What the NAI's Factor Analysis Means for Healthcare Advertising
March 10, 2026 · Jeremy Mittler
Last week, the NAI released guidelines to help determine if companies are processing sensitive health data. It's called a Factor Analysis for Health-Related Sensitive Personal Information —a structured way to test if a company is processing HSPI. I read all 24 pages. Here is what it means for healthcare advertising.
1. This Was Needed
State privacy laws are inconsistent, ambiguous, and growing. Nearly two dozen of them now. A common framework reduces a lot of guesswork.
The NAI's factor analysis gives the industry something it hasn't had: a structured, repeatable way to assess whether a data practice crosses into sensitive health territory. That kind of clarity matters when the legal landscape changes every legislative session.
2. Inferences Are Called Out as a Key Factor
This is the part that matters most. The framework makes it explicit. The examples reinforce it hard.
If your audience methodology creates health inferences—predictions about a person's condition, treatment status, or diagnosis—the NAI's framework treats that as a factor that moves you toward HSPI. Not a gray area. A key factor.
3. Inferences and Predictions Create HSPI
The document uses a "pregnancy prediction score" as the example. Take safe data. Run a model. Output a health likelihood score. Target.
That approach lights up most of the factors. The input data may be innocuous. But the output—a prediction tied to a health condition—is what creates the problem. The method is what matters, not just the source.
4. Evidence-Based Cohorting Does Not Create HSPI
Blueprint's evidence-based cohorting (EBC) approach does not create HSPI.
Population prevalence is a cohort property. It is not an inference about an individual. The document says it directly: it would be unreasonable and unnecessary to infer a condition from population prevalence alone.
That is exactly how EBC works.
5. Most Vendors in Our Industry Are Creating HSPI Today
Not because they touch claims data. Because they create inferences.
Washington requires opt-in consent for HSPI. Maryland restricts its sale altogether. If your method produces health inferences, the regulatory exposure is real and growing.
One More Thing for Specialty and Rare Disease Marketers
The "safe" examples in the document rely on demographic correlations. This is fine for broad conditions but not good enough when you are trying to find a needle in a haystack.
EBC delivers precision without creating HSPI. That is the whole point.
The Bottom Line
This framework does not say stop health advertising. It says be honest about what you are creating.
If you work in health advertising, the NAI's Factor Analysis is worth 30 minutes of your time. Run your strategy through it. Then run your vendor's story through the examples.
Want to Understand How EBC Avoids Creating HSPI?
Blueprint's evidence-based cohorting was designed from the ground up to deliver precision without creating sensitive health inferences. If you want to see how your current strategy holds up under the NAI's framework, we're happy to walk through it.
About the Author: Jeremy Mittler is the founder of Blueprint Audiences, a privacy-first healthcare audience company that helps pharmaceutical and healthcare marketers reach patients without compromising on compliance or performance.