CDC PLACES: Local Data for Better Health (ZCTA Level)
- Rows
- 24
- Columns
- 33
- Data range
- 70714 to 70820
Model-based ZIP Code Tabulation Area (ZCTA) level estimates for 27+ chronic disease measures, health risk behaviors, preventive services, and health outcomes across the US. Produced by the CDC Division of Population Health. Numerically rich with prevalence rates per zip code.
AI analysis
Dataset Analysis: CDC PLACES ZCTA-Level Health Data (Baton Rouge Area)
1. Dataset Overview & Analytical Value
This dataset represents model-based prevalence estimates for 27+ chronic disease, behavioral, and preventive health measures aggregated at the ZIP Code Tabulation Area (ZCTA) level, sourced from the CDC Division of Population Health via the Baton Rouge open data portal (data.brla.gov). The source URL's brla.gov domain confirms this is a locally scoped extract — almost certainly covering Baton Rouge, Louisiana ZIP codes (confirmed by the ZIP range 70714–70820) — rather than a national file, which is a critical scoping note for any analysis. With 24 ZCTAs and 33 columns, this is a numerically dense, cross-sectional snapshot ideal for intra-city spatial health correlation studies. Its value lies in connecting socioeconomic vulnerability indicators (like lack of health insurance) to disease burden and preventive care uptake across Baton Rouge neighborhoods, supporting public health targeting and equity research.
---
2. Data Quality Observations
Overall data quality is exceptionally clean for a public health administrative dataset. Across 24 rows and 33 columns (792 total cells), only 1 null cell exists, confined entirely to MEN CORE PREVENTATIVE SERVICES (4.2% missing — 1 of 24 ZCTAs), which is negligible and likely reflects a suppressed model estimate for a small-population ZCTA. No type mismatches are reported, all ZIP codes are non-null with 24 distinct values (confirming no duplicate geographies), and the join key column is fully populated. However, two columns warrant scrutiny for distributional data quality issues rather than missingness: DEPRESSION has only 2 distinct values (0.2 and 0.3) across all 24 ZCTAs — strongly suggesting a rounding artifact or model suppression behavior rather than true variation — and GENERAL HEALTH returns only 4 distinct values, similarly suspicious. These columns should be treated as ordinal categories, not continuous measures, for any regression or correlation work. Duplicate row detection is listed as pending Phase D recomputation, but the 24 distinct ZIP codes make duplicates structurally unlikely.
---
3. Key Column Distributions & Notable Patterns
Several columns reveal striking distributional stories worth highlighting:
- BLOOD PRESSURE MEDICINE (mean=0.709, σ=0.119, skew=-3.69, 2 outliers): The extreme negative skew and tight upper clustering (Q1=0.721, Q3=0.754) suggest most ZCTAs have high medication adherence, but 2 outliers pull sharply left — potentially low-income ZCTAs with access barriers. This is the most skewed continuous variable in the dataset. - ARTHRITIS (mean=0.233, median=0.246, skew=-2.23, 2 outliers): The negative skew with a high median indicates most ZCTAs have moderate-to-high arthritis prevalence, with 2 outliers representing substantially healthier (or younger-population) ZCTAs. - LACK OF HEALTH INSURANCE (mean=0.147, range=0.095–0.233, σ=0.038): Moderate variation across ZCTAs with a mild positive skew (0.51), suggesting a few ZCTAs face disproportionately high uninsured rates — a key equity signal. - DIABETES (mean=0.114, range=0.018–0.195, σ=0.042, 24 distinct values): One of the most granularly distributed variables with near-zero skew, making it an excellent continuous dependent variable for correlation modeling. - NO LEISURE-TIME PHYSICAL ACTIVITY (mean=0.307, range=0.217–0.463, σ=0.066, 24 distinct values): Wide spread across ZCTAs with 1 high-end outlier at 46.3% — a potential flag for a severely underserved neighborhood. - SENIORS WITH ALL TEETH LOST (mean=0.178, range=0.093–0.336, skew=0.815): High variance (σ=0.066) and right skew point to concentrated dental health deprivation in specific ZCTAs, often a proxy for long-term poverty.
