Being in Private Practice as an Internal Medicine doctor, I usually take time from my office schedule to take advantage of the holiday season to be with my family and get energized for another year of patient care. A unique practice offering for my patients is my Inpatient Advocate Service™. During the Christmas week I had a couple of admitted patients to assist. The Inpatient Advocacy program was designed to allow me to reconnect my traditional role as my patient’s main doctor when hospitalized for acute or critical illness.
As I was rounding on these cases I began to reminisce about the Hospitalist movement and how this has impacted patient care during my career. I wanted to share my observations and thoughts on this topic as well as provide some facts and statistics.
Some History
I am now entering my 27th year in medicine. Many of the current practicing physicians don’t know nor ever experienced how we took care of patients prior to 1996/7. Let me paint the picture. My day started early in the morning and my first patients encountered were the ICU cases at the hospital. I would actually start in the radiology department and viewed all inpatient imaging tests on the x-ray Rolodex machine. From there I went to the sickest patients (ICU or step down unit). I then maneuvered through the hospital until all patients were seen. I actually had to cover two location facilities BEFORE I entered the medical office to see my full book of office patients. Charts were actually paper files. I carried a beeper for phone pages and I had a fancy Motorola car phone with a hand free voice and speaker system feature.
Throughout the typical office day I might get called by one of the facility emergency room physicians to admit an acutely sick patient. The ER doctor triaged and evaluated the patient and determined admission was warranted. Often upon discussion we would agree that with initiation of therapy a next day office evaluation could appropriately save the patient a hospital stay. Sometimes I needed to go to the emergency room and consult there to make a final triage decision. It was a very collegial process. I might also admit an acutely ill patient directly to the hospital from my office, with either hand written orders in hand or give verbal nurse intake orders.
I would get urgent calls from the floor/ward nursing staff if a patient’s condition warranted my assistance. Most evenings I would have to stop back at the hospital to check on the sicker or unstable patients and outstanding test results. Obviously this was a very rough life for your doctor- but the care burden was optimized for this setting. We doctors partnered with other private physicians of similar training in order to share night coverage and weekend duties. We were less efficient in the office setting BUT we actually made the admissions process very efficient for patients. We also truly knew our patients. This meant with a truly traditional and wholesome physician patient relationship the doctor wasn’t surprised by the need for admission and could transition care smoothly and effectively.
In the pre-Hospitalist era, hospital admissions were cared for by the doctor community. We shared equally in rotation for admitting patients including those considered non pay or indigent. This fact is not commonly known and is one of the arguments for Hospitalist arrangements with hospitals. The hospitals claim they need these doctors (and the subsidies paid to them are justified) in order to care for the “indigent” and uninsured.
A recent review regarding the Era of Hospitalists indicated that 1996 was the year that the concept of utilizing a full time on site physician to attend patients in the hospital took off. It appears to have been fostered as a concept by Dr. Robert Watchter MD who is in academic medicine and apparently motivated by hospital safety according to his background. He now is writing and assisting policy in the area of Telemedicine- see the AND NOW section below.
What Happened?
It is imperative for the public to understand why it is they rarely get to see their primary/main doctor at the most sick moment of their care. All of the prior mentioned tradition of practical inpatient care went out the window when Medicare’s fixed payment scheme called the Relative Value Unit payment system was made the law of the land. That legislative decision started the price-fixing of physician reimbursement for Medicare patients. Doctors were paid based upon a (mostly) time labor analysis for their services and care. The RVU program didn’t account for my drive time nor holding the phone overnight time and on call commitment. The scheme ignored my office overhead expense into the inpatient admission and patient rounding codes. With the passage of the 105th Congress Balanced Budget Act, it became illegal for physicians to collect any additional payment beyond what the program stated was the payment rate.
The RVU payment structure and its price-fixing caused physicians to leave their hospital duties and allow the Hospitalist Model to move forward as “common accepted practice”. There were less than 1000 hospitalists in 1996 and as of 2014 there were over 44,000. Based upon the number of new hospitalists being introduced into my local hospital system in the past few years I wouldn’t be surprised if that number isn’t 80,000 as of the end of 2019.
As the Hospitalist movement took hold, it only took a few years to then witness the end of specialty admissions even for specialty related acutely ill patients. Now routinely acute kidney stones, joint fractures, acute abdomens and even heart attack cases are turfed from the Emergency Room to the Hospitalist team for admission and primary management. Never mind that virtually all Hospitalist by definition are Internal Medicine, Pediatric or Family Medicine doctors and so can’t do anything therapeutically to treat your blocked painful (extreme) ureter, acute appendicitis, hip fracture or your life-threatening blocked coronary artery!
And Now?
The main positive claim made by the Hospitalist Model is that using Hospitalists resulted in lowering of patient’s length of stay. Improving this metric might be indirectly correlated to safety and outcomes but to my knowledge that isn’t proven. Given that hospitals get paid a flat rate by Medicare and insurers based upon a scheme called Diagnosis Related Group billing, the lower length of stay actually means higher profit and lower expense. Less days in the facility means more opportunity to fulfill a new payment transaction (admission) noting the fixed cost of running the hospital.
