Short-term variability of self report of incontinence in older persons

Virtually all estimates of the prevalence and incidence of incontinence in the community rely on self-reported continence status. The goal of this study was to assess the reliability of this measure in older adults.

Telephone interviews administered approximately 2 weeks apart.

Community-based congregate living facility.

A convenience sample of approximately 100 residents was contacted by letter; 48 of 51 (94%) who indicated their willingness to participate were interviewed. They included eight men and 40 women > 70 years old (79% > 80 years old), virtually all of whom were independent in basic ADLs and 83% of whom reported their health as good or excellent.

Responses to a structured questionnaire.

The prevalence of urinary incontinence was 40% at baseline and 44% on re-interview; the prevalence of fecal incontinence was 17% on both occasions. All Spearman correlations for items related to urinary incontinence characteristics were between .80 and .86, except for a question related to stress incontinence (r = .62); correlations for fecal incontinence were .67-.69.

Prevalence estimates of incontinence are stable over a 2-week period. However, the variability of individual responses, while relatively low, was within the range previously reported for estimates of incidence and remission rates of incontinence in community-dwelling elderly. This, variability should be taken into consideration when interpreting previous studies and designing future ones.