# CED Cannabis Science Digest: 3 Public-Health Cannabis Signals Worth Watching

> **Author:** Benjamin Caplan, MD (ABFM Board Certified, MA Lic. #240562, NPI: 1962722637)
> **Publisher:** CED Clinic (https://cedclinic.com)
> **Canonical URL:** https://cedclinic.com/cannabis-public-health-signals-2026-digest/
> **Published:** 2026-06-20T19:14:30-04:00 | **Last Modified / Clinically Audited:** 2026-09-28T10:51:09-04:00
> **Clinical Summary:** Explore cannabis public health signals 2026 with insights on pregnancy outcomes, psychosis follow-up, and e-scooter safety in this evidence-based digest.

---

CED Clinical Relevance #58 Notable Clinical Interest June 20, 2026 surfaced several cannabis papers, but the strongest human-facing items were already live elsewhere on CED Clinic or too overlap-heavy to justify repeat standalone treatment. Three fresh public-health signals still warranted preservation in digest form.

Clinical Insight | CED Clinic
No newly verified cannabis paper on June 20, 2026 offered a strong enough combination of human evidence, novelty, and distinctiveness to justify its own full-length evidence feature. The most defensible publication was a public-health digest that keeps three smaller but still useful signals in view: a mixed-substance pregnancy-outcomes analysis that included cannabis exposure, a retrospective psychosis cohort in which THC was the most common substance present, and a forensic road-safety study showing that cannabis-related intoxication questions now extend to e-scooter use as well as cars. None of these studies proves cannabis-specific causation for an individual patient, but all three sharpen clinical counseling.

Digest[Public Health](https://cedclinic.com/cannabis-mold-contamination-risks-recall-safety/)PregnancyPsychosisRoad Safety

Audience
Patients, caregivers, cannabis clinicians, obstetric clinicians, psychiatrists, primary-care clinicians, and public-health readers

Primary Topic
Three verified cannabis-related public-health signals spanning pregnancy outcomes, psychosis follow-up, and e-scooter road safety

Source
[Read the full study](https://pubmed.ncbi.nlm.nih.gov/42312764/)

CED Clinic’s strongest same-day cannabis items were already covered elsewhere or carried too much overlap to justify repeat standalone treatment, but three additional public-health signals still deserved careful preservation: one on pregnancy outcomes in hospitalized patients with psychological conditions and substance use, one on follow-up risk after substance-induced psychosis, and one on cannabis detection in e-scooter intoxication cases.

What This Study Teaches Us
This digest shows why cannabis risk counseling often depends on imperfect but still useful public-health datasets. One paper grouped cannabis with other substances in pregnancy-related hospitalizations, one followed patients after substance-induced psychosis where THC was commonly present, and one tracked real-world intoxication offenses involving e-scooters. None offers a simple consumer takeaway, but together they refine screening, follow-up, and harm-reduction conversations.

Why This Matters
[Cannabis medicine](https://cedclinic.com/essential-oils-industrial-hemp-antimicrobial-benefits/) conversations increasingly intersect with pregnancy, psychiatric vulnerability, and newer transportation settings. Clinicians need language that distinguishes signal from proof. These studies help identify where counseling should become more specific, even when the underlying data are retrospective, mixed-substance, or jurisdiction-specific.

Study Snapshot

Post Type
Evidence digest using the canonical CED layout

Batch ID
c898476e56da9d14

Items Reviewed
3 verified, nonduplicate, digest-eligible items

Editorial Decision
Useful public-health signals, but no new single-study cannabis paper justified separate full-length treatment after duplicate and overlap review

Item 1
Pregnancy outcomes with psychological conditions and substance use

Item 2
Substance-induced psychosis progression with cannabis prominence

Item 3
Cannabis and e-scooter road-risk toxicology

Primary Dates
June 2026; June 15, 2026; June 17, 2026

Content Lanes
Safety Signal; Safety Signal; Safety Signal

Digest Standard
Signals preserved with limitations, uncertainty, and non-causal framing made explicit

Related Reading
3 verified live CED Clinic internal links

Clinical Bottom Line
This digest is useful because it sharpens screening questions rather than offering a new treatment claim. Ask about pregnancy and co-occurring mental health conditions, treat cannabis-associated psychosis as a follow-up problem rather than a one-time event, and remember that cannabis-related road-risk counseling applies to micromobility as well as driving.

