Table 3.
Ordinal logistic regression estimates for level of agreement that health is linked to housing affordability (N = 400).a,b
| The cost of housing matters for people’s health | The affordability of people’s homes affects their health, not just their pocketbook | |||
|---|---|---|---|---|
| Independent Variable | Unadjusted OR [95% CI] | aOR [95% CI]c | Unadjusted OR [95% CI] | aOR [95% CI]d |
| Political Affiliation | ||||
| Independent (Ref) | 1.00 | 1.00 | 1.00 | 1.00 |
| Democrat | 2.06 [1.15, 3.71] | 2.09 [1.18, 3.70] | 1.59 [0.92, 2.77] | 1.52 [0.87, 2.66] |
| Republican | 0.67 [0.35, 1.27] | 0.71 [0.37, 1.35] | 0.75 [0.40, 1.42] | 0.72 [0.38, 1.36] |
| Annual Household Income | ||||
| 0 to 24,999 K (Ref) | 1.00 | 1.00 | 1.00 | 1.00 |
| 25Kto 49 K | 1.44 [0.72, 2.90] | 1.38 [0.67, 2.85] | 1.51 [0.73, 3.12] | 1.42 [0.67, 3.00] |
| 50 K to 99 K | 0.66 [0.33, 1.30] | 0.50 [0.22, 1.16] | 0.60 [0.35, 1.48] | 0.68 [0.30, 1.53] |
| 100 K to 149 K | 0.77 [0.39, 1.53] | 0.64 [0.28, 1.48] | 0.60 [0.29, 1.24] | 0.61 [0.25, 1.47] |
| > 150 K | 0.29 [0.11, 0.70] | 0.22 [0.07, 0.71] | 0.62 [0.28, 1.37] | 0.53 [0.20, 1.42] |
| Geopolitical Context | ||||
| Congressional Rep - Republican (Ref) | 1.00 | 1.00 | 1.00 | 1.00 |
| Congressional Rep - Democrat | 1.85 [1.15, 2.99] | 1.72 [1.03, 2.89] | 1.38 [0.87, 2.18] | 1.45 [0.87, 2.42] |
| Severe Housing Burden | ||||
| Bottom 75th Percentile - Rent (Ref) | 1.00 | 1.00 | 1.00 | 1.00 |
| Top 25th Percentile - Rent | 0.89 [0.52, 1.55] | 0.82 [0.46, 1.46] | 1.07 [0.59, 1.93] | 1.03 [0.57, 1.86] |
| Bottom 75th Percentile - Mortgage (Ref) | 1.00 | 1.00 | 1.00 | 1.00 |
| Top 25th Percentile - Mortgage | 0.86 [0.52, 1.43] | 0.84 [0.50, 1.40] | 0.76 [0.45, 1.30] | 0.71 [0.43, 1.19] |
aOR indicates adjusted odds ratio; CI, confidence interval
a
Estimates reflect how strongly respondents agreed or disagreed with the following statements:“The cost of housing matters for people's health” and “If we want to improve people's health, we need to make affordable housing a priority”
b
Estimates are weighted using population weights.
c
Models adjust for gender, age, race, marital status, educational attainment, homeownership status, and rural residence
d
Models adjust for gender, age, race, marital status, educational attainment, homeownership status, and rural residence