Summary of findings of QIC studies conducted in a hospital setting (39 studies)
| Reference and study design | Topic | Main results |
| Dirks et al
32 Cluster RCT | Acute thrombolysis for ischaemic stroke | Thrombolysis 13.1% intervention patients versus 12.2% control patients. Adjusted OR (adj OR) 1.25; 95% CI 0.93 to 1.68. Thrombolysis within 4 hours of stroke onset: 44.5% intervention patients versus 39.3% control patients (adj OR 1.58; 1.11–2.27) |
| Horbar et al
20 Cluster RCT | Surfactant for preterm infants | Surfactant given to 54.7% QIC infants vs 18.2% control infants (adj OR 5.38; 2.84–10.20); first dose surfactant treatment >2 hours after birth given to 9.4% QIC infants versus 24.9% control infants (adj OR 0.35; 0.24–0.53); median time to surfactant QIC infants 21 min (IQR 10–128) versus 78 min (IQR 29–410) (adj HR 1.57; 1.42–2.07). Patient outcomes no statistical significant differences. |
| Kritchevsky et al
37 Cluster RCT | Preoperative antibiotic prophylaxis | Preoperative antibiotic dose within appropriate time; 83.2% intervention versus 85.3% control. No statistical significant difference after adjustment. |
| Lee et al
33 Cluster RCT | Bronchopulmonary dysplasia and hospital-associated infection | Absolute reduction bronchopulmonary dysplasia 0.0006 (95% CI 0.0011 to 0.0001) versus infection group Absolute reduction infection 0.0020 (95% CI 0.0007 to 0.0004) versus pulmonary group |
| Newhouse et al
38 Cluster RCT | Heart failure care | No statistical significant improvements found for any of the four primary effect measures. |
| Power et al
34 Cluster RCT | Stroke acute care | Compliance with early hours bundle 10.9%; 1.3%–20.6% (OR 1.56; 1.06–2.31) Compliance with rehabilitation bundle 11.2%; 1.4%–21.5% (OR 1.61; 1.07–2.42) |
| Russell et al
35 Cluster RCT | Lung cancer services | Active treatment increased QIC group by 5.2% versus 1.2% in the control group, mean difference 4.1%, 0.1%–8.2% (p=0.055). Other processes had no statistical significant differences. |
| Benning et al
62 CBA | Patient safety (two phases: SPI 1 and SPI 2) | SPI 1 and SPI 2 reported no differences in clinical processes or adverse events. No statistical significant differences in mortality found SPI 1, but for SPI 2 hospital mortality fell from 10.3% to 6.1% but increased from 17.3% to 21.4% in control hospitals (p=0.043). |
| Brush et al
43 CBA | AMI and heart failure | AMI and HF measures increased by 61%–77% versus 51%–60% control hospitals but not statistically significant. |
| Campbell et al
57 CBA | Surgical mortality and morbidity | No significant differences in mortality. Significant reduction in unadjusted morbidity rate for all cases from 10.7% to 9.7% in intervention sites versus 12.4% in controls that did not change over time. Adj OR 0.898 versus 1.00 (p=0.004) |
| Carlhed et al
22 60 CBA | Acute myocardial infarction | Mean absolute increase in QIC group for angiography 6.2% versus 16.8% (p=0.027), ACEI 1.4% versus 12.6% (p=0.002), lipid lowering therapy 2.3% versus 7.2% (p=0.065), clopidogrel 26.3% versus 41.2% (p=0.01), heparin 5.3% versus 16.3 (p=0.01) |
| Horbar et al
46 Rogowski et al 21 CBA | Hospital-associated infections, oxygen use | Absolute decrease in all infections −5.5% versus −1.6% (p=0.058), coagulase negative infections −5.4% versus −0.8% (p=0.26) and oxygen supplementation −12.1% versus −0.1% (p=0.045) but not mortality 1.7% versus −2.1% (p=0.44) |
| Howard et al
47 CBA | Organ donation | Organ donation increased from 52% to 60% QIC group vs static at 51% in control hospitals (absolute change 8%; 2%–13%). |
| Lee et al
24 CBA | Breast milk feeding | Breast milk feeding rates for very low birthweight infants increased QIC sites from 54.6% to 61.7% (adj OR 1.31; 1.05–1.64) versus control sites 64.2%−65.7% (adj OR 1.03; 0.92–1.16). Improvement for QIC sites sustained (adj OR 1.44; 1.09–1.91). |
| Lee et al
50 CBA | Delivery room practice | Hypothermia rates declined in the QIC sites from 39% to 21% (adj OR 0.37; 0.31–43) versus control site 42%−34% (adj OR 0.67; 0.57–0.79). |
| Schouten et al
12 CBA | Stroke length of stay | Length of stay poststroke reduced 27% QIC hospitals from 18.3 to 13.3 days versus 5.7% reduction in control hospitals (19.2 to 18.1 days). |
| Wirtschafter et al
56 CBA | Nosocomial infections* | Nosocomial infection (very low birthweight infants) reduced from 15.6% to 13.5% versus 19.4% to 16.4% controls (adj OR 0.81; 0.68–0.96). |
| Battersby et al
