Study analysis · The Journals of Gerontology Series A: Biological Sciences and Medical Sciences · 2023
In older adults, walking speed and chair-rise ability begin a steep terminal decline about 2.5 years before death — and the last year is when the cliff drops hardest.
In 702 older adults, physical function dropped slowly for years, then 6–8 times faster (relative) in the last 1–2.5 years before death.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
In simple terms
This study watched a group of older people over many years and measured their physical abilities until they died. It shows that physical abilities tend to drop faster in the last couple of years before death, but it cannot prove that dying causes the drop or that anything can stop it. It is like noticing a pattern, not proving what causes it.
What’s the bottom line?
Researchers followed 702 older adults until death and measured walking, balance, and chair-standing ability. They found that physical function declines slowly for years, then drops much faster in the last 1 to 2.5 years of life.
How strong is this study?
The study is strong for this kind of research because it followed many people for a long time, used real physical tests, and almost everyone's death was recorded. But it was still just one group in one area, so we should be careful about assuming the same pattern holds for all older adults everywhere.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
37 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control groupno control group
- Sample size (n=702)+19.4/20
- Follow-up+10/10
100 / 100
54 / 100
- P-valuesno p-values reported
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 546 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. This is an observational cohort study without randomization or a control group. Although it uses repeated measures and time-to-death trajectories, it cannot rule out confounding, reverse causation, selection effects, or unmeasured factors. It can describe temporal patterns and associations, but it cannot prove that impending death causes the acceleration in physical function decline or that modifying any factor would change the trajectory.
No Conflicts
No conflicts of interest identified
No conflicts of interest identified; no funding or COI statement was provided in the text.
The provided text lacks a conflict of interest or funding disclosure section. The study is an academic cohort (Yale PEP Study) and no industry ties are evident. Absence of disclosure prevents full assessment of undisclosed interests.
Key takeaways
- 01
Terminal decline was 6–8 times steeper than pre-terminal decline (relative comparison).
- 02
Absolute declines: overall physical function (SPPB) fell about 0.4 points/year before terminal decline and about 2.9 points/year during terminal decline; gait speed worsened about 0.4 vs 2.7 seconds/year; chair rise worsened about 0.3 vs 1.4 seconds/year.
- 03
Terminal decline started about 1.0 year before death for SPPB, 2.5 years for chair rise, and 2.6 years for gait speed.
- 04
Compared with frailty deaths, dementia deaths had up to 6 months earlier onset and cancer deaths up to 3 months later onset for SPPB (absolute differences in months).
- 05
For a human, this means the average older adult in this study lost function slowly for years, then much faster in the final year or two of life.
- 06
The absolute differences are concrete: about 2.5 extra SPPB points lost per year in the terminal phase versus pre-terminal, and about 2.3 extra seconds per year for gait speed.
- 07
The study did not report an absolute risk increase for death; it reports absolute decline rates and absolute differences in when decline began by cause of death.
Surprising findings
- Chronic disease count was not associated with the onset of terminal decline.People often assume that more chronic conditions means earlier physical collapse, but the study found no link to when the steep decline began.
- Gait speed and chair rise detected terminal decline about 2.5–2.6 years before death, while SPPB detected it only 1.0 year before death.The composite SPPB is widely used, yet the simpler continuous measures picked up the acceleration much earlier — likely because SPPB is coarser and hits a floor effect.
- Cancer deaths had a later onset of terminal decline than frailty deaths, by up to 3 months for SPPB and 5 months for chair rise.Cancer is often thought to cause the sharpest and earliest physical decline, but this study found the opposite for onset timing.
- Physical terminal decline followed cognitive terminal decline by a few months.Many assume body and mind decline together, but cognition tipped into rapid decline about 2.7 years before death, with physical decline starting a few months later.
- People born before 1920 had a later onset of terminal decline by about 1 month for gait speed and 4 months for chair rise compared with those born in 1920 or later.Later-born cohorts are often assumed to be healthier, but here they entered the steep decline phase earlier.
Practical takeaways
For older adults, track gait speed and chair-rise time over time — not just a single snapshot. A sudden acceleration may signal a terminal decline phase.
This study was retrospective in deceased older adults and cannot predict individual death. Measurements near death are often missing.
medium confidenceDo not rely on the number of chronic diseases alone to estimate when physical decline will accelerate.