According to a graph documented on Statista, the average Hospital length of stay in 1993 was 7 days. It had dropped significantly by 1997 to 6.1 (recall the Hospitalist Model only had 1000 doctors then). It plateaued to 5.4 days by 2009 and has hovered between 5.4 to 5.5 since. If you assumed 100% correlation of improved lengths of stay to hospitalist doctors, this means growing the hospitalist physician population by 40 times resulted in one half a day improved hospital bed turn around time per admission on a national average!!
In a more recent review of the hospitalist situation I have learned that hospitals are struggling with a flattening of work output from their hospitalist physicians. In fact it turns out that there has been a dramatic increase in subsidizing expenses to support hospitalist with ancillary staff (meaning physician assistants, registered Nurse Practitioners, advanced Registered Nurses etc). I find this ironic because I was very vocally opposed to the system pushing your primary doctor out of your hospital attendance. Knowing the Hospitalist by definition wasn’t your primary doctor it was obvious they couldn’t be efficient in total patient care. This same recent review article states now the administrators of hospitals are backtracking on the biggest pro for staffing with hospitalists. Perhaps the whole reason for the hospitalist movement (24/7 in house physician attendance) isn’t necessary!!!
Their solution? Get this- Telemedicine! A quote from the article goes like this; “Maybe we don’t need physicians to be in the hospital 24/7 if we have access to telehealth, or a partnership with the emergency department, or greater use of advanced care practice providers,” Ms. Himebaugh said. Holy smokes- wasn’t that what my pager and phone were for back when I functioned for my patients as a traditional Internal Medicine Physician?
Most doctors in the community have electronic health record systems capable of uploading current clinical content data files. Despite a massive investment in the Epic Hospital health record our community has yet to have a bidirectional health data sharing interface. This means the emergency room doctors and the hospitalists do not receive current outpatient medical records. This data is vital to understanding where a patient is in their disease story. They are not privy to the most recent care prior to presenting to the emergency department. Nearly always I find medication errors on my patients admitted due to this lack of communication. I never receive a call from the various hospital sanctioned Hospitalist groups unless I have initiated the request to communicate. I have had to show emergency room nurses where to look on the Epic dashboard to see potential care locations the patient has had (including my office location). The staff literally don’t know they can click a folder tag and see all electronic documentation locations a patient has experienced by date and location. Pharmacy contacts, any other electronic record providing doctor or clinic shows up in this readily available link.
What Am I Witnessing?
From my vantage point of walking on both sides of the hospital signpost, I continue to have a re-enforced opinion that the Hospitalist Model has been an overall failed project. We certainly haven’t seen the cost of health care go down. From other recent data we have more Hospitalists now than Emergency Room doctors in the United States. If we assumed most of Hospitalist are Internists, the 44,000 Hospitalist represent 38% of the total Internal Medicine workforce! So in effect we have 38% of our Internists whose full time job is only rounding on hospitalized patients. It is an interesting side-note to see that over 32% of our total health care dollars go to hospitals. There are only 254.7 beds per 100,000 people in the state of Florida. This likely represents a national average.
Hiring all these Hospitalist hasn’t improved patient access nor physician productivity. I have interviewed several hospitalist in our community. Despite salaries that often exceed the full time outpatient physician equivalent, hospitalists work 2 weeks per month and have 2 weeks off.
Due to the shift work nature of hospitalist medicine, it is common to transfer primary attendance duty several times in a week long stay. Having to relearn a patient case takes much more physician time to manage the patient then continuity care. Trends can’t be appreciated whether they be favorable or unfavorable. Order error potential is massively higher with more and more cooks in the kitchen. The paradox of the hospitalist model in actual action is one of more fragmented and potentially erroneous care. This is happening when a person is most vulnerable in their physiology.
There is patient conflict of interest built into the currently working hospitalists model. Due to exclusive labor contracts and hospital controlled/employed physicians, hospitalists are incentivized to admit patients. The benefit of admission versus outpatient management for many patients is a marginal call. Admission has become the path of least resistance for the emergency room. I have never seen a ER hospitalist consultation in the department where a home disposition with close followup with the patient’s primary doctor was the course of action.
The model cuts both ways impacting outpatient physician care and decisions. Knowing the hospitalist is on standby and that personally directing a patient admission is a difficult path, there is little reason the office physicians will avoid directing their acute illness appointment requests to the emergency room. Problems that could be professionally and efficiently evaluated in the office are turfed to an already overfull emergency ward. It would be very interesting to see statistics on the rate of ER admissions pre-Hospitalist movement to post.
What Is One To Conclude?
None of the theory has been born out in practice.
The hospitalist model in theory would streamline patient hospital care. Going further in theory it would improve physician efficiency, disease management and patient outcomes. It would provide for better patient communication and wholesome transitions of care in and out of the hospital. With improved efficiencies the quality of life for your physicians would improve and health costs would improve. Since technology should improve more outpatient and home health care access the need for hospital growth and thus hospital employed doctors should go down.
Another conclusion is that the Inpatient Advocate Service™ is important to patients who want truly comprehensive physician care and a true doctor patient relationship. It assists the hospitalist model to achieve better patient communication and ease patient confusion when one is admitted to a hospital bed. The program is paid for directly by patients and isn’t under the control of any third party administrator or payer. It represents what can be possible if we allow doctors to function without conflicting interests that have been forced on unwitting Americans. It may well be a service offered by more doctors especially if arbitrary payment incentives are changed by Medicare that force Hospitals to end their physician employed consolidation trends.