Doctor’s Notes in America: The States Where a Sick Day Comes With Paperwork

Research

An analysis of 93,633 DrHouse telehealth visits finds a 20-point state-by-state gap in how often a visit ends with a doctor’s note.

Getting sick is universal. The paperwork that comes with it is not.

Among completed DrHouse telehealth visits, 27.9% of visits from Delaware ended with a doctor’s note. From Maine, 7.7% did. Same platform, same kind of appointment, and a twenty-point gap in whether the patient left with a document in hand.

With paid sick leave now law in twenty states and the District of Columbia, we analyzed 93,633 completed telehealth visits from February 29, 2024, through August 17, 2026, and put a number on the other half of the sick day: the paperwork.

What follows ranks 46 states and districts by how often a visit ends with a note, tests the obvious explanations, and says plainly which questions this dataset cannot answer.

Key Findings

• Among DrHouse telehealth visits, the share ending with a doctor’s note runs from 27.9% in Delaware to 7.7% in Maine. That is a 20-point spread, and the top rate is 3.6 times the bottom one.
• The most solid figures in the table are Maryland at 27.4% (2,115 visits) and Texas at 24.9% (16,406 visits, the largest and most precisely measured sample in the study). Both sit well above the overall rate of 21.4%.
• Maine, New Hampshire (7.8%) and Montana (8.1%) have the lowest rates, each under one visit in twelve. Connecticut (12.2%), Idaho (12.7%) and Minnesota (14.2%) come next, on larger samples.
• Patient age does not explain the gap. Recalculating every state against one common age profile leaves the ranking almost unchanged: the two versions correlate at 0.98, and no state moves more than eight places.
• The South runs highest by region at 22.5%, against 20.2% in the Northeast, 20.0% in the Midwest and 19.9% in the West. The spread between states is roughly eight times wider than the spread between regions.
• Places with a statewide paid-leave law come in two points lower on visit counts (20.1% vs 22.1%), but the gap is not statistically distinguishable when each state counts once, and the law states include both Maryland near the top and Maine at the bottom.
• Note counts tell a different story from note rates. Texas issued the most notes (4,085) and Florida the second most (2,542) because they send the most patients; by rate, Florida ranks 25th of 46.

What This Measures, and What It Doesn’t

Three things sound alike here, and they are not the same.

• The note rate is the share of completed DrHouse telehealth visits at which a doctor’s note was issued. It is the number in every ranking and table in this study. Delaware’s 27.9% means 109 of its 391 visits ended with a note.
• The note count is the raw number of notes, and it mostly tracks how many patients a state sends. Texas issued 4,085 notes, Florida 2,542 and California 1,320. By rate, Texas ranks sixth, Florida 25th and California 28th. A large state can issue the most notes while its patients are no more likely than average to walk away with one.
• Calling in sick is how often people miss work. No visit dataset measures that, and this one does not try. A state near the top of the table is not sicker and does not skip work more often; its patients came away with written confirmation of a visit more often. This is also not a list of where people fake sick days: the data records that a note was issued, not why anyone wanted it. It is a measure of paperwork, not illness, and every figure in it describes DrHouse visits, not the population of any state.

The Ranking

Forty-six places make the table: 45 states plus the District of Columbia. Five states had fewer than 100 visits over the period and are left out, because a rate built on a few dozen visits can swing on a handful of notes.

Those five (Alaska, North Dakota, South Dakota, Vermont and Wyoming) account for 335 visits between them, less than half of one percent of the dataset.

Beside each rate we show the age-adjusted rate, which is the number a state would post if it had the same age mix as the dataset as a whole, and a 95% confidence interval, which is the right way to read the smaller states.

