Outpatient Practice

Why Do People In Ohio Smoke So Much?

This month, the CDC released the preliminary results of the National Health Interview Survey for 2017. The great news is that the percentage of American adults who smoke dropped nearly 2 percentage points from last year, from 15.8% to 13.9%. This is the lowest number in more than 50 years. But in Ohio, the percentage of people who smoke did not change; in fact, it is going up.

Smoking has notable demographics. Men are more likely to smoke than women (15.8% versus 12.2%). White Americans are more likely to smoke (15.7%) than Black Americans (15.1%) or Hispanic Americans (9.7%). People living in a rural area are more likely to smoke (21.5%) than people living in small cities (15.6%) or large cities (11.4%).

On January 11, 1964, Dr. Luther Terry, the Surgeon General of the United States, released a report concluding that smoking causes lung cancer and chronic bronchitis. That year, 42% of Americans smoked and the average per capital U.S. cigarette consumption was 4,300 per year (another way of stating this is the average American smoked 215 packs of cigarettes per year). The number of cigarettes smoked immediately began to fall. The science behind this report was solid but there were a lot of “smoking-deniers” who refused to believe it and so in the mid 1970’s, the per capita cigarette consumption in the U.S. rose. Over time, most of the deniers either changed their mind or died so that now, 54 years later, only a few Americans doubt the relationship and the per capita cigarette consumption has been steadily falling every since the mid-1970’s.

Over the past 20 years, the percentage of Americans who smoke has fallen an average of 0.5% per year. At that rate, in 26 years, no Americans will smoke. Therefore, in 26 years, three jobs that you don’t want to have are: tobacco farmer, cigarette maker, and pulmonologist. Alas, my medical specialty is going to go the way of the blacksmith, slide rule maker, and telephone operator.

From a health economics standpoint, cigarettes are actually a very effective way of government cost-control. The average woman who smokes a pack a day will live 11 years less than a non-smoking woman and the average man who smokes lives 12 years less than a non-smoker. If you do the math, it works out to losing 14 minutes of life for every cigarette a person smokes. Since smokers die so much earlier than non-smokers, they consume a lot less Social Security benefits. With more Americans not smoking, we’ll see more people living long enough to be on Medicare and Social Security.

But Ohio is different. From a report by the United Health Foundation, for the past three years, the prevalence of smoking has gone up rather than down and in 2016 (the most recent data year available for the state), 22.5% of Ohioans smoked. So then, why is the prevalence of smoking going up in Ohio whereas it is going down in the rest of the country? If you examine the data carefully, then demographically, Ohio smokers are different than the rest of the country:

  1. Age. Similar to the rest of the country, the older you are in Ohio, the less likely you are to smoke. However, young Ohioans (age 18 – 44) are far more likely to smoke (26.4%) than in the rest of the country (18.1%).
  2. Race. Nationwide, Hispanic Americans are the least likely racial/ethnic group to smoke but 22.3% of Hispanic Ohioans smoke, double the national average. Ohio has the dubious distinction of being one of the most equal opportunity states for smokers with about the same percentage of white, black, and hispanic Ohioans smoking.
  3. Education. For many years, there has been a significant inverse relationship between education and the prevalence of smoking – the more education a person has, the less likely they are to smoke. But in Ohio, for those people with less than a college degree, these differences are amplified. Thus, if a person doesn’t finish high school in Ohio, that person is far more likely to smoke than a high school dropout in the rest of the country.
  4. Income. Not surprisingly, the relationship of smoking with income tracks that of education. But once again, in Ohio, the effects of income are amplified.
  5. Rural versus urban. Nationwide, Americans living in rural areas are less likely to smoke than those in large cities. But Ohio is different. People living in Ohio’s large cities have the same prevalence of cigarette smoking as those who live in rural Ohio.

So what does all this mean about Ohio? For starters, 25 years from now, Ohio will be one of the last hold-outs for smokers in the United States. If you live in a large city, have lower income, or have a lower education level in Ohio, then you are more likely to smoke than in the rest of the country. With the average price of cigarettes at about $6.50 in Ohio, the Ohioan who smokes spends about $2,300/year on cigarettes – for the 39% of Ohioans earning less than $25,000/year who smoke, than means that they are spending 1/10th or more of their entire annual income on cigarettes. Many in Ohio complain about the loss of good-paying jobs over the past 15 years. Part of our problem in Ohio is that because we spend so much more money on cigarettes than the rest of the country, once you factor in the cost of those cigarettes, our take-home income is less for the same job as the rest of the country. Clearly, we have a lot of work still to do in Ohio when it comes to tobacco education and control. Our high rate of tobacco smoking puts Ohio at a competitive disadvantage to other states

As a pulmonologist, this unfortunately means that I will still have a job in Ohio when I am well into my 80’s, long after my peer pulmonologists in the rest of the country are out of work.

June 28, 2018

Outpatient Practice

Should Men Be Screened For Prostate Cancer With PSA?

Identification of cancers in early stages when they are surgically curable has resulted in an overall reduction in cancer death rates in the past 20 years. Where cancer screening has been most effective is in colorectal cancer, cervical cancer, and breast cancer. Prostate cancer screening has been much more controversial with several recent studies suggesting that screening with serum PSA levels may not have much impact on overall survival. The most recent study showing no benefit by PSA screening was published in this month’s JAMA. This was a powerfully large study that involved 415,357 men age 50 – 69 years old followed for 10 years. The overall conclusion of the study was:

“…there was no significant difference in prostate cancer mortality after a median follow-up of 10 years but the detection of low-risk prostate cancer cases increased. The findings do not support single PSA testing for population-based screening.”

In other words, they found more cancers but finding them did not result in men living any longer. The idea is that by screening, you find very low grade cancers that would never pose a risk to a man with a normal life expectancy. And the implication is that by screening, we’d do a lot of unnecessary prostatectomies on men who really don’t need them. So, does this mean that we should stop screening men for prostate cancer? Well, we shouldn’t condemn the PSA too fast.

First, the study took place in England and Wales – notable because one of the major risk factors for prostate cancer is being African American (at least in the United States). In the U.S., 12.3% of the population is black whereas in the U.K., only 3.0% of the population is black. Therefore a significant at-risk population was underrepresented in the study. It is likely that different screening strategies should be employed for higher risk men, such as those who are black and those with a family history of prostate cancer. This would be analogous to doing mammography at an earlier age for women with BRCA genes and strong family history of breast cancer or doing more frequent colonoscopies in people with familial polyposis.

Second, the study protocol was a single PSA test with prostate biopsy for those men with a PSA > 3.0 ng/ml. I would question the value of a single PSA test, especially if it is only minimally elevated, for example 3.0 – 5.0 ng/ml. Instead, it may be more predictive to look at the PSA trajectory over time with men at the highest risk of more aggressive prostate cancer having the steepest trajectory in PSA levels.

Third, if you look carefully at the data from the recent study, although there was no overall significant difference in the mortality from prostate cancer between the men who got usual care (control) versus those that got the single screening PSA (intervention), by 14 years, the curves were starting to separate and I question whether there would be a significant difference in mortality if the groups had been followed for 18 or 20 years. The implication of this is that for men with a long life expectancy, PSA screening may be appropriate.

So, what is my personal take on whether we should be doing PSA screening for prostate cancer? I believe that we should do it but that screening should be personalized based on life expectancy and risk factors. Here are my thoughts on it:

  1. Be sure that the PSA test is accurate. There are many things other than prostate cancer that can cause a moderately elevated PSA level, such as age, prostatitis, urinary tract infections, and benign prostatic hypertrophy (BPH). But there are also other conditions that can cause mildly elevated PSA levels and these need to be taken into consideration when considering the timing of the blood draw for the PSA test: constipation, riding bicycles (particularly those with the sharp-angled, pointed seats), recent ejaculation, recent foley catheter placement or urinary tract surgery, and digital rectal exam. Because the half-life of PSA in the blood is 2-3 days, we should caution men to avoid these things for 3-4 days before getting their PSA level drawn so that the the risk fo false positive tests are minimized.
  2. Screen only those men who are otherwise likely to live at least 12 years. There is no need to get a PSA test on a patient who has advanced COPD, who has severe heart failure, or who is 85 years old. On the other hand, it may be reasonable to screen a very healthy 60-year old man with a family history of everyone living into their 90’s.
  3. Men with risk factors, such as having family histories of prostate cancer or being black, should be managed differently, with a lower threshold for doing PSA screening.
  4. Follow the PSA trajectory. Every child has their height plotted on a growth curve as they get older – we should do the same with PSA and then those men whose curve deviates from the expected increase in PSA with age should be the ones that we target for doing more detailed cancer testing such as prostate biopsy. A single PSA is just one point in time and what may be more important is the change in PSA over time.

