Asking older people about fear of falling did not have a negative effect

To assess whether completing a questionnaire on risk of falling could affect outcome measures: fear of falling, reported falls, and health service contacts in older people (panel conditioning).

We used a postal questionnaire to assess the effect on falls risk of implementing falls injury prevention guidelines within a single locality in outer London, UK. We compared responses for the baseline and 6-month follow-up surveys with those for a fresh survey. The latter was sent to a new pool of subjects drawn from the same population, and was sent only once; timing coincided with the follow-up survey.

At baseline, we received 498 responses for 1,000 (50%) surveys sent; of these, 358 (72%) subsequently returned the follow-up survey. For the fresh survey, we received 1,261 out of 2,000 (61%) responses to the fresh survey. The odds ratio for the effect of panel conditioning on fear of falling was 0.92 (95% confidence interval CI = 0.64-1.33), within our predefined limit for equivalence. Odds ratios for the effect on reported falls and health service contacts were 0.87 (95% CI = 0.59-1.29) and 0.75 (95% CI = 0.55-1.02), respectively.

The proportions of subjects who feared falling in the follow-up survey and in the fresh survey were equivalent. Reduced reporting of falls and health service use in the follow-up survey suggest that the potential for panel effects cannot be ignored.