Table 1. Doctor’s note rate by state, DrHouse telehealth visits, February 29, 2024 – August 17, 2026

#	State	Visits	Notes	Note rate	Age-adjusted	95% CI
1	Delaware	391	109	27.9%	27.0%	23.7–32.5%
2	Maryland	2,115	579	27.4%	27.2%	25.5–29.3%
3	Nevada	1,217	324	26.6%	27.3%	24.2–29.2%
4	Kentucky	939	248	26.4%	25.9%	23.7–29.3%
5	Pennsylvania	2,476	631	25.5%	25.1%	23.8–27.2%
6	Texas	16,406	4,085	24.9%	24.1%	24.2–25.6%
7	Hawaii	113	28	24.8%	23.5%	17.7–33.5%
8	New Mexico	310	76	24.5%	26.3%	20.1–29.6%
9	Oklahoma	896	218	24.3%	24.9%	21.6–27.2%
10	Missouri	1,252	292	23.3%	23.1%	21.1–25.7%
11	Alabama	1,334	298	22.3%	21.6%	20.2–24.7%
12	Mississippi	924	206	22.3%	22.8%	19.7–25.1%
13	Ohio	2,388	530	22.2%	22.2%	20.6–23.9%
14	Louisiana	964	213	22.1%	22.2%	19.6–24.8%
15	Oregon	195	43	22.1%	23.8%	16.8–28.4%
16	Tennessee	3,119	684	21.9%	21.8%	20.5–23.4%
17	Michigan	1,845	404	21.9%	21.8%	20.1–23.8%
18	South Carolina	1,949	419	21.5%	22.7%	19.7–23.4%
19	Washington	1,247	267	21.4%	21.0%	19.2–23.8%
20	New Jersey	3,499	740	21.1%	21.6%	19.8–22.5%
21	Virginia	2,672	561	21.0%	21.5%	19.5–22.6%
22	Iowa	539	113	21.0%	19.8%	17.7–24.6%
23	Georgia	3,712	777	20.9%	21.7%	19.7–22.3%
24	Rhode Island	212	44	20.8%	21.2%	15.8–26.7%
25	Florida	12,515	2,542	20.3%	20.3%	19.6–21.0%
26	Arizona	3,349	675	20.2%	20.2%	18.8–21.5%
27	Illinois	3,078	610	19.8%	20.1%	18.4–21.3%
28	California	6,747	1,320	19.6%	20.0%	18.6–20.5%
29	North Carolina	3,208	621	19.4%	19.9%	18.0–20.8%
30	Wisconsin	703	134	19.1%	19.2%	16.3–22.1%
31	New York	4,227	800	18.9%	18.6%	17.8–20.1%
32	West Virginia	315	59	18.7%	18.4%	14.8–23.4%
33	Kansas	1,022	191	18.7%	18.1%	16.4–21.2%
34	Utah	707	131	18.5%	16.9%	15.8–21.6%
35	Indiana	1,427	263	18.4%	18.9%	16.5–20.5%
36	Arkansas	203	35	17.2%	16.7%	12.7–23.0%
37	Massachusetts	446	75	16.8%	17.7%	13.6–20.6%
38	Colorado	1,647	270	16.4%	16.0%	14.7–18.3%
39	District of Columbia	269	43	16.0%	12.4%	12.1–20.8%
40	Nebraska	405	62	15.3%	15.4%	12.1–19.1%
41	Minnesota	932	132	14.2%	14.8%	12.1–16.5%
42	Idaho	323	41	12.7%	13.1%	9.5–16.8%
43	Connecticut	617	75	12.2%	13.3%	9.8–15.0%
44	Montana	173	14	8.1%	8.0%	4.9–13.1%
45	New Hampshire	115	9	7.8%	9.1%	4.2–14.2%
46	Maine	156	12	7.7%	9.4%	4.5–13.0%

Rates are the share of completed visits with a doctor’s note issued. Age-adjusted rates are standardized to the age profile of the full dataset. Confidence intervals are Wilson 95% intervals. Overall rate across all DrHouse visits: 21.4%.

The middle of the table is crowded. Nineteen of the 46 places sit within two points of the overall 21.4% on unrounded rates (North Carolina, twentieth, just misses the band), and a position in that zone is a rough guide rather than a verdict. Twenty-four places have intervals that exclude the overall rate, though with 46 comparisons two or three would do so by chance; eleven states clear a far stricter p < 0.001 threshold.

Small states can move several places with a year of new data, which is one reason the table will be refreshed annually.

The Top of the Table

Delaware leads at 27.9%, but on 391 visits its interval runs from 23.7% to 32.5%, and its rate has moved: 21.0% across 143 visits in 2025, then 30.0% across 203 in 2026. It belongs in the top group. Its exact position is soft.

