A UC San Francisco team has developed a tool that can help determine -- and perhaps influence -- senior citizens' 10-year survivability rates.
The simple checklist helps doctors assess health risks that influence the longevity of older adults, and according to the authors, could be an opportunity for seniors to really engage with their primary care provider in having informed discussions about their health care maintenance.
The UCSF team created a 12-item "mortality index" based on data of more than 20,000 adults over the age of 50 from 1998 until 2008, from the Health and Retirement Study (HRS), a nationally-representative sample of independently living U.S. adults. The point system was based on their risk factors and survival rate at the end of 10 years.
Their findings will be published March 5, in the Journal of the American Medical Association (JAMA).
Calculating medical risk can be an inexact science, especially for older adults. Many factors from environmental to chronic diseases can help determine how long a person lives.
"The most important thing we found was the risk factors that go into estimating shorter intermediate survival are very similar to risk factors that go into estimating the likelihood of longer-term survival," said first author Marisa Cruz, MD, a clinical fellow with the UCSF School of Medicine. "We also found that building a tool that clinicians can use to estimate that likelihood of longer-term survival requires considering many different types of risk factors.
"Not one particular risk factor tells you whether or not you are likely to survive but a host of attributes about your life and your medical conditions will give you a clearer picture," she said.
Points for Risk Factors
The clinical tool operates on a point system, and the total determines a patient's 10-year risk of mortality. For example, age, gender and medical conditions were given specific points. Adults between the ages of 60 and 64 received one point, for example, compared to those over the age of 85 who received seven points. Health risks such as current tobacco use, non-skin cancers, chronic lung disease and heart failure each were assigned two points.
"It's an easily usable simple model with only 12 factors. The clinician can ask the patient 'yes' or 'no' questions about his or her health and functional status and then can then go over how the patient could benefit from certain medical interventions," Cruz said.
The ability to complete cognitive or motor skills such as managing one's finances or walking several blocks also was factored into the equation. The difficulty in performing each aptitude generated one to two points.
"The goal of the study is to build a clinical tool that can be used in any setting," Cruz said. "So you want to have these kinds of models to be able to be generally applicable to a primary care provider no matter where they are."
This current research builds on a study that senior author Sei Lee, MD, MAS, assistant professor with the UCSF Division of Geriatrics, did in 2006. He and his team looked at factors that were most likely to influence intermediate term life expectancy.
"We looked at things like what factors are most predictive of someone's likelihood of surviving for four years," he said. "The new study takes that research even further by helping doctors be able to counsel patients on what clinical interventions are most likely to benefit them."
A U.S. National Sample
Researchers could not determine any difference in outcome based on geography or ethnicity. Theoretically, someone from Louisiana would have the same outcome as someone from Maine with similar risk factors.
"The goal of health care in general is to improve people's lives but I think that this project and this model in particular are meant to be able to make sure that people live the highest quality life they can," Cruz said. "Any interventions that we do in the name of promoting health must yield more benefit than harm."
"There are national guidelines that most provider s are very familiar with like getting a colonoscopy every 10 years or getting a mammogram every two years. I think those are excellent and well-founded and well-studied guidelines," Cruz said. "But I do think that there are some variability in tailoring that to your particular situation, so there's benefit to engaging your primary care provider and asking, 'Does this make sense for me?'"
Cruz is the first author of the paper; Lee is the senior author; co-authors include Kenneth Covinsky, MD, MPH; Eric W. Widera, MD; and Irena Stijacic-Cenzer of the UCSF Division of Geriatrics.
This study was supported by the American Federation for Aging Research and the National Institute on Aging through the Beeson Career Development Award (K23AG040779) and a grant (K24AG029812).
- Marisa Cruz, Kenneth Covinsky, Eric W. Widera,Irena Stijacic-Cenzer, Sei J. Lee. Predicting 10-Year Mortality for Older Adults. JAMA, 2013 DOI: 10.1001/jama.2013.1184
Cite This Page: