A new U.K. study that correlates longer duration of breastfeeding with higher academic test scores at age 16 is fanning the flames of the perennial infant feeding debate.

Parents often feel tremendous pressure to breastfeed after birth and continue to feel that way when they return to work, which presents a new slew of challenges and obstacles. If parents choose not to breastfeed or are unable to exclusively breastfeed, they often feel shame and guilt. New parents are particularly vulnerable to accusations that they are not doing the right thing or acting in their child’s best interest, and touting the benefits of breastfeeding in front of them can feel unnecessary and cruel.

Is breastfeeding so beneficial that promoting it is worth alienating families who don’t breastfeed or don’t breastfeed for the two years recommended by the World Health Organization and the American Academy of Pediatrics?

Experts aren’t in agreement. Breastfeeding advocates believe we should highlight all of its potential benefits not only to encourage families to breastfeed but also to help shore up support for nursing, leading to policies like the PUMP Act, which legally recognizes workers’ need for time and an appropriate place to pump breast milk while on the job.

Critics of such studies say their findings are of limited value and do a disservice to parents struggling with guilt over their feeding choices.

Economist and parenting data expert Emily Oster, who has written previously about how she believes that the benefits of breastfeeding are frequently overblown, posted a video on June 6 for her Instagram followers in which she questions the validity of the U.K. study’s findings on the same grounds that she has criticized previous similar studies.

“It’s so easy to find this correlation because the characteristics of families who breastfed their kids are totally different,” Oster, who recently joined an advisory board for a formula company, told viewers.

Breastfed children are more likely to come from wealthier families and to have parents with higher levels of educational attainment. It is virtually impossible to untangle the impact of breastfeeding alone from all these other confounding factors.

Oster calls the study a classic example of correlation, not causation. While breastfed babies may have become higher-scoring adolescent test-takers, it doesn’t necessarily mean that breastfeeding was what led to their higher scores.

The Millennium Cohort Study in the U.K. followed a group of approximately 5,000 children born between 2000-2002. The children’s parents were asked whether or not the children were breastfed and, if so, for how long. This data was then analyzed with the students’ scores on their GCSE exams, which all British youth take around age 16.

Researchers found that students’ scores rose in relation to how many months they had been breastfed (less than two months, two to four months, four to six months, six to 12 months, or more than 12 months).

Those who breastfed for 12 or more months failed their English GCSE at a rate of 19.2%, compared to 41.7% of those who never breastfed. And among those who were breastfed for 12 or more months, 28.5% scored high (a “high pass”), compared to 9.6% among those who were not breastfed.

The same correlation did not hold for the Mathematics GCSE exam.

The paper’s authors refer to these test score gains as “modest.”

“The differences in the exam results between those children who were never breastfed, and those children who were breastfed for longer durations, those differences are not large. That’s why we used the word modest,” Dr. Reneé Pereyra-Elías, the study’s lead author, told HuffPost.

Pereyra-Elías also explained that this observational study was subject to confounding factors such as maternal economic status and educational attainment.

“Moms who have more years of formal education and moms who are better off are more likely to breastfeed their babies for longer,” said Pereyra-Elías, echoing Oster’s critique.

He and his coauthors did consider such factors in their analysis, including parental education levels and occupations, which Pereyra-Elías admits is not necessarily a perfect proxy for income.

“We have tried to control for the differences in socioeconomic circumstances and for the differences in maternal cognitive ability as well,” she said. “However, because it is an observational study, there may be still some confounding [factor] that was not accounted for.”

Pereyra-Elías continued, “If we were able to account perfectly for all confounding factors, there may still be some difference between those children who were not breastfed and those who were breastfed for longer durations. But these differences would be small, perhaps smaller than what it is now, which is already modest.”

Other studies that examined the relationship between breastfeeding and brain function, for example, IQ scores, have shown “conflicting” results, said Pereyra-Elías. Some find no association after controlling for confounding factors, but others do.

For example, one 2015 study from Brazil found that people who breastfed as infants had higher IQs by about 3.7 points (on a scale from 40-160). Again, this effect is small.

One theory behind such findings is that substances found in breast milk, such as polyunsaturated fatty acids, “promote neurodevelopment,” said Pereyra-Elías, leading to increased intelligence or academic achievement — but we do not understand exactly how this process works. We also do not know if these neuro-developmental benefits are derived from what’s in the milk, the act of nursing at the breast (which promotes bonding) or a mix of the two.

Overall, “breastfeeding has multiple potential benefits, not only academic results, so it should be encouraged whenever possible. And families, especially mothers, should receive all the possible support to do this. However, if [it’s] not possible, for whatever reason, mothers should not be judged, and they should not worry because the differences in academic results are not large,” Pereyra-Elías concluded.

When considering how realistic breastfeeding recommendations are for working families, it may be worth noting that only 9.5% of children in the study were breastfed for 12 months or more. Despite the recommendations to breastfeed for two years or more, breastfeeding for 12 months or beyond is not the norm, and perhaps it is counterproductive to suggest that there may be cognitive benefits of doing so when so many new parents are struggling just to make it to six months, or even two.

Janine Zee-Cheng is a pediatrician practicing in Indiana. They are also a lactation consultant and a parent of two.

“Yes, I recommend breastfeeding,” they told HuffPost. Zee-Cheng cites the health benefits of breastfeeding listed on the Centers for Disease Control and Prevention website: “You have reduced risk — not zero risk — but reduced risk of asthma, autoimmune disease, allergies, pneumonia, ear infections, SIDS.” NICU babies fed breastmilk have a decreased risk of necrotizing entercolitis (a life-threatening infection for preterm babies.) Breast cancer is also a decreased risk for breastfeeding people, and they have reduced chances of other health issues as well.

However, Zee-Cheng continued, “I don’t recommend breastfeeding at the cost of the parents’ mental health — or any health. And I don’t recommend breastfeeding to the exclusion of everything else.”

While breastfeeding has benefits, each family must weigh their unique needs when considering whether or how long to breastfeed. Zee-Cheng does not discuss the intellectual or academic benefits of breastfeeding with their patients, feeling that the research in this area is not as conclusive as the health benefits mentioned above.

“I don’t think you can divorce money from the ability to breastfeed,” they said.

“When you see a study like this, I just remind people that there’s a lot of different things that go into [it],” said Zee-Cheng. For example, breastfeeding may decrease a child’s asthma risk, but so will a lack of exposure to cigarette smoke, living in a low-pollution area and having a good air filtration system at school. And when it comes to cognitive ability, there is still disagreement about how accurately standardized tests like the GCSE measure kids’ academic potential, as a number of these tests have been shown to have a racial bias.

When discussing breastfeeding with families, Zee-Cheng emphasizes that parents should prioritize their relationship with their child.

“You will always do the best thing for the relationship that you have with your child,” they said. “And there are times when breastfeeding will harm the relationship rather than help it.”