Why These Items Belong Together

The strongest fresh cannabis papers from today’s discovery cycle were already live on CED Clinic in other formats, including a CBD safety digest and separate coverage of a cancer-symptom extract trial and an epidermolysis bullosa pain study. Rather than repackage those same stories, this digest preserves three newer public-health signals that remained distinct but individually too limited for their own feature.

All three items are useful for counseling, screening, or follow-up. None reaches the kind of clean, cannabis-specific treatment-evidence standard that would justify a stronger headline on its own.

Digest Card 1 | Pregnancy Outcomes and Mixed Substance Exposure

Authors / source / date / lane: Nriagu and colleagues, JACC Advances, June 2026, Safety Signal. This paper was preserved in digest form rather than expanded into its own full-length article because the exposure group pooled cannabis with alcohol, opioids, cocaine, and amphetamines instead of isolating cannabis-specific risk estimates.

What was investigated: a propensity-matched analysis of 2,040,635 pregnancy- and delivery-related hospitalizations among patients with psychological conditions, comparing hospitalizations with substance use against matched hospitalizations without substance use.

What it appeared to find: substance use in that high-risk population was associated with higher odds of composite adverse pregnancy outcomes, hypertensive disorders, preterm delivery, fetal growth restriction, and abruptio placenta.

Limitations and uncertainty: this was cross-sectional hospital data, not a randomized study; cannabis was only one of several included substances; and the study does not tell readers what the cannabis-only effect size would be.

Why it is noteworthy: clinicians who counsel pregnant patients often face mixed real-world substance-use patterns rather than neat single-exposure scenarios, and this paper supports integrated screening rather than cannabis-only assumptions.

Digest Card 2 | Psychosis Follow-Up After Cannabis-Linked Presentations

Authors / source / date / lane: Ifteni, Miron, Necula, Dima, and Teodorescu, Psychiatry Research, June 15, 2026, Safety Signal. This item stayed in digest form because it was retrospective, involved substantial polysubstance use, and offers a risk signal rather than a clean causal estimate for cannabis alone.

What was investigated: a hospital-based review of substance-induced psychosis admissions between 2014 and 2024, examining which substances were involved and how often patients later returned or converted to a major psychiatric diagnosis.

What it appeared to find: THC was present in 82.95% of cases, 44.96% of patients reported multiple-drug use, nearly half required later psychiatric readmission, and 11.62% eventually received a schizophrenia- or bipolar-spectrum diagnosis.

Limitations and uncertainty: this was one retrospective cohort from a single psychiatric center, and the analysis cannot separate the independent contribution of cannabis from other vulnerability factors or co-used substances.

Why it is noteworthy: the paper reinforces that a cannabis-linked psychosis presentation should not be treated as a short-lived episode with no need for follow-up, especially when ongoing psychiatric risk may unfold over years.

Digest Card 3 | Cannabis, E-Scooters, and Road-Risk Counseling

Authors / source / date / lane: Städter, Kuntze, Dreßler, and Becker, International Journal of Legal Medicine, June 17, 2026, Safety Signal. This paper remained a digest card because it is region-specific forensic surveillance rather than a broadly generalizable clinical trial or national policy analysis.

What was investigated: 411 police-recorded traffic offenses involving e-scooter riders in Leipzig and surrounding areas from October 2021 through June 2024, with toxicology patterns and injury details reviewed retrospectively.

What it appeared to find: cannabis was predominantly detected in administrative intoxication offenses, alcohol played the larger role in criminal offenses, most cannabis-positive riders were non-chronic consumers, and many injuries reflected self-inflicted loss of balance with head, face, and upper-extremity trauma.

Limitations and uncertainty: the study reflects one German setting, policing patterns shape which cases are captured, and the data do not prove that any single THC threshold maps neatly onto real-world impairment for all riders.

Why it is noteworthy: patient counseling about cannabis and transportation risk should now include scooters and other micromobility devices, not just cars.

How Strong Is This Evidence?
All three items use real-world human data and address clinically relevant safety questions, which makes them worth preserving. The strength lies in practical signal detection, not in definitive proof.

Where This Paper Deserves Skepticism
Each paper also has a ceiling. One mixes multiple substances, one is a retrospective psychiatric cohort with heavy polysubstance exposure, and one is a regional forensic road-safety dataset. These are counseling papers, not product-endorsement papers.

