Table 2.
Social determinants of health (SDoH) concepts identified from the lung cancer patient cohort.
| SDoH category | Total number of concepts detected by NLP | Total number of patients has at least one SDoH | Percentage of patients has at least one SDoH for current category (%) |
|---|---|---|---|
| Gender | 843,066 | 9,552 | 98.7 |
| Alcohol use | 223,214 | 9,195 | 95.0 |
| Drug use | 180,309 | 8,756 | 90.5 |
| Marital status | 167,457 | 8,655 | 89.5 |
| Education | 167,018 | 8,463 | 87.5 |
| Occupation | 142,306 | 8,345 | 86.3 |
| Smoking | 132,833 | 7,639 | 79.0 |
| Race | 144,980 | 7,376 | 76.2 |
| Ethnicity | 86,789 | 5,231 | 54.1 |
| Language | 83,539 | 5,173 | 53.5 |
| Physical activity | 55,842 | 3,092 | 32.0 |
| Transportation | 24,191 | 2,877 | 29.7 |
| Financial constraint | 113,220 | 2,766 | 28.6 |
| Social cohesion | 9,170 | 2,727 | 28.2 |
| Employment status | 843,066 | 2,110 | 21.8 |