Table 3.

Characteristics of the newly developed instruments (n=7)

Name of the tool	First author, year of publication (reference no)	Point of view	Language of validation	Inpatients and/or outpatients	Number of dimensions and items of SDM	Response-scale	Methodology	Sample	Reliability	Validity	Model and items generation process
SDM-Q-Doc	Scholl I, 2012 [34]	Professional	German	Inpatients and outpatients	1 dimension; 9 items	6-point scale	Real consultations	PHYSICIANS: N = 29/General practitioners 51.7 %, orthopaedists 13.8 %, psychiatrists 13.8 %, diabetologists 20.7 % PATIENTS: N = 324/external patients of primary and secondary care with a chronic back pain, type 2 diabetes, or depression	Cronbach’s alpha = 0.88 ICC = 0.35–0.76	Content validity: unknown Face validity: Yes Construct validity: Yes Convergent/Discriminant validity: Yes	Pre-existent tool (SDM-Q, 2006) and theory-driven: Nine practical steps of the SDM process defined by the authors: disclosure that a decision needs to be made, formulation of equality of partners, presentation of treatment options, informing on the benefits and risks of the options, investigation of patient’s understanding and expectations, identification of both parties’ preferences, negotiation, reaching a shared decision, arrangement of follow-up
Mappin’SDM	Kasper J, 2012 [35]	Patient, professional, observer	German	Inpatients and outpatients	15 items	5-point Likert scale	Real consultations	(video recording) 40 consultations physician-patient videorecorded (Hambourg)/average duration 19.5 min (2.5–51 min)/average duration of decision sequence 15 min (2.5 to 38.8 min)// PHYSICIANS: N = 10/neurologists and internal medicine 40 %, dentists 30 %, general practitioners 30 % PATIENTS: N = 40/55 % of men	Cronbach’s alpha = 0.91–0.94 ICC = Yes	Face validity: unknown Construct validity: No Convergent/Discriminant validity: Yes	Theory-driven (created by the authors): three perspectives, two constructs, three units and seven focus result in a set of three tools, each of them measuring the same fifteen items
Informed decision making instrument	Leader A, 2012 [36]	Observer	English	Inpatients and outpatients	3 dimensions 9 items Patient empowerment (1) Information sharing (4) Active engagement in preference clarification (4)	2 point-scale	Real consultations audio recorded N = 146	PHYSICIANS: N = 22 PATIENTS: N = 146 men candidates screening of prostate cancer	Cronbach’s alpha = 0.80 ICC = 0.81	Construct validity: use of an existing instrument	Theory-driven: Nine elements of Informed Decision Making developed by Dr Braddock [56]: the patient’s role in decision making, the impact of the decision on the patient’s daily life (context of decision), the nature of the decision or clinical issue, alternatives, pros and cons surrounding alternatives, uncertainties regarding alternatives, physician assessment of the patient’s understanding, physician assessment of the patient’s desire for input from trusted others, physician solicitation and exploration of the patient’s preference
SDM’Mass (SDM Meeting its concept’s ASSumptions)	Geiger F, 2012 [37]	Patient, professional, observer	German	Inpatients and outpatients	15 items	5-point Likert scales	Real consultations video recorded N = 40	Average duration 20 min (2.5–51 min; SD = 11)/Average duration of decision sequence 15 min (3 to 39 min; SD = 8)// PHYSICIANS: N = 10/neurologists and internal medicine 40 %, dentists 30 %, general practitioners 30 % PATIENTS: N = 40/55 % of men	Cronbach’s alpha = 0.94 ICC = 0.74–0.87	Face validity: unknown Construct validity: No Convergent/discriminant validity: No	Theory-driven (created by the authors): Three perspectives, two constructs, three units and seven focus result in a set of three tools, each of them measuring the same fifteen items
CICAA-Decision	Ruiz Moral R, 2010 [38]	Observer (on professional ‘s behaviour)	Spanish	Outpatient	3 dimensions; 17 items Identifying and understanding problems (2) Reach an agreement and help to act (11) Decisions with options (4)	3-point scale	Real and fictional consultations N = 111 real patients and N = 50 simulated patients	(Video recording) 161 consultations videorecorded: 61 consultations between “professional” and patient with chronic disease (diabetes et chronic pain) + 100 consultations between last year’s residents and new patients (50) or simulated patients (50)// Then selection of 32 consultations (20 % where item 25 is positive = a bit of participation is detected)/average duration = 11.3 min (SD = 5.6; IC95 = 9.2-13.3)	Cronbach’s alpha = 0.60–0.51 (1st and 2nd encounter) ICC global = 0.96	Content validity: Yes Face validity: Yes Construct validity: Yes Convergent/discriminant validity: No	Pre-existent tool (CICAA-CP) and literature research (review of pre-existing conceptual frameworks)
Dyadic measure of SDM	Légaré F, 2012 [33]	Patient, professional	English and Quebec French (patient and doctors recruited in Ontario and Québec)	Outpatient	7dimensions; 30 items Information giving (9) Values clarification (3) Doctor recommendations (5) Self-efficacy (3) Feeling uninformed (3) Information verifying (4) Uncertainty (3)	5-point scale and 10-point scale (different subscales)	Real consultations	PHYSICIANS: N = 272/english language N = 109, french N = 163 PATIENTS: N = 269/english language N = 108, french language N = 161/69 % of women/average age 49 (SD = 18) Complete DYAD: N = 259 (after consultation)	Cronbach’s alpha = 0.90 ICC = 0.43–0.82	Face validity: No Construct validity: Yes Convergent/discriminant validity: No Criterion validity: correlation with OPTION scale: Yes sensitivity analysis (AUC and ROC) Agreement across raters: ICC	Pre-existent tools and theory-driven (created by the authors): Based on Makoul and Clayman model, creation of a dyadic model that conceptualized the interpersonal and interdependent elements of the relationship between physicians and patients Then identification of instruments tested on both physicians and patients. Finally cross-cultural adaptation of the identified subscales that mapped the essential elements of SDM included in their dyadic model
Collaborate	Barr PJ, 2014 [39]	Observer (citizen; has to put in patient’s place)	English	Inappropriate	3 dimensions; 3 items Explanation of the health issue Elicitation of patient preferences Integration of patient preferences	2 versions: 5-point scale, and 10-point scale	Fictional consultations N = 6 simulated videos of encounters physician-patient	OBSERVERS: N = 1341 in study 1/N = 251 in study 2 (1–2 weeks after first answer)/On N = 1341: 46 % of men/47 % of 18-44 years, 33 % of 45-64 years, 20 % of 65 years and more/public in general population acting as the observer. Recruited by the 2010 US Census/representative sample of general population of USA	Cronbach’s alpha: No ICC = 0.76–0.90	Convergent/Discriminant validity: Yes Criterion validity: Yes Sensitivity to change: Yes	Theory-driven (created by the authors): the “talk model” developed by authors in a previous study [25]