Maryland is the anchor. Its 27.4% rests on 2,115 visits, it posted 26.8% in 2025 and 28.8% in 2026 (its smaller 2024 sample sat at the overall rate), and it is the second-strongest signal in the table after Texas.

Texas is the largest sample in the study, 16,406 visits or 17.5% of the dataset, and the most precisely measured: 24.9%, with a margin under a point either way. When the largest sample lands three and a half points above the overall rate, that is not noise.

Maryland, Nevada (26.6%), Kentucky (26.4%), Pennsylvania (25.5%) and Texas all differ from the overall rate at p < 0.001; Delaware comes in at p = 0.002. Hawaii (24.8%), New Mexico (24.5%), Oklahoma (24.3%) and Missouri (23.3%) complete the top ten, though Hawaii’s 113 visits make its rate the least certain in the group. The top ten together run at 25.2%, one note for every four visits.

Table 2. The ten highest and ten lowest note rates

Rank	Highest	Rate	Rank	Lowest	Rate
1	Delaware	27.9%	37	Massachusetts	16.8%
2	Maryland	27.4%	38	Colorado	16.4%
3	Nevada	26.6%	39	District of Columbia	16.0%
4	Kentucky	26.4%	40	Nebraska	15.3%
5	Pennsylvania	25.5%	41	Minnesota	14.2%
6	Texas	24.9%	42	Idaho	12.7%
7	Hawaii	24.8%	43	Connecticut	12.2%
8	New Mexico	24.5%	44	Montana	8.1%
9	Oklahoma	24.3%	45	New Hampshire	7.8%
10	Missouri	23.3%	46	Maine	7.7%
	Top 10 combined	25.2%		Bottom 10 combined	14.4%

The Bottom of the Table

Maine (7.7%), New Hampshire (7.8%) and Montana (8.1%) post the three lowest rates, each under one visit in twelve, and each on fewer than 200 visits, so their intervals overlap the states just above them. Connecticut (12.2%), Idaho (12.7%) and Minnesota (14.2%) come next on larger samples, and all three sit far enough below the overall rate to clear p < 0.001. Nebraska (15.3%), the District of Columbia (16.0%), Colorado (16.4%) and Massachusetts (16.8%) round out a bottom ten that combines to 14.4%.

On a map, northern New England and the northern Mountain West sit at the bottom, with Minnesota and Nebraska not far above. We won’t try to explain that here. What the data does show is that the bottom is not a one-year fluke: Connecticut was bottom-five in both 2025 and 2026 whichever way the smaller states are handled, and Montana’s 2026 rate, on 101 visits, was 7.9%.

It Isn’t the Ages

The first objection any careful reader should raise: doctor’s notes are a young person’s document. In our data, 38.7% of visits by 18-to-24-year-olds end with a note. For patients 65 and over, it is 2.5%. A state whose patients skew young should rank high for that reason alone.

So we checked. Every state’s rate was recalculated as if it had the age profile of the full dataset, using direct standardization across seven age bands. The adjusted ranking correlates with the raw one at 0.98. No state moves more than eight places. Five of the top six stay in the top six, with Texas slipping to eighth (24.1%) as New Mexico climbs to fourth (26.3%). The bottom three stay the bottom three.

The states that do move show the adjustment doing its job. The District of Columbia has the youngest patients in the study, 61% under 35 against 46% across all visits, and its rate falls from 16.0% to 12.4% once that is accounted for.

Utah drops from 18.5% to 16.9% for the same reason. Maine and New Hampshire have older patient mixes, and their adjusted rates rise to 9.4% and 9.1%. Still in the bottom three; the adjustment just hands last place to Montana, at 8.0%. Whatever separates Delaware from Maine, it is not the birthdays.

