Table 1 Outcomes, variables, measures and methods of analysis
Outcome | Hypothesis | Outcome Measure | Analysis Methods |
|---|---|---|---|
Primary | |||
Deathsa | Intervention decreases overall mortality | All-cause mortality [binary] | Chi-squared test |
Secondary | |||
Nurse evaluation | Intervention enhances rate | Having a nurse evaluation within a year [binary] | Chi-squared test |
Dietitian visit in low BMI (<23 kg/m2) participants | Intervention enhances rate | Having a dietitian visit within a year [binary] | Chi-squared test |
Dispensed anti-diabetic drugs in criterion b patients | Intervention reduces drug-dispensing | WHO DDD of anti-diabetics in first 3 months [continuous] | T-test/ Wilcoxon |
Dispensed cholesterol-lowering drugs in criterion c patients | Intervention reduces drug-dispensing | DDD of cholesterol-lowering drugs in first 3 months [continuous] | T-test/ Wilcoxon |
Cost | Intervention reduces medical costs | Total CHS expense per patient [continuous] | T-test/ Wilcoxon |
Total drug use | Intervention reduces use | Yearly average total DDD [continuous] | T-test/ Wilcoxon |
Emergency room visits | Intervention reduces visits | Annual ER visits [continuous] | T-test/ Wilcoxon |
BMI - in low BMI (<23 kg/m2) participants | Intervention decreases weight-loss | BMI change in a year [continuous] | T-test/ Wilcoxon |
HbA1c% - in criterion b patients | Intervention increases HbA1c% | HbA1c in a year [continuous] | T-test/ Wilcoxon |
Hypoglycemia - in criterion b patients | Intervention decreases hypoglycemia risk | Number of glucose measures <70 mg/dL [continuous] | T-test/ Wilcoxon |
Hypocholesterolemia - in criterion c patients | Intervention decreases hypocholesterolemia | Number of cholesterol measures <160 mg/dL in a year [continuous] | T-test/ Wilcoxon |
Subgroup analyses will include | |||
Tertiles of BMI, HbA1c% and cholesterol | Participants with lower indexes may benefit more | Regression methods with appropriate interaction | |
Tertiles of Age | Older participants may benefit more | ||
Gender | A difference may be found | ||
Prior cardiovascular diagnosis vs. none | A difference may be found | ||
North vs. South district | Outcomes may be better in the North due to the direct participation of nurses | ||