In the last 12 years, 18 children in Maine have been killed in homes where state child welfare officials knew that the children or their siblings were subjected to abuse or neglect, sometimes over a period of years.

An additional 34 deaths that were ruled accidental or of natural causes occurred in homes where abuse or neglect was substantiated.

Those revelations raise questions about whether any of the deaths could have – or should have – been prevented.

“The short answer is that it’s impossible to tell on a case-by-case basis because you just don’t have all the information,” said Christine Alberi, the state’s child welfare ombudsman. “But you could certainly make the case that some might have been preventable.”

The Maine Department of Health and Human Services released detailed information this week, in response to a request from the Press Herald, about 117 deaths dating to 2007, the first year the department automated its child protective system. The list includes three categories: deaths as a result of homicide, regardless of whether DHHS was involved with the family; deaths with findings of abuse or neglect made by the department as a result of death, again regardless of whether there was child protective history; and deaths where a child’s family had DHHS involvement before or during the child’s life, even if the cause of death was natural, accidental or undetermined.

The records show that, long before the high-profile deaths of 4-year-old Kendall Chick in December 2017 and 10-year-old Marissa Kennedy in February 2018 , there were many child deaths that occurred after clear warning signs were present.


“These deaths may not all be preventable, but we could really lower the rate,” said Dr. Stephen Meister, an Augusta pediatrician and member of the state’s Child Death and Serious Injury Review Panel. “One of the biggest things is changing behavior and identifying the families that have multiple risk factors, which dramatically increases the likelihood of death, accidental or otherwise. So a death may not be the result of abuse, but can be linked to an increased number of hazards.”

It’s difficult to determine how Maine fits into the national picture on child deaths. The most recent federal report by the U.S. Department of Health and Human Services’ Administration for Children & Families, from 2017, includes state child fatalities and the rate of fatality per 100,000 children. Maine was the only state that did not submit data.

Department of Health and Human Services spokeswoman Jackie Farwell said Commissioner Jeanne Lambrew was unavailable for an interview Wednesday.

Although the information provided by the department does not include any names, locations or exact date of death, some cases were identifiable based on the age of the child and public reporting about them.

In May 2012, 2-month-old Ethan Henderson died from brain injuries caused by his father, Gordon Collins-Faunce. Information provided by DHHS shows that a child protective caseworker had conducted an assessment, the process to determine whether a report of abuse or neglect has occurred, one month earlier. Before that assessment was complete, the boy was killed.

In July 2012, 2-year-old Brooklyn Foss-Greenway of Fairfield died while in the care of a 10-year-old baby sitter, who later was charged with manslaughter. State officials noted substantiated neglect at the time of death, but the girl’s family had a child protective services history, although all contacts occurred before Brooklyn was born.


In January 2011, 2-year-old Brantin Webster of Searsmont died of internal injuries he suffered while rough-housing with an older brother. Although DHHS did not find neglect or abuse related to the death, the family had a long case file dating to 2001.

In October 2015, 3-month-old Leo Josephs died from injuries consistent with shaken baby syndrome. His father, Eugene Martineau, was later sentenced to 15 years in prison. Martineau admitted that he violently shook his son. Police investigators had testified that Leo had suffered from ongoing abuse, including fractured ribs and wrist. The DHHS summary said that a case was opened in September 2015, and there was still an ongoing investigation at the time of the child’s death. Services offered to the family included medical care, public health nursing, home visiting and housing. The department concluded that there were substantiated incidents of abuse and neglect.

In June 2013, a 4-month-old girl died on the same day of an assessment, according to DHHS. There was a report of suspected abuse or neglect made three months before then, but it was not assigned for an assessment. The girl’s death was determined to be natural, but substantiated neglect also was noted.

There have been 10 child homicides since 2007 where there was no prior family involvement with child protective services, the most recent being in March, when 14-month-old Quinten Leavitt of Presque Isle was shot and killed by his father, Matthew Leavitt, who then killed himself.

The information provided this week by DHHS also included more detailed information about Kendall Chick, who was killed by her primary caretaker, Shawna Gatto, the fiancée of her paternal grandfather. That information revealed that Kendall’s parents, or at least one parent, had contact with child protective services dating to 2008 and involving two other children. It also showed that two children were removed from that home – one in 2011, the other in 2013 – and did not return.