Table 3. Where age adjustment moves the number most

State	Raw rate	Age-adjusted	Change	Rank (raw → adjusted)	What drives the move
District of Columbia	16.0%	12.4%	-3.6 pts	39 → 43	18-to-24-year-olds are 25% of its visits, against 14% across all visits
Utah	18.5%	16.9%	-1.6 pts	34 → 36	Young patient mix: 53% under 35, against 46% across all visits
Hawaii	24.8%	23.5%	-1.3 pts	7 → 10	18-to-24s are 21% of visits; small sample (113 visits)
Iowa	21.0%	19.8%	-1.2 pts	22 → 29	18-to-24s are 20% of visits, against 14% across all visits
Texas	24.9%	24.1%	-0.8 pts	6 → 8	More under-18 visits than the full-dataset profile (8% vs 6%)
Alabama	22.3%	21.6%	-0.7 pts	11 → 19	Slightly younger mix; the biggest rank move in the table, because the middle is crowded
New Mexico	24.5%	26.3%	+1.8 pts	8 → 4	Fewer 18-to-24s than the full-dataset profile (10% vs 14%)
Oregon	22.1%	23.8%	+1.7 pts	15 → 9	Small sample (195 visits): three under-18 visits, two with notes, carry outsized weight once standardized
Maine	7.7%	9.4%	+1.7 pts	46 → 44	Older patient mix: 18-to-24s are 8% of visits
New Hampshire	7.8%	9.1%	+1.3 pts	45 → 45	Older patient mix: 10% of visits are patients 65+, against 6% across all visits

Positive change means the state’s rate rises when standardized to the age profile of the full dataset. Adjusted rates for all 46 places are in Table 1.

Regions: The South Stands Apart, but the States Matter More

Grouped into the four Census regions, the South comes out on top at 22.5%, ahead of the Northeast (20.2%), the Midwest (20.0%) and the West (19.9%). On visit counts, the South’s lead over each of the other three clears p < 0.001. Counting each of the 46 places once, the South’s 17 still run higher than the other 29 (p = 0.02). The other three regions cannot be told apart.

Table 4. Doctor’s note rate by Census region

Region	Visits	Notes	Note rate	Lowest reported state	Highest reported state
South	51,931	11,697	22.5%	District of Columbia, 16.0%	Delaware, 27.9%
Northeast	11,834	2,393	20.2%	Maine, 7.7%	Pennsylvania, 25.5%
Midwest	13,716	2,748	20.0%	Minnesota, 14.2%	Missouri, 23.3%
West	16,152	3,213	19.9%	Montana, 8.1%	Nevada, 26.6%

Regional totals include all 93,633 visits, including the five states too small to report individually.

Regions flatten the story, though. The South’s own places run from 16.0% in the District of Columbia to 27.9% in Delaware, and the Northeast holds both Pennsylvania at 25.5% and Maine at 7.7%. The spread between regions is under three points. Between states it is twenty.

Do Paid Sick-Leave Laws Explain It?

Twenty states and the District of Columbia now require paid sick leave or general-purpose paid leave, by the count of the compliance tracker GovDocs; other trackers land a state or two either side, depending on how the general-purpose laws in Illinois, Maine and Nevada are classified.

Some of those laws limit when an employer may ask for a note, and Connecticut’s is the clearest example: since January 1, 2025, employers covered by Public Act 24-8 may not require any documentation for paid sick leave, a doctor’s note included, according to the advocacy group A Better Balance and employer-side summaries of the law. Connecticut ranks fourth from the bottom at 12.2%.

That looks like a smoking gun until you check the timing. Connecticut’s rate did not fall after the rule took effect (11.2% in 2025, 13.0% in 2026), and the 41 visits that predate it, too few to report, show no sign of a higher rate. The other leave laws point in different directions. Nevada’s lets workers at employers with 50 or more staff use paid leave “without providing a reason,” and Nevada ranks third at 26.6%. Maine’s allows leave for any reason at employers with more than 10 workers, and Maine ranks last.

Across the 19 reportable places with a statewide law, 20.1% of visits ended with a note, against 22.1% in the 27 without one. That two-point gap comes mostly from high-volume Texas and Florida sitting in the no-law group. Counting each state once, it is not distinguishable from chance (p = 0.2), and the range inside each group is ten times wider than the gap between them. Nothing here says the law is what sorts the states. It does not prove the laws change nothing, either; it means this dataset cannot show it one way or the other.

