Derivation and validation of a simple, accurate and robust prediction rule for risk of mortality in patients with Clostridium difficile infection

Butt, Emma; Foster, Jane A.; Keedwell, Edward; Bell, Julia; Titball, Richard W.; Bhangu, Aneel; Michell, Stephen L.; Sheridan, Ray

doi:10.1186/1471-2334-13-316

Cited by 29 publications

(28 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2,3 Over the last decade, substantial increases in the incidence of CDI have been observed, 4,5 likely related to the emergence of more virulent strains. 6,7 Approximately 50% of CDI cases develop during the course of an inpatient hospital episode of care, 8 which may result in significantly longer lengths of stay (LOS) and, consequently, in greater healthcare costs.…”

mentioning

confidence: 99%

Excess Length of Stay Attributable to Clostridium difficile Infection (CDI) in the Acute Care Setting: A Multistate Model

Stevens

Khader

Nelson

et al. 2015

Infect. Control Hosp. Epidemiol.

View full text Add to dashboard Cite

Background. Evaluation of the economic impact of interventions designed to reduce the incidence of Clostridium difficile infection (CDI) depends on accurate estimates of healthcare resource use attributable to infection. Studies that do not account for the timing of infection overestimate attributable length of stay (LOS). The purpose of this study was to evaluate the excess LOS due to CDI using a multi-state model Methods. We conducted a retrospective cohort study of all patients hospitalized on a general medical, surgical, or intensive care unit within the US Department of Veterans Affairs (VA) health care system between January 1, 2005 and December 31, 2012. A diagnosis of CDI was based on a positive laboratory result by enzyme immunoassay or cytotoxin test. A multi-state approach, implemented through the etmpackage in R, was used to estimate the excess LOS attributable to CDI while accounting for the timing of infection. Confidence intervals were estimated using bootstrapping techniques. A composite outcome of discharge or death was used. Estimates from the multistate model were compared to traditional estimates, including crude comparisons, ordinary least squares (OLS), and a generalized linear model (GLM) with a log link and gamma distributionResults. During the study period, 4.2 million Veterans were followed up for nearly 22 million patient-days of observation. CDIs were observed in 43,661 (1%) of hospitalizations. The median LOS among patients with and without CDI was 10 (IQR: 17) and 3 (IQR: 4) days, respectively, for a crude difference of 7 days ( p < .0001). OLS regression returned an estimated excess LOS of 14.3 (95%CI: 14.0 -14.6) days. The estimated increase in LOS from the GLM model was 7.0 (95% CI: 6.9 -7.1) days. When accounting for the timing of infection in a multi-state model, the attributable LOS was estimated to be 2.28 (95% CI: 2.14 -2.41) daysConclusion. CDI significantly contributes to LOS in the acute care setting but the magnitude of its estimated impact is substantially smaller with methods that account for the time-varying nature of infection. Further efforts to control for time-varying confounding will refine these estimates. Estimates of CDI costs should be based on more accurate measures of attributable LOS to avoid overstating the benefit of interventions to reduce CDI

show abstract

mentioning

confidence: 99%

Excess Length of Stay Attributable to Clostridium difficile Infection (CDI) in the Acute Care Setting: A Multistate Model

Stevens

Khader

Nelson

et al. 2015

Infect. Control Hosp. Epidemiol.

View full text Add to dashboard Cite

show abstract

“…We identified 12 papers presenting a prediction model for a complicated course of CDI (Table 1) [8][9][10][11][12][13][14][15][16][17][18] , of which eight studies were excluded because they included a selected patient group 18 , used nonquantitative parameters (eg, altered mental status, Horn's index) [10][11][12] , or used variables that were not available in our cohort on the day of CDI diagnosis (eg, serum albumin, radiological findings). [12][13][14][15][16]19 Age, white blood count, and albumin level were used in the prediction model in >50% of the studies. All other 24 predictors were used in only 10% or 20% of models.…”

Section: Selection Of Prediction Modelsmentioning

confidence: 99%

“…Several scoring systems to predict a complicated course of CDI have been developed (Table 1). [8][9][10][11][12][13][14][15][16][17][18][19] However, none has gained widespread clinical acceptance due to lack of external validation, the retrospective study design, or the limited numbers of patients on which they are based. 20 To our knowledge, to date, only one study has been externally validated.…”

Section: Introductionmentioning

confidence: 99%

External Validation of Three Prediction Tools for Patients at Risk of a Complicated Course of Clostridium difficile Infection: Disappointing in an Outbreak Setting

Beurden

Hensgens

Dekkers

et al. 2017

Infect. Control Hosp. Epidemiol.

