Volume 30, Number 7—July 2024
Research
Electronic Health Record Data for Lyme Disease Surveillance, Massachusetts, USA, 2017–2018
Table 1
Sensitivity and positive predictive values of candidate criteria for a Lyme disease algorithm at 2 clinical practice groups in Massachusetts, USA, 2017–2018*
| Strata | Algorithm components |
Charts reviewed/ flagged |
No. Lyme disease cases† |
Summary performance |
|||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ICD codes | + EIA | + WB | ABX | PPV, % (95% CI)‡ | No. projected Lyme disease cases§ | Sensitivity, % (95% CI)# | |||||||||
| Site 1 | Site 2 | C | P | S | |||||||||||
| 1 |
1 |
18/90 |
20/52 |
3 |
1 |
0 |
10.5 (1.5–21.6) |
14.9 |
6.5 (1.7–13.2) |
||||||
| 2 |
1 |
1 |
20/53 |
20/105 |
22 |
3 |
0 |
62.5 (47.4−77.0) |
98.8 |
43.1 (35.8–49.2) |
|||||
| 3 |
1 |
12/12 |
19/65 |
0 |
0 |
1 |
3.2 (0.0–12.0) |
2.5 |
1.1 (0.0–4.3) |
||||||
| 4 |
1 |
1 |
0 |
1/3 |
0 |
0 |
0 |
0.0 (0.0–85.0) |
0.0 |
0.0 (0.0 −1.3) |
|||||
| 5 |
1 |
1 |
5/5 |
12/18 |
0 |
0 |
0 |
0.0 (0.0–10.4) |
0.0 |
0.0 (0.0–1.2) |
|||||
| 6 |
1 |
1 |
1 |
2/2 |
7/11 |
2 |
0 |
0 |
22.2 (0.0–51.8) |
2.9 |
1.3 (0.0 −3.0) |
||||
| 7 |
1 |
1 |
9/9 |
15/15 |
2 |
6 |
16 |
100.0 (92.5–100.0) |
24.0 |
10.5 (8.3–10.9) |
|||||
| 8 |
1 |
1 |
1 |
6/6 |
3/5 |
1 |
3 |
5 |
100.0 (81.5–100.0) |
11.0 |
4.8 (3.7–5.4) |
||||
| 9 |
1 |
1 |
1 |
12/15 |
5/9 |
1 |
9 |
7 |
100.0 (89.6–100.0) |
24.0 |
10.5 (8.8–11.8) |
||||
| 10 | 1 | 1 | 1 | 1 | 8/29 | 15/22 | 14 | 5 | 4 | 100.0 (92.2–100.0) | 51.0 | 22.3 (19.2–25.1) | |||
*ABX, antibiotics; C, confirmed; EIA, enzyme immunoassay; ICD, International Classification of Diseases, 10th Revision, Clinical Modification; P, probable; PPV, positive predictive value; S, suspected; WB, Western blot. †Based on criteria from the Council of State and Territorial Epidemiologists and Centers for Disease Control and Prevention. ‡PPV for each stratum was calculated by summing up the total number of confirmed, probable, and suspected cases for the stratum and then dividing by the total number of charts reviewed for the stratum. §The total number of projected Lyme disease cases for each stratum was calculated by multiplying the total number of charts flagged in the stratum by the positive predictive value for the stratum. #Sensitivity was calculated as the total number of projected Lyme disease cases for the criteria of interest divided by the total number of projected Lyme disease cases for the entire population. The total number of projected Lyme disease cases for the entire population is 229.1, which is the sum of the projected Lyme disease cases from each of the 10 unique strata in the top half of the table using unrounded numbers.