What This Paper Does Not Show
This digest does not prove that cannabis alone caused every measured risk, that any one THC threshold defines impairment for every rider, or that a patient with a substance-induced psychosis presentation will inevitably transition to schizophrenia or [bipolar disorder](https://cedclinic.com/cannabis-induced-appetite-satiety-bypass-thc-brain/). It also does not argue against all medical-cannabis use in pregnancy, mental-health care, or chronic disease management in a blanket way.

How This Fits With the Broader Clinical Conversation

[Cannabis science](https://cedclinic.com/biphasic-effects-of-cannabis/) often advances through messy public-health observations rather than ideal randomized trials. That is especially true for pregnancy, emergency psychiatry, and transportation safety.

For clinicians, the task is to convert these studies into better questions, not to pretend they answered everything. Which substance was used, in what combination, in what population, and with what follow-up support still matters more than any one headline.

For patients, the key lesson is not panic but specificity. Pregnancy, psychosis history, and transportation choices are contexts where cannabis-related risk conversations should become more careful and more individualized.

Dr. Caplan’s Take

*When a paper mixes cannabis with other substances, I do not treat it as a cannabis-only verdict. I treat it as a reminder to ask better questions about the full exposure picture and the patient’s baseline vulnerability.*

*What catches my attention here is not a single dramatic effect size. It is the pattern: pregnancy counseling needs integration, psychosis follow-up needs endurance, and transportation counseling needs to catch up with how people actually move through the world.*

What a Careful Reader Should Take Away
These papers matter because they tighten clinical attention around three risk-heavy contexts. They should improve screening, documentation, and follow-up conversations, but they should not be used as shorthand proof that cannabis alone explains every measured harm.

Evidence Interpretation Guide

Table of Contents

- [How to Read Mixed Public-Health Cannabis Papers Without Overstating Them](#How_to_Read_Mixed_Public-Health_Cannabis_Papers_Without_Overstating_Them)[A Reading Order for Public-Health Cannabis Signals](#A_Reading_Order_for_Public-Health_Cannabis_Signals)
- [The Same Study Can Mean Different Things Depending on the Question Being Asked](#The_Same_Study_Can_Mean_Different_Things_Depending_on_the_Question_Being_Asked)[Context Matters More Than Headlines](#Context_Matters_More_Than_Headlines)
- [Screening Quality Is the Main Clinical Gain](#Screening_Quality_Is_the_Main_Clinical_Gain)
- [Pregnancy Risk Counseling Needs Mixed-Exposure Thinking](#Pregnancy_Risk_Counseling_Needs_Mixed-Exposure_Thinking)
- [Psychosis Presentations Deserve Longitudinal Attention](#Psychosis_Presentations_Deserve_Longitudinal_Attention)
- [Road Safety Is Moving Beyond Cars](#Road_Safety_Is_Moving_Beyond_Cars)
- [These Are Signals, Not Verdicts](#These_Are_Signals_Not_Verdicts)
- [Law, Labeling, and Counseling All Interact](#Law_Labeling_and_Counseling_All_Interact)
- [What Better Cannabis Safety Research Still Needs](#What_Better_Cannabis_Safety_Research_Still_Needs)
- [Frequently Asked Questions](#Frequently_Asked_Questions)[Related](#Related)

## How to Read Mixed Public-Health Cannabis Papers Without Overstating Them

Many [cannabis safety](https://cedclinic.com/respiratory-effects-inhalational-marijuana/) papers do not study one clean cannabinoid exposure in one tightly defined population. They study what clinicians actually face: mixed substance use, psychiatric comorbidity, and real-world injury patterns.

That makes them useful, but only if readers stay disciplined about what the design can and cannot support.

### A Reading Order for Public-Health Cannabis Signals

**Start With the Exposure Definition**Ask whether the paper isolates cannabis, groups cannabis with other substances, or treats THC as one marker inside a broader intoxication picture. Exposure definition controls the size of the claim.

**Check the Outcome Type**Hospital complications, psychiatric readmission, and police-recorded traffic offenses are not interchangeable outcomes. Each tells you something different about risk.

**Separate Association From Counseling Value**A study can be too limited for a broad causal conclusion and still be clinically useful because it tells you where screening or follow-up should become more specific.

