Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Assessment of Rates of Child Maltreatment in States With Medicaid Expansion vs States Without Medicaid Expansion

Assessment of Rates of Child Maltreatment in States With Medicaid Expansion vs States Without... Key Points Question Is the state expansion of IMPORTANCE Physical abuse and neglect affect a significant number of children in the United Medicaid associated with rates of child States. The 2014 Medicaid expansion, in which several states opted to expand their Medicaid physical abuse and neglect? programs, is associated with parental financial stability and access to mental health care. Findings In this ecological study comparing pre– and post–Medicaid OBJECTIVE To determine whether Medicaid expansion is associated with changes in physical abuse expansion state-level rates of child and neglect rates. physical abuse and neglect from the National Child Abuse and Neglect Data DESIGN, SETTING, AND PARTICIPANTS This ecological study used state-level National Child Abuse Systems, after adjusting for and Neglect Data Systems (NCANDS) data from January 1, 2010, through December 31, 2016, to confounders, there were fewer cases of compare the change in physical abuse and neglect rates in states that chose to expand Medicaid vs reported neglect (422 fewer per those that did not. All cases of physical abuse and neglect of children younger than 6 years during the 100 000 younger than 6 years) in states study period that were referred to state-level Child Protective Services and screened in for further that expanded Medicaid than during intervention after having met a maltreatment risk threshold were included. Cases with only that time in nonexpansion states, which documented sexual or emotional abuse were excluded. A difference-in-difference analysis was had a baseline rate of 3944 cases per conducted from April 12, 2018, through March 26, 2019. 100 000 children younger than 6 years in 2013. EXPOSURES State-level Medicaid expansion status. Meaning These results suggest that MAIN OUTCOMES AND MEASURES Incidence rate of screened-in referrals for physical abuse or Medicaid expansion may serve as a neglect per 100 000 children younger than 6 years per year by state. means to prevent child neglect. RESULTS Data were analyzed for 31 states and the District of Columbia that expanded Medicaid and Invited Commentary 19 states that did not during the study period, with baseline neglect counts of 646 463 and 388 265, respectively. After Medicaid expansion, 422 fewer cases of neglect per 100 000 children younger Supplemental content than 6 years (95% CI, −753 to −91) were reported each year after adjusting for confounders for Author affiliations and article information are comparison of postexpansion and preexpansion rates in states that expanded Medicaid contrasting listed at the end of this article. with the change during that time in nonexpansion states. From 2013 to 2016, Medicaid coverage for adults with dependent children increased a median 1.9% (interquartile range, 0.4% to 4.3%) in the states that did not expand Medicaid and 4.2% (interquartile range, 0.9% to 6.0%) in the states that did. No associations were found between Medicaid coverage or Medicaid eligibility criteria and physical abuse or neglect rates. CONCLUSIONS AND RELEVANCE Medicaid expansion was associated with a reduction in the reported child neglect rate, but not the physical abuse rate. These findings suggest that expanding Medicaid may help prevent child neglect. JAMA Network Open. 2019;2(6):e195529. Corrected on July 5, 2019. doi:10.1001/jamanetworkopen.2019.5529 Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 1/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion Introduction Multiple studies have shown that individuals who experience childhood adverse events are at greater 1-3 risk of developing several diseases in adulthood, including heart disease and cancer. Five of the 10 events included on the current Adverse Childhood Experience screening questionnaire developed from these studies relate to child maltreatment, a broad term encompassing physical, sexual, and emotional abuse and neglect. The estimated total monetary cost of confirmed child maltreatment cases occurring in a single year in the United States, including lifetime medical care and productivity losses, is $124 billion to 5-7 $428 billion. The lifetime cost per person who experiences maltreatment is $210 012, more than $43 000 of which represents direct medical costs. These numbers only account for confirmed cases. Child Protective Services (CPS) receives more than 3 million referrals annually, representing about 5% of all children. Once unreported cases are added, an estimated 15.2% of children in the United States experience caregiver maltreatment each year. Much work has been done to develop programs to reduce child maltreatment’s significant burden on children, families, and society. Few programs, however, have been found to be 10-13 consistently effective at reducing maltreatment rates. Policies and programs addressing some of the risk factors for maltreatment (eg, poverty, limited parental access to mental health care) have 14 15 been proposed as a means to prevent child maltreatment indirectly. One study demonstrated that California’s 2004 paid family leave policy was associated with a significant reduction in the rate of abusive head trauma compared with states without similar leave policies. Another study showed that child care access and continuity of child health care policies were also associated with decreased child maltreatment rates. On January 1, 2014, 24 states and the District of Columbia expanded their Medicaid programs to include all adults with an annual income to 138% of the federal poverty level (FPL) as part of the 17,18 Affordable Care Act (ACA) (Figure 1). Seven other states expanded their programs at a later date, through July 1, 2016. Although the expansion was intended to roll out nationwide, the US Supreme Court determined that states could opt out, and 19 states did. The Medicaid expansion was credited in 2015 with expanding coverage to 11 million adults. This increase in medical insurance Figure 1. Medicaid Expansion Status by State as of December 31, 2016 WA VT ME MT ND MN NH OR MA WI ID SD NY RI WY MI CT IA PA NE NV OH NJ IN IL UT CA DE CO WV VA KS MO MD KY NC DC TN OK AZ NM SC AR MS AL GA LA TX AK FL HI Medicaid expansion on January 1, 2014 Medicaid expansion after January 1, 2014 No Medicaid expansion JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 2/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion coverage had a positive effect on financial stability and several health outcomes in the lower-income 22-25 population. The proportion of uninsured parents living in states with Medicaid expansions dropped 33% from 2013 to 2014. Studies have shown that the ACA Medicaid expansion resulted in 22,25 improved mental health status for low-income parents. The findings from these studies suggest that expanding Medicaid was associated with parents’ financial stability and access to mental health care. Because these are important risk factors for maltreatment, Medicaid expansion may have also been associated with child maltreatment outcomes, including physical abuse and neglect. Because Medicaid expansion did not happen in all states or at the same time, this allows for a comparison not just of maltreatment outcomes within states that expanded Medicaid before and after the expansion occurred, but also the same maltreatment outcomes between states that expanded Medicaid and those that did not. We conducted this ecological study to examine and quantify the association between state Medicaid expansion and the rates of child abuse and neglect. We hypothesized that states and years with expanded Medicaid would have lower rates of child physical abuse and neglect than would states and years without Medicaid expansion. Methods We conducted an ecological study because the exposure of interest (ie, Medicaid expansion) was inherently at the group level. The goal was not to understand individual-level associations but rather the associations of a policy shift at the state level. Moreover, decisions by policy makers regarding Medicaid expansion, at least in the context of this investigation, would likely revolve around changes in aggregated state-level outcomes rather than changes for specific individuals, so the adopted design is most appropriate. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. This study was exempt from full review by the Seattle Children’s Hospital and the University of Washington institutional review boards because data were aggregated and no identifiable information was recorded; deidentification was not indicated. Data and Sample We used demographic and maltreatment data from the National Child Abuse and Neglect Data System (NCANDS) Child Files (eMethods in the Supplement). This administrative data set includes child-level data for all child maltreatment reports investigated by CPS in all 50 states and the District of Columbia. We included data for 2010 through 2016 to capture trends in maltreatment rates for several years before and after Medicaid expansion. The year 2016 was used as the study cutoff because this was the most recent year for which complete data were available for the exposure, outcomes, and covariates as described below. Exposure The exposure of interest was whether a state expanded Medicaid on or after January 1, 2014. State- level exposure status was determined based on Medicaid expansion information available through the Henry J. Kaiser Family Foundation (KFF). The KFF Medicaid expansion data have been used 19,24 previously in other studies evaluating expansion outcomes by state. Information about the income eligibility for Medicaid for parents of dependent children expressed as a percentage cutoff of the FPL in each state and for each year was obtained from the KFF. States expanding Medicaid after January 1, 2014, were only included in the expanded group for the period in which they actually had expanded Medicaid. Outcome The primary outcome using NCANDS data was the incidence rate of screened-in referrals for physical abuse or neglect per 100 000 children younger than 6 years per year in each state. These referrals made to CPS have met criteria at intake, indicating that an investigation is warranted for 1 or several JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 3/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion types of maltreatment. Once an investigation is complete, the concern for maltreatment may be categorized as substantiated, unsubstantiated, or several disposition categories between. If several children are included in 1 referral, each child counts as a separate case. If the same child is referred to CPS more than once per year, that child is counted once for each referral to CPS. We used the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research population estimates to determine the child population by state and year. We included all screened-in referrals instead of only substantiated cases. This decision was based on prior research demonstrating that the decision process for substantiation is inconsistent and subject to policy changes and biases and that little difference in parental recidivism and other outcomes seems apparent between children whose cases were substantiated and those whose cases 32-35 16,36-39 were not. This decision is consistent with other recent studies using CPS records. The screened-in referrals were stratified by maltreatment type (physical abuse or neglect) and included cases screened-in for an alternative response. Cases with concern for both physical abuse and neglect were included in both groups. Although some of the risk factors for sexual abuse are the same as for physical abuse or neglect, we chose to exclude cases in which sexual abuse was the only maltreatment type; perpetrators in cases of sexual abuse are less likely to be parents or guardians than in cases of physical abuse or neglect, so Medicaid expansion was not thought to affect these cases in the same 40-42 way, even if the perpetrators were eligible for Medicaid through the expansion. We also excluded cases in which maltreatment type or disposition was missing or emotional abuse was the only maltreatment type, because this category is inconsistently captured in CPS records owing to the 43,44 absence of a consensus definition, because some states do not include emotional abuse as a separate