A practice administrator hires a new medical assistant, spends eight weeks training them, and watches them hand in a resignation letter five months later. The reasons an employee like that gives for leaving are rarely simple. Pay might be part of it. So might the schedule, the manager relationship, a lack of any clear path forward, or the day-to-day reality of the job itself.
That complexity is exactly why retention is hard to fix with a single lever. But it does not mean practice leaders are powerless. Research on healthcare turnover consistently identifies a specific set of factors, and one of them, workload design, is something a practice administrator can actually influence without waiting on a budget cycle.
A 2025 study of nurses' reasons for leaving the profession, published in the International Journal of Nursing Studies Advances, identified five recurring factors: compensation and fairness, the psychosocial work environment, career development constraints, non-clinical workload, and support and safety. Compensation and fairness explained the largest share of variance in the model, followed by the psychosocial work environment, meaning pay concerns and workplace culture both outweighed workload on their own, but none of the five factors stood alone.
A separate NCSBN 2024 National Nursing Workforce Study, the most recent completed edition of NCSBN's ongoing workforce survey, found stress and burnout selected as the root cause by roughly 41.5% of nurses reporting an intent to leave the workforce, followed by workload, understaffing, and inadequate salary. NCSBN's 2026 survey launched in March 2026 and is still collecting responses through September 2026, so the 2024 figures remain the latest published results as of this writing.
A large cross-sectional study of nurses who had already left health care employment found the top contributing factors were planned retirement (39%), burnout (26%), insufficient staffing (21%), and family obligations (18%).
Across this research, workload consistently shows up as one of several meaningful drivers of turnover, alongside compensation, career growth, management, and scheduling. It is rarely the only reason someone leaves. It is, however, one of the few reasons a practice can address directly and relatively quickly, without a compensation overhaul or a change in market conditions.
Turnover cost is easiest to see at hospital scale, where the numbers get tracked closely. The 2026 National Health Care Retention and RN Staffing Report puts the average cost of replacing one hospital-employed registered nurse at roughly $60,090, with hospitals losing an average of about $5.19 million a year to RN turnover. Those figures describe hospital RNs specifically, not independent-practice medical assistants or front-desk staff, so they should not be read as a direct benchmark for a smaller practice. The underlying cost categories, recruiting, training time, lost productivity during ramp-up, and the burden on remaining staff, apply at any scale, but the dollar figures themselves do not transfer directly.
What does transfer is where the strain concentrates. An MGMA Stat poll from May 2026 found that while overall practice-level turnover has stabilized somewhat, 28% of medical group leaders still reported higher turnover than the year before, concentrated in medical assistants, nursing and clinical support staff, and front-desk and administrative roles. Those are exactly the roles most likely to absorb extra work when a practice is short-staffed elsewhere.
Role creep happens gradually. A medical assistant who is good at troubleshooting the EHR becomes the unofficial IT contact. A front-desk lead who handles one insurance appeal well becomes the default person for every appeal. None of this shows up in a job description, and none of it gets reversed on its own, because each individual addition feels small at the time.
The cumulative effect is a role that no longer resembles what the person was hired to do, doing work that may not be the best use of their training, while the tasks they were actually hired for get squeezed into less time. That mismatch is one contributor to burnout and turnover, separate from raw workload volume. An employee can be busy and still feel like they are failing at their actual job, because so much of their day is spent on things outside it.
An employee can be busy and still feel like they are failing at their actual job, because so much of their day is spent on things outside it.
Most conversations about documentation burden focus on physicians, and for good reason. Physicians have long spent roughly two hours on EHR and desk work for every hour of direct patient care, a pattern documented in a 2016 time-and-motion study that later research continues to confirm. AMA's most recent national data shows burnout affecting 41.9% of physicians in 2025, with meaningful variation by specialty.
But documentation and administrative burden do not stop at the physician. Nurses, medical assistants, and administrative staff absorb their own version of it, chart prep, coding support, insurance documentation, and follow-up paperwork that expands to fill whatever time is left after direct patient tasks. When that burden grows faster than the team's capacity to absorb it, workload becomes one more pressure point on top of whatever else is already driving someone toward the door.
This does not require a full reorganization. It starts with a few concrete steps most practices can take without waiting on a new hire or a compensation review:
Workload design will not fix every reason someone leaves a practice. Compensation, scheduling, management, and career growth all matter, and a practice serious about retention should look at all of them. But workload is the lever a practice administrator can pull without approval from finance or a full leadership review, which makes it worth taking seriously on its own.
Where documentation burden specifically is driving the strain, Clinical Informatics support, in-person or AI-enabled virtual scribes with human oversight, or staffing help where a role genuinely needs a dedicated person, is one way to relieve that load without asking already-stretched staff to keep absorbing more. The Clinical Informatics Playbook walks through how to diagnose that kind of capacity gap before deciding what kind of support fits.
That same instinct to hire is often slower and more expensive than fixing the capacity problem underneath it, a question Do You Really Need to Hire? Rethinking Healthcare's Staffing Reflex takes head on.
Research points to several recurring factors, including compensation and fairness, career development, management and workplace culture, and workload. No single factor explains most turnover on its own, but workload is one of the few a practice can influence directly and relatively quickly.
Start by mapping what each role actually does day to day, separating documentation and administrative work from clinical work where possible, and asking staff directly about workload and role fit rather than relying on general satisfaction surveys alone.
Role creep is when a job quietly absorbs tasks it wasn't originally meant to include, until the employee is doing work outside their training while the tasks they were hired for get squeezed out. It's a distinct, often-missed contributor to burnout and turnover.
Nurses, medical assistants, and administrative staff absorb their own version of documentation burden, chart prep, coding support, insurance paperwork, and it becomes one more pressure point on top of whatever else is already driving someone toward leaving.
According to MGMA's May 2026 data, turnover is currently concentrated among medical assistants, nursing and clinical support staff, and front-desk and administrative roles, the same roles most likely to absorb extra work during a staffing gap.
Rarely. Research consistently shows compensation is one of several factors, alongside workplace culture, career development, and workload, so pay increases without addressing workload or role design often don't fix the underlying problem.
It's how work is actually distributed across roles, day to day, as opposed to what a job description says. Poor workload design is one of the few retention levers a practice administrator can influence directly, without a compensation overhaul.
Map what each role actually does versus its job description, and ask current staff directly about workload and role fit rather than relying only on general satisfaction surveys, which often don't surface workload issues specifically.
Ask specifically about workload and role fit, not just overall satisfaction. General exit questions tend to miss role creep and workload strain, which often sit underneath more generic complaints.
Burnout in one part of a practice tends to push work downstream. When physicians are overloaded, tasks often shift to nurses and administrative staff, spreading the same documentation and administrative strain across more roles.