Main characteristics of included studies, stratified by targeted type of HAI
| Total | SSI | BSI | UTI | Pneumonia | Other | |
|---|---|---|---|---|---|---|
| N studies | 57 | 34 | 24 | 15 | 14 | 2 |
| (N comparisons) | (71) | (44) | (29) | (15) | (15) | (2) |
| Device-associated | 20 | – | 12 | 7 | 7 | 1 |
| ICU only | 5 | 1 | 3 | 2 | 3 | 0 |
| Type of reference standard | ||||||
| NHSN | 35 | 26 | 9 | 6 | 7 | 2 |
| (VA)SQIP | 6 | 2 | 6 | 2 | 3 | 0 |
| Clinical | 4 | 1 | 3 | 1 | 1 | 0 |
| Other | 12 | 5 | 6 | 6 | 3 | 0 |
| Application of administrative data | ||||||
| External quality assessment | 24 | 9 | 19* | 6 | 8 | 0 |
| Within hospital surveillance | 18 | 13 | 3 | 7 | 4 | 1 |
| Combined with other HAI indicators | 15 | 12 | 3 | 2 | 2 | 1 |
| Specific quality metric | ||||||
| PSI | 9 | 1 | 10 | 0 | 2 | 0 |
| HAC | 3 | 0 | 2 | 1 | 0 | 0 |
| PHC4 | 4 | 4 | 3 | 3 | 4 | 0 |
| Region of origin | ||||||
| USA | 44 (55) | 22 (29) | 19 (24) | 10 (10) | 9 (10) | 1 (1) |
| Europe | 8 (10) | 8 (9) | 4 (4) | 4 (4) | 4 (4) | 1 (1) |
| Other | 4 (6) | 4 (6) | 1 (1) | 1 (1) | 1 (1) | 0 (0) |
| High risk of bias on QUADAS domain | ||||||
| Patient selection | 1 (1) | 1 (1) | 1 (1) | 0 (0) | 1 (1) | 0 (0) |
| Index test | 0 (3) | 0 (1) | 0 (1) | 0 (0) | 0 (0) | 0 (0) |
| Reference standard | 19 (27) | 11 (18) | 6 (7) | 4 (4) | 2 (2) | 1 (1) |
| Flow | 19 (29) | 10 (18) | 8 (11) | 4 (4) | 3 (4) | 1 (1) |
| Verification pattern | ||||||
| Complete or random sample | 37 (42) | 23 (26) | 16 (18) | 11 (11) | 10 (10) | 1 (1) |
| Complete with discrepant analysis | 3 (6) | 3 (6) | 1 (2) | 1 (1) | 1 (2) | 0 (0) |
| Partial, based on index test only | 8 (8) | 2 (4) | 5 (7) | 2 (2) | 2 (2) | 0 (0) |
| Partial, based on index and other tests | 8 (12) | 6 (6) | 1 (1) | 1 (1) | 1 (1) | 1 (1) |
| Other or unclear | 1 (3) | 0 (2) | 1 (1) | 0 (0) | 0 (0) | 0 (0) |
| Data availability | ||||||
| Complete 2×2 table, by HAI type | 29 | 20 | 10 | 6 | 6 | 1 |
| Complete 2×2 table, HAI combined | 3 | 3 | 2 | 4 | 3 | 0 |
| Positive predictive value only, by HAI | 9 | 3 | 6 | 1 | 2 | 0 |
| Other | 9 | 2 | 5 | 3 | 3 | 0 |
| No data extraction possible | 7 | 6 | 1 | 1 | 0 | 1 |
Some studies presented multiple comparisons and/or assessed more than 1 type of HAI; the number of comparisons is shown in brackets.
*One study targeting external quality assessment using administrative data combined with other sources of data.
BSI, bloodstream infections; HAC, Healthcare-associated condition as defined by the Centers for Medicare and Medicaid Services; HAI, healthcare-associated infections; ICU, intensive care unit; NHSN, National Healthcare Safety Network; PHC4, Pennsylvania Healthcare Cost Containment Counsel code selection; PSI, Patient Safety Indicator; QUADAS, Quality assessment for diagnostic accuracy studies; SSI, surgical site infection; UTI, urinary tract infection; (VA)SQIP, (Veteran's Administration) Surgical Quality Improvement Project.