Table 2.
Study outcomes
| Study | Shared decision‐making measure used | Shared decision making Intervention vs Control | Outcomes: Intervention (I), Control (C) Blood pressure Hypertension knowledge Adherence | Other outcomes |
|---|---|---|---|---|
| Watkins | Not measured | Not measured |
Blood pressure: At 1 y SBP (mmHg) Adjusted mean (SE): I = 149.8 (2.6), C = 149.2 (2.6), P (for test of difference between two comparison groups) <0.001 DBP (mmHg): I = 95.3 (1.7), C = 94.9 (1.7) NS P (for test of difference between two comparison groups) <0.001 Hypertension knowledge Knowledge score at 1 y: Adjusted mean (SE): I = 25.95 (0.21), C = 25.08 (0.21) P (for test of difference between two comparison groups) <0.001 |
|
|
Montgomery (2003) Emmett (2005) (3‐y follow‐up) |
Decisional Conflict Scale (DCS) A 16‐item patient self‐report scale. The 16‐item scale comprises five subscales, assessing the quality of the decision, and the extent to which a patient reports being informed, clear about their values, supported and certain about their choice. |
Decisional conflict scale (DCS) score Primary follow‐up (mean of 14 days after intervention) Total DCS score Decision analysis vs no decision analysis: Adjusted difference in means, (95% CI), P = −9.4 (95% CI −13.0 to −5.8), P < 0.001 Decision analysis subscales: Adjusted difference in means (95% CI) −5.4 (95% CI −10.6 to −0.2) for uncertainty subscale −15.7 (95% CI −20.2 to −11.2) for uninformed subscale −13.1 (−18.0 to −8.1) for unclear values subscale −8.7 (−12.8 to −4.7) for the unsupported subscale −1.7 (−6.0 to 2.5) for the decision quality subscale |
Antihypertensive treatment: At 3 mo: Prescription of blood pressure‐lowering medication: Decision analysis vs no decision analysis OR (adjusted) 1.13 (95% CI 0.59, 2.19) P = 0.71 Blood pressure: At 3 y: (mean follow‐up 2.8 y, range 2.2‐3.4 y) Decision analysis vs no decision analysis: Mean SBP in mmHg (SD): 149 (14), 147 (15) Adjusted difference (95% CI): 0.94 (−3.2 to 5.1) P = 0.65 DBP(mmHg): 85 (8), 85 (10) Adjusted difference (95% CI): −0.76 (−3.1 to 1.6) P = 0.53 Hypertension knowledge: At primary follow‐up (mean 14 d after randomization): % of answers correct, mean (SD) 73 (15), 67 (16) Difference (95% CI) 5 (2‐9), P = 0.003 Change, from primary to 3‐mo follow‐up, in % of answers correct, mean (SD) −1.9 (10.6), −1.2 (16.6) Difference (95% CI) −1.2 (−5.1 to 2.7), P = 0.53 Adherence Proportion reporting taking all their medication 91% overall 90% decision analysis OR = 1.56, 95% CI 0.49‐4.96 |
Emotional state anxiety (range 20‐80): At primary follow‐up (mean 14 days after randomization), mean (SD) Decision analysis 34.8 (10.3), No decision analysis 36.8 (13.8) Adjusted difference (95% CI): Decision analysis ‐no decision analysis = −2.8 (−5.6 to 0.1), P = 0.055 Change from primary to 3‐mo follow‐up mean (SD) Decision analysis −0.3 (10.9), No decision analysis −2.2 (11.10 Adjusted difference (95% CI) Decision analysis ‐ no decision analysis = 2.4 (−1.2 to 6.0) P = 0.19 intention to start antihypertensive treatment, after decision analysis session (mean 14 days after randomization): yes vs unsure, Decision analysis vs no decision analysis RR (CI) 1.19 (0.59‐2.4) No vs unsure, RR (CI) Decision analysis vs no decision analysis 3.15 (0.91‐10.98) Proportion prescribed antihypertensives Decision analysis vs no decision analysis OR (adjusted) 0.93 (95% CI 0.46‐1.96) P = 0.85 Clinical: Decision analysis vs no decision analysis Mean 10‐y cardiovascular risk (SD) 22 (11), 23 (12) Adjusted difference (95% CI) (−0.02 to 0.05) P = 0.54 Mean consultations per year (SD) 4.5 (2.5), 4.7 (2.1) Adjusted difference (98% CI) 0 (−0.61 to 0.6) P = 0.90 Mean consultations in which change to medication was made (SD) 3.1 (3.4), 3.4 (2.0) Adjusted difference (95% CI) −0.2 (−1.0 to 0.6) P = 0.61 |
| Deinzer |
Autonomy Preference Index (API) A patient self‐report measure of preference for participation in decision making and for information Modified COMRADE scale A 20‐item patient report scale measuring satisfaction with communication and confidence in decision made |
Autonomy Preference Index (API) At 1 y “showed no difference between the SDM and control group at baseline (P = 0.60) and did not change after 1 y (P = 0.83)” (no figures reported) Modified COMRADE scale At 1 y “Both groups showed an increase in SDM” (no figures reported) |
