This discrepancy may be due to our study being cross-sectional and perhaps including individuals with less severe disease; previous studies were longitudinal and likely included more homogenous populations of individuals with clinically verified chronic conditions