Anyone who has worked in home-based care knows documentation takes time away from the visit. What is surprising, once you go looking for it, is how little of that experience has been quantified — and how much of what circulates as evidence turns out, on inspection, to be borrowed from a different setting entirely.
This page sets out what is actually established, what is not, and where the widely repeated figures come from. We publish it because we are building documentation tools for this workforce and we would rather state the evidence plainly than repeat a number we cannot defend.
The one number the federal government publishes
Every Medicare-certified home health agency completes OASIS assessments. Because OASIS is a federal information collection, the Centers for Medicare & Medicaid Services is required under the Paperwork Reduction Act to estimate what it costs in staff time, and to publish that estimate in the rule that authorizes it.
The full set of estimates, by assessment type:
| Assessment | Estimated clinician minutes |
|---|---|
| Start of Care | 57.3 |
| Resumption of Care | 48.0 |
| Discharge | 40.2 |
| Follow-Up | 11.1 |
| Transfer of Care | 6.6 |
| Death at Home | 2.7 |
Source: CMS, CY 2023 HH PPS Final Rule. Wage inputs are drawn from BLS Occupational Employment Statistics for May 2020, so the dollar figures below understate current cost.
Aggregated across all Medicare-certified agencies, CMS puts the annual burden of OASIS assessments at roughly 14.6 million clinician hours and $1.17 billion in clinician time. Those totals are the sum of the per-assessment-type figures CMS publishes; CMS itself publishes the components rather than the total.
What that number does not include. The CMS estimate covers the OASIS instrument and nothing else. It does not cover visit notes, plans of care, coordination notes, communication with families, supervisory documentation, or anything an aide records. A clinician’s real documentation day is some larger number, and it is a number nobody has published.
What has never been measured
We could find no peer-reviewed study, and no AHRQ, ASPE, or HHS publication, measuring total documentation time for home health clinicians in the United States — not per visit, not per day, not as a share of the working week. The only direct measurements that exist are European, small, and now several years old.
| Study | Finding | Sample and year |
|---|---|---|
| Home healthcare nurses, Finland | Median 50 minutes of documentation per 8-hour shift (range 14–170); 38% of the shift in direct patient contact | 18 nurses, 196 observation sheets, data collected autumn 2017 |
| Community nurses, Netherlands | 8.0 hours per week on clinical documentation plus 3.6 hours on organizational documentation, self-reported | 195 respondents, surveyed June–July 2019 |
| Home care staff, Norway | Registered nurses spent 32% of time on direct patient care against 28% on administration and documentation | 20,964 observations, three rural municipalities |
These are the strongest direct measurements available anywhere. All three are observational or self-reported, all three are non-US, and the newest underlying data is from 2019.
Note what the Finnish study implies: roughly a tenth of the shift, not a third. The frequently cited claim that nurses spend about a third of their time on documentation comes from hospital and inpatient time-and-motion research, where the work, the systems, and the staffing are different. Carrying that figure across to home care is an assumption, not a finding, and we do not make it.
One more claim that does not survive checking. It is often said that OASIS made assessments longer. The one peer-reviewed study we could locate examining that question found the opposite — that reported time to complete an assessment with OASIS did not differ from a comparable assessment without it. That study is more than twenty years old and should not be treated as current. But it means the claim is unsupported rather than merely unproven.
The scale this sits inside
For context, the most recent MedPAC report to Congress records 12,234 Medicare-participating home health agencies serving roughly 2.7 million fee-for-service beneficiaries, at $16.0 billion in Medicare spending. The same report notes that in-person visits fell to 8.3 per 30-day period, down from 10.2 in 2020.
That last figure is worth sitting with. Fewer in-person visits per episode means each visit carries more of the clinical picture — and each minute inside it is worth more. Documentation time does not come out of an abstraction. It comes out of that visit.
Why we are publishing this rather than a savings claim
The obvious thing for a company building AI documentation tools to say is that it saves clinicians time. We are not going to say that, because it has not been demonstrated. The strongest study of ambient AI documentation to date — a controlled evaluation of 112 clinicians against 103 controls — found no statistically significant difference in EHR-related metrics. That was ambulatory care rather than home care, but it is the best available evidence and it is null. Vendor claims of eighty and ninety percent reductions circulate without published methodology, denominator, or sample, and we would not repeat them.
What follows from all of this is not that documentation burden is imaginary. The federal estimate is real, the clinicians describing the problem are describing something real, and the absence of measurement is itself the finding. It means that anyone proposing to reduce documentation burden in home-based care is proposing to change a quantity that has never been established — and that establishing it is a prerequisite, not a footnote.
Measuring it is part of what our 90-day pilot with a Medicare-certified home health partner is designed to do. We would rather report a number we collected than repeat one we did not.
Common questions
How much time does home health documentation take?
The only federal estimate covers the OASIS assessment instrument alone. CMS puts a single OASIS-E Start of Care assessment at 57.3 clinician minutes, and the annual total across all Medicare-certified home health agencies at roughly 14.6 million clinician hours and $1.17 billion in clinician time. Visit notes, care plans, and care coordination are not included in that figure.
Has anyone measured total documentation time in US home health?
No. No published United States study measures total documentation time in home-based care. The available time-and-motion evidence comes from European studies conducted in different health systems, and does not transfer cleanly to the US regulatory context.
Did electronic assessment make home health documentation take longer?
The claim does not survive checking. A study in Home Health Care Services Quarterly reported that the time to complete an assessment with OASIS did not differ from the time to complete a comparable assessment without it.
Sources
- Centers for Medicare & Medicaid Services, Medicare Program; Calendar Year 2023 Home Health Prospective Payment System Rate Update, Collection of Information Requirements. Federal Register 2022-23722.
- Vehko T. et al., time-motion observation of home healthcare nurses, Scandinavian Journal of Caring Sciences, published December 2019 (data collected 2017). PMC7754451.
- De Groot K. et al., “Nurses’ perceptions of the administrative burden…”, BMC Nursing, January 2022 (survey 2019). BMC Nursing 21:22.
- Time-use study of home care staff in three Norwegian municipalities, International Journal of Environmental Research and Public Health, October 2021. IJERPH 18(19):10504.
- Medicare Payment Advisory Commission, Report to the Congress: Medicare Payment Policy, Chapter 8, Home Health Care Services, March 2026. MedPAC.
- Study of OASIS and assessment time, Home Health Care Services Quarterly. PubMed 29016255. Twenty-two years old; cited here only to show that the opposite claim is unsupported.
Related
How long does an OASIS-E assessment take?The full CMS per-time-point table, and where the 57.3 minute figure comes from. The direct care workforce, in numbersWages, turnover, and the projection that ranks first among all US occupations. What the evidence shows about technology for dementia caregiversThe Cochrane review found little or no effect. That is the starting point, not the objection.Published July 31, 2026. Corrections to rupakshah@pyramid.care.