Background
Aboriginal Australian health
Chronic kidney disease (CKD); the problem
CKD; the measurement
CKD among Aboriginal Australians
Periodontal disease
Periodontal disease and Aboriginal Australians
CKD and periodontal disease
Can CKD progression be reduced by periodontal therapy?
Can CVD progression be reduced by periodontal therapy?
Aims and hypotheses
-
Aim 1: To describe the extent and severity of periodontal disease in a group of Aboriginal adults with CKD.
-
Hypothesis: The extent and severity of periodontal disease in an Aboriginal population with CKD will be high compared with national-level indicators.
-
Aim 2: To examine the effects of a comprehensive periodontal intervention among an Aboriginal population with CKD.
-
Hypothesis: Exposure to a comprehensive periodontal intervention will reduce progression of CKD, non-invasive measures of CVD or CKD/CVD-related death amongst those with CKD.
Methods
Study design
Ethical approval
Inclusion criteria
Randomisation
Periodontal intervention
Location
Engagement, recruitment and retention
Follow up
Sample size
Measures
Data collection techniques
CKD/CVD and cardio-metabolic-related assessments
Reimbursement for time
Data analysis
-
Aim 1: Extent and severity of periodontal disease compared with national-level counterparts: Extent of periodontal disease will be defined by the number of periodontal sites in the mouth with a given pocket depth (PD) divided by the number of teeth in the mouth x 100. The National Survey of Adult Oral Health dataset will enable comparison with both general and Aboriginal population-level estimates [49]. General analysis will comprise Chi-square and Student’s t-test within the study sample and non-overlapping 95 % confidence intervals when comparing with population estimates.
-
Aim 2: Changes in cIMT, measures of renal function and CKD/CVD-related death: The primary analysis will compare changes in cIMT and 95 % confidence intervals. Secondary analyses will examine each outcome variable using separate mixed models, to allow inclusion of all participants without having to impute data. We will use generalised linear mixed models for non-continuous outcomes. Models will include treatment group as a fixed effect. Analyses will be conducted to identify subgroups that modify the response to the intervention (eg. active/passive smoking, socio-economic status). Model assumptions will be checked and appropriate adjustments made. We will conduct sensitivity analyses to assess uncertainty in key parameters. Death and cardiovascular events, both binary outcomes, will be analysed using the chi-square test with frequencies and percentages per treatment arm and odds-ratios or other measures of treatment effect reported along with summary measures. The Greenhouse-Geisser correction for the F-test will be used to adjust the degrees of freedom for deviation from sphericity. Logarithmic transformation of the data will be performed where appropriate. A post hoc correlation analysis by the Spearman rank-correlation method will be performed to evaluate the relationship between the changes in renal function and cIMT from baseline to 6, 12 and 24 months following periodontal therapy. The same will also be done for measures of periodontal health and cIMT. Effect sizes will be calculated by dividing the mean of change scores by the pooled estimate of the standard deviation. Effect measures will also be presented through calculation of number-needed-to-treat and its associated 95 % confidence interval. The statistically significant level will be at α < 0.05.