Table 2.
Logistic regression models examining adjusted odds of a High‐quality summary score (4 stars or higher)
| High‐quality summary score | High‐quality score: process measures a | High‐quality score: satisfaction measures b | |
|---|---|---|---|
| Odds Ratio (95% CI) | Odds Ratio (95% CI) | Odds Ratio (95% CI) | |
| MA HHI/100c | 1.02 (1.01‐1.03)** | 1.01 (0.99‐1.02) | 1.02 (1.00‐1.03)* |
| Hospital HHI/100d | 1.02 (1.01‐1.03)*** | 1.02 (1.01‐1.03)** | 1.02 (1.01‐1.02)** |
| Hospital beds per 1000 pop. | 1.24 (1.04‐1.47)* | 1.24 (1.05‐1.48)* | 1.22 (1.04‐1.43)* |
| MDs per 1000 pop. | 0.93 (0.77‐1.12) | 1.03 (0.85‐1.26) | 1.06 (0.87‐1.28) |
| MA benchmark payment rate relative to average FFS Medicare spending | 6.27 (1.43‐27.6)* | 2.61 (0.58‐11.81) | 1.40 (0.30‐6.48) |
| Log county‐level population over 65 | 1.30 (1.08‐1.56)** | 1.34 (1.11‐1.63)** | 0.80 (0.67‐0.96)* |
| County median income (in 1000s) | 1.01 (0.99‐1.02) | 1.02 (0.99‐1.04) | 0.99 (0.98‐1.01) |
| County‐level proportion of Medicare beneficiaries in fair or poor health | 0.99 (0.98‐1.00) | 0.98 (0.97‐0.99)* | 0.99 (0.97‐1.00) |
| PPO plan | 0.51 (0.37‐0.71)*** | 0.49 (0.35‐0.69)** | 0.78 (0.56‐1.08) |
| PFFS plan | 0.20 (0.11‐0.38)*** | 0.37 (0.20‐0.67)*** | 0.27 (0.15‐0.49)*** |
| For‐profit plan | 0.18 (0.14‐0.25)*** | 0.28 (0.21‐0.38)*** | 0.22 (0.16‐0.29)*** |
| n | 1232 | 1232 | 1232 |
| Pseudo R 2 | 0.14 | 0.10 | 0.13 |
aA high‐quality score on process measures is operationalized as a total score of 4 stars or higher for both of the two categories “Staying Healthy: Screening, Tests and Vaccines” and “Managing Chronic (Long Term) Conditions.”
bA high‐quality score on satisfaction measures is operationalized as a total score of 4 stars or higher for each of the three categories “Ratings of Health Plan Responsiveness and Care,” “Health Plan Members’ Complaints and Appeals,” and “Health Plans’ Telephone Customer Service.”
cMA HHI is Medicare Advantage health insurance market concentration, as measured by the Herfindahl‐Hirschman Index (HHI), with counties serving as the relevant geographic market and each parent insurance company's county‐level MA enrollment serving as the measure of market share. The unit of analysis is the MA plan level, so for these analyses, county MA HHI was enrollment‐weighted and aggregated to the plan level to calculate mean HHI across each plan's service area. For the above analyses, MA HHI was divided by 100 to allow for easier interpretation of coefficients.
dHospital market concentration was measured by calculating HHIs with geographic markets set at the core‐based statistical area (CBSA) level and with market shares defined using each hospital's total annual Medicare days. As above, CBSA‐level hospital HHI was enrollment‐weighted and aggregated to the plan level to calculate mean hospital HHI across each plan's service area. For the above analyses, hospital HHI was divided by 100 to allow for easier interpretation of coefficients.
***P < .001; **P < .01; *P < .05.