Saturday, November 20, 2010

Desktop Medicine: Here to Stay

I want to draw readers attention to a brilliant essay in JAMA by Jason Karlawish. He describes a new concept of medical practice that he calls “desktop medicine.”

Historically, we have used “bedside medicine” to diagnose pathological diseases. Bedside medicine incorporates the standard elements of the history and physical: the chief complaint, the history, review of systems, physical exam, and diagnostic studies. Bedside medicine is focused on diagnosing the presence or absence of disease: does this patient with dyspnea have heart failure or COPD? Is this pain due to cancer or osteoarthritis? Answers to these questions inform treatment decisions (e.g. lasix or albuterol, opioids or nsaids).

Desktop medicine, in contrast, uses different tools for different ends. The focus of desktop medicine is management of risk. Clinicians gather evidence from the patient and laboratory values, and use these to generate risk estimates, often with the help of a desktop computer. These risk estimates are the basis of treatment decisions aimed at minimizing risk. The questions addressed are different: what is the 10 year risk of heart attack? What is the 10 year risk of fracture? Answers to these questions also influence treatment decisions (e.g. statins or no statins, bisphosphonates or no bisphosponates).

These concepts are not mutually exclusive. Initial diagnosis of illness is made using traditional bedside medicine techniques. But most of the care we provide in geriatrics is management of chronic conditions, minimizing the risk that these conditions will result in catastrophic disease: heart attack, stroke, hip fracture, or cancer.

As Dr. Karlawish notes, Desktop Medicine is already what we do in most of geriatrics:

Desktop medicine does not so much change medicine as explain the way it is. Educating and training physicians to practice desktop medicine is especially important for the care of elderly patients who have competing risks.
Training in desktop medicine requires a different skill set from bedside medicine. Desktop medicine relies on understanding epidemiology, statistics as they apply to individuals, and communication of risk to patients. 

How does this apply to palliative care? We increasingly rely on desktop medicine to help us decide hospice eligibility. Does this patient with dementia residing in a nursing home have less than 6 months to live if the disease takes its usual course? Let’s use a prognostic index to help (desktop medicine). Palliative medicine providers will also begin to use desktop medicine more and more to make every day decisions about treatment: Should we continue this statin? Should we anticoagulate for this deep venous thrombosis?  These decisions are based on risk of an event (e.g. heart attack, pulmonary embolism), in light of the patient’s expected prognosis and goals.

How can we do a better job at our desktops? We need better training in the clinical skills mentioned above. We also need better tools for estimating risk.

Here is a teaser: look out for the launch of a new desktop tool for GeriPal users soon!!!

by: Alex Smith

6 comments:

Anonymous said...

Librarians can help clinicians understand how to most effectively use the desktop tools that will enhance their practice. Partner with a medical librarian for training and personalized information updates to help you best use desktop medicine.

Anonymous said...

Heaven help us all if what we do most in geriatrics is desk top medicine. Why would we need physicians if that were so? . It is the combination of the art and science of medicine that makes us different as geriatricians and what makes geriatric medicine most responsive to the needs and the care of our patients as people...not a sum of their lab values nor the mere ascertainment of their risk scores based upon those numbers. There really are not enough data to give us an evidence-based way to treat our patients...especially the frail octo- and nonogenarians...please let's not be relegated to the faceless way so much medicine is practiced

Lindsey Yourman said...

To anonymous, I wholeheartedly agree with you that less time at the bedside is a bad thing, especially in Geriatrics. So how should we decide who gets screened for colon cancer, which patients with Diabetes need tight glycemic control, and who should be eligible for hospice (all decisions with life-expectancy-based guidelines)? Should those decisions be based solely on the patient's preference and the physician's clinical judgment? Is that enough to best make sure that the tests and treatments with a lag-time to benefit are more helpful than harmful to our patients?

One frustrating problem with leaving database technologies out of the equation is that objective population-based models tend to more accurately predict a patient's life expectancy than does a physician (especially when the physician knows the patient well, ironically), and physicians' predictions are greatly improved by the use of these models (Knaus et al., 1995, Annals of Internal Medicine; The SUPPORT Prognostic Model: Objective Estimates of Survival for Seriously Ill, Hospitalized Adults) . . .

Hmmm, maybe what we need is Ipad medicine!, evidence-based prediction tools to inform a physician while at the patient's bedside (applied in conjunction with their clinical judgment and a patient's preferences :)? . . . If only there were a technology to facilitate quick and easy use of these prediction optimizers without replacing bedside time with desktop time? . . .

Phone Sex said...

Pretty good post. I just stumbled upon your blog and wanted to say that I have really enjoyed reading your blog posts. Any way I'll be subscribing to your feed and I hope you post again soon.

Christian Sinclair, MD said...

A new desktop tool for Geripal, what is it, what is it?

Eric Widera said...

The new tool is still in beta testing but you can find it at eprognosis.org