Table 2

Summary of findings of QIC studies conducted in an ambulatory care or general practice setting (20 studies)

Reference and study designTopicMain results
Barceló et al 31
Cluster RCT
Diabetes careQIC sites increased foot (p<0.01) and eye examinations (p<0.01) the proportion of patients reaching three or more treatment goals (p<0.01), HbA1c control (p<0.01), TC<200 mg/dL (p<0.01) but no difference in BP control.
Homer et al 36
Cluster RCT
Asthma careNo statistical significant differences found
Shaw et al 39
Cluster RCT
Colorectal cancer screeningNo statistical significant differences found
Asch et al 40
Chronic heart failureQIC sites showed greater improvement in 11 of 23 indicators, achieving 17% versus 1% improvement in composite process bundle (p<0.0001). No statistical significant differences for patient outcomes.
Benedetti et al 42
Diabetes careQIC sites showed greater improvement in 7 of 12 diabetes indicators (p<0.05).
Franx et al 23
Antidepressant prescribingAntidepressant prescription rates decreased in QIC sites by 23% versus no decrease (OR 0.44; 0.21–0.92) control sites.
Haggstrom et al 45
Implementation of chronic care modelOverall improvement in CCM model implementation (p=0.002)
Landon et al 49
HIV careNo statistical significant differences
Landon et al 48
Diabetes, asthma or hypertensionFor all three conditions, compared with external and internal control sites, QIC sites reported 4.5% and 4.9% improvements, respectively (p<0.001). By condition, asthma and diabetes care but not hypertension significantly improved. No statistical significant differences for patient outcomes.
Mangione-Smith et al 51
Schonlau et al 54
Asthma careQIC sites improved composite paediatric process score 13% versus 0% (p<0.0001) and two of five self-management indicators. QIC sites improved composite adult process score 10% versus 1% (p=0.003) and one of six self-management indicators.
2 of 9 paediatric outcomes statistically significant and 1 of 5 adult outcomes statistically significant
Margolis et al 52
Practice delivery systemsMean number of care delivery systems increased from 12.9 (SD 4.6) to 19.4 (SD 3.87) in QIC group with no change for controls. No statistical significant difference for composite quality indicator (four aspects of parental reported care). RR 1.25; 0.93–1.75
Peterson et al 61
Diabetes careQIC practices reduced mean HbA1c −3.7 mmol/mol versus −1.7 mmol/mol (p<0.001).
Powell et al 53
Follow-up positive colorectal cancer screening testQIC sites increased 60-day follow-up colonoscopy from 27% to 39% versus control site decrease from 45% to 29% (p<0.02). Mean days to colonoscopy decreased for QIC sites from 129 to 103 days and increased for control sites from 81 to 103 days (p=0.001). Differences in 1-year follow-up not statistically significant.
Schouten et al 29
Diabetes careQIC sites improved 4 of 19 process measures; dietitian visit (17.8% vs 9.9%, p<0.01); glucose monitoring advice (61.7% vs 55.8%, p<0.05); advice to examine feet (75.2% vs 69.4%, p<0.05); instruction on foot examination (66% vs 59.5%, p<0.05) and improved 2 of 9 patient outcomes; mean SBP 139.3 (SD 17.4) versus 141.8 (SD 16.5), p<0.05; mean HDL 14.4 (SD 0.4) versus 1.3 (SD 0.4).
Vernacchio et al 55
Asthma careAsthma exacerbations declined greater than controls but not statistically significant.
Youngleson et al 69
ITS with control sites
Mother-to-child HIV transmissionHIV-exposed infants testing positive declined from 7.6% to 5%. Antenatal AZT increased from 74% to 86%. Intrapartum AZT increased from 43% to 84% and HAART from 10% to 25%. Postnatal HIV testing increased from 79% to 95%.
Crandall et al 72
Inflammatory bowel diseaseCrohn’s disease (CD) and ulcerative colitis (UC), completion of assessment bundle (CD 55%–93%; UC 62%–89%, p<0.01). TPMT measure (CD 60%–80%; UC 50%–73%, p<0.01). Appropriate thiopurine dose (CD 48%–56% not statistically significant; UC 23%–64%, p<0.01). Remission (CD 55%–68%; UC 61%–72%). Steroid-free treatment (CD 86%–90%; UC, unchanged). No change in nutritional status or growth
Patel et al 80
Bloodstream infections (BSI)Mean BSI and access-related BSI rates reduced from 1.09 and 0.73 events/100 patient-months to 0.89 and 0.42 events/100 patient-months. Modelled BSI rates decreased 32% (p=0.01) and 54% (p=0.001) for access-related BSI.
Pierce-Bulger et al 81
Infant mortality50% reduction in infant mortality with days between infant deaths increasing from mean 100 days to over 300 days
Webster et al 89
HIV careHAART initiations increased from 179/month (SD 17.2) to 511/month (SD 44.9); a 185% increase.
  • AZT, azidothymidine; BP, blood pressure; CBA, controlled before-after study; CCM, chronic care model; HAART, highly active antiretroviral treatment; HDL, high-density lipoprotein; ITS, interrupted time series study; QIC, quality improvement collaborative; RCT, randomised controlled trial; RR, relative risk; SBP, systolic blood pressure; TC, total cholesterol; TPMT, thiopurine methyltransferase.