category of maltreatment, and because studies have shown that emotional abuse is significantly underreported. Maltreatment deaths were also excluded because state identifier data are masked in NCANDS for cases resulting in death such that we were unable to determine their exposure status. Intermediate Outcome We used KFF state-level data on the estimated number of adults with dependent children who have insurance coverage to determine the association of Medicaid expansion policies with Medicaid coverage among parents in each state, a possible intermediate step in the association between Medicaid expansion policies and child maltreatment outcomes. The KFF estimates are based on additional analyses of the yearly US Census Bureau’s Current Population Survey and the Annual Social and Economic Supplements. Covariates We included the following state-level variables as possible confounders based on factors shown to be associated with the outcomes of physical abuse and neglect and thought to be associated with the exposure, Medicaid expansion. Paid family leave policy status was determined from the National Partnership for Women & Families data. Unemployment proportion was obtained from the US Bureau of Labor Statistics. Teenage birth rate data were obtained from the National Center for Health Statistics National Vital Statistics System. The percentage of families with children younger than 5 years living below the FPL was determined using the American Community Survey estimates from the US Census Bureau. Finally, child care wait list data were obtained from the National Women’s Law Center. Statistical Analysis We compared prespecified baseline characteristics of expansion and nonexpansion states in 2013. To compare the change in physical abuse and neglect reporting rates in states that expanded Medicaid vs those that did not during the 7-year period of interest, we conducted difference-in-difference analyses with general linear regression models with state and year as fixed effects (eMethods in the JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 4/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion Supplement) in which the regression coefficient for Medicaid expansion is interpretable as a generalized form of difference-in-difference. Difference-in-difference analyses allow for the comparison of a change in mean outcome between a group that has experienced a treatment and a group that has not experienced the treatment before and after that treatment occurs. We calculated descriptive statistics, including median and interquartile range (IQR) for the change in Medicaid eligibility. A general linear regression model with state and year as fixed effects was used to examine the association between the Medicaid expansion exposure and the change in Medicaid eligibility that was treated as a continuous outcome. To further analyze the association between the intermediate outcome of change in Medicaid eligibility and the final outcomes of interest, physical abuse and neglect, general linear regression analyses were performed. The same analyses were conducted for the change in parental Medicaid coverage. For all the models, to identify influential data points, the difference-in-fits approach was used. This diagnostic technique omits each observation, one at a time, and refits the regression model to determine whether particular observations have an outsized influence on the fitted model. This approach identified an unexplained near doubling of West Virginia’s physical abuse rates, possibly due to a CPS intake change, in 2014. West Virginia was therefore excluded from analyses that included physical abuse or neglect outcomes, although secondary analyses including West Virginia were conducted to evaluate the exclusion’s potential impact. Analyses were also conducted excluding several other potential outliers. The number of averted cases after expansion was calculated based on the estimate for the outcome rate and the known child population in the nonexpansion states for the years 2014 through 2016. All analyses were conducted using SAS, version 9.4 (SAS Institute, Inc). The analyses were conducted from April 12, 2018, through March 26, 2019. All 2-sided P values were considered statistically significant at P < .05. Results We analyzed NCANDS data for 31 states and the District of Columbia that expanded Medicaid and 19 states that did not during the period from January 1, 2010, through December 31, 2016. Table 1 shows that before Medicaid expansion (in 2013), the median teen birth rate and the proportion of families with children younger than 5 years living below the FPL were higher in the states that did not expand Medicaid than in the states that did. Conversely, the proportion of parents with Medicaid coverage was lower in the states that did not go on to expand Medicaid. From 2013 to 2016, Medicaid coverage for adults with dependent children increased a median 1.9% (IQR, 0.4% to 4.3%) in the states that did not expand Medicaid and 4.2% (IQR, 0.9% to 6.0%) in the states that did (Table 2). In the states that did not expand Medicaid, the Medicaid eligibility cutoff for parents decreased 6 percentage points of the FPL (IQR, −17.0 to 2.5), whereas in the states that did expand, the eligibility cutoff increased by 42.0 percentage points of the FPL (IQR, 5.0 to 81.0). Medicaid coverage for adults with 1 or more dependent children was found to increase (0.6 Table 1. Comparison of the States With vs Without Medicaid Expansion in 2013 States b c Baseline Characteristic With Medicaid Expansion (n = 31) No Medicaid Expansion (n = 20) No. of children aged <6 y by state 279 291 (79 339-535 097) 354 489 (198 220-550 928) Abbreviation: FPL, federal poverty level. Total No. of children aged <6 y 14 008 018 9 845 257 Unless otherwise indicated, data are expressed as Unemployment rate, % 7.3 (6.6-8.1) 6.7 (5.3-7.4) median (interquartile range). Teen birth rate per 1000 women 23.6 (17.7-30.3) 29.6 (24.8-34.5) Includes Washington, DC. US population below FPL, % 16.8 (13.5-21.2) 20.7 (18.3-23.6) Louisiana began expansion in July 2016 and is Parents with Medicaid coverage, % 14.6 (12.3-19.5) 9.2 (8.0-13.8) included among the states that did not expand States with childcare wait list, No. (%) 12 (39) 7 (35) Medicaid. States with paid leave, No. (%) 2 (6) 0 Includes parents with children younger than 5 years. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 5/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion percentage points per year; 95% CI, 0.2 to 1.0) and was positively associated with Medicaid expansion status (1.6 percentage points higher for a given year; 95% CI, 0.1 to 3.0) (eTable 1 in the Supplement). In comparing the years 2013 and 2016, Medicaid expansion was found to be associated with a significant increase in percentage of FPL Medicaid eligibility, as would be expected (48.6 percentage points higher; 95% CI, 22.5 to 74.6). In 2013, the baseline count of physical abuse cases was 176 591 in expansion states and 112 047 in nonexpansion states, or 1261 and 1138 per 100 000 children younger than 6 years, respectively. The baseline count of neglect cases was 646 463 in expansion states and 388 265 in nonexpansion states, or 4615 and 3944 per 100 000 respectively. From 2013 to 2016, many individual states in the Medicaid expansion and nonexpansion groups experienced decreased rates of physical abuse and neglect (Figure 2). When comparing the entire preexpansion and postexpansion periods, the states with Medicaid expansion as a whole saw the physical abuse rate decrease by 68 cases per 100 000 children younger than 6 years and the neglect rate decrease by 336 cases per 100 000 children younger than 6 years. In comparison, the group of states without Medicaid expansion had a decreased rate of physical abuse of 35 cases per 100 000 and an increased rate of neglect by 90 cases per 100 000 children younger than 6 years. Table 3 shows the estimates for net change in the physical abuse and neglect rates after Medicaid expansion in the states that expanded Medicaid relative to the states that did not in any year. After Medicaid expansion, states that expanded Medicaid had fewer reported cases of neglect after adjusting for confounders (−422 per 100 000 children younger than 6 years; 95% CI, −753 to −91) compared with the change in the rate during that time in nonexpansion states, which had a baseline rate of 3944 per 100 000 children younger than 6 years in 2013 (Table 3). No statistically significant association of Medicaid expansion with physical abuse rates was found. Findings were similar when including West Virginia and excluding possible outlier state data (eTable 2 and eTable 3 in the Supplement). Neither Medicaid coverage itself nor Medicaid eligibility criteria were significantly associated with physical abuse or neglect rates, with or without adjusting for potential confounders (eTable 4 and eTable 5 in the Supplement). Based on these results, we estimated that had the nonexpansion states expanded Medicaid, a total of 124 981 (95% CI, 26 921-223 041) fewer cases of neglect would have been reported in the United States from 2014 through 2016, or a mean of 41 660 fewer cases per year. Discussion This study suggests an association between the ACA Medicaid expansion and reductions in the rate of screened-in reports made to CPS for concerns of neglect in children younger than 6 years. This association was seen even after controlling for other factors, including state-level policies and measures that might affect the reporting rate of child neglect and be associated with the status of Medicaid expansion in a state. The results add to the small but growing body of evidence demonstrating a positive association 15,16 between a social policy and child maltreatment outcomes. Neglect represents approximately three-quarters of reported child maltreatment cases in the United States. Despite being the most Table 2. Net Change From 2013 to 2016 of Medicaid Coverage and Medicaid Eligibility Criteria for Adults With Dependent Children Medicaid Expansion, Median (IQR) Net Change Present (n = 31) Absent (n = 20) Medicaid coverage, % 4.2 (0.9 to 6.0) 1.9 (0.4 to 4.3) Medicaid eligibility, % of FPL 42.0 (5.0 to 81.0) −6.0 (−17.0 to 2.5) Abbreviations: FPL, federal poverty level; IQR, interquartile range. Louisiana began expansion in July 2016 and is included among the states that did not expand Medicaid. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 6/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion common form of child maltreatment, efforts to prevent neglect have proven difficult in part because of a lack of consensus regarding what constitutes neglect and how to best stratify risk for 57,58 neglect. Our findings offer a promising avenue for future child neglect prevention research, especially as new states expand Medicaid. The absence of a significant association between Medicaid expansion and rates of reported physical abuse could be due to one of several possibilities. The first is that there may truly not be any Figure 2. Change in the Physical Abuse and Neglect Rates Compared by State for 2013 vs 2016 A Physical abuse in expansion states B Neglect in expansion states MI MI KY RI DE WA MT IA NV DC AK AK AR AZ ND KY NH NM NJ NY AZ AR OH MD WA NH CA OH NY NJ NM DE VT CA RI ND IA CT CT CO OR HI CO IL IL VT HI PA MA NV MD MN PA OR IN MA MN MT DC IN –2500 –2000 –1500 –1000 –500 0 500 1000 1500 2000 2500 3000 3500 –4000 –3000 –2000 –1000 0 1000 2000 No. of Cases of Physical Abuse No. of Cases of Neglect per 100 000 Children C Physical abuse in nonexpansion states D Neglect in nonexpansion states OK TN KS ME FL NC SD FL WI SD TX WY WY KS UT TX LA VA VA NE NE UT NC WI MO LA GA ID MS MO AL AL ME OK ID SC SC MS TN GA –2500 –2000 –1500 –1000 –500 0 500 1000 1500 2000 2500 3000 3500 –4000 –2000 –2000 –1000 0 1000 2000 No. of Cases of Physical Abuse per 100 000 Children No. of Cases of Neglect per 100 000 Children Physical abuse and neglect rates are expressed as cases per 100 000 children younger in this analysis. For West Virginia (not included), the change in physical abuse rates is than 6 years. Negative changes indicate the rate decreased from 2013 to 2016. Louisiana thought to be an outlier not due to an actual change but to data unreliability concerns did not expand Medicaid until July 2016 and is classified with the nonexpansion states possibly related to a change in 2014 to the Child Protective Services intake system. