Skip to main content

Table 4 Individual trial estimates not combined in meta-analyses

From: The effects of shared decision-making compared to usual care for prostate cancer screening decisions: a systematic review and meta-analysis

First author & publication year Outcome Measurement point Intervention Control Effect estimate
    SDM patients (n) Total (N) Usual Care patients (n) Total (N) RR (95 % CI)
BINARY DATA
Patient-reported ordering of screening
Krist, 2007 [43, 44] (Woolf, 2005) patient-reported PSA tests ordered (patients' exit questionnaires) immediately after consultation 1) web-based DA 176 226 no pre-visit educational material and no DA during discussions with physicians 60 75 0.97 (0.85 to 1.11)
2) paper version of DA in 1) 151 196 60 75 0.96 (0.84 to 1.10)
Actual ordering of screening
Landrey, 2013 [42] PSA tests order by clinicians (chart-documented) following doctor's appointment flyer 85 136 no flyer 86 147 1.07 (0.88 to 1.29)
Krist, 2007 [43, 44] (Woolf, 2005) physician-reported PSA tests ordered (chart-documented) immediately after consultation 1) web-based DA 176 205 no pre-visit educational material and no DA during discussions with physicians 66 70 0.91 (0.84 to 0.99)
2) paper version of DA in 1) 155 182 66 70 0.90 (0.83 to 0.98)
Physicians' recommendations: towards screening
Wilkes, 2013 [41] doctor's recommendations towards PSA screening: unannounced standardised patients (physicians’ questionnaires) after clinic visitb 1) MD-Ed + A 16 36 CDC educational brochures on PC 34 43 0.56 (0.38 to 0.84)
2) MD-Ed 24 41 34 43 0.74 (0.55 to 1.00)
Physicians' recommendations: neither nor against screening
Wilkes, 2013 [41] doctors neither suggested nor recommended for or against PSA test: unannounced standardised patients (physicians’ questionnaires) after clinic visitb 1) MD-Ed + A 18 36 CDC educational brochures on PC 6 43 3.58 (1.59 to 8.06)
2) MD-Ed 14 41 6 43 2.45 (1.04 to 5.76)
Patient-estimates of lifetime risks
Gatellari, 2003 [45] how likely men were to give a correct estimate (within 2%) of the lifetime risk of dying from PC (correct answers over incorrect answers) unclear (questionnaires mailed 3 days post-consultations) 32-page (3085-word) evidence-based booklet 55 104 968-word pamphlet by the Australian government 3 75 13.22 (4.30 to 40.66)
how likely men were to give a correct estimate (within 10%) of the lifetime risk of developing PC (correct answers over incorrect answers) 59 104 18 108 3.40 (2.16 to 5.36)
    SDM mean (SD) Total (N) Usual Care mean (SD) Total (N) SMD (95 % CI)
CONTINUOUS DATA
Satisfaction with the visit
Wilkes, 2013 [41] patient-reported satisfaction with the visit: planned visits (sum of 5 satisfaction items: 5 = least satisfied, 20 = most satisfied) after clinic visitb MD-Ed + A 18 (3.00) 102 CDC educational brochures on PC 18 (3.00) 291 0.00 (-0.23 to 0.23)
patient-reported satisfaction with the visit: clinic visits by patients (sum of 5 satisfaction items: 5 = least satisfied, 20 = most satisfied) MD-Ed 18 (2.00) 188 18 (3.00) 291 0.00 (-0.18 to 0.18)
Men's views towards screening
Gatellari, 2003 [45] men’s views weighted towards or against reasons for having PSA testing (Scoring -5 to 5. Positive: weighting for; Higher: stronger weighting for; Negative: weighting against; Lower: stronger weighting against)b unclear (questionnaires mailed 3 days post-consultations) 32-page (3085-word) evidence-based booklet 1.70 (1.58) 106 968-word pamphlet by the Australian government 1.4 (1.59) 108 0.19 (-0.08 to 0.46)
Decisional conflict
Gatellari, 2003 [45] decisional conflict (9-item factors contributing to uncertainty scale; higher scores = greater decisional conflict) unclear (questionnaires mailed 3 days post-consultations) 32-page (3085-word) evidence-based booklet 21.60 (4.73) 106 968-word pamphlet by the Australian government 24.3 (4.77) 108 -0.57 (-0.84 to -0.29)
  1. PC Prostate Cancer, SDM Shared Decision-Making, MD-Ed + A Physician Education and patient Activation, MD-Ed Physician Education, DA Decision Aid, CDC Centers for Disease Control and Prevention, PSA Prostate Specific Antigen, n number of patients with events or number of events, N total number of patients per group, RR Relative Risk, SD Standard Deviation, SMD Standard Mean Difference, CI Confidence Intervals
  2. aQuestionnaire adapted from an attitudinal measure of the mammography screening instrument
  3. bMen followed-up in 6-16 weeks depending on the timing of the standardised visit: about 6 weeks after the intake survey for control physicians, 6-10 weeks for MD-Ed physicians, and 6-16 weeks for MD-Ed+A physicians