Estimated fixed effects of panel data regression
| Variable | Coefficient | 95% CI (lower to upper) | P value |
| Intercept | 8.00 | 7.81 to 8.18 | 0.00 |
| Years diagnosed | −0.001 | −0.002 to –0.001 | 0.04 |
| Pre-AIP | 0.95 | 0.85 to 1.05 | 0.00 |
| Post-AIP | 0.14 | 0.05 to 0.23 | 0.00 |
| Male sex | −0.11 | −0.20 to –0.02 | 0.02 |
| SE group 2 | −0.14 | −0.32 to 0.03 | 0.10 |
| SE group 3 | −0.17 | −0.35 to 0.02 | 0.08 |
| SE group 4 | −0.22 | −0.38 to –0.06 | 0.01 |
| SEgGroup 5 | −0.35 | −0.52 to –0.19 | 0.00 |
| Pump therapy | −0.70 | −0.81 to –0.57 | 0.00 |
| MDI regimen | −0.15 | −0.29 to –0.00 | 0.04 |
| Three times a day regimen | −0.07 | −0.15 to 0.02 | 0.13 |
| Age | 0.05 | 0.04 to 0.06 | 0.00 |
| Goodness of fit | Value | ||
| R2 | 0.69 | ||
| Adjusted R2 | 0.69 | ||
| BIC | 43 617 |
Pre-AIP indicates a dummy variable equal to 1 if there was an AIP in the previous 6 months prior to the data point and equal to 0 otherwise. Post-AIP is a dummy variable which is equal to 1 if an AIP occurs in the 6 months after the data point and 0 otherwise. SE group is a categorical variable which indicates the expected difference between a patient in a specific SE group relative to the lowest SE group, which is taken as the index group.
AIP, active intervention programme; MDI, multiple daily injection; SE, socioeconomic.