Age and Mental Disability Predict Survival with Dementia


By John Gever
CAMBRIDGE, England, Jan. 10 -- After the onset of dementia of patients in their late 60s, their median survival is nearly 11 years, researchers here have calculated.
Point out that this study found other factors, such as educational attainment, affluence, gender, and living and marital situation, significantly predicted survival on univariate analysis but failed to show significance under multivariate analysis.
The median survival after dementia onset ranged from 10.7 years (interquartile range, 5.6 to not available) for those 65 to 69, to 3.8 years (interquartile range, 2.3 to 5.2) for those in their 90s, Carol Brayne, M.D., M.Sc., of the University of Cambridge, and colleagues reported online in BMJ.
Median survival for those developing dementia in their 80s was 4.3 years (interquartile range, 2.8 to 7.0), she reported, and 5.4 years (interquartile range, 3.4 to 8.3) for patients with onset in their 70s.
Functional mental disability, as measured by the Blessed dementia scale, also markedly affected survival. Median survival among participants classified as most impaired was 3.3 years (interquartile range, 2.2 to 5.3) from dementia onset versus 6.4 years (interquartile range, 4.0 to 10.4) for the least impaired.
Dr. Brayne and colleagues acknowledged that, qualitatively, the findings would not surprise many clinicians. Rather, they suggested the study's value was in putting numbers to the expected survival after the onset of dementia as "a useful measure for individuals, families, and society."
They pointed out that "some of these results may seem self evident, but they answer questions asked by those caring for and advising people with dementia."
The Cambridge group analyzed health records of 438 patients who developed dementia during the first years of a larger prospective study of cognitive function and aging that began in 1991. Following initial interviews and exams, 14 years of follow-up data on mortality were available. As of the current analysis, 81% of the 438 individuals had died.
Compared with people 65 to 69 years old, those in their 90s had a univariate hazard ratio of death of 6.4 (95% CI: 2.7 to 15.0) during a 14-year follow-up period.
In the 80-to-89 age range, the hazard ratio was 4.5 (95% CI: 2.0 to 10.2). For those in their 70s, it was 3.3 (95% CI: 1.4 to 7.5).
In addition to age and disability, univariate analysis revealed significant hazard ratios for death for these factors:
Living in an elderly-care facility versus private residence: 1.6 (95% CI: 1.3 to 2.0, P<0.001).
Marriage or cohabitation versus widowed: 0.7 (95% CI: 0.7, P=0.005).
Scores of 22 to 25 on the Mini-Mental State Examination versus scores of 17 or below: 0.6 (95% CI: 0.4 to 0.8, P<0.001).
But these survival differences became insignificant under multivariate analysis in which all factors were considered simultaneously. Only age and Blessed dementia scale scores remained strongly predictive of survival in multivariate analysis.
Educational attainment, social class, self-reported health, and affluence had little influence on survival in the study.
Dr. Brayne and colleagues wrote, "Clinicians and those providing care for patients with dementia are often asked to provide a sense of how long the patient might survive. There is often a delay in recognition of dementia and formal diagnosis, which clinicians need to consider when using the best estimate in each clinical case. Knowing which factors influence the length of survival is also important."
In an accompanying editorial, Murna Downs, Ph.D., of the University of Bradford in England, and Barbara Bowers, Ph.D., R.N., of the University of Wisconsin in Madison, commented that the study provided "clear evidence that people with dementia need coordinated care and support from a range of professionals and practitioners from diagnosis to death to ensure maximum quality of life and prevent unnecessary disability and suffering."
They noted that well organized care has been shown to reduce disability, and dementia is no longer a "living death" for which custodial care is the only option.
"In planning care and support, doctors need to pay as much attention to the essential human worth of a person with dementia and their retained capacity for relationships, pleasure, communication, and coping as they do to deficits and dysfunction," Drs. Downs and Bowers wrote. "They also need to be aware of the growing evidence base for therapeutic intervention and effective support to minimize disability and promote optimal quality of life."
Limitations in the Cambridge study included the exclusion of some participants who may have developed dementia but died before a diagnosis could be made, causing the study to overestimate survival times. Study dropouts, who were also excluded from the analysis, tended to have impaired cognition and higher mortality, the researchers said, which may have skewed the findings as well.
The Cambridge study was funded by the U.K. Medical Research Council and Department of Health.
No financial conflicts of interest were reported by study investigators or the editorialists.
Additional source: BMJSource reference: Xie J, et al "Survival times in people with dementia: analysis from population based cohort study with 14 year follow-up" BMJ 2008; DOI: 10.1136/bmj.39433.616678.25. Additional source: BMJSource reference: Downs M, et al "Caring for people with dementia" BMJ 2008; DOI: 10.1136/bmj.39433.616678.25.