Reliability of Interview Data in Lung Cancer Study
Author Information
Author(s): Huang Desheng, Guan Peng, Shi Hailong, He Qincheng, Zhou Baosen
Primary Institution: China Medical University
Hypothesis
The study investigates the reliability of epidemiological data obtained in a non-smoking female lung cancer case-control study.
Conclusion
The study identified variability in interview data collection and proposed strategies to improve data quality.
Supporting Evidence
- Kappa values were all more than 0.5 in all studied indexes.
- Errors in collecting and classifying data were common for complicated clinical events.
- Agreement was excellent for family history of cancer and history of passive smoking.
Takeaway
This study looked at how reliable answers from interviews are when asking women about lung cancer, and found some answers were better than others.
Methodology
Fifty-six pairs of cases and controls were re-interviewed to compare responses and calculate Kappa values for agreement.
Potential Biases
Interviewer misinterpretation and subject disagreement were noted as sources of variability.
Limitations
Errors in data collection and coding were identified, particularly for complex clinical events.
Participant Demographics
Participants were non-smoking females aged 35–74, with cases newly diagnosed with lung cancer.
Statistical Information
P-Value
0.03
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website