I’m not ready to give up on recommending prostate cancer screening on the basis of the most recent study. But I am ready to commit to screening more strategically and making sure that my patients are better informed about the test.

March 22, 2018

Outpatient Practice

Making Your Outpatient Practice More Efficient

It is hard to make a huge salary with a purely outpatient medical practice, in fact, it is hard to just break even without support from a hospital or some other institution. American medical financing is structured in order to preferentially reward physicians who do procedures, who have high-volume inpatient practices, or who are subsidized by a hospital. But there are some tangible things that the outpatient physician can do to improve the operational efficiency of an outpatient medical practice to optimize revenue and improve income.

Why you can’t make a living off of wRVUs alone.

First, you have to understand the current medium of exchange in physician financing, the RVU (relative value unit). Every service and procedure in medicine is assigned a number of RVUs based on the average time it takes to perform that service or procedure, along with the complexity and degree of training required to do it. Each total RVU is composed of 3 subcomponents: the work RVU (that represents physician effort), the expense RVU (that represents the typical overhead expense to provide that service or procedure), and the malpractice RVU (that represents the average expense to cover the malpractice premium to perform that service or procedure). Medicare pays physician based on the number of RVUs that they bill – currently about $36 per RVU. Medicaid typically pays less – around $22 per RVU and commercial insurers pay more (around $40 – $60 per RVU, depending on individual physician contracts with specific commercial insurance companies). Uninsured patients typically pay little to nothing per RVU. We often focus on the work RVU (wRVU) as a measure of physician productivity. Therefore, in an outpatient medical practice with an average overhead expense, the amount of money that the physician takes home should equate to the number of wRVUs that he or she generates per year times the average revenue per RVU generated by Medicare, Medicaid, and all of the various commercial insurance companies that cover the patients in that medical practice.

So, in theory, if you generate 1,000 wRVUs and you get paid an average of $36/wRVU, then your income should be $36,000. But reality doesn’t work that way. The 2017 Medscape salary survey indicates that the average pulmonologist makes $310,000 per year. If you add to that 25% benefits, then you need $388,000 to cover a year’s worth of salary and benefits. To generate this much money, you’d need to generate 10,764 wRVUs per year. However, the average pulmonary/critical care physician only generates 5,768 wRVUs per year. The MGMA does a deeper analysis of compensation per wRVU and found that the average pulmonary/critical care physician brings in $58 per wRVU.

So, how do you improve your compensation per wRVU?

The simplest way to get to the $58/wRVU number is to restrict your practice by not taking uninsured patients and Medicaid patients and putting a limit on the number of Medicare patients that you see in order to load up on the higher paying commercially insured patients. There are a lot of physicians who do just this. But most physicians can’t or won’t limit their practices to patients with commercial insurance. However, there are some things that the physician can do to improve their average compensation per wRVU:

  1. Negotiate a higher compensation per wRVU. If you are a solo practitioner, you have zero negotiating power with commercial insurance companies but if you are in a large group practice or affiliated with a hospital with a lot of financial clout, then you can negotiate for higher rates then the “standard fee schedule” that the insurance company offers. Also, if you are affiliated with a hospital, then you can often negotiate a higher payment per wRVU from the hospital with the assumption that the hospital will make enough in downstream revenue from your patients that they can afford to subsidize your practice.
  2. Buy your own testing equipment. If you have sufficient procedural volume to justify purchasing an EKG machine, a point-of-care lab testing machine, or a spirometer, then you can add to your overall revenue. As an example, if you purchase a spirometer for $1,500 and you get paid $34/test (Medicare rate), then you need to do 44 tests to cover your initial costs and every test over that is profit.
  3. Negotiate coverage of hospital activities. Hospitals are required to have a physician available for things like cardiac rehabilitation, anticoagulation clinics, and hyperbaric oxygen treatments. Most of the time, these require very little physician work except on the rare occasions that a patient has a medical problem while participating in these programs. So, negotiate for an hourly coverage fee – you can still be seeing patients in the office at the same time as long as the office is in the same building as the hospital-based program.
  4. Reduce your clinic expenses by improving your operational efficiency.

Improving operational efficiency in the outpatient clinic

Medicare doesn’t pay you by the wRVU, it pays you by the total RVU. So, if you can reduce the amount of money that you have to spend on overhead expense, you increase the amount of money that goes to the physician’s salary. Lets take a hypothetical service with a total RVU of 2.22. Of that total RVU, there is 1.00 wRVU, 1.00 peRVU (practice expense RVU), and 0.22 mRVU (malpractice RVU). At $36/RVU Medicare rates, the assumption is that you will need $36 to cover the overhead expense of providing that service and there will be $36 toward the physician salary for that service. But, if you can reduce your expense to $18 for that service, then you will have $54 to go toward the physician’s salary, thus converting $36/wRVU into $54/wRVU.

Optimizing clinic efficiency can reduce the overhead expense for the medical services you provide. Most of the overhead expenses that a medical practice has on any given day that the practice is open are fixed expenses: the cost of rent, the cost of the nurses and registration staff, and the cost of your billing staff. If the physician had to do everything on their own (registration, scheduling, checkout, billing, answering the telephone, vital signs, etc.), then the physician would probably only be able to see a very small number of patients per day. Strategically, the physician should just do the components of medical practice that only a physician can uniquely do and the practice should employ other people to do the more routine things. By hiring office staff and nursing staff, you can improve the number of patients seen per day, up to a point. The trick is to hire just enough staff to match the physician’s peak productivity – too few staff, and the physician either has to do non-physician tasks (reducing productivity) or cut back the number of patients seen and too many staff, and the physician can’t generate enough work to keep all of the staff busy. According to the MGMA, the average number of staff in a physician-owned medical practice is 1.24 business operations support staff, 1.48 front office support staff, 1.82 clinical support staff, and 0.78 ancillary support staff per physician in a multi specialty practice. In a hospital-based clinic, the numbers of staff per physician are generally lower. It is a lot like the Starling curve that defines cardiac output and preload.

As an example, lets take a look at what happens if you change from scheduling return visits at 20 minutes per visit to scheduling them at 15 minutes per visit. Lets assume that you bill all of your return visits as level III returns and get paid Medicare rates of $36 per RVU. And lets say that your overhead expense for your practice is $272,000 per year. By simply dropping from 20 minutes per visit to 15 minutes per visit, you can double the physician income:

  • 3 patients/hour increases 33% to 4 patients/hour
  • 24 patients/day increases 33% to 32 patients/day
  • Net revenue $408,000 increases 33% to $544,000
  • Physician income $136,000 increases 100% to $272,000

This is because if the overhead expense is fixed, everything that the physician generates above the overhead expense is profit to the physician. In reality, overhead is made of of both fixed and variable expenses, particularly if things like billing are outsourced where billing expenses are usually charge based on the volume of bills rather than a fixed annual amount. Nevertheless, you can see that small changes to clinic visit times can have a huge impact on physician salary revenue.

Other things that can reduce expenses and improve clinic efficiency:

