Utility of self-reported oral health as a surveillance measure

Objective: Dental examination-based survey methods provide useful, detailed measures of observed oral health status, but are costly for larger surveys. Self-reported measures of oral disease provide less detail, but offer a less costly option for monitoring oral health in interview surveys. Several large health interview surveys ask a self-reported general health question. We examine the utility of self-reported poor oral health (SRPOH) as a measure of unmet or urgent oral health needs among adults (aged 20-44, 45-64, and 65+ years) using National Health and Nutrition Examination Survey (NHANES) 1999-2004 data for oral examination and recommendations for care.

Methods: We report sensitivity (sn) and specificity (sp) of SRPOH as a proxy measure for unmet or urgent treatment need for tooth decay, gum disease, oral hygiene, pain; number of missing teeth >= 90th percentile (MT) for specific age groups, and prevalence of each condition. Multivariate logistic regression models are used to determine which of these conditions contribute to SRPOH independently of self-reported general health, smoking, gender, age, race/ethnicity, and income. Analyses were limited to dentate adults Analyses were conducted using SAS v9.0 and SUDAAN v9.0 to account for the complex sample design.

Results: Sensitivity was highest and most stable across age groups for tooth decay (sn 64.17-73.12%, sp 67.64-74.85%). Urgent (OR=7.43, 95%CI=[4.35,12.69]) and unmet (OR=3.30, 95%CI=[2.94,3.70]) treatment need for caries were the strongest predictors of SRPOH, followed by poor general health (OR=2.94, 95%CI=[2.56,3.37]), urgent need for periodontal treatment (OR=2.57, 95%CI=[1.68,3.94]), missing teeth (OR=2.36, 95%CI=[1.99,2.80]), and unmet need for periodontal treatment (OR=2.02, 95%CI=[1.75,2.34]). Oral pain was not a significant predictor of SRPOH in this model (prevalence 1-2% across age groups).

Conclusions: Including a self-reported oral health question in health interview surveys may be a useful and relatively inexpensive measure for tracking unmet to urgent oral health needs at national, state and local levels.