Table 5. Statewide paid-leave laws and note rates

Group	Reportable states	Visits	Notes	Note rate	Lowest state	Highest state
Statewide paid sick-leave or paid-leave law in force	19	32,513	6,551	20.1%	Maine, 7.7%	Maryland, 27.4%
No statewide law	27	60,785	13,452	22.1%	New Hampshire, 7.8%	Delaware, 27.9%

Law group: Arizona, California, Colorado, Connecticut, District of Columbia, Illinois, Maine, Maryland, Massachusetts, Michigan, Minnesota, Nebraska, Nevada, New Jersey, New Mexico, New York, Oregon, Rhode Island, Washington. Alaska and Vermont also have statewide laws but fall below the 100-visit threshold. Missouri’s 2025 law was repealed effective August 28, 2025 and is not counted; Nebraska’s took effect October 1, 2025, part-way through the study period. Local ordinances (for example in Philadelphia and Pittsburgh) are not counted. Law list verified against GovDocs (February 2026) and Paylocity (June 2026) trackers and the relevant state statutes.

If It Isn’t Age, Region or Law, What Is It?

The honest answer: the data can rank the states, and it cannot say why they rank where they do. 

• The mix of jobs. A doctor’s note gets written because someone wants proof, usually an employer, and employers differ in what they ask for. A salaried employee working remotely can message a manager and take the day; someone on an hourly shift, an attendance-point system or a direct-care schedule may be expected to bring paper back. If documentation-heavy jobs cluster in some states, those states would post higher note rates. Our data does not record occupation or employer, so the study cannot test this. For scale, federal data measures absence rather than paperwork: in 2025, 3.2% of full-time wage and salary workers missed some work in a typical week, 2.2% of them for illness or injury, per the Bureau of Labor Statistics (an eleven-month average, since October’s survey was not collected). That is a different measure from ours, and the two should not be compared.
• Who books a telehealth visit, and why. This is a visit base, not a population sample. Who uses DrHouse in each state depends on how people find the service there, and a state where more visits are prescription follow-ups or routine questions would show a lower note share for reasons that have nothing to do with sick days. The data does not record the reason for visit, so we cannot rule this in or out. It is the first thing we would test next.
• Employers, not states. State law is one layer of the sick-day experience. Company policy, industry practice, union contracts and individual managers set the rest, and two workers on the same street can live under entirely different rules: one texts a supervisor, the other calls in before the shift and hands over documentation after. State-level data cannot see any of that. What a twenty-point spread does give researchers is a reason to keep asking what sits underneath it.

The Same Sick Day, 46 Different Paper Trails

Twenty points is a lot of daylight for something as small as a note. Among people who book a telehealth visit for one, the expectation that a sick day comes with paperwork looks less like a national rule than a local habit, and one that changes at the state line. In Delaware or Maryland, a DrHouse visit ends with documentation more than one time in four. In Maine, New Hampshire and Montana, fewer than one in twelve. Same visit, different paper trail.

The ranking will be refreshed each year as the dataset grows, so the states that ranked on small samples will get bigger ones. Next year’s table will tell us how much of this persists and how much was noise.

How to Read, and Report, These Numbers

State rankings are easy to overstate, so a few ground rules for anyone citing this study.

• Bigger samples deserve more confidence. Maryland’s 27.4% rests on 2,115 visits and Texas’s 24.9% on 16,406; Hawaii’s 24.8% rests on 113. The small-state numbers are not wrong, they are less precise, and the confidence intervals in Table 1 say by how much.
• The middle of the table is crowded. A state at No. 19 is not meaningfully different from one at No. 24.
• These are shares of DrHouse telehealth visits, not of state residents. The accurate sentence is “27.9% of completed DrHouse telehealth visits from Delaware ended with a doctor’s note,” not “27.9% of Delawareans get doctor’s notes.”
• A higher rate is not better or worse. The data shows a note was issued. It does not show whether anyone needed it, required it, or resented it.

