Chantal Marie Couris

Updating and Validating the Charlson Comorbidity Index and Score for Risk Adjustment in Hospital Discharge Abstracts Using Data From 6 Countries

With advances in the effectiveness of treatment and disease management, the contribution of chronic comorbid diseases (comorbidities) found within the Charlson comorbidity index to mortality is likely to have changed since development of the index in 1984. The authors reevaluated the Charlson index and reassigned weights to each condition by identifying and following patients to observe mortality within 1 year after hospital discharge. They applied the updated index and weights to hospital discharge data from 6 countries and tested for their ability to predict in-hospital mortality. Compared with the original Charlson weights, weights generated from the Calgary, Alberta, Canada, data (2004) were 0 for 5 comorbidities, decreased for 3 comorbidities, increased for 4 comorbidities, and did not change for 5 comorbidities. The C statistics for discriminating in-hospital mortality between the new score generated from the 12 comorbidities and the Charlson score were 0.825 (new) and 0.808 (old), respectively, in Australian data (2008), 0.828 and 0.825 in Canadian data (2008), 0.878 and 0.882 in French data (2004), 0.727 and 0.723 in Japanese data (2008), 0.831 and 0.836 in New Zealand data (2008), and 0.869 and 0.876 in Swiss data (2008). The updated index of 12 comorbidities showed good-to-excellent discrimination in predicting in-hospital mortality in data from 6 countries and may be more appropriate for use with more recent administrative data.

Sensibilité et spécificité de deux méthodes d'identification des cancers du sein incidents dans les services spécialisés à partir des données médico-administratives

Position du probleme Les bases de donnees medico-administratives du secteur hospitalier de court sejour disponibles au niveau national englobent la majorite des patients cancereux au debut de leur maladie. L’objectif est de determiner la possibilite de denombrer a partir de ces donnees les patientes atteintes de cancer du sein incident en utilisant des methodes d’identification. Deux methodes d’identification a partir de ces donnees ont ete evaluees dans trois services specialises d’un Centre Hospitalo-Universitaire. Methodes La premiere methode d’identification permettait de selectionner les femmes ayant eu au moins un sejour avec cancer du sein en diagnostic principal. La deuxieme, plus restrictive, selectionnait les femmes ayant au moins un sejour avec cancer du sein en diagnostic principal et un acte chirurgical specifique du traitement curatif du cancer du sein. Ces deux methodes ont ete appliquees aux 4 588 femmes d’au moins 20 ans hospitalisees en 2000 dans trois services specialises des Hospices Civils de Lyon. Afin de les classer en deux groupes : cancer du sein incident ou absence de cancer du sein incident, 150 femmes ont ete tirees au sort dans chacun des groupes ainsi constitues et leurs dossiers medicaux ont ete utilises comme reference. Resultats La sensibilite et la specificite assorties de leur intervalle de credibilite etaient respectivement 99,4 % (84-99,9) et 91,7 % (90,3-93,3) pour la premiere methode ; 93,8 % (76,2-98,7) et 97,3 % (96,1-98) pour la deuxieme methode. Parmi les femmes identifiees a tort avec un cancer du sein incident en 2000, 75,4 % (43/57) avaient un cancer du sein non incident dans l’annee avec la premiere methode et 96 % (24/25) avec la deuxieme methode. Pour ces patientes identifiees a tort avec un cancer du sein incident, des erreurs de codage du diagnostic principal ont ete retrouvees dans 24,6 % (14/57) des cas avec la premiere methode et dans 4 % (1/25) des cas avec la deuxieme methode. Leur correction a conduit a des sensibilite et specificite de 99,2 % (86,5-99,9) et 92,9 % (91,4-94,6) pour la premiere methode et de 94,2 % (76,5-98,7) et 97,3 % (96,2-98,1) pour la deuxieme methode. Conclusions La deuxieme methode utilisant des actes de chirurgie apparait plus specifique avec une perte en sensibilite qui reste faible. Les conditions d’utilisation de methodes d’identification pour estimer le nombre de cancer incident reste a preciser.

A Literature Review to Assess the Use of Claims Databases in Identifying Incident Cancer Cases

AbstractBackground: Claims databases that include standardized medical and administrative information routinely collected on patients are becoming increasingly prevalent in many industrialized countries. Some of them have a national scope and are an efficient source of information on large heterogeneous patient populations. These databases can help to describe cancer incidence in the absence of national cancer registries. A literature review was performed to describe the use of these databases to identify incident cancer cases. Methods: References (1990 and 2000) were extracted from MEDLINE. A final selection of the articles to be analyzed was made by appraising the references using defined criteria. Results: Most of the studies retrieved were based on the Medicare databases. Identification of new cases was based on the detection of cancer diagnosis codes or on the finding of cancer-specific diagnosis and procedure codes. The standards used to evaluate identification were either cancer registries or medical records. Evaluation was based on comparison of incidence rates or carried out on a person-level basis according to a cross-linkage of claims and standard databases. Only studies that reported sensitivity and specificity for identification were able to define the best compromise between a high identification rate and a low false positive rate. These studies were based on breast cancer and reported sensitivity and specificity varying from 75 to 90% and from 99.75 to 99.86%, respectively. Conclusions: Identification methods can be used to obtain an accurate number of incident cancer cases if an appropriate statistical method is applied. We proposed a method that take into account lack of sensitivity and specificity.

French Claims Data as a Source of Information to Describe Cancer Incidence: Predictive Values of Two Identification Methods of Incident Prostate Cancers

Claims data from the “Programme de Médicalisation du Système d’Information” (PMSI) have been commonly used for several years to complement cancer registries and describe cancer incidence in France. It is less clear whether or not it is possible to use these data as an independent source of information to assess cancer incidence, in the absence of a regional cancer registry. Following a similar study on breast cancer, we present a study which aimed to evaluate two methods of identifying incident prostate cancer using claims data. These methods were developed using claims data from the Hospices Civils de Lyon (HCL) and their validity was tested against medical records. The first method (M1) identified incident patients as those who had at least one stay with a principal diagnosis of prostate cancer. The second method (M2) had a prostate cancer treatment code in addition to the criteria for the first method. Both methods of identification had similar results, indicating a low rate of false negatives (negative predictive values: M1=100 [CI95: 93.8–100], M2=98.6 [CI95: 90.1–99.6]) and a high rate of false positives (positive predictive values: M1=33.3 [CI95: 23.2–42.1], M2=33.7 [CI95: 24.2–43.2]). The sample size did not allow us to produce consistent estimates of sensitivity and specificity. Our results showed that an estimation of the number of incident cases of prostate cancer using both methods of identification would be biased because of the high rate of false positives. Statistical methods that correct identification errors should be used.