Inter-observer Variability in Performance Status Assessment in Cancer Patients
Author Information
Author(s): J.B. Sørensen, M. Klee, T. Palshof, H.H. Hansen
Primary Institution: Finsen Institute/Rigshospitalet
Hypothesis
The study aims to evaluate the reliability of the ECOG Scale of Performance Status by measuring the non-chance agreement between three oncologists.
Conclusion
The reliability of the ECOG Scale of Performance Status is fair, with higher agreement in patients with good performance status compared to those with poor performance status.
Supporting Evidence
- Total unanimity was observed in 40 cases among the three observers.
- Unanimity between two observers occurred in 53 cases.
- Total disagreement was noted in seven cases.
- The overall Kappa statistic was 0.44, indicating moderate agreement.
- Agreement was higher for patients with ECOG scores 0-2 compared to those with scores 3-4.
Takeaway
Doctors sometimes disagree on how well cancer patients are doing, but they usually agree when patients are doing well.
Methodology
Three oncologists assessed the performance status of 100 consecutive cancer patients using the ECOG Scale, with evaluations made on the same day and without knowledge of each other's ratings.
Potential Biases
Potential systematic bias due to the order of patient interviews, although the distribution of scores did not indicate significant bias.
Limitations
The study only included hospitalized patients, which may not represent the general cancer patient population.
Participant Demographics
The study included 100 cancer patients with an equal sex distribution (49 males and 51 females) and various cancer types.
Statistical Information
P-Value
0.44
Confidence Interval
(0.38-0.51)
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website