Welcome to the early spring installment of “Ask the Expert!” Many of us have been presented with dashboards, charts/tables, and proposals relating to the organization and economics of our practices, questioning “Why do you need so many people?” and “Why can’t you cover all these locations?” Many such inquiries, especially when coming from unenlightened administrators, can seem counterintuitive or nonsensical. No big deal: This month we welcome Dr. Amr Abouleish, an expert on practice management and perioperative medicine, who will educate us on a proper approach to staffing.
Amr, thank you for joining us. Please describe your current position and responsibilities.
At my university, I am full-time clinical faculty specializing in pediatric anesthesiology. I have administrative roles as chief of pediatric anesthesiology as well as chair of our promotions committee. Outside my university, I am secretary of the Texas Society of Anesthesiologists and am on several ASA committees, including the Committee on Anesthesia Care Team and the Committee on Economics. I also represent ASA as an alternate member of the AMA/Specialty Society RVS Update Committee.
How did you get interested in practice management?
Around the time that I started my MBA program (mid-1990s, classmates with Asa Lockhart), our medical school started an initiative called “Mission Based Management.” I was appointed to represent anesthesiology on the clinical mission workgroup. The workgroup focused on units per full-time equivalent (FTE) as the core measurement, to be compared to MGMA surveys. In 1995, my chair and I co-wrote a white paper on why “units per FTE” does not work for measuring anesthesia clinical productivity. This manuscript started off 25+ years of work on anesthesia productivity measurements, anesthesia departmental economics, and staffing models. This includes being a member and previous chair of the ASA Committee on Practice Management.
How should a practice ideally structure itself?
This is a $64,000 (or more!) question, usually asked by facility administrators or medical school deans, but not anesthesia practices. The reason groups don’t usually ask this question is because anesthesiologists understand that the primary determinants of daily staffing needs are the number of sites to be covered, the hours of operation, and the location of those sites. For most groups, their staffing system is based on experience, safety, geographic limitations, and the type of staff they have. However, since the current economic reality is that health systems and medical schools have to provide funds to anesthesiology groups/departments for staffing, these entities are going to be delving into personnel costs.
An aside: These payments should not be called “stipends,” which has the connotation of a “handout.” Group support is really to cover uncompensated (or poorly compensated) clinical commitments. The payments have increased recently as facilities want more non-OR anesthetizing locations sites covered, while these are not as financially lucrative as OR sites. In academic departments, payments have increased due to an escalating number of sites to be covered, without a concomitant increase in residents. So, new sites will be covered via anesthesiologist-only care, or anesthesia care team with CRNA/CAA, both of which are more expensive than supervising residents.
Back to your original question: Understand that there are staffing models and then there are staffing models. An illuminating analogy is what type of car is ideal to drive. The answer will be different in cost-minimization versus cost-benefit analysis. For cost-minimization, it doesn’t matter what the journey is like, you only care about reaching the endpoint, and all endpoints are equal. In that case, you will obtain the cheapest car (e.g., a Chevrolet Spark). It might not be the safest in an accident, it might not be the most comfortable, and it might not last long, but it will reach the desired destination. In reality, most people don’t drive this type of car; they choose something more expensive. Why? Because we believe that what happens on the journey is as important as the endpoint. We do think that safety, comfort, and reliability are important. We value the extras, so we spring for the Tesla.
So which anesthesiology group is most desirable, the Spark or the Tesla equivalent? If the former, then all you are doing is covering OR-based anesthesia with little other value provided. You may not have enough staff to cover emergencies or to react when extra help is needed in an OR. You won’t provide meaningful PACU coverage, and maybe won’t do many blocks. On the other hand, if you are the “Tesla group,” you provide comprehensive perioperative care (day and night), including optimization/evaluation via a pre-anesthesia clinic (PAT), postoperative pain blocks, PACU coverage, flexible staffing to cover additional unscheduled site(s), and involvement as physician leaders in hospital committee and administrative work.
In our 2010 article, we looked at a cost-minimization analysis for moving from MD-only to medical direction (ASA Monitor 2010;74:30-51). It was evident that this move would not necessarily save money and that there were unintended, hidden costs. Further, in academics, the staffing cost per hour was similar for an assistant professor and a CRNA/CAA.
So to answer your question, it depends. Often the most economical staffing model is a mixture of MD-only and medical direction. The reality of which way to go is based on who the group can recruit (really, who they can hire): anesthesiologists, CRNAs, or CAAs?
In staffing, what are the “hidden” or forgotten costs that you mentioned?
