A recent study shows alarmingly high attrition of first-time parents from STEM careers: 43 percent of new mothers and 23 percent of new fathers leave STEM annually in the United States. This high departure rate has sparked great concern about the cultural expectations and lack of support for new parents. To stop this massive leak from STEM fields, we must have an honest conversation about what parents in science face.


Early career scientists are in a precarious position during their most fertile years and women who want to start families frequently face a biological timeline that doesn’t neatly align with their career paths. A PhD in STEM fields takes an average of five years after a BS or MS, and the transition between a graduate or postdoctoral or longer-term position is tight, occurring over two to four years. That also happens to be the time when many STEM parents are starting families. Although a few federal agencies do have parental leave policies, early career scientists are often at the mercy of their supervisor when it comes to negotiating any type of leave. The attitudes of senior scientists toward parental leave and parenting as a scientist vary widely—which can lead to heated debate.

U.S. scientific institutions fall short across the board. Approximately 50 percent of OECD nations offer at least 14 weeks parental leave, as advocated by UN International Labor Standards since 1952. In contrast, the United States is one of only four countries (also Liberia, Papua New Guinea, Eswatini) lacking mandated leave for new mothers. Our Canadian neighbors qualify for 35 weeks of paid leave, with an optional extension. In Europe, parental leave is a minimum of 14 weeks to a year or more, funded by employer/employee contributions similar to those supporting U.S. unemployment benefits.

The only federal mandate in the U.S. is the Family Medical Leave Act (FMLA), providing up to 12 weeks of job protection. However, one must be employed for a year (24 hours/week) with at least 50 employees (within 75 miles). Additional limitations may be placed on FMLA use by individual institutions. At Montana State University, only six weeks of sick leave may be used for the birth of a child, with added time coming from annual /vacation leave or unpaid. As is, only 12 percent of eligible employees actually utilize the FMLA benefit and are predominately upper middle class, college-educated and white.

Because early career scientists are expected to move frequently, accruing sufficient leave prior to giving birth is often impossible. FMLA-based leave is expensive because an employee must continue paying health insurance premiums to retain coverage, a hefty price tag. The outdated expectation that women will wait until after tenure to start a family is unrealistic, particularly as the list of qualifications needed to obtain a permanent position continues to lengthen.

As of 2019 only six states (CA, NJ, WA, RI, NY and MA) and D.C. have paid leave laws. Even as a few states step up to the plate, these benefits are far from universal and vary between academia, biotech and the scientific federal workforce. The private sector seems faster to respond to cultural shifts that may influence employee retention. Across the majority of U.S.-based biotech companies, six paid weeks are extended to both parents.

Large academic institutions are demonstrably slower to change, and differences between them are striking, creating a patchy landscape for new parents to navigate. A recent analysis found that paid leave ranges from zero to the bare minimum (e.g., 60 hours!), to more generous policies supporting up to 22 weeks off. In some cases, full-time faculty are offered reduced teaching loads and paid leave regardless of start date, negating the need to accrue leave. However, this is not the norm for most U.S. institutions.

Variability exists even within the same state. University of California faculty enjoy months-long leave, but in the separate C.S.U. (California State University) system, parents get only 30 days. North Carolina is similarly contrary, with state universities in Chapel Hill and Wilmington getting a semester paid, but other campuses receiving none (e.g., U.N.C. Charlotte). Lecturers and non–tenure-track faculty usually don’t receive the same benefits as their tenure-track colleagues.

The support gap in U.S. academic institutions is huge. On the whole, the majority of institutional parental leave remains unpaid, with few opportunities to supplement with accrued leave or to compensate lost salary. At 120 out of 197 PhD-granting institutions where women do receive parental leave, the average 14.2 or 11.6 weeks (women, men) is covered by accrued vacation or sick leave. Federal workers (NIH, DOE, NOAA, EPA, etc.) have the same basic leave policies but overall, access to paid family leave for scientists across institutional levels and sectors is incredibly confusing, not guaranteed, geographically spotty, and overly reliant on accrued sick and annual leave time.

If funded by the NIH or NSF, tenure-track faculty can take advantage of the Family Friendly Initiative, but only under certain award types. NIH K (career) award recipients are considered institution employees, while NRSA fellows (either parent) may receive up to eight weeks. Awardees of the R01 grant (or higher) can fund a temporary replacement of project personnel while trainees or the principal investigator are on leave. However, the Montana State university grant office and an NIH program officer, asked separately, had never seen this benefit requested, suggesting that it sounds good on paper but is impractical and underutilized.

