similar results were found in the validation set. The proportion of admissions at high-risk for return in 30 days was 18.2%, compared to 18.0% estimated by the HOSPITAL score.

The study was supported in part by grants from the Swiss Science National Foundation and the SICPA Foundation.

[email protected]

Seven data points may help predict which discharged patients are at risk for readmission in the next 30 days – critical information now that Medicare penalizes hospitals with excess readmissions.

The HOSPITAL score assesses factors that independently contribute to readmission: hemoglobin at discharge; discharge from an oncology service; sodium level at discharge; procedures employed during index admission; type of index admission; number of admissions in the last 12 months; and length of stay.

The tool scores risk on a 0-13 scale. Patients who score 0-4 points are considered at low risk for a readmission, those with scores of 5-6 are at intermediate risk, and those with scores of 7 or more are considered high risk, according to a study published March 25 in JAMA Internal Medicine (2013 [doi:10.1001/jamainternmed.2013.3023]).

"The use of this simple score before discharge may help target transitional care for patients who might benefit the most and consequently reduce the rate of avoidable readmission," wrote the researchers from Brigham and Women’s Hospital in Boston and the Bern (Switzerland) University Hospital.

The tool does not include functional status, health literacy, degree of social support, or previous medication adherence – all of which are predictors of potentially avoidable readmission, the researchers wrote. They said that they chose not to include these data points because they are infrequently measured and difficult to obtain.

"The study’s goal was to derive a model that could be easily and widely used," they wrote.

To develop the tool, the researchers performed a retrospective cohort study of consecutive adult patient discharges from all of the medical services at the Brigham and Women’s Hospital between July 1, 2009, and June 30, 2010. They compared patients with a potentially avoidable readmission to those who were not readmitted at 30 days. They excluded unavoidable readmissions, such as planned readmissions for any reason, and readmissions for newly developed, unrelated conditions.

Two-thirds of the admissions were randomized into a derivation set (6,141) and one-third was included in a validation set (3,071). The validation set showed good calibration. The Hosmer-Lemeshow goodness-of-fit statistics were P = .28 and P = .15 in the derivation and validation sets, respectively. Nonsignificant P values on the test indicate a "good fit."

The discriminatory power of the HOSPITAL score was considered fair, with a cross-validated C statistic of 0.69 in the derivation set and 0.71 in the validation set.

For instance, in the derivation set, 18.7% of admissions were observed to be at high risk for readmission, compared with 18.3% estimated by the HOSPITAL score. Similar results were found in the validation set. The proportion of admissions at high-risk for return in 30 days was 18.2%, compared to 18.0% estimated by the HOSPITAL score.

The study was supported in part by grants from the Swiss Science National Foundation and the SICPA Foundation.

[email protected]

Seven data points may help predict which discharged patients are at risk for readmission in the next 30 days – critical information now that Medicare penalizes hospitals with excess readmissions.

The HOSPITAL score assesses factors that independently contribute to readmission: hemoglobin at discharge; discharge from an oncology service; sodium level at discharge; procedures employed during index admission; type of index admission; number of admissions in the last 12 months; and length of stay.

The tool scores risk on a 0-13 scale. Patients who score 0-4 points are considered at low risk for a readmission, those with scores of 5-6 are at intermediate risk, and those with scores of 7 or more are considered high risk, according to a study published March 25 in JAMA Internal Medicine (2013 [doi:10.1001/jamainternmed.2013.3023]).

"The use of this simple score before discharge may help target transitional care for patients who might benefit the most and consequently reduce the rate of avoidable readmission," wrote the researchers from Brigham and Women’s Hospital in Boston and the Bern (Switzerland) University Hospital.

The tool does not include functional status, health literacy, degree of social support, or previous medication adherence – all of which are predictors of potentially avoidable readmission, the researchers wrote. They said that they chose not to include these data points because they are infrequently measured and difficult to obtain.

"The study’s goal was to derive a model that could be easily and widely used," they wrote.

To develop the tool, the researchers performed a retrospective cohort study of consecutive adult patient discharges from all of the medical services at the Brigham and Women’s Hospital between July 1, 2009, and June 30, 2010. They compared patients with a potentially avoidable readmission to those who were not readmitted at 30 days. They excluded unavoidable readmissions, such as planned readmissions for any reason, and readmissions for newly developed, unrelated conditions.

Two-thirds of the admissions were randomized into a derivation set (6,141) and one-third was included in a validation set (3,071). The validation set showed good calibration. The Hosmer-Lemeshow goodness-of-fit statistics were P = .28 and P = .15 in the derivation and validation sets, respectively. Nonsignificant P values on the test indicate a "good fit."

The discriminatory power of the HOSPITAL score was considered fair, with a cross-validated C statistic of 0.69 in the derivation set and 0.71 in the validation set.

For instance, in the derivation set, 18.7% of admissions were observed to be at high risk for readmission, compared with 18.3% estimated by the HOSPITAL score. Similar results were found in the validation set. The proportion of admissions at high-risk for return in 30 days was 18.2%, compared to 18.0% estimated by the HOSPITAL score.

The study was supported in part by grants from the Swiss Science National Foundation and the SICPA Foundation.

[email protected]