Table 2.
Single-group interrupted time series analysis of monthly relative-value units per employee
| Periods | Parameters | All PHIs (95% CI) |
Central town and street PHIs (95% CI) |
Remote countryside PHIs (95% CI) |
Other PHIs (95% CI) |
|---|---|---|---|---|---|
|
Period I (Before the pilot reform#, n = 24 months) |
Baseline level (β0) | 159.61 (132.94–186.27)*** | 150.94 (114.90–186.98)*** | 166.32 (113.28–219.37)*** | 187.80 (156.62–218.99)*** |
| Baseline trend (β1) | 4.40 (2.40–6.40)*** | 4.40 (1.86–6.93)*** | 2.83 (−0.91–6.56) | 3.79 (1.59–5.98)*** | |
|
Period II (After the pilot reform, n = 36 months) |
Level change after reform (β2) | 25.20 (−11.22–61.60) | 38.82 (−7.38–85.01) | 10.72 (−57.28–78.70) | −15.92 (−55.88–24.05) |
| Trend change after reform (β3) | −1.55 (−3.78–0.68) | −2.29 (−5.12–0.54) | 7.58 (3.42–11.75)*** | 0.76 (−1.69–3.21) | |
| Trend after reform (β1 + β3) | 2.86 (1.98–3.74)*** | 2.53 (1.61–3.45)*** | 11.04 (8.57–13.51)*** | 4.93 (3.62–6.24)*** | |
|
Period III (After COVID-19 pandemic, n = 36 months) |
Level change after the pandemic (β4) | −19.55 (−51.55–12.45) | −4.40 (−45.00–36.20) | −96.14 (−155.90–−36.39)** | −23.21 (−58.34–11.91) |
| Trend change after the pandemic (β5) | −1.74 (−3.23 to −0.24)* | −1.19 (−3.09–0.71) | −9.35 (−12.14 to −6.55)*** |
−3.472 (−5.116 to −1.828)*** |
|
| Trend after pandemic (β1 + β3 + β5) | 1.12 (−0.47–2.71) | 1.34 (−0.53–3.20) | 1.69 (−1.76–5.13) | 1.46 (−0.86–3.77) |
Note: * p < 0.05, **p < 0.01, *** p < 0.001; #Pre-reform (2015–2016) monetary and RVU values were generated by a static micro-simulation model; seasonal pattern was adjusted by fitting Fourier terms in the regression model by two sine/cosine pairs