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 7/13 States Not Expanded States Expanded States Not Expanded States Expanded JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion association between Medicaid expansions and child physical abuse, suggesting that at least some of the pathways resulting in physical abuse are different from those resulting in neglect. An association may also exist between Medicaid expansion and physical abuse, but this association is seen only after more time has lapsed or only for a younger subset of children when compared with neglect. This possibility suggests that the association is delayed relative to that seen with neglect and was therefore not captured in the relatively short study period. After state expansion occurred, some states saw much larger changes in eligibility and coverage than others. Therefore, we conducted further analyses to evaluate not only the association between Medicaid expansion and child maltreatment outcomes but also the associations between Medicaid expansion and these intermediate outcomes. We found that expanding Medicaid was associated with increases in the percentage of FPL eligibility cutoff as well as the proportion of parents covered by Medicaid. This finding is in line with those of prior research demonstrating that the expansion of 22,26,59 Medicaid had the intended effect of increasing Medicaid coverage. Although Medicaid expansion was independently associated with the percentage of FPL eligibility cutoff and the proportion of parents covered by Medicaid and with neglect rates, no significant association was found between the former and the latter. This finding may suggest that the association between Medicaid expansion and neglect is complex and does not simply involve increased coverage. Another possibility is that a direct association exists between increased coverage and neglect but was not adequately detected in our analyses. Medicaid coverage increased even in states that did not expand Medicaid. Other studies have attributed this finding to a “welcome-mat” effect among people who had been eligible for Medicaid before the expansions took place but only applied for Medicaid coverage after learning more about health care coverage through the national health care discussions. Despite this change, we found that the increase in Medicaid coverage for parents was significantly larger in the states that expanded Medicaid than in the states that did not. Limitations This study had several limitations. Considering the Medicaid expansion as a binary exposure may not have fully captured some of the more complex changes that occurred within states. Medicaid expansion may be associated with other health care changes that improved patients’ ability to access care and services differently in different states, although they all opted to expand Medicaid. Medicaid delivery varies by state. Some of this variation is due to Medicaid waiver programs, which allow states to modify federal Medicaid rules to test different delivery and payment models. For example, as Table 3. Changes in Annual Child Abuse and Neglect Rates in States That Expanded Medicaid Relative to States That Did Not Variable Estimate of Rate per 100 000 Children Aged <6 y for Each Study Year (95% CI) Variable Physical Abuse Neglect Medicaid expansion −57 (−213 to 99) −422 (−753 to −91) Family policy −101 (−661 to 459) −1092 (−2282 to 98) Unemployment rate −62 (−125 to 1) −85 (−219 to 49) Families living below FPL 9 (−10 to 27) 31 (−9 to 71) Teen births 5 (−21 to 31) −8 (−63 to 48) Childcare wait list 96 (−110 to 303) 153 (−287 to 592) Abbreviation: FPL, federal poverty level. The unadjusted change for physical abuse was −33 (95% CI, −186 to 119) per 100 000 children and for neglect was −427 (95% CI, −752 to −102) per 100 000 children (P < .05). The adjusted model controlled for the following state-level covariables: paid family leave policy, unemployment, teen birth rate, proportion of families living in poverty, and presence of a child care wait list. Medicaid expansion indicates states with relative to those without this expansion. The analyses exclude West Virginia because its outcome data were found to be unreliable. P < .05. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 8/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion of December 7, 2018, there were 46 approved and 25 pending Section 1115 waivers pertaining to matters as varied as Medicaid work requirements and healthy behavior incentives. In addition to increasing Medicaid coverage, Medicaid expansion also involved states receiving federal funds to cover 100% of newly eligible Medicaid recipients. These additional funds may have allowed states to cover additional services affecting parents’ ability to provide for their children. Studying state- level measures of Medicaid delivery such as the number of health care professionals per capita accepting Medicaid, evaluating Medicaid waiver programs themselves, or looking at how states used the additional federal Medicaid funds may help identify specific pathways through which the association occurs between Medicaid expansion and decreased child neglect. When considering Medicaid eligibility for parents based on income, considerable heterogeneity between states existed even before the ACA Medicaid expansion took place. We controlled for these baseline differences by conducting difference-in-difference analyses in which states were compared with themselves before and after Medicaid expansion. In addition, we were unable to capture and include state-level changes to child physical abuse and neglect policies and reporting laws that might have confounded our results. We attempted to control for these changes indirectly by controlling for other state-level variables that could have a bearing on them, although future research could evaluate the effect of these changes directly by collecting and including data on state-level policy changes over time. It has been previously demonstrated that only a subset of maltreated children are 63,64 reported to CPS and therefore included in the NCANDS database used for this study. Including the entire population of maltreated children might alter the results by strengthening or weakening the association seen herein. Conclusions This study found that ACA Medicaid expansion was associated with a reduction in the reported child neglect rate but with no significant change to the physical abuse rate. It appears from these findings that future studies should focus on assessing what the mechanism for this association is to determine whether expanding Medicaid prevents child neglect and whether further expansions could be designed to bolster this secondary association. ARTICLE INFORMATION Accepted for Publication: April 22, 2019. Published: June 14, 2019. doi:10.1001/jamanetworkopen.2019.5529 Correction: This article was corrected on July 5, 2019, to fix a typographical error in the running title. Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Brown ECB et al. JAMA Network Open. Corresponding Author: Emily C. B. Brown, MD, MS, Seattle Children’s Research Institute, Center for Child Health, Behavior, and Development, 4800 Sand Point Way NE, PO Box 5371, Seattle, WA 98105 (emily.brown@ seattlechildrens.org). Author Affiliations: Seattle Children’s Research Institute, Center for Child Health, Behavior, and Development, Seattle, Washington (Brown, Garrison, Bao, Qu); Department of Pediatrics, University of Washington School of Medicine, Seattle (Brown, Jenny, Rowhani-Rahbar); Division of Child Psychiatry, University of Washington School of Medicine, Seattle (Garrison); Department of Epidemiology, University of Washington School of Public Health, Seattle (Rowhani-Rahbar); Harborview Injury Prevention & Research Center, University of Washington, Seattle (Rowhani-Rahbar). Author Contributions: Dr Brown had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Brown, Jenny, Rowhani-Rahbar. Acquisition, analysis, or interpretation of data: Brown, Garrison, Bao, Qu, Rowhani-Rahbar. Drafting of the manuscript: Brown, Bao, Qu. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 9/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion Critical revision of the manuscript for important intellectual content: Brown, Garrison, Jenny, Rowhani-Rahbar. Statistical analysis: Brown, Bao, Qu. Obtained funding: Brown. Administrative, technical, or material support: Jenny, Rowhani-Rahbar. Supervision: Brown, Garrison, Jenny, Rowhani-Rahbar. Conflict of Interest Disclosures: Drs Brown and Jenny reported providing expert testimony, as part of their clinical responsibilities, in cases of alleged child maltreatment. Dr Brown reported receiving grants from Helfer Society during the conduct of the study. Dr Rowhani-Rahbar reported receiving funding from the Centers for Disease Control and Prevention outside the submitted work. No other disclosures were reported. Funding/Support: This study was funded by a career development grant from the Helfer Society. Role of the Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Additional Contributions: Kathryn B. Whitlock, MS, Seattle Children’s Research Institute, Seattle, Washington, aggregated the data and conducted the preliminary analyses. She was compensated for this work through the biostatistical services core at Seattle Children’s Research Institute. REFERENCES 1. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: the Adverse Childhood Experiences (ACE) Study. Am J Prev Med. 1998;14(4): 245-258. doi:10.1016/S0749-3797(98)00017-8 2. Felitti VJ. Adverse childhood experiences and adult health. Acad Pediatr. 2009;9(3):131-132. doi:10.1016/j.acap. 2009.03.001 3. Nemeroff CB. Paradise lost: the neurobiological and clinical consequences of child abuse and neglect. Neuron. 2016;89(5):892-909. doi:10.1016/j.neuron.2016.01.019 4. National Council of Juvenile and Family Court Judges. Adverse Childhood Experience (ACE) questionnaire. http:// www.ncjfcj.org/sites/default/files/Finding%20Your%20ACE%20Score.pdf. Published October 24, 2006. Accessed December 5, 2016. 5. US Department of Health and Human Services. Child abuse and neglect cost the United States $124 billion. http:// www.cdc.gov/media/releases/2012/p0201_child_abuse.html. Published February 1, 2012. Accessed December 5, 6. Fang X, Brown DS, Florence CS, Mercy JA. The economic burden of child maltreatment in the United States and implications for prevention. Child Abuse Negl. 2012;36(2):156-165. doi:10.1016/j.chiabu.2011.10.006 7. Peterson C, Florence C, Klevens J. The economic burden of child maltreatment in the United States, 2015. Child Abuse Negl. 2018;86:178-183. doi:10.1016/j.chiabu.2018.09.018 8. US Department of Health and Human Services. Child Maltreatment 2014. https://www.acf.hhs.gov/sites/default/ files/cb/cm2014.pdf. Published January 25, 2016. Accessed May 9, 2019. 9. Finkelhor D, Turner HA, Shattuck A, Hamby SL. Prevalence of childhood exposure to violence, crime, and abuse: results from the National Survey of Children’s Exposure to Violence. JAMA Pediatr. 2015;169(8):746-754. doi:10. 1001/jamapediatrics.2015.0676 10. Mikton C, Butchart A. Child maltreatment prevention: a systematic review of reviews. Bull World Health Organ. 2009;87(5):353-361. doi:10.2471/BLT.08.057075 11. Scribano PV. Prevention strategies in child maltreatment. Curr Opin Pediatr. 2010;22(5):616-620. 12. Wood JN. Challenges in prevention of abusive head trauma. JAMA Pediatr. 2015;169(12):1093-1094. doi:10. 1001/jamapediatrics.2015.3023 13. Curry SJ, Krist AH, Owens DK, et al; US Preventive Services Task Force. Interventions to prevent child maltreatment: US Preventive Services Task Force recommendation statement. JAMA. 2018;320(20):2122-2128. doi:10.1001/jama.2018.17772 14. Klevens J, Whitaker DJ. Primary prevention of child physical abuse and neglect: gaps and promising directions. Child Maltreat. 2007;12(4):364-377. doi:10.1177/1077559507305995 15. Klevens J, Luo F, Xu L, Peterson C, Latzman NE. Paid family leave’s effect on hospital admissions for pediatric abusive head trauma. Inj Prev. 2016;22(6):442-445. doi:10.1136/injuryprev-2015-041702 16. Klevens J, Barnett SB, Florence C, Moore D. Exploring policies for the reduction of child physical abuse and neglect. Child Abuse Negl. 2015;40:1-11. doi:10.1016/j.chiabu.2014.07.013 JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 10/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion 17. The Henry J. Kaiser Family Foundation. Status of state action on the Medicaid expansion decision. http://kff.org/ health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care- act/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D. Published November 7, 2018. Accessed November 16, 2018. 