Skip to content
Kestrel Ridge HealthKestrel Ridge Health

Reference

Ophthalmic Technician Staffing Ratios: What the Benchmarks Actually Say, and What They Cannot Decide

The published ophthalmic staffing benchmarks give you two usable numbers: roughly 4 to 8 full-time-equivalent staff per FTE provider at the 25th to 75th percentile, and roughly 3 to 3.5 patients worked up per technician per hour. Both are sound for the job they were built for, which is sizing an annual headcount budget and sanity-checking payroll as a share of revenue. Neither can tell you whether next Tuesday afternoon's glaucoma session is staffed correctly, because neither one knows what is on that schedule. This article states where each benchmark came from, how old it is, and exactly which decisions it can and cannot settle.

The three numbers everyone quotes, and where they came from

4 to 8 FTE staff per FTE provider

This range appears in CRSToday’s “What Is the Right Ratio?” (October 2016), attributed to American Academy of Ophthalmology benchmarking data and reported as the 25th to 75th percentile band.

Two things about this number are routinely lost in retelling. First, it counts all staff, clinical and administrative, not technicians alone. If you are trying to size a technician team specifically, this range does not answer your question. Second, a 25th-to-75th-percentile band is a description of what practices do, not a statement of what works. Half of all practices sit outside it, and the article gives no evidence that the ones inside it perform better.

3 to 3.5 patients per technician per hour

Also from the CRSToday piece, attributed to American Society of Ophthalmic Administrators technician benchmarking studies conducted in 2009 with a follow-up in 2015. The article is explicit that this assumes “a balanced schedule combination of long and short exams,” and notes that technicians in retina and oculoplastics may work up more per hour.

The conditional clause is the whole point and it is almost always dropped. The number is an average across a mix. It tells you nothing about a session that is not that mix, which is most sessions.

1 physician to 1.26 technical staff

This is the most-repeated ratio in ophthalmology practice management, and it is the oldest. It originates in Review of Ophthalmology’s “Benchmarking A Better Payroll” by Sandra E. D. McGraw and Kathryn E. Kocevar, published 15 January 2003, drawing on 2002 ASOA benchmarking survey data.

That data is from 2002. Since then ophthalmology practices have adopted electronic health records, added substantial diagnostic imaging volume, and changed their scribing arrangements. The 2002 ratio predates all of it. It is still quoted as though it were current.

What the benchmarks are genuinely good for

Ratios are the right tool for a small, specific set of decisions.

  • Annual headcount budgeting. If you need a defensible starting number for next year’s technician FTE line, a ratio applied to projected provider FTE is a reasonable first pass.
  • Detecting a gross outlier. If you are running 12 staff per provider or 2, the benchmark is telling you something real and you should go find out what.
  • Payroll as a share of revenue. Ophthalmology Management’s “Do I Have the Right Number of Staff?” uses gross wages at roughly 25 percent of practice revenues as a check. That is a financial control, and it works as one.
  • Arguing with a physician about a hiring request. An external benchmark is a neutral third party in a conversation that otherwise runs on anecdote.

What ratios structurally cannot decide

A ratio has one input, provider count, and one output, staff count. Anything that varies while provider count stays constant is invisible to it. In an ophthalmology clinic, almost everything varies while provider count stays constant.

It cannot see appointment mix

Ten patients is not ten patients. A ten-patient session of postoperative cataract checks and a ten-patient session of new glaucoma workups with visual fields, OCT, and pachymetry consume very different amounts of technician time. Both are ten patients. A ratio sees ten patients.

It cannot see diagnostic testing load

Diagnostic testing is where technician minutes actually go, and testing load varies by subspecialty, by provider habit, and by individual patient. The CRSToday piece quotes Maureen Waddle of BSM Consulting directly on this: she states that despite the existence of employees-per-physician benchmarks, she does not believe it is the best way to plan a staffing model, citing specialty type, diagnostic test volume, and provider type as material factors.

Waddle made the same argument at length in the 2011 Ophthalmology Management article, proposing that staffing be modelled from three work units, number of office visits, number of diagnostic tests, and number of surgeries, rather than from provider count. That article is now fifteen years old and remains the strongest published statement of the case.

It cannot see credentialing constraints

Technician minutes are not fungible. A COA, a COT, and a COMT are different credentials with different scopes, documented by IJCAHPO, which also certifies ophthalmic surgical assisting, ophthalmic ultrasound biometry, diagnostic ophthalmic sonography, and ophthalmic scribes. Site privileges, device competencies, and hour restrictions layer on top. A ratio that says you need 6.4 technicians has said nothing about whether the six you have can legally and competently do the specific work on Tuesday’s schedule.

It cannot see the variance it is averaging over

This is the deepest limitation. A ratio is a mean. Staffing to the mean means being right on average and wrong most days, over on the light ones and under on the heavy ones. The cost of being wrong is asymmetric: an overstaffed session costs you a technician’s hourly wage, and an understaffed session costs you patient wait time, provider idle time, technician burnout, and sometimes a rescheduled patient.

For context on that hourly wage, O*NET reports a 2025 median of $21.91 hourly for ophthalmic medical technicians, with 78,800 employed as of 2024. The BLS Occupational Employment and Wage Statistics for May 2022 put employment at 66,060 with a mean annual wage of $42,480 and a 10th-to-90th-percentile band of $29,830 to $58,970. The point of the wage figure is not that overstaffing is cheap. It is that the two failure modes have different units, and a single ratio cannot balance them.

The practical division of labour

The honest framing is that ratios and forecasts answer different questions and you need both.

Which staffing question each tool can answer
QuestionRight tool
How many technician FTEs should be in next year’s budget?Ratio benchmark
Is our payroll out of line with comparable practices?Ratio benchmark, plus payroll as a share of revenue
Are we structurally over or understaffed as a practice?Ratio benchmark, as a first flag
How many technician minutes does Tuesday’s 2pm glaucoma session need?Session-level forecast from your own historical data
Which of our five sites is short next week?Session-level forecast plus an allocation model
Can Maria cover that session given her credentials and site privileges?Explicit constraint rules, not a ratio and not a forecast

What to do with this

If you are using ratios today, three changes make them more useful without adopting anything new.

  1. Date every benchmark you quote. Write the vintage next to the number in your own documents. A 2002 ratio and a 2015 throughput study are not the same evidence.
  2. Separate technicians from total staff. The 4-to-8 band is all staff. Build your own technician-only ratio from your own payroll data before comparing yourself to anything external.
  3. Look at the spread, not just the mean. Pull twelve months of realized technician hours by session from your practice management system and plot the distribution. If the spread is wide, your ratio is averaging over a problem, and you now know how large that problem is in hours.

That third step is the one that changes the conversation, because it is measured on your own operation rather than borrowed from a survey. It is also the input a forecast needs. If you want to see how session-level forecasting differs from ratio staffing in method, Kestrel Ridge Health documents its approach on how it works and states its current validation status plainly on evidence.

Sources

Related