Billing maze
For every hour a physician spends with patients, how many go to the computer?
The US jumped from about 13,000 ICD-9 codes to 68,106 ICD-10 diagnosis codes in 2015, and physicians now spend nearly two hours on records and desk work for every hour of direct patient care.
The problem
Billing and documentation complexity is an operating-cost problem across the hospital and physician landscape. It consumes clinician time, adds back-office labor, slows revenue-cycle work, and turns coding requirements into a daily tax on patient-facing capacity.
The recommendation
Make documentation simplification a capacity and workforce strategy. The recommended operating model is to reduce duplicative coding and prior-documentation requirements, deploy responsible automation and team documentation support, and evaluate success by clinician time returned to patient care.
The code explosion
How big and how fast the coding system grew, from the ICD-9 to ICD-10 diagnosis jump through the separate procedure book and the still-unadopted ICD-11.
Diagnosis codes before and after the 2015 switch
Diagnosis codes before and after October 1, 2015. The count jumped about 5.2 times overnight. The reference line marks the roughly 13,000 ICD-9-CM baseline.
Read it this way The two bars show the code count on either side of the October 2015 switch, 68,106 versus roughly 13,000, a jump of about 5.2 times overnight. The reference line makes the size of that jump visible at a glance, but it says nothing about whether the added codes changed diagnosis accuracy. Use this chart to see which part of administrative burden is being quantified and why the recommendation treats simplification as a way to recover care capacity, not just reduce paperwork.
⊞ data table⬇ CSV
| Code set | Diagnosis codes |
|---|---|
| ICD-9-CM (before 2015) | 13000 |
| ICD-10-CM (2015) | 68106 |
JAMIA (Steindel) and Advances in Wound Care (Cartwright) · 2015 · source
The code set clinicians must navigate keeps growing
Codebooks are not directly comparable: ICD-9-CM and ICD-10-CM values are diagnosis codes, ICD-10-PCS is inpatient procedure codes, and ICD-11 is codable terms after postcoordination. The US has not adopted ICD-11 for billing.
Read it this way Reading left to right, the figure climbs each time a new system arrives: about 13,000 diagnosis codes in 1979, 68,106 diagnosis codes and 72,589 procedure codes in 2015, and roughly 120,000 ICD-11 codable terms adopted by WHO in 2022. Because those four figures measure different things, diagnoses, procedures, and postcoordinated terms, this shows growth in coding complexity generally, not one consistent code count over time, and the US has not adopted the 120,000-term ICD-11 for billing. Use this chart to see which part of administrative burden is being quantified and why the recommendation treats simplification as a way to recover care capacity, not just reduce paperwork.
⊞ data table⬇ CSV
| Year | Codebook | Codes or terms |
|---|---|---|
| 1979 | ICD-9-CM diagnosis codes | 13000 |
| 2015 | ICD-10-CM diagnosis codes | 68106 |
| 2015 | ICD-10-PCS inpatient procedure codes | 72589 |
| 2022 | ICD-11 codable terms (postcoordinated) | 120000 |
JAMIA, descriptive overview of the ICD-10-CM/PCS HIPAA code sets (Steindel), and WHO ICD-11 Fact Sheet · 2015 · source
The time tax
What the complexity costs in physician hours: the roughly 2-to-1 desk-to-patient ratio, and where the paperwork lands across the day and after hours at home.
Direct patient time versus EHR and desk work
Across the whole office day, physicians spent nearly two hours on EHR and desk work for every hour of direct patient face time. Inside the exam room the balance is closer.
Read it this way Compare the two grouped pairs: across the whole office day, EHR and desk work (49.2 percent) outweighs direct patient time (27.0 percent), but inside the exam room itself the balance flips toward patients, 52.9 versus 37.0 percent. The gap between the two settings is where documentation done outside the room accumulates, not evidence that any single visit is unbalanced. Use this chart to see which part of administrative burden is being quantified and why the recommendation treats simplification as a way to recover care capacity, not just reduce paperwork.
Caveat Time and motion study of 57 physicians in 4 ambulatory specialties. Office-day shares do not sum to 100 percent because other work is excluded.
⊞ data table⬇ CSV
| Setting | Direct patient face time % | EHR and desk work % |
|---|---|---|
| Whole office day | 27 | 49.2 |
| In the exam room | 52.9 | 37 |
Annals of Internal Medicine, Allocation of Physician Time (Sinsky) · 2016 · source
Why this matters
The code explosion and the time tax reinforce each other: more codes to select from means more documentation decisions per encounter, and that documentation happens disproportionately outside the exam room, where the split is closer to even (52.9 percent patient time versus 37.0 percent EHR and desk work). Some physicians close the gap only by working after hours, with 21 of the 57 study physicians self-reporting 1 to 2 additional hours of EHR work at home per night. This is drawn from a single 57-physician time and motion study in 4 ambulatory specialties, not a national census, so the exact magnitude will vary by specialty and EHR vendor.
Recommended actions
- Fund scribe support and ambient documentation tools for high-volume specialties, since documentation time outside the exam room is the largest recoverable hour.
- Automate routine coding assistance so physicians spend less time searching a 68,106-code set for the correct diagnosis.
- Hold off on any move toward ICD-11's roughly 120,000 codable terms until documentation burden from ICD-10 is demonstrably reduced.
- Track after-hours EHR time as a standard practice metric, not just office-day averages, since pajama time is where burden hides.
The recommendation
Therefore, make documentation simplification a capacity and workforce strategy. The recommended operating model is to reduce duplicative coding and prior-documentation requirements, deploy responsible automation and team documentation support, and evaluate success by clinician time returned to patient care.
Demographic slice none. Coding and documentation-time studies are practice or setting-level.
Sources