Methodology

• Data. 93,633 completed DrHouse telehealth visits with a visit date from February 29, 2024, through August 17, 2026. Each visit ID appears once. Canceled and no-show bookings are removed before the data reaches this analysis. A visit is counted as ending with a doctor’s note when at least one note was recorded for that visit; 20,051 visits met that definition, an overall rate of 21.4%.
• Outcome. The note rate is notes issued divided by completed visits. It is not a count of notes, not a measure of how often patients miss work, and not a measure of how ill they were.
• States. Each visit is assigned to the state recorded for the patient. Only states and the District of Columbia with at least 100 visits are reported individually. Five states fell below that threshold (Alaska, North Dakota, South Dakota, Vermont, Wyoming) and are excluded from the state table; together they hold 335 visits, 0.4% of the total. The four regional figures use Census Bureau regions and include all 93,633 visits. The same 100-visit floor applies to every year-level figure quoted in the text.
• Age adjustment. Direct standardization. Each state’s note rate was recalculated by applying its age-band-specific rates to the age distribution of the full dataset, in seven bands: under 18, 18–24, 25–34, 35–44, 45–54, 55–64, and 65 and over. Age is computed at the date of the visit. Forty-two visits with inconsistent dates of birth were excluded from the adjustment only.
• Statistical tests. Confidence intervals are Wilson 95% intervals. Each state was tested against the overall rate with a two-sided exact binomial test. The eleven states that differ from the overall rate at p < 0.001 also clear a Bonferroni-corrected threshold for 46 comparisons (0.05 divided by 46, roughly 0.001); Delaware (p = 0.002) does not. Significance is assessed against the overall rate, not state against state. An omnibus chi-square test across the 46 reportable places rejects the hypothesis that they share one rate (chi-square 487.0, 45 degrees of freedom, p < 0.001). Regional and law-group comparisons are reported two ways: on visit counts (two-proportion z-tests) and with each state as a single observation (Mann-Whitney tests on the 46 state rates), because states, not visits, are the unit at which a region or a law applies. On the state-level test the South differs from the other three regions combined at p = 0.02; the law group does not differ from the no-law group (p = 0.2). These intervals and tests treat visits as independent; repeat visits by the same patient and clinician-to-clinician differences in note-writing are not modeled, so the true uncertainty is somewhat larger than the intervals show, most of all in the smaller states.
• Robustness. Year-to-year stability is moderate. Comparing 2025 (full year) with 2026 to date, state rates correlate at 0.66 across the 38 places with at least 100 visits in both (Spearman, p < 0.001). Restricting both years to the same January-to-mid-August window leaves 29 places and a weaker correlation of 0.27, which is not statistically significant, largely because early-2025 volume was under a third of 2026’s; this should firm up as the dataset grows. Random split-halves of the full dataset correlate at 0.71 to 0.82 across five draws. Maryland, Nevada and Pennsylvania were top-five in both 2025 and 2026 however the smaller states are handled; Connecticut was bottom-five in both. The overall rate was 22.6% in 2024 (from February 29), 19.9% in 2025 and 22.4% in 2026 (to August 17); on a like-for-like January-to-mid-August window, 2025 was 19.2% and 2026 is 22.4%. This study does not analyze the trend.
• Paid-leave grouping. The list of statewide paid sick-leave and paid-leave laws was verified against two employer compliance trackers (GovDocs, February 2026; Paylocity, June 2026) and the state statutes cited in the text. Grouping is by statewide law in force at the end of the study period (August 17, 2026), which is why Missouri’s repealed law is not counted and Nebraska’s, effective October 1, 2025, is. The comparison is descriptive; alternative groupings (excluding the three any-reason states, or excluding Nebraska) move the law-group figure by no more than 0.2 points.
• What this data is not. It is a record of DrHouse visits, not a random sample of any state’s residents or workforce. Who uses DrHouse in each state, and what they use it for, depends on how people find the service there, so a state’s figure describes its DrHouse patients rather than its population. It measures whether a note was issued, not whether the patient was ill, how ill, or whether anyone required the note. It does not record occupation, employer or reason for visit. State comparisons are descriptive and are not evidence that any policy or industry causes any outcome.
• Privacy. All figures are aggregated and de-identified. No patient-level information is published and no personally identifiable information leaves DrHouse’s systems. Visit and note counts, confidence intervals and test statistics for every figure in this study are available on request. This study was reviewed for compliance prior to publication.