Table 3.
Characteristics of the 13 articles on using SDoH for risk prediction
| Characteristics | n | % |
|---|---|---|
| Publication year | ||
| 2017-2020 | 11 | 85 |
| 2012-2016 | 2 | 15 |
| Major SDoH data source | ||
| ACS or Census | 6 | 46 |
| Other health surveys or databases | 2 | 15 |
| EHRs | 5 | 38 |
| SDoH level | ||
| Neighborhood | 6 | 46 |
| Individual | 5 | 38 |
| Both | 2 | 15 |
| Sample size | ||
| <10 000 | 5 | 38 |
| Between 10 000 and 100 000 | 2 | 15 |
| >100 000 | 6 | 46 |
| Outcome | ||
| Disease incidence/diagnosis/health outcome | 3 | 23 |
| Use of healthcare services | 8 | 62 |
| Referral/intervention/prevention of health services | 2 | 15 |
| SDoH contribution to improved prediction | ||
| Contributed | 8 | 62 |
| Minimal contribution | 5 | 38 |
ACS: American Community Survey; EHR: electronic health record; SDoH: Social Determinants of Health.