**Notice the Missing Counterfactual**If a study lacks cannabis-only estimates, randomized exposure, or multi-site replication, that absence sets the upper limit on what you should claim publicly.

The Question Editors Needed to Answer
Were these public-health signals useful enough to preserve for readers even though each paper still carried design limits that mattered clinically?

The Question Patients Usually Need Answered
Does this study tell me cannabis definitely caused harm for people like me, or does it mainly tell me that my clinician should ask more detailed questions about pregnancy, psychiatric history, and transportation safety?

The Bottom Line
These papers support sharper counseling and follow-up, not blanket claims or blanket reassurance.

CED Perspective Lens

## The Same Study Can Mean Different Things Depending on the Question Being Asked

Scientific papers rarely answer a single question. Patients, clinicians, researchers, and critics can read the same data differently. These evidence-based lenses show where this trial is useful, where it remains uncertain, and how easily it can be overstated.

Overview
The same digest can matter differently to an obstetric clinician, a psychiatrist, a patient with prior cannabis exposure, or a public-health reader tracking transportation injury risk. These lenses keep the evidence grounded in the actual question being asked.

#### Context Matters More Than Headlines

These studies do not say that every cannabis user faces the same level of pregnancy, psychiatric, or road-safety risk. They do say that those three contexts deserve more detail than a casual conversation usually gives them.

If any of these contexts applies to you, the right next step is not self-diagnosis from a digest. It is a more specific conversation about your full exposure pattern, other substances, mental-health history, and transportation habits.

Lens takeaway
Use this digest to improve the quality of your questions, not to replace individualized medical advice.

#### Screening Quality Is the Main Clinical Gain

The practical value here is screening specificity. Ask pregnant patients about co-occurring substances and mental-health diagnoses, document psychosis follow-up plans rather than only the acute episode, and include scooters in transportation counseling.

None of these papers supports a new prescription habit, but each supports a better documentation and counseling habit.

Lens takeaway
The actionable output is better history-taking, follow-up, and counseling language.

#### Pregnancy Risk Counseling Needs Mixed-Exposure Thinking

Pregnancy counseling often breaks down when cannabis is treated in isolation despite real-world overlap with alcohol, nicotine, stimulants, or untreated psychiatric illness. The JACC Advances paper is useful precisely because it reflects that complexity.

Its weakness is the same as its strength: because it groups substances together, it cannot tell you what cannabis alone would have done.

Lens takeaway
Integrate cannabis into broader pregnancy risk screening rather than extracting it from the clinical context.

#### Psychosis Presentations Deserve Longitudinal Attention

A patient can look clinically improved after discharge and still remain at meaningful psychiatric risk over the next several years. That is the signal preserved by the psychosis paper.

The study does not prove a deterministic transition, but it strongly argues against minimizing follow-up after a cannabis-linked psychosis presentation.

Lens takeaway
Treat substance-induced psychosis as a follow-up issue, not only an acute-management issue.

#### Road Safety Is Moving Beyond Cars

Cannabis-related road-risk discussions often stay focused on automobile driving. The Leipzig e-scooter paper shows why that framing is already outdated.

Micromobility devices create new combinations of intoxication, balance-related injury, and legal exposure that clinicians and policymakers both need to understand.

Lens takeaway
Transportation counseling should include scooters and other micromobility devices.

#### These Are Signals, Not Verdicts

A careful skeptic should notice how much inferential space remains in all three items. Mixed-substance exposure, retrospective design, and jurisdiction-specific enforcement patterns all constrain causal interpretation.

That skepticism is not a reason to ignore the papers. It is the discipline that keeps them clinically useful instead of rhetorically inflated.

Lens takeaway
The right response is calibrated attention, not dismissal and not overstatement.

#### Law, Labeling, and Counseling All Interact

Pregnancy warning language, psychosis follow-up infrastructure, and intoxication rules for emerging forms of transport are all policy questions, not just bedside questions.

The evidence here is not final enough to settle those debates, but it is relevant enough that policymakers should not ignore it.

Lens takeaway
Public-health evidence often shapes policy before it reaches the certainty of a trial.

#### What Better Cannabis Safety Research Still Needs

The next step is not more vague cannabis-risk rhetoric. It is better design: cannabis-only analyses where possible, prospective psychiatric follow-up, and multi-jurisdiction road-safety data that connect toxicology with functional impairment.

