Skip to content

Privacy controls

Essential

Required for security and remembering your privacy choices. These cannot be switched off.

Always on
AnalyticsMicrosoft Clarity and Google Analytics help us understand site use and improve the experience.
MarketingMeta Pixel helps us measure campaigns and understand whether our advertising is useful.
FunctionalGoHighLevel provides our waitlist form, demo survey, calendar and optional chat.
Blog

Caregiver Turnover Rate: Measure the First 90 Days Separately or You Will Fix the Wrong Thing

One number on the owner's dashboard says turnover is 60%. It hides the real story: most of the people leaving have been with you less than three months.

Nuraflow Editorial · Workforce and Compliance · November 14, 2026 · 6 min read

The owner asks what turnover was last year. Someone pulls payroll, counts the people who left, divides by headcount and announces a number. Everyone nods, the number goes in the board deck, and the agency spends the next quarter on a retention bonus for long-tenured caregivers who were never planning to leave. The caregiver turnover rate was real. The decision built on it was wrong, because a single annual figure blends two very different problems: people who leave in their first weeks, and people who leave after years.

This guide gives you a consistent formula, a separate 90-day measure that shows where the losses really are, and public benchmarks to compare against.

Why Caregiver Turnover Rate Deserves a Real Definition

Turnover is a workforce number, but it is also a care number. Each departure is a client who gets a new face, a schedule that has to be rebuilt and an orientation that has to be repeated. For personal care in particular, hiring never stops. The scheduling and matching guide explains why continuity matters to clients; turnover is the force working against it.

The first problem is that agencies calculate it differently from month to month. One counts people who never worked a shift; another excludes anyone terminated; a third divides by year-end headcount after a growth spurt. Write the definition down once and keep it.

The Basic Formula

Use the same structure the Bureau of Labor Statistics uses for separations: separations during the period divided by average employment during the period.

  1. 1.Define who counts as a caregiver. For example, every field employee in an aide, attendant or homemaker role who worked at least one paid shift.
  2. 2.Count separations. Everyone in that group whose employment ended during the period, whether they quit, were let go or simply stopped accepting shifts and were removed. Pick a rule for the last group, for example no shift worked in 30 days, and apply it every time.
  3. 3.Calculate average headcount. Add active caregivers at the start and end of each month and divide by two, then average the months.
  4. 4.Divide and multiply by 100.

Annual caregiver turnover rate = separations in the year ÷ average monthly headcount × 100.

Report three versions every time: total, voluntary (quits and job abandonment) and involuntary (terminations). They point to different fixes.

Benchmarks: What the Public Data Shows

BLS's Job Openings and Labor Turnover Survey publishes monthly quits rates by industry. For health care and social assistance, the not seasonally adjusted JOLTS quits rate ran between 1.6% and 2.5% a month during 2025, and between 1.5% and 2.4% in the months of 2026 published so far. Add up the twelve monthly quits rates for 2025 and you get roughly 24 quits per 100 employees over the year. That is a broad industry that includes hospitals and physician offices, so treat it as a floor for comparison, not a home care benchmark.

For the occupation itself, BLS Occupational Employment and Wage Statistics for home health and personal care aides (31-1120) put national employment at about 4.3 million in May 2025, with a mean hourly wage of $17.36 and a median of $17.21. Roughly 1.08 million worked in the home health care services industry, at a mean of $16.69 an hour, and about 2.27 million in individual and family services.

PHI's Direct Care Workers in the United States: Key Facts 2025 estimates 9.7 million total job openings in direct care from 2024 to 2034 once occupational transfers and labor force exits are included, and reports median annual earnings for direct care workers of just under $26,000, which it attributes to low wages and part-time hours. In other words, every agency is recruiting from the same constrained pool, and the competition for each caregiver will not ease on its own.

Benchmarks are useful for context. Your own trend is more useful for decisions.

Why the First 90 Days Need Their Own Measure

An annual turnover figure treats a caregiver who left after six days the same as one who left after six years. They are not the same problem.

Early departures usually trace to things the agency controls in the first weeks: a job that did not match what the recruiter described, too few hours, travel the caregiver did not expect, a first client who was a poor match, or no one calling to check in after the first shift. Later departures trace to pay, scheduling stability, supervision and life changes. If 70% of your separations happen in the first 90 days, a tenure bonus at year two cannot move the number.

So measure early turnover as a cohort, not a rate:

  1. 1.Group caregivers by hire month. Everyone who started in January is one cohort.
  2. 2.Track each cohort for 90 days from their start date.
  3. 3.Count how many separated within 90 days.
  4. 4.90-day turnover = separations within 90 days ÷ cohort size × 100.

Also count hired but never worked: offers accepted, onboarding started or finished, no shift worked. This is a recruiting and onboarding problem, and it disappears from most turnover reports because the person was never on the schedule. Track it as its own line.

A Worked Example

Illustrative figures for a personal care agency:

  • Average monthly caregiver headcount for the year: 180.
  • Separations during the year: 108.
  • Annual turnover: 108 ÷ 180 × 100 = 60%.

Now split it by tenure at separation:

  • Separated within 90 days of hire: 65.
  • Separated after 90 days: 43.

And by cohort:

  • Caregivers hired during the year: 120.
  • Of those, separated within 90 days: 65.
  • 90-day turnover: 65 ÷ 120 × 100 = 54%.
  • Hired but never worked a shift (not in the 120): 22.

The headline 60% suggested a general retention problem. The split shows that more than half of new hires leave in their first three months and another 22 people accepted offers but never started. Turnover among caregivers past 90 days is 43 separations on a much smaller at-risk group, a problem worth working on, but not the first one.

What to Do With the Split

Match the fix to the window:

  • Hired but never worked: shorten time from offer to first shift, keep in touch during onboarding, and give a confirmed first assignment date at offer.
  • First 90 days: describe hours and travel honestly at interview, match the first client carefully, have a supervisor call after shifts one, three and ten, and schedule enough hours to make the job worth keeping.
  • After 90 days: look at schedule stability, pay compared with local alternatives, recognition and how supervisors handle problems.

Record a reason code for every separation from a short fixed list, and ask departing caregivers directly. Then review 90-day turnover by recruiter, by source and by first client. Patterns show up quickly.

Calculation Mistakes That Distort the Number

Four errors show up again and again when agencies compare their turnover with someone else's:

  • Dividing by year-end headcount. A fast-growing agency looks better than it is; a shrinking one looks worse. Use the monthly average.
  • Leaving out job abandonment. A caregiver who stops answering the phone is a separation. Apply your inactivity rule consistently, or voluntary turnover will look low.
  • Counting office staff. Schedulers and nurses have different labor markets. Report field caregivers on their own line.
  • Counting rehires as new hires without a rule. Decide whether a caregiver who returns within 90 days is a continuation or a new cohort member, write it down, and apply it every month.

Report It Monthly, the Same Way

A one-page monthly view keeps the conversation honest:

  1. 1.Annual caregiver turnover rate, rolling 12 months, total, voluntary and involuntary.
  2. 2.90-day turnover for each hire cohort that has completed 90 days.
  3. 3.Hired but never worked, by month.
  4. 4.Top three separation reasons.
  5. 5.Open shifts caused by departures.

Nuraflow's Workforce Management is designed to keep one file from application through separation, including retention and separation records, and Reports and Analytics is designed to calculate measures like these from current source records rather than a payroll export rebuilt each month.

This article is general information. Retention practices and separation handling can carry employment law obligations that vary by state, so check policy changes with HR or employment counsel.

Founding Agencies

One Record, From the Referral to the Paycheck.

Bring the operating question behind the article and help test how the record should move through the agency.