Thursday, February 08, 2007

Little ideas work, too

As a follow-up to yesterday's posting, I want to give two examples of changes instituted in one of our departments, neurology, that are indicative of ideas that bubble up and help improve access and results. Please understand, I am not saying these are necessarily state-of-the-art or haven't been done elsewhere or are world-shattering in scope. They are simply two examples of creative thinking that originated with the medical leadership that were put into place with no fuss, and that work well for patients.

First, we noticed that people who wanted to see neurologists on a semi-emergent basis, i.e., in a day or two, were finding an appointment lag of several weeks. Why? Because all the doctors were booked well in advance. So our chief of neurology instructed his faculty to leave their schedules open for a portion of each session, to be available for last-minute patient requests. The doctors were worried, though, that they would have wasted clinical time during their days, that those reserved open hours would never be filled.

What happened? Just the opposite. The reserved open hours were always filled with patients with more immediate needs. And, those patients had a much better record of actually showing up for their appointments than people who had made appointments several weeks in advance. And, doctors had a chance to see patients while they were freshly suffering from neurological symptoms, instead of hearing about those symptoms weeks after the fact. End result: Happier patients, better clinical diagnoses, and more productive doctors.

Second, every hospital has a "mortality and morbidity" conference procedure to review cases with adverse outcomes. But what about the normal, day-to-day cases? How do you audit for quality control? I don't mean questions about proper documentation. I mean review of the doctor's decision-making. Even the best of doctors will make mistakes and omissions in the course of treating a patient, most of which are not crucial, but many of which can be instructive if they are pointed out. Here, too, our chief of neurology put in place a simple idea, which he calls a "biopsy" of the medical record. Here's how it works.

Each faculty member in the department, from most junior to most senior, is asked to anonymously review the patient record of a colleague. He or she then offers a "grade" on the quality of the diagnosis and treatment, with minimal or extensive commentary depending on what he or she finds. That written review is then shared with the attending physician on the case.

What is going on here? Let's remember, first, that these doctors are extremely well intentioned and quite expert and really don't need an incentive to treat patients as well as possible. Through this gentle, non-threatening, but direct, peer review process, they are told by an equally expert colleague how they can do better. The reviewer, too, benefits by thinking through an interesting case and reflecting on his or her own practice. Since everybody gets to be a judge and part of the review team, the likelihood of defensive behavior or denial is reduced.

Not a big deal, you say? Maybe not. On the other hand, it is a thoughtful and effective process that is respectful of the expertise of the faculty while providing a gentle nudge towards more consistent clinical excellence. I like how it works, and I like what it stands for: A underlying value system of collegial behavior in service to the patients.


Ileana said...

A sismilar idea emerged in software programming a few years back: pair programming - you would put two people to code together instead of each of them writing their own code. The productivity and, eevn more important, the quality of the code increased significantly this way: less errors, two people know the code, you can think through issues with a peer, standards are used more when someone is watching... lots of good stuff. As you said: managing well in one field helps you manage well in all fields.

I always wondered if, with a bit of effort, those appointments that are made weeks or even months in advance cannot be handled differently. I learned now that you make an appointment when you are healthy and keep those follow up appointments religiously, but it took some time and frustration to do that, and at the end of the day maybe sometimes I am wasting my doctors' time this way. I am all for preventive medicine, but I would love to get an appointment with a good doctor within a week when I need one.

Anonymous said...

Those are great examples - I enjoyed reading about them.

I'm curious - have you ever read the HBR case on Virginia Mason Medical Center? If not, I recommend it for some more insightful examples.

Thanks for sharing!

Anonymous said...

Chart review, awesome.

Its being going on for a long time within software development in the form of design reviews and code reviews. It is amazing how motivated developers get to write clean, well designed code with liberal comments when they know that their peers will be pouring over the detail, looking for something insightful to say.

Some people reject it of course but the majority leap at the chance to do their best - and then learn something from the process.

I never imagined the application would be broad enough for a hospital - very cool.

Unknown said...

As an early careerist in hospital administration, I find your blog refreshing. Thank you for your openness and honesty!

It's wonderful to know your staff are empowered to bring their ideas forth, and that BIDMC administration is responsive to patient-centered suggestions!

Anonymous said...

We do something similar where I work (social science research). Every report we write gets fact-checked extensively before it goes out the door. This gives everyone an incentive to be as accurate as possible in whatever statistical analysis or data manipulation they are doing and to leave a clear paper trail of all their work so that someone else can replicate it. We've found that assigning our more junior staff the fact-checking role also gives them a chance to learn more effective coding and data analysis techniques from their more experienced peers, as well as getting them in the habit of doing high-quality and replicable work.

In addition, we do a project debrief after every major (and not-so-major) project to learn from what went well and what went wrong along the way. This has helped us to identify the circumstances under which projects go well, so that we can learn to repeat them!

Good management is time-consuming in the short run, but in the long run it's worth the investment.