Do epidemiological results replicate? The prevalence and health-economic consequences of neck and back pain in the general population.

Current estimates of the prevalence and consequences of neck and back pain vary greatly between studies. It is not known whether this variance is due to differences in methodology, or if it depends on the dynamics of the problem over time. The aim of this study was consequently an attempt to replicate and extend the findings of a previous epidemiological study using the same methodology on a new population. A survey of 3000 35-45 year olds, selected at randon, was conducted to determine the prevalence, site, frequency and intensity of the pain as well as any work loss or health-care utilization. The response rate was 69% and an analysis of non-responders showed that they were very similar to responders, but had a slightly lower prevalence. The results replicated the original study: 73% reported back pain during the past year and the consequences included considerable suffering and functional impairment. Moreover, 17% of those reporting pain had utilized sick leave during the past year for the problem, while an additional 14% had been off work but had not used sick leave. Sufferers averaged 3.5 health-care visits during the past year. However, the consumption of resources was highly skewed and about 6% of the sufferers accounted for over 50% of the costs. It was concluded that when the same selection criteria and assessment techniques are employed, the results found are quite similar. This implies that much of the huge variation in reported prevalence rates and consequences of back pain may be due to methodological differences. This underscores the need for standardized methods.