The Lancet — Health Longevity
Chih-Sung Liang, MD †; Dian-Jeng Li, MD †; Fu-Chi Yang, PhD; Ping-Tao Tseng, MD; Andre F Carvalho, PhD; Brendon Stubbs, PhD; Trevor Thompson, PhD; Christoph Mueller, MD; Prof Jae Il Shin, MD; Joaquim Radua, PhD; Prof Robert Stewart, MD; Tarek K Rajji, PhD; Prof Yu-Kang Tu, PhD; Tien-Yu Chen, MD; Ta-Chuan Yeh, MD; Chia-Kuang Tsai, PhD; Chia-Ling Yu, MMeD; Chih-Chuan Pan, PhD; Che-Sheng Chu, MD;
July 20, 2021
Credit to the image: medical net
Key messages:
Evidence before this study:
- Alzheimer’s disease is the most common form of dementia.
- Alzheimer’s disease was the sixth leading cause of death in the USA and the fifth leading cause of death among US Americans aged 65 years and older in 2019.
- However, differences in mortality rates between different types of dementia remain unknow.
Findings from the study
- The prevalence of dementia, also classified as neurocognitive disorder, has been increasing, and is one of the major causes of disability and dependency among older people worldwide.
- Alzheimer’s disease is the most common type of dementia and one of the major causes of mortality worldwide.
- This results in high costs for health and social care systems, and people with dementia face substantial health challenges and might have at least twice as high a mortality risk as people without dementia.
- However, the findings from the current study suggest that non-Alzheimer’s disease dementias were associated with higher morality rates and shorter life expectancy than Alzheimer’s disease.
- Our findings suggest that Alzheimer’s disease has the most favourable survival-related outcomes compared with non-Alzheimer’s dementias, and Lewy body dementias have the highest mortality rates.
- Understanding differences in mortality between different types of dementia is important for both physicians and policy makers to develop tailored treatment and rehabilitation programmes and for patients and their families to facilitate future care planning.
- Developing tailored treatment and rehabilitation programmes for different types of dementia is important for mental health providers, patients, and their families.
Summary
Background
- People with dementia die prematurely.
- Identifying differences in mortality rates between different types of dementia might aid in the development of preventive interventions for the most vulnerable populations.
- The aim of this study was to compare the difference in mortality rates between individuals without dementia and individuals with various types of dementia.
Methods
- For this systematic review and meta-analysis, we did a systematic search of MEDLINE, PubMed, Embase, and Cochrane Library from inception to July 11, 2020, for cross-sectional or cohort studies that assessed mortality and survival-related outcomes among people with different types of dementia compared with people without dementia.
- Single-arm studies without comparison groups and autopsy studies or family studies that used a selected sample were excluded.
- The Newcastle-Ottawa Scale was used by two authors (D-JL and C-SC) independently to measure the methodological quality of included studies, and two authors (F-CY and P-TT) independently extracted data.
- We assessed differences in all-cause mortality rate and survival time from dementia diagnosis between individuals without dementia, individuals with Alzheimer’s disease, and individuals with non-Alzheimer’s disease dementias.
- The secondary outcomes were age at death and survival time from disease onset. Random-effects meta-analyses were done.
- Effect sizes included hazard ratios (HRs) and mean differences (MDs) with 95% CIs. Potential moderators, including age-associated moderators, were identified through meta-regression and subgroup analyses.
- This study is registered with PROSPERO, CRD42020198786.
Findings
- Our database search identified 11 973 records, and we included 78 eligible studies in our analyses, encompassing 63 125 individuals with dementia and 152 353 controls.
- Individuals with any type of dementia had a higher mortality rate than individuals without dementia (HR 5·90, 95% CI 3·53 to 9·86), and the HR for all-cause mortality was highest for Lewy body dementia (17·88, 5·87 to 54·46).
- After diagnosis, the mean survival time for people with Alzheimer’s disease was 5·8 years (SD 2·0).
Compared with people with Alzheimer’s disease, a diagnosis of any non-Alzheimer’s disease dementia was associated with
- a higher risk of all-cause mortality (HR 1·33, 1·21 to 1·46), a shorter survival time from diagnosis (MD −1·12 years, 95% CI −1·52 to −0·72), and
- a younger age at death (−1·76 years, −2·66 to −0·85).
