New Telehealth Study Underreports Medication Abortion Risks


The journal Nature recently published a study that reported on the effectiveness and safety of telehealth medication abortion in the USA. The study concluded that “telehealth medication abortion is effective, safe, and comparable to published rates of in-person medication abortion care.” The authors analyzed 6,034 electronic medical records from three for-profit medication abortion businesses.

The results have been widely praised by the New York TimesScientific AmericanNPR, and many other media organizations. However, the study has serious flaws that underreport medication abortion risks.

Eight adverse event types excluded from study

The authors included four adverse event types in their analysis: “blood transfusion, abdominal surgery, hospital admission requiring overnight stay, or death.” However, this list is missing eight additional adverse event types:

  1. sepsis
  2. hemorrhage (without blood transfusion)
  3. pain
  4. gastrointestinal symptoms
  5. patient experience (e.g., anxiety, depression, regret)
  6. drug toxicity/overdose
  7. homicide
  8. suicide

The excluded adverse event types are not hypothetical or unusual. They are valid medication abortion adverse event types recognized in these reports:

The study underreports medication abortion adverse events because it does not include eight widely recognized adverse event types.

Financial conflict of interest not disclosed

The study analyzed medical records from three companies: ChoixHeyJaneand Abortion on Demand. These organizations are for-profit start-up businesses. They have a financial incentive to report “safe and effective” telehealth abortion. Choix was founded in 2020 and ceased operations in October 2023, citing “significant fundraising challenges.” HeyJane and AbortionOnDemand were founded in 2021. The report did not disclose these financial conflicts of interest.

Patient underreporting of adverse events

Almost three-quarters of the patients, 72 percent, communicated with the clinic using mobile app messaging (i.e., “asynchronous care”). The clinics sent patients messages and surveys after they ingested the medication abortion pills. Clinicians could not see or listen to each patient to assess whether the patient had underreported or misreported symptoms. Clinicians assumed none if a patient did not explicitly state their adverse reaction. The authors similarly assumed that lack of a response is a positive indicator of a healthy outcome. The authors provide no explanation of why this is a valid assumption. 

Author conflict of interest not disclosed

All five authors work for pro-abortion organizations. Four authors work for Advancing New Standards in Reproductive Health (Upadhyay, Koenig, Ko, Biggs). Meckstroth is a paid staffer for Danco Laboratories, a drug company that sells only one drug (mifepristone). Valladares works for California Latinas for Reproductive Justice. Aside from Meckstroth, the report did not disclose these conflicts of interest.

The report misleads women by underreporting medication abortion risks.


Please enter your comment!
Please enter your name here