Table 2.
Logit Model of Access and Utilization Measures by Race on %MHMO in MSA, Race × MHMO Interactions with SES, Demographic, and Health Status Variables Used as Independent Controls
| Doctor Visit in Last Year | Usual Source of Care | ER Use in Last Year | ||||
|---|---|---|---|---|---|---|
| Coefficient | Standard Error | Coefficient | Standard Error | Coefficient | Standard Error | |
| Race | ||||||
| Black | −0.60* | 0.25 | −0.04 | 0.30 | 0.07 | 0.12 |
| Hispanic | −0.82* | 0.27 | 0.05 | 0.28 | −0.12 | 0.16 |
| %Medicaid HMO | ||||||
| MHMO | 0.42 | 0.67 | 0.43 | 0.85 | 0.43 | 0.41 |
| MHMO × black | 1.25* | 0.59 | 0.31 | 0.80 | 0.19 | 0.50 |
| MHMO × Hispanic | 0.70 | 0.77 | −0.26 | 0.85 | 0.76 | 0.45 |
| Poverty status | ||||||
| <50% FPL | 0.07 | 0.15 | −0.30 | 0.16 | 0.002 | 0.10 |
| <50% FPL × black | −0.29 | 0.42 | 0.18 | 0.41 | −0.27 | 0.27 |
| <50% FPL × Hispanic | −0.47 | 0.39 | 0.06 | 0.47 | −0.50* | 0.22 |
| <50% FPL × MHMO | −1.10* | 0.55 | −0.12 | 0.73 | 0.30 | 0.38 |
| Gender | ||||||
| Female | 1.13* | 0.19 | 1.23* | 0.23 | 0.15 | 0.10 |
| Female × black | −1.12* | 0.37 | −0.15 | 0.45 | −0.46* | 0.24 |
| Female × Hispanic | −0.58 | 0.37 | 0.26 | 0.35 | −0.41 | 0.26 |
| Female × MHMO | 0.38 | 0.75 | −0.76 | 0.95 | −0.47 | 0.57 |
| Nativity | ||||||
| Foreign-born | 0.19 | 0.23 | −0.38* | 0.18 | −0.60* | 0.12 |
| Foreign × black | −0.37 | 0.58 | −0.93 | 0.68 | 0.02 | 0.34 |
| Foreign × Hispanic | −0.62 | 0.67 | −0.84 | 0.65 | 0.51 | 0.33 |
| Foreign × MHMO | 0.63 | 0.95 | −0.64 | 0.81 | −1.46* | 0.41 |
| Education | ||||||
| High school graduate | 0.38* | 0.19 | 0.02 | 0.19 | −0.10 | 0.13 |
| High school graduate × black | −0.28 | 0.54 | −0.73 | 0.42 | 0.15 | 0.37 |
| High school graduate × Hispanic | −0.37 | 0.49 | −0.75 | 0.65 | −0.25 | 0.31 |
| High school graduate × MHMO | −0.88 | 0.92 | −0.43 | 0.59 | −0.07 | 0.54 |
| Some college | 0.51* | 0.19 | 0.03 | 0.15 | 0.24 | 0.14 |
| Some college × black | −0.50 | 0.61 | −0.81 | 0.57 | 0.34 | 0.30 |
| Some college × Hispanic | −0.67 | 0.47 | −0.37 | 0.45 | −0.06 | 0.37 |
| Some college × MHMO | 0.81 | 0.75 | 0.22 | 0.65 | −0.02 | 0.60 |
| Age (referent 35–44) | ||||||
| 18–24 | 0.48* | 0.17 | −0.43 | 0.25 | 0.37* | 0.17 |
| 25–34 | 0.14 | 0.15 | 0.15 | 0.21 | 0.19 | 0.11 |
| 45–54 | 0.07 | 0.22 | −0.24 | 0.25 | −0.09 | 0.19 |
| 55–64 | −0.02 | 0.29 | 0.37 | 0.71 | −0.37* | 0.17 |
| Marital status | ||||||
| Married | 0.15 | 0.16 | −0.14 | 0.16 | 0.13 | 0.16 |
| MSA level variables | ||||||
| Income ×$10,000 | −0.27* | 0.11 | 0.09 | 0.17 | 0.13 | 0.08 |
| HMO penetration | −0.46 | 1.03 | 0.69 | 1.33 | −0.58 | 0.75 |
| HMO competition | −0.12 | 1.17 | 1.66 | 1.16 | 0.13 | 0.59 |
| Health status | ||||||
| Health status≥fair | −0.16 | 0.23 | −0.40 | 0.25 | −0.46* | 0.13 |
| Limitation of activity | 0.65* | 0.17 | −0.02 | 0.29 | 0.18 | 0.14 |
| Health conditions | ||||||
| Hypertension | 0.58* | 0.20 | 0.12 | 0.20 | 0.23* | 0.11 |
| Congenital heart disease | 0.46 | 0.40 | 1.38 | 1.14 | 0.25 | 0.36 |
| Heart attack | −0.04 | 0.51 | 0.64 | 0.68 | 0.69* | 0.33 |
| Stroke | 0.69 | 0.82 | 0.04 | 0.72 | −0.06 | 0.42 |
| Asthma | 0.30 | 0.40 | 1.03* | 0.52 | 0.52* | 0.21 |
| Ulcer | 0.48 | 0.33 | 0.49 | 0.39 | 0.00 | 0.17 |
| Diabetes | 1.27* | 0.47 | 0.68 | 0.42 | 0.41* | 0.21 |
| Sinusitis | 0.58* | 0.26 | 0.38 | 0.39 | 0.45* | 0.14 |
| Bronchitis | 0.61 | 0.47 | 0.11 | 0.40 | 0.38* | 0.17 |
| Pregnant | 0.34 | 0.36 | 0.25 | 0.36 | 0.30 | 0.20 |
| Back pain in last 3 months | 0.30 | 0.17 | −0.16 | 0.19 | 0.24 | 0.14 |
| Migraine in last 3 months | 0.38* | 0.16 | 0.22 | 0.20 | 0.42* | 0.10 |
| Health behaviors | ||||||
| Smoker | −0.02 | 0.18 | −0.24 | 0.20 | 0.36* | 0.12 |
| Exercise | 0.13 | 0.15 | 0.21 | 0.19 | −0.08 | 0.10 |
| Drink greater than five drinks | 0.15 | 0.21 | 0.10 | 0.21 | 0.19 | 0.14 |
| Overweight | 0.11 | 0.10 | 0.08 | 0.14 | −0.03 | 0.12 |
Logit coefficients and standard errors take into account sampling weights and stratification used to make NHIS sample representative of U.S. population.
Race, gender, and age coefficients are centered on the means so that regression coefficients on a given characteristic and their significance can be directly interpreted as the difference by race from the overall mean of the characteristic.
Managed care variable used is the percentage of Medicaid beneficiaries (23 state and 4-year dummy indicators were in model but are not shown).
FPL, federal poverty level; MHMO, Medicaid Health Maintenance Organization; HMO, health maintenance organization; SES, socioeconomic status; NHIS, National Health Interview Survey.
p <.05.