18. Ayanian JZ, Ehrlich GM, Grimes DR, Levy H. Economic effects of Medicaid expansion in Michigan. N Engl J Med. 2017;376(5):407-410. doi:10.1056/NEJMp1613981 19. Anderson ME, Glasheen JJ, Anoff D, Pierce R, Lane M, Jones CD. Impact of state Medicaid expansion status on length of stay and in-hospital mortality for general medicine patients at US academic medical centers. JHosp Med. 2016;11(12):847-852. doi:10.1002/jhm.2649 20. Lyon SM, Douglas IS, Cooke CR. Medicaid expansion under the Affordable Care Act: implications for insurance- related disparities in pulmonary, critical care, and sleep. Ann Am Thorac Soc. 2014;11(4):661-667. doi:10.1513/ AnnalsATS.201402-072PS 21. Rudowitz R, Artiga S, Young K. What coverage and financing is at risk under a repeal of the ACA Medicaid expansion? http://kff.org/report-section/what-coverage-and-financing-is-at-risk-under-a-repeal-of-the-aca- medicaid-expansion-issue-brief/. Published December 6, 2016. Accessed March 15, 2017. 22. Antonisse L, Garfield R, Rudowitz R, Artiga S. The effects of Medicaid expansion under the ACA: updated findings from a literature review. https://www.kff.org/medicaid/issue-brief/the-effects-of-medicaid-expansion- under-the-aca-updated-findings-from-a-literature-review-march-2018/. Published March 28, 2018. Accessed October 10, 2018. 23. Bhatt CB, Beck-Sagué CM. Medicaid expansion and infant mortality in the United States. Am J Public Health. 2018;108(4):565-567. doi:10.2105/AJPH.2017.304218 24. Wherry LR, Miller S. Early coverage, access, utilization, and health effects associated with the Affordable Care Act Medicaid expansions: a quasi-experimental study. Ann Intern Med. 2016;164(12):795-803. doi:10.7326/ M15-2234 25. McMorrow S, Gates JA, Long SK, Kenney GM. Medicaid expansion increased coverage, improved affordability, and reduced psychological distress for low-income parents. Health Aff (Millwood). 2017;36(5):808-818. doi:10. 1377/hlthaff.2016.1650 26. Kenney G, Anderson N, Long S, et al. Taking stock: health insurance coverage for parents under the ACA in 2014. http://hrms.urban.org/briefs/Health-Insurance-Coverage-for-Parents-under-the-ACA-in-2014.html. Published September 9, 2014. Accessed May 9, 2019. 27. Jenny C. Child Abuse and Neglect: Diagnosis, Treatment, and Evidence. St Louis, MO: Saunders/Elsevier; 2011. 28. US Department of Health and Human Services, Administration for Children and Families, Administration on Children, Youth and Families, Children’s Bureau. National child abuse and neglect data system (NCANDS) child file, FFYs 2010-2016 [dataset]. http://www.ndacan.cornell.edu. Updated 2018. Accessed March 1, 2019. 29. Kaiser Family Foundation. Medicaid income eligibility limits for parents, 2002-2018. https://www.kff.org/ medicaid/state-indicator/medicaid-income-eligibility-limits-for-parents/?currentTimeframe=0& sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D. Updated 2018. Accessed May 10, 2018. 30. Centers for Disease Control and Prevention (CDC). Bridged-race population estimates. https://wonder.cdc.gov/ bridged-race-population.html. Published June 30, 2016. Accessed March 29, 2017. 31. Font SA, Maguire-Jack K. Decision-making in child protective services: influences at multiple levels of the social ecology. Child Abuse Negl. 2015;47:70-82. doi:10.1016/j.chiabu.2015.02.005 32. Kohl PL, Jonson-Reid M, Drake B. Time to leave substantiation behind: findings from a national probability study. Child Maltreat. 2009;14(1):17-26. doi:10.1177/1077559508326030 33. Hussey JM, Marshall JM, English DJ, et al. Defining maltreatment according to substantiation: distinction without a difference? Child Abuse Negl. 2005;29(5):479-492. doi:10.1016/j.chiabu.2003.12.005 34. Drake B, Jonson-Reid M, Way I, Chung S. Substantiation and recidivism. Child Maltreat. 2003;8(4):248-260. doi:10.1177/1077559503258930 35. Leiter J, Others A. Substantiated and unsubstantiated cases of child maltreatment: do their consequences differ? Soc Work Res. 1994;18(2):67-82. doi:10.1093/swr/18.2.67 36. Hammond I, Eastman AL, Leventhal JM, Putnam-Hornstein E. Maternal mental health disorders and reports to child protective services: a birth cohort study. Int J Environ Res Public Health. 2017;14(11):1320. doi:10.3390/ ijerph14111320 JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 11/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion 37. Parrish JW, Shanahan ME, Schnitzer PG, Lanier P, Daniels JL, Marshall SW. Quantifying sources of bias in longitudinal data linkage studies of child abuse and neglect: measuring impact of outcome specification, linkage error, and partial cohort follow-up. Inj Epidemiol. 2017;4(1):23. doi:10.1186/s40621-017-0119-6 38. Putnam-Hornstein E. Report of maltreatment as a risk factor for injury death: a prospective birth cohort study. Child Maltreat. 2011;16(3):163-174. doi:10.1177/1077559511411179 39. Drake B, Lee SM, Jonson-Reid M. Race and child maltreatment reporting: are blacks overrepresented? Child Youth Serv Rev. 2009;31(3):309-316. doi:10.1016/j.childyouth.2008.08.004 40. Westat I, McDonald WR; US Administration for Children and Families Office of Planning, Research, and Evaluation. Fourth National Incidence Study of Child Abuse and Neglect (NIS-4): Report to Congress. Washington, DC: US Dept of Health and Human Services, Administration for Children and Families, Office of Planning, Research and Evaluation and the Children's Bureau; 2010. 41. Leclerc B, Chiu YN, Cale J. Sexual violence and abuse against children: a first review through the lens of environmental criminology. Int J Offender Ther Comp Criminol. 2016;60(7):743-765. doi:10.1177/ 0306624X14564319 42. Finkelhor D, Shattuck A, Turner HA, Hamby SL. The lifetime prevalence of child sexual abuse and sexual assault assessed in late adolescence. J Adolesc Health. 2014;55(3):329-333. doi:10.1016/j.jadohealth.2013.12.026 43. Hibbard R, Barlow J, Macmillan H; Committee on Child Abuse and Neglect; American Academy of Child and Adolescent Psychiatry; Child Maltreatment and Violence Committee. Psychological maltreatment. Pediatrics. 2012;130(2):372-378. doi:10.1542/peds.2012-1552 44. Kimber M, MacMillan HL. Child psychological abuse. Pediatr Rev. 2017;38(10):496-498. doi:10.1542/pir. 2016-0224 45. US Department of Health and Human Services. Child maltreatment 2016. https://www.acf.hhs.gov/sites/default/ files/cb/cm2016.pdf. Published 2018. Accessed May 9, 2019. 46. Henry J. Kaiser Family Foundation. Health insurance coverage of adults with dependent children. http://kff.org/ other/state-indicator/nonelderly-adults-with-dependents/?currentTimeframe=0&sortModel=%7B%22colId% 22:%22Location%22,%22sort%22:%22asc%22%7D. Published 2017. Accessed March 15, 2017. 47. US Census Bureau. Current Population Survey (CPS). https://www.census.gov/programs-surveys/cps.html. Accessed May 8, 2019. 48. US Census Bureau. Current Population Survey (CPS): data. https://www.census.gov/programs-surveys/cps/ data-detail.html. Revised March 6, 2018. Accessed May 8, 2019. 49. National Partnership for Women & Families. State paid family and leave insurance laws. http://www. nationalpartnership.org/research-library/work-family/paid-leave/state-paid-family-leave-laws.pdf. Published February 2017. Accessed March 29, 2017. 50. US Department of Labor, Bureau of Labor Statistics. Local area unemployment statistics. https://www.bls. gov/lau/. Updated 2017. Accessed March 29, 2017. 51. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics. National Vital Statistics System. https://www.cdc.gov/nchs/nvss/. Reviewed March 14, 2019. Accessed March 29, 2017. 52. US Census Bureau. American Community Survey. https://www.census.gov/acs/www/data/data-tables-and- tools/index.php. Updated 2016. Accessed March 29, 2017. 53. National Women’s Law Center. Persistent gaps: state child care assistance policies. https://nwlc.org/resources/ persistent-gaps-state-child-care-assistance-policies-2017/. Updated 2017. Accessed September 9, 2017. 54. Wing C, Simon K, Bello-Gomez RA. Designing difference in difference studies: best practices for public health policy research. Annu Rev Public Health. 2018;39:453-469. doi:10.1146/annurev-publhealth-040617-013507 55. SAS Institute. Influence statistics. SAS/STAT(R) 9.22 User's Guide https://support.sas.com/documentation/cdl/ en/statug/63347/HTML/default/viewer.htm#statug_reg_sect040.htm. Accessed March 12, 2019. 56. US Department of Health and Human Services. Child Maltreatment 2015. https://www.acf.hhs.gov/sites/default/ files/cb/cm2015.pdf. Published 2017. Accessed May 9, 2019. 57. Rebbe R. What is neglect? state legal definitions in the United States. Child Maltreat. 2018;23(3):303-315. doi: 10.1177/1077559518767337 58. Dubowitz H, Christian CW, Hymel K, Kellogg ND. Forensic medical evaluations of child maltreatment: a proposed research agenda. Child Abuse Negl. 2014;38(11):1734-1746. doi:10.1016/j.chiabu.2014.07.012 59. Frean M, Gruber J, Sommers BD. Premium subsidies, the mandate, and Medicaid expansion: coverage effects of the Affordable Care Act. J Health Econ. 2017;53:72-86. doi:10.1016/j.jhealeco.2017.02.004 JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 12/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion 60. Medicaid.gov. State waivers list. https://www.medicaid.gov/medicaid/section-1115-demo/demonstration-and- waiver-list/index.html. Updated 2018. Accessed December 10, 2018. 61. Henry J. Kaiser Family Foundation. Medicaid waiver tracker: approved and pending Section 1115 waivers by state. https://www.kff.org/medicaid/issue-brief/medicaid-waiver-tracker-approved-and-pending-section-1115- waivers-by-state/. Published December 7, 2018. Accessed December 10, 2018. 62. Snyder L, Rudowitz R. Medicaid financing: how does it work and what are the implications? https://www.kff. org/medicaid/issue-brief/medicaid-financing-how-does-it-work-and-what-are-the-implications/. Published May 20, 2015. Accessed January 22, 2019. 63. Fallon B, Trocmé N, Fluke J, MacLaurin B, Tonmyr L, Yuan YY. Methodological challenges in measuring child maltreatment. Child Abuse Negl. 2010;34(1):70-79. doi:10.1016/j.chiabu.2009.08.008 64. Sumner SA, Mercy JA, Dahlberg LL, Hillis SD, Klevens J, Houry D. Violence in the United States: status, challenges, and opportunities. JAMA. 2015;314(5):478-488. doi:10.1001/jama.2015.8371 SUPPLEMENT. eMethods. Data Source and Main Outcome Models eTable 1. Change in Parental Medicaid Coverage and Medicaid Eligibility Criteria From 2013 to 2016 as a Function of Medicaid Expansion Status eTable 2. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Including West Virginia eTable 3. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Excluding Possible Outlier States eTable 4. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Coverage eTable 5. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Eligibility Criteria JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 13/13 Supplementary Online Content Brown ECB, Garrison MM, Bao H, Qu P, Jenny C, Rowhani-Rahbar A. Assessment of rates of child maltreatment in states with Medicaid expansion vs states without Medicaid expansion. JAMA Netw Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 eMethods. Data Source and Main Outcome Models eTable 1. Change in Parental Medicaid Coverage and Medicaid Eligibility Criteria From 2013 to 2016 as a Function of Medicaid Expansion Status eTable 2. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Including West Virginia eTable 3. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Excluding Possible Outlier States eTable 4. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Coverage eTable 5. