I embarked on a mission several years ago to stream line our CPT medical coding process. Data is knowledge. And knowledge is power. The power to improve the process. This was no easy task. For an entire year I kept intricate details of my billing practices for every single patient I saw for an entire year. About 2500 encounters. I created a spread sheet program full of fancy formulas and incredibly useful information for our group. From what type of shift it was (long shift, short shift, night shift). What hospital, what the billing code was (the CPT code), how many daily encounters for every shift, how many new hospital admissions and consults for all the different shifts, total numbers of RVUs for all different shifts worked, numbers of shifts, numbers of procedures, daily RVUs, monthly RVUs. I developed numerous calculations. I developed my own hospitalist efficiency "efficiency ratio" (the total number of follow-up codes/total number of admission codes) where as we could evaluate the practicing characteristics of all 17 of our hospitalists. Because shift work hospitalists don't "own" a patient from start to finish, it is often hard to tell who hangs on to patients longer and who is efficient with discharges. In most hospitalist programs across the country, hospitalist salaries combine with productivity based models of care. These characteristics can become important when dissecting out those partners that see more new encounters from those that simply see more patient encounters. A form of "length of stay" equalizer.
My learning curve in this whole process was remarkable. When I first started tracking my data I found HUGE discrepancies between my data and the data of our contracted billing company. We hand code our encounters in a very simple, cheap and highly effective system. When I first started my data collecting mission, I was shocked at how many errors in the process were occurring. Many patients I had seen where not billed under my name. And many patients I didn't see were billed under my name. The major problem was discovered in the way our billing cards were created. With one very simple stroke of the brilliance key, the problem was solved. Several process changes were made in the connection between me documenting my work and the billing company filing a claim.
As a result of this endeavor we went from an error rate of several percentage points to a personally calculated error rate of approximately 0.1% in 2007 That my friends is about 2 encounters out of 2500 encounters. . A remarkable feat. And something I am quite proud of. The system we have now is remarkable in its ease, efficiency, cost effectiveness and accuracy.
In addition to revamping our billing practices, the data I have gathered in the last 3 years, along with a partner of mine, have allowed us to create a very accurate picture of work effort as it relates to the many different types of shifts we have. As a result , we have been able to calculate expected encounters AND RVU's, for our night shifts, long shifts, short shifts and swing shifts. This has helped immensely in determining fair compensation models for our part time docs and has helped everyone feel the love.
I am not aware of any published hospitalist data that has been able to break down expectations of productivity for all the different shifts in a shift model practice. The data we have gathered has been, without a doubt, highly influential in establishing fair compensation models for the flexibility we offer in our program. In essence, we have been able to develop, almost an RVU/hour or RVU/shift based on past data, and have been able to use this data to establish forward looking compensation models and expectations for new hires. We have been able to tell, through our data, when census numbers point towards additional staffing needs. I believe we have been able to anticipate the growth towards the future.
I am quite proud of what this data gathering initiative has done for our group. It has allowed each individual doc to evaluate themselves in terms of productivity and efficiency within the confines of our group average. I am not aware of any published data or hospitalist programs out there that has given a group the ability to establish such intricate details and comparisons. That can tell you exactly how many encounters per shift, or RVU/shift the group as a whole is experiencing. That can give you an "efficiency" rating of your docs.
Data is a very powerful thing. I think I'll pat myself on the back.
Allow me to pat you on the back. THis is an evolving field, and for you to get a grasp on the numbers early on is a remarkable feat.
ReplyDeleteI too, love data. I can tell you how many patients I've seen, the percentage of medicare and self pay, collection rates, etc for the past 10 years of my practice. I've discovered that managed care companies were not paying our contracted rates, before our physician organization knew. A discrepancy unearthed on one month's data pointed to a charge entry clerk who later that year was found to have been stealing. Again, no one recognized it till I found the data.
Our incomes depend on our productivity, and no one is motivated as much as we are to get it right. The best advice I can give to a young doc starting out is to track their own charges and visit volume.
Great post.
Dear Happy Hospitalist,
ReplyDeleteFirstly let me tell you that I love your website (I just recently discovered it). Very informative and educational. I have a question for you. We are a hybrid Hospitalist group. Some of us are full time hospitalists who work week on week off and every third "on" week is night shifts. Some of us have out patient practices and work as hospitalists part time. They don't cover nights. Now our employer wants us to go on productivity model. Will FTP hospitalist who do night calls be at a disadvantage as compared to the ones who don't cover night shifts and work just day shifts? Thank you.