CFSA, DCPS and the safety of children not in school buildings

Ever since the onset of the COVID-19 pandemic and the attendant lockdowns and school closures, there has been widespread concern about the safety of children who might be abused or neglected and are now spending school days at home. Losing school meant the loss of both a safe space for six hours a day and a set of caring eyes to report on any concerns. In FY 2019, school personnel made 42.9 percent of calls to the Child and Family Services Agency (CFSA) hotline in the District of Columbia. Data on hotline calls show that indeed the number fell greatly in the wake of the pandemic but began to recover in the current school year. The newest data, from January through March of 2021, show that hotline calls were well below 2020 levels in January and February 2021, then jumped above these levels when schools opened in March. These data suggest that there is still reason for concern about the safety of children who are not attending school buildings, and a need for CFSA and DCPS to make a plan to check on the safety of all children who have not recently been seen.

Current DC law requires that students five through 13 years old who have 10 or more full school-day
unexcused absences within a school year be referred to CFSA for “educational neglect.” Younger children are not required to attend school nor are their absences required to be reported. Schools are not required to report absences of youths aged 14-17, which are presumed to be the result of truancy rather than educational neglect, except under special circumstances.

Table One below shows the monthly number of calls to the CFSA hotline, known as “referrals,” from March 2020, when the pandemic emergency began, through March 2021, compared to the same dates of the previous year. (Note that March 2020 is included in both 13-month periods as the pandemic closures began in the first half of that month). DC’s sudden closure of schools for an extended spring break in March 2020 was followed by a chaotic virtual reopening, as schools and nonprofits strove to get children connected with computers and internet service. Not surprisingly the first three months of the pandemic produced a drastic drop in hotline calls, investigations, and substantiated cases of maltreatment. Reports stayed at basically the same level from April to August, unlike a normal year, when reports drop after schools close.

Chart One

Source: Data from CFSA Dashboard and data provided by CFSA for January-March 2020.
Note: March 2020 data are shown twice in this table

On August 28, CFSA issued new operating procedures for schools in response to attendance concerns under COVID-19. The guidance stressed that child welfare involvement can be an “invasive and traumatic” experience for children and should be a last resort, with schools taking the lead on dealing with chronic absenteeism. CFSA also noted the unique challenges that remote learning provides for parents and stressed that these problems are generally not best addressed by CFSA. Accordingly, CFSA provided a list of steps to try to reach families and recommended that schools should take “as many of these as possible” before calling the hotline. In the guidance, CFSA warned that a report would be screened out “when the reporting school has not demonstrated exhaustive efforts to make contact with a student and family to resolve attendance matters, and/or the report is missing critical information.”

We do not know what impact these guidelines had on referrals from schools once school started on August 31, 2020. There was a modest uptick in in reports in September with the opening of virtual schooling, though it was much less than what occurred in the previous year. In mid-September, CFSA completed a webinar and related documents to be used to train teachers about how to spot abuse and neglect in a virtual environment. These materials were submitted to the Deputy Mayor for Education for distribution to LEAs in mid-September with the direction that all teachers assigned to virtual classrooms complete the trainings. 

Perhaps in part due to the new guidance, hotline calls rose In October, and began to approach the previous year’s level in November and December. Unfortunately, the newest data shows that the improving trend did not continue in January 2021, when only 1,224 referrals were made compared to 1,862 in January of 2020. Reports did increase in February and especially in March, after about 20 percent of the school system’s population returned to school early in February. In fact, there were many more calls (2,253) in March 2021 than in March 2019 (1858), when all schools were open.

The big uptick in March referrals is concerning in view of the fact that only about 20 percent of students returned to school buildings that month. One wonders how many calls would have come in if all students had been back in school buildings. It is also interesting in light of a change in practice that CFSA announced early that month. As quoted in a letter from CFSA Brenda Donald to DC Council Member Christina Henderson, CFSA a sent a message to DC Public Schools (DCPS) and the Public Charter School Board early in March 2021 describing a new practice in screening referrals for educational neglect “due to the influx of reports for potential educational neglect and furthermore the city-wide attendance issues caused by the COVID-19 pandemic.” Under the new procedure, CFSA would reject any educational neglect referral for a family with whom the school or CFSA had been in contact within the previous 10 days of school. Exceptions would be allowed for families with a recent substantiation for educational neglect, a recent concern for other types of abuse or neglect, or a “clear demonstration of parental disregard for student’s education preceded by exhaustive school/community based efforts.”

