Executive Summary
What the data shows
Kalawao County has unusually high barrier levels and high diabetes prevalence, strongly influencing the apparent relationships. Patterns among the other four counties are mixed. These are exploratory associations—not evidence of causation.
Diabetes range12.0%–22.1%Across all five counties.
Primary associationNo Insurance (18–64)r = 0.97 with diabetes prevalence.
Handle with caren = 5 countiesResults are highly sensitive to Kalawao.
County Comparison
Kalawao Has the Highest Estimated Diabetes Prevalence
Healthcare Access and Diabetes
Kalawao Drives the Apparent Healthcare-Access Relationship
r = 0.97R² = 0.94n = 5
Exploratory only. The relationship changes substantially when Kalawao is removed, showing that one observation strongly influences the result.
| Analysis | Pearson r | R² | n |
|---|---|---|---|
| All 5 counties (primary) | 0.97 | 0.94 | 5 |
| Excluding Kalawao (sensitivity check) | −0.17 | 0.03 | 4 |
Factors Associated With Diabetes Prevalence
Strong Correlations Require Caution at n = 5
Strong values are largely influenced by Kalawao. Sensitivity checks weaken—and sometimes reverse—several correlations, which indicates instability rather than protective effects.
Conclusions and Recommended Next Steps
What this means for practice
- Use these findings to identify questions for more detailed assessment.
- Examine healthcare access and social barriers together in future research.
- Repeat the analysis using census-tract, ZIP-code, or individual-level data.
- Do not use this five-county analysis alone for resource-allocation decisions.
Methodology
About the Data and Methods
Source: CDC PLACES County Data, 2025 release. Values are modeled county-level crude prevalence estimates.
- Pearson correlations are unweighted; each county is one observation.
- The regression is exploratory, not predictive or causal.
- County-level estimates do not establish individual-level relationships.
- Results are highly sensitive to Kalawao County.