Until then, clinicians will keep working with imperfect but meaningful signals like these.

Lens takeaway
Better exposure definition and longer follow-up would raise the evidentiary ceiling substantially.

Join the Conversation

Have a question about how this applies to your situation? [Ask Dr. Caplan](https://benjamincaplan.com/ask-dr-caplan)

Want to discuss this topic with other patients and caregivers? [Join the forum discussion](https://cedclinic.com/interact/discussion-forums/forum/cannabis-industry/)

[https://cedclinic.com/contact/](https://cedclinic.com/contact/)

Physician-Led, Whole-Person Care
A doctor who takes the time to truly understand you.
Personal care that starts with listening and is guided by experience and ingenuity.

Health, Longevity, Wellness
One-on-One Cannabis Guidance
Metabolic Balance

[Leave a Message](https://cedclinic.com/contact/)
[Metabolic Care](https://cedclinic.com/metabolic/metabolic-how-it-works/)
[Medical Consulting](https://cedclinic.com/book/)
[Cannabis Care](https://cedclinic.com/book/)

[https://cedclinic.com/contact/](https://cedclinic.com/contact/)

Have thoughts on this? Share it:

[Share on X](https://twitter.com/intent/tweet?text=CED+Cannabis+Science+Digest%3A+3+Public-Health+Cannabis+Signals+Worth+Watching&url=https://cedclinic.com/cannabis-science-digest-june-20-2026-public-health-signals/)[Share on LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https://cedclinic.com/cannabis-science-digest-june-20-2026-public-health-signals/)[Share on Bluesky](https://bsky.app/intent/compose?text=CED+Cannabis+Science+Digest%3A+3+Public-Health+Cannabis+Signals+Worth+Watching%20https://cedclinic.com/cannabis-science-digest-june-20-2026-public-health-signals/)

Source: [Primary sources are listed inside each digest card; this lead source link opens the pregnancy-outcomes analysis on PubMed.](https://pubmed.ncbi.nlm.nih.gov/42312764/)

Related Reading at CED Clinic
Continue exploring the evidence

[Dartmouth Health targets treatment gap for pregnant patients with SUD](https://cedclinic.com/dartmouth-health-targets-treatment-gap-for-pregnant-patients-with-sud/)

Useful companion reading for the pregnancy paper because it frames how cannabis counseling fits into broader prenatal substance-use screening and treatment access.

[Read the pregnancy context](https://cedclinic.com/dartmouth-health-targets-treatment-gap-for-pregnant-patients-with-sud/)

[Relationships between cannabis use and mental disorders: assessing the coherence of evidence from studies with different methodologies.](https://cedclinic.com/relationships-cannabis-mental-disorders-evidence-coherence/)
Helpful context for the psychosis paper because it shows the broader evidence base linking heavier cannabis exposure with psychiatric risk while still separating association from certainty.

[Read the psychiatric context](https://cedclinic.com/relationships-cannabis-mental-disorders-evidence-coherence/)

[CED Cannabis Science Digest: 3 Behavior and Safety Signals Worth Watching](https://cedclinic.com/cannabis-behavior-safety-signals-june-2026/)
Relevant follow-on reading for the road-safety card because it preserves adjacent CED coverage on driving capability, cross-border crash risk, and cannabis-related behavior signals.

[Read the safety context](https://cedclinic.com/cannabis-behavior-safety-signals-june-2026/)

Similar Studies on CEDClinic.com
Earlier coverage on closely related evidence
When a new paper overlaps with earlier CED Clinic coverage, we preserve the chain instead of hiding the overlap. These links point to older related posts so readers can compare what is new, what is repeated, and how the evidence has moved.

Digest Card 1 | Safety Signal | Pregnancy outcomes
[Dartmouth Health targets treatment gap for pregnant patients with SUD](https://cedclinic.com/dartmouth-health-targets-treatment-gap-for-pregnant-patients-with-sud/)

Earlier CED coverage on pregnancy-related screening and treatment access for patients with substance use disorders.

Similar because both pieces address pregnancy risk and substance-use screening. Different because the new paper is a large mixed-substance hospital analysis rather than a care-delivery article.