When Kendall was born in November 2013, DHHS was notified. The agency opened an initial case and provided parenting education services, housing assistance and substance use disorder treatment. That case was closed in August 2014.


In May 2016, when Kendall was 2½ years old, there was a report of suspected abuse or neglect that was substantiated. Kendall was removed from the home and placed with her grandfather Stephen Hood and his fiancée, Gatto, who had sometimes baby-sat for the girl.

That placement was what the department calls a safety plan, a loosely structured voluntary agreement that provides parents an opportunity to obtain services with the goal of getting the child back. That didn’t happen in Kendall’s case. The state closed its case involving Kendall in February 2017, advising Hood and Gatto that they should consider seeking legal guardianship of the girl. They didn’t pursue that, though. Ten months later, she was killed in their Wiscasset home.

Alberi, who as ombudsman has access to Kendall’s entire case file, said it was clear that policies and procedures were not followed in her case. Specifically, she said, the infrequency of in-home visits was problematic. Caseworkers visited Kendall in Hood and Gatto’s home only twice in a span of 10 months, even though agency rules mandate monthly visits.

But Alberi said it’s also entirely possible that when the state closed its case on Kendall in early 2017, there were no signs that the girl was in danger.

“I think the standard is that caseworkers are not closing cases unless they are reasonably sure that a child is safe,” she said.

Farwell, the DHHS spokeswoman, said the agency doesn’t comment on personnel matters. However, she said the caseworker assigned to the Kendall Chick case no longer works for the department.


Kendall’s case, and others included in the historical data released this week, coincided with a prolonged period of tumult within the Office of Child and Family Services. High staff turnover at the caseworker level, and among administrators, led to a leadership vacuum and also helped create an overburdened system. Workers were routinely responsible for far more cases than they could reasonably manage and were further bogged down by burdensome paperwork.

From 2008 through 2016, confirmed cases of physical abuse of children in Maine increased by 52 percent even though the overall number of abuse and neglect cases declined slightly during that time.

Many of the caseworker concerns didn’t come out until after the deaths of Kendall and Marissa Kennedy, the 10-year-old Stockton Springs girl who was killed three months after Kendall.

The information released this week does not include Marissa’s case because that is still tied up in court. Her mother, Sharon Carrillo, and stepfather, Julio Carrillo, have each been charged with murder and await separate trials.

Alberi said if policies weren’t followed in Kendall’s case – the frequency of in-home visits, for instance – it’s possible that happened in other cases, too. It’s impossible to determine that from the summary information provided by DHHS.

But she also said some deaths, even in cases where abuse or neglect had taken place, were accidents. Of the 117 child deaths reported to OCFS since 2007, 29 were the result of co-sleeping.


Co-sleeping deaths are far more likely to occur in homes where one or more parent struggles with substance use disorder. Meister, the Augusta pediatrician, said the fact that co-sleeping deaths increased during the height of the opioid crisis was not an accident. In fact, since 2014, in 30 percent of all child deaths reported to DHHS, the child’s family received substance use disorder treatment services.

State officials in April launched a Safe Sleep educational campaign to warn parents about the dangers of co-sleeping. Meister said sometimes the reality is that low-income families don’t have access to cribs or bassinets, even though there are programs to provide them to families in need.

Many changes have been made within the Office of Child and Family Services dating to last year, the biggest of which has been increased funding for new positions and salary increases for existing workers. The office also is in the process of implementing a new record keeping system that it hopes will cut down on paperwork.

Lambrew, the DHHS commissioner, and new OCFS director Todd Landry both have said more changes are likely, but they want to hear from workers to make sure they have buy-in.

Already, caseworkers have stopped relying on safety plans, using them only when a child stays in his or her own home.

Alberi said that although there have been many improvements over the last year, some of the changes take time to implement.

“One thing I think the department has recognized is that child welfare employees need more training on how to conduct assessments, or investigations,” she said.

Meister said from what he’s seen, caseworkers have been working harder to take out the subjective parts of their decision making and use data more effectively. He said that might result in more children being removed from their homes, but that could be a trade-off to a safer system that keeps more children alive.

Staff Writer Joe Lawlor contributed to this story.

Related Headlines

Comments are no longer available on this story