Beyond Rurality: Individual SES and Chronic disease prevalence

This study of over 455,000 adults in the Upper Midwest reveals that individual socioeconomic status is a stronger predictor of chronic disease prevalence than rurality, as rural-urban disparities largely disappear after adjusting for socioeconomic factors.

The Gist

Imagine you are trying to figure out why some people in a town get sick more often than others. For a long time, doctors and researchers have looked at the map and said, "Ah, the people living in the countryside (rural areas) get sick more than the people living in the city (urban areas)." They thought the location itself was the main culprit—maybe the air is different, or the hospitals are too far away.

But this new study says, "Hold on a minute. We need to look closer at the people, not just the map."

Here is the story of what they found, explained simply:

The Big Mystery: Is it the Address or the Wallet?

The researchers looked at nearly half a million adults in the Midwest (covering 27 counties). They wanted to know: Is getting sick because you live in a rural area, or is it because of your socioeconomic status (SES)—basically, your money, education, and resources?

To measure "money and resources," they didn't just ask people, "How much do you make?" (which people often lie about or forget). Instead, they used a clever trick called the HOUSES Index.

The Analogy: The House as a Mirror
Think of a person's house as a mirror reflecting their life. The researchers looked at public records to see:

• How big is the house?
• How many bedrooms and bathrooms does it have?
• How much is the house worth?

If you have a big house with lots of rooms, it's like a mirror showing you have a lot of resources. If you have a small, modest home, the mirror shows fewer resources. This gave them a very accurate, objective score for each person's "wealth" without needing to ask them directly.

The Experiment: The Two Filters

The researchers ran the data through two different "filters" to see what mattered more.

Filter 1: The "Location" Filter
First, they just looked at Rural vs. Urban.

• The Result: It looked like rural people had more diabetes, high blood pressure, and heart disease. It seemed like living in the country was the problem.

Filter 2: The "Wealth" Filter
Then, they added the HOUSES Index (the house-size score) into the mix. They asked: "If we compare a rich person in the city to a poor person in the city, and a rich person in the country to a poor person in the country, who gets sick more?"

• The Big Reveal: When they accounted for the house/wealth score, the "Rural vs. Urban" difference disappeared.
• The rural people who got sick were mostly the ones with smaller, less valuable houses (lower wealth).
• The urban people who got sick were mostly the ones with smaller, less valuable houses (lower wealth).
• The Rich People: Whether they lived in a mansion in the city or a big farm in the country, they were generally healthier.

The "Dose-Response" Ladder

The study found a very clear pattern, like climbing a ladder.

• Bottom Rung (Lowest Wealth/Houses): Highest risk of getting sick.
• Middle Rungs: Medium risk.
• Top Rung (Highest Wealth/Houses): Lowest risk.

This pattern was true for every disease they studied (Asthma, Diabetes, High Blood Pressure, Heart Disease, and Mood Disorders). The "ladder of wealth" was a much stronger predictor of health than the "ladder of geography."

The Takeaway: It's Not Where You Live, It's What You Have

The authors use a great metaphor in their conclusion: Rurality is often just a disguise for poverty.

Think of it like this:

If you see a group of people struggling to walk up a hill, you might blame the hill (the rural location). But if you look closer, you realize they are all carrying heavy backpacks full of rocks (poverty/lack of resources). If you take the backpacks off, the hill doesn't matter as much anymore.

Why does this matter?
For a long time, health programs have focused on "fixing rural areas" by building more clinics or sending mobile doctors. While that helps, this study suggests we are missing the bigger picture.

The Real Solution:
Instead of just looking at the zip code, doctors and policymakers need to look at the individual's resources.

• If a patient has a small house and low income, they are at high risk for disease, whether they live in a city or a farm.
• We need to help people climb that "wealth ladder" (better housing, better income, better access to food) because that is what actually keeps them healthy.

In short: Don't just blame the map. Look at the wallet. The size of your house (and what it represents) tells us more about your health than the name of your town.

Technical Summary

1. Problem Statement

Rural-urban health disparities are well-documented, with rural populations often exhibiting higher burdens of chronic disease. However, the extent to which these disparities are driven by geographic location (rurality) versus individual-level socioeconomic status (SES) remains unclear.

• Limitations of Current Metrics: Traditional definitions of rurality (e.g., RUCA codes, Census criteria) are inconsistent and often mask significant socioeconomic variations within rural and urban areas.
• Limitations of SES Measures: Existing studies often rely on area-level SES proxies (e.g., Area Deprivation Index, census tract income) which assign a uniform score to all residents of a geographic block, failing to capture individual-level vulnerability.
• Research Gap: There is a lack of large-scale studies comparing the predictive power of individual-level SES against geographic rurality for chronic disease prevalence, specifically using a validated, housing-based individual SES metric.

2. Methodology

Study Design & Population

• Type: Retrospective, population-based, cross-sectional study.
• Setting: 27 counties in the Upper Midwest (19 in Minnesota, 8 in Wisconsin) served by the Expanded Rochester Epidemiology Project (E-REP).
• Participants: 455,802 adults (≥18 years) with available HOUSES index data and valid Rural-Urban Commuting Area (RUCA) codes.
• Timeframe: Five-year prevalence period (April 1, 2014 – April 1, 2019).