[Compare related coverage](https://cedclinic.com/dartmouth-health-targets-treatment-gap-for-pregnant-patients-with-sud/)

Digest Card 2 | Safety Signal | Psychosis follow-up
[Relationships between cannabis use and mental disorders: assessing the coherence of evidence from studies with different methodologies.](https://cedclinic.com/relationships-cannabis-mental-disorders-evidence-coherence/)
Earlier CED review of the broader evidence linking cannabis exposure with psychiatric outcomes.

Similar because both pieces address cannabis and psychiatric risk. Different because the new paper tracks real-world hospital outcomes after substance-induced psychosis rather than synthesizing multiple methodologies.

[Compare related coverage](https://cedclinic.com/relationships-cannabis-mental-disorders-evidence-coherence/)

Digest Card 3 | Safety Signal | Road safety
[CED Cannabis Science Digest: 3 Behavior and Safety Signals Worth Watching](https://cedclinic.com/cannabis-behavior-safety-signals-june-2026/)
Earlier CED digest preserving adjacent driving- and behavior-related [cannabis safety](https://cedclinic.com/respiratory-effects-inhalational-marijuana/) signals.

Similar because both items discuss cannabis-associated transportation or behavior risk. Different because the new paper focuses specifically on e-scooter toxicology and injury patterns in one jurisdiction.

[Compare related coverage](https://cedclinic.com/cannabis-behavior-safety-signals-june-2026/)

## Frequently Asked Questions

Why is this a digest instead of a full-length feature on one paper?

Because the most useful new items from this run were smaller public-health signals rather than one clearly dominant, broadly generalizable cannabis study. Grouping them together preserves their value without overstating any single paper.

Does the pregnancy paper prove cannabis alone caused worse outcomes?

No. The exposure group combined cannabis with other substances, so the study supports integrated risk screening but does not provide a cannabis-only effect estimate.

What did the pregnancy study actually compare?

It compared matched pregnancy-related hospitalizations among patients with psychological conditions, separating those with documented substance use from those without documented substance use.

What is the main clinical lesson from the psychosis study?

A cannabis-linked or broader substance-induced psychosis presentation should trigger meaningful follow-up planning because later readmission and conversion to major psychiatric diagnoses can still occur.

Does the psychosis paper prove cannabis causes schizophrenia?

No. It is a retrospective cohort with substantial polysubstance use, so it shows association and follow-up risk, not deterministic causation.

Why include an e-scooter paper in a cannabis digest?

Because cannabis-related impairment counseling increasingly applies to micromobility devices as well as cars, and clinicians need contemporary examples of where those safety questions arise.

Does the e-scooter study settle what THC legal limit should apply to riders?

No. The paper argues that current limits deserve review, but a region-specific forensic dataset cannot by itself define the correct legal threshold for every setting.

What should clinicians do differently after reading this digest?

Ask more specific questions about pregnancy, psychiatric vulnerability, co-used substances, and micromobility behavior, and avoid turning mixed datasets into oversimplified cannabis-only claims.

What should patients do differently after reading this digest?

Bring up pregnancy plans, mental-health history, other substance use, and transportation habits openly with a clinician instead of assuming cannabis risk is uniform across all situations.

Are any of these studies evidence that cannabis is a treatment for these problems?

No. These are safety and public-health papers. They should guide counseling and follow-up, not be used as proof of treatment benefit.

 

### Related

---

### Verification & Medical Governance

- **Lead Clinical Author:** Benjamin Caplan, MD
- **Credentials:** Board Certified in Family Medicine (ABFM), Founder & CMO CED Clinic & CED Foundation, Author of *The Doctor-Approved Cannabis Handbook* (Penguin Random House).
- **Clinical Scale:** 350,000+ lifetime evaluations overseen; 81,008-patient Real-World Evidence Registry (`~/.ced_rwe_vault.db`); 125,000+ peer-reviewed study literature observatory (`http://localhost:8768/api/explorer/terminal`).
- **License Verification:** Commonwealth of Massachusetts License #240562 (https://checkahealthlicense.mass.gov/).
- **NPI Registry:** 1962722637 (https://npiregistry.cms.hhs.gov/provider-view/1962722637).
- **Ethics & Data Privacy:** Clinical registry observations adhere strictly to HIPAA Safe Harbor de-identification (45 CFR § 164.514(b)) and criteria for secondary research exemption under 45 CFR § 46.104(d)(4).