Cost-effectiveness of periodontal management in diabetes patients

Background + Aims

• Periodontitis and type 2 diabetes are linked, but whether periodontal management is associated with lower healthcare use and expenditure remains uncertain.
• This study examined periodontal-care records over two years in relation to healthcare costs, hospital admissions and insulin initiation in the following year.

Materials + Methods

• This observational study used insurance claims from one Japanese prefecture. It included 16,583 adults aged at least 35 who received medication for type 2 diabetes in fiscal year 2015 and had at least three years of records.
• Participants were grouped by periodontal care in both initial years, either initial year, other dental treatment only, or no dental treatment. Claims-defined periodontal management included examination, scaling, root planing, curettage, surgery and supportive care.
• The primary outcome was total medical, dental and pharmacy expenditure in year three. Secondary outcomes were all-cause hospitalisation, admissions for ischaemic stroke or acute myocardial infarction, and insulin initiation.
• The insulin analysis included 13,222 people who had not received insulin in the first two years. Initiation was used as a proxy for diabetes progression, not a direct measure of glucose control.
• Models adjusted for age, sex and recorded comorbidities; expenditure models also accounted for previous costs. Sensitivity analyses examined care frequency and intensity.

Results

• 4,127 participants (24.9%) received periodontal management in both years, while 9,165 (55.3%) received no periodontal management. The latter group included people receiving other dental care.
• Of those receiving care in both years, 1,830 (44.3% of that group) had care in each six-month period.
• Compared with no dental treatment, care in both years was associated with approximately 4% lower year-three healthcare expenditure: adjusted multiplier 0.96 (95% CI 0.92–1.00, rounded).
• Adjusted odds ratios for this group were 0.90 for all-cause hospitalisation, 0.60 (95% CI 0.44–0.81) for ischaemic-stroke admission, and 0.77 (95% CI 0.64–0.92) for insulin initiation. These represent associations with lower odds, not proven treatment effects.
• There was no significant association with myocardial-infarction admission: aOR 1.06 (95% CI 0.54–2.06).

Limitations

• Claims lacked HbA1c and clinical periodontal severity, preventing direct assessment of glucose improvement or effects by actual disease severity.
• Residual confounding by socioeconomic circumstances, health behaviours and healthcare engagement remains possible.
• The insurance schemes and single-prefecture setting limit generalisability. Medication-based diabetes identification could include some people with type 1 diabetes.
• Insulin initiation depends on clinical decisions and is not inherently an adverse outcome; the study cannot show that periodontal care safely avoids or delays necessary insulin.

Conclusion

• Regularly recorded periodontal care was associated with modestly lower subsequent expenditure and some healthcare-use outcomes. The findings support further investigation of integrated care, but do not establish causal savings, stroke prevention or reduced need for diabetes medication.