  1. Don’t hire an registered nurse to do a medical assistant’s job. RNs are expensive and MAs salaries are lower. Make sure that you are not over loading the clinic staff with people who are over qualified for typical duties. You don’t need an RN to do vital signs and schedule tests.
  2. Work the staff at the top of their license. This is an extension of #1 above. The physician should delegate those tasks such as maintaining the past medical history and calling in prescriptions rather than trying to do everything him/herself.
  3. Avoid canceling clinics. Even though the physician and the patients aren’t there, the office staff are still there and you still have to pay rent. In large practices or hospital practices, rather than having all of the staff sit around idle, a different physician should be seeing patients if the normally scheduled physician is on vacation.
  4. Convert phone calls into paid office visits. On a typical Monday, I’ll have around 15 – 20 patient phone calls and email message via our electronic medical record. I prescribe a lot of medications over the phone and give out a lot of advice over the phone. And all of it is free since phone calls and emails are not reimbursable. By saving 15 – 30 minutes at the end of each day for sick call visits, you can convert a lot of that free medical care that you provide into paid medical care that you provide.
  5. Use the entire office hour time. I know some physicians who don’t start seeing patients until 9:00 even though the office staff start at 7:30. That is a lot of idle time. Ideally, the physician should be present and seeing patients during the entire time that the clinic is open and then reserve meetings, phone calls, etc. to a time of day that you are not paying your office staff to be present.
  6. Spend 15 minutes prepping for the day’s patients. By scanning the patients’ charts you can identify things for your office staff to do in order to improve the efficiency of the clinic visit, whether that is locating test results that aren’t in the chart or pre-ordering an office EKG or spirometry that you know that you are going to need. The goal is to have the patient spend only the amount of time needed for the office visit in an exam room and to avoid having the patient waiting in the exam room while the physician or nurse tracks down the results of the CT scan that the patient is there to discuss with the physician.
  7. Do a pre-clinic huddle with the clinic staff. This follows from #6 above. If the staff know what to expect with the patients scheduled for that day, they can prepare and optimize flow.
  8. Double book strategically. As a rule, I don’t like to double book – if both patients show up, then both of them only get half of the time that they deserve or I end up putting the rest of the day’s schedule behind. However, if you can anticipate which patients are less likely to show up, then double booking those patients can average out to a normal schedule in the long run. In my practice, new patients referred from the emergency department or referred as hospital follow-ups frequently don’t show. Also, return patients with a history of not showing in the past are likely to not show in the future. Demographically, in many practices, uninsured patients and Medicaid patients are more likely to not show.
  9. Schedule patients strategically. In every medical practice, there are some patients that you know are going to take longer than others. For example, in my practice, a pulmonary nodule follow up or an annual asthma check up take less time than a return visit for a patient with multiple medical problems who is close to entering hospice. Also, there are some patients that always seem to need more time – maybe because they have lower health literacy, because they require a translator, or because they just need to talk more. Schedule these patients at the end of the day.
  10. Organize patient flow efficiently. You need to really look at the patient movement through the clinic: where the physician charting area is located, where the nursing charting area is located, where registration and check out are located. Often, patient flow is determined more by where the rooms were located when you first rented your office space – by making small adaptations, you can often greatly improve efficient flow.
  11. Make your electronic medical record work for you. I’ve posted a lot about EMRs but the reality is that I like our EMR. But it is easy to not take advantages of the efficiencies that it can bring to the table. Getting refresher training can improve your utilization of note templates, increase your use of smart phrases, reduce your number of mouse clicks per encounter, and improve how you navigate between different EMR windows.
  12. Don’t forget reimbursable services that you probably already perform. Remember that smoking cessation counseling is an add-on billable service to the regular office visit service. Also, if you prescribe inhalers for asthma or COPD, inhaler technique training is a billable service (but it should be done by the office staff, not the physician).
  13. Get the right number of exam rooms per physician. Once again, there is the equivalent of the Starling curve for exam rooms – you can increase productivity up to a point but beyond that, productivity does not increase with increasing number of exam rooms (but your square foot rental expense does continue to increase). The right number of rooms varies by medical specialty and by individual physician. A cardiologist may only need 2 exam rooms whereas a dermatologist may need 4.

Even with peak efficiency, it is hard to get a physician in an office practice up to the $50 – $60 per wRVU that it takes to cover the physician’s salary so physicians are increasingly becoming hospital-employed in order to get access to hospital subsidies that would not be available in a physician-owned practice due to Stark laws and anti-kickback laws. As the medical director of a hospital, my responsibility is to ensure that the hospital-associated outpatient medical practices continue to use the tenets of practice efficiency to prevent the hospital from running a deficit and going out of business.

February 24, 2018

Inpatient Practice Outpatient Practice

Should Your Hospital Hire Locum Tenens Physicians?

It is a vexing question that most hospitals and medical practices face eventually: when you have an expected or unexpected physician vacancy, do you bring in a temporary substitute? Locum tenens is a huge business – rarely a day goes by that I don’t get a phone call or an email from a locum tenens company asking if I’d be willing to cover a pulmonary practice for a few weeks or do a few ICU shifts in some community in the Midwest.

There are a lot of very legitimate reasons why bringing in a locum tenens physician makes sense. The local physician could be out on an extended medical leave or maternity leave. A doctor in the National Guard could be called up for active duty. Maybe there was an unexpected death or retirement and the replacement physician can’t start for a few months. There could be unexpected physician resignations or an unexpected growth in clinical demand and hiring just can’t keep up.

In nursing, we call those temporary workers “travelers”. In business, they are often called “temporaries”. When it comes to physicians, we call them “locum tenens”. The word comes from Latin meaning “placeholder”. I’ve seen these employees work both ways. For example, when I was the treasurer of our Department of Internal Medicine, we brought in a temporary administrative director to oversee our revenue cycle department and she brought in a wealth of knowledge and experience – we were able to capitalize on her objective analysis of our operations as an outsider looking in so that we were able to greatly improve our billing and collection efficiency. On the other hand, I have seen hospitals bring in temporary physicians who did not perform clinically as expected and resulted in a deterioration in the quality of care in those hospitals.

Many hospital medical directors and hospital credentials committees remember the bad experiences and forget about the good experiences and thus try to avoid locum tenens doctors. This month, in JAMA, we finally have some objective data to help us to decide whether bringing in locum tenens physicians is safe. In an article from Harvard, investigators looked at 1.8 million Medicare admissions covered by an internist between 2009-2014. They found that 2.1% of the admissions were covered entirely by a locum tenens physician. In addition, 9.3% of the non-locum tenens primary admitting physicians were covered by a locum tenens physician at some point during the hospitalizations.

The key finding was that there was no difference in 30-day mortality between the locum tenens physicians (8.83%) and the non-locum tenens physicians (8.70%). There were some interesting differences, however. Patients treated by locum tenens physicians had higher costs of hospitalization ($1,836 versus $1,712), longer length of stays (5.64 days versus 5.21 days), and lower readmission rates (22.80% versus 23.83%) compared to patients treated by non-locum tenens physicians. In summary:

  • No difference in mortality
  • Longer length of stay
  • Higher cost of hospitalization
  • Lower readmission rates

There were also some interesting demographic differences. Locum tenens physicians were more common in Southern and Western United States. They were also more common in smaller rural and suburban hospitals (as opposed to urban hospitals) and in public hospitals (as opposed to private hospitals).

A recent survey of healthcare organizations indicated that 85% use locums tenens at some time. Although you might think that locums tenens primarily attracts younger, more mobile physicians, it turns out that 75% of locum tenens physicians are over age 51, that is, toward the end rather than at the beginning of their careers. When considering hiring a locum tenens physician in your hospital, it is important to understand the doctor’s motivation to do locum tenens work – there are “good” reasons and there are “bad” reasons:

The Good Reasons For Being Locum Tenens:

  1. Desire to visit different parts of the United States or to visit family members residing in different parts of the country.
  2. Desire to try out a hospital practice or a part of the country before committing permanently (like dating before committing to marriage).
  3. Wanting to scale back clinical practice as a bridge to retirement.
  4. Flexibility in scheduling and ability to have extended time off during the year.
  5. Preferred lifestyle.
  6. Enrichment of professional experiences by practicing in multiple locations to improve clinical abilities.
  7. Better pay.

The Not So Good Reasons For Being Locum Tenens:

  1. The physician does not get along with other people and cannot maintain lasting relationships.
  2. The physician has substandard practice and cannot hold a job for a long period of time.
  3. The physician does not want to be invested in the long-term success of an organization.
  4. Better pay (this can be either a good reason or a bad reason, depending on the circumstances.
  5. The physician has become burned out and is just trying to stay employed.

When considering a locum tenens physician, the hospital will most often be dealing with a locum tenens company as an intermediary. It is important that the hospital stick with its usual credentialing process to vet the physicians – there can be a tendency to delegate some of the credentialing steps to the locum tenens company with the assumption that they are as thorough or have as high of standards as the hospital’s credentials committee. However, that is not necessarily the case. The locum tenens company is selling you a product, namely the doctors on their list and they will be motivated by trying to make a sale rather than by trying to improve your hospital’s quality of care. There is also the belief that because the locum tenens physician will only be there a short while, that the hospital can live with lower standards than they would require for a regular permanent physician. However, remember, it is always harder to get someone off of the medical staff than to put them on in the first place and if that physician decides to stay in the area and now has regular medical staff privileges, he/she could decide to continue to practice at the hospital long-term. Also, a bad doctor can do a lot of damage to your hospital in a short period of time. There is also the issue of cost – locum tenens physicians will usually be more expensive than a regular, permanent physician.

Some large physician groups will have an internal locum tenens group in order to fill temporary vacancies at the various practice locations that they have. Although you can be a bit more sure of what you are getting in this circumstance (since these locum tenens physicians are actually employed by the physician group), they still require the regular vetting you use for any other new physician, including contacting references from past locations where they have practiced.

The use of locum tenens physicians is becoming more and more common. Most hospitals in the U.S. will be using at least some locum tenens physicians in the next year. The new study indicates that the overall quality of care by locum tenens physicians is good. However, the hospital still need to be sure that the individual locum tenens physician’s motivations for his/her career choice will mesh with the hospital’s culture and goals.