---
4. Recommended Join Key Columns
| Column | Role | Notes | |---|---|---| | ZIP CODE | Primary join key | 24 distinct integers, no nulls, range 70714–70820; standard 5-digit ZCTA format | | GEOLOCATION | Spatial join key | 24 distinct strings, no nulls; enables GIS-based joins to census shapefiles or point-of-interest data | | 2010 POPULATION | Weighting key | Use as a population weight in weighted correlation analyses; range 1,335–43,194 (σ=11,994) suggests significant size variation requiring weighted rather than unweighted aggregation |
For cross-dataset joins, ZIP CODE is the most portable key, though analysts should be aware that ZCTAs and USPS ZIP codes are not perfectly congruent — a known CDC PLACES caveat. GEOLOCATION enables spatial joins to parcel, census tract, or infrastructure datasets when ZIP-level granularity is insufficient.
---
5. Recommended Pairing Datasets for Correlation Discovery
| Dataset | Source | Correlation Hypothesis | |---|---|---| | ACS 5-Year Estimates (ZIP/ZCTA level) | Census Bureau API | Correlate poverty rate, median household income, and educational attainment with diabetes, obesity, and insurance gaps | | USDA Food Access Research Atlas | USDA ERS | Test whether food desert designation predicts higher obesity, diabetes, and physical inactivity rates by ZCTA | | Baton Rouge Parish Tax Assessor / Property Data | data.brla.gov | Link housing value and density to chronic disease burden as a neighborhood socioeconomic proxy | | EPA EJScreen Environmental Justice Data | EPA API | Correlate air quality indices and toxic release proximity with asthma (mean=0.095) and COPD (mean=0.067) prevalence | | CMS Medicare Claims Data (ZIP-level) | data.cms.gov | Validate CDC model-based estimates against actual claims for hypertension, diabetes, and coronary heart disease | | Baton Rouge Crime/311 Service Data | data.brla.gov | Explore whether neighborhood stress indicators (crime density, code violations) predict poor mental health and depression rates | | Louisiana LAHIP / Medicaid Enrollment | LA DHH | Cross-reference uninsured rates against Medicaid enrollment gaps to identify coverage cliff ZCTAs |
The strongest immediate pairing recommendation is ACS income + poverty data joined on ZIP CODE, which would likely surface statistically significant correlations across at least a dozen of this dataset's health outcome columns — particularly LACK OF HEALTH INSURANCE, OBESITY, DIABETES, NO LEISURE-TIME PHYSICAL ACTIVITY, and DENTAL VISIT, all of which show sufficient continuous variation (24 distinct values each) to support meaningful Pearson or Spearman correlation analysis at this ZCTA level.
Columns
- ZIP CODE (integer)
- 2010 POPULATION (integer)
- LACK OF HEALTH INSURANCE (decimal)
- ARTHRITIS (decimal)
- BINGE DRINKING (decimal)
- HIGH BLOOD PRESSURE (decimal)
- BLOOD PRESSURE MEDICINE (decimal)
- CANCER (EXCLUDING SKIN CANCER) (decimal)
- ASTHMA (decimal)
- CERVICAL CANCER SCREENING (decimal)
- CORONARY HEART DISEASE (decimal)
- ROUTINE CHECKUP (decimal)
- CHOLESTEROL SCREENING (decimal)
- COLON SCREENING (decimal)
- COPD (decimal)
- MEN CORE PREVENTATIVE SERVICES (decimal)
- WOMEN CORE PREVENTATIVE SERVICES (decimal)
- CURRENT SMOKER (decimal)
- DENTAL VISIT (decimal)
- DEPRESSION (decimal)
- DIABETES (decimal)
- GENERAL HEALTH (decimal)
- HIGH CHOLESTEROL (decimal)
- CHRONIC KIDNEY DISEASE (decimal)
- NO LEISURE-TIME PHYSICAL ACTIVITY (decimal)
- MAMMOGRAPHY USE FOR OLDER WOMEN (decimal)
- POOR MENTAL HEALTH (decimal)
- OBESITY (decimal)
- POOR PHYSICAL HEALTH (decimal)
- LESS THAN 7 HOURS SLEEP (decimal)
- STROKE (decimal)
- SENIORS WITH ALL TEETH LOST (decimal)
- GEOLOCATION (string)