Blood pressure: Blood pressure at 1 y Unadjusted mean change from baseline (SD) SBP in mmHg: −9.26 (10.2), −6 (11.8) P = 0.24 DBP (mmHg): −5.33 (9.5), −3.0 (8.3) P = 0.19 Hypertension knowledge “After 1 y both groups showed similar levels of knowledge “(no figures given) |
Health‐related quality of life No figures are given: “There were no differences between the 2 groups concerning health‐related quality of life measured with the 8 scales of SF‐36” |
| Cooper |
Physicians’ Participatory Decision‐Making Style (PDM) This patient report measure is an aggregate score of three items (each scored 0 to 5): (i) If there were a choice between treatments, how often would this doctor ask you to help make the decision? (ii) How often does this doctor give you some control over your treatment? 3) How often does this doctor ask you to take some of the responsibility for your treatment? Higher scores reflect more participatory visits Patients’ Perceived Involvement in Care Scale (PICS) Patient self‐report measure with 3 subscales (doctor facilitation of patient involvement, information exchange between patient and physicians and patient participation in medical decision making) each scored from 1 to 5, with higher scores reflecting more involvement in care |
Patient rating of clinician's participatory decision‐making style (PDM) At 12 mo Change from baseline (coefficient and 95% CI from mixed‐effects regression controlling for nesting within physician) Physician + patient intensive 6.2 (−0.5, 12.9) Physician minimal/patient intensive 3.2 (−4.8, 11.3) 13 Physician intensive/patient minimal 3.1 (−3.9, 10.2) Physician + patient minimal −5.2 (−13.0, 2.5) P values for the comparison of the change in PDM at 1 y between each intervention group and the reference group (physician and patient minimal): physician intensive/patient intensive group P = 0.03; physician minimal/patient intensive group P = 0.13; physician intensive/patient minimal P = 0.12 Patients’ Perceived Involvement in Care Scale (PICS) subscales At 12 mo Change from baseline (coefficient and 95% CI from mixed‐effects regression controlling for nesting within physician) Subscale: doctor facilitation Physician + patient intensive 0.22 (0.08, 0.56) Physician minimal/patient intensive 0.12 (−0.15, 0.39) Physician intensive/patient minimal 0.09 (−0.14, 0.33) Physician + patient minimal −0.17 (−0.43,0.09) Subscale: information exchange Physician + patient intensive 0.32 (0.08, 0.56) Physician minimal/patient intensive 0.16 (−0.14, 0.45) Physician intensive/patient minimal 0.13 (−0.13, 0.38) Physician + patient minimal −0.22 (−0.51, 0.07) Subscale: patient decision making Physician + patient intensive 0.21 (−0.03, 0.44) Physician minimal/patient intensive 0.07 (−0.23, 0.36) Physician intensive/patient minimal 0.16 (−0.10, 0.41) Physician + patient minimal −0.13 (−0.42, 0.16) P values for the comparison of the change in PICS subscales at one year between each intervention group and the reference group (physician and patient minimal): Subscale: doctor facilitation Physician intensive/patient intensive group P = 0.03 physician minimal/patient intensive group P = 0.11 physician intensive/patient minimal P = 0.14 Subscale: information exchange Physician + patient intensive P = 0.005 Physician minimal/patient intensive P = 0.08 Physician intensive/patient minimal P = 0.08 Subscale: patient decision making Physician + patient intensive P = 0.08 Physician minimal/patient intensive P = 0.35 Physician intensive/patient minimal P = 0.14 |
Blood pressure At 12 mo: Change from baseline (coefficient and 95% CI from mixed‐effects regression controlling for nesting within physician) SBP in mmHg: Physician + patient intensive −2.8 (−9.5, 2.8) P = 0.58 Physician minimal/patient intensive −6.5 (−14.2, 1.2) P = 0.24 Physician intensive/patient minimal −2.3 (−9.7, 4.0) P = 0.65 Physician + patient minimal −0.1 (−0.75, 7.4) P values for the comparison of the change in SBP at 1 y between each intervention group and the reference group (physician and patient minimal): Physician + patient intensive P = 0.58 Physician minimal/patient intensive P = 0.24 Physician intensive/patient minimal P = 0.65 DBP in mmHg: Physician + patient intensive 0.2 (−3.7, 4.1) Physician minimal/patient intensive −0.9 (−5.4, 3.6) Physician intensive/patient minimal −1.4 (−5.1, 2.3) Physician + patient minimal 0.2 (−4.1,4.6) P values for the comparison of the change in DBP at 1 y between each intervention group and the reference group (physician and patient minimal): Physician + patient intensive P = 1.0 Physician minimal/patient intensive P = 0.72 Physician intensive/patient minimal P = 0.57 Adherence (Morisky scale): At 12 mo Change from baseline (predicted probability and 95% CI from logistic mixed‐effects regression controlling for nesting within physician) Physician + patient intensive 0.75 (0.62, 0.84) Physician minimal/patient intensive 0.80 (0.65, 0.90) Physician intensive/patient minimal 0.66 (0.53, 0.77) Physician + patient minimal 0.77 (0.63, 0.87) P for comparison with reference group (physician and patient minimal) Physician + patient intensive P = 0.75 Physician minimal/patient intensive P = 0.76 Physician intensive/patient minimal P = 0.22 |
