TABLE 1.
Summary of main differences between explanatory and pragmatic trials
| Study Characteristic | Explanatory Trials | Pragmatic Trials |
|---|---|---|
| Objective | Test of efficacy Tests a specific component of treatment |
Test of effectiveness Tests a package of care rather than individual components contributing to that care |
| Inclusion/exclusion criteria | Many; narrowly defined | Few; broadly defined |
| Patient group | Selected and homogeneous | Diverse and heterogeneous Participants reflect the population for which the treatment is intended |
| Randomization | Usually randomization by participant | Often quasi-experimental designs with no randomization |
| Blinding | Often double-blind | Participants blinded to treatment allocation when possible Data collectors and analysts often blinded to treatment allocation |
| Intervention | Standardized Simple interventions; often a discrete single activity |
May be a discrete single activity, but often involves a complex intervention |
| Control | Often placebo-controlled | Standard care (clinically relevant interventions or no treatment) |
| Ancillary therapy | Rarely present | Often present Reflects clinical practice |
| Outcomes | Single-objective, often laboratory-based, outcomes | Wider spectrum; measures that are familiar to prescribing clinicians and relevant to everyday life, such as function and quality of life |
| Setting | Experimental setting | Routine care setting |
| Technical skill and experience of practitioners | Usually experts | Wide variation |
| Compliance | Maximized Measured to assure high level |
Not essential, often low Often measured as an outcome Includes non-compliers and dropouts |
| Sample size | Small Standard statistical determination of sample size |
Large |
| Co-morbidities | Often none | Often present |
| Informed consent | Lengthy | Brief |
| Follow-up | Usually short-term | Longer-term |
| Data collection | Extensive | Limited |
| Confounding | Controlled where possible | Not controlled |
| Internal validity (the extent to which the study design is likely to have precluded bias) | Higher | Lower |
| External validity (degree of generalizability of the results) | Lower | Higher |