Often the facility administrator asking the question about your staffing is focused on why you need so many clinicians. For example, if you have only 10 sites to cover and you cover them at 1:2, why can’t you do it with five anesthesiologists and 10 CRNAs/CAAs total? To answer the question, you require more than a 10 minute informal meeting. Ask for an hour and bring handouts – you will need to explain many issues.
- Coverage of anesthesia sites:
- The anesthesiologist running your daily operations is the person who is called first to help out or cover something when a colleague is busy and can’t leave the patient. He/she also deals with administrative duties (schedules, relief assignments, movement of staff to meet demand), and may cover PACU and support PAT. For this anesthesiologist, I typically assign one room; some groups don’t assign any, but that may change during the day.
- For every 10 sites at a full-service hospital, I typically assign one anesthesiologist to work 1:1 with a resident or CRNA/CAA. The room can be complex (e.g., cardiothoracic or neonatal cases) or can be a non-complex surgery in a complicated, sick patient.
- 10-hour or 12-hour CRNA/CAAs working 40 hours/week do not work five days/week.
- Geographic challenges mean some sites cannot be covered at the same ratio as the OR due to physical distance.
- The number of sites doesn’t always lend itself to an efficient staffing ratio, e.g., five sites when you ideally cover 1:3.
- Non-anesthesia sites/duties are many, and you need to communicate the value of those. Examples include PACU, PAT, acute pain, chronic pain, ICU, emergency airway, and OB (if no dedicated staff).
- Not all staff are available for “first starts.”
- Late arrival staff, such as those starting shifts later in the day, on-call staff, post-all staff.
- Break CRNAs/CAAs are the most ubiquitous hidden costs overlooked by facilities. You will need one for every four (maybe five) CRNAs/CAAs. If they start at 11 a.m., they often are evening shift staff.
- Physicians unavailable due to non-clinical duties (administrative, educational).
- Paid time off means we need to account for vacation, meeting time, and sick time. The first two are calculated based on earned time. For the latter, estimate that you will have one FTE out sick (including FMLA) for every 20-30 FTEs on staff. If your staff is younger, use the more conservative 1:20 benchmark (increased maternal/paternal leaves).
For more information, see: Developing a Staffing Model: Estimating the Number of Anesthesia Providers (ASA Monitor 2013;77:10-3).
What is your view on using CRNAs or CAAs versus using both, or neither?
Anesthesia care is the practice of medicine, and I am a strong believer in anesthesiologist personally provided care, in the anesthesia care team (ACT) and also in teaching attendings and working with residents. CRNAs and CAAs are valuable members of the ACT, and for staffing models, they are treated the same. The compensation is comparable and their duties and responsibilities are the same.
How do you address those who claim that solo MD practice is prohibitively expensive?
I addressed this already, but allow me to elaborate. As mentioned, hidden costs of moving from MD-only to medical direction are not often considered. Let’s say you have 15 anesthesiologists in an MD-only group who take in-house call and cover three late rooms every day. The physicians are on call 2x/month and cover late rooms 4x/month. If we move to medical direction and reduce to seven anesthesiologists, the on-call burden increases to 4x/month, and the late-stay days increase to 6-7/month. This increased work will lead to a direct increase in staffing costs per anesthesiologist and indirectly to probable dissatisfaction and turnover. In addition, note that physician-only practices are more able to facilitate invaluable clinical services such as pain blocks as well as professional services such as peer-to-peer communications with surgeons.
Can we objectively assess group productivity? Is there a role for ASA units (aggregate or per FTE)?
As noted, my professional journey in medical economics began with this question. ASA units per FTE does not meaningfully compare individual or group productivity. This number is highly dependent on staffing model and OR utilization/workload. It also is very dependent on the anesthesiologist’s opportunity to bill ASA units. For anesthesiologists who work in areas that don’t produce ASA units (e.g., PAT, PACU, pain), or for those with administrative/educational duties, ASA units/FTE will be decreased. Further, an anesthesiologist who takes disproportionate in-house call will suffer as he/she will accrue fewer billable units in the evening hours, as compared to during the regular workday. Drs. Charles Whitten, Mark Hudson, and I authored a review article, which covers this topic in detail, as well as how to measure productivity in a group setting (Anesthesiology 2019;130:336-48).
Any parting words for our readers?
Basic marketing is a three-step process: identify the customer’s needs, develop a product to meet the needs, and communicate the products to the customer. This simple process is the best way to communicate the value you provide to the hospital and make sure the hospital understands what they need – especially the non-surgical anesthesia work.
Finally, my favorite final slide of any talk: “Healing is Art, Medicine is a Science, and Health Care is Business.” To be successful, you need to be good at all three.