Skip to main content
. 2024 Apr 3;271(7):4057–4066. doi: 10.1007/s00415-024-12328-x

Table 3.

Multivariate logistic regressions

All questions Question 1 Question 2 Question 3 Question 4
IRR (95% CI) p-value OR (95% CI) p-value OR (95% CI) p-value OR (95% CI) p-value OR (95% CI) p-value
Age, years 0.99 (0.99–1.00) 0.381 0.98 (0.97–0.99) 0.021 1.00 (0.99–1.02) 0.273 0.98 (0.97–1.00) 0.066 1.00 (0.98–1.01) 0.796
Sex (Ref. other)
 Female 1.48 (0.68–3.19) 0.316 3.57 (0.50–25.60) 0.204 0.59 (0.06–5.83) 0.659 2.31 (0.20–26.06) 0.497 0.92 (0.70–1.21) 0.575
 Male 1.53 (0.71–3.31) 0.277 4.23 (0.58–30.47) 0.152 0.55 (0.05–5.46) 0.617 2.73 (0.24–31.03) 0.416 0.82 (0.81–1.35) 0.574
Education (Ref. high school graduate)
 Elementary school 0.85 (0.49–1.48) 0.576 1.46 (0.26–8.15) 0.662 0.29 (0.06–1.39) 0.124 2.64 (0.54–12.85) 0.229 0.17 (0.02–1.47) 0.108
 Middle school 1.02 (0.90–1.17) 0.675 1.16 (0.74–1.79) 0.506 0.84 (0.56–1.25) 0.398 1.19 (0.80–1.76) 0.379 1.11 (0.75–1.66) 0.087
 College graduates 0.87 (0.79–0.95)  < 0.01 0.64 (0.48–0.85)  < 0.01 0.72 (0.55–0.95) 0.019 0.80 (0.61–1.05) 0.108 0.78 (0.59–1.03) 0.584
Duration of MS disease, years 1.00 (0.99–1.00) 0.711 1.00 (0.98–1.01) 0.736 0.99 (0.98–1.01) 0.711 1.01 (0.99–1.02) 0.066 0.99 (0.98–1.01) 0.502
MS clinical descriptors (Ref. “PPMS”)
 RRMS 0.86 (0.72–1.03) 0.114 0.45 (0.22–0.90) 0.025 0.66 (0.37–1.19) 0.169 0.78 (0.44–1.37) 0.460 0.93 (0.52–1.64) 0.804
 SPMS 0.90 (0.72–1.13) 0.383 0.36 (0.16–0.80) 0.012 0.83 (0.41–1.68) 0.610 0.47 (0.23–0.96) 0.038 2.19 (1.10–4.35) 0.025
 “I don’t know” 0.91 (0.74–1.11) 0.365 0.44 (0.20–0.93) 0.032 0.78 (0.41–1.48) 0.451 0.79 (0.42–1.47) 0.460 1.31 (0.70–2.47) 0.393
Depression 0.94 (0.86–1.02) 0.173 0.67 (0.51–0.89)  < 0.01 1.08 (0.84–1.40) 0.526 0.81 (0.63–1.05) 0.127 0.93 (0.71–1.27) 0.624
Subjective memory and attention deficit 1.03 (0.94–1.13) 0.423 0.79 (0.56–1.04) 0.097 1.39 (1.06–1.81) 0.014 0.84 (0.64–1.09) 0.203 1.45 (1.11–1.90)  < 0.01

The main outcome (likelihood of selecting the ChatGPT response) was evaluated through a stratified analysis employing logistic regression models, by assessing the frequency of selecting the ChatGPT option divided by the total number of questions. This approach was adopted to address variations in the nature of the questions. Covariates were age, sex, treatment duration, clinical descriptors, self-reported cognitive deficit, depressive symptoms and educational attainment

CI Confidence interval, IRR incidence rate ratio, MS multiple sclerosis, OR odds ratio, PPMS primary progressive multiple sclerosis, RRMS relapsing–remitting multiple sclerosis, SPMS secondary-progressive multiple sclerosis; the use of bold formatting within the table was employed to highlight significant values