IADR Abstract Archives

Association between periodontal disease and HbA1c: A nonparametric assessment

Objectives: Our study performed cross-sectional medical/dental assessment on Gullah-speaking adolescent and adult type-2 diabetic African Americans living on the “Sea Islands” of South Carolina. Our aim was comparing periodontal health, as determined from probed pocket depth (PPD), clinical attachment loss (CAL) and bleeding on probing (BOP), among two groups, viz. those with good glycemic control (HbA1c < 7%, henceforth GGC) and those with poor control (HbA1c ³ 7%, henceforth PGC)(Source: ADA), controlling possible confounding due to smoking.

Methods: Clustered by subjects, the response measures are the periodontal outcomes averaged over 6 sites of a tooth to yield multiple tooth level data. For comparing measures of location for the two groups, a nonparametric Wilcoxon rank sum test (henceforth WRST) assuming underlying distributions to be unknown is infeasible here owing to the subject level clustering. Following Rosner et. al. (Biometrics, 2003), we incorporate clustering effects into the Wilcoxon test (henceforth CWRST) and also extend it to control for possible confounding effects by covariates. A large sample approach is used to test the null hypothesis of equality of location measures.

Results: For PPD and BOP, there was no significant difference between the two groups using either test, however, the WRST underestimated the variance by several fold. Ignoring clustering, WRST showed higher CAL value for abnormal glycemic group but CWRST (accounting for clustering) indicated no difference between the two groups. Considering all the three scores and accounting for confounding due to smoking habits which are coded as ‘ever/never' or ‘never/past/present' smoker, it is seen that the CWRST supports more severe periodontal disease for the PGC over the GGC (p value < 0.05).

Conclusions: Accounting for both the clustering and the confounding (smoking) effect, periodontal disease is more severe among the people with PGC in our population.

Support: NIH/NCRR P20 RR-017696


IADR/AADR/CADR General Session
2007 IADR/AADR/CADR General Session (New Orleans, Louisiana)
New Orleans, Louisiana
2007
74
Oral Health Research
  • Bandyopadhyay, Dipankar  ( Medical University of South Carolina, Charleston, SC, USA )
  • Wiegand, Ryan  ( Medical University of South Carolina, Charleston, SC, USA )
  • Slate, Elizabeth  ( Medical University of South Carolina, Charleston, SC, USA )
  • Fernandes, Jyotika  ( Medical University of South Carolina, Charleston, SC, USA )
  • London, Steven  ( Medical University of South Carolina, Charleston, SC, USA )
    • Oral Session
    Oral Health and Quality of Life Issues
    03/21/2007