Skip to main content

Table 3 Meta-analyses (n = 4)

From: An overview of systematic reviews on the collaboration between physicians and nurses and the impact on patient outcomes: what can we learn in primary care?

Aubin et al., 2012
Intervention Control Outcome Number of studies Number of patients Median effect sizea % (95% BCI)b Hetero-geneity Quality of evidence: GRADE
Interdisciplinary teams (targeting informational continuity) Usual care Functional status 11 3057 0 (−3.40, 2.70) NAV Very low
Physical status 16 3589 0 (−0.50, 0.50) NAV Very low
Psychological status 13 3228 −0.24 (−3.00, 0.02) NAV Very low
Social status 4 589 −0.01 (−10.70, 0.30) NAV Very low
Global quality of life 9 2472 2.0 (−0.03, 3.20) NAV Very low
Interdisciplinary teams (targeting management continuity) Usual care Functional status 11 2612 0 (−3.40, 2.00) NAV Very low
Physical status 18 3439 0 (−0.50, 0.03) NAV Very low
Psychological status 15 3687 −1.1 (−6.30, 0.00) NAV Very low
Social status 4 528 −0.7 (−7.00, 0.30) NAV Very low
Global quality of life 7 1717 2.0 (−1.90, 3.20) NAV Very low
Health Quality Ontario, 2014
Intervention Control Outcome Number of studies Number of patients Effect size (95% CI) Hetero-geneity Quality of evidence: GRADE
Home team-based model of care Medicare guidelines for home health care Home death (number of people) 1 310 Odds ratio 2.20 (1.30, 3.72) NA Low
Hospital admission (number of people admitted to hospital) 1 310 Odds ratio 0.39 (0.24, 0.62) NA Low
Home (indirect) team-based model of care Usual care by a management care organization Advanced care planning (number of people) 1 190 Odds ratio 1.30 (0.58, 2.90) NA Very low
Hospital team-based model of care Hospital care/primary care team only Advanced care planning 2 616 Odds ratio
2.77 (0.48, 16.11)
I-square
48%
Very low
Comprehensive team-based model of care Usual care Home death (number of people) 1 434 Odds ratio
1.89 (1.13, 3.16)
NA Moderate
Nursing home death (number of people) 1 434 Odds ratio 0.37 (0.20, 0.67) NA Moderate
Hospital admission 1 434 Odds ratio 0.90 (0.42, 1.89) NA Moderate
Comprehensive, early start, team-based model of care Routine oncologic care Hospital admission 1 151 Odds ratio 0.84 (0.34, 2.03) NA Low
Shaw et al., 2014
Intervention Control Outcome Number of studies Number of patients Effect size (95% CI) Hetero-geneity Quality of evidence- risk of bias
Nurse-managed protocols Usual care Systolic blood pressure (difference in mmHg) 12 Intervention:5244
Control:4980
Weighted mean difference − 3.68 (−6.31, −1.05) I-square 75.1% According to the approach recommended by the Agency for Healthcare Research and Quality:
4 articles: Low risk of bias/good quality
12 articles:
Moderate risk of bias/fair quality
2 articles: High risk of bias/poor quality
Usual care Diastolic blood pressure (difference in mmHg) 12 Intervention:5244
Control:4980
Weighted mean difference − 1.56 (−2.76, −0.36) I-square 75.1%
Usual care Total cholesterol (difference in mg/dL) 9 Intervention:1879
Control:1615
Weighted mean difference − 9.37 (−20.77, 2.02) I-square 90.8%
Usual care Low-density lipoprotein cholesterol (difference in mg/dL) 6 Intervention:564
Control:555
Weighted mean difference − 12.07 (−28.27, 4.13) I-square 89.1%
Usual care Hemoglobin A1c level 8 Intervention:1444
Control: 1189
Weighted mean difference − 0.40 (−0.70, −0.10)% I-square
69.8%
Snaterse et al., 2016
Intervention Control Outcome Number of studies Number of patients Effect size (95% CI) Hetero-geneity Quality of evidence– risk of bias
Nurse-coordinated care Usual care Blood pressure (difference in mmHg) 7 3514 Weighted mean difference −2.96 (−4.40, −1.53) I-square 37.1% p = 0.146 Cochrane Collaboration’s risk of bias tool: low/unclear risk of bias.
Usual care Low-density lipoprotein cholesterol (difference in mmol/L) 8 3441 Weighted mean difference −0.23 (−0.36, −0.10) I-square
74.3%
p = 0.000
Usual care Smoking cessation rates (Relative risk of quitting) 8 3265 Relative risk
1.25 (1.09, 1.43)%
I-square
0.0%
p = 0.459
  1. Table 3 presents the results of the meta-analyses of four of the included systematic reviews. The different ‘collaboration interventions’ are presented, followed by the control group, patient outcomes, number of studies, number of patients, effect size, a measure of heterogeneity (if available) and a measure of quality of evidence/risk of bias (if available). The improved patient outcomes are written in bold
  2. NA ‘not applicable’, NAV ‘not available’
  3. aTo handle the diverse set of outcomes within each individual study, the median value was computed of all the measured effects across all the outcomes of the same class. To pool the results from multiple studies, the median effect size was calculated for each class of outcome, by computing the median from all the median effects in outcomes obtained from individual studies. The researchers chose this pooling strategy to be consistent with the median approach used in other reviews [45,46,47]
  4. bnon-parametric bootstrap confidence intervals