Skip to main content

Table 3 Assumed values of sensitivity and specificity for each analytic sample using pre-randomization data considered in the power evaluation

From: Addressing identification bias in the design and analysis of cluster-randomized pragmatic trials: a case study

 True OUD prevalence (π)
 0.010.020.04
SampleaAssumptions / NotesSpecificitySensitivityb
1Documented OUD• Assumes all individuals with documented OUD do in fact have true OUD (specificity = 1)
• Sensitivity selected to be consistent with the observed proportion of patients with documented OUD based on Phase 1 data (0.5%) and the specific choice of the true prevalence (π)
10.50.250.125
2All patientsBy definition, sensitivity = 1 and specificity = 00111
3aHigh specificitySelected to have slightly higher sensitivity than Sample 1 (1.2 times the value), at the cost of slightly reduced specificity0.950.60.30.15
3bHigh sensitivity• Sensitivity was selected based on a previously developed algorithmc to identify individuals with opioid abuse and addiction, among patients on long-term opioid therapy
• We considered a lower specificity (0.5 vs 0.64c) given that our initial sample is the entire site population, not restricted to long-term opioid users
0.50.850.850.85
3cEqual sens./spec.Selected to have lower sensitivity and higher specificity than Sample 3b0.60.60.60.6
  1. OUD opioid use disorder
  2. a All options identify the study population using baseline (pre-randomization) data; see Table 2
  3. b For some of the options, sensitivity was allowed to vary across the assumed prevalence of true OUD (π)
  4. c Different cut-points for the developed risk score [39] achieved different values of sensitivity and specificity. We considered the “high sensitivity” scenario that achieved a sensitivity of 0.85 and a specificity of 0.64 in the validation sample (D. Carrell, personal communication, 5 July 2017)