Multimorbidity kept accumulating into the late nineties, and mostly not at the end of life

Design

nationwide register-based cohort analysis of population-level disease accumulation with survivorship and proximity-to-death decomposition

Population

all individuals born in Sweden 1920-1922 who survived to age 70, followed to age 100

Primary outcome

trajectory of disease accumulation between ages 70 and 100

Effect

accumulation accelerated to 80, stabilised 80-90, decelerated thereafter; mean 3.9 diseases in men and 3.5 in women at 90; more than half of those ≥95 had five or more; most accumulation in non-terminal years

Swedish nationwide registers allow something few datasets do: following an entire birth cohort - everyone born between 1920 and 1922 who survived to 70 - through to age 100, and separating how much of the observed disease accumulation is real accumulation and how much is an artefact of who survives.

Disease burden rose steeply to age 80, plateaued in pace between 80 and 90, and decelerated after that - but never stopped. At 90 the mean was 3.9 diseases in men and 3.5 in women, and more than half of those aged 95 and over carried five or more. Cardiovascular disease was the dominant contributor throughout, while musculoskeletal and neurosensorial conditions became progressively more important with age. The finding that reframes the rest is that most accumulation occurred during non-terminal years, not in a terminal decline.

That last point cuts against a comfortable assumption. If disease accumulation were concentrated in a final year or two, the service implication would be palliative. It is not: it is spread across a long stretch of life in which people are living with five conditions rather than dying of them. Combined with the shift towards musculoskeletal and sensory disease at advanced ages, that argues for a model built around function, hearing, vision and mobility rather than around organ-specific disease control - which is the opposite of how most specialty clinics are organised, in Sweden and considerably more so in India, where an 85-year-old with five conditions typically attends five separate outpatient departments.

• Expect four or more conditions in a patient over 90, and five or more in more than half of those over 95
• Build the review around function - mobility, hearing, vision, continence - rather than around each organ system in turn
• Do not treat multimorbidity in extreme old age as a terminal-phase phenomenon; most accumulation happened outside the final years
• Consolidate follow-up where possible; five conditions managed in five clinics is five drug charts and no overview
• Note this is a Swedish cohort born in the early 1920s - the absolute numbers reflect that population's survival and diagnostic patterns

Why it matters

It says the long stretch of life lived with five conditions, not the final decline, is where multimorbidity actually sits.

Don't overread it

Register-based disease counts in one Swedish birth cohort - the trajectory transfers better than the absolute numbers do.

The statistics, in plain English

A cohort followed from 70 to 100 loses most of its members along the way, and everything measured at 95 describes an unusually robust survivor group - so the plateau in accumulation after 90 partly reflects who is left, which is exactly what the survivorship analysis here is designed to separate. Mean disease counts depend heavily on how diseases were defined in register data: a register that codes more conditions produces a higher count without anyone being sicker. The comparison across ages within one consistent coding system is what makes the trajectory interpretable.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.