Table 2.

 Effect sizes of interventions to encourage patients to ask questions in consultations

Outcome	No of studies	No of patients	Standardised mean difference (95% CI)
All studies
Question asking*	14	2020	0.27 (0.19 to 0.36)
Patients’ satisfaction*	17	3316	0.09 (0.03 to 0.16)
Anxiety before consultations	3	372	−1.56 (−7.10 to 3.97)†
Anxiety after consultations	6	809	−0.08 (−0.22 to 0.06)
Patients’ knowledge‡	5	378	−0.34 (−0.94 to 0.25)
Consultation length	13	3406	0.10 (−0.05 to 0.25)
Written materials v coaching
Question asking:
Written materials	6	563	0.42 (0.26 to 0.59)
Coaching	5	414	0.36 (0.16 to 0.56)
Patients’ satisfaction:
Written materials	10	2354	0.08 (0.00 to 0.16)
Coaching	6	722	0.23 (0.08 to 0.38)
Consultation length:
Written materials	10	2534	0.13 (0.05 to 0.21)
Coaching	3	872	0.07 (−0.07 to 0.20)
Clinicians’ training
Patients’ satisfaction:
Clinicians’ training	3	821	−0.01 (−0.15 to 0.12)
No clinicians’ training	15	2569	0.13 (0.05 to 0.21)
Consultation length:
Clinicians’ training	2	682	0.17 (0.01 to 0.32)
No clinicians’ training	12	2798	0.17 (0.10 to 0.24)

*Two assumptions were made about data from study by Roter^w29—that number analysed in intervention and control groups for outcomes of question asking and patients’ satisfaction were equal and that means for patient’s satisfaction in the two groups were 1.46 and 1.37 and not 146 and 1.37 as stated in text.

†Weighted mean difference.

‡In two studies^{w4 w18} intervention for control group could increase patients’ knowledge. Analysis was repeated with remaining three studies and a small and not statistically significant increase in knowledge was found (0.17, −0.09 to 0.43).