behind the

About This Research

This research report and podcast is the result of a 9 month study into the hidden, uncensored emotional experience of pregnancy – what we are calling the Invisible Weight –  and how the U.S healthcare system fares under the needs of that weight. This story is told through the voices of pregnant Americans and experts on the frontlines of care.

Our hope for this research is to shine a light on the hidden side of pregnancy and the dire lack of emotional care in the U.S.

This research was conducted by Nonfiction Research and commissioned by Twill a digital-first healthcare company focusing on mental health and chronic illness who’s tools support over 18 million lives. Twill was contractually prohibited from influencing the results of this study

About Twill

Twill (formerly Happify Health) shortens the gap between need and care with digital-first tools that are as connected as the body and mind. Twill intelligently guides each person to the care they need, when they need it, in the way they want. Twill builds Sequences which combine evidence-based digital therapeutics, well-being products, community-based care, clinician-trained AI, and live coaching and telebehavioral health to address specific mental and physical health conditions.

In an effort to address the invisible weight of pregnancy that you’ve learned about in this research, Twill has developed a Sequence focused on pregnancy that offers a breath of maternal health support.

This Sequence brings together digital therapeutics, community and coaching. Learn more about Twill’s Pregnancy Sequence here.


This research was conducted between January 2022 and September 2022 and consisted of in-depth interviews with Amreicans pregnant in the last two years, experts and a nation-wide survey. The nationwide survey was conducted with Americans who have been pregnant in the last two years (including those who were pregnant at the time of the survey). Adhering to standard practices for quantitative representation, Nonfiction collected 1,029 survey responses. The one-on-one interviews were conducted with members of the same audience. To account for historically underrepresented experiences, the demographic makeup of this sample emphasized the presence of persons of color and those insured by Medicaid – but all findings were tested for broader applicability in the survey. Experts interviewed included OB/GYNs, certified doulas, midwives, nurse practitioners, clinical therapists and clinical social workers.


Who did this study?

Nonfiction Research, a market research firm based in New York.

Who paid for this study?

Costs were shared between Nonfiction Research and Twill, a digital health company focused on mental health and chronic illness. Twill was contractually obliged to not influence the findings of this research.

When did this research happen?

Research began in January of 2022 and concluded in September. The first nationwide survey was run in April of 2022. In response to the decision to overturn Roe v. Wade, the survey was run again in August of 2022. The results of the survey remained largely the same, not meaningfully impacting the narrative.

How many people were surveyed?

1,029 people completed our survey, all of whom were Americans and had been pregnant in the last two years (including those who were also pregnant at the time of the survey)

How can a little over 1,000 people represent every single pregnant American from the last two years?

They can’t, but neither can anyone interview every single pregnant American, no matter their time and resources–as much as we would like to.  If you survey enough people within a specific audience, however, you can feel confident your results represent the largest possible number of that audience. That confidence comes in the form of a number used widely in survey research, called the margin of error.

What is the “margin of error”?

The margin of error tells you just how close the results of your survey are to representing the total audience. The lower the margin of error, the closer your results are to representing your audience. In the case of this study, 1,029 people took our survey, out of an estimated 12 million pregnant people since 2020. We’ll spare you the mathematics, but that gives us a margin of error of 3%.

I’ve seen some surveys say, “respondents weighted to reflect the US Census”: what does that mean? Does your survey reflect the US Census?

The US Census is the most (but not completely) accurate picture of the demographic makeup of the United States. If we want survey responses to reflect the nation, we’d want the proportions of the people who take it to match the Census. If the subject of a study is small enough, however, like our audience, matching proportionally isn’t possible. Rather than matching this percentage of people from that state, or this percentage from that background, we focused on one thing and one thing only: their experience with pregnancy.

the invisible weight
the invisible weight

If you have interest in hearing more from the authors of this work, please reach out to

If you just want to chat, reach out to