Skip to main content

Table 4 Number of studies with improvements by types of co-interventions and study designs

From: Reporting feedback on healthcare outcomes to improve quality in care: a scoping review

 

Controlled designsa

Non-controlled designsa

Examples

Without co-intervention (N = 66)

18/27 (67%)

33/39 (85%)

 

With co-intervention(s) (N = 169)

47/68 (69%)

94/101 (93%)

 

Co-intervention Types

   

Facilitative Approaches:

 Education 74/112 (84%)

24/38 (63%)

61/64 (95%)

Topics:

• Strategies for improvement (N = 32)

• Reflection on findings (N = 12)

• Current guidelines (N = 27)

• Importance of improvement goals (N = 6)

• Data literacy (N = 3)

• Unclear (N = 8)

Mode:

• In-person (N = 53)

• Hybrid (N = 24)

• Distance learning (N = 18)

• Outreach visits (N = 7)

• Unclear (N = 10)

 Post-feedback consultations (N = 24/29)

9/12 (75%)

15/17 (88%)

• Counselling actions post-feedback (N = 17)

Clarifying details about feedback (N = 12)

 Reminders (N = 24/29)

10/13 (77%)

14/16 (88%)

• Via emails/ electronic alerts (N = 24):

• Via text messages (N = 1)

• Verbal (N = 4)

Coercive Approaches

 Decision Aids (N = 31/39)

15/21 (71%)

16/18 (89%)

• Clinician decision support systems (CDSS)/ action toolbox (N = 18)

• Pocket size guidelines/ cue cards/ posters (N = 17)

• Clinician care pathways (N = 4)

 Social influence (N = 5/8)

0/3 (0%)

5/5 (100%)

• Engaging patients in QI (N = 7)

Publicly posting outcomes (N = 1)

 Incentives (N = 9/13)

3/5 (60%)

6/8 (75%)

• Financial incentive (N = 9)

• Continuing medical education (CME) credits (N = 4)

• Light refreshments/ gift cards (N = 1)

  1. The section prior to "Co-intervention Types" separates studies with and without co-interventions and their reported outcomes. Below we describe which co-intervention were used in different studies and number of studies reporting improvements
  2. aControlled designs include randomized controlled trials, stepped wedge trials and other quasi experimental designs with a control group, non-controlled designs include quantitative studies with historic control only