December 7, 2017


Electronic Medical Records Outpatient Practice

Why The Medication List In Your Electronic Medical Record is Wrong

The importance of an accurate medication list for every outpatient cannot be overstated. In theory, the electronic medical record should improve the accuracy of the medication list but in reality, medication lists are very often inaccurate. The act of verifying the medication list is called “medication reconciliation” which sounds so easy on the surface but is so hard in reality.

The Agency for Healthcare Research and Quality (an agency of the U.S. Department of Health and Human Services) says that there should be a single medication list that is the “one source of truth” for the patient and that medical practices should standardize and simplify the medication reconciliation process in order to make the right thing to do the easiest thing to do. But in most outpatient practices, mediation reconciliation is neither standardized, simple, or easy. Pull up any patient’s electronic medical record and if that patient sees more than 1 physician, there is a pretty good chance that the medication list is not accurate. Here are some of the reasons why:

  1. No stop date on short-term medications. Recently, I opened a patient’s chart and found that the medication list included amoxicillin. I asked the patient who said she wasn’t taking amoxicillin. So I pulled up the medication history and found that she had gotten a 7-day course of amoxicillin in 2009 (8 years ago) but it had never been taken off of her list. With electronic prescribing, a physician can set a duration of therapy (or set a stop date) and after that time, the medication falls off of the medication list. However, even if only 14 pills are prescribed with no refills, if the physician does not set the duration or stop date, then that medication stays on the list in perpetuity. I prescribe a lot of short courses of prednisone and antibiotics and so I have my own prescriptions for these commonly prescribed short-term medications in my “preference list” in our electronic medical record and these include a fixed number of days duration. But many physicians don’t have a preference setting for every antibiotic or other short-term medication that they prescribe and so if they don’t manually enter the stop date every time they write a prescription, that medication will continue to appear to be a long-term maintenance medication forever.
  2. Restricted hospital formularies. Hospital pharmacies cannot stock every single medication that is on the market. First, it is too expensive to maintain that wide of an inventory. Second, each hospital negotiates with various pharmaceutical companies or medication wholesalers for the least expensive of therapeutically equivalent medications in order to keep their costs down. As a consequence, when, for example,  a patient with asthma taking the inhaler Advair gets admitted to a hospital that does not have Advair on the formulary but does have the similar inhaler, Symbicort, then the admitting hospitalist will prescribe Symbicort while the patient is in the hospital. When that patient gets discharged, it is very easy for Symbicort to show up on the discharge medication list but since the patient does not know that Advair and Symbicort are equivalent drugs, that patient will start taking both inhalers. When that patient then sees a primary care physician, the medication list will include Symbicort instead of Advair even though the patient is now taking both drugs.
  3. “Don’t mess with my medication list”. Frequently, the responsibility for maintaining an accurate medication list falls to the primary care physician and in large, multi-group practices, specialists are often told not to take anything off of the patient’s medication list unless checking with the primary care physician first in order to insure that the list is, in fact, correct. The problem is that nobody has time to check with a patient’s primary care physician every time a patient says that they are not taking a medication that appears on their list so the path of least resistance is for specialist to never delete a medication from the list, only add new medications that they prescribe.
  4. Leaving medication reconciliation only up to the doctor. Medicare sets the rules for what nurses or medical students can document in the medical record and what the physician has to document. The “past medical history” is a chart component that nurses and medical students are permitted to document. The past medical history is supposed to include the patient’s medication list. Consequently, in many practices, the first time a patient is seen in that practice, the nurse will record the past medical history, including the patient’s allergies, previous surgeries, and current medications. Because it takes a lot of time to enter a long list of medications (including dose, frequency of administration, etc.), it makes sense to have a nurse or medical assistant do all of that documentation, thus freeing the physician’s time up to see more patients. But in many parts of the country, medication reconciliation is considered to be equivalent to prescribing a medication and so physicians are the only ones who can do medication reconciliation which includes taking medications off of the list when a patient reports that he/she is no longer taking them. Thus, the nurses can add to the list but only the physicians can delete from the list. This creates duplicate work and confusing responsibilities and as a consequence, errors occur. In the hospital, deleting a medication from the medication list directly affects the patient’s treatment and should only be done by a physician; however, in the outpatient setting, deleting a medication from the medication list is more justifiably done by a nurse if the patient says that they do not take that medication any longer. It is very important to distinguish the role and responsibility of the nurse in adding/deleting medications from the medication list in the inpatient setting versus the outpatient setting. Your practice has to decide whether maintaining an accurate medication list in the outpatient electronic medical record is part of recording the past medical history or a part of the medication prescription process – if you consider it part of the past medical history, then empower the nurses to do it.
  5. Medication lists cluttered with non-medication orders. In most electronic medical records, a lot of stuff that gets ordered ends up in the medication list because the EMR doesn’t know what else to do with it. Thus, disability parking placards, the influenza vaccine that the patient received 3 years ago, home oxygen, and wheelchairs will show up in the medication list. This ends up making the list excessively long and more confusing for anyone who is looking at it.
  6. What is the truth, really? Is the “one true source” a list of what the patient says that he or she is actually taking or what the doctor thinks that they are supposed to be taking? If you ask 100 doctors, 50 of them will say it is what the patient is actually taking and the other 50 will say it is what was actually prescribed. So, if a patient was prescribed a medication for gout, but never filled the prescription because it was too expensive, should that medication be on the patient’s list? One the one hand, it is the medication prescribed to treat the gout so you want to have that as part of your on-going medical record. But on the other hand, if the patient isn’t taking it, then should it keep appearing in the medical record? This is controversial and there is no perfect answer.
  7. Samples. If you give a patient a sample of a medication in the office, then you want to have documentation of them getting that medication, for example, a small tube of a steroid cream that a dermatologist gives a patient for their poison ivy. But once again, if there is not a stop date when you enter the sample in the medication list, then it stays on the list until someone else takes it off of the list at a later date.
  8. Bad data from other institutions. Many EMRs have the ability to interrogate other hospital systems that the patient has visited to pull in medications from that hospital system’s medical record. Even if the physicians in your hospital system are tediously compulsive about keeping an accurate medication list, unless the other hospital system’s doctors are equally compulsive, you can pull in errors into the patient’s medication list.
  9. There just isn’t enough time. When a patient gets admitted to the hospital, usually there is a nurse, a pharmacist, and a physician, all reviewing the patient’s medications. The patient is usually in the hospital for a few days so there is ample time for a thorough review and reconciliation with checks and double checks. But in the outpatient physician office, the responsibility for medication documentation often all falls solely the physician. Even if the office nursing staff participate, the time that the patient is in the office is so short that it is hard to get everything done. A typical physician’s return office visits are scheduled every 15 minutes. That means that in those 15 minutes, the physician has to review the patient’s history and any new test results, take an interval history from the patient, do a physical exam, discuss their findings and recommendations with the patients, order any new tests or medications, document a note, create a letter to the referring physician, enter their billing charges, and do the medication reconciliation. Often, there simply just isn’t enough time to do all of that and frequently the first corner to get cut is to skip the medication reconciliation. Some people would say that the answer is to schedule patients every 20 minutes, rather than every 15 minutes. However, over the course of a full day, that would equate to 25% fewer patients being seen that day and the reduction in revenue associated with 25% fewer patients would put most practices out of business.
  10. Poorly designed EMR workflows. A physician will naturally go to three windows in the electronic medical record: the progress note window, the test results window, and the orders window. Anything else and they will need a prompt. If the nurses in the office are permitted to mark medications for deletion but are not actually permitted to delete those medications from the medication list, then there has to be a prompt for the physician to go to that medication list window to approve those deletions. Otherwise, they will close that encounter without ever deleting those medications. Either the nurses should be empowered to delete medications that the patient says they are no longer taking or there needs to be a hard-stop to go to the medication list window before the physician can close the encounter.

The good news is that there are some concrete things you can do to reduce the inaccuracy of the medication list. Here are a few:

  1. Be sure that all short-term medications and samples have a  stop date in the initial prescription.

  2. Develop an agreed-upon consensus among all of the physicians in the organization about whether they define the medication list as the list of medications prescribed for the patient or what the patient says that they are actually taking.

  3. Develop an institutional policy that all physicians are allowed to delete medications from the list – the primary care physician or any specialist.

  4. Allow office staff to work at the top of their license with respect to adding or deleting medications from the outpatient’s list.

  5. Develop a simple practice to ensure that therapeutic substitutions made for inpatient formulary reasons at the time of admission to the hospital are substituted back to the patient’s normal outpatient equivalent drug.

  6. Create workflows in the office that are consistent so that everyone knows their role in maintenance of the medication list and performs that role on every patient.