View full text Add to dashboard Cite

OBJECTIVE Estimating the risk of a complicated course of Clostridium difficile infection (CDI) might help doctors guide treatment. We aimed to validate 3 published prediction models: Hensgens (2014), Na (2015), and Welfare (2011). METHODS The validation cohort comprised 148 patients diagnosed with CDI between May 2013 and March 2014. During this period, 70 endemic cases of CDI occurred as well as 78 cases of CDI related to an outbreak of C. difficile ribotype 027. Model calibration and discrimination were assessed for the 3 prediction rules. RESULTS A complicated course (ie, death, colectomy, or ICU admission due to CDI) was observed in 31 patients (21%), and 23 patients (16%) died within 30 days of CDI diagnosis. The performance of all 3 prediction models was poor when applied to the total validation cohort with an estimated area under the curve (AUC) of 0.68 for the Hensgens model, 0.54 for the Na model, and 0.61 for the Welfare model. For those patients diagnosed with CDI due to non-outbreak strains, the prediction model developed by Hensgens performed the best, with an AUC of 0.78. CONCLUSION All 3 prediction models performed poorly when using our total cohort, which included CDI cases from an outbreak as well as endemic cases. The prediction model of Hensgens performed relatively well for patients diagnosed with CDI due to non-outbreak strains, and this model may be useful in endemic settings. Infect Control Hosp Epidemiol 2017;38:897-905.

show abstract

“…2 Several studies have been published identifying risk factors for C. difficile infection and developing prediction rules with the aim to improve management of individuals deemed to be at the highest risk. 6,[8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25] Most of these studies target patients admitted to hospitals or living in long-term healthcare facilities. [9][10][11][17][18][19][22][23][24][26][27][28] or C. difficile infection cases risking recurrence.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of a risk score to predict future Clostridium difficile disease using UK primary care and hospital data in Clinical Practice Research Datalink

Marley¹,

Hahi

Ferreira³

et al. 2019

Human Vaccines & Immunotherapeutics

View full text Add to dashboard Cite

We evaluated the applicability of a Clostridium difficile infection (CDI) risk index developed for patients at hospital discharge to identify persons at high-risk of CDI in a primary care population. This retrospective observational study used data from the UK Clinical Practice Research Datalink, linked with Hospital Episodes Statistics. The risk index was based on the following patient characteristics: age, previous hospitalizations, days in hospital, and prior antibiotics use. Individual risk scores were calculated by summing points assigned to pre-defined categories for each characteristic. We assessed the association of risk factors with CDI by multivariate logistic regression. The estimated CDI incidence rate was 4/10,000 and 2/10,000 person-years in 2008 and 2012, respectively. On an index with a maximal risk of 19, a cut-off for high risk of ≥7 had sensitivity, specificity and positive predictive values of 80%, 87% and 12%, respectively. A high-risk person had a ~ 35% higher risk of CDI than a low-risk person. Multivariate risk factor analysis indicated a need to reconsider the relative risk scores. The CDI risk index can be applied to the UK primary care population and help identify study populations for vaccine development studies. Reassessing the relative weights assigned to risk factors could improve the index performance in this setting.

show abstract

Derivation and validation of a simple, accurate and robust prediction rule for risk of mortality in patients with Clostridium difficile infection

Cited by 29 publications

References 19 publications

Excess Length of Stay Attributable to Clostridium difficile Infection (CDI) in the Acute Care Setting: A Multistate Model

Excess Length of Stay Attributable to Clostridium difficile Infection (CDI) in the Acute Care Setting: A Multistate Model

External Validation of Three Prediction Tools for Patients at Risk of a Complicated Course of Clostridium difficile Infection: Disappointing in an Outbreak Setting

Evaluation of a risk score to predict future Clostridium difficile disease using UK primary care and hospital data in Clinical Practice Research Datalink

Contact Info

Product

Resources

About