For NSF researchers, funds can support temporary personnel replacement, dependent care and pay for accrued leave. NSF CAREER awardees can request up to $12,000 for staff support while the PI is on leave; however, at time of writing, neither agency allows salary support during leave. A postdoc at Oregon State University ran into this predicament when expecting her second child in 2017. Her supervisor permitted three months of paid leave, funded by a grant she helped write. However, to comply with funding guidelines, the university required she take a month unpaid. Because federal agencies require frequent progress reports and there is no accommodation for parental leave, reporting zero percent effort for a grant period can be a violation.

Graduate students face more hurdles as they are often paid by grants or supported with teaching assistantships, and do not accrue sick or annual leave time. Only those holding specific fellowships (e.g., F31 or GRFP) have access to some form of paid leave. Institutional policies about classwork and enrollment are just as varied.

Transitioning back to work. Transitioning back to full-time employment can be hard, especially when short leaves that are partially or completely unpaid must be patched together, alongside the added stress of having less income. The lack of supported parental leave in the U.S. is perplexing considering that a number of studies demonstrate the numerous benefits associated with paid leave for both men and women. Sufficient leave paves the way for a smoother transition back to work as well as better outcomes for parents and their babies. Paid leave reduces infant mortality by up to 10 percent, and women are less likely to experience depression even later in life. Children receive higher rates of immunization and increased breastfeeding duration. Paternity leave is becoming more common, and men who take paid leave continue to share in child-rearing responsibilities years later, changing the long term dynamics of their families as well as shaping their children’s chances of succeeding in school.

Finding suitable care can be challenging and time-consuming, especially without sufficient parental leave. Over half of Americans live in “childcare deserts,” with an insufficient number of licensed facilities. Even if childcare is available, care is unaffordable for more than 70 percent of American families. But early career scientists are often financially vulnerable or in debt and even just a year at home with children can impact a science career, particularly early on when establishing one’s reputation. A new mother can quickly find herself with a hard-fought graduate degree, but without good employment prospects or affordable childcare. Making things more complicated, dual career couples are often coordinating the shift to the next career stage simultaneously. As such, even the best intentions of highly egalitarian couples often don’t result in two successful careers.


A number of scientific organizations recognize these challenges and are providing grants in support of care during conferences, so parents don’t miss out on important networking opportunities. However, only 68 percent of conferences currently provide child care support, leaving many parents opting to forego conferences altogether. Some institutions are also making childcare more accessible to postdoctoral scholars and students. For example, Oregon State University has two high quality centers, with subsidized tuition for trainees’ children. Similarly, Montana State University provides two preschools for ages 3 and up, but not younger. An increase in childcare programs such as these will go a long way towards improving the retention of parents in STEM.

There is momentum for paid family leave and 12 additional states are considering paid leave policies. Even President Trump’s 2020 budget calls for six weeks of paid family leave. Both Republicans and Democrats in congress have proposed paid family leave legislation, although the details differ considerably. As mothers, we fully welcome this progress at the state and federal level and its importance for the next generation of women scientists to succeed. In the meantime, we have some concrete ideas for how to support new STEM parents.

Federal agencies, especially NIH and NSF, should fully support family initiatives. Specifically this means allowing salary support during parental leave, even at a reduced rate (e.g., 75 percent) for trainees (postdocs, graduate students, etc.). This level of support is not unprecedented. Without similar efforts in the U.S., the so-called family-friendly and worklife balance initiatives fall far short of actually supporting STEM parents, particularly women.

The United States should adopt the United Nations International Labor Standard of 14 weeks (minimum) paid leave for parents. U.S. Senator Kirsten Gillibrand is currently sponsoring legislation to establish a national paid leave program, which supports a wide spectrum of caregiving needs that is similar to programs in place in the majority of other countries.

Institutions should support extended sick leave to fund time off following the birth or adoption of a child. Arbitrary limits are detrimental to staff and faculty and place an undue burden on new parents.

Institutions should invest in on-site childcare services and drop-in care that can be utilized during the first few months after a child is born. This will allow new parents to check in with team members, attend meetings and continue mentorship of trainees, allowing a gradual transition back to work that would greatly reduce the stress associated with the arrival of a new child. Childcare should be subsidized for those with lower incomes, particularly students and postdocs.

Both parents should be given full support to take sufficient family leave. Fathers are taking more time off now compared to 20 years ago, but we must continue to facilitate paternal leave and support it 100 percent to promote better gender equality.

Last but not least, we must use our voices to advocate for paid family leave. Check if your state is considering a paid family leave policy and call your representatives to voice your support Join the nearest 500 Women Scientists pod to get started. Our voices are finally starting to be heard, but we must shout far louder.