Virtual Library

Start Your Search

A. Gilbert



Author of

  • +

    P3.22 - Poster Session 3 - Epidemiology, Etiology (ID 168)

    • Event: WCLC 2013
    • Type: Poster Session
    • Track: Prevention & Epidemiology
    • Presentations: 1
    • +

      P3.22-007 - Determining completeness of case ascertainment to a lung cancer registry:<br /> A single institutional study. (ID 2585)

      09:30 - 09:30  |  Author(s): A. Gilbert

      • Abstract

      Background
      The Victorian Lung Cancer Registry (VLCR) pilot project was established in January 2011 and aims to recruit all newly diagnosed lung cancer cases across participating sites in Victoria. Case ascertainment for the registry is derived from institutional ICD-10 coding with subsequent assessment against inclusion criteria. Incomplete case ascertainment threatens data validity and several methods have been proposed for estimating this in cancer registries.

      Methods
      A quantitative, case finding audit was employed to evaluate the VLCR’s case ascertainment methodology at a major metropolitan hospital between 01/07/2011 and 30/06/2012. ICD-codes determined that 125 new cases were registered at the hospital. Lists of patients recorded or suspected to have a diagnosis of lung cancer were requested from the following institutional and external departments: Radiotherapy, Day Procedure Unit, Oncology Lung Multidisciplinary Team Meeting (MDM), Cardiothoracic Surgery (CTS), Victorian Cancer Registry (VCR) and Pathology. Comparisons were made between patients included in the registry and departmental lists provided. Medical records were then assessed to check eligibility of outstanding patients for inclusion in the registry.

      Results
      Six patient lists were compared with the VLCR. Excluding duplications and exclusions a total of 10 eligible patients had not been recruited by the registry. Investigations indicated that the underreporting of these cases was largely attributed to the use of the ICD10 R91 Code. This code is not a primary lung malignancy code and is assigned by clinical coders for abnormal findings on diagnostic imaging of the lung where lung cancer is suspected but not confirmed. Of the 10 patients eligible for inclusion in the registry, 7 were discharged with the R91 code and pending clinical confirmation were later included in the registry. The remaining 3 patients were not included in hospital data extracts as they were non admitted day patients and therefore not coded. A capture – recapture methodology is used to evaluate ascertainment completeness.

      Conclusion
      The completeness of cancer registry data – the extent to which all of the incident cancers occurring in the population are included in the registry database – is an extremely important attribute of a cancer registry. Only a high degree of completeness in case-finding procedures will ensure cancer incidence rates and survival proportions are close to their true value.[i] It was identified that a possible 7.5% of total lung cancer patients were not being captured by the VLCR recruitment method. The inclusion of the R Code in hospital ICD code extracts will increase the VLCR ascertainment rate. [i] Parkin, D. M. and F. Bray (2009). "Evaluation of data quality in the cancer registry: Principles and methods Part II. Completeness." European Journal of Cancer 45(5): 756-764.