Dr Luke Wee
November 19, 2026

Where the Gap Actually Sits

Life expectancy counts the endpoint. It tells you when a life stopped. It says nothing about the condition of the years preceding it, and it is in those years that the deeper inequality sits.
Found this valuable? Share it forward.

Whenever the health gap between Māori and non-Māori is discussed, the same number does the work. Around 7 years of life expectancy. It is a serious number, it is reported carefully, and it deserves the attention it gets.

It is also the wrong instrument for the question it is usually asked to answer.

Diagram comparing life expectancy, which counts only the endpoint, with health expectancy, which counts the years lived in good health
Image generated by Gemini

Life expectancy counts the endpoint. It tells you when a life stopped. It says nothing about the condition of the years preceding it, and it is in those years that the deeper inequality sits.

Health expectancy is the alternative — years lived free of functional limitation, or with limitation that does not require assistance. It is the measure longevity medicine has spent a decade insisting is the one that matters. Not how long, but how well, and for how long before decline sets in. Healthspan over lifespan is the premise of the entire field, and it is a premise I accept.

Applied to a population rather than to an individual, that premise produces a result the field has not reckoned with.

The Gap Behind The Gap

On the NZ Ministry of Health's 2013 estimates, Māori men could expect 54.3 years of independent life against 66.7 years for non-Māori men. That is a gap of 12.4 years, set against a lifespan gap of roughly seven. Māori men lived 74.4% of their lives independently; non-Māori men, 83%. Between 1996 and 2013, years lived in good health rose for Māori women, for non-Māori women and for non-Māori men. For Māori men, the figure fell by 1.7 years.

Chart: Māori men 54.3 years of independent life out of 73.4, non-Māori men 66.7 out of 81.0; a lifespan gap of 7.6 years and a healthspan gap of 12.4 years

Something should be said immediately about what has happened since, because the news is better than that paragraph suggests. Between 2005–2007 and 2022–2024, life expectancy rose for every ethnic group in New Zealand, and Māori gained the most — 3.3 years for men and 2.9 for women. The male lifespan gap has narrowed from just under nine years to about seven and a half. That is real progress and it should be said plainly.

It also tells us nothing about the years themselves, which is the whole difficulty.

Thirteen Years Without Looking

Because the 2013 estimates are still the most recent figures we have. The Ministry's Health and Independence Report 2025, published this month, reports healthy life expectancy using the Global Burden of Disease measure, and states plainly that the available estimates are only reported for males, females and the total population, and do not show differences by ethnicity or other population groups. The independent life expectancy series that produced those 2013 numbers has not been updated. It has been replaced by a measure that cannot answer the question at all.

So we are in the position of knowing, with annual precision, how the lifespan gap is moving, and having no current idea whatsoever how the healthspan gap is moving. Thirteen years, and counting.

What We Choose to Count

This is worth sitting with, because it concerns what we choose to count.

Mortality is registered continuously. Every death in this country is recorded, coded and available, which is why life expectancy can be reported with confidence and updated regularly. Health expectancy depended on a household disability survey conducted roughly once every ten years — and the ethnicity-disaggregated version of it is now simply not produced. A measure taken every year shapes policy. A measure taken every decade does not. A measure that has been discontinued shapes nothing at all.

The asymmetry is not neutral, and it has consequences for what can be claimed. The 2024 edition of the same report described the gap in healthy life expectancy between Māori and non-Māori as decreasing. For life expectancy, that is supportable. For healthy life expectancy, there is no published ethnicity-split figure from which the claim could have been derived, and the 2025 report says so in its own text. The measure gets invoked while remaining unmeasured, which is the position a field ends up in when it decides something matters and then does not count it.

The same problem appears in clinical practice, at a smaller scale. A field organised around healthspan ought to be measuring function directly — strength, cardiorespiratory fitness, gait, the capacity to get up from the floor unassisted at 80 — and treating biochemical markers as proxies that earn their place by predicting those outcomes. Much of longevity medicine has this the wrong way round. The panel is the object of attention and function is inferred from it, which reverses the order of the claim the field makes about itself.

Diagram: healthspan as usually practised, with the biomarker panel first and function inferred, against what the premise implies, with function measured directly and biomarkers as proxies
Illustration by Gemini

The Decade Before Death

What the arithmetic points to is a difference in kind, not only in degree. A shortened life is an absence. It is grief, and it is borne by the people who remain. Morbidity is not an absence. It is a presence — years of dependency lived through in sequence, by the person and by the household that reorganises itself around him. Life expectancy renders none of that visible. It counts the death and passes over the decade.