  7. Design the electronic medical record so that the physician’s tasks for outpatient medication reconciliation becomes a natural part of the workflow or so that medication reconciliation is required before the encounter can be closed.

  8. Involve the patient by printing a copy of the medication list when the patient arrives to the clinic and having him/her review the list for additions/deletions.

November 6, 2017

Outpatient Practice

Medications To Prevent COPD Readmissions

The number of inhalers available by prescription in the United States grew enormously this year. If it is hard for me as a pulmonologist to keep up with all of them, then it has got to be nearly impossible for the hospitalist or primary care physician to keep up. But with all of the new medications available, which ones actually work to help reduce readmission to the hospital after an admission for a COPD exacerbation? In my own practice, I have a lot of patients come in who are on the wrong medications – and these not only can add unnecessary expense but in some cases, they can actually make patients worse.

In 2015, the American College of Chest Physicians and the Canadian Thoracic Society partnered to create a clinical practice guideline: Prevention of Acute Exacerbations of COPD.  There is a lot of good information in it but there have been a number of important studies since its publication that are furthering our knowledge of the best ways to prevent COPD exacerbations and readmissions for COPD. The key recommendations are for a step-wise treatment for patients with COPD, starting at the lowest step and then moving up the steps if the patient’s condition warrants:

  1. Short-acting combination anti-cholinergic plus short-acting beta agonist should be the initial PRN rescue inhaler (e.g., Combivent).
  2. A LAMA (long-acting muscarinic antagonist) should be the first line maintenance therapy. Once a LAMA is started, the short-acting PRN rescue inhaler should be changed to a short-acting beta agonist alone (e.g., albuterol).
  3. A LAMA + LABA (long-acting beta agonist) should be the second line maintenance therapy
  4. A LAMA + LABA + ICS (inhaled corticosteroid) should be the third line maintenance therapy
  5. Inhaled corticosteroid alone is not recommended
  6. Azithromycin, roflumilast, and N-acetylcysteine can also reduce readmissions.

Inhalers are expensive and the least expensive, albuterol, will still run the patient about $60 per inhaler if they buy out of pocket. Inhaler prices, like prices for all medications, are hard to pin down because each insurance company will take price bids from different pharmaceutical companies for drugs like inhalers so the insurance companies almost always contract for a price that is much lower than the sticker price that an uninsured person would pay. The following is the price on for common rescue inhalers:

Once a patient is on a maintenance inhaler, the cost goes up. For many years, Spiriva had a monopoly on the l0ng-acting anti-muscaric market but now, there are additional competitors.


The newest products are the combined long-acting muscarinic antagonists and long-acting beta agonists (LAMA/LABAs). Although these can be prescribed individually as a LAMA inhaler plus a second LABA inhaler, the combination inhalers are less expensive than purchasing the LAMA plus the LABA individually.


The LABAs can also be combined with an inhaled corticosteroid. For many years, the major player in the market was Advair but now there are several competitors available, including the newest, Airduo, which is essentially a generic form of Advair.

The problem with inhaled corticosteroids is that although they can reduce the risk of COPD exacerbation, they can increase the risk of pneumonia. Therefore, over-prescription of inhaled corticosteroids can actually increase readmissions for COPD. It appears that the best predictor appears to be the blood eosinophil percentage. If the eosinophil count is above 4%, then the benefit in COPD exacerbation reduction exceeds the harm from pneumonia – in the figure to the right, a group of 100 COPD patients with >4% blood eosinophils will have 2 excess hospitalizations for pneumonia but 5 fewer hospitalizations for COPD exacerbation if treated with an inhaled corticosteroid. However, if the eosinophil count is less than 200, the harm from pneumonia exceeds the benefit from reducing COPD exacerbations – in the figure to the right, a group of 100 COPD patients with <2% blood eosinophils will have 2 excess hospitalizations for pneumonia and only 1 fewer hospitalization for COPD exacerbation. Blood eosinophil counts between 200-400 are a gray area where the risk/benefit ratio is more uncertain but risk is probably greater than the benefit (2 excess hospitalizations for pneumonia versus 1.5 fewer hospitalizations for COPD exacerbation). The bottom line is that COPD patients with normal blood eosinophil counts should not routinely be prescribed inhaled corticosteroids.

Three drugs have been shown to reduce COPD readmissions: azithromycin, roflumilast, and N-acetylcysteine. Roflumilast tends to get all of the publicity since it is a non-generic prescription medication that gets a lot of advertising; however, the monthly cost is very high. Azithromycin is a generic prescription medication that is considerably cheaper per month. N-acetylcysteine is an over-the-counter generic medication that is very inexpensive on a per-month basis and thus gets no advertising. At this time, we don’t have head-to-head comparisons of these medications to know which one is better so I tend to start with N-acetylcysteine first, then go to azithromycin (as long as the QTc interval on the EKG is OK), and then go to roflumilast (last).

The most recent study looking at strategies to reduce COPD readmissions looked at home nocturnal mechanical ventilation using a BiPAP device. The main entry criteria was a PCO2 > 53 mm Hg. Those patients started on home BiPAP stayed out of the hospital for an average of 4.3 months whereas control patients stayed out for an average of 1.4 months. Based on this study, the use of home BiPAP at night appears to be another effective way of reducing COPD readmissions, at least in those patients with fairly severe COPD.

So, when discharging a patient with COPD, be sure that they have a PRN rescue inhaler. Pretty much anyone with COPD who has it severe enough to get hospitalized with an exacerbation warrants a maintenance inhaler, generally a LAMA. By judiciously stepping-up patients and selecting add-on medications based on the medical literature rather than on personal practice convention, patients can be maintained on the medications that are most likely to keep them out of the hospital without adding unnecessary costs.

September 28, 2017

Outpatient Practice

Prescribing Oxygen

It is pretty common to have to start a patient on home oxygen after discharge from the hospital but ordering oxygen for home use is much different and much more complicated than ordering oxygen in the hospital. The good news is that our patients have more options than ever before but the bad news is that with more options comes a lot more complexity. In this post, I’ll summarize what you need to know about prescribing home oxygen.

The first thing to understand is how Medicare covers home oxygen. Most insurance companies follow whatever Medicare decides, so knowing the Medicare rules will generally apply to other insurance companies. When you order home oxygen through a respiratory therapy company, Medicare will pay that company to provide oxygen for up to 36 months. Medicare pays 80% of a fixed amount each month that is supposed to cover the rental of the oxygen equipment (concentrators, oxygen tanks, etc.) plus supplies (oxygen tubing, etc.). This amount varies depending on the geographic location in the country and the specific billing codes used but the Medicare reimbursement is generally in the $150-$200/month range. The patient (or the patient’s supplemental insurance) is responsible for the other 20%. After 36 months, the company has to provide the equipment for 24 months (for free) and Medicare will only reimburse that company a small amount for a “service charge”. After 5 years, the cycle starts all over again. The key operative here is that Medicare will pay to rent equipment but not to buy equipment. This is important because after 3 years, the total amount that a respiratory therapy company can get for rental of expensive equipment, such as a portable oxygen concentrator (total cost about $2,500) is often not enough to cover the cost of that equipment and since the company is then required to supply that equipment for free for the next 2 years, by the time 5 years is up, the equipment is fully depreciated and can’t be re-rented to another patient. The Medicare pricing for the various oxygen HCPCS codes for each state can be found at the CMS Noridian website. Prescribing oxygen requires the physician (or NP/PA) to fill out the Medicare oxygen CMN (certificate of medical necessity) form – this form is usually sent to the patient’s outpatient physician after discharge from the hospital; therefore, having all of the required oxygen qualification information documented in the inpatient medical record or in the discharge summary can make it much easier for the patient’s primary care physician to gather the information necessary to complete the CMN.

In order for Medicare to cover oxygen, the patient has to have a qualifying oxygen level. Furthermore, there has to be documentation that the patient’s oxygen level improves with supplemental oxygen. Therefore, you have to have in the chart the following:

  1. Documentation that the oxygen level is low
  2. Documentation that the oxygen level can be restored to a normal level with supplemental oxygen
  3. The amount of oxygen (in liters per minute) that are necessary to return the oxygen level to an acceptable oxygen saturation level.

There are several ways that a patient can be qualified for coverage for home oxygen. The oxygen level can be low at rest, while sleeping, or with exertion. The following are the thresholds that Medicare requires:

  1. Resting PO2 < 55 or oxygen saturation < 89%
  2. Oxygen saturation < 89% during exercise (for example, walking in the hall)
  3. Oxygen saturation < 89% for > 5 minutes while sleeping

Testing for home oxygen is most commonly done with an “oxygen titration study” (sometimes called an oxygen qualification test). During this test, the oxygen saturation measured using an oximeter and then the patient walks until their oxygen saturation drops below 89%. The patient is then given oxygen and it is turned up to the flow rate necessary to keep the oxygen saturation above 89% while walking.

