What would a health system designed around prevention actually measure?

Health systems tend to measure what they do: operations performed, appointments held, beds occupied. A system built around prevention would need to measure something else: how many people stay well, how early problems are found, and how much of the population's health depends on things outside the clinic.

The World Health Organization highlights noncommunicable diseases, such as heart disease, cancers, diabetes and chronic respiratory illness, as a major share of illness and death worldwide, and many of these are influenced by long-term factors that prevention can change.

We are asking readers to propose measures. A good answer would suggest three to five indicators, explain what each would reveal that current measures do not, say who would collect the data and how, and describe how to avoid perverse incentives, such as hiding illness or avoiding difficult patients.

Examples from health systems you know, including where prevention has worked, are especially welcome. This is not medical advice.