- Survival time from disease onset was also shorter in people with non-Alzheimer’s dementia, across types, compared with people with Alzheimer’s disease, but the subgroup analysis revealed that this difference was only significant for vascular dementia (MD −1·27 years, −1·90 to −0·65) and dementia with Lewy bodies (MD −1·06 years, −1·68 to −0·44).
- The interactions between age and several survival-related outcomes were significant. 39 (50%) of the 78 included studies were rated as good quality, and large heterogeneity (I2>75%) was observed for most of the study outcomes.
Interpretation
- Alzheimer’s disease is the most common type of dementia and one of the major causes of mortality worldwide.
- However, the findings from the current study suggest that non-Alzheimer’s disease dementias were associated with higher morality rates and shorter life expectancy than Alzheimer’s disease.
- Developing tailored treatment and rehabilitation programmes for different types of dementia is important for mental health providers, patients, and their families.
Funding
None.
Introduction
The prevalence of dementia, also classified as neurocognitive disorder, has been increasing, and is one of the major causes of disability and dependency among older people worldwide.
This results in high costs for health and social care systems, and people with dementia face substantial health challenges and might have at least twice as high a mortality risk as people without dementia.,
Differences in mortality rates across different types of dementia might affect prevalence rates and health-care service needs.
Previous studies addressing life expectancy and survival have largely focused on Alzheimer’s disease.
- The life expectancy is about 7–10 years in individuals diagnosed with Alzheimer’s disease in their 60s and early 70s, but findings for other types of dementia have been inconsistent.
- For example, in some studies, people with vascular dementia were found to have a poorer prognosis and a shorter survival time after diagnosis than people with Alzheimer’s disease,, whereas other studies have reported opposite findings.
People with Parkinson’s disease dementia or dementia with Lewy bodies were found to have a three times higher risk of mortality compared with individuals without dementia, and people with dementia with Lewy bodies have also been shown to have poorer health outcomes and higher mortality rates compared with people with Alzheimer’s disease.
However, several studies have found little differences in mortality between different dementia types.
Research in context
Evidence before this study
- Alzheimer’s disease is the most common form of dementia.
- Alzheimer’s disease was the sixth leading cause of death in the USA and the fifth leading cause of death among US Americans aged 65 years and older in 2019.
However, differences in mortality rates between different types of dementia remain unknow.
- Previous studies addressing life expectancy and survival have largely focused on Alzheimer’s disease, but findings for other types of dementia have been inconsistent. Identifying differences in survival between different types of dementia might aid in the development of preventive interventions for these most vulnerable populations.
- To our knowledge, no meta-analyses have been done to examine mortality rates across different types of dementias compared with individuals without dementia. We searched MEDLINE, PubMed, Embase, and Cochrane Library for studies done in humans from database inception until July 11, 2020, for studies published in English that provided sufficient data on survival or mortality parameters for any types of dementia versus individuals without dementia or any non-Alzheimer’s types of dementia versus Alzheimer’s disease.
- The search terms used were: (“mortality” OR “death” OR “survival”) AND (“Alzheimer’s disease” OR “Frontotemporal dementia” OR “Frontotemporal lobar degeneration” OR “behavioral variant frontotemporal dementia” OR “dementia with Lewy body” OR “Lewy body” OR “Lewy bodies” OR “Lewy” OR “dementia with Lewy bodies” OR “LBD” OR “Parkinson’s disease dementia” OR “Parkinson’s disease dementia” OR “vascular dementia” OR “Arteriosclerotic Dementia” OR “Dementia Multi-Infarct”) AND (“normal cognition” OR “non-dementia” OR “control”). Specific database search terms are outlined in the appendix (p 170).
- We aimed to examine which type of dementia is associated with the highest mortality rate and the shortest survival.
Added value of this study
- To our knowledge, this study is the first systematic review and meta-analysis to compare mortality rates and survival outcomes between people with Alzheimer’s disease, people with non-Alzheimer’s dementias, and people without dementia based on all the available published evidence.