Once oxygen is prescribed, there are several choices you will need to make. Often, oxygen equipment is obtained from a home health or home respiratory therapy company. These companies will usually have several different options for home oxygen and can help the patient select the oxygen option that is best for them. However, because of the way that Medicare pays these companies for home oxygen (monthly payments for rental of oxygen equipment), some oxygen delivery devices, such as portable oxygen concentrators, are not cost-effective and thus not offered by all home oxygen companies. Patients can also purchase oxygen concentrators directly from the manufacturers (more on that below).

There are three equipment options for home oxygen: oxygen concentrators, compressed oxygen gas, and liquid oxygen.

  1. Stationary oxygen concentrators. These are electrical devices that concentrate the oxygen that is in the regular air. Stationary oxygen concentrators are large machines about the size of a large suitcase that are meant to be kept in one location in the home. Patients can connect oxygen tubing up to 50 feet in length for mobility within their house. There are two general types of stationary oxygen concentrators: those that go to 5-6 L flow rate and those that go up to 8-10 L flow rate (“high flow” concentrators). Although these can be purchased directly from manufacturers, most patients rent a stationary oxygen concentrator from a home respiratory therapy company.
  2. Portable oxygen concentrators. These are smaller devices that are about the size of a toaster and weigh about 4-7 pounds. They are electrically powered and can be run off of batteries, a regular wall socket, or an automobile’s electrical system. The batteries are rechargable and can last up to 10 hours depending on the oxygen flow rate setting. Most portable oxygen concentrators will provide up to 4 L oxygen flow but some of the larger devices will go up to 6 L oxygen flow. Most portable oxygen concentrators will deliver oxygen by either “continuous flow” or “oxygen conserving flow”. Continuous flow means that the oxygen flows continuously whether you are breathing in or out. Oxygen conserving flow means that the oxygen only flows when you are breathing in. Because it takes energy to make the oxygen, the battery will last longer when set to oxygen conserving flow. Portable oxygen concentrators are a good option for patients who are frequently out of the house, who have a house big enough that 50 feet of oxygen tubing is insufficient, who travel by car, or who travel by airplane (portable oxygen concentrators are the only oxygen equipment permitted by commercial oxygen companies). Some home respiratory therapy companies will offer at least one brand of portable oxygen concentrator but there are many companies that make oxygen concentrators and for many patients, it is easier to purchase their own portable oxygen concentrator directly from the company in order to get the specific brand that they prefer. Most portable oxygen concentrators cost about $2,500 to $3,000. Some of the popular brands include: Inogen, Airsep, Inova, Respironics, and Invacare.
  3. Compressed oxygen gas. These are supplied by a home oxygen supply company or home health company. There are many different sizes of the oxygen cylinders. Very large stationary tanks (H cylinders) are designed to be back-ups to the stationary oxygen concentrators (for example, in event of a power outage) and are meant to be left in one place in the home (they are not portable). Medium sized tanks (E cylinders) are designed to be pulled in a wheeled cart that looks a little like a cart used to carry a golf bag. They are difficult to get in and out of cars and are difficult to go up and down stairs with; depending on the flow rate, these will usually last 2-8 hours. Small sized tanks (M-6 cylinders) are designed to be carried over the shoulder or in a backpack. The small sized tanks are usually used with an oxygen conserving device to make them last longer; depending on the flow rate, these will usually last 1-4 hours. Many of the small sized oxygen tanks can be recharged from a stationary oxygen concentrator.
  4. Liquid oxygen. Oxygen in liquid form is much more dense than oxygen in gas form. Liquid oxygen reservoirs are devices that are about 2-3 feet tall and plug into the wall electrical socket. The reservoirs are very heavy and cannot be moved around the house easily. They will usually last several weeks and then are replaced by a home oxygen supply company. One advantage of liquid oxygen reservoirs is that they can be run at a very high oxygen flow rate (10-15 L) for patients who require a lot of oxygen. Liquid oxygen can also be used with small portable tanks that weigh 3.5 to 5.5 pounds (for example, Helios brand). An advantage of the portable liquid oxygen tanks is that they last much longer than compressed oxygen gas tanks (usually 4-22 hours depending on the flow rate and size of the tank). The portable oxygen tanks can be quickly refilled from the liquid oxygen reservoir.

Practical tips for using oxygen for patients:

  1. Be sure that you plan ahead when traveling outside of the house so that you do not run out of oxygen. Always take extra tanks or batteries (for portable concentrators) in case your traveling takes longer than anticipated.
  2. Do not take your oxygen off in the shower – water will not hurt the oxygen tubing. Keep any electrical equipment (such as oxygen concentrators) outside the shower so the electrical equipment does not get wet.
  3. If you use a higher flow rate with exercise than when you are at rest, turn the oxygen to the higher flow rate a minute or two before you begin your activities. For example, if you need 2 L flow at rest but 5 L with exertion and you are getting ready to walk up the stairs, be sure to turn the oxygen up to 5 L before you get up to walk up the stairs.
  4. You can purchase a finger oximeter at your pharmacy or on-line to help guide your oxygen flow rate. These cost as little as $35.
  5. Do not smoke in a house with oxygen equipment in it. Also, do not use your oxygen when you are around open flames (such as a gas oven or gas grill). Above all, never smoke while using oxygen – it will cause serious or even life-threatening burns to the face and nose.
  6. Let your power company know that you are on oxygen so that your home can be targeted to have power restored earlier in event of a power shortage.
  7. Let your local fire department know that you have oxygen in the home. Consider placing an “oxygen in use” sign on the door.
  8. Oxygen can dry out the nose and this can result in the nose feeling dry, nose bleeds, or sometimes an excessively watery nose (as the nose tries to counterbalance the drying effect of the oxygen). If this happens, try using over-the-counter saline nose spray every few hours to keep the nasal passages moist. You can also use water-based lubricants (such as K-Y jelly) but do not use oil-based lubricants (such as vasoline).
  9. You cannot get addicted to supplemental oxygen. Either you need it or you do not need it but you will not become dependent on it just by wearing it. Sometimes, you may need to use the oxygen more often that usual, for example, if you get pneumonia. Once the pneumonia clears up, you can usually go back to your previous oxygen flow rate.

My Bias:

Personally, I like the portable oxygen concentrators. If money is not a barrier, I think purchasing (or renting) a stationary concentrator for primary home use plus also purchasing a portable concentrator for use outside of the home gives patients the most freedom and convenience. With home oxygen options, you get what you pay for and even though the better-made portable concentrators are more expensive and are often not be covered by Medicare, they can make life much easier. For patients who don’t have $3,000 to buy a stationary plus portable concentrator, a home respiratory therapy company can help tailor the equipment that would be covered by Medicare to the patient’s personal needs.

September 23, 2017

Outpatient Practice

Size Matters: What Is The Best Patient Panel Size For A Primary Care Physician?

In the past, we assessed how productive a primary care physician was by how many RVUs he or she generated. With the move toward accountable care organizations, quantity of visits is less important than value of care. Primary care physicians are often employed by hospitals or, in the case of academic medical centers, they are subsidized by the health system. So, we need to find another way other than RVUs to determine if the primary care physician is working at 100% effort.