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Eligibility Criteria This supplementary material has been provided by the authors to give readers additional information about their work. ©2019 Brown ECB et al. JAMA Netw Open. eMethods. Data Source and Main Outcome Models The analyses presented in this publication were based on data from the National Child Abuse and Neglect Data System (NCANDS) Child File, FFYs 2010v5, 2011v5, 2012v4, 2013v3, 2014v2, 2015v1, and 2016v1. These data were provided by the National Data Archive on Child Abuse and Neglect at Cornell University, and have been used with permission. The data were originally ureau. Funding was provided by Children, Youth and Families, Administration for Children and Families, U.S. Department of Health and Human Services. The collector of the original data, the funding agency, NDACAN, Cornell University, and the agents or employees of these institutions bear no responsibility for the analyses or interpretations presented here. The information and opinions expressed reflect solely the opinions of the authors. Main outcome models Unadjusted: Y expansion + + it 0 1 it i t it Adjusted: Y = + expansion + leave + unemp + teen + fpl + wait + + + it 0 1 it 2 it 3 it 4 it 5 it 6 it i t it In these models, Y denotes the aggregated outcome rate calculated from individual-level data as the number of it screened-in referrals for physical abuse or neglect per 100,000 children under the of 6 years for each state and each year, expansion denotes whether or not a state had a Medicaid expansion at that year (0, 1, or a fraction for the it states that expanded Medicaid in the middle of a year), and and respectively denote the state and year fixed i t effects. In addition, the vector of covariates for each state in each year included leave denoting paid family leave policy status, unemp denoting the unemployment proportion, teen denoting the teen pregnancy rate, fpl denoting the percentage of families with children under the age of 5 living below the FPL, wait denoting whether there was a child care wait list, and denoting the error term. Based on this model, 1 is the effect of Medicaid expansion after accounting for covariates as well as state and year effects. ©2019 Brown ECB et al. JAMA Netw Open. eTable 1. Change in Parental Medicaid Coverage and Medicaid Eligibility Criteria From 2013 to 2016 as a Function of Medicaid Expansion Status Medicaid Medicaid coverage eligibility Medicaid 1.56 (0.10, 48.60 expansion 3.02)* (22.50, estimate (95% 74.60)* CI ) Medicaid coverage is expressed as a percentage and Medicaid eligibility represents the percent of federal poverty level CI = confidence interval *p-value < 0.05 ©2019 Brown ECB et al. JAMA Netw Open. eTable 2. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Including West Virginia Variable Variable estimate (95% CI ) Physical abuse Neglect Medicaid expansion 140 (-103 , 382) -399 (-724, -74)* Family policy 91 (-789 , 971) -1070 (-2250, 110) Unemployment rate 76 (-20, 173) -69 (-199, 61) Below FPL 46 (17, 74)* 35 (-3, 73) Teen births -41 (-81, 0)* -13 (-67, 41) Waitlist 29 (-296, 353) 145 (-291, 581) The unadjusted changes for physical abuse and neglect were 119 (-123, 360) and -407 (-728 , -87)* per 100,000 children respectively. The adjusted model presented above controlled for the following state-level covariables: paid family leave policy, unemployment, teen birth rate, proportion of families living in poverty, and presence of a child care wait list. CI = confidence interval FPL = federal poverty level *p-value < 0.05 ©2019 Brown ECB et al. JAMA Netw Open. eTable 3. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Excluding Possible Outlier States Variable Variable estimate (95% CI ) Physical abuse Neglect Medicaid expansion 26 (-78, 130) -354 (-673, -35)* Family policy -170 (-535, 196) -1303 (-2427, -180)* Unemployment rate -63 (-107, -20)* -163 (-297, -29)* Below FPL 1 (-11, 14) 36 (-2, 74) Teen births -4 (-21, 14) 5 (-48, 58) Waitlist 99 (-36, 233) 170 (-243, 583) The states excluded were Michigan, Oklahoma, Tennessee, and West Virginia. The unadjusted changes for physical abuse and neglect were 32 (-71, 136) and -347 (-667, -27)* per 100,000 children respectively. The adjusted model presented above controlled for the following state-level covariables: paid family leave policy, unemployment, teen birth rate, proportion of families living in poverty, and presence of a child care wait list. CI = confidence interval FPL = federal poverty level *p-value < 0.05 ©2019 Brown ECB et al. JAMA Netw Open. eTable 4. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Coverage Variable Variable estimate (95% CI ) Model 1 Model 2 Physical abuse Neglect Physical abuse Neglect Medicaid -51 (-104, 2) 16 (-65, 97) -46 (-100, 9) 25 (-58, 108) coverage Family policy 134 (-1470, 1737) -1461 (-3913, 167 (-1444, -1398 (-3854, 992) 1779) 1057) Unemployment -32 (-242, 178) 45 (-277, 366) -50 (-264, 165) 12 (-315, 340) rate Below FPL 14 (-28, 56) 24 (-40, 88) 15 (-27, 58) 27 (-37, 92) Teen births -6 (-109, 97) -11 (-169, 147) 3 (-103, 109) 5 (-156, 166) Waitlist 230 (-650, 1110) 430 (-916, 1776) 231 (-653, 1114) 431 (-915, 1777) Medicaid -196 (-663, 271) -364 (-1076, expansion 347) The unadjusted changes for physical abuse and neglect were -48 (-97, 1) and 16 (-60, 92) per 100,000 children respectively. Model 1 controlled for paid family leave, unemployment, teen birth rate, proportion of families with children under age 5 living below the federal poverty level, and presence of a child care wait list and model 2 controlled for the same covariates with the addition of Medicaid expansion. The analyses above exclude West Virginia as its outcome data was found to be unreliable. Medicaid coverage is expressed as a percentage and Medicaid eligibility represents the percent of federal poverty level CI = confidence interval FPL = federal poverty level ©2019 Brown ECB et al. JAMA Netw Open. eTable 5. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Eligibility Criteria Variable Variable estimate (95% CI ) Model 1 Model 2 Physical abuse Neglect Physical abuse Neglect Medicaid -4 (-8, 1) -3 (-10, 4) -3 (-8, 3) -1 (-10, 7) eligibility Family policy -297 (-1951, 1356) -1598 (-4075, -211 (-1920, -1435 (-3991, 880) 1498) 1121) Unemployment -56 (-270, 159) 35 (-286, 356) -64 (-283, 155) 20 (-307, 348) rate Below FPL 11 (-32, 54) 20 (-44, 85) 13 (-31, 57) 24 (-43, 90) Teen births -2 (-107, 102) -27 (-184, 130) 5 (-106, 116) -13 (-178, 153) Waitlist 191.92 (-702, 595 (-744, 1934) 175 (-731, 1080) 563 (-791, 1917) 1086) Medicaid -132 (-686, 423) -248 (-1077, 581) expansion The unadjusted changes for physical abuse and neglect were -3 (-7, 1) and -1 (-8, 5) per 100,000 children respectively. Model 1 controlled for paid family leave, unemployment, teen birth rate, proportion of families with children under age 5 living below the federal poverty level, and presence of a child care wait list and model 2 controlled for the same covariates with the addition of Medicaid expansion. The analyses above exclude West Virginia as its outcome data was found to be unreliable. Medicaid coverage is expressed as a percentage and Medicaid eligibility represents the percent of federal poverty level CI = confidence interval FPL = federal poverty level ©2019 Brown ECB et al. JAMA Netw Open. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA Network Open American Medical Association

Assessment of Rates of Child Maltreatment in States With Medicaid Expansion vs States Without Medicaid Expansion

Download PDF
20 pages

Loading next page...
 
/lp/american-medical-association/assessment-of-rates-of-child-maltreatment-in-states-with-medicaid-Jqf5j74F6L

References (105)

Publisher
American Medical Association
Copyright
Copyright 2019 Brown ECB et al. JAMA Network Open.
eISSN
2574-3805
DOI
10.1001/jamanetworkopen.2019.5529
Publisher site
See Article on Publisher Site

Abstract

Key Points Question Is the state expansion of IMPORTANCE Physical abuse and neglect affect a significant number of children in the United Medicaid associated with rates of child States. The 2014 Medicaid expansion, in which several states opted to expand their Medicaid physical abuse and neglect? programs, is associated with parental financial stability and access to mental health care. Findings In this ecological study comparing pre– and post–Medicaid OBJECTIVE To determine whether Medicaid expansion is associated with changes in physical abuse expansion state-level rates of child and neglect rates. physical abuse and neglect from the National Child Abuse and Neglect Data DESIGN, SETTING, AND PARTICIPANTS This ecological study used state-level National Child Abuse Systems, after adjusting for and Neglect Data Systems (NCANDS) data from January 1, 2010, through December 31, 2016, to confounders, there were fewer cases of compare the change in physical abuse and neglect rates in states that chose to expand Medicaid vs reported neglect (422 fewer per those that did not. All cases of physical abuse and neglect of children younger than 6 years during the 100 000 younger than 6 years) in states study period that were referred to state-level Child Protective Services and screened in for further that expanded Medicaid than during intervention after having met a maltreatment risk threshold were included. Cases with only that time in nonexpansion states, which documented sexual or emotional abuse were excluded. A difference-in-difference analysis was had a baseline rate of 3944 cases per conducted from April 12, 2018, through March 26, 2019. 100 000 children younger than 6 years in 2013. EXPOSURES State-level Medicaid expansion status. Meaning These results suggest that MAIN OUTCOMES AND MEASURES Incidence rate of screened-in referrals for physical abuse or Medicaid expansion may serve as a neglect per 100 000 children younger than 6 years per year by state. means to prevent child neglect. RESULTS Data were analyzed for 31 states and the District of Columbia that expanded Medicaid and Invited Commentary 19 states that did not during the study period, with baseline neglect counts of 646 463 and 388 265, respectively. After Medicaid expansion, 422 fewer cases of neglect per 100 000 children younger Supplemental content than 6 years (95% CI, −753 to −91) were reported each year after adjusting for confounders for Author affiliations and article information are comparison of postexpansion and preexpansion rates in states that expanded Medicaid contrasting listed at the end of this article. with the change during that time in nonexpansion states. From 2013 to 2016, Medicaid coverage for adults with dependent children increased a median 1.9% (interquartile range, 0.4% to 4.3%) in the states that did not expand Medicaid and 4.2% (interquartile range, 0.9% to 6.0%) in the states that did. No associations were found between Medicaid coverage or Medicaid eligibility criteria and physical abuse or neglect rates. CONCLUSIONS AND RELEVANCE Medicaid expansion was associated with a reduction in the reported child neglect rate, but not the physical abuse rate. These findings suggest that expanding Medicaid may help prevent child neglect. JAMA Network Open. 2019;2(6):e195529. Corrected on July 5, 2019. doi:10.1001/jamanetworkopen.2019.5529 Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 1/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion Introduction Multiple studies have shown that individuals who experience childhood adverse events are at greater 1-3 risk of developing several diseases in adulthood, including heart disease and cancer. Five of the 10 events included on the current Adverse Childhood Experience screening questionnaire developed from these studies relate to child maltreatment, a broad term encompassing physical, sexual, and emotional abuse and neglect. The estimated total monetary cost of confirmed child maltreatment cases occurring in a single year in the United States, including lifetime medical care and productivity losses, is $124 billion to 5-7 $428 billion. The lifetime cost per person who experiences maltreatment is $210 012, more than $43 000 of which represents direct medical costs. These numbers only account for confirmed cases. Child Protective Services (CPS) receives more than 3 million referrals annually, representing about 5% of all children. Once unreported cases are added, an estimated 15.2% of children in the United States experience caregiver maltreatment each year. Much work has been done to develop programs to reduce child maltreatment’s significant burden on children, families, and society. Few programs, however, have been found to be 10-13 consistently effective at reducing maltreatment rates. Policies and programs addressing some of the risk factors for maltreatment (eg, poverty, limited parental access to mental health care) have 14 15 been proposed as a means to prevent child maltreatment indirectly. One study demonstrated that California’s 2004 paid family leave policy was associated with a significant reduction in the rate of abusive head trauma compared with states without similar leave policies. Another study showed that child care access and continuity of child health care policies were also associated with decreased child maltreatment rates. On January 1, 2014, 24 states and the District of Columbia expanded their Medicaid programs to include all adults with an