High-quality early childhood programs can reduce the number of children diagnosed with certain learning disabilities by third grade, according to a study published Tuesday in the Educational Evaluation and Policy Analysis journal.

The study, conducted by Clara G. Muschkin, Helen F. Ladd and Kenneth A. Dodge of Duke University, could have significant implications for reducing the financial burden special education services place on municipal budgets.

The researchers explored how two early childhood initiatives in North Carolina affected the likelihood that children would be placed in special education by the end of third grade. It focused on a preschool program for four-year-olds from at-risk families and a program that provides child, family, and health services for children from birth through age five. The study tracked 871,000 children who were born between 1988 and 2000 and were enrolled in third grade between 1995 and 2010.

Children who participated in the More at Four preschool program, now called NC Pre-K, were 32 percent less likely to be placed in special education by third grade, compared to peers who did not participate in the preschool program, the study found.

Those enrolled in the Smart Start program for children from birth through age five were 10 percent less likely to be receiving special education services by third grade, the researchers found.

Together, both programs reduced the likelihood of third-graders receiving special education services by 39 percent, the researchers said. That could translate into significant savings, as special education in the U.S. costs almost twice as much as regular classroom education.

“It shows a level of benefit not only in academic terms but also financially, because special education services are so expensive,” Muschkin said. “This gives policy makers useful evidence that investments in early childhood education are a source of significant cost savings for the state.”

Some small studies have suggested children enrolled in high-quality preschool gain lifelong benefits, such as better jobs and higher wages and are less likely to be involved in the criminal justice system and to receive social services.

But in 2013, a large-scale study of nearly 5,000 children enrolled in Head Start, the federal government’s early childhood program for low-income children from birth to age five, found that progress in literacy among those children fades by the end of third grade. Critics of President Obama’s push to make preschool universal for low-income four-year-olds have pointed to the Head Start study in arguing that heavy investment in preschool might not be cost effective.

The Duke study suggests that high-quality preschool might not only benefit children who would otherwise need special education services, but that it has a “spillover” effect that benefits other children, Muschkin said.

Children in the same high-quality preschool programs as those who were enrolled through “More at Four” also benefitted, because the program was high-quality, Muschkin said. “The state funding goes to the slots in the classroom, but because the quality has to be high, the other children in the classroom also benefitted, even though they didn’t qualify for More at Four,” she said.

The children who attended state-funded early childhood programs also were ready for kindergarten and beyond, which made their public school classrooms better for their peers, she said. “When kids come to school ready to learn, it’s better for all kids,” she said.