The cohort was initially nondisabled, regional, and mostly White; findings may not generalize.
medium confidenceFor care planning, people dying from dementia may enter the steep decline phase earlier — up to 6 months earlier than frailty deaths — potentially requiring earlier support.
Cause-of-death classification was based on death certificates and last assessment, and differences are absolute months, not relative risks.
medium confidenceStaying physically active in later life may be linked to a slightly later onset of terminal decline — about 3 months later for chair rise.
Observational association only; cannot prove that increasing activity delays terminal decline.
low confidenceWhy this study matters
The 6–8x terminal decline cliff
In 702 deceased adults aged 70+, physical function declined 6–8 times faster in the terminal phase than pre-terminal (relative comparison). Absolute declines: SPPB fell about 0.4 points/year before terminal decline vs 2.9 points/year during it; gait speed worsened about 0.4 vs 2.7 seconds/year; chair rise worsened about 0.3 vs 1.4 seconds/year.
It suggests death casts a measurable shadow years before it arrives, and the final years are not just more of the same gradual aging.
Gait speed is an early warning system
Terminal decline began about 1.0 year before death for SPPB, 2.5 years for chair rise, and 2.6 years for gait speed. Gait speed and chair rise are continuous measures, while SPPB is coarser and hit a floor effect from the sixth wave onward.
A simple walking test may flag decline earlier than a composite physical battery — and it’s cheap, low-tech, and non-invasive.
Physical decline follows cognitive decline by a few months
Cognitive function (MMSE) began terminal decline about 2.7 years before death with a 15-fold steepening. Physical terminal decline began about 1.0–2.6 years before death and was 6–8 times steeper. The physical terminal decline generally followed cognitive terminal decline by a few months.
The mind may tip into rapid decline before the body does, challenging the idea that physical and cognitive decline are simultaneous.
Cause of death shifts the timeline
Compared with frailty deaths, dementia deaths had up to 6 months earlier onset of terminal decline in SPPB, while cancer deaths had up to 3 months later onset. For gait speed, dementia was 3 months earlier and cancer 3 months later; for chair rise, dementia 1 month earlier and cancer 5 months later.
Not all dying trajectories are the same. Dementia means a longer period of low function; cancer often means a later but possibly sharper drop.
Chronic disease count did not predict when decline starts
Baseline count of chronic diseases showed no association with estimated change points for SPPB, gait speed, or chair rise time after adjustment. This was surprising because more diseases are often assumed to mean earlier decline.
It suggests the timing of terminal decline is not simply a tally of diagnoses — something else, perhaps underlying physiological reserve, may drive it.
Lifestyle and obesity are linked to timing
High baseline physical activity was associated with about 3 months later onset of terminal decline in chair rise time. Obesity was associated with about 2 months earlier onset in gait speed and chair rise time. These are associations, not proven causes.
It offers a hopeful message that activity might delay the steep phase, while obesity might hasten it — but the study cannot prove cause and effect.
Want the whole report?
Detailed mode opens the full scientific breakdown — every score component, the methodology, conflicts of interest, the evidence analysis behind each claim, and the raw study data.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Researchers followed 702 older adults until death and measured walking, balance, and chair-standing ability. They found that physical function declines slowly for years, then drops much faster in the last 1 to 2.5 years of life.
Research results
Terminal decline was 6–8 times steeper than pre-terminal decline (relative comparison). Absolute declines: overall physical function (SPPB) fell about 0.4 points/year before terminal decline and about 2.9 points/year during terminal decline; gait speed worsened about 0.4 vs 2.7 seconds/year; chair rise worsened about 0.3 vs 1.4 seconds/year. Terminal decline started about 1.0 year before death for SPPB, 2.5 years for chair rise, and 2.6 years for gait speed. Compared with frailty deaths, dementia deaths had up to 6 months earlier onset and cancer deaths up to 3 months later onset for SPPB (absolute differences in months).
What this means - more context
For a human, this means the average older adult in this study lost function slowly for years, then much faster in the final year or two of life. The absolute differences are concrete: about 2.5 extra SPPB points lost per year in the terminal phase versus pre-terminal, and about 2.3 extra seconds per year for gait speed. The study did not report an absolute risk increase for death; it reports absolute decline rates and absolute differences in when decline began by cause of death.