In response to a letter from Council Member Henderson expressing concern about the new screening practice, Donald replied in her letter that in the second half of the school year CFSA often receives “a high volume of reports of children who have accumulated more than 10 absences.” She also stated that “on average, approximately 87 percent of the educational neglect referrals that CFSA receives are screened out and 64 percent of screened in referrals are unsubstantiated.” And she explained that the change was made in order to “meet report triage timeliness and ensure that the agency resources are used most effectively by ….responding to only those reports that contain allegations of parental neglect and/or document the impact of the absences on the student’s academic performance or general well-being; while screening out those reports that are submitted by schools for compliance purposes only.”

We do not have data on the proportion of educational neglect referrals that have been screened out during the pandemic. But it is interesting to look at the proportion of all referrals that were screened out during the pandemic, as shown in Chart Two. The screened-out percentage dropped sharply in May and June 2019, perhaps because of the drop in reports by teachers, who according to CFSA make many reports “for compliance purposes only.” After those months, the percentage of referrals that was screened out during the pandemic was pretty similar to pre-pandemic levels, except for March 2021. In that month, CFSA screened out 78.6 percent of calls, compared to 67.0% in March 2020 and 71.8 percent in March 2019. Perhaps this change reflects CFSA’s new screening procedures.

Chart Two

Source: Data from CFSA Dashboard and data provided by CFSA for January-March 2020

CFSA’s intent to keep a lid on educational neglect referrals is understandable. Administrators are afraid of being overwhelmed by referrals of educational neglect. Moreover, there has been considerable pushback by activists in jurisdictions like New York City about reports of parents being investigated for educational neglect when they were not able to obtain computers or internet service. However, it is important to note that while categorized as “educational neglect,” referrals from schools about absences often serve a much broader purpose than ensuring that children are going to school. Chronic absence is often the first indicator that the child is not safe. It may even be an indicator that the child is missing. In the case of Relisha Rudd, who disappeared in 2014 and was never found, 18 days of absences did not trigger a report to CFSA because the absences were excused with the help of a bogus “doctor” who was probably Relisha’s abductor.

So what can be done? In this extraordinary time of virtual education, CFSA ought to change its emphasis to protecting children (its actual mission) rather than avoiding referrals that it considers to be a nuisance. Rather than threatening schools with rejecting inappropriate referrals, CFSA should work with DCPS to make sure all students are safe. CFSA should first modify its recent screening changes. If a family has been in touch with a school within the past ten days, that does not mean the school has been in touch with the child, who may still be in danger. More important, CFSA and DCPS need to make a plan for checking up on all schoolchildren who have not been seen recently in the classroom. If the issue relates to technology, CFSA can facilitate solving the problem without substantiating the parents for educational neglect.

DCPS welcomed more students to school buildings on April 19, with the beginning of the fourth quarter, but most of these students were in wealthier parts of the city, where abuse and neglect are less prevalent. Without much hope for more in-person student contact before the fall, it is crucial that CFSA and DCPS collaborate to assure the safety of our most vulnerable children.

Hotline calls, investigations, substantiations, and services still gravely suppressed by pandemic

In August, I wrote about the drastic decline in reports to CFSA’s hotline, investigations, and substantiated abuse and neglect allegations in the wake of the Covid-19 emergency, which closed schools and resulted in increased social isolation for children. CFSA has released data from June through October 2020, and the same trends are continuing. Reports, investigations, substantiations, and in-home case openings are down drastically. Foster care entries have decreased as well, but that drop seems to have occurred mainly before the pandemic. With plans for reopening schools up in the air, it is long past time to worry about the children who are not being seen in school or child care and to find alternative ways to reach them until schools and childcare centers are fully open. Moreover, these frightening data lend support to those calling for schools to reopen soon–especially for children of elementary-school age or younger, who are incapable or less capable of seeking help.

In the District, Mayor Muriel Bowser declared a public health emergency on March 11, 2020 and schools closed for in-person classes on March 13. After a two-week spring break, online learning began on March 24 and the stay-at-home order went into effect on April 1. Online learning ended on May 29, nearly a month early, and the same day that the District entered Phase One of reopening. Schools started on a virtual basis again on August 31. In its public-facing Dashboard, CFSA posts data for each quarter 45 days after the quarter ended. The data for the July through September quarter were posted on November 13, 2020. Using this data, we compared the numbers of reports, investigations, dispositions, in-home case openings and foster care placements since the pandemic with the numbers during the same months of 2019.

