E-Prognosis: Prognostic Models Clinicians Can Use

vssalexsmith_121218_flyerLast month, the Institute on Aging (IOA) hosted Dr. Alex Smith, Associate Professor of Medicine in the Division of Geriatrics, University of California, San Francisco, for their final Visiting Scholar Series lecture of the year.

Dr. Smith began his training in palliative medicine originally focusing on individuals, many of them being younger in age, with cancer. However, after experiencing consults with older adults, many of them with dementia or multiple chronic conditions, he recognized a disparity between his training and the need for a better understanding of how to improve care for these patients.

With this goal in mind, Dr. Smith and his team decided to focus their efforts on E-Prognosis, online prognostic tools that care providers can use in the clinic. During his lecture, and in his video interview below, Dr. Smith explains how E-Prognosis works and the influence that it has on treating older adults with chronic conditions or disabilities.


Dr. Smith says that the next steps for his team at UCSF would be to make E-Prognosis even easier for clinicians to use. In order to do so, they are working on breaking through several barriers that using certain prognostic models tend to involve. First, he explains that most models currently estimate mortality risk over a given time, i.e. “63% risk of death in four years,” however, most people don’t tend to think that way. Most think in terms of “how long do I have left to live?” which Dr. Smith and his team are now able to calculate using a Gompertz equation.

Another important consideration is to determine what the clinical implications are for clinicians who are using this model and are making decisions based on the results of the prognosis. They are working to find a way to tie the prognosis information in with common interventions that clinicians use that either have delayed time to benefit, or have some sort of prognostic requirement such as hospice or breast cancer screenings – common medical decisions that are highly dependent on how long someone might live.

As for future goals, Dr. Smith would like to figure out a way to integrate prognostic models into electronic medical records. The major challenges with this are 1.) not all electronic medical records are the same so more than one model would need to be developed, and 2.) electronic medical records unfortunately do not capture a patient’s instrumental activities of daily living (IADL), which are tremendously important predictors in an individuals prognosis.

While E-Prognosis is primarily intended for use by clinicians, it is publicly accessible here.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s