Banner
banner

Winter 2016 (Volume 26, Number 4)

Little Data

By Philip A. Baer, MDCM, FRCPC, FACR

Download PDF

“When you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely, in your thoughts advanced to the stage of science.”
- Lord Kelvin

Everyone is talking about Big Data and its potential impact on medicine and business in general. Mining terabytes of data with supercomputers could lead to research breakthroughs, a cure for cancer, extended lifespans and personalized medicine. Closer to home, the investments physicians and governments have made in electronic medical records (EMRs) are partly premised on their eventual ability to provide a source of rich information. This could be combined with other existing databases to improve clinical care and allow every patient’s clinical information to contribute to answering key unsolved questions in medicine.

The potential is staggering, as is the hype. This has been pointed out very recently in an editorial in the Canadian Medical Association Journal (CMAJ)1. Perhaps in the meantime, we can all try our hands at answering questions about our own practices using the data at hand. This would be very much in the spirit of practice self-audits, which are a key undertaking of the CRA, and are one of the best ways to obtain the newly required Royal College Section 3 credits for maintenance of certification.

Let me illustrate with a couple of examples from my own practice. I currently review all new referral requests personally. Most are accepted, some are rejected, and some result in requests for more information to determine their appropriateness and priority. What is the ideal ratio of these outcomes? No textbook will tell you. However, one can assess the possibilities with some simple analysis. I decided to count the number of spots available each week for new patients, and to compare this to the number of new referrals received per week. To obtain a reasonable sample, I collected these two data points over a four-week period not including any vacations, and then combined them into a single ratio of referrals received/referral spots available. As a thought experiment, consider a situation where 200 new referrals are received but only 10 referral appointments are available in a given period. The ratio would be 20. Stress would likely be high in such a practice. In this situation, I would suggest a strategy of accepting only those patients in greatest need of a rheumatologist (e.g., those with inflammatory arthritis, connective tissue diseases and vasculitis). Simple management suggestions and alternative referral possibilities could be provided to the primary care physicians of those patients not being accepted for consultation, along the lines of the rheumatology triage program operating in Calgary. On the other hand, if only 20 new referrals are received for 40 available spots, the ratio is 0.5. This could perhaps occur for someone newly in practice. In this situation, one might consider marketing their availability to the referring physician audience through a practice portal, providing CME lectures, or getting involved in a local medical association chapter. A ratio close to 1 would provide the least grief long-term, if achievable. It allows you to broaden your practice to whatever areas interest you outside of core inflammatory rheumatic diseases, including gout, osteoporosis, osteoarthritis and regional rheumatic disorders. Do you know your ratio? If you don’t, are you making the most informed decisions possible about how you run your practice? The information is easily available and could be calculated on a running basis by your office staff.

Similarly, how do you decide on the ideal length of appointments for both new patients and follow-ups? Do you use tradition, guesswork or data? There is no average patient, but having 10 different lengths of appointments is also not practical. Say you allot 30 minutes for new referrals and 15 minutes for follow-ups. You may also know that more complex referrals actually end up taking 45 minutes. With a referral request/referral availability ratio of 20, almost all your referrals will be complex, and there is a mismatch between your appointment slots and the time actually required. With a ratio of 1, and knowing the prevalence of inflammatory diseases in Canada, you can be virtually certain that only 40-50% of your referrals will be so complex. Assuming simpler rheumatology problems can be handled in 20 minutes, especially with pre-office review of the patient’s documentation, a 30-minute consult appointment slot now makes sense.

Follow-ups tend to be dominated by more complicated patients, as those with simpler conditions are best sent back to primary care for ongoing treatment. I like to run right on time in my office, so any day I finish late provides an opportunity for reflection. I usually find that one particularly complex patient has required extra time. If I feel this will be an ongoing issue, I then assign that patient 30 minutes for their next follow-up. Much better to feel the office is moving along as scheduled next time, than to be running out of chairs in the waiting area. If the patient is doing better on the next visit, I can always use a few extra free minutes in the day, and they can return to a 15-minute spot thereafter. With the complexity of rheumatic diseases and their therapies, as well as an increasingly older follow-up population with multiple comorbidities, the number of patients permanently requiring longer follow-up spots will only grow. Empirically, I would also venture that requiring a longer follow-up appointment correlates with higher five-year mortality, but I am certainly not going to reveal that possibility to my patients in that situation. More research is required, both of the “little data” and “Big Data” varieties.

If you would like to contribute any examples of “little data” you have found useful in your practice, please send them to us at CRAJ for possible future publication.

1. Kirsten Patrick. Harnessing big data for health. CMAJ May 17, 2016 188:555; doi:10.1503/cmaj.160410


Philip A. Baer, MDCM, FRCPC, FACR
Editor-in-chief, CRAJ
Scarborough, Ontario

Issue