932
the electronic program and a comprehensive audit of all the cases performed by the infection preventionists (IPs). Mean rates of voluntary reporting for the 38 categories of surgery during the study period were 16.7% in 2013 and 19.6% in 2014. To detect cases of SSI, IPs reviewed the electronic medical records of all patients who underwent target surgical procedures during the study period, including physicians’ progress notes, nursing records, microbiology and radiology reports, antimicrobials administered, reoperation records and discharge summaries. The medical records of readmissions, emergency visits, and outpatient visits after discharge were also reviewed. The CDC NHSN criteria for SSIs were used, and SSIs were classified as superficial inci- sional, deep incisional, or organ/space SSIs.16 The surveillance periods after surgery were 30 days for surgeries without prosthetic implants and 90 days for surgeries with prosthetic implants. Surveillance data were collected prospectively on all patients during the surveillance period. For each case of SSI identified by the IP audit, queries were sent to the surgeon responsible for the operation and further information was collected. However, the surgeons did not have a chance to override the decision based on the NHSN criteria.
Semiautomated SSI surveillance using electronic screening algorithms
To reduce the chart review workload of the IPs, electronic screening algorithms were developed. The algorithms for detec- tion of probable SSI events comprised 3 criteria: (1) antibiotics were ordered after postoperative day 5; (2) microbial cultures were done; and (3) an inpatient infectious disease (ID) specialist consultation was requested during the postoperative surveillance period. To exclude prolonged administration of antibiotics for prophylactic purposes, only the cases in which antibiotics were administered 5 days after surgery were included in the criteria. For the microbial culture criteria, the specimens obtained at the time of the surgery were excluded. All data used by screening algorithms were automatically extracted from a hospital information system. Cases that met at least 1 of the criteria were flagged in the SSI Surveillance Registry and were subse- quently reviewed by the IPs. Since July 2016, this semiautomated SSI surveillance system has been in place for 38 categories of surgery.
Validation of the semiautomated SSI surveillance system
The dataset for all surgical procedures under conventional SSI surveillance between January 2013 and December 2014, stored in the SSI Surveillance Registry, was used to verify the validity of the semiautomated SSI surveillance system. The conventional SSI surveillance was used as a reference for the calculation of sensi- tivity, positive predictive value (PPV), and specificity of the new surveillance method. The validity of the semiautomated SSI sur- veillance system was analyzed according to wound class, type of SSI, and SSI risk index.17 The impact of the new surveillance method on workload was determined by counting the number of flagged cases requiring chart review by the IPs and by calculating the estimated time spent on chart review for all flagged cases. The estimated time spent on chart review for each case of surgical procedure requiring a 30-day postoperative surveillance period was calculated by applying the following criteria based on the actual experiences of the IPs: (1) 15 minutes for SSI cases in clean
Sun Young Cho et al
or clean-contaminated surgery; (2) 3 minutes for non-SSI cases in clean or clean-contaminated surgery; (3) 25 minutes for SSI cases in contaminated or dirty-contaminated surgery; and (4) 5 min- utes for non-SSI cases in contaminated or dirty-contaminated surgery. For surgical procedures requiring a 90-day postoperative surveillance period, the estimated time required for chart review was calculated at twice the aforementioned values.
Results
Among the 40,516 surgical procedures, a total of 575 SSI events (1.42%) were identified by conventional SSI surveillance. The SSI rate was 0.67% (131 of 19,672) among clean surgeries; the SSI rate was 2.09% (433 of 20,736) among clean-contaminated surgeries; and the SSI rate was 10.19% (11 of 108) among contaminated or dirty-contaminated surgeries. Using the electronic screening algorithms, 13,511 of the 40,516 surgeries (33.3%) were flagged for chart review by the IPs, and 556 (1.37%) were confirmed as SSI cases (Table 1). The probable SSI cases flagged by the algo- rithms included 5,599 cases (28.5%) among clean surgeries, 7,820 cases (37.7%) among clean-contaminated surgeries, and 92 cases (85.2%) among contaminated or dirty-contaminated surgeries. The sensitivity of the semiautomated SSI surveillance system was 96.7% (556 of 575 cases). The sensitivity of SSI detection was 98.5% for clean surgeries; the sensitivity of SSI detection was 96.1% for clean-contaminated surgeries; and the sensitivity of SSI detection was 100% for contaminated or dirty-contaminated surgeries. The sensitivity, specificity, and PPV for each procedure are shown in Supplementary Table 1. The SSI cases that had not been identified by the electronic screening algorithms included 17 cases of superficial incisional SSI among clean-contaminated sur- geries and 2 cases of superficial incisional SSI among clean surgeries. In the 17 unidentified cases of SSI, treatment depended only on wound dressing, without using antibiotics. In addition, 2 other cases of SSI were diagnosed at other hospitals after discharge. The first criterion (antibiotic prescription) used by the
screening algorithms flagged a total of 10,910 cases, including 516 SSI cases. The second criterion (microbial culture) flagged an additional 2,474 cases, including 40 SSI cases. The third criterion (consultation with an ID specialist) flagged an additional 127 cases but no SSI cases (Fig. 1). The performance of each criterion for the detection of probable SSI events is shown in Table 2. The SSI detection sensitivity of antibiotic prescription was 89.7% (516 of 575 cases); the SSI detection sensitivity of microbial culture was 89.0% (512 of 575 cases); and the SSI detection sensitivity of ID specialist consultation was 53.7% (309 of 575 cases). Using screening algorithms, the PPV was 4.1%. The third criterion showed the highest PPV (15.1%) among the 3 criteria. The cases flagged by the electronic screening algorithms
included 16 of 17 surgical cases (94.1%) with a risk index of 3, 833 of 1,108 (75.2%) cases with a risk index of 2; 5,626 of 10,262 (54.8%) cases with a risk index of 1; and 7,036 of 29,129 (24.2%) cases with a risk index of 0 (Table 3). The SSI detection sensitivity was 100% for surgery with a risk index of 2 or 3, 98.8% for surgery with a risk index of 1, and 94.4% for surgery with a risk index of 0.
Semiautomated SSI surveillance using electronic screening algorithms reduced the number of cases for chart review from 20,258 to 6,756 per year (a 66.7% decrease) and reduced the chart review workload of the IPs from 1,283 to 482 person hours per year (a 62.4% decrease).
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124 |
Page 125 |
Page 126 |
Page 127 |
Page 128 |
Page 129 |
Page 130 |
Page 131 |
Page 132 |
Page 133 |
Page 134 |
Page 135 |
Page 136 |
Page 137 |
Page 138 |
Page 139 |
Page 140
