Table 1.
Total texts | Texts for medical doctors | Texts for patients | p-value | |
---|---|---|---|---|
N | 128 | 64 | 64 | |
Humanness, median (range) | 5 (4–5) | 5 (4–5) | 5 (4–5) | 0.130 |
Correctness, median (range) | 4.25 (2–5) | 4.5 (2–5) | 4 (2–5) | 0.976 |
Relevance, median (range) | 4 (3–5) | 4 (3–5) | 4 (3–5) | 0.808 |
Flesch-Kincaid readability index, mean (SD) | 39.19 (25.83) | 15.26 (8.86) | 63.14 (10.11) | < 0.001 |
Fleisch-Kincaid grade level, mean (SD) | 12.83 (4.02) | 16.45 (2.01) | 9.23 (1.50) | 0.003 |
Gunning fog, mean (SD) | 15.99 (4.62) | 20.17 (2.18) | 11.83 (1.75) | 0.001 |
Coleman-Liau index, mean (SD) | 14.68 (4.06) | 18.31 (2.05) | 11.04 (1.47) | 0.001 |
Simple Measure of Gobbledygook index, mean (SD) | 11.50 (3.48) | 14.57 (1.76) | 8.44 (1.50) | < 0.001 |
Note: this table reflects the obtained evaluations of humanness, correctness and relevance of multiple sclerosis texts generated by ChatGPT as evaluated by three medical doctors and readability scores