The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Hospital & medical-tech payroll: where compliance quietly gets expensive
Healthcare and medical-tech hire like startups and get regulated like hospitals. That tension shows up first in payroll.
Where risk concentrates:
1. Worker classification — locums, 1099 billers, fractional CMOs, clinical advisors. Misclassification = back withholding, FICA, unemployment, multi-year penalties.
2. Multi-state people nexus — remote educators/engineers create withholding + SUI where work is performed. Registering nine months late is a common expensive pattern.
3. Differentials, on-call, stipends — taxability + overtime interaction. Taxable stipends labeled as nontaxable “reimbursements” without an accountable plan is a recurring exam issue.
4. Benefits timing — ACA / plan eligibility when headcount crosses thresholds; payroll codes must match plan docs.
5. Mixed entities — MSO + locum + W-2 + production bonuses need clear pay-element mapping.
Controls that help: classification memos before engagement, multi-state checklist on hire day-1, accountable plans in writing, and a CPA/EA who reads payroll like an exam agent would.
Sophie Master, CPA & IRS Enrolled Agent — sophiecpa.com
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Trending
retrofuturism
Share your favorite designs from the past exploring the distant future. Or your own take on the aesthetic 🚀
halftone
Dots. Texture. Grain. Let's see your designs using the halftone technique.
Midjourney
Creatives are showing off what they're making with Midjourney.
colorgrade
Video editors sound off: How do you enhance the mood of your video in color grading?
Related posts
Think of a construction site where every subcontractor gets a blueprint, their own fenced-off area, and an inspector who checks the work before it joins the building. That's manager-mode: my open-source framework for running many AI coding agents at once without them stepping on each other or cutting corners. This page is the inspection record from one real build: 13 pieces, all passed, none torn out, $29.62 in AI cost. Faster delivery, and every step is on the record.
Two years ago I was working as a Clinical Officer, not a data analyst. I saw hospital numbers every day — appointments, billing, readmissions — but I didn't have the tools to actually dig into them.
Now I do. And here's what I've learned: a lot of hospitals and insurers are sitting on numbers that could save them real money, if someone actually analyzed them properly.
Here's a sample of what turned up across three recent projects:
📊 Hospital operations — I analyzed appointment and billing data and found that 51.5% of appointments ended in a no-show or cancellation, and 33% of payments failed outright. Chemotherapy and MRI turned out to be the top revenue-generating services — which changes where you'd prioritize fixing the leaks.
📊 Patient readmissions — I found that a single prior readmission more than doubled a patient's risk, from 27.79% to 69.50%. Patients with four or more comorbidities crossed 75%. Numbers like that can actually change how discharge planning gets done.
📊 Insurance claims — I analyzed 4,500 claims worth $22.5M and found something unexpected: patient age, income, and gender weren't significantly associated with claim cost. Pediatrics was the real cost driver at $4.9M — the specialty mattered, not the demographics.
I do this work with SQL, Python, Excel, Power BI, and Tableau — but the tools alone aren't the point. My clinical background is what helps me understand what's actually happening behind each number, not just what's sitting in a spreadsheet column.
If you've got operations, claims, or readmission data you haven't fully dug into — send me a sample. I'll tell you honestly what it can (and can't) show before you commit to anything.
