Interval Estimation under Two Study Designs for Kappa with Binary Classifications

Hale, Cecilia A.; Fleiss, Joseph L.

doi:10.2307/2532564

Cited by 74 publications

(53 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fifty-five percent of the cases respondent were spouses, as were 47% of control respondents. There was no difference between spouse and other respondents with respect to distributions of the quality of the [Hale and Fleiss, 1993]. b 95% CI by the logit method with 1 /2 correction [Gart and Zweifel, 1962;Cox, 1970].…”

Section: Resultsmentioning

confidence: 99%

“…The Mantel and Haenszel [1959] procedure was used to adjust odds ratios for type of next-of-kin (spouse and other) interviewed and for quality of interview (highly reliable, generally reliable, and questionable/unreliable). To measure agreement among exposure assessment methods, separately for cases and controls, we computed pair-wise kappa statistics [Cohen, 1960;Fleiss, 1981] and corresponding 95% confidence intervals [Hale and Fleiss, 1993], and a three-way kappa measure of assessment [Shrout and Fleiss, 1979]. To assess agreement, we also computed marginal odds ratios among the exposure measurements as well as a pair-wise conditional odds ratios, holding the third exposure constant.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of asbestos exposure assessments by next-of-kin respondents, by an occupational hygienist, and by a job-exposure matrix from the National Occupational Hazard Survey

2005

View full text Add to dashboard Cite

Expert assessments were the most plausible, but the data indicate that disease associations could also be detected with the other exposure assessment methods. Using some combination of the proxy respondent's assessment and the JEM assessment, one can predict the expert's assessment. A strategy that relied on the respondent's assessment when it was positive and otherwise obtained an expert assessment could reduce costs with little error, compared to expert assessment on all subjects.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Comparison of asbestos exposure assessments by next-of-kin respondents, by an occupational hygienist, and by a job-exposure matrix from the National Occupational Hazard Survey

2005

View full text Add to dashboard Cite

show abstract

“…Kappa-values were assessed as: k#0.20, minor agreement; k50.21-0.40, fair agreement; k50.41-0.60, moderate agreement; k50.61-0.80, high agreement; and k50.81-1.00, almost perfect agreement [9,10]. Discrepancies regarding these evaluations between the two reviewers were resolved by consensus.…”

Section: Discussionmentioning

confidence: 99%

Thoracic duct and cisterna chyli: evaluation with multidetector row CT

Kiyonaga¹,

Mori²,

Matsumoto³

et al. 2012

The British Journal of Radiology

View full text Add to dashboard Cite

Objectives: The aim of this study was to evaluate the normal anatomy of the thoracic duct and cisterna chyli obtained by axial and multiplanar reformation (MPR) images of 1 mm slice thickness using multidetector row CT (MDCT). Methods: We evaluated the ability of MDCT to examine the normal anatomy of the thoracic duct and cisterna chyli. The axial and coronal images of thoracoabdominal MDCT images obtained in 50 patients (20 females and 30 males; mean age, 63.5 years; range, 32-81 years) were reviewed between January and October 2005. We excluded patients with malignant neoplasms, inflammation or vascular diseases (e.g. aortic aneurysm, aortic dissection) and those with a history of thoracoabdominal surgery. The thoracic duct was divided into three anatomical sections: the upper, middle and lower. We evaluated the degree of visualisation and the maximum size of the thoracic duct. We also evaluated the degree of visualisation, maximum size, configuration and location of the cisterna chyli. Results: Visualisation of the thoracic duct and cisterna chyli was almost 100% on axial and coronal images. The lower section of the thoracic duct was most clearly visualised among the three sections. There was little difference in the maximum size of the thoracic duct among the three sections. The cisterna chyli was most frequently located at the Th12 or L1 level, and the most common type was the ''straight thin tube type''. Conclusion: Axial and MPR images of 1 mm slice thickness using MDCT can clearly depict the thoracic duct and cisterna chyli.

show abstract

“…The reliability of ISH and Consensus PCR was determined using Fleiss' intra-class correlation coefficient (ICCC) [47,48]. The Kappa coefficients were divided into categories as described by Landis and Kock [49].…”

Section: Reliabilitymentioning

confidence: 99%

Evaluation of a combined triple method to detect causative HPV in oral and oropharyngeal squamous cell carcinomas: p16 Immunohistochemistry, Consensus PCR HPV-DNA, and In Situ Hybridization

Pannone

Rodolico

Santoro

et al. 2012

Infectious Agents and Cancer

View full text Add to dashboard Cite

Background: Recent emerging evidences identify Human Papillomavirus (HPV) related Head and Neck squamous cell carcinomas (HN-SCCs) as a separate subgroup among Head and Neck Cancers with different epidemiology, histopathological characteristics, therapeutic response to chemo-radiation treatment and clinical outcome. However, there is not a worldwide consensus on the methods to be used in clinical practice. The endpoint of this study was to demonstrate the reliability of a triple method which combines evaluation of: 1. p16 protein expression by immunohistochemistry (p16-IHC); 2. HPV-DNA genotyping by consensus HPV-DNA PCR methods (Consensus PCR); and 3 viral integration into the host by in situ hybridization method (ISH). This triple method has been applied to HN-SCC originated from oral cavity (OSCC) and oropharynx (OPSCC), the two anatomical sites in which high risk (HR) HPVs have been clearly implicated as etiologic factors. Methylation-Specific PCR (MSP) was performed to study inactivation of p16-CDKN2a locus by epigenetic events. Reliability of multiple methods was measured by Kappa statistics. Results: All the HN-SCCs confirmed HPV positive by PCR and/or ISH were also p16 positive by IHC, with the latter showing a very high level of sensitivity as single test (100% in both OSCC and OPSCC) but lower specificity level (74% in OSCC and 93% in OPSCC). Concordance analysis between ISH and Consensus PCR showed a faint agreement in OPSCC ( = 0.38) and a moderate agreement in OSCC ( = 0.44). Furthermore, the addition of double positive score (ISHpositive and Consensus PCR positive) increased significantly the specificity of HR-HPV detection on formalin-fixed paraffin embedded (FFPE) samples (100% in OSCC and 78.5% in OPSCC), but reduced the sensitivity (33% in OSCC and 60% in OPSCC). The significant reduction of sensitivity by the double method was compensated by a very high sensitivity of p16-IHC detection in the triple approach.

show abstract

Interval Estimation under Two Study Designs for Kappa with Binary Classifications

Cited by 74 publications

References 15 publications

Comparison of asbestos exposure assessments by next-of-kin respondents, by an occupational hygienist, and by a job-exposure matrix from the National Occupational Hazard Survey

Comparison of asbestos exposure assessments by next-of-kin respondents, by an occupational hygienist, and by a job-exposure matrix from the National Occupational Hazard Survey

Thoracic duct and cisterna chyli: evaluation with multidetector row CT

Evaluation of a combined triple method to detect causative HPV in oral and oropharyngeal squamous cell carcinomas: p16 Immunohistochemistry, Consensus PCR HPV-DNA, and In Situ Hybridization

Contact Info

Product

Resources

About