Page 131 - AIH-1-2

Artificial Intelligence in Health SDoH in clinical narratives

Figure 3. Adjusted odds ratios for the probability of mentioning marital status based on clinical case type, journal specialty, journal’s geographic region,
and author’s geographic region. The figure was plotted with Matplotlib.

(AOR: 0.23; 95% CI: 0.15 – 0.65), and journals specializing 3.4.5. Homelessness
in ophthalmology (AOR: 0.12; 95% CI: 0.02 – 0.92) and Mentions of homelessness were strongly associated with
dermatology (AOR: 0.28; 95% CI: 0.09 – 0.90). Diagnoses journals in the field of forensic medicine (AOR: 14.92; 95%
pertaining to cardiovascular diseases (AOR: 0.43; 95% CI: 5.48 – 40.64). Other strongly correlated factors included
CI: 0.27 – 0.69) and neoplasms (AOR: 0.43; 95% CI: 0.31 journals in the areas of pathology (AOR: 3.95; 95% CI: 1.39
– 0.63) also displayed marked negative associations with – 11.28) and infectious diseases (AOR: 3.75; 95% CI: 1.77 –
immigrant status mentions. Both general medicine journals 7.94), publications from Eastern Europe (AOR: 4.76; 95%
(AOR: 1.75; 95% CI: 1.34 – 2.29) and authors from North CI: 1.88 – 12.03), and diagnoses related to infections (AOR:
America (AOR: 1.53; 95% CI: 1.17 – 2.01) demonstrated 6.36; 95% CI: 3.57 – 11.32), mental disorders (AOR: 5.80;
moderate positive associations with mentions of immigrant 95% CI: 2.26 – 14.89), and injuries (AOR: 4.73; 95% CI: 2.29
status. Further details on immigrant status mentions are – 9.77). Further information on homelessness mentions is
available in Figure 5 and Table S4. available in Figure 6 and Table S5.

Volume 1 Issue 2 (2024) 125 doi: 10.36922/aih.2737