- We identified 78 studies encompassing 63 125 individuals with dementia and 152 353 controls. We found that people with Lewy body dementia were associated with the highest all-cause mortality rate compared with individuals without dementia.
- The mean survival time was 7·6 years from Alzheimer’s disease onset and 5·8 years from diagnosis.
- Compared with people with Alzheimer’s disease, people with non-Alzheimer’s dementias had higher mortality rates, shorter survival times from diagnosis, and younger ages at death.
- Although our meta-regression analyses identified several factors that moderated individual study findings, the change in point estimates of mortality rate was less than 10% when adjusted for potential moderators.
- Group differences between vascular dementia, Lewy body dementias, and frontotemporal lobe degeneration were not significant.
Implications of all the available evidence
- Our findings suggest that Alzheimer’s disease has the most favourable survival-related outcomes compared with non-Alzheimer’s dementias, and Lewy body dementias have the highest mortality rates.
- Understanding differences in mortality between different types of dementia is important for both physicians and policy makers to develop tailored treatment and rehabilitation programmes and for patients and their families to facilitate future care planning.
- Future studies should explore risks and potential contributing factors affecting these differences between different types of dementia.
Studies on non-Alzheimer’s types of dementia have commonly used people with Alzheimer’s disease as a reference group, and considerable uncertainty exists regarding mortality rates in non-Alzheimer’s types of dementia compared with the individuals without dementia after controlling for confounders (eg, age and co-occurring medical conditions). Thus, a systematic review and meta-analysis might provide more robust evidence to inform treatment plans and advice to those affected. To our knowledge, there has been no meta-analysis focusing on mortality rates across different types of dementias compared with the general population. The aim of this study was to compare mortality rates and other survival-related outcomes among individuals without dementia, people with Alzheimer’s disease, and people with non-Alzheimer’s dementias using a meta-analysis to synthesise all available evidence.
Methods & Data Analysis
See the original publication
Discussion
This meta-analysis compared the mortality rate and survival outcomes between individuals with Alzheimer’s disease, with non-Alzheimer’s dementias, and without dementia on the basis of all the available published evidence.
The main findings are that people living with dementia showed a 5·90 times larger HR for all-cause mortality rate compared with individuals without dementia, and the HR for all-cause mortality increased to 17·88 in people living with Lewy body dementia.
With respect to the risk posed by different types of dementia, people living with non-Alzheimer’s dementia (all types grouped together) showed a 1·33 times greater HR for all-cause mortality and a 1·12 year shorter survival after diagnosis compared with people with Alzheimer’s disease, but there were no significant differences between the vascular dementia, Lewy body dementia, and frontotemporal degeneration subgroups. In brief, although Alzheimer’s disease is the most common type of dementia and has been reported to be one of the leading causes of mortality, it has better survival outcomes than non-Alzheimer’s dementias.
To date, most studies addressing mortality risk in people with dementia focused on individuals with Alzheimer’s disease versus individuals without dementia.
Our study found that people living with Alzheimer’s disease had a 3·70 times larger HR for all-cause mortality compared with individuals without dementia, indicating that Alzheimer’s disease contributed to a shortened life expectancy.
We further found that people living with Lewy body dementia had a 17·88 times greater HR for all-cause mortality compared with individuals without dementia and a 1·45 times greater HR for all-cause mortality compared with individuals with Alzheimer’s disease.
The subtypes of Lewy body dementia (Parkinson’s disease dementia and dementia with Lewy bodies) were also associated with higher HRs for all-cause mortality against the Alzheimer’s disease reference, which strengthens the evidence of a poor prognostic profile in these neurodegenerative conditions.
A previous meta-analysis indicated that the RR for all-cause mortality was 2·2 in people with Parkinson’s disease versus people without dementia, and the subgroup analysis showed that people with Parkinson’s disease dementia had a particularly high risk of mortality compared to people without dementia (RR 3·78, 95% CI 2·06–6·92).
A population-based cohort study suggested that part of the increased mortality risk in patients with Parkinson’s disease can be ascribed to their increased risk of developing dementia.
