03 July 2025 : Database Analysis
Can ChatGPT Provide Patient-Friendly and Reliable Information on Cervical Cancer Screening? A Study of ChatGPT-Generated Information in Polish
Adriana A. Michalska ABCDEF 1, Małgorzata M. Stefaniak
BCDF 2, Joanna Gotlib-Małkowska ADE 1*
DOI: 10.12659/MSM.947992
Med Sci Monit 2025; 31:e947992
Table 4 Ordinal regression parameters for predictors of readability.
| Predictor variable | b | SE | Z | p | OR | 95% CI | |
|---|---|---|---|---|---|---|---|
| Lower | Upper | ||||||
| Percentage of verbs | −0.413 | 0.089 | −4.63 | <0.001 | 0.662 | 0.551 | .783 |
| Number of difficult words | 0.070 | 0.031 | 2.27 | 0.023 | 1.073 | 1.01 | 1.142 |
| Percentage of nouns | 0.291 | 0.055 | 5.27 | <0.001 | 1.337 | 1.205 | 1.498 |
| Mean word length [in syllables] | 9.792 | 2.270 | 4.31 | <0.001 | 1.788 | 2.550 | 1.94E+06 |
| Percentage of difficult words | 0.221 | 0.103 | 2.16 | 0.031 | 1.248 | 1.025 | 1.536 |
| FOG Index: complex headwords | 0.562 | 0.118 | 4.76 | <0.001 | 1.753 | 1.405 | 2.237 |
| b – non-standardised regression coefficient; SE – standard error; OR – odds ratio; 95% CI – 95% confidence interval; FOG Index – readability index. | |||||||