And the direction of travel at population level is not encouraging. The gap between life expectancy and health-adjusted life expectancy in New Zealand widened from 11.3 years in 1990 to 13.1 years in 2023. Health-adjusted life expectancy itself has stalled and, for both sexes, declined since 2020. The Ministry attributes the widening to expansion of morbidity — the pattern in which better treatment postpones death more reliably than it postpones disease — and notes that New Zealand sits among a small group of countries where a relatively high proportion of life is spent living with illness or disability. That is the national picture. We simply cannot see how it is distributed.

The Work We Already Know How to Do

Much of what produces the gap is amenable. Cardiovascular disease, diabetes, lung cancer and respiratory disease. These are conditions we know how to detect earlier and treat better, and the interventions are neither novel nor expensive.

It is worth being concrete about what that work actually consists of, because it is unglamorous enough to be routinely underestimated. Blood pressure identified and then controlled, rather than recorded at successive visits and left. Lipids treated to target in men who are not naturally inclined to return. Glycaemia caught during the decade in which it is drifting rather than the year in which it announces itself. Smoking cessation offered again after it has been declined, and again after that, because the evidence on repeated brief intervention is unambiguous and the single conversation is not where the benefit lies. Spirometry performed in a man who has been breathless for two years and has attributed it to age.

None of this is at the frontier. All of it rests on evidence considerably stronger than most of what the longevity field currently discusses with enthusiasm.

The obstacle is not knowledge. It is continuity — the same clinician, seeing the same person, often enough to notice a trajectory rather than a reading. That is a structural feature of how care is funded and delivered, not a deficit in the person who do not return.

What Arrives in the Consulting Room

But I want to be accurate about how much of this sits inside a consulting room, because overstating the physician's reach is its own kind of dishonesty.

The disease presenting in a man of 55 was set in motion long before he sat down. Rheumatic heart disease is a housing problem that arrives at cardiology thirty years late, and that chain of causation is about as well established as anything in New Zealand medicine. Type 2 diabetes at 45 carries the imprint of what was affordable in a household decades earlier — a looser claim, but a well-supported one. Smoking-related disease reflects an exposure landscape shaped when the man was a teenager, and by the time the emphysema is audible that landscape has finished its work.

What arrives in the room is the residue of conditions that are already historical.

I can alter a trajectory. I cannot reach back and change the years that set it.

Timeline: housing in childhood, household income and food environment, and tobacco exposure in adolescence, all operating decades before the consultation at 55
Illustration by Gemini

There is a temptation, in discussions of health inequity, to locate the responsibility with clinicians. I understand the appeal of it. The consultation is visible, the physician is nameable, and it is easier to ask something of an individual than of a system. But the gap is not created in a consulting room. It arrives there. I inherit the outcomes of housing, income and commercial environments that shaped a man's health long before he became my patient, and no amount of skill within a fifty-minute appointment reverses thirty years of exposure. Governments design the conditions in which health is achievable. Physicians work within conditions already designed.

This is not an argument for therapeutic pessimism. It is an argument for advocacy about where the leverage sits with the clinician now.

Where the Health System Still Matters

It is also an argument with a limit, and the limit is worth stating precisely. Five-year net cancer survival for Māori in 2022–23 was 58.5%, against 69.6% for non-Māori. That is an 11-point difference measured from the point of diagnosis onward — after the housing, after the income, after the exposures, at the stage where the health system has already taken the person in hand. The Ministry's 2024 report made a related observation: Māori living in areas of high deprivation have lives on average 7.4 years shorter than non-Māori, non-Pacific peoples in those same areas, and the determinants of health explain some of the difference but not all of it. Whatever accounts for the remainder is operating a good deal closer to where I work than housing policy is.

Which suggests something about timescale, and here I am reasoning by analogy rather than from evidence. The changes that closed earlier health gaps — housing, income, sanitation, tobacco control — took decades to register in mortality data, because they altered exposures in children who then had to live long enough to become statistics. A gap produced across generations will most likely take generations to close. That is not pessimism. It is a statement about latency, and it argues for beginning rather than for waiting.

The measure remains the measure. If healthspan is the right instrument, and the field insists that it is, then the last time anyone in this country measured it by ethnicity, the gap for Māori men was 12 years against a lifespan gap of 7 years, and their years of good health had gone backwards while everyone else's improved.

That was 2013. We have not looked since.

Tane Mahuta. Photo by Yathursan G on Unsplash

Dr Luke Wee is a general practitioner working across the Tasman in Aotearoa New Zealand. He thinks about medicine, health equity, and what clinical practice can learn from lived experiences of patients. This includes health, whānau and the conditions in which people live.

First published in The Longer Consultation on LinkedIn, 11 September 2026. Cover photograph: Suad Kamardeen on Unsplash.

Related reading