Table 1. Patient Characteristics*.
| All (N = 12933) | NLP Available (N = 9783) | Validation Sample (N = 500) | ||||
|---|---|---|---|---|---|---|
| HIV+ | HIV- | HIV+ | HIV- | HIV+ | HIV- | |
| N | 3487 | 9446 | 2868 | 6915 | 250 | 250 |
| Age, mean (SD) | 44.5 (10.6) | 43.2 (10.7) | 44.7 (10.6) | 43.3 (10.5) | 44.4 (10.3) | 43.4 (9.2) |
| Female gender, N (%) | 1121 (32) | 3511 (37) | 956 (33) | 2817 (41) | 96 (38) | 91 (36) |
| Race | ||||||
| Caucasian, N (%) | 1802 (52) | 4565 (48) | 1485 (52) | 3293 (48) | 122 (49) | 107 (43) |
| African-American, N (%) | 722 (21) | 1955 (21) | 598 (21) | 1473 (21) | 58 (23) | 56 (22) |
| Hispanic, N (%) | 622 (18) | 1695 (18) | 534 (19) | 1364 (20) | 51 (20) | 62 (25) |
| Other/Unknown, N (%) | 341 (10) | 1231 (13) | 251 (9) | 785 (11) | 19 (8) | 25 (10) |
| Cardiovascular risk factor, N (%) | 1880 (54) | 4268 (45) | 1669 (58) | 3793 (55) | 123 (49) | 147 (59) |
| Hypertension, N (%) | 975 (28) | 2895 (31) | 888 (31) | 2595 (38) | 59 (24) | 93 (37) |
| Diabetes, N (%) | 612 (18) | 1325 (14) | 538 (19) | 1203 (17) | 47 (19) | 45 (18) |
| Dyslipidemia, N (%) | 1394 (40) | 2899 (31) | 1264 (44) | 2637 (38) | 96 (38) | 104 (42) |
| Mood disorder, N (%) | 1481 (43) | 3034 (32) | 1361 (47) | 2701 (39) | 119 (48) | 104 (42) |
| Depression, N (%) | 1224 (35) | 2247 (24) | 1128 (39) | 2025 (29) | 101 (40) | 77 (31) |
| Anxiety, N (%) | 857 (25) | 2094 (22) | 801 (28) | 1910 (28) | 68 (27) | 69 (28) |
| Bipolar disorder, N (%) | 261 (7) | 504 (5) | 242 (8) | 435 (6) | 21 (8) | 17 (7) |
| Schizophrenia, N (%) | 102 (3) | 245 (3) | 94 (3) | 199 (3) | 7 (3) | 13 (5) |
| Pharmacologic smoking cessation, N (%) | 266 (8) | 350 (4) | 255 (9) | 341 (5) | 19 (8) | 9 (4) |
| Nicotine replacement therapy use, N (%) | 131 (4) | 215 (2) | 127 (4) | 211 (3) | 7 (3) | 8 (3) |
| Varenicline use, N (%) | 169 (5) | 195 (2) | 161 (6) | 189 (3) | 13 (5) | 3 (1) |
| ART use, N (%) | 1705 (49) | — | 1523 (53) | — | 139 (56) | — |
| CD4 cell count, mean (SD) | 501 (326) | — | 497 (327) | — | 442 (316) | — |
| CD4 cell count <200/mm3, N (%) | 330 (16) | — | 292 (17) | — | 34 (22) | — |
| HIV RNA (log-transformed), mean (SD) | 8.6 (2.7) | — | 8.5 (2.7) | — | 8.5 (2.6) | — |
| HIV RNA <400 copies/ml, N (%) | 2975 (85) | — | 2432 (85) | — | 210 (84) | — |
| Encounters/year, median (IQR) | 6.3 (3.0–11.5) | 6.5 (3.1–13.3) | 6.8 (3.5–11.6) | 6.6 (3.5–12.0) | 7.1 (3.3–12.6) | 6.6 (4.0–12.5) |
| Inpatient | 0.1 (0–0.2) | 0.1 (0–0.3) | 0.1 (0–0.3) | 0.1 (0–0.4) | 0.1 (0–0.3) | 0.1 (0–0.3) |
| Outpatient | 6.0 (2.7–11.0) | 5.9 (2.8–11.7) | 6.4 (3.2–11.1) | 6.1 (3.1–11.2) | 6.4 (3.0–12.2) | 6.3 (3.6–11.5) |
| Years in health care system, median (IQR) | 8.4 (3.6–12.1) | 7.7 (2.8–11.9) | 8.9 (4.5–12.4) | 9.2 (4.6–12.4) | 8.4 (3.6–12.1) | 9.0 (3.7–12.1) |
* ICD codes: hypertension = 401.xx; diabetes = 250.xx; dyslipidemia = 272.xx; depression = 311.xx, 296.2, 296.3; anxiety = 300.xx; bipolar disorder = 296.0, 296.1, 296.4–296.8; schizophrenia = 295.xx. NLP = natural language processing; SD = standard deviation; ART = antiretroviral therapy; IQR = inter-quartile range