How AI saves doctors time: what the evidence shows

Doctors lose more time to paperwork than to patients — and the latest evidence shows AI is starting to give that time back. Across large real-world deployments and randomised trials, clinicians using AI to draft notes, letters and forms are reclaiming thousands of hours and reporting markedly lower burnout. This guide looks at what the data actually says, and where AI saves time in a real South African practice.

How much time does admin actually cost doctors?

The administrative load on clinicians is well documented. A landmark 2016 time-and-motion study published in the Annals of Internal Medicine found that for every hour of direct clinical face time with patients, physicians spent nearly two additional hours on the electronic health record (EHR) and desk work during the clinic day — and another one to two hours of after-hours work at home.

That after-hours work has a name: “pajama time”. The American Medical Association, drawing on EHR event-log research, reports that family physicians spend on average around 86 minutes of “pajama time” in the record each night. More recent log analyses put physicians at roughly 5.9 hours of an 11.4-hour workday inside the EHR. This documentation burden is one of the most consistently cited drivers of clinician burnout.

Most of this research comes from large US health systems, but the underlying squeeze — more documentation, more forms, more after-hours catch-up — is exactly what South African clinicians describe too.

What does the latest evidence say about AI and efficiency?

The most striking numbers come from The Permanente Medical Group, which rolled ambient AI scribes out across its physicians. Over roughly a year (October 2023 to December 2024), about 7,260 physicians used the technology for around 2.5 million patient encounters, and the scribes were associated with an estimated 15,700+ hours of documentation time saved — the equivalent of nearly 1,800 working days. In surveys, 88% of clinicians said the AI scribe had a positive impact on their interactions with patients.

Other recent studies point the same way:

The pattern is consistent: AI doesn’t just shave a few seconds off a note — it removes enough cumulative admin to change how a clinician’s week feels.

Where AI saves time in a real practice

Efficiency gains don’t come from one feature; they come from automating the repetitive work scattered across a consult. Here is where it adds up, and how GreenNotes handles each piece.

Clinical notes — the AI scribe

The biggest single drain is writing up consults. The GreenNotes AI Scribe records the visit and produces a structured clinical note — SOAP or your preferred format — in seconds, tuned for South African accents and Afrikaans. You review and sign; the typing is gone. Instead of carrying ten consults’ worth of notes into the evening, the record is essentially done by the time the patient leaves.

Letters and reports — auto-generation

Referral letters, sick notes, medical reports and medical-aid motivations are formulaic but time-consuming. GreenNotes auto-generates letters from your consult notes onto your custom letterhead — you select the relevant consults, the AI drafts the letter, and you edit and sign. A task that used to mean dictating or typing a page now takes a minute.

Forms and intake — auto-completion

Every practice drowns in repetitive forms: patient intake, consent, and statutory paperwork like workers’ compensation (COIDA / Injury-on-Duty) reports. Because GreenNotes already holds the patient demographics and consult details, it can auto-populate these forms rather than make you re-type the same information repeatedly. Remote intake links let patients complete their own admin and consent before they arrive, so the front desk isn’t the bottleneck.

Coding and billing — assisted billing (coming soon)

Billing is where lost time becomes lost revenue. GreenNotes already exports consultation data for your billing workflow, and GreenNotes Billing — launching soon — will add AI-assisted ICD-10 coding, real-time medical-aid claims and benefit verification, so fewer claims are rejected and fewer hours go into chasing them.

Does faster documentation mean worse notes?

It’s the right question to ask. The answer is that speed should never come at the cost of clinical accuracy — which is why the clinician stays in control. In a sound AI workflow, the model produces a draft; the doctor reviews, corrects and approves it before it enters the record. Used this way, AI removes the mechanical effort of documentation while leaving clinical judgement exactly where it belongs. For more on using AI responsibly in a clinical setting, see our note on AI data-processing terms.

How to capture these gains in your practice

• Start with the single biggest drain — usually consult notes — and adopt an AI scribe there first.
• Move letters, sick notes and reports to auto-generation so they’re a review step, not a writing task.
• Push intake and consent to the patient before the visit with remote forms.
• Automate the repetitive statutory forms (COIDA, referrals) you complete every week.
• Keep a human review step on everything — efficiency and accuracy are not a trade-off when the clinician signs off.
• Choose POPIA-compliant software so the time savings don’t create a compliance risk.

The headline from the evidence is simple: the admin burden is real and measurable, and AI is now demonstrably reducing it. For a South African practice, that means fewer evenings lost to paperwork — and more attention for the patient in the room.

Sources

1. Sinsky C, et al. Allocation of Physician Time in Ambulatory Practice: A Time and Motion Study in 4 Specialties. Annals of Internal Medicine, 2016.
2. American Medical Association. Family doctors spend 86 minutes of “pajama time” with EHRs nightly.
3. Tierney AA, et al. Ambient Artificial Intelligence Scribes: Learnings after 1 Year and over 2.5 Million Uses. NEJM Catalyst, 2025.
4. UCLA Health. UCLA study finds AI scribes may reduce documentation time and improve physician well-being, 2025.
5. American Medical Association. AI scribes save 15,000 hours — and restore the human side of medicine.