Websites were retrospectively analysed to study the effects that technological advances in web design have had on accessibility for persons with disabilities. A random sample of general websites and a convenience sample of US government websites were studied and compared for the years 1997-2002. Web accessibility barrier (WAB) and complexity scores were calculated. Analysis of Variance (ANOVA) and Tukey's HSD were used to determine differences among years for general sites. Repeated measures of ANOVA were used to analyse trends in US government sites, and Pearson's correlation coefficient (r) was computed to evaluate the relationship between accessibility and complexity. Random websites become progressively inaccessible as complexity increases. US government websites remain accessible while increasing in complexity. Increasing complexity, often caused by adding complex components to a Webpage, doesn't have to contribute to increasing barriers to accessibility: US government websites remain accessible despite increasing complexity by limiting the number of scripts used in Webpage design. (Contains 1 table and 6 figures.)