% with BP controlled at 12 mo conditional probability (95% CI) from mixed‐effects regression controlling for nesting within physician Physician + patient intensive 0.53 (0.38, 0.68) Physician minimal/patient intensive 0.61 (0.43, 0.77) Physician intensive/patient minimal 0.65 (0.50, 0.78) Physician and patient minimal 0.55 (0.37,0.71) P for comparison of the % with BP controlled to reference group (physician and patient minimal): Physician + patient intensive P = 0.92 Physician minimal/patient intensive P = 0.58 Physician intensive/patient minimal P = 0.35 Patient ‐ physician communication i. Change in verbal dominance ratio (first study visit with actual patient, compared with videotaped simulated visit), by physician intervention group: Change (coefficient and 95% CI from mixed‐effects regression controlling for nesting within physician) Physician intensive group: −1.67 (−2.06, −1.28) Physician minimal group: −1.94 (−2.36, −1.53) P for comparison of change between physician intensive and physician minimal groups P = 0.35 ii. Change in patient‐centredness ratio (first study visit with actual patient, compared with videotaped simulated visit), by physician intervention group: Change (coefficient and 95% CI from mixed‐effects regression controlling for nesting within physician) Physician intensive group: −0.52 (−0.71, −0.32) Physician minimal group: −0.82 (−1/02, −0.61) P for comparison of change between physician intensive and physician minimal groups P = 0.04 |
| Tinsel | The SDM‐Q‐936 is a 9‐item self‐report scale in which the patient reports the extent to which shared decision making occurred; raw scores are transformed into a scale from 1‐100 in which higher scores indicate that patients perceive more shared decision making to occur. |
Shared Decision Making Q‐9 (SDM‐Q‐9) Difference in average means change (from baseline to 18 mo) from mixed‐effects model adjusted for baseline values of outcomes: Intervention vs control 3.1182, 97.5% CI −2.3730; 8.6093, P = 0.2029* *not statistically significant at 2.5% level [Bonferroni correction applied due to multiple outcome measures] |
Blood pressure: Difference in average means change over 1 y from T1 to T3, from mixed‐effects model adjusted for baseline values of outcomes: SBP (mmHg): +1.75 mmHg (97.5% CI [−0.189; 3.69], P = 0.043*) *not statistically significant at 2.5% level [Bonferonni correction applied due to multiple outcome measures] DBP (mmHg): + 0.9377 (95% CI −0.0381; 1.9134 P = 0.0596) Hypertension knowledge: Difference in average means change from T0 to T3 (CI), from mixed‐effects model adjusted for baseline values of outcomes: 1.3267 (95% CI −4.3272; 6.9806), P = 0.65454 Adherence (MARS‐D) Difference in average means change from T1 to T3 (CI) from mixed‐effects model adjusted for baseline values of outcomes: 0.670 (95% CI −0.3748; 1.7166), P = 0.2084 |
Cardiovascular risk score Difference in average means change from T1 to T3 (CI) from mixed‐effects model adjusted for baseline values of outcomes: −0.4891 (95% CI −1.4307; 0.4526), P = 0.3084 |
| Denig | Not measured |
Antihypertensive treatment Proportion of patients eligible for intensification of blood pressure treatment (SBP >=140 mmHg) Who had blood pressure treatment intensified OR (intervention, control), P value intervention: intensification in 17 (16%) control: Intensification in 8 (17%) OR 0.93 (0.37‐2.34) P = 0.882 |
(S/D) BP, systolic/diastolic) blood pressure; API, Autonomy Preference Index; CI, confidence interval; DA, decision analysis; DCS, Decisional Conflict Scale; NS, nonsignificant; OR, odds ratio; PDM, Participatory Decision‐Making Score; PICS, Patients’ Perceived Involvement in Care Scale RCT, randomized controlled trial; RR, relative risk; SD, standard deviation; SDM‐Q‐9, 9‐item Shared Decision Making Questionnaire; SE, standard error.
Unadjusted results unless otherwise stated.