Table 3.
Multilevel multivariable logistic regression analysis for prehospital delay.
| Variable | Unadjusted | Basic | Model 1 | Model 2 | Model 3 | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | p | Model | p | aOR | 95% CI | p | aOR | 95% CI | p | aOR | 95% CI | p | |
| Fixed effects | ||||||||||||||
| Area-level SES (Affluent) | 1 | 1 | 1 | |||||||||||
| Middle | 1.06 | (0.98–1.16) | 0.14 | 1.01 | (0.85–1.18) | 0.95 | 1.04 | (0.90–1.20) | 0.62 | |||||
| Deprived | 1.27 | (1.17–1.38) | <0.01 | 1.25 | (1.08–1.45) | 0.003 | 1.03 | (0.89–1.20) | 0.7 | |||||
| Individual-level SES | ||||||||||||||
| High school graduate (below high school) | 0.88 | (0.82–0.95) | <0.01 | 0.91 | (0.84–0.98) | 0.02 | 0.92 | (0.86–1.00) | 0.04 | 0.93 | (0.85–1.32) | 0.07 | ||
| Occupation (non-manual) | 1 | 1 | 1 | 1 | ||||||||||
| Manual | 1.14 | (0.95–1.36) | 0.15 | 1.09 | (0.90–1.30) | 0.39 | 1.08 | (0.90–1.29) | 0.4299 | 1.09 | (0.90–1.32) | 0.39 | ||
| Other | 1.00 | (0.83–1.21) | 1.00 | 0.95 | (0.78–1.16) | 0.61 | 0.96 | (0.78–1.16) | 0.64 | 1.04 | (0.84–1.27) | 0.73 | ||
| Inactive | 1.12 | (0.94–1.33) | 0.20 | 1.07 | (0.90–1.28) | 0.44 | 1.07 | (0.90–1.29) | 0.44 | 1.17 | (0.97–1.42) | 0.11 | ||
| National Health Insurance (Medicaid) | 0.99 | (0.84–1.16) | 0.88 | 1.00 | (0.85–1.18) | 1.00 | 1.01 | (0.86–1.18) | 0.93 | 0.94 | (0.79–1.11) | 0.45 | ||
| Covariates | ||||||||||||||
| EMS use (no) | 0.32 | (0.30–0.35) | <0.001 | 0.39 | (0.36–0.42) | <0.001 | ||||||||
| Interhospital transfer (no) | 2.29 | (2.11–2.48) | <0.001 | 1.85 | (1.69–2.03) | <0.001 | ||||||||
| Age group (years) (19–39) | 1 | 1 | ||||||||||||
| 40–65 | 1.12 | (0.90–1.40) | 0.30 | 1.13 | (0.90–1.43) | 0.31 | ||||||||
| ≥65 | 1.17 | (0.94–1.45) | 0.15 | 1.29 | (1.01–1.63) | 0.04 | ||||||||
| Male (female) | 0.99 | (0.92–1.06) | 0.70 | |||||||||||
| Previous diabetes (no) | 1.21 | (1.12–1.31) | <0.001 | 1.25 | (1.14–1.35) | <0.001 | ||||||||
| Previous hypertension (no) | 1.01 | (0.94–1.08) | 0.76 | |||||||||||
| Previous cardiovascular disease (no) | 0.79 | (0.74–0.85) | <0.001 | 0.83 | (0.77–0.90) | <0.001 | ||||||||
| Exercise (no) | 0.88 | (0.81–0.96) | <0.001 | 0.92 | (0.83–1.01) | 0.08 | ||||||||
| Smoker (never) | 1 | |||||||||||||
| Former | 1.08 | (1.00–1.18) | 0.05 | |||||||||||
| Current | 0.99 | (0.90–1.09) | 0.79 | |||||||||||
| Alcohol consumption (never) | 1.05 | (0.80–0.92) | 0.21 | |||||||||||
| Onset (day) | 0.86 | (0.80–0.92) | <0.001 | 0.85 | (0.79–0.92) | <0.001 | ||||||||
| Loss of consciousness (no) | 0.55 | (0.50–0.60) | <0.001 | 0.58 | (0.52–0.64) | <0.001 | ||||||||
| Recognition of symptoms (no) | 1.28 | (1.17–1.40) | <0.001 | 1.04 | (0.94–1.15) | 0.48 | ||||||||
| Urbanization | 0.74 | (0.69–0.80) | <0.001 | 0.90 | (0.78–1.04) | 0.17 | ||||||||
| Random effects | ||||||||||||||
| Area-level variance | 0.147 | <0.001 | 0.143 | <0.001 | 0.133 | <0.001 | 0.08 | <0.001 | ||||||
| Intraclass correlation | 4.27 | 4.15 | 3.88 | 2.37 | ||||||||||
| AUC | 0.599 | (0.599–0.609) | 0.599 | (0.589–0.608) | 0.693 | (0.684–0.702) | ||||||||
| AUC change | <0.001 | 0.095 | ||||||||||||
| Model fitness | ||||||||||||||
| AIC | 18,310.3 | 18,308.2 | 18,300.6 | 17,104.3 | ||||||||||
| BIC | 18,317.1 | 18,332.2 | 18,331.4 | 17,172.7 | ||||||||||
Abbreviations: SES—socioeconomic status; OR—odds ratio; CI—confidence interval; aOR—adjusted odds ratio; EMS—emergency medical service; AUC—area under curve; AIC—Akaike Information Criterion; BIC—Bayesian Information Criterion. All references are shown in parentheses. Basic—no predictor variables, random effect model; Model 1—basic model with individual-level SES included; Model 2—as per Model 1 but with area-level SES included; Model 3 = included both area-level and individual-level SES adjusted for age, exercise, diabetes, history of cardiovascular disease, stroke onset time, EMS use, transfer, loss of consciousness, recognition of symptoms, and urbanization status.