Yet another previous study showed that the survival advantage of Alzheimer’s disease over dementia with Lewy bodies persisted after adjusting for age at onset, gender, comorbidity, and cognitive function.
It has been reported that people with Alzheimer’s disease had a better prognosis than people with dementia with Lewy bodies. Moreover, psychosis is more common in people with dementia with Lewy bodies than in people with Alzheimer’s disease, resulting in increased mortality risk.
Compared with people with Alzheimer’s disease, people with dementia with Lewy bodies are reported to have an accelerated cognitive decline, more comorbid conditions, greater health-care service use, and poorer quality of life, which leads to a higher mortality rate. Consequently, Lewy body dementia (including Parkinson’s disease dementia and dementia with Lewy bodies) was associated with a higher mortality rate and greater reduction in life expectancy compared with Alzheimer’s disease.
We found that compared with people with Alzheimer’s disease, people living with vascular dementia had a 1·26 times larger HR for all-cause mortality and a 1·33 year shorter survival time after diagnosis, whereas there were no significant differences in age at death.
An increased occurrence of vascular risk factors and higher rates of circulatory-associated death have been implicated in the increased mortality risk and reduced life-span survival time in vascular dementia compared with Alzheimer’s disease.
The high frequency of mortality from circulatory system diseases in vascular dementia might reflect that vascular dementia is part of a general cardiovascular disease.
In our study, frontotemporal degeneration was associated with a reduced life expectancy but not with an increased mortality rate compared with Alzheimer’s disease, although a higher mortality rate was observed when compared with individuals without dementia.
A study published in 2021 reported that motor symptoms were associated with reduced survival in patients with frontotemporal degeneration, including parkinsonism, dystonia, and apraxia. Besides, rapid eating and dysphagia are common in patients with frontotemporal degeneration, and these symptoms might increase the risk of choking, aspiration pneumonia, and mortality. Importantly, people with frontotemporal degeneration might have a younger age at onset and diagnosis than people with Alzheimer’s disease, and thereby the mortality rate after diagnosis might be attenuated during the follow-up period.
For example, Gerritsen and colleagues studied people with young-onset dementia who experienced their first symptoms before the age of 65 years and reported a longer survival time for people with frontotemporal degeneration than for people with Alzheimer’s disease.
To date, studies investigating the survival outcomes of frontotemporal degeneration and its subtypes are scarce. The comparison of mortality rate between frontotemporal degeneration is debated.
Thus, for survival time from diagnosis, age at diagnosis might be a moderator; for survival time from onset, age at onset might be the moderator.
Our study indeed found significant interactions between age and several survival-related outcomes.
However, the interaction effects were only observed for particular types of non-Alzheimer’s dementias compared with Alzheimer’s disease. Moreover, most of the adjusted effect sizes on these outcomes had less than 10% of change-in-estimates. Importantly, the findings of mortality rate (HR and RR) were robust for all non-Alzheimer’s dementias versus Alzheimer’s disease, without significant age-related moderator effects.
Furthermore, we did not find any small-study effects for the outcomes of survival time from onset and diagnosis.
However, the interactions between age and mortality were significant for Lewy body dementia and its subtypes; the adjusted effect sizes for age at death had significant changes in point estimate when adjusted for age at onset or diagnosis (range 0·9–30·5%).
Therefore, age at onset and diagnosis might play an important role in the difference in survival time between Lewy body dementia (or its subtypes) and Alzheimer’s disease.
This study has several limitations.
First, the sample size and the number of eligible studies for some subtypes of Lewy body dementia and frontotemporal degeneration were limited; therefore, we could not detect a difference though a significant difference might have existed for some comparisons.
Second, heterogeneity was high in most analyses. We addressed this issue by using random-effects meta-analysis models as well as meta-regression and subgroup analyses. To further reduce heterogeneity, we did not include single-arm studies, and all the effect sizes were calculated against individuals without dementia or Alzheimer’s disease reference groups. We also estimated the adjusted effect sizes and the proportion of change in point estimates for the significant moderators.
Third, the data on age at onset and survival time from disease onset might be subject to recall bias.