One way of doing this is by the physician’s panel size. This is the number of patients that that physician is the primary care provider for. But there are a lot of problems with using panel size as a productivity metric because it is very easy to game the system. There are a lot of numbers thrown around about the optimal patient panel size. Historically, an often quoted number is 2,500 but this value was calculated on some flawed assumptions. More recently, many studies suggest optimal panel sizes from 1,300 to 1,900 per physician. So, what is the correct number? As with most things in medicine… it depends. There are a myriad of variables that affect the number of patients that a primary care physician should have in his/her panel. Here are a few:

  1. Panel maturity. By that, I don’t mean how old the patients are but rather how long the physician has been seeing those patients. For a physician straight out of residency, every patient is a new patient and requires a lot more effort to get them plugged into specialists, a preventative medicine calendar, etc. Those new patient visits take a lot more time and often need shorter interval initial follow up visits. For a first year primary care physician to achieve a panel size of 1,900 in the first year out would require him/her to see 8-9 new patient visits per day and this is really not achievable.
  2. Physician maturity. The more experienced the physician, the more efficient they get in the office. The same thing happens with ER physicians, hospitalists, and surgeons. A physician who has 20 years of practice experience can comfortably manage more patients than a physician straight out of residency.
  3. Clinic maturity. Efficiency in patient flow and management is not just dependent on the experience of the physician but the experience of the nurses and office staff and how well they function together as a team.
  4. Patient maturity. The intensity of primary care is age-dependent with the greatest intensity at either extreme of age. For a pediatrician, taking care of 1,500 children all under age 2 years old is a lot different than taking care of 1,500 13-year olds. Similarly for a family physician: a panel made up mostly of 25-year olds is much easier to manage than a panel the same size made up of 75-year olds. Geriatricians take care exclusively of the elderly and these are patients with more medical problems and psychosocial needs than working age adults; consequently, the geriatrician will have a smaller patient panel than a typical general internist.
  5. Illness complexity. I know some primary care practices composed of a mixture of family practitioners, pediatricians, and internists. The children with more complex health needs, such as cerebral palsy and diabetes, are followed by the pediatricians. The adults with more complex health needs, such as advanced COPD and severe systolic heart failure, are followed by the internists. The rest of the patients are followed by the family medicine physicians. In this model, each patient gets assigned to the physician best equipped to take care of that patient but the pediatricians and the internists care for the most complex patients and generally require a smaller panel size than the family medicine physicians. Similarly, in a large primary care practice that has one physician who manages all of the patients with autism (who require more total physician time per patient per year), that physician will need to have a smaller panel size but taking care of all of the autistic patients, the other physicians will be able to handle larger panel sizes.
  6. Use of advance practice providers. These are usually nurse practitioner or physician assistants. They may see patients independently, in which case they have their own panel of patients – usually smaller than the physicians. Or they may be used to amplify the productivity of the physician by performing less complex clinical tasks that free the physician up to manage more complex tasks. For example, a primary care physician may work with a nurse practitioner who does all of the routine health maintenance visits and preventive care visits thus freeing the physician up to see new patients and visits for new or acute problems. In this model, the physician can maintain a much larger panel size than the primary care physician who is a solo practitioner without a nurse practitioner. There is a solo family physician in our community who manages thousands of patients in an enormous multistory clinic building. He does it by employing an army of nurse practitioners and physician assistants and then he only sees those patients that the NPs or PAs have questions about.
  7. Use of health specialists. Yes, the physician can spend an hour counseling a patient on a diabetic diet. But a dietician’s time is less expensive than a physician’s time. Having clinic staff who can teach patients how to use inhalers, monitor INRs to adjust Coumadin dose, do smoking cessation counseling can all free up the physician to do other activities. Therefore, dietitians, pharmacists, and health educators can equate to a larger panel size.
  8. Health literacy of the patient population. The greater the health literacy of the patients, the easier it is to manage those patients and the easier it is to maintain a higher patient panel size. The same applies to access to healthcare of the patient – it is much simpler to manage a patient with good health insurance than it is to manage a patient with no insurance when it comes to getting medications, tests, and consultants.
  9. The patients’ language. In my own practice, I take care of a lot of elderly Russian-speaking immigrants, French speaking patients from Africa, and Somalis. Each patient encounter takes longer because of additional time that it takes to communicate through a translator. Furthermore, the physician practice is generally responsible for arranging translation and this adds practice expense. The more non-English speaking patients in the panel, the smaller the panel size should be.
  10. Office design. The physical structure of the clinic can have a huge impact on efficiency of patient flow and on how many patients can be seen. This translates to more patients that the primary care physician can comfortably manage in the panel. Having a sufficient number of exam rooms per physician plus easy access to basic testing such as x-ray, EKG, and lab equates to higher panel size.
  11. Clinical education. Medical students represent the next generation of physicians and it is our professional obligation to ensure that the next generation is well-trained. However, medical students can slow you down since, for Medicare compliance purposes, the attending physician has to re-take the history, do a complete exam, and then do nearly all of the progress note documentation. That means that the clinical care time from the physician’s perspective is about the same whether there is a medical student or not, but once you factor in additional time to actually teach the medical student, the time costs add up. Fortunately, most physicians get inherent reward by educating students and so we would accept a smaller panel size, even it it meant a smaller salary, in order to be clinical educators.
  12. The definition of who is in the panel. Our electronic medical record is great – it is very easy for me to see how many individual patients I saw in the office last year. However, if you use the EMR to calculate the panel size of a primary care physician by calculating how many individual patients that physician saw in the past year, you’ll get bad data. Inevitably, there will be times when the primary care physician will be cross-covering for a partner or will be taking acute illness calls or will be the physician in the office on the day that all of the practice’s patients come in for their flu shots. You have to have an accurate way of identifying those patients who are truly associated with each individual primary care physician as their physician of record.
  13. Coordination of care with specialists. In a health system with good inter-physician communication with an electronic medical record and where specialists assume longitudinal management of specific medical problems, the time and effort required by the primary care physician for coordination of care is less and this translates to an ability to manage a larger patient panel. For example, if a patient is found to have pulmonary nodule and the primary care physician can refer that patient to a pulmonologist who will assume responsibility for further testing and follow-up of the nodule, then that frees up a lot of the primary care physician’s time. Some primary care physicians have a practice philosophy of “a consult is a sign of weakness” but these physicians can get tied up spending more time trying to manage highly specialized or complex diseases that could be managed more efficiently by coordinating management with a specialist.
  14. Disease diversity. By this, I mean the homogeneity of the spectrum of health problems that the primary care physician sees. For example, being a general family physician is very challenging because you have to be familiar with the health maintenance requirements and preventive health needs for children all the way through the elderly. You have to be facile with gynecology (and sometimes obstetrics) as well as minor surgery. That is a lot of areas to keep up on from an education standpoint. This can limit the panel size. On the other hand, a primary care physician who limits their practice to one age group or one gender has a more focused patient population and in theory, can often manage a larger number of patients.

In summary, there is not a “one size fits all” when it comes to primary care physician patient panel size. My own take on panel size is that more than 2,500 is probably too many for the typical primary care physician and less than 1,000 is probably too few for the typical primary care physician. Whether an individual physician should have a panel size closer to 1,000 or closer to 2,500 depends on all of the variables above.

August 27, 2017

Inpatient Practice Outpatient Practice

The Future Of Telemedicine Is Now

Telemedicine. It is an idea that just seems to make so much sense but it has had a hard time getting much traction in the United States. But recently, economic forces are causing telemedicine to become a viable option in a number of areas. Here is how we are using telemedicine in our hospital:

Prison medicine.

For more than 25 years, we have had a telemedicine program with the Ohio Department of Corrections. Inmates with medical conditions requiring more than basic primary care go to a video camera set up in the infirmary of Ohio prisons where there is 2-way video communication with a specialist at the Ohio State University Wexner Medical Center. There is a nurse at the prison-end who can do physical examination directed by the physician at the OSU-end and there is an electronic stethoscope for auscultation.

Why it works: It is cost effective for the Department of Corrections. It costs a lot to transport an inmate across the state to be seen by a physician in person and there is security risks. It is far cheaper for the Department of Corrections to pay our medical center for telemedicine visits. Patients that do need in-hospital care for medical or surgical conditions are admitted to a locked-down prison unit at University Hospital.


Patients who come into the emergency department at University Hospital East are evaluated by a two-way video camera system by a psychiatrist located in a remote location. The psychiatrist can determine if the patient needs inpatient hospitalization in a psychiatric hospital or if they can be safely discharged from the ER.

Why it works: Our psychiatrists are based out of the main University Hospital campus where the inpatient psychiatry hospital is. My hospital, University Hospital East, is 9 miles away. A psychiatrist seeing emergency department and psychiatric observation patients would need to leave those patients and drive over to our hospital to see psychiatry patients in the emergency department. Telepsychiatry permits the psychiatrist to evaluate patients remotely, thus keeping the number of patients being boarded in the emergency department for psychiatric evaluation down.


Patients presenting to the emergency department with an acute stroke get a quick screening head CT and then a stroke neurologist (a neurologist who has done an advanced fellowship in stroke medicine) evaluates the patient by a video system. The stroke neurologist then makes a decision about whether the patient should receive t-PA (a blood clot dissolving medication).

Why it works: From the time a patient with an acute stroke arrives in the emergency department, we only have an hour for the emergency department physician to do his/her assessment, get a head CT to be sure that there is not a tumor or bleeding in the brain, be assessed by the stroke neurologist, and then get the t-PA. If you go much beyond an hour, the stroke becomes irreversible. There are not enough stroke neurologists to have one in every emergency department 24-hours a day so the solution is to have a central “hub” (our main University Hospital) and then a few dozen “spoke” hospitals that can use telemedicine to connect to the stroke neurologist at the hub. My hospital, University Hospital East, is one of many spoke hospitals affiliated with Ohio State.


