S'abonner

P13 - Trends in the yearly evolution of the performance metrics of a regional breast cancer screening program and assessment of their bias based on a linkage to French national medico-administrative data : methodology and first results - 10/05/24

Doi : 10.1016/j.jeph.2024.202453 
E. Jemelen 1, 2,
1 Inria, Inserm, Université Paris Cité, Laboratoire HEKA, Paris, France 
2 Epiconcept Paris, Unité projets, Paris, France 

Auteur correspondant

Résumé

Background

The French national screening program for breast cancer (DOCS) was designed to increase the healthcare system's ability to detect early forms of breast cancer through mammograms reading and clinical examination for women aged 50 to 74 years old without known risk factors. Inner metrics for assessing the performance of the program are based on the screening tests results and the identification of breast cancer through complementary exams (tissue analyses mostly, which are not part of the screening program and are not performed on patients with a negative screening result). The DOCS has a follow-up policy which makes it possible to retrieve the complementary exams results and thus assess the validity of the screening tests. We claim that this inner assessment is incomplete. Indeed, in the case of cancer cases appearing after a negative screening exam, the screening program cannot identify the cancer diagnosis, for no complementary exam was performed. In 2019, a report by Santé publique France estimated that 17% of cancer cases identified in the screened population were detected less than 24 months after a negative screening. Under the assumption that some of these cancer cases were detectable during the screening exam and could be considered as false negatives, this result underlines the importance of finding methods to better identify breast cancer and assess the performance measures of the DOCS in a more acute manner.

Methods

After chaining the regional DOCS data of Occitanie collected between 2007 and 2018 to the medico-administrative database of the national health insurance system (SNDS), we compared the sensitivity and predictive positive value (PPV) obtained by inner identification of breast cancer (DOCS data only) with the same metrics obtained when identifying breast cancer with screening records linked to the SNDS data. Identification of cancer by chaining DOCS screening records to the SNDS data was considered the gold standard in our study. The yearly evolutions of the sensitivity and PPV of the DOCS with the two types of cancer identification (DOCS inner identification and gold standard identification) were then compared. Inclusion criteria: screening exams following the DOCS guidelines between 2007 and 2018 were included in our study (a total of 0,7% of screening exams were excluded).

Results

In Occitanie, between 2007 and 2018, the inner sensitivity and PPV of the DOCS evolved similarly to the gold standard sensitivity and PPV but with an offset. Over the period, the average inner sensitivity of the DOCS was 98.9% (SD = 0.7%) and our gold standard sensitivity was 78.5% (SD = 2.6%), resulting in an average difference of 20.3% (SD = 2.7%). The average inner PPV of the DOCS was 13.7% (SD = 2.8%) and our gold standard PPV was 18.2% (SD = 3.3%), resulting in an average difference of -4.4% (SD = 1.7%). Over the period, the rate of cancers identified with the SNDS data less than 24 months after a negative screening exam was 19.7% (higher than the 17% announced in 2019 by Santé publique France).

Conclusion

Through data linkage of the DOCS Occitanie screening database to the SNDS, we were able to identify more incident cancer diagnoses than with the DOCS data alone. Our findings show that the inner sensitivity of the program is overestimated and the inner PPV underestimated, which proves the usefulness of data linkage to the SNDS medico-administrative database to better estimate the performance of the DOCS screening program.

Le texte complet de cet article est disponible en PDF.

Keywords : Sensitivity and PPV, Organized screening program, Data linkage, Medico-administrative data, SNDS



© 2024  Publié par Elsevier Masson SAS.
Ajouter à ma bibliothèque Retirer de ma bibliothèque Imprimer
Export

    Export citations

  • Fichier

  • Contenu

Vol 72 - N° S2

Article 202453- mai 2024 Retour au numéro
Article précédent Article précédent
  • P12 - Algorithme d'identification des patients atteints d'Hémoglobinurie paroxystique nocturne (HPN) à partir des données du PMSI
  • R. Peffault de Latour, L. Chillotti, S. Bénard, J. Vercruyssen, N. Grandfils, ML. Bellais, S. Pibre, D. Pau
| Article suivant Article suivant
  • P14 - Facteurs individuels et géographiques associées aux usagers fréquents des urgences : une étude rétrospective nationale sur base de données
  • J. Chrusciel, D. Parise, R. Hellmann, A. Etienne, S. Sanchez

Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.

Déjà abonné à cette revue ?

Mon compte


Plateformes Elsevier Masson

Déclaration CNIL

EM-CONSULTE.COM est déclaré à la CNIL, déclaration n° 1286925.

En application de la loi nº78-17 du 6 janvier 1978 relative à l'informatique, aux fichiers et aux libertés, vous disposez des droits d'opposition (art.26 de la loi), d'accès (art.34 à 38 de la loi), et de rectification (art.36 de la loi) des données vous concernant. Ainsi, vous pouvez exiger que soient rectifiées, complétées, clarifiées, mises à jour ou effacées les informations vous concernant qui sont inexactes, incomplètes, équivoques, périmées ou dont la collecte ou l'utilisation ou la conservation est interdite.
Les informations personnelles concernant les visiteurs de notre site, y compris leur identité, sont confidentielles.
Le responsable du site s'engage sur l'honneur à respecter les conditions légales de confidentialité applicables en France et à ne pas divulguer ces informations à des tiers.


Tout le contenu de ce site: Copyright © 2024 Elsevier, ses concédants de licence et ses contributeurs. Tout les droits sont réservés, y compris ceux relatifs à l'exploration de textes et de données, a la formation en IA et aux technologies similaires. Pour tout contenu en libre accès, les conditions de licence Creative Commons s'appliquent.