To determine whether and when physical function in older adults shows a terminal decline phase—an acceleration of decline in the last years before death—and to compare it with terminal cognitive decline.
In 702 deceased adults aged 70+ from the Yale PEP Study, physical function declined 6–8 times faster in the terminal phase than pre-terminal (relative comparison). Absolute declines: SPPB about 0.4 points/year pre-terminal vs 2.9 points/year terminal; gait speed worsened about 0.4 vs 2.7 seconds/year; chair rise about 0.3 vs 1.4 seconds/year. Terminal decline began about 1.0 year before death for SPPB, 2.5 years for chair rise, and 2.6 years for gait speed.
Methods Used
Longitudinal cohort of 702 deceased adults aged 70+ from the Yale PEP Study, with 4,133 SPPB measurements up to 20 years before death plus continuous gait speed and chair rise times. Random change point mixed models estimated onset and steepness of terminal decline, adjusted for age at death.
Main Finding
All three physical function measures showed terminal decline. Onset was about 1.0 year before death for SPPB, 2.5 years for chair rise, and 2.6 years for gait speed. Terminal decline was 6–8 times steeper than pre-terminal decline (relative comparison); absolute terminal slopes were 2.9 SPPB points/year, 2.7 seconds/year for gait speed, and 1.4 seconds/year for chair rise. Dementia death was associated with up to 6 months earlier onset and cancer death with up to 3 months later onset for SPPB compared with frailty death (absolute differences in months).
Confidence Level
Moderate to high. Strengths include long follow-up, validated objective measures, repeated assessments, and near-complete mortality. Limitations include observational design, single regional initially nondisabled cohort, retrospective analysis of deceased participants, and no prospective individual prediction.
Study Flags
Red Flags
- •Observational study cannot establish causality.
- •Single regional cohort of initially nondisabled older adults may not generalize to all older populations.
- •Retrospective analysis of deceased participants cannot prospectively predict individual terminal decline.
Surprising Findings
Chronic disease count was not associated with the onset of terminal decline.
People often assume that more chronic conditions means earlier physical collapse, but the study found no link to when the steep decline began.
Practical Takeaways
For older adults, track gait speed and chair-rise time over time — not just a single snapshot. A sudden acceleration may signal a terminal decline phase.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 546 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study watched a group of older people over many years and measured their physical abilities until they died. It shows that physical abilities tend to drop faster in the last couple of years before death, but it cannot prove that dying causes the drop or that anything can stop it. It is like noticing a pattern, not proving what causes it.
The study has a COI section but no disclosure was found. A small penalty has been applied.
Strengths
- Longitudinal cohort with up to 20 years of follow-up and repeated objective physical performance assessments.
- Large sample of 702 deceased older adults with 4,133 SPPB measurements and thousands of gait and chair rise measurements.
- Near-complete mortality follow-up (93%), reducing attrition bias for mortality-related outcomes.
Weaknesses
- Observational cohort design cannot establish causality.
- No control group and no randomization.
- Single regional cohort limits generalizability.
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Researchers followed 702 older adults until death and measured walking, balance, and chair-standing ability. They found that physical function declines slowly for years, then drops much faster in the last 1 to 2.5 years of life.
Research results
Terminal decline was 6–8 times steeper than pre-terminal decline (relative comparison). Absolute declines: overall physical function (SPPB) fell about 0.4 points/year before terminal decline and about 2.9 points/year during terminal decline; gait speed worsened about 0.4 vs 2.7 seconds/year; chair rise worsened about 0.3 vs 1.4 seconds/year. Terminal decline started about 1.0 year before death for SPPB, 2.5 years for chair rise, and 2.6 years for gait speed. Compared with frailty deaths, dementia deaths had up to 6 months earlier onset and cancer deaths up to 3 months later onset for SPPB (absolute differences in months).
What this means - more context
For a human, this means the average older adult in this study lost function slowly for years, then much faster in the final year or two of life. The absolute differences are concrete: about 2.5 extra SPPB points lost per year in the terminal phase versus pre-terminal, and about 2.3 extra seconds per year for gait speed. The study did not report an absolute risk increase for death; it reports absolute decline rates and absolute differences in when decline began by cause of death.
