Skip to main content
. 2022 Dec 7;8(4):e42033. doi: 10.2196/42033

Table 4.

Effectiveness of the training programs according to the measures reported.

Training format and articles SMDa, EMDb, or RDc (95% CI) Narrative results Effectived
Interactive format (including reflexivity strategies)

Krones et al [32], 2008; Hirsch et al [64], 2010 0.40 (0.28 to 0.52) N/Ae Yes

Murray et al [33], 2010 3.75 (2.46 to 5.03) N/A Yes

Sanders et al [56], 2017 0.85 (0.54 to 1.16); 0.93 (0.62 to 1.25) N/A Yes

Elwyn et al [34], 2004; Edwards et al [67], 2004; Longo et al [68], 2006 −0.30 (−1.19 to 0.59); 0.05 (−0.17 to 0.27) N/A No

Fossli et al [41], 2011 0.38 (−0.17 to 0.94) N/A No

Kennedy et al [42], 2013; Kennedy et al [69], 2010 −0.05 (−0.12 to 0.01) N/A No

Epstein et al [55], 2017; Butow et al [62], 2015 0.00 (−0.24 to 0.24) N/A No
Interactive format (not including reflexivity strategies)

Bieber et al [35], 2006; Bieber et al [65], 2007 N/A “An ANOVA for repeated measurements comparing the SDM group with the information group revealed that patients’ appraisal of the interaction quality was higher in the SDM group” Yes

Stacey et al [36], 2008 2.07 (1.26 to 2.87) N/A Yes

Loh et al [37], 2007 N/A “In the intervention group, significantly higher patient participation from pre- to postintervention was found.” Yes

Haskard et al [38], 2008 N/A “Training significantly improved physicians’ health behaviour counseling of their patients.” Yes

Deinzer et al [39], 2009 N/A “The degree of SDM was significantly higher in the SDM group at baseline and after one-year visits. The results of the SDM sum score on actually practiced SDM exhibited in both groups significantly increased, but the control group did not reach the score of the study group after one year.” Yes

Feng et al [43], 2013 N/A “Significant difference in favour of the intervention group, high risk of bias.” Yes

Tinsel et al [44], 2013 0.32 (0.17 to 0.46) N/A Yes

Hamann et al [3], 2007 0.16 (−0.28 to 0.61) N/A No

Bernhard et al [45], 2011; Butow et al [62], 2015 N/A “There was no effect for this variable for SGAf doctors (estimated population mean difference 0.52, SE 1.39, ESg=0.04; P=.71)” “After the training workshop, doctors in the experimental group within the ANZh cohort displayed more behaviours designed to establish the SDM framework than doctors in the control group (estimated population mean difference=3.42, SE 1.50, ES=0.30, P=.03). However, the ES was small” “There was considerable variation in patient outcomes between the SGA and ANZ cohorts and no substantial training effect” No

Cooper et al [54], 2011 0.11 (−0.30 to 0.51); 0.03 (−0.15 to 0.20); 0.16 (−0.23 to 0.56) N/A No

Légaré et al [52], 2012; Allaire et al [70], 2012; Légaré et al [71], 2013 0.01 (−0.03 to 0.06) N/A No

Cooper et al [53], 2013 0.70 (0.30 to 1.90)5 N/A No

Wilkes et al [51], 2013 −0.13 (−0.32 to 0.05) N/A No

Härter et al [57]i, 2015; Bieber et al [66], 2018 0.54 (0.35 to 0.74); −0.07 (−0.26 to 0.12); 0.11 (−0.10 to 0.31) N/A No

Tai-Seale et al [58]j, 2016; Dillon et al [63], 2017 0.35 (−0.53 to 1.24); 0.51 (0.19 to 0.84); −0.29 (−1.17 to 0.60); 0.00 (−0.32 to 0.32) N/A No

Ampe et al [59], 2017 −0.10 (−0.96 to 0.76) N/A No

Cox et al [60], 2017 0.11 (−0.21 to 0.42) N/A No
Unidirectional

Hess et al [46], 2012 2.82 (2.43 to 3.21) N/A Yes

O’Cathain et al [40], 2002 −0.02 (−0.05 to 0.01) N/A No

Koerner et al [47], 2014 −0.08 (−0.26 to 0.11) N/A No

Mathers et al [48], 2012 −0.09 (−0.23 to 0.05) N/A No

Rise et al [49], 2012; Rise et al [72], 2016 0.13 (−0.32 to 0.58) N/A No

Sheridan et al [50], 2014 −0.17 (−0.35 to 0.00) N/A No

Coylewright et al [61], 2016 0.51 (−0.05 to 1.07) N/A No

aSMD: standardized mean difference.

bEMD: effect size mean difference.

cRD: risk difference.

dScored as “Yes” if the 95% CI reported in the Cochrane review did not include 0 for the SMD, RD, and MD values or when the 95% CI did not include 1 for the OR values. In some studies, with ≥2 scales, we referred to the conclusion of the authors.

eN/A: not applicable.

fSGA: Switzerland, Germany, and Australia.

gES: effect size.

hANZ: Australia and New Zealand.

iThis study found no effect of shared decision-making training on the primary outcomes, which were similar between both the groups. However, training did contribute to improved observer-rated shared decision-making skills in physicians and reduced anxiety and depression in patients, particularly in women with breast cancer.

jThe primary outcome measure was CollaboRATE, a patient-reported experience with care. While the odds ratios (ORs) from the ASK (Ask Share Know) clinic (OR 1.417) and the OpenComm plus ASK clinic (OR 1.134) were greater than 1, their 75% CIs included 1, which suggests no difference from the usual care clinic. Our findings suggest that something could be done to improve the patient experience. We view the results as promising evidence of the intervention’s efficacy and as meaningful signals of its likely effects on patient experience.