- Design
- Secondary cohort analysis of a randomised trial population (ASPREE) with latent class analysis
- Population
- 16,095 community-dwelling Australians aged 70 and over without dementia or cardiovascular disease
- Primary outcome
- All-cause mortality by multimorbidity cluster over median 10.9 years
- Effect
- Frailty-depression-osteoporosis cluster aHR 1.56 (men) and 1.68 (women); gout-anaemia-kidney cluster in women aHR 1.96
This secondary analysis of the ASPREE trial cohort included 16,095 Australians aged 70 and over, free of dementia and cardiovascular disease at baseline, followed for a median of nearly 11 years; 4,217 died. Fifteen chronic conditions were grouped by statistical clustering.
88% had multimorbidity. The commonest cluster, hypertension with raised lipids, was not the most dangerous. A cluster combining frailty, depressive symptoms and osteoporosis was associated with about 56% higher mortality in men and 68% in women. Clusters centred on gout, kidney disease and metabolic conditions in men, and gout, anaemia and kidney disease in women, carried similar or higher risk.
The patterns differed by sex, which suggests that a simple count of conditions misses important information about which combinations matter.
The authors describe the findings as hypothesis-generating. But clinically they point to a recognisable group, the frail, low-mood older person with fragile bones, who is easily treated condition by condition when a comprehensive assessment is what is needed.
- Look for frailty, low mood and osteoporosis together in older patients; they clustered with higher mortality.
- Consider comprehensive geriatric assessment for patients with this combination.
- Check kidney function and haemoglobin in older patients with gout.
- Do not rely on a simple count of conditions to judge risk.
- Screen for depression in frail older adults as a routine part of review.
Why it matters
It shows which combinations of conditions, not just how many, identify the older patients at highest risk.
Don't overread it
Observational and hypothesis-generating; clusters may not reproduce in other cohorts.
The statistics, in plain English
An adjusted hazard ratio of 1.68 means the death rate was about two-thirds higher than in the reference group. Clusters are produced by a statistical method and may look different in other populations; they describe patterns, not causes.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for geriatrics, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free