annual income to 138% of the federal poverty level (FPL) as part of the 17,18 Affordable Care Act (ACA) (Figure 1). Seven other states expanded their programs at a later date, through July 1, 2016. Although the expansion was intended to roll out nationwide, the US Supreme Court determined that states could opt out, and 19 states did. The Medicaid expansion was credited in 2015 with expanding coverage to 11 million adults. This increase in medical insurance Figure 1. Medicaid Expansion Status by State as of December 31, 2016 WA VT ME MT ND MN NH OR MA WI ID SD NY RI WY MI CT IA PA NE NV OH NJ IN IL UT CA DE CO WV VA KS MO MD KY NC DC TN OK AZ NM SC AR MS AL GA LA TX AK FL HI Medicaid expansion on January 1, 2014 Medicaid expansion after January 1, 2014 No Medicaid expansion JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 2/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion coverage had a positive effect on financial stability and several health outcomes in the lower-income 22-25 population. The proportion of uninsured parents living in states with Medicaid expansions dropped 33% from 2013 to 2014. Studies have shown that the ACA Medicaid expansion resulted in 22,25 improved mental health status for low-income parents. The findings from these studies suggest that expanding Medicaid was associated with parents’ financial stability and access to mental health care. Because these are important risk factors for maltreatment, Medicaid expansion may have also been associated with child maltreatment outcomes, including physical abuse and neglect. Because Medicaid expansion did not happen in all states or at the same time, this allows for a comparison not just of maltreatment outcomes within states that expanded Medicaid before and after the expansion occurred, but also the same maltreatment outcomes between states that expanded Medicaid and those that did not. We conducted this ecological study to examine and quantify the association between state Medicaid expansion and the rates of child abuse and neglect. We hypothesized that states and years with expanded Medicaid would have lower rates of child physical abuse and neglect than would states and years without Medicaid expansion. Methods We conducted an ecological study because the exposure of interest (ie, Medicaid expansion) was inherently at the group level. The goal was not to understand individual-level associations but rather the associations of a policy shift at the state level. Moreover, decisions by policy makers regarding Medicaid expansion, at least in the context of this investigation, would likely revolve around changes in aggregated state-level outcomes rather than changes for specific individuals, so the adopted design is most appropriate. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. This study was exempt from full review by the Seattle Children’s Hospital and the University of Washington institutional review boards because data were aggregated and no identifiable information was recorded; deidentification was not indicated. Data and Sample We used demographic and maltreatment data from the National Child Abuse and Neglect Data System (NCANDS) Child Files (eMethods in the Supplement). This administrative data set includes child-level data for all child maltreatment reports investigated by CPS in all 50 states and the District of Columbia. We included data for 2010 through 2016 to capture trends in maltreatment rates for several years before and after Medicaid expansion. The year 2016 was used as the study cutoff because this was the most recent year for which complete data were available for the exposure, outcomes, and covariates as described below. Exposure The exposure of interest was whether a state expanded Medicaid on or after January 1, 2014. State- level exposure status was determined based on Medicaid expansion information available through the Henry J. Kaiser Family Foundation (KFF). The KFF Medicaid expansion data have been used 19,24 previously in other studies evaluating expansion outcomes by state. Information about the income eligibility for Medicaid for parents of dependent children expressed as a percentage cutoff of the FPL in each state and for each year was obtained from the KFF. States expanding Medicaid after January 1, 2014, were only included in the expanded group for the period in which they actually had expanded Medicaid. Outcome The primary outcome using NCANDS data was the incidence rate of screened-in referrals for physical abuse or neglect per 100 000 children younger than 6 years per year in each state. These referrals made to CPS have met criteria at intake, indicating that an investigation is warranted for 1 or several JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 3/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion types of maltreatment. Once an investigation is complete, the concern for maltreatment may be categorized as substantiated, unsubstantiated, or several disposition categories between. If several children are included in 1 referral, each child counts as a separate case. If the same child is referred to CPS more than once per year, that child is counted once for each referral to CPS. We used the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research population estimates to determine the child population by state and year. We included all screened-in referrals instead of only substantiated cases. This decision was based on prior research demonstrating that the decision process for substantiation is inconsistent and subject to policy changes and biases and that little difference in parental recidivism and other outcomes seems apparent between children whose cases were substantiated and those whose cases 32-35 16,36-39 were not. This decision is consistent with other recent studies using CPS records. The screened-in referrals were stratified by maltreatment type (physical abuse or neglect) and included cases screened-in for an alternative response. Cases with concern for both physical abuse and neglect were included in both groups. Although some of the risk factors for sexual abuse are the same as for physical abuse or neglect, we chose to exclude cases in which sexual abuse was the only maltreatment type; perpetrators in cases of sexual abuse are less likely to be parents or guardians than in cases of physical abuse or neglect, so Medicaid expansion was not thought to affect these cases in the same 40-42 way, even if the perpetrators were eligible for Medicaid through the expansion. We also excluded cases in which maltreatment type or disposition was missing or emotional abuse was the only maltreatment type, because this category is inconsistently captured in CPS records owing to the 43,44 absence of a consensus definition, because some states do not include emotional abuse as a separate category of maltreatment, and because studies have shown that emotional abuse is significantly underreported. Maltreatment deaths were also excluded because state identifier data are masked in NCANDS for cases resulting in death such that we were unable to determine their exposure status. Intermediate Outcome We used KFF state-level data on the estimated number of adults with dependent children who have insurance coverage to determine the association of Medicaid expansion policies with Medicaid coverage among parents in each state, a possible intermediate step in the association between Medicaid expansion policies and child maltreatment outcomes. The KFF estimates are based on additional analyses of the yearly US Census Bureau’s Current Population Survey and the Annual Social and Economic Supplements. Covariates We included the following state-level variables as possible confounders based on factors shown to be associated with the outcomes of physical abuse and neglect and thought to be associated with the exposure, Medicaid expansion. Paid family leave policy status was determined from the National Partnership for Women & Families data. Unemployment proportion was obtained from the US Bureau of Labor Statistics. Teenage birth rate data were obtained from the National Center for Health Statistics National Vital Statistics System. The percentage of families with children younger than 5 years living below the FPL was determined using the American Community Survey estimates from the US Census Bureau. Finally, child care wait list data were obtained from the National Women’s Law Center. Statistical Analysis We compared prespecified baseline characteristics of expansion and nonexpansion states in 2013. To compare the change in physical abuse and neglect reporting rates in states that expanded Medicaid vs those that did not during the 7-year period of interest, we conducted difference-in-difference analyses with general linear regression models with state and year as fixed effects (eMethods in the JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 4/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion Supplement) in which the regression coefficient for Medicaid expansion is interpretable as a generalized form of difference-in-difference. Difference-in-difference analyses allow for the comparison of a change in mean outcome between a group that has experienced a treatment and a group that has not experienced the treatment before and after that treatment occurs. We calculated descriptive statistics, including median and interquartile range (IQR) for the change in Medicaid eligibility. A general linear regression model with state and year as fixed effects was used to examine the association between the Medicaid expansion exposure and the change in Medicaid eligibility that was treated as a continuous outcome. To further analyze the association between the intermediate outcome of change in Medicaid eligibility and the final outcomes of interest, physical abuse and neglect, general linear regression analyses were performed. The same analyses were conducted for the change in parental Medicaid coverage. For all the models, to identify influential data points, the difference-in-fits approach was used. This diagnostic technique omits each observation, one at a time, and refits the regression model to determine whether particular observations have an outsized influence on the fitted model. This approach identified an unexplained near doubling of West Virginia’s physical abuse rates, possibly due to a CPS intake change, in 2014. West Virginia was therefore excluded from analyses that included physical abuse or neglect outcomes, although secondary analyses including West Virginia were conducted to evaluate the exclusion’s potential impact. Analyses were also conducted excluding several other potential outliers. The number of averted cases after expansion was calculated based on the estimate for the outcome rate and the known child population in the nonexpansion states for the years 2014 through 2016. All analyses were conducted using SAS, version 9.4 (SAS Institute, Inc). The analyses were conducted from April 12, 2018, through March 26, 2019. All 2-sided P values were considered statistically significant at P < .05. Results We analyzed NCANDS data for 31 states and the District of Columbia that expanded Medicaid and 19 states that did not during the period from January 1, 2010, through December 31, 2016. Table 1 shows that before Medicaid expansion (in 2013), the median teen birth rate and the proportion of families with children younger than 5 years living below the FPL were higher in the states that did not expand Medicaid than in the states that did. Conversely, the proportion of parents with Medicaid coverage was lower in the states that did not go on to expand Medicaid. From 2013 to 2016, Medicaid coverage for adults with dependent children increased a median 1.9% (IQR, 0.4% to 4.3%) in the states that did not expand Medicaid and 4.2% (IQR, 0.9% to 6.0%) in the states that did (Table 2). In the states that did not expand Medicaid, the Medicaid eligibility cutoff for parents decreased 6 percentage points of the FPL (IQR, −17.0 to 2.5), whereas in the states that did expand, the eligibility cutoff increased by 42.0 percentage points of the FPL (IQR, 5.0 to 81.0). Medicaid coverage for adults with 1 or more dependent children was found to increase (0.6 Table 1. Comparison of the States With vs Without Medicaid Expansion in 2013 States b c Baseline Characteristic With Medicaid Expansion (n = 31) No Medicaid Expansion (n = 20) No. of children aged <6 y by state 279 291 (79 339-535 097) 354 489 (198 220-550 928) Abbreviation: FPL, federal poverty level. Total No. of children aged <6 y 14 008 018 9 845 257 Unless otherwise indicated, data are expressed as Unemployment rate, % 7.3 (6.6-8.1) 6.7 (5.3-7.4) median (interquartile range). Teen birth rate per 1000 women 23.6 (17.7-30.3) 29.6 (24.8-34.5) Includes Washington, DC. US population below FPL, % 16.8 (13.5-21.2) 20.7 (18.3-23.6) Louisiana began expansion in July 2016 and is Parents with Medicaid coverage, % 14.6 (12.3-19.5) 9.2 (8.0-13.8) included among the states that did not expand States with childcare wait list, No. (%) 12 (39) 7 (35) Medicaid. States with paid leave, No. (%) 2 (6) 0 Includes parents with children younger than 5 years. