Figure 1.
Map of China, Chongqing and Qianjiang: geographic distribution of trial places.
Figure 2.
The study hypothesis, assumptions and proof paradigm.
Figure 3.
MDT: re-construction of the health human resource in three levels.
Figure 4.
MIP: re-enactment of the organisation collaboration and patient flow.
Table 2.
Minimum cases needed under index “continuity of clinical service”
Summary: group cases of 120/96 or 120/120 in treatment/control group could achieve over 80% power to detect a 0.2000 difference between group proportions.
aConsidering the intervention effect, we assumed the relation of conforming cases between control/treatment group varied from 0.8–1.0/1.0.
bAssumed the detectable proportion of cases conformed to the defined “continuity of clinical service” is 0.5–0.7 in the treatment group and 0.3–0.5 in the control group.
Table 4.
Result of grouping and patient sampling
Note: ① Type A for better cluster: larger and richer; type B for worse cluster: smaller and poorer. ② H for hypertension; D2 for diabetes 2. ③ Population, patient population and sample size are not averaged under “total average”. ④ All variables are compared with treatment group 1. ⑤ Distance is transferred into traffic time by public transportation. Time varies in accordance to the cluster terrain instead of linear map distance.
