IADR Abstract Archives
Nutrition mediates the relationship between periodontal status and general health
Objectives: Our objective is to investigate the association of periodontal status and general health and to test whether intake of different food groups mediates this relationship.
Methods: Data were collected in 2004–06, using a computer-assisted telephone interview, followed by oral examination, mailed questionnaire and a food frequency questionnaire (FFQ) in New South Wales and Queensland.
Multivariate linear regression was conducted to assess relations between variables. Self-rated general health and periodontal status were used as outcome and explanatory variables, food groups (dairy, bread-cereal, meat-fish-eggs, sweet-snacks, mixed-vegetables, vegetables and fruits) were the mediators. Age, sex, smoking-status, brushing-habits, diabetes, alcohol-consumption and social-support were the control variables. Baron and Kenny’s mediation analysis was initially performed, followed by Sobel’s test for mediation. Lastly bootstrapping for standard error and Structural Equation Modelling (SEM) were conducted to assess consistency of the mediation model to the data.
Results: A total of 1,202 persons responded to the FFQ (98.7% response rate), with 62.6% aged 45+ years. Adults with none/mild and moderate periodontal problems compared to severe periodontal problems rated their general health better (β1=0.13 with p<0.001 and β2=0.09 with p<0.001). Baron and Kenny and Sobel-tests showed the associations were partially mediated by food intake (Sobel test: for all mediators dairy, bread & cereal, meat-fish-eggs, sweet-snacks, mixed-vegetable, vegetables and fruits, p<0.05). Multiple mediation bootstrap results showed bias corrected confidence intervals (-0.0091, 0, 0052) for the mediators: dairy, (-0.0012, 0.0347) bread-cereal, (-0.0017, 0.0303) fish-meat-eggs, (-0.0028, 0.0287) sweet-snacks, (-0.0036, 0.0126) mixed-vegetables, (-0.0064, 0.0132) vegetables, and (-0.00205, 0.0022) fruits, indicative of no mediation. SEM analysis for mediation showed p= 0.76, p=0.045, p=0.050, p=0.015, p=0.73, p=0.42 and p= 0.30 for dairy, bread-cereal, meat-fish-eggs, sweet-snacks, mixed-vegetables, vegetables and fruits.
Conclusions: Less severe periodontal problems predicted better general health. SEM indicated that this association was mediated by consumption of bread-cereal, meat-fish-eggs and sweet-snacks.
Methods: Data were collected in 2004–06, using a computer-assisted telephone interview, followed by oral examination, mailed questionnaire and a food frequency questionnaire (FFQ) in New South Wales and Queensland.
Multivariate linear regression was conducted to assess relations between variables. Self-rated general health and periodontal status were used as outcome and explanatory variables, food groups (dairy, bread-cereal, meat-fish-eggs, sweet-snacks, mixed-vegetables, vegetables and fruits) were the mediators. Age, sex, smoking-status, brushing-habits, diabetes, alcohol-consumption and social-support were the control variables. Baron and Kenny’s mediation analysis was initially performed, followed by Sobel’s test for mediation. Lastly bootstrapping for standard error and Structural Equation Modelling (SEM) were conducted to assess consistency of the mediation model to the data.
Results: A total of 1,202 persons responded to the FFQ (98.7% response rate), with 62.6% aged 45+ years. Adults with none/mild and moderate periodontal problems compared to severe periodontal problems rated their general health better (β1=0.13 with p<0.001 and β2=0.09 with p<0.001). Baron and Kenny and Sobel-tests showed the associations were partially mediated by food intake (Sobel test: for all mediators dairy, bread & cereal, meat-fish-eggs, sweet-snacks, mixed-vegetable, vegetables and fruits, p<0.05). Multiple mediation bootstrap results showed bias corrected confidence intervals (-0.0091, 0, 0052) for the mediators: dairy, (-0.0012, 0.0347) bread-cereal, (-0.0017, 0.0303) fish-meat-eggs, (-0.0028, 0.0287) sweet-snacks, (-0.0036, 0.0126) mixed-vegetables, (-0.0064, 0.0132) vegetables, and (-0.00205, 0.0022) fruits, indicative of no mediation. SEM analysis for mediation showed p= 0.76, p=0.045, p=0.050, p=0.015, p=0.73, p=0.42 and p= 0.30 for dairy, bread-cereal, meat-fish-eggs, sweet-snacks, mixed-vegetables, vegetables and fruits.
Conclusions: Less severe periodontal problems predicted better general health. SEM indicated that this association was mediated by consumption of bread-cereal, meat-fish-eggs and sweet-snacks.