TY - JOUR
T1 - Development and Validation of a Risk Score for Clostridium difficile Infection in Medicare Beneficiaries
T2 - A Population-Based Cohort Study
AU - Zilberberg, Marya D.
AU - Shorr, Andrew F.
AU - Wang, Li
AU - Baser, Onur
AU - Yu, Holly
N1 - Publisher Copyright:
© 2016, Copyright the Authors Journal compilation © 2016, The American Geriatrics Society
PY - 2016/8/1
Y1 - 2016/8/1
N2 - Objectives: To create a risk stratification score for Clostridium difficile infection (CDI) in elderly adults. Design: A sample from the Medicare 5% data set linked to Medicaid and Minimum Data Set (MDS) files from 2008 to 2009. Risk score was derived via a split-cohort through logistic regression model used to assign numerical values to each retained covariate. Score characteristics were tested using a threshold analysis. Setting: Community, long-term, and acute care settings. Participants: Population-based sample Medicare beneficiaries aged 65 and older on January 1, 2008, with continuous Medicare coverage from January 1, 2008, through December 31, 2009. Measurements: The primary outcome was incident CDI, defined as International Classification of Diseases, Ninth Revision, Clinical Modification code 008.45 or CDI according to the MDS, and no CDI in the preceding 12 months. Results: The cohort consisted of 6,838 participants with CDI and 1,158,327 without. Logistic regression modeling (hospitalization, nursing home stay, or antibiotics in prior year; inflammatory bowel, chronic liver, chronic kidney, or cardiac disease; aged ≥75, Northeast residence; c-statistic = 0.858) was used to determine to a score out of 22 possible points. A score of seven points (found in 18.8% of the total population) had a negative predictive value of 98.7%. Conclusion: CDI risk is high and age dependent in the population aged 65 and older. By stratifying risk, this score should help ensure efficient allocation of prevention resources.
AB - Objectives: To create a risk stratification score for Clostridium difficile infection (CDI) in elderly adults. Design: A sample from the Medicare 5% data set linked to Medicaid and Minimum Data Set (MDS) files from 2008 to 2009. Risk score was derived via a split-cohort through logistic regression model used to assign numerical values to each retained covariate. Score characteristics were tested using a threshold analysis. Setting: Community, long-term, and acute care settings. Participants: Population-based sample Medicare beneficiaries aged 65 and older on January 1, 2008, with continuous Medicare coverage from January 1, 2008, through December 31, 2009. Measurements: The primary outcome was incident CDI, defined as International Classification of Diseases, Ninth Revision, Clinical Modification code 008.45 or CDI according to the MDS, and no CDI in the preceding 12 months. Results: The cohort consisted of 6,838 participants with CDI and 1,158,327 without. Logistic regression modeling (hospitalization, nursing home stay, or antibiotics in prior year; inflammatory bowel, chronic liver, chronic kidney, or cardiac disease; aged ≥75, Northeast residence; c-statistic = 0.858) was used to determine to a score out of 22 possible points. A score of seven points (found in 18.8% of the total population) had a negative predictive value of 98.7%. Conclusion: CDI risk is high and age dependent in the population aged 65 and older. By stratifying risk, this score should help ensure efficient allocation of prevention resources.
UR - http://www.scopus.com/inward/record.url?scp=84982272522&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84982272522&partnerID=8YFLogxK
U2 - 10.1111/jgs.14236
DO - 10.1111/jgs.14236
M3 - Article
C2 - 27295521
AN - SCOPUS:84982272522
SN - 0002-8614
VL - 64
SP - 1690
EP - 1695
JO - Journal of the American Geriatrics Society
JF - Journal of the American Geriatrics Society
IS - 8
ER -