# Physician Assistants

> Provide healthcare services typically performed by a physician, under the supervision of a physician. Conduct complete physicals, provide treatment, and counsel patients. May, in some cases, prescribe medication. Must graduate from an accredited educational program for physician assistants.

- **SOC code:** 29-1071.00
- **Canonical URL:** https://singulariki.com/roles/role-29-1071-00
- **Also known as:** Certified Physician Assistant (PA-C), Family Practice Physician Assistant, Physician Assistant (PA), Physician's Assistant, Cardiology Physician Assistant, Emergency Medicine Physician Assistant (Emergency Medicine PA), Orthopaedic Physician Assistant, Orthopedic Physician Assistant
- **Frame:** "AI exposure" means task overlap (how codifiable the work is), not jobs lost or a forecast. Every figure below is traced to a named public dataset.

## What this work is

**Core tasks** (O*NET):
- Make tentative diagnoses and decisions about management and treatment of patients.
- Examine patients to obtain information about their physical condition.
- Interpret diagnostic test results for deviations from normal.
- Obtain, compile, and record patient medical data, including health history, progress notes, and results of physical examination.
- Prescribe therapy or medication with physician approval.
- Administer or order diagnostic tests, such as x-ray, electrocardiogram, and laboratory tests.
- Instruct and counsel patients about prescribed therapeutic regimens, normal growth and development, family planning, emotional problems of daily living, and health maintenance.
- Perform therapeutic procedures, such as injections, immunizations, suturing and wound care, and infection management.
- Visit and observe patients on hospital rounds or house calls, updating charts, ordering therapy, and reporting back to physician.
- Provide physicians with assistance during surgery or complicated medical procedures.
- Supervise and coordinate activities of technicians and technical assistants.
- Order medical and laboratory supplies and equipment.

**Emerging tasks** (O*NET):
- Refer patients to other healthcare providers.

## Skills, tools, capabilities

**Knowledge, skills & abilities** (O*NET, highest importance first):
- Medicine and Dentistry _(knowledge)_
- Biology _(knowledge)_
- English Language _(knowledge)_
- Psychology _(knowledge)_
- Therapy and Counseling _(knowledge)_
- Customer and Personal Service _(knowledge)_
- Reading Comprehension _(essential_skill)_
- Critical Thinking _(essential_skill)_
- Oral Comprehension _(ability)_
- Problem Sensitivity _(ability)_
- Deductive Reasoning _(ability)_
- Inductive Reasoning _(ability)_

**Skills in demand:**
- Biology _(Specialized Skill)_
- English Language _(Common Skill)_
- Psychology _(Specialized Skill)_
- Reading Comprehension _(Common Skill)_
- Inductive Reasoning _(Common Skill)_
- Deductive Reasoning _(Common Skill)_
- Critical Thinking _(Common Skill)_
- Writing _(Common Skill)_
- Speech Recognition _(Specialized Skill)_
- Social Perceptiveness _(Common Skill)_
- Information Ordering _(Specialized Skill)_
- Active Listening _(Common Skill)_

**Tools & technology:**
- Epic Systems _(hot technology, in demand)_
- eClinicalWorks EHR software _(hot technology)_
- MEDITECH software _(hot technology)_
- Microsoft Excel _(hot technology)_
- Microsoft Office software _(hot technology)_
- Microsoft Outlook _(hot technology)_
- Microsoft Word _(hot technology)_
- ChartWare EMR
- Electronic medical record EMR software
- Healthcare common procedure coding system HCPCS
- Medical condition coding software
- Medical procedure coding software

## AI exposure & outlook

- **AI task-overlap index:** 38th percentile (Moderate) across all occupations — composite of current-era exposure studies (ai-exposure-index-v1).
- **Overall AI exposure (Felten et al.):** 51st percentile (Moderate) — source: felten_aioe.
- **LLM task exposure, γ (OpenAI / Eloundou):** 59th percentile (Moderate) — source: eloundou_gamma.
- **AI assistant applicability (Microsoft):** 11th percentile (Low) — source: microsoft_applicability.
- **Frey–Osborne (2013, historical computerization estimate):** 31st percentile — kept separate from current-era studies.
- **Remote-capable (Dingel–Neiman):** no — task structure, not who actually works remote.
- **Projected employment (BLS 2024–34):** 20.4% growth (Growing fast); 12k annual openings; 162.7k → 195.8k jobs.
- **Pay & employment (BLS OEWS, May 2024):** median $133,260; 155,540 employed.

## How people actually use AI here

Anthropic Economic Index — measured AI conversations mapped to this occupation's tasks:

- **Automation vs augmentation:** 18% automation, 73% augmentation (usage-weighted).
- **Autonomy median:** 4.0 (higher = AI acts more independently).
- **Dominant collaboration mode:** learning.

**Tasks most handed to AI here:**
- Interpret diagnostic test results for deviations from normal. _(4.0% of measured AI use; learning)_
- Instruct and counsel patients about prescribed therapeutic regimens, normal growth and development, family planning, emotional problems of daily living, and health maintenance. _(0.8% of measured AI use; learning)_
- Examine patients to obtain information about their physical condition. _(0.6% of measured AI use; learning)_

**Example prompts (honest phrasings of the tasks above — starting points, not endorsed instructions):**
- Help me interpret diagnostic test results for deviations from normal.
- Help me instruct and counsel patients about prescribed therapeutic regimens, normal growth and development, family planning, emotional problems of daily living, and health maintenance.
- Help me examine patients to obtain information about their physical condition.

## Sources

- **O*NET** (30.3) — U.S. Department of Labor / National Center for O*NET Development. https://www.onetcenter.org/database.html
- **BLS Occupational Employment and Wage Statistics (OEWS)** (May 2024) — U.S. Bureau of Labor Statistics. https://www.bls.gov/oes/
- **BLS Employment Projections** (2024–2034) — U.S. Bureau of Labor Statistics. https://www.bls.gov/emp/
- **Anthropic Economic Index** (v4 (2026-01-15) + v2 (2025-03-27)) — Anthropic. https://www.anthropic.com/economic-index
- **Microsoft “Working with AI”** (working-with-ai) — Microsoft Research. https://www.microsoft.com/en-us/research/
- **“GPTs are GPTs” (Eloundou et al.)** (arXiv 2303.10130) — OpenAI / academic. https://arxiv.org/abs/2303.10130
- **AI Occupational Exposure (AIOE)** (Felten, Raj & Seamans) — academic. https://github.com/AIOE-Data/AIOE
- **Frey & Osborne (2013)** (frey-osborne-automation) — academic. https://www.oxfordmartin.ox.ac.uk/publications/the-future-of-employment/
- **Dingel & Neiman (2020)** (dingel-neiman-workathome) — academic. https://github.com/jdingel/DingelNeiman-workathome

---
_Generated from Singulariki's joined dataset; data snapshot 2026-06-02T21:00:32.945303+00:00. https://singulariki.com/roles/role-29-1071-00_