Regional Inequality in the Distribution of Dental Caries among Michigan Third Grade Children

1
Regional Inequality in the Distribution of Dental
Caries among Michigan Third Grade Children

• By Michael Paustian, M.S.
• Michigan Department of Community Health
• Division of Genomics, Perinatal Health, and
  Chronic Disease Epidemiology MCH Section
• Michigan Epidemiology Conference, April 12, 2007

2
Rationale

• Obtain regional estimates on the prevalence and
  severity of dental caries
• Compare/contrast inequality measures (Gini
  coefficient vs. DHII)
• Interpretation of inequality measures

3
Count Your Smiles, 2005-06 population

• Inclusion
• Public schools with at least 20 third grade
  students
• Exclusion
• Private schools and home-schooled children
• Special education and accelerated classrooms
• Replacement schools
• School closure
• School refusal or poor response rate

4
Study protocol

• Active consent from parents
• Survey of parents
• Access to care use of services, insurance
• Demographic data
• Visual examination by volunteer hygienists
• Caries experience, untreated decay
• Sealants

5
Study-defined regions
Upper Peninsula
Northern Lower Peninsula
Southern Lower Peninsula
Suburban Detroit and the City of Detroit
6
Defining caries experience

• At least 0.5mm loss of tooth structure
• Visible upon gross examination
• Conservative approach likely underestimates
  disease

7
Central measures of disease

• Prevalence
• Untreated decay
• Caries experience
• Mean
• dft primary teeth
• DFMT permanent teeth
• Significant Caries Index most diseased
  one-third of the population

8
Lorenz curve

• Plot of cumulative disease versus cumulative
  population
• What do these curves provide?
• Statement of cumulative disease burden
• Inequality measures Gini coefficient
• Concentration index

9
The Sinking Ship of inequality
10
Gini coefficient and DHII

• Equals the proportion of area under the line of
  equality that falls between the line of equality
  and Lorenz curve.
• Ranges between 0 and 1
• 0 represents perfect equality
• 1 represents perfect inequality

11
Calculating DHII

• Expected distribution based upon the probability
  of an event (caries)
• P(caries) teeth with caries/ teeth examined
• Number of teeth examined in this analysis were
  not collected but based upon eruption patterns
  for a child 8 to 9 years old.
• Binomial calculation to obtain expected
  proportion of population with 0, 1, 2, etc teeth
  with caries
• Lorenz curve generated for expected distribution
  DHII line of equality

12
Regional inequality in the total dental caries
burden among Michigan 3rd grade children, 2005-06
13
Dentition-specific regional inequality in the
dental caries burden
14
Regional inequality in the total untreated decay
burden among Michigan 3rd grade children, 2005-06
15
Regional inequality in the primary teeth caries
experience versus untreated decay
16
Regional inequality in the permanent teeth
caries experience versus untreated decay
17
What does inequality mean?

• Is a value near 1 bad?
• Disease is distributed unequally
• Targeted public health approach?
• Is a value near 0 good?
• Disease is distributed equally
• Population-based public health approach?

18
Conclusions

• There are regional differences in dental caries
  disease prevalence and severity.
• Inequality measures can supplement other disease
  measures in determining how to tailor disease
  prevention to a population.
• Measurement of health inequality should use an
  appropriate reference line of equality.

19
Limitations

• Different types of teeth have a different
  probability of developing caries
• Assumed of teeth examined, based upon eruption
  patterns
• Exposure duration may impact inequality measures

20
Public Health Implications

• Inequality measures can help validate cumulative
  disease statements as statements of inequality.
• Inequality measures can help in determining an
  appropriate intervention method.
• The DHII methodology could be applied to other
  event data where probability of that event is
  known or can be estimated.

21
Acknowledgments

• Shannon Zackery, MPH
• Sheila Semler, RDH, PhD
• Michael Manz, DDS, MPH