Reports to the Hotline

Hotline calls are generally a family’s first contact with CFSA and thus the earliest indicator of the effects of the pandemic. Figure 1 shows the number of hotline reports per month in FY 2019 and 2020. In FY 2019, the number of reports increased every month until May, dropped to a much lower level in July and August when schools were closed, and then bounced up in September with the opening of schools. The pandemic year of 2020 looked very different. The number of calls fell from February to March with the closure of schools, followed by a much larger drop in April, with the number of calls staying fairly flat until a modest rise in September with the opening of school. It’s as if summer vacation started in March, with a slight increase of reports when school started again.

Comparing the months across years shows that in January and February 2020, before the pandemic shutdown began, there were actually more hotline calls than there were in the same months of 2019, suggesting that the year would have seen increased reports if not for the pandemic. But in every month of the pandemic, the number of hotline calls in 2020 was considerably less than its counterpart in 2019. (The actual numbers are provided in Table 2 at the bottom of this post.) The biggest year-to-year differences were in April and May 2020, after the pandemic emergency began, when calls were down by 64 percent and 67 percent respectively over the same months in the previous year. In July and August, the differences between 2019 and 2020 were less drastic, which is to be expected because schools are normally closed during the summer months. But still reports were down by nearly a quarter in both July and August 2020 from the previous year. The total number of hotline calls received between March and September (roughly the period affected by covid-19) fell from 14,245 in 2019 to 9,780 in FY 2020, a decrease of 31 percent.

It is likely that the school closures were a major reason for the drop in hotline reports. Many schools, especially in the poorest areas, struggled to engage many of their students in distance learning last spring; some students were missing for the entire quarter. In addition, signs of maltreatment may be harder for teachers to observe online, although a virtual platform does allow them to observe incidents in the home that might otherwise go unreported. Moreover, DCPS closed several weeks early, offering not even virtual education in June. The lower level of reports even in the summer months may be due to families being more isolated due to the pandemic, spending less time with friends, neighbors, and extended family members who might report suspected abuse or neglect, as well as putting off visits to doctors, another common reporting source.

While monthly data on reporting source are not available, annual data shown in Table 1 support the hypothesis that a decline in reports from school personnel was a major factor behind the fall in reports overall. In 2019, presumably a fairly “normal” year, school and childcare personnel made 42.9 percent of all reports to the hotline. This is larger than the national percentage of 21 percent for these groups in FY 2018, probably because the District requires schools to report to the hotline all students who have more than 10 unexcused absences during a school year. These data show that these groups made a smaller share of reports in the recently completed Fiscal Year 2020 (in which more than half of the school year was virtual) than in FY 2019. In FY 2019, childcare and school personnel made 7,704 reports, which was 43 percent of all reports to the hotline. But in 2020 they made only 5,006 reports, which was only 36 percent of all reports. Counselors, therapists and social workers also made fewer reports in FY 2020, probably reflecting services that were suspended and or transitioned to a virtual platform. Their reports dropped from 2,342 to 1,702, or 13 percent of reports, which was not a large percentage change. On the other hand, law enforcement and medical personnel made an increased number and share of reports. Law enforcement personnel made 1,938 reports in 2020, a slight increase over the 1,891 reports they made in FY 2019. Their share of total reports increased from about 11 percent to 14 percent of all reports. Medical personnel made 965 reports in FY 2020, a big increase from the 866 reports they made the previous year, and their share of reports increased from five to seven percent. Friends and neighbors made fewer reports, but these reports were a slightly higher fraction–about 14 percent) of those received. Clearly no group could make up for the missing reports from teachers, so the total number of reports for the year was 14,046, down from 17,960 in 2019. The increasing number and share of reports due to law enforcement and medical personnel reflects their status as essential workers who have continued to see District residents in person. There have been anecdotal reports from many areas that maltreatment cases that do come to the attention of child welfare tend to be more severe; this may reflect the increased role of these reporters, who are more likely to see children who are seriously injured, and the reduced role of teachers, who are often said to report concerns that do not rise to the level of abuse or neglect.

Table 1: Hotline Reports by Source, 2019 and 2020

Reports accepted for investigation

A hotline call can be screened out as not meeting the requirements for an investigation, referred to another agency, or accepted for investigation. The number and percentage of reports accepted for investigation is shown below in Table 2.  The percentage of hotline calls accepted was higher each month during the pandemic period (especially in May and June) than in that same month of the previous year. This pattern suggests that the reports made during the lockdown tended to be more serious, with the less serious reports less likely to be made. Such a trend has been observed in other jurisdictions where data have been analyzed in more depth, as reported in my national blog, Child Welfare Monitor.  As mentioned above, some commentators have suggested that teachers in particular make many reports that do not merit investigation. Virtual schooling may have screened out some of this “noise.” CFSA data provide evidence for a small winnowing effect but one that was far from enough to make up for the drastic drop in the number of reports.

