Influence of number of surfaces and observers on statistical power in a multiobserver ROC radiographic caries detection study.

The aim of the study was to evaluate the influence of the number of surfaces (N(SURF)) and the number of observers (N(OBS)) on the statistical power of a study comparing the diagnostic accuracies of radiographic systems used for approximal caries lesion detection. A data set consisting of 338 surfac...

Descripción completa

Detalles Bibliográficos
Publicado en:Caries Research Vol. 37; no. 3; pp. 200 - 206
Autores principales: Hintze H, Frydenberg M, Wenzel A, Hintze, Hanne, Frydenberg, Morten, Wenzel, Ann
Formato: equations & formulas research tables/charts Journal Article
Publicado: Karger AG May/Jun2003
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=106883133&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 106883133
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        00086568
        DEU
      jtl: Caries Research
      issn: 00086568
      maglogo: N
    pubinfo:
      dt: May/Jun2003
      vid: 37
      iid: 3
      pid: 2485
      pub: Karger AG
    artinfo:
      ui:
        106883133
        NLM12740544
        2003151614
        10.1159/000070445
        NLM12740544
        106883133
      ppf: 200
      ppct: 6
      formats:
      tig:
        atl: Influence of number of surfaces and observers on statistical power in a multiobserver ROC radiographic caries detection study.
      aug:
        au:
          Hintze H
          Frydenberg M
          Wenzel A
          Hintze, Hanne
          Frydenberg, Morten
          Wenzel, Ann
        affil: Department of Oral Radiology, Faculty of Health Sciences, University of Aarhus, Aarhus, Denmark
      sug:
        subj:
          Dental Caries Radiography
          Radiography, Dental Methods
          Research Methodology Evaluation
          Comparative Studies
          Confidence Intervals
          Descriptive Statistics
          Evaluation Research
          Reproducibility of Results
          Two-Way Analysis of Variance
          Funding Source
          Human
      ab: The aim of the study was to evaluate the influence of the number of surfaces (N(SURF)) and the number of observers (N(OBS)) on the statistical power of a study comparing the diagnostic accuracies of radiographic systems used for approximal caries lesion detection. A data set consisting of 338 surfaces examined by 10 independent observers using four radiographic systems was available. The presence of a caries lesion was assessed from a 5-point confidence scale. The true lesion diagnosis was established by histological validation. ROC curve areas (A(z)s) were used to express the diagnostic accuracy of the observers with the radiographic systems. Assuming that the A(z)s were tested by a two-way analysis of variance, we performed a simulation study in order to evaluate how the power of this statistical analysis depended on N(SURF) and N(OBS). As a measure of the statistical power we used the standard error of the difference between the expected A(z)s of two systems. The simulations were made with N(SURF) in the range from 25 to 338 and N(OBS) from 2 to 10. The simulations showed that the power increased as a function of the total number of evaluations per system (N(SURF) x N(OBS)), but how this number was attained in relation to the number of surfaces and observers had only marginal influence on the power. Thus, from a statistical point of view it may be concluded, provided that data are analyzed by a two-way analysis of variance, that study designs for comparing the accuracy of several systems can be composed freely in relation to the number of surfaces and observers as long as the total number of evaluations per system are identical.
      pubtype: Academic Journal
      doctype:
        equations & formulas
        research
        tables/charts
        Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N