Hospitalist at University Hospital East can take a photo of a rash or skin lesion and load it into our electronic medical record. A dermatologist at a remote location can then look at the photo and determine if there is enough information to make a diagnosis, if a punch or shave biopsy of the skin is required (which can be done by our family physicians or general internists), or if an in-person consult is required.

Why it works: We don’t have a lot of dermatology consults in our hospital. The cost of a dermatologist driving 15-20 minutes each way to come to our hospital to see a consult is not cost effective – they can see a lot of patients in the office in that 30-40 minutes. Because most of the time, the dermatologist can make a reasonably confident diagnosis from the photograph, it allows the dermatologist to see more patients rather than wasting time in the car. We have also extended the teledermatology program to Ohio State primary care physicians who can take a photo of a rash using their smartphone, load that image into the electronic medical record, and then the dermatologist will give an opinion within 24 hours – far simpler than waiting 2 weeks for a new patient appointment; the medical center is currently funding this program since it is not covered by private insurance.


This one is a natural product of the computerization of radiology images. A radiologist can pull up an x-ray or CT scan from any computer, anywhere in the world, and read the radiology study.

Why it works: The images on the computer are every bit as good (and usually better) than the images on old-fashioned x-ray films. We have x-ray sites all over the city and this allows the radiologist to be in one location and give an immediate interpretation of an x-ray done at any of these sites and report them on our electronic medical record.


At University Hospital East, we have a general pathologist on site. But sometimes, you need a pathology specialist to review and interpret the microscope slides of a biopsy. Also, we have a procedure called “endobronchial ultrasound” where a needle aspirate of a suspicious lymph node or lung mass is taken during a bronchoscopy using a tiny ultrasound at the tip of the bronchoscope to guide the needle to the right place. Normally, there a cytopathologist (a pathologist who specializes in looking at cytology slides) does an immediate stain of the cells removed from each pass of the needle to tell the bronchoscopist whether there is enough cells to make a cancer diagnosis or whether additional passes of the needle are required. In telepathology, the microscope slide is loaded into a special microscope that can be manipulated by a pathologist viewing the images from that microscope over the internet, miles away.

Why it works: The cytopathologist can be at one location and review slides being done simultaneously from multiple other locations without having to travel to multiple sites in a single hospital or to multiple hospitals within a hospital system. Also, if a surgical procedure runs late, an on-call pathologist at one hospital location can do a frozen section analysis of a specimen at a different location so that a second pathologist does not need to be called in to do the frozen section.

In the past, telemedicine has always been constrained by reimbursement: Medicare and insurance companies did not want to pay for something seen as less than a “real” face-to-face encounter. But there are other economic factors in play – it is too costly for hospitals to have a stroke neurologist, psychiatrist, cytopathologist, dermatologist, or radiologist physically located at all sites within a large medical center at the same time. So even though many of the telemedicine services mentioned in this post are not reimbursable, they are financially viable because they obviate the need for a high-cost specialist to have a lot of down time driving from one location to another.

If we move toward a capitated health care model, we will likely see telemedicine incorporated into other practices. Just imagine how happy patients would be to get a ring on their Skype program when their doctor is ready to see them, rather than having to drive to their doctor’s office, park, wait in the waiting room, and then wait again in the examination room to finally see the doctor who is (as all doctors always seem to be) running late. The only thing holding us back is economics.

June 6, 2017

Outpatient Practice

CG-CAHPS Survey Demystified

The HCAHPS survey is designed to let patients rate their hospital based on their satisfaction with the care provided during an inpatient stay. The CG-CAHPS survey lets patients rate their outpatient doctor or group practice. CG-CAHPS stands for “Clinician and Group Consumer Assessment of Healthcare Providers and Systems”. The information is publicly reported on Medicare’s Physician Compare website. In my case, the CG-CAHPS for all of Ohio State University Department of Internal Medicine outpatient physicians is reported in aggregate. However, internally, we get reports from Press Ganey that allows us to drill down on individual practice sites.

Here are the questions on the GC-CAHPS survey:

  1. If this was a visit for an illness, injury, or condition for care you needed right away, did you get this appointment as soon as you thought you needed it?
    • Yes
    • No
  2. If this was a visit for a check-up or routine care, did you get this appointment as soon as you thought you needed?
    • Yes
    • No
  3. In the last 3 months, if you phoned this provider’s office with a medical question during regular office hours, how often did you get an answer to your medical question that same day?
    • Never
    • Sometimes
    • Usually
    • Always
  4. In the last 3 months if you phoned this providers office with a medical question after regular office hours, how often did you get an answer to your medical question as soon as you needed?
    • Never
    • Sometimes
    • Usually
    • Always
  5. In the last 3 months, if this provider ordered a blood test, x-ray, or other test, how often did someone from this provider’s office follow-up to give you the results?
    • Never
    • Sometimes
    • Usually
    • Always
  6. During this visit, did you see this provider within 15 minutes of your appointment time?
    • Yes
    • No
  7. During this visit, did this provider explain things in a way that was easy to understand?
    • Yes, definitely
    • Yes, somewhat
    • No
  8. During this visit, did the is provider listen carefully to you?
    • Yes, definitely
    • Yes, somewhat
    • No
  9. During this visit, if you talked to this provider about any health questions or concerns, did this provider give you easy to understand information about health questions or concerns?
    • Yes, definitely
    • Yes, somewhat
    • No
  10. During this visit did this provider seem to know the important information about your medical history?
    • Yes, definitely
    • Yes, somewhat
    • No
  11. During this visit, did this provider have your medical records?
    • Yes
    • No
  12. During this visit, did this provider show respect for what you had to say?
    • Yes, definitely
    • Yes, somewhat
    • No
  13. During this visit, did this provider spend enough time with you?
    • Yes, definitely
    • Yes, somewhat
    • No
  14. Using any number from 1 to 10, where 0 is the worst provider possible and 10 is the best provider possible, what number would you use to rate this provider?
  15. Would you recommend this provider’s office to your family and friends?
    • Yes, definitely
    • Yes, somewhat
    • No
  16. During this visit, were clerks and receptionists at this provider’s office as helpful as you thought they should be?
    • Yes, definitely,
    • Yes, somewhat
    • No
  17. During this visit, did the clerks and receptionists treat you with courtesy and respect?
    • Yes, definitely
    • Yes, somewhat
    • No
  18. In the last 3 months, if you took any prescription medications, how often did you and anyone on your health team talk about all the prescriptions medicines you were taking?
    • Never
    • Sometimes
    • Usually
    • Always

There are additional questions about the patient’s demographics, education level, overall perception of health, and race.

The data is aggregated into different topics and then reported as the percent of patients rating that provider as “top box”. So, for example, in the case of question #17, the top box would be ratings of 9 or 10. Similar to the HCAHPS survey, the CG-CAHPS survey suffers from grade inflation. In our specific clinic location, 93.5% of patients rate us as top box, with either a 9 or 10 overall rating. You can see the average scores at the Agency for Healthcare Research and Quality website where you can see averages by region of the country, physician specialty, type of practice, etc.

Some of the questions can be misleading:

When a patient is seeing a consultant the, questions about whether the provider had your medical records is a reflection of whether the referring physician sent records – if the consultant did not receive records before the patient’s initial visit, they cannot request those records until after the patient has come for the office visit and signed a release form allowing that consultant to request outside records. Therefore, if the referring physician failed to send records, the consultant gets a bad rating for question #11.

Many lab tests can take 2-3 weeks before the results come back. If the GC-CAHPS survey is sent out 2 days after the office visit, the lab test results will not be available and so of course no one from the office will have contacted the patient with the results. You have been set up for failure on question #5.

A specialist is generally not going to talk with the patient about all of their prescription medications every visit. Lets face it, you don’t want your orthopedic surgeon talking with you about the inhalers your pulmonologist prescribed. Similarly, do you really expect your dermatologist, who you are seeing for a mole, to review with you every one of the 25 medications you are taking for heart failure, diabetes, and hypertension? You are going to take a hit on question #18.

Right now, the data is mainly just accumulating and being reported for large group practices (such as the OSU Department of Internal Medicine). I get my own personal patient satisfaction scores internally from our department. However, some health systems now publicly report individual physician scores from the CG-CAHPS (for example, University of Utah). For physicians who are aghast that this information could be made public, search yourself on the internet – you are already being ranked by your patients at websites such as Vitals, Angie’s List, and Healthgrades. That data is unfiltered and subject to error (for example, in Healthgrades, I am listed as a neurologist, not a pulmonologist).

So get ready, you’re being rated just like restaurants on Yelp or hotels on Trip Advisor.

March 14, 2017