Table 2: Hotline Calls Accepted for Investigation

Completed and Substantiated Investigations

In addition to the number of calls accepted for investigation each month, CFSA also reports on the number of investigations completed each month. This is a different number because investigations can take a month or sometimes even longer to complete. So we can expect a time lag in observing the effects of the pandemic on the number of investigations completed. Also there is a small number of investigations that are not completed, as explained on the dashboard. These of course are included as part of total, but not completed, investigations.

Figure 2 shows a similar pattern of differences in completed investigations as shown in Figure 1 for reports, with the time lag delaying the effect by a month. In 2019, completed investigations mostly increased from January through June and then fell through September. In the pandemic year of 2020, the number of completed investigations fell precipitously in April and May, with the onset of the pandemic, drifted slightly upward through August, and fell slightly in September.

Turning to the numbers, in the pre–pandemic months of January and February 2020, the number of investigations was 24 percent and 36 percent higher than it was in January and February 2019. (Table 2 below the text shows the numbers and percentage changes.) It appears that the agency was on track to have more reports and investigations in 2020 than in 2019, if not for the pandemic. Completed investigations fell in April but it is not until May that the number of investigations completed in FY 2020 dropped below that of FY 2019. Once the number of investigations began to fall, however, it dropped like a stone. In May, only 230 investigations were completed compared to 565 in 2019–a drop of 59 percent. There were 56 percent fewer investigations completed in June 2020 than in June 2019. The gap between FY 2019 and FY 2020 lessened in July but was still large at 39 percent, reflecting the fact that many of these investigations would have begun in the first half of June, when school was still in session in 2019 but not 2020. It was only in August and September that the year-to-year difference declined to 21 percent and 18 percent respectively, reflecting the time lag between reports and completed investigations. It is worth noting that the year-to-year difference was still significant even in those months. This presumably reflects the continuing suppression of hotline reports during the summer of the pandemic. The total number of investigations completed between April and September (roughly the period affected by covid-19) fell from 2,716 in 2019 to 1,787 in FY 2020. That was a decrease of 25 percent–somewhat less than the difference in reports, reflecting the fact that a higher fraction of reports was investigated in 2020.

An investigation can have several possible results. It can result in a finding of “inconclusive,” meaning the evidence is insufficient to prove maltreatment despite some indications it occurred; “unfounded,” which means “there was not sufficient evidence to conclude or suspect child maltreatment has occurred,” or “substantiated,” indicating that the evidence supports the allegation of maltreatment. (See the CFSA Data Dashboard for the full definitions of these terms as well as of “incomplete investigations.”) As shown in Figure 3, the monthly trends and yearly differences were very similar to those for investigations in general. The total number of substantiated investigations for April through September dropped from 643 in FY 2019 to 420 in FY 2020, a decrease of 35 percent. The percentage of investigations that was substantiated during that period stayed almost exactly the same from year to year at approximately 24 percent.

In-home cases opened

When an abuse or neglect allegation is substantiated, several things may happen, depending on the perceived level of risk to the child or children in the home. The agency may take no action, refer the family to a community-based collaborative, open an in-home case, or place the child or children in foster care. As Figure 4 shows, there was a drastic drop in in-home cases opened between March and April, coinciding with the closing of schools and the pandemic emergency. Case openings dropped 48 percent in April, 54 percent in May and 74 percent in June compared with the previous fiscal year.[2] Data were not available for the quarter ending September 30, 2020. Presumably the trend in new case openings reflects in large part the trend in substantiated allegations; the patterns appear similar but not identical in the two measures. The total number of in-home cases opened in the pandemic months of March to June dropped from 533 in March-June 2019 to 267 in the same months of 2020–a decrease of 50.0 percent.

Source: CFSA Data Dashboard, available at https://cfsa.dc.gov/service/cfsa-data-dashboard
Note: In-Home Data not available for July-September 2020

Foster Care Entries

It is not surprising that hotline calls, investigations, and substantiations all declined in the wake of the pandemic and associated closures. The big surprise is that foster care entries do not display the same pattern. There was a big drop in foster care entries in February 2020–before the pandemic closures hit. Foster care entries actually rose in March, April and May of 2020 before dropping sharply in June and a bit more in September. True, monthly entries into foster care were always lower in 2020 than in 2019, though only by one child during the month of July. The total number of children placed in foster care declined from 193 in March through September of 2019 to 119 in March through September of 2020.