Fourth, we examined the mortality rate and survival time at a single point of clinical diagnosis, and some cases of dementia might be underdiagnosed or diagnosed late. A single cutoff of mortality on a particular date might lose information about when patients die over time. Local variation in practice groups (in terms of diagnosis) might add uncertainty to our estimates. Further studies could pool the prevalence and incidence data or infer the survival differences between Alzheimer’s disease and non-Alzheimer’s dementia.
Fifth, in clinical studies, the diagnosis of dementia was based on clinical assessment, which lacks specificity. Patients with dementia might have co-pathologies of Alzheimer’s disease, Lewy bodies, or vascular lesions. Finally, we only included peer-reviewed studies published in English. Therefore, our analyses did not include grey literature (eg, government reports) that might report vital statistics on dementia mortality.
In this systematic review and meta-analysis, we comprehensively compared non-Alzheimer’s dementias with Alzheimer’s disease and with individuals without dementia.
Non-Alzheimer’s dementias were associated with higher mortality rates and shorter life expectancy than Alzheimer’s disease. Most of all, Alzheimer’s disease appeared to have the most favourable survival-related outcomes, and Lewy body dementia appeared to have the highest mortality rates.
Higher mortality rates might also imply a higher likelihood of morbidity and disability. Discovering potential sources of divergence in mortality risks for distinct dementia types is important both for physicians and policy makers to develop tailored treatment and rehabilitation programmes for different types of dementia, and for patients and their families to facilitate future care planning.
Further epidemiological research is warranted to investigate the specific risk factors of early mortality at different levels of morbidity across different types of dementia.
Contributions
C-SL and C-SC led the conception and design of the study. T-YC and T-CY led the data collection and quality assessment. C-SL, D-JL, C-CP, and C-SC did the statistical analysis, interpreted the data, and wrote and revised the Article. F-CY, P-TT, AFC, BS, TT, CM, JIS, JR, RS, TKR, Y-KT, C-KT, and C-LY contributed to study design, assisted in data interpretation, and revised the Article. All authors revised and approved the final version of the Article. All authors had full access to all the data reported in the study. C-SL and C-SC accessed and verified the data. All authors had final responsibility for the decision to submit for publication.
Data sharing
Our study is based on published data, and all data are retrieved from original papers. Therefore, there are no primary data to be shared. The data supporting the findings of our study are available within this Article and the appendix. The statistical plan and code for analyses are available on request from the corresponding author without any access criteria.
Declaration of interests
We declare no competing interests.
Acknowledgments
This work was not funded. BS is supported by a Clinical Lectureship ( ICA-CL-2017–03–001 ) jointly funded by Health Education England and the UK National Institute for Health Research (NIHR). BS is part-funded by the NIHR Biomedical Research Centre at South London and Maudsley NHS Foundation Trust. BS also holds active grants with the Medical Research Council (and the GCRF global multimorbidity seed-funding call) and Guy’s & St Thomas’ Charity (GSTT). The views expressed are those of the author(s) and not necessarily those of the (partner organisation), the NHS, NIHR, Department of Health and Social Care, MRC, or GSTT. RS is part-funded by the NIHR Biomedical Research Centre at the South London and Maudsley NHS Foundation Trust and King’s College London; and the NIHR Applied Research Collaboration South London (NIHR South London) at King’s College Hospital NHS Foundation Trust. TKR has received research support from Brain Canada, Brain and Behavior Research Foundation, BrightFocus Foundation, Canada Foundation for Innovation, Canada Research Chair, Canadian Institutes of Health Research, Centre for Aging and Brain Health Innovation, National Institutes of Health, Ontario Ministry of Health and Long-Term Care, Ontario Ministry of Research and Innovation, and the Weston Brain Institute. TKR also received in-kind equipment support for an investigator-initiated study from Magstim, and in-kind research accounts from Scientific Brain Training Pro.
Originally published at https://www.thelancet.com.
PDF version
chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/viewer.html?pdfurl=https%3A%2F%2Fwww.thelancet.com%2Fcms%2F10.1016%2FS2666-7568(21)00140-9%2Fattachment%2Fb126da12-ba95-4573-a8b0-cca48db76db8%2Fmmc1.pdf&clen=19867022
