Table 4.

Details of included studies

Source/condition	Design	Population/denominator	Timescale	Case ascertainment method	Diagnostic method	Outcome	Potential bias/methodological limitations
Nath et al. (2001) [9] PSP	Population-based study of prevalence	3 substudies; –National study: population of 59,236,500 –North East study: population of 2,589,240 –Community study (35 GP catchments in Newcastle-upon-Tyne): population of 259,998; populations all based on 1998 census figures for the UK	Prevalence on January 1, 1999	–National study: case ascertainment not complete –Regional study: direct referral of cases from all neurologists in the region, correspondence reviews, database screening, inpatient hospital sources –Community study: screening of GP records for potential patients followed by review of records and structured interview and clinical examination	Possible or probable cases based on NINDS- SPSP criteria	–National study: prevalence of 1.0/100,000 –Regional study: prevalence of 3.1/100,000 –Community study: prevalence of 6.5/100,000	Smaller population denominator at each level of study led to more extensive capture methods, hence the greater prevalence in smaller denomination studies; national study did not use active case ascertainment methods and thought up to 81% of cases unidentified

PSP = Progressive supranuclear palsy; GP = general practitioner; NINDS-SPSP = National Institute of Neurological Disorders and Society for Progressive Supranuclear Palsy.