Volume 8, Number 12—December 2002
Research
Using Automated Health Plan Data to Assess Infection Risk from Coronary Artery Bypass Surgery
Table 1
Characteristics of the population of patients of five health-care plans in Massachusetts, January 1996–March 1999
| Plan 1 | Plan 2 | Plan 3 | Plan 4 | Plan 5 | All | p valuea | |
|---|---|---|---|---|---|---|---|
| Procedures | 161 | 584 | 635 | 363 | 210 | 1,953 | |
| Age (range)a | 67 (61–73) | 59 (53–64) | 60 (54–65) | 59 (53–63) | 72 (68–75) | 61 (55–67) | <0.0001 |
| Sex, % male | 77 | 78 | 83 | 80 | 60 | 78 | <0.0001 |
| Prescriptions (range) during the 180 days before surgeryb | 14 (9-22) | 18 (10-28) | 18 (12-27) | 9 (6-12) | 14 (7-21) | 15 (9-24) | 0.0001 |
| Chronic disease score (range)b | 2,474 (1,774–3,704) | 2,211 (1,403– 3,456) | 2,378 (1,506–3,662) | 2,156 (1,372–3,350) | 2,369 (1,432–3,376) | 2,283 (1,446–3,511) | 0.01 |
| Prescriptions (range) during the 30 days after surgery b | 5 (3–7) | 6 (4–8) | 4 (3–6) | 5 (3–7) | 3 (0–5) | 5 (3–7) | <0.0001 |
| Days (range) with ambulatory care claims b | 1 (1–2) | 10 (7–13) | 6 (3–8) | 11 (8–16) | 14 (10–18) | 8 (5–13) | 0.0001 |
| Diagnoses on days with ambulatory claims (average) | 2.6 | 2.6 | 2.2 | 1.9 | 1.5 | 2.2 | |
| Patients with specified indicator of infection % (no. of patients): | |||||||
| Diagnosis during index hospitalization | 6 (9) | 3 (17) | 2 (14) | 2 (7) | 1 (2) | 3 (49) | 0.05 |
| Diagnosis in ambulatory setting (excludes emergency room) | 12 (19) | 15 (89) | 17 (105) | 11 (39) | 13 (28) | 14 (280) | NS |
| Diagnosis in emergency rooms | 2 (3) | 3 (19) | 2 (12) | 2 (6) | 1 (3) | 2 (43) | NS |
| Diagnosis during second hospitalization, includes extended care facilities | 9 (14) | 7 (41) | 2 (15) | 2 (6) | 0 (1) | 4 (77) | <0.01 |
| Antistaphylococcal antibiotic in ambulatory setting | 16 (26) | 18 (107) | 13 (81) | 14 (52) | 12 (25) | 15 (291) | 0.05 |
| Wound culture performed | 2 (3) | 2 (12) | 5 (33) | 6 (22) | 1 (2) | 4 (72) | <0.01 |
| Wound care |
0 (0) |
1 (4) |
2 (15) |
1 (5) |
1 (2) |
1 (26) |
0.052 |
| aKruskal-Wallis or chi-square tests of the null hypothesis of no difference between plans. bMedian (interquartile range). | |||||||