To determine whether and when physical function in older adults shows a terminal decline phase—an acceleration of decline in the last years before death—and to compare it with terminal cognitive decline.
In 702 deceased adults aged 70+ from the Yale PEP Study, physical function declined 6–8 times faster in the terminal phase than pre-terminal (relative comparison). Absolute declines: SPPB about 0.4 points/year pre-terminal vs 2.9 points/year terminal; gait speed worsened about 0.4 vs 2.7 seconds/year; chair rise about 0.3 vs 1.4 seconds/year. Terminal decline began about 1.0 year before death for SPPB, 2.5 years for chair rise, and 2.6 years for gait speed.
Methods Used
Longitudinal cohort of 702 deceased adults aged 70+ from the Yale PEP Study, with 4,133 SPPB measurements up to 20 years before death plus continuous gait speed and chair rise times. Random change point mixed models estimated onset and steepness of terminal decline, adjusted for age at death.
Main Finding
All three physical function measures showed terminal decline. Onset was about 1.0 year before death for SPPB, 2.5 years for chair rise, and 2.6 years for gait speed. Terminal decline was 6–8 times steeper than pre-terminal decline (relative comparison); absolute terminal slopes were 2.9 SPPB points/year, 2.7 seconds/year for gait speed, and 1.4 seconds/year for chair rise. Dementia death was associated with up to 6 months earlier onset and cancer death with up to 3 months later onset for SPPB compared with frailty death (absolute differences in months).
Confidence Level
Moderate to high. Strengths include long follow-up, validated objective measures, repeated assessments, and near-complete mortality. Limitations include observational design, single regional initially nondisabled cohort, retrospective analysis of deceased participants, and no prospective individual prediction.
Study Flags
Red Flags
- •Observational study cannot establish causality.
- •Single regional cohort of initially nondisabled older adults may not generalize to all older populations.
- •Retrospective analysis of deceased participants cannot prospectively predict individual terminal decline.
Surprising Findings
Chronic disease count was not associated with the onset of terminal decline.
People often assume that more chronic conditions means earlier physical collapse, but the study found no link to when the steep decline began.
Practical Takeaways
For older adults, track gait speed and chair-rise time over time — not just a single snapshot. A sudden acceleration may signal a terminal decline phase.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 546 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study watched a group of older people over many years and measured their physical abilities until they died. It shows that physical abilities tend to drop faster in the last couple of years before death, but it cannot prove that dying causes the drop or that anything can stop it. It is like noticing a pattern, not proving what causes it.
The study has a COI section but no disclosure was found. A small penalty has been applied.
Strengths
- Longitudinal cohort with up to 20 years of follow-up and repeated objective physical performance assessments.
- Large sample of 702 deceased older adults with 4,133 SPPB measurements and thousands of gait and chair rise measurements.
- Near-complete mortality follow-up (93%), reducing attrition bias for mortality-related outcomes.
Weaknesses
- Observational cohort design cannot establish causality.
- No control group and no randomization.
- Single regional cohort limits generalizability.
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
The study is strong for this kind of research because it followed many people for a long time, used real physical tests, and almost everyone's death was recorded. But it was still just one group in one area, so we should be careful about assuming the same pattern holds for all older adults everywhere.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
37 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control groupno control group
- Sample size (n=702)+19.4/20
- Follow-up+10/10
100 / 100
54 / 100
- P-valuesno p-values reported
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 546 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. This is an observational cohort study without randomization or a control group. Although it uses repeated measures and time-to-death trajectories, it cannot rule out confounding, reverse causation, selection effects, or unmeasured factors. It can describe temporal patterns and associations, but it cannot prove that impending death causes the acceleration in physical function decline or that modifying any factor would change the trajectory.
No Conflicts
No conflicts of interest identified
No conflicts of interest identified; no funding or COI statement was provided in the text.
The provided text lacks a conflict of interest or funding disclosure section. The study is an academic cohort (Yale PEP Study) and no industry ties are evident. Absence of disclosure prevents full assessment of undisclosed interests.
Standing
The people behind it
The researchers who wrote the study this analysis is built on.
Authored by
4 researchersIf this is your work, this is how we attribute it on Fit Body Science. Erwin Stolz is listed as the lead author.