65 ITS with control sites | Maternal breast milk | Exclusive and any maternal breast milk at discharge for preterm infants increased from 26% to 33% and 50% to 57%, respectively, at QIC sites. Exclusive breast milk increase per month 0.22%; 0.11−0.34 versus control sites 0.05%; 0.01−0.09 (p=0.007) |
| Power et al
67 ITS with control sites | Clostridium difficile infection | QIC infection rates dropped from 2.60 (2.11−3.17) cases/1000 bed days to 0.69 (0.50−0.60), a 73% (0.69, 0.50−0.91) reduction. Control wards dropped from 1.15 (1.03−1.29) cases/1000 bed days to 0.51 (0.44−0.60), 56% (0.51, 0.44−0.60) reduction. |
| Shafer et al
68 ITS with control sites | Organ donation | Organ donation for QIC sites increased by 14.1% versus 8.3% control sites in the first year and over the 3-year period increased 22.5% from baseline compared with a 5.5% increase for the same period of time preceding the collaborative. |
| Bonello et al
70 ITS | CLABSI and VAP | VAP bundle compliance improved from 50% to 82%. CLABSI bundle compliance increased from 58% to 74%. |
| Broughton et al
19 ITS | Third stage of labour | Active management compliance went from 0% to over 95%. Vaginal deliveries with postpartum haemorrhage reduced from 83% to 5%. |
| Bundy et al
71 ITS | CLABSI | Mean baseline CLABSI rate of 2.85 per 1000 line days reducing to 2.04 per 1000 line days, a 28% reduction (RR 0.71; 0.55–0.92) |
| DePalo et al
73 ITS | CLABSI and VAP | CLABSI rates reduced 74% from 3.73/1000 days to 0.97. VAP rate reduced 15% from 3.44/1000 to 2.92/1000 ventilator days. |
| Donovan et al
74 ITS | Scheduled births | Proportion inappropriate scheduled births between 36 and 39 weeks’ gestation declined from 25% to less than 5% (p<0.05). |
| Glasgow et al
26 ITS | Discharges before noon, length of stay | Length of stay improved for 35% hospitals; 60% of these sustained improvements 2 years post-QIC. Discharges before noon improved for 46% hospitals; 32% of these sustained improvement. |
| Hayes et al
75 ITS | Paediatric ‘codes’ | Code rate decreased by 3% (not statistically significant). |
| Jeffries et al
28 ITS | CLABSI | Median CLABSI rate dropped from 6.3/1000 line days (IQR 5.0–8.9) to 4.3 (IQR 2.6–7.6). |
| Kaplan et al
76 ITS | Bloodstream infection | Mean prevalence of late-onset bloodstream infections in preterm infants reduced from 18.2% at baseline to 14.3%. |
| Koll et al
77 ITS | CLABSI | Mean CLABSI rate reduced from 4.85 infections/1000 line days to 2.24; a 54% reduction. |
| Miller et al
27 ITS | CLABSI | Mean CLABSI rate reduced from 5.4/1000 line days to 3.1 infections/1000 line days; a 43% reduction (p<0.0001). Two years postintervention, mean CLABSI rate was 2.3/1000 line days (RR 0.44; 0.37–0.53). |
| O’Connor et al
79 ITS | CABG surgery | Mortality rate postcoronary artery bypass graft surgery reduced by 24% (p=0.001). |
| Quigley et al
82 ITS | Falls | Mean fall-related injury rates remained the same, and mean % falls with serious injury increased. |
| Ralston et al
83 ITS | Acute bronchiolitis care | Absolute decrease in bronchodilator dose/patient by 3.4 (95% CI 1.4 to 5.8). Exposure to any dose of bronchodilator decreased by 12% (95% CI 5% to 25%). Reduction in chest physiotherapy by 10% (95% CI 3% to 18%), but no decrease for steroids, chest radiography or viral testing. |
| Rosen et al
84 ITS | Rapid response | The rate of ICU and non-ICU arrests/1000 discharges decreased but not statistically significant. |
| Toltzis et al
85 ITS | Surgical site infections | Surgical site infections reduced from 4.48/100 procedures to 1.89/100 procedures, a 58% reduction. |
| Weeks et al
86 ITS | Central line days | Mean and median central line days reduced from 516 (SD 403) to 481 (SD 420) and from 441 (IQR 225–688) to 400 (IQR 187–652). |
| Wheeler et al
87 ITS | CLABSI | CLABSI/1000 line days reduced from 3.0/1000 line days to <1.0/1000 line days. |
| Wirtschafter et al
88 ITS | CLABSI | CLABSI/1000 line days reduced from 4.32/1000 line days to 3.22 during follow-up; a 25% reduction. |
*Nosocomial infections are also known as hospital-associated infections.
ACEI, ACE inhibitor; AMI, acute myocardial infarction; CABG, coronary artery bypass grafting; CBA, controlled before-after study; CLABSI, central line associated bloodstream infection; HF, heart failure; ICU, intensive care unit; ITS, interrupted time series study; QIC, quality improvement collaborative; RCT, randomised controlled trial; RR, relative risk; SPI, safer patients initiative; VAP, ventilator-associated pneumonia.