Table 1.
Summary results of the association between 111 call characteristics and the probability of making an avoidable type 1 ED attendance within 24 hours
| Variable | Column 1 | Column 2 |
| Logistic regression, ORs (95% CI) |
OLS regression, percentage point changes (95% CI) |
|
| Patient characteristics | ||
| Patient aged 0–4 (vs all other ages) | 1.34*** (1.33 to 1.35) |
1.72*** (1.68 to 1.75) |
| Female (vs male) patient | 1.07*** (1.06 to 1.08) |
0.34*** (0.31 to 0.37) |
| Geographic characteristics | ||
| Quintile of area deprivation (1=least deprived, 5=most deprived) |
1.03*** (1.03 to 1.03) |
0.17*** (0.16 to 0.18) |
| Patient’s distance to hospital relative to GP | 1.00 (1.00 to 1.00) |
0.00 (−0.01 to 0.01) |
| Patient in rural (vs urban) area | 0.91*** (0.90 to 0.92) |
−0.44*** (−0.48 to to 0.40) |
| GP practice characteristics | ||
| Effect of 10 percentage point increase in % of patients saying they can typically get an appointment at GP practice | 0.87*** (0.83 to 0.90) |
−0.79*** (−0.97 to to 0.61) |
| Effect of 10 additional FTEs at GP practice |
0.99*** (0.99 to 0.99) |
−0.03*** (−0.04 to −0.03) |
| Call characteristics | ||
| Call received clinical input (vs no input) | 0.52*** (0.51 to 0.53) |
−3.65*** (−3.76 to to 3.54) |
| Call duration in minutes | 0.99*** (0.99 to 0.99) |
−0.02*** (−0.02 to to 0.02) |
| Bank holiday | 0.83*** (0.81 to 0.84) |
−0.09*** (−0.10 to to 0.08) |
| Christmas period (December 24–26) | 1.03* (1.00 to 1.06) |
0.13 (−0.02 to 0.27) |
| Additional controls? | Yes | Yes |
| Baseline | 0.059/1 | 5.6% |
| Goodness of fit | AUC=0.70 | R2=3.4% |
| Observations | 10 954 783 | 10 954 783 |
Additional controls=hour of day, day of week, month of year, year, region of England, 111 site and disposition code assigned to the call.
*p<0.05, **p<0.01, ***p<0.001.
AUC, area under curve; FTE, full-time equivalent; OLS, ordinary least squares.