Characteristics of included studies
| Topic, disease or condition | Reference | Study design (control condition) | Number of participants in collaborative/number of participants providing data for evaluation* | Data collection method (number of participants) | Effectiveness: short-term or long-term (sustained changes or spread of changes) | Study set out to test determinant (yes/no) |
|---|---|---|---|---|---|---|
| Emergency care | Bartlett et al9 | UBA | 18 emergency departments 18 (I) | Self-reported progress data (18 teams) | Short-term effectiveness | No |
| Asthma (children) | Homer et al10 | RCT | 22 primary care practices 22 (I) 21 (C) | Patient telephone interview (530 (I), 591 (C)) | Short-term effectiveness | No |
| Acute myocardial infarction | Carlhed et al11 | CBA | 19 hospitals 19 (I) 19 (C) | National quality registry (3786 patients (I), 2940 (C)) | Short-term effectiveness | Yes |
| Fall prevention in nursing facilities | Colon-Emeric et al12 | BA with reference data | 42 nursing facilities 36 (I) 353 (reference group) | Patient chart review (832 (I)) Minimum dataset (falls) Self-reported progress data (36 facilities) | Short-term effectiveness | No |
| Patient safety | Leape et al13 | Cross sectional, post intervention | 90 teams 90 (I) | Self-reported progress data Questionnaire (58 hospital CEOs, 90 team leaders) | Short-term effectiveness | Yes |
| Paediatric preventive services in primary care | Young et al14 | UBA | 14 paediatric practices | Patient chart review 2 year olds: initial sample 267 (B) final sample 265 (A) 4 year olds: initial sample 277 (B) final sample 252 (A) | Short-term effectiveness | No |
| Catheter-related bloodstream infections in the ICU | Amarasingham et al15 | UBA | 108 ICUs 19 (I) | Patient chart review Questionnaire (19 directors) | Short-term effectiveness | No |
| Diabetes care in community health centres | Chin et al16 | UBA followed by an RCT studying long-term effects | 34 community health centres 16 (standard intensity) 16 (high intensity) | Patient chart review 1174 (I)/1190 (C) (B) 1249 (I)/1168 (C) (A) 2212 (C and I) (follow up) | Short-term effectiveness | Yes |
| Preventive services for healthy newborns during birth hospitalisation | Mercier et al17 | UBA | 12 hospitals 12 (I) | Patient chart review 359 newborns (B) 360 newborns (A) | Short-term effectiveness | No |
| Paediatric preventive and developmental primary care | Lannon et al18 | UBA | Clinics (16) 15 (I) | Patient chart review 171 charts (B; 15 practices) 305 charts (A; 8 practices) Questionnaire (15 teams) | Short-term effectiveness | No |
| Stroke | Schouten et al19 | BA with reference data | 23 stroke services 23 (I) | Self-reported progress data, 4549 patients (I) LOHS registries (81 hospitals, reference data) Questionnaire (21 stroke services; 73 team members) | Short-term effectiveness | Yes, theory based |
| Four collaboratives: • Access and efficiency • Primary care • Complications in ICU • Adverse drug events • Surgical site infections | Nembhard20 | Cross sectional, post intervention | 78 teams 53 (I) | Questionnaire (217 team members) Self-reported progress data Interview (9 randomly selected team leaders) | Short-term effectiveness | Yes |
| Five VHA collaboratives: • Adverse drug events • Patient safety • Home-based dementia primary care • Compensation and pension examination • Falls | Mills and Weeks21† | UBA | 134 teams 134 (I) | Questionnaire (93% of the teams) Self-reported progress data (134 teams) | Short-term effectiveness | Yes, theory based |
| VHA Adverse drug events | Weeks et al22† | UBA with follow-up data | 27 teams | Self-reported progress data Questionnaire (teams) Telephone interview (27 teams) | Short-term effectiveness 6 months after the end of the collaborative: Sustained changes Spread | Yes, theory based |
| VHA Patient safety | Mills et al23† | UBA with follow-up diffusion data | 40 teams 22 (I) | Questionnaire Self-reported progress data (22 teams) Telephone Interview (chief nurses inside and outside participating centres) | Short-term effectiveness 12 months after the end of the collaborative: Spread | Yes, theory based |
| VHA Falls | Neily et al24† | Follow up data from a UBA | 37 teams | Telephone interview (34 primary contacts) | Short-term effectiveness 12 months after the end of the collaborative: Sustained changes Spread | Yes, theory based |
| HIV | Landon et al25‡ | CBA | 62 clinics 44 (I) 25 (C) | Patient chart review 6406 (I) 3580 (C) Questionnaire (clinics) | Short-term effectiveness | No |
| HIV | McInnes et al26‡ | CBA | 62 clinics 54 (I) 37 (C) | Questionnaire: Clinicians 104 (I) 90 (C) Medical directors 34 (I) 28 (C) Program administrators 45 (I) 35 (C) | Short-term effectiveness | No |
| Four collaboratives on chronic illness: • Asthma • Diabetes • Congestive heart failure • Depression | Shortell et al27§ | Cross sectional, post intervention | 88 teams 40 (I) | Questionnaire (40 teams) | Short-term effectiveness | Yes, theory based |
| Four collaboratives on chronic illness: • Asthma • Diabetes • Congestive heart failure • Depression | Marsteller et al28§ | Cross sectional, post intervention | 95 teams 73 (I) | Self-reported progress data Telephone interview (team contact person) | Short-term effectiveness | Yes, theory based |
| Diabetes | Daniel et al29§ | UBA | 47 teams 47 (I) | Patient registry (method unclear) | Short-term effectiveness | No |
| Chronic heart failure | §Baker et al30 | Controlled, cross sectional | 13 clinics 6 (I) 6 (C) | Patient telephone survey [367 (I), 414 (C)] | Short-term effectiveness | Yes |
| Depression | Meredith et al31§ | Cross sectional, post intervention | 23 primary care organisations 17 (I) | Self-reported progress data Telephone interview (team leader) | Short-term effectiveness 18 months after the end of the collaborative: Sustained changes Spread | Yes, theory based |
*Teams participating in a collaborative quality improvement project do not always participate in an evaluation of determinants of success of that collaborative—for this reason column 4 reports both figures.
†Papers by Weeks et al,22 Mills et al23 and Neily et al24 describe three different Veteran's Health Affairs collaboratives. The paper by Mills and Weeks21 includes these collaboratives and two others. While most of Mills et al's23 determinant and effect parameters overlap those of Mills and Weeks21 the paper also contains information on spread for safety that is not included in the Mills and Weeks paper.
‡Papers by Landon et al25 and McInnes et al27 report on the same HIV collaborative but use different effect parameters.
§Papers by Daniel et al,29 Baker et al30 and Meredith et al31 describe collaboratives undertaken as part of the Improving Chronic Illness Care Program. All evaluate different determinants and different effect parameters. The papers by Shortell et al27 and Marsteller et al28 include these collaboratives. They focus on different aspects of team functioning and interaction as potential determinants.
BA, before–after study; (C), control group; CBA, controlled before–after study; HIV, human immunodeficiency virus; (I), intervention group; ICU, intensive care unit; LOHS, length of hospital stay; RCT, randomised controlled trial; UBA, uncontrolled before–after study; VHA, Veterans Health Affairs.