Efficiency vs. Effectiveness in Healthcare

The author of a recent NY Times opinion article, Gilbert Welch, argues that we are not putting enough effort into challenging standard practices in medicine, but instead focus too much on accommodating and streamlining those practices already established, and on coming up with ever new diagnostic and screening tools, treatments and procedures to spend health care money on. This is despite prominent examples like hormonal replacement therapy after menopause or PSA screening, that should remind us to keep a certain amount of skepticism.

While arguments constantly circle around who should pay for all the blessings of modern medicine, it is often not so clear which existing or newly approved treatments are actually worth it (another example is the revoked fast track approval of bevacizumab for breast cancer). The responsibility however is still mostly on the physician, including that of dealing with insurers and insecure patients.

With clinical to-do lists and guidelines growing longer, one important distinction to make is that between efficiency (performing a given task in the most economical manner) and effectiveness (capability of producing a desired result), nicely depicted by Tim Ferriss in “The 4-Hour Workweek”:

• Doing something unimportant well [meaning: efficient] does not make it important.
• What you do is infinitely more important than how you do it.
• Being busy is a form of laziness – lazy thinking and indiscriminate action.

‘Being busy’ sounds familiar. When my father did part of his physician training in a small rural hospital in the south of Austria, about thirty-five years ago, the average inpatient hospital stay was somewhere between three to four weeks, even though diagnostic and treatment options were relatively limited. Most of the time he was able to sleep comfortably during night shifts. The introduction of pagers was averted by the young colleagues at the lower end of the command chain, by feigning bad signal and other technical difficulties with the unloved devices. For sure they had their stressful times as well, but they didn’t indulge in stress like we do today.

Thirty-five years later, I happened to work in that very same hospital for a little while, and it is still relatively relaxed in comparison. But now, with an added focus on acute geriatrics, most patients stay for an average of four days. During that time, they might have daily blood works and infusions, ECG, X-ray, CT, ultrasound, stress-echocardiography, 24h blood pressure monitoring, tilt table test, plethysmography, thyroid scintigraphy, colonoscopy and gastroscopy, diabetes management optimization, professional dietary advice and physiotherapy. The product of this whole machine is a medical report, proving that everything remotely applicable has been done in a pretty efficient way.

It would be taking it too lightly to just refer to law #13 of The House of God (“The delivery of good medical care is to do as much nothing as possible”), when it is not certain which of the growing number of choices can safely be eliminated from the menu (“A third of our health care dollars go to therapies that do not improve our health“). One attempt to get there is the establishment of the Patient-Centered Outcomes Research Institute.

But another solution might be approaching on the fast lane, an alternative to the time consuming, costly and sometimes biased process of randomized controlled trials: Data from various sources including electronic health records and online patient communities, as well as genetic data, will be merged and utilized, allowing detection of relevant correlations and prediction of outcomes. The clear advantages are much larger patient numbers and only a fraction of the time and cost of a conventional clinical trial. What are currently often exclusion criteria for clinical trials, e.g. old age or unrelated illness, could be turned into tailored selection criteria to answer more specific questions. Here is one recent example of this approach, answering whether lithium can be used to treat ALS or not.

Aside from obvious positive effects on health care costs and patient care, better data on what works and what doesn’t would presumably also help to relieve physicians from some of their tasks and responsibilities, clean up guidelines, prioritize remaining options, improve doctor- patient relationships and joint decision making, and generally make the job more meaningful and rewarding again.

About the author

Marco Treven, MD studied medicine in Vienna, Austria, and Bristol, UK. He has since been working in hospitals in Germany and Austria, and is currently a Predoctoral Fellow at the Center for Brain Research in Vienna. Passionate about neuroscience, education, public health, elegant ideas and thoughtful design. Loves outdoor sports and triathlon.