Table A5.
Intervention analysis of the costs and DID series by active principle. Real values from January 2009 to June 2015 and predictions from July 2015 to June 2016 with dummy variable V1.
| Active Substance | Amoxicillin | Amoxicillin and Inhibitors | Azithromycin | |||
|---|---|---|---|---|---|---|
| Group | “access” | “access” | “watch” | |||
| Chosen by | High prescription (23% of recipes) | High prescription (21% of recipes) | High prescription (12% of recipes) | |||
| Series | Costs (euros) | DID | Costs (euros) | DID | Costs (euros) | DID |
| ARIMA model | (2,0,0) (2,1,0) |
(2,1,2) (0,1,1) |
(0,1,2) (0,1,1) |
(1,0,0) (2,1,1) |
(0,1,2) (0,1,1) |
(0,1,2) (0,1,1) |
| AR1 | 0.438 *** | −0.441 *** | - | 0.560 *** | - | - |
| AR2 | 0.098 * | −0.292 * | - | - | - | - |
| MA1 | - | −0.151 * | 0.270 *** | - | −0.052597 * | −0.040777 * |
| MA2 | - | −0.841 * | 0.580 *** | - | −0.547567 *** | −0.638842 *** |
| SAR1 | −0.547 *** | - | - | −0.194 ** | - | - |
| SAR2 | −0.405 *** | - | - | −0.374 *** | - | - |
| SMA1 | - | 0.845 *** | 0.811 *** | 0.376 * | −0.521205 *** | −0.577372 *** |
| V1 | −0.047 * | −0.041 * | −0.045 * | −0.040 * | - | - |
| Q test (p-value, delay 18) | 15.412 (0.3951) |
9.606 (0.7260) |
17.420 (0.3214) |
15.011 (0.377) |
16.517 (0.223) |
17.906 (0.1611) |
| AIC | −7.222 | −6.511 | −7.182 | −7.147 | −18.42 | −12.45 |
| Residual sum of squares | 0.078 | 0.088 | 0.091 | 0.087 | 0.0615 | 0.0655 |
| Standard error of the regression | 0.017 | 0.026 | 0.016 | 0.018 | 0.0807 | 0.0822 |
| Effect on the series (calculations of savings in Table A6) |
−9.26% | −0.277 DID (−6.69%) | −10.65% | −0.193 DID (−4.19%) |
−14.29% | −0.174 DID (−12.30%) |
Legend: (*) significant level equal to or less than 0.05; (**) significance level equal to or less than 0.01; (***) level of significance equal to or less than 0.001; ARIMA (integrated autoregressive moving average model); AR1 (first autoregressive term); AR2 (second autoregressive term; MA1 (first moving average term), MA2 (second moving average term); SAR1 (first seasonal autoregressive term); SMA1 (first seasonal moving average term); V1 (dummy variable that takes the value 1 in June 2013 and −1 in July 2013); AIC (Akaike information criterion); DID (daily defined doses per thousand people). Source: Own elaboration from data provided by the Pharmacy Directorate of the Basque Government.