But the big decrease in foster care entries appears to have occurred before the onset of the pandemic, as Figure 6 shows. The number of entries into foster care was falling throughout FY 2019, aside from slight rises in May and August. There was actually somewhat of a rebound in the spring pandemic quarter, while the summer quarter entries fell back to the January-March levels.

The downward trend in foster care placements that began before the pandemic may have been due to policy or practice changes. As we mentioned in an earlier post, the data suggest that there was a renewed push to “narrow the front door” of foster care starting in the third quarter of Fiscal Year 2019. As we reported then, CFSA’s Communications Director indicated that the fall in foster care entries reflected CFSA’s “continued commitment to keep children out of foster care by supporting families in their homes.” Of course, we do not know if placements would have been higher in the absence of the pandemic; perhaps the school closures and pandemic orders did suppress foster care entries somewhat. But clearly, the pandemic has had a more obvious effect on the opening of in-home cases, resulting in a big decrease in in-home case openings. This makes sense if we assume that CFSA had become more likely before the pandemic to open in-home cases in lieu of foster care cases.

Exits from Foster Care

Commentators around the country have expressed concern that covid-19 would create delays in the achievement of permanency for foster youth. Family reunifications could be delayed by court closures, cancellation of in-person parent-child visits and increased difficulty facing parents needing to complete services in order to reunify with their children. Court delays could also hamper exits from foster care due to adoption and guardianship. The monthly data in Figure 7 show that fewer children did exit foster care every month from March to September, especially in May and June, than in the same months in 2019. The difference between years declined in July and August and almost disappeared by September, so perhaps the agency and court were able to clear the backlog. The total number of children exiting foster care declined from 261 during the period from March through September 2019 to 178 in the same months of 2020, a difference of 83 children or 32 percent. A small part of the reduction is due to the City Council’s decision to allow older youth to stay in foster care instead of aging out during pandemic. There were 27 exits by emancipation in the third and fourth quarters of FY 2019 and only 10 in the second half of FY 2020. But this difference does not account for most of the drop in foster care exits during the pandemic period.

Figure 8 shows the trends in total foster care caseload over time, which reflects both entries and exits. The total number of children in foster care on the last day of Fiscal Year 2020 was 695, compared to 798 on September 30, 2019, for a decrease of 13 percent. This reflects a total of 217 entries and 321 exits during the year. [1] (The percentage declines are listed in Table 7 at the bottom of this post). Combining Dashboard data with previously published data shows that the number of children in foster care on the last day of the fiscal year has declined every year since FY 2012. However, the 13 percent fall during 2020 was greater than in any other year since FY 2014. It appears that this decline happened mostly before the pandemic. Comparing data from March to September of 2019 and 2020 shows that 74 fewer children entered foster care and 68 fewer children exited it, suggesting that the pandemic had little effect on the total foster care caseload.

Conclusion

The release of fourth-quarter FY 2020 data on CFSA’s data dashboard shows a continuation of the trends that were displayed in the second-quarter data released in August. The pandemic and its associated closures had the expected downward effect on calls to the child abuse hotline, investigations, substantiated maltreatment reports, and in-home case openings. What was not expected was that although foster care entries fell in FY 2020, most of that fall occurred before the pandemic and appeared to be due to other factors, such as policy and practice changes. It appears that by suppressing reports and investigations, the pandemic probably suppressed the opening of in-home cases more than it reduced removals into foster care. But in any case, it resulted in a loss of protective services for children. Child Welfare Monitor DC has long been urging CFSA to do more to reach the children who may be suffering behind closed doors through means like working with schools to track down no-shows, launching a public awareness campaign about child abuse and neglect, educating non traditional reporters to spot signs of abuse and neglect, and reaching out to at-risk families known to the system. These ideas are discussed in my national blog, Child Welfare Monitor. With no clear plan for reopening schools, it is hoped that CFSA will now make this their top priority. Moreover, the data are sobering enough to support a return to school soon for children of elementary school age or younger as soon as possible because school is such a crucial safety net for abused and neglected children.

[1] The total is 694 rather than 695, which presumably reflects a small inconsistency in counting of entries and exits.

[2] There was an anomaly in the in-home case data that casts some doubt on the accuracy of the case opening numbers. We are awaiting clarification from CFSA.

Additional Tables

Table 2: CFSA Hotline Reports by Month, FY 2019 and FY 2020

Table 3: Investigations by Month, FY 2019 and FY 2020

Table 4: In-Home Cases Opened by Month

Note: Data not available for July-September 2020

Table 5: Foster Care Entries by Month

Table 6: Foster Care Exits by Month, FY 2019 and FY 2020

Figure 7: Children in Foster Care on Last Day of Fiscal Year