Article type
Abstract
BACKGROUND:
RCTs and other well-designed studies often apply strict inclusion criteria. In contrast, clinical guidelines typically apply to a broader patient population. If guidelines are primarily informed by studies with strict inclusion criteria, they may suffer from poor external validity or be inapplicable to certain patient groups.
OBJECTIVES:
This project formally evaluates the eligibility of real-world patients who have had a stroke for 8 key studies that informed the National Institute for Health and Care Excellence's stroke rehabilitation guideline.
METHODS:
Characteristics of patients in the Sentinel Stroke National Audit Programme (SSNAP) in England were compared with the inclusion criteria of 8 studies. The cohort of real-world patients potentially affected by the findings of each study was identified. Then, for each study considered, the percentage of potentially affected real-world patients who met all the inclusion criteria was calculated. The percentage excluded by each individual criterion was also calculated.
RESULTS:
Characteristics of 391,144 real-world patients who have had a stroke were evaluated. Seven of 8 studies assessed excluded patients with severe aphasia, which affected 21% of real-world patients who have had a stroke. Four studies excluded patients with advanced age or a previous stroke or transient ischemic attack (TIA), although in England, 42% of real-world patients were aged older than 80 years and 26% previously had a stroke or TIA.
CONCLUSIONS:
Patients with older age, previous stroke, and communication difficulties were frequently excluded from studies. This may limit the applicability of these to real-world patients and mean some patients' treatment is determined by evidence that is less applicable to their circumstances. Our analysis demonstrates the usefulness of real-world evidence to assess the external validity of studies presented to guideline committees and to help identify evidence gaps. The final published guideline on this topic recommends more research studies including patient groups who are typically excluded from stroke rehabilitation RCTs and other studies.
RCTs and other well-designed studies often apply strict inclusion criteria. In contrast, clinical guidelines typically apply to a broader patient population. If guidelines are primarily informed by studies with strict inclusion criteria, they may suffer from poor external validity or be inapplicable to certain patient groups.
OBJECTIVES:
This project formally evaluates the eligibility of real-world patients who have had a stroke for 8 key studies that informed the National Institute for Health and Care Excellence's stroke rehabilitation guideline.
METHODS:
Characteristics of patients in the Sentinel Stroke National Audit Programme (SSNAP) in England were compared with the inclusion criteria of 8 studies. The cohort of real-world patients potentially affected by the findings of each study was identified. Then, for each study considered, the percentage of potentially affected real-world patients who met all the inclusion criteria was calculated. The percentage excluded by each individual criterion was also calculated.
RESULTS:
Characteristics of 391,144 real-world patients who have had a stroke were evaluated. Seven of 8 studies assessed excluded patients with severe aphasia, which affected 21% of real-world patients who have had a stroke. Four studies excluded patients with advanced age or a previous stroke or transient ischemic attack (TIA), although in England, 42% of real-world patients were aged older than 80 years and 26% previously had a stroke or TIA.
CONCLUSIONS:
Patients with older age, previous stroke, and communication difficulties were frequently excluded from studies. This may limit the applicability of these to real-world patients and mean some patients' treatment is determined by evidence that is less applicable to their circumstances. Our analysis demonstrates the usefulness of real-world evidence to assess the external validity of studies presented to guideline committees and to help identify evidence gaps. The final published guideline on this topic recommends more research studies including patient groups who are typically excluded from stroke rehabilitation RCTs and other studies.