Checkpoint: What Is the Number Actually Measuring?

This checkpoint connects the three healthy-life indicators already established—HALE, HLY, and Dis-FLE—so the reader can make one practical judgment: whether a comparison is measuring the same thing at all. The useful task is not to repeat their definitions, but to place each number on four dimensions that travel with it: the health state, the evidence stream, the estimation rule, and the population represented.

Four dimensions that travel with every number

The first dimension is the health state. HALE represents years in “full health” after weighting functional-health loss across causes. HLY represents expected years without long-term activity limitation as reported through the GALI survey instrument. Dis-FLE represents expected time without a new event from the French study’s set of severe clinically recorded conditions. All three use the language of healthy years, but the boundary around a healthy year is different in each case.

That difference is not a semantic footnote. It changes what counts as evidence of improvement or deterioration. A population could experience fewer severe hospital events while many people still report activity limitations. It could also report fewer limitations while clinical diagnoses become more complete. Those outcomes might coexist because they refer to different parts of health. A measure is not contradicted simply because another measure sees a different dimension.

The second dimension is the evidence stream. HALE combines a life-table mortality framework with information about disease, injury, disability, or functional-health loss. HLY uses self-evaluated long-term activity limitation. Dis-FLE uses hospital discharge dates and standardized diagnoses to follow severe events in individual trajectories. The evidence source determines not only what is visible, but also what can be missing: mortality registration, morbidity prevalence, survey interpretation, hospital access, and diagnoses that do not describe functional consequences all enter at different points.

The third dimension is the estimation rule. HALE applies a health-loss fraction to years lived and averages equivalent healthy years among survivors at a specified age. Dis-FLE estimates a conditional remaining duration from a disease-free survival curve. HLY is also an expectancy, but its underlying health state is obtained from survey responses rather than the clinical event transition used by Dis-FLE. “Expected years” is therefore not one common mathematical object. The conditioning and weighting behind the expectation matter.

The fourth dimension is the population. An indicator may describe a general population, a sex-specific group, people who have reached an age, or a selected group visible in a health system. The French hospital study’s post-exclusion cohort is large, but the multivariable results remain relative to people represented in the hospitalized population. Its sex-specific whole-population adjustment is a separate construction with explicit assumptions. A result cannot inherit a broader denominator merely because the source dataset is large.