Key Variables

• Exposure 1 (Rurality): Defined using RUCA codes. Rural was defined as primary codes 4–10; Urban as codes 1–3.
• Exposure 2 (SES): Measured using the HOUSES index.
  • Mechanism: An individual-level metric derived from public property assessment records (assessed value, square footage, bedrooms, bathrooms).
  • Scoring: Components are standardized to z-scores and summed. Scores are categorized into quartiles (Q1 = lowest SES, Q4 = highest SES).
• Outcomes: Five high-burden chronic diseases identified via ICD-9/ICD-10 codes:
  1. Asthma
  2. Mood Disorders
  3. Hypertension
  4. Diabetes Mellitus
  5. Coronary Heart Disease (CHD)
• Covariates: Age, sex, race/ethnicity, and healthcare utilization (proxy: receipt of at least one General Medical Examination [GME]).

Statistical Analysis

• Models: Mixed-effects logistic regression with county as a random effect.
• Model 0: Unadjusted association between rurality and disease.
• Model 1: Adjusted for demographics and GME (key independent variable: RUCA).
• Model 2: Adjusted for demographics, GME, RUCA, and HOUSES index (key independent variables: RUCA and HOUSES).
• Interaction: Tested for interactions between rurality and SES quartiles.

3. Key Contributions

1. Validation of Individual-Level SES: The study demonstrates that the HOUSES index, a housing-based individual-level metric, is a superior predictor of chronic disease prevalence compared to geographic rurality.
2. Disentangling Geography and Economics: The research provides empirical evidence that observed rural-urban health disparities are largely confounded by socioeconomic factors. When individual SES is controlled for, the statistical significance of rurality disappears for most conditions.
3. Dose-Response Relationship: The study establishes a clear, graded dose-response relationship between lower SES (lower HOUSES quartiles) and higher prevalence of all five chronic diseases, independent of location.
4. Scalable Methodology: It highlights the utility of the HOUSES platform for integrating individual-level SES data into Electronic Health Records (EHR) for risk stratification, offering a solution to the limitations of area-level indices like the ADI.

4. Key Results

Demographics

• Sample: 455,802 adults (42.8% rural, 53.8% female, 87.4% non-Hispanic White).
• SES Distribution: Rural residents were more likely to be in the lowest SES quartile (Q1: 21.7% vs. 19.8% urban) and less likely to be in the highest (Q4: 26.6% vs. 32.1% urban).

Prevalence & Associations

• Unadjusted Analysis: Rural residents had significantly higher crude prevalence of Mood Disorders, Hypertension, Diabetes, and CHD. Asthma was slightly more prevalent in urban areas.
• Adjusted for Demographics (Model 1): After adjusting for age, sex, race, and healthcare utilization, rural residence remained associated with slightly lower odds of Mood Disorders and Diabetes, but associations for other diseases were attenuated.
• Adjusted for SES (Model 2):
  • Rurality: Once the HOUSES index was included, rural-urban differences were no longer statistically significant for any of the five diseases.
  • SES (HOUSES Index): A strong, independent, dose-response association was observed. Individuals in the lowest SES quartile (Q1) had significantly higher odds of all diseases compared to the highest quartile (Q4).
    • Example (Hypertension): Adjusted OR for Q1 vs. Q4 was 1.53 (95% CI: 1.50–1.56).
    • Example (Diabetes): Adjusted OR for Q1 vs. Q4 was 1.42 (95% CI: 1.37–1.46).
    • Example (Mood Disorders): Adjusted OR for Q1 vs. Q4 was 1.59 (95% CI: 1.56–1.61).

Interactions

• Significant interactions between rurality and SES were found for Asthma and Mood Disorders, suggesting the effect of SES on these conditions varies slightly by location, though the dominant driver remains SES.

5. Significance and Implications

• Paradigm Shift in Health Disparities: The findings suggest that "rurality" is often a proxy for socioeconomic disadvantage. Public health interventions targeting "rural" areas may be more effective if they specifically target low-SES individuals, regardless of whether they live in rural or urban settings.
• Clinical Application: Integrating the HOUSES index into EHR systems could enable precision population health. Clinicians could identify patients at high socioeconomic risk for chronic disease earlier, facilitating targeted screening, care coordination, and resource allocation.
• Policy Relevance: Policies aimed at reducing chronic disease burdens should prioritize addressing structural socioeconomic determinants (housing, wealth, income) rather than focusing solely on geographic access or rural-specific funding streams.
• Limitations: The study is cross-sectional (precluding causal inference), relies on EHR data (potential selection bias toward healthcare users), and the cohort is predominantly White, limiting generalizability to more diverse populations.

Conclusion: Individual-level socioeconomic status is a more potent determinant of chronic disease prevalence than geographic rurality. The HOUSES index offers a robust, scalable tool to operationalize this insight for improved risk stratification and health equity strategies.