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 5/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion percentage points per year; 95% CI, 0.2 to 1.0) and was positively associated with Medicaid expansion status (1.6 percentage points higher for a given year; 95% CI, 0.1 to 3.0) (eTable 1 in the Supplement). In comparing the years 2013 and 2016, Medicaid expansion was found to be associated with a significant increase in percentage of FPL Medicaid eligibility, as would be expected (48.6 percentage points higher; 95% CI, 22.5 to 74.6). In 2013, the baseline count of physical abuse cases was 176 591 in expansion states and 112 047 in nonexpansion states, or 1261 and 1138 per 100 000 children younger than 6 years, respectively. The baseline count of neglect cases was 646 463 in expansion states and 388 265 in nonexpansion states, or 4615 and 3944 per 100 000 respectively. From 2013 to 2016, many individual states in the Medicaid expansion and nonexpansion groups experienced decreased rates of physical abuse and neglect (Figure 2). When comparing the entire preexpansion and postexpansion periods, the states with Medicaid expansion as a whole saw the physical abuse rate decrease by 68 cases per 100 000 children younger than 6 years and the neglect rate decrease by 336 cases per 100 000 children younger than 6 years. In comparison, the group of states without Medicaid expansion had a decreased rate of physical abuse of 35 cases per 100 000 and an increased rate of neglect by 90 cases per 100 000 children younger than 6 years. Table 3 shows the estimates for net change in the physical abuse and neglect rates after Medicaid expansion in the states that expanded Medicaid relative to the states that did not in any year. After Medicaid expansion, states that expanded Medicaid had fewer reported cases of neglect after adjusting for confounders (−422 per 100 000 children younger than 6 years; 95% CI, −753 to −91) compared with the change in the rate during that time in nonexpansion states, which had a baseline rate of 3944 per 100 000 children younger than 6 years in 2013 (Table 3). No statistically significant association of Medicaid expansion with physical abuse rates was found. Findings were similar when including West Virginia and excluding possible outlier state data (eTable 2 and eTable 3 in the Supplement). Neither Medicaid coverage itself nor Medicaid eligibility criteria were significantly associated with physical abuse or neglect rates, with or without adjusting for potential confounders (eTable 4 and eTable 5 in the Supplement). Based on these results, we estimated that had the nonexpansion states expanded Medicaid, a total of 124 981 (95% CI, 26 921-223 041) fewer cases of neglect would have been reported in the United States from 2014 through 2016, or a mean of 41 660 fewer cases per year. Discussion This study suggests an association between the ACA Medicaid expansion and reductions in the rate of screened-in reports made to CPS for concerns of neglect in children younger than 6 years. This association was seen even after controlling for other factors, including state-level policies and measures that might affect the reporting rate of child neglect and be associated with the status of Medicaid expansion in a state. The results add to the small but growing body of evidence demonstrating a positive association 15,16 between a social policy and child maltreatment outcomes. Neglect represents approximately three-quarters of reported child maltreatment cases in the United States. Despite being the most Table 2. Net Change From 2013 to 2016 of Medicaid Coverage and Medicaid Eligibility Criteria for Adults With Dependent Children Medicaid Expansion, Median (IQR) Net Change Present (n = 31) Absent (n = 20) Medicaid coverage, % 4.2 (0.9 to 6.0) 1.9 (0.4 to 4.3) Medicaid eligibility, % of FPL 42.0 (5.0 to 81.0) −6.0 (−17.0 to 2.5) Abbreviations: FPL, federal poverty level; IQR, interquartile range. Louisiana began expansion in July 2016 and is included among the states that did not expand Medicaid. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 6/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion common form of child maltreatment, efforts to prevent neglect have proven difficult in part because of a lack of consensus regarding what constitutes neglect and how to best stratify risk for 57,58 neglect. Our findings offer a promising avenue for future child neglect prevention research, especially as new states expand Medicaid. The absence of a significant association between Medicaid expansion and rates of reported physical abuse could be due to one of several possibilities. The first is that there may truly not be any Figure 2. Change in the Physical Abuse and Neglect Rates Compared by State for 2013 vs 2016 A Physical abuse in expansion states B Neglect in expansion states MI MI KY RI DE WA MT IA NV DC AK AK AR AZ ND KY NH NM NJ NY AZ AR OH MD WA NH CA OH NY NJ NM DE VT CA RI ND IA CT CT CO OR HI CO IL IL VT HI PA MA NV MD MN PA OR IN MA MN MT DC IN –2500 –2000 –1500 –1000 –500 0 500 1000 1500 2000 2500 3000 3500 –4000 –3000 –2000 –1000 0 1000 2000 No. of Cases of Physical Abuse No. of Cases of Neglect per 100 000 Children C Physical abuse in nonexpansion states D Neglect in nonexpansion states OK TN KS ME FL NC SD FL WI SD TX WY WY KS UT TX LA VA VA NE NE UT NC WI MO LA GA ID MS MO AL AL ME OK ID SC SC MS TN GA –2500 –2000 –1500 –1000 –500 0 500 1000 1500 2000 2500 3000 3500 –4000 –2000 –2000 –1000 0 1000 2000 No. of Cases of Physical Abuse per 100 000 Children No. of Cases of Neglect per 100 000 Children Physical abuse and neglect rates are expressed as cases per 100 000 children younger in this analysis. For West Virginia (not included), the change in physical abuse rates is than 6 years. Negative changes indicate the rate decreased from 2013 to 2016. Louisiana thought to be an outlier not due to an actual change but to data unreliability concerns did not expand Medicaid until July 2016 and is classified with the nonexpansion states possibly related to a change in 2014 to the Child Protective Services intake system. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 7/13 States Not Expanded States Expanded States Not Expanded States Expanded JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion association between Medicaid expansions and child physical abuse, suggesting that at least some of the pathways resulting in physical abuse are different from those resulting in neglect. An association may also exist between Medicaid expansion and physical abuse, but this association is seen only after more time has lapsed or only for a younger subset of children when compared with neglect. This possibility suggests that the association is delayed relative to that seen with neglect and was therefore not captured in the relatively short study period. After state expansion occurred, some states saw much larger changes in eligibility and coverage than others. Therefore, we conducted further analyses to evaluate not only the association between Medicaid expansion and child maltreatment outcomes but also the associations between Medicaid expansion and these intermediate outcomes. We found that expanding Medicaid was associated with increases in the percentage of FPL eligibility cutoff as well as the proportion of parents covered by Medicaid. This finding is in line with those of prior research demonstrating that the expansion of 22,26,59 Medicaid had the intended effect of increasing Medicaid coverage. Although Medicaid expansion was independently associated with the percentage of FPL eligibility cutoff and the proportion of parents covered by Medicaid and with neglect rates, no significant association was found between the former and the latter. This finding may suggest that the association between Medicaid expansion and neglect is complex and does not simply involve increased coverage. Another possibility is that a direct association exists between increased coverage and neglect but was not adequately detected in our analyses. Medicaid coverage increased even in states that did not expand Medicaid. Other studies have attributed this finding to a “welcome-mat” effect among people who had been eligible for Medicaid before the expansions took place but only applied for Medicaid coverage after learning more about health care coverage through the national health care discussions. Despite this change, we found that the increase in Medicaid coverage for parents was significantly larger in the states that expanded Medicaid than in the states that did not. Limitations This study had several limitations. Considering the Medicaid expansion as a binary exposure may not have fully captured some of the more complex changes that occurred within states. Medicaid expansion may be associated with other health care changes that improved patients’ ability to access care and services differently in different states, although they all opted to expand Medicaid. Medicaid delivery varies by state. Some of this variation is due to Medicaid waiver programs, which allow states to modify federal Medicaid rules to test different delivery and payment models. For example, as Table 3. Changes in Annual Child Abuse and Neglect Rates in States That Expanded Medicaid Relative to States That Did Not Variable Estimate of Rate per 100 000 Children Aged <6 y for Each Study Year (95% CI) Variable Physical Abuse Neglect Medicaid expansion −57 (−213 to 99) −422 (−753 to −91) Family policy −101 (−661 to 459) −1092 (−2282 to 98) Unemployment rate −62 (−125 to 1) −85 (−219 to 49) Families living below FPL 9 (−10 to 27) 31 (−9 to 71) Teen births 5 (−21 to 31) −8 (−63 to 48) Childcare wait list 96 (−110 to 303) 153 (−287 to 592) Abbreviation: FPL, federal poverty level. The unadjusted change for physical abuse was −33 (95% CI, −186 to 119) per 100 000 children and for neglect was −427 (95% CI, −752 to −102) per 100 000 children (P < .05). The adjusted model controlled for the following state-level covariables: paid family leave policy, unemployment, teen birth rate, proportion of families living in poverty, and presence of a child care wait list. Medicaid expansion indicates states with relative to those without this expansion. The analyses exclude West Virginia because its outcome data were found to be unreliable. P < .05. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 8/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion of December 7, 2018, there were 46 approved and 25 pending Section 1115 waivers pertaining to matters as varied as Medicaid work requirements and healthy behavior incentives. In addition to increasing Medicaid coverage, Medicaid expansion also involved states receiving federal funds to cover 100% of newly eligible Medicaid recipients. These additional funds may have allowed states to cover additional services affecting parents’ ability to provide for their children. Studying state- level measures of Medicaid delivery such as the number of health care professionals per capita accepting Medicaid, evaluating Medicaid waiver programs themselves, or looking at how states used the additional federal Medicaid funds may help identify specific pathways through which the association occurs between Medicaid expansion and decreased child neglect. When considering Medicaid eligibility for parents based on income, considerable heterogeneity between states existed even before the ACA Medicaid expansion took place. We controlled for these baseline differences by conducting difference-in-difference analyses in which states were compared with themselves before and after Medicaid expansion. In addition, we were unable to capture and include state-level changes to child physical abuse and neglect policies and reporting laws that might have confounded our results. We attempted to control for these changes indirectly by controlling for other state-level variables that could have a bearing on them, although future research could evaluate the effect of these changes directly by collecting and including data on state-level policy changes over time. It has been previously demonstrated that only a subset of maltreated children are 63,64 reported to CPS and therefore included in the NCANDS database used for this study. Including the entire population of maltreated children might alter the results by strengthening or weakening the association seen herein. Conclusions This study found that ACA Medicaid expansion was associated with a reduction in the reported child neglect rate but with no significant change to the physical abuse rate. It appears from these findings that future studies should focus on assessing what the mechanism for this association is to determine whether expanding Medicaid prevents child neglect and whether further expansions could be designed to bolster this secondary association. ARTICLE INFORMATION Accepted for Publication: April 22, 2019. Published: June 14, 2019. doi:10.1001/jamanetworkopen.2019.5529 Correction: This article was corrected on July 5, 2019, to fix a typographical error in the running title. Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Brown ECB et al. JAMA Network Open. Corresponding Author: Emily C. B. Brown, MD, MS, Seattle Children’s Research Institute, Center for Child Health, Behavior, and Development, 4800 Sand Point Way NE, PO Box 5371, Seattle, WA 98105 (emily.brown@ seattlechildrens.org). Author Affiliations: Seattle Children’s Research Institute, Center for Child Health, Behavior, and Development, Seattle, Washington (Brown, Garrison, Bao, Qu); Department of Pediatrics, University of Washington School of Medicine, Seattle (Brown, Jenny, Rowhani-Rahbar); Division of Child Psychiatry, University of Washington School of Medicine, Seattle (Garrison); Department of Epidemiology, University of Washington School of Public Health, Seattle (Rowhani-Rahbar); Harborview Injury Prevention & Research Center, University of Washington, Seattle (Rowhani-Rahbar). Author Contributions: Dr Brown had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Brown, Jenny, Rowhani-Rahbar. Acquisition, analysis, or interpretation of data: Brown, Garrison, Bao, Qu, Rowhani-Rahbar. Drafting of the manuscript: Brown, Bao, Qu. JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 9/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion Critical revision of the manuscript for important intellectual content: Brown, Garrison, Jenny, Rowhani-Rahbar. Statistical analysis: Brown, Bao, Qu. Obtained funding: Brown. Administrative, technical, or material support: Jenny, Rowhani-Rahbar. Supervision: Brown, Garrison, Jenny, Rowhani-Rahbar. Conflict of Interest Disclosures: Drs Brown and Jenny reported providing expert testimony, as part of their clinical responsibilities, in cases of alleged child maltreatment. Dr Brown reported receiving grants from Helfer Society during the conduct of the study. Dr Rowhani-Rahbar reported receiving funding from the Centers for Disease Control and Prevention outside the submitted work. No other disclosures were reported. Funding/Support: This study was funded by a career development grant from the Helfer Society. Role of the Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Additional Contributions: Kathryn B. Whitlock, MS, Seattle Children’s Research Institute, Seattle, Washington, aggregated the data and conducted the preliminary analyses. She was compensated for this work through the biostatistical services core at Seattle Children’s Research Institute. REFERENCES 1. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: the Adverse Childhood Experiences (ACE) Study. Am J Prev Med. 1998;14(4): 245-258. doi:10.1016/S0749-3797(98)00017-8 2. Felitti VJ. Adverse childhood experiences and adult health. Acad Pediatr. 2009;9(3):131-132. doi:10.1016/j.acap. 2009.03.001 3. Nemeroff CB. Paradise lost: the neurobiological and clinical consequences of child abuse and neglect. Neuron. 2016;89(5):892-909. doi:10.1016/j.neuron.2016.01.019 4. National Council of Juvenile and Family Court Judges. Adverse Childhood Experience (ACE) questionnaire. http:// www.ncjfcj.org/sites/default/files/Finding%20Your%20ACE%20Score.pdf. Published October 24, 2006. Accessed December 5, 2016. 5. US Department of Health and Human Services. Child abuse and neglect cost the United States $124 billion. http:// www.cdc.gov/media/releases/2012/p0201_child_abuse.html. Published February 1, 2012. Accessed December 5, 6. Fang X, Brown DS, Florence CS, Mercy JA. The economic burden of child maltreatment in the United States and implications for prevention. Child Abuse Negl. 2012;36(2):156-165. doi:10.1016/j.chiabu.2011.10.006 7. Peterson C, Florence C, Klevens J. The economic burden of child maltreatment in the United States, 2015. Child Abuse Negl. 2018;86:178-183. doi:10.1016/j.chiabu.2018.09.018 8. US Department of Health and Human Services. Child Maltreatment 2014. https://www.acf.hhs.gov/sites/default/ files/cb/cm2014.pdf. Published January 25, 2016. Accessed May 9, 2019. 9. Finkelhor D, Turner HA, Shattuck A, Hamby SL. Prevalence of childhood exposure to violence, crime, and abuse: results from the National Survey of Children’s Exposure to Violence. JAMA Pediatr. 2015;169(8):746-754. doi:10. 1001/jamapediatrics.2015.0676 10. Mikton C, Butchart A. Child maltreatment prevention: a systematic review of reviews. Bull World Health Organ. 2009;87(5):353-361. doi:10.2471/BLT.08.057075 11. Scribano PV. Prevention strategies in child maltreatment. Curr Opin Pediatr. 2010;22(5):616-620. 12. Wood JN. Challenges in prevention of abusive head trauma. JAMA Pediatr. 2015;169(12):1093-1094. doi:10. 1001/jamapediatrics.2015.3023 13. Curry SJ, Krist AH, Owens DK, et al; US Preventive Services Task Force. Interventions to prevent child maltreatment: US Preventive Services Task Force recommendation statement. JAMA. 2018;320(20):2122-2128. doi:10.1001/jama.2018.17772 14. Klevens J, Whitaker DJ. Primary prevention of child physical abuse and neglect: gaps and promising directions. Child Maltreat. 2007;12(4):364-377. doi:10.1177/1077559507305995 15. Klevens J, Luo F, Xu L, Peterson C, Latzman NE. Paid family leave’s effect on hospital admissions for pediatric abusive head trauma. Inj Prev. 2016;22(6):442-445. doi:10.1136/injuryprev-2015-041702 16. Klevens J, Barnett SB, Florence C, Moore D. Exploring policies for the reduction of child physical abuse and neglect. Child Abuse Negl. 2015;40:1-11. doi:10.1016/j.chiabu.2014.07.013 JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 10/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion 17. The Henry J. Kaiser Family Foundation. Status of state action on the Medicaid expansion decision. http://kff.org/ health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care- act/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D. Published November 7, 2018. Accessed November 16, 2018. 18. Ayanian JZ, Ehrlich GM, Grimes DR, Levy H. Economic effects of Medicaid expansion in Michigan. N Engl J Med. 2017;376(5):407-410. doi:10.1056/NEJMp1613981 19. Anderson ME, Glasheen JJ, Anoff D, Pierce R, Lane M, Jones CD. Impact of state Medicaid expansion status on length of stay and in-hospital mortality for general medicine patients at US academic medical centers. JHosp Med. 2016;11(12):847-852. doi:10.1002/jhm.2649 20. Lyon SM, Douglas IS, Cooke CR. Medicaid expansion under the Affordable Care Act: implications for insurance- related disparities in pulmonary, critical care, and sleep. Ann Am Thorac Soc. 2014;11(4):661-667. doi:10.1513/ AnnalsATS.201402-072PS 21. Rudowitz R, Artiga S, Young K. What coverage and financing is at risk under a repeal of the ACA Medicaid expansion? http://kff.org/report-section/what-coverage-and-financing-is-at-risk-under-a-repeal-of-the-aca- medicaid-expansion-issue-brief/. Published December 6, 2016. Accessed March 15, 2017. 22. Antonisse L, Garfield R, Rudowitz R, Artiga S. The effects of Medicaid expansion under the ACA: updated findings from a literature review. https://www.kff.org/medicaid/issue-brief/the-effects-of-medicaid-expansion- under-the-aca-updated-findings-from-a-literature-review-march-2018/. Published March 28, 2018. Accessed October 10, 2018. 23. Bhatt CB, Beck-Sagué CM. Medicaid expansion and infant mortality in the United States. Am J Public Health. 2018;108(4):565-567. doi:10.2105/AJPH.2017.304218 24. Wherry LR, Miller S. Early coverage, access, utilization, and health effects associated with the Affordable Care Act Medicaid expansions: a quasi-experimental study. Ann Intern Med. 2016;164(12):795-803. doi:10.7326/ M15-2234 25. McMorrow S, Gates JA, Long SK, Kenney GM. Medicaid expansion increased coverage, improved affordability, and reduced psychological distress for low-income parents. Health Aff (Millwood). 2017;36(5):808-818. doi:10. 1377/hlthaff.2016.1650 26. Kenney G, Anderson N, Long S, et al. Taking stock: health insurance coverage for parents under the ACA in 2014. http://hrms.urban.org/briefs/Health-Insurance-Coverage-for-Parents-under-the-ACA-in-2014.html. Published September 9, 2014. Accessed May 9, 2019. 27. Jenny C. Child Abuse and Neglect: Diagnosis, Treatment, and Evidence. St Louis, MO: Saunders/Elsevier; 2011. 28. US Department of Health and Human Services, Administration for Children and Families, Administration on Children, Youth and Families, Children’s Bureau. National child abuse and neglect data system (NCANDS) child file, FFYs 2010-2016 [dataset]. http://www.ndacan.cornell.edu. Updated 2018. Accessed March 1, 2019. 29. Kaiser Family Foundation. Medicaid income eligibility limits for parents, 2002-2018. https://www.kff.org/ medicaid/state-indicator/medicaid-income-eligibility-limits-for-parents/?currentTimeframe=0& sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D. Updated 2018. Accessed May 10, 2018. 30. Centers for Disease Control and Prevention (CDC). Bridged-race population estimates. https://wonder.cdc.gov/ bridged-race-population.html. Published June 30, 2016. Accessed March 29, 2017. 31. Font SA, Maguire-Jack K. Decision-making in child protective services: influences at multiple levels of the social ecology. Child Abuse Negl. 2015;47:70-82. doi:10.1016/j.chiabu.2015.02.005 32. Kohl PL, Jonson-Reid M, Drake B. Time to leave substantiation behind: findings from a national probability study. Child Maltreat. 2009;14(1):17-26. doi:10.1177/1077559508326030 33. Hussey JM, Marshall JM, English DJ, et al. Defining maltreatment according to substantiation: distinction without a difference? Child Abuse Negl. 2005;29(5):479-492. doi:10.1016/j.chiabu.2003.12.005 34. Drake B, Jonson-Reid M, Way I, Chung S. Substantiation and recidivism. Child Maltreat. 2003;8(4):248-260. doi:10.1177/1077559503258930 35. Leiter J, Others A. Substantiated and unsubstantiated cases of child maltreatment: do their consequences differ? Soc Work Res. 1994;18(2):67-82. doi:10.1093/swr/18.2.67 36. Hammond I, Eastman AL, Leventhal JM, Putnam-Hornstein E. Maternal mental health disorders and reports to child protective services: a birth cohort study. Int J Environ Res Public Health. 2017;14(11):1320. doi:10.3390/ ijerph14111320 JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 11/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion 37. Parrish JW, Shanahan ME, Schnitzer PG, Lanier P, Daniels JL, Marshall SW. Quantifying sources of bias in longitudinal data linkage studies of child abuse and neglect: measuring impact of outcome specification, linkage error, and partial cohort follow-up. Inj Epidemiol. 2017;4(1):23. doi:10.1186/s40621-017-0119-6 38. Putnam-Hornstein E. Report of maltreatment as a risk factor for injury death: a prospective birth cohort study. Child Maltreat. 2011;16(3):163-174. doi:10.1177/1077559511411179 39. Drake B, Lee SM, Jonson-Reid M. Race and child maltreatment reporting: are blacks overrepresented? Child Youth Serv Rev. 2009;31(3):309-316. doi:10.1016/j.childyouth.2008.08.004 40. Westat I, McDonald WR; US Administration for Children and Families Office of Planning, Research, and Evaluation. Fourth National Incidence Study of Child Abuse and Neglect (NIS-4): Report to Congress. Washington, DC: US Dept of Health and Human Services, Administration for Children and Families, Office of Planning, Research and Evaluation and the Children's Bureau; 2010. 41. Leclerc B, Chiu YN, Cale J. Sexual violence and abuse against children: a first review through the lens of environmental criminology. Int J Offender Ther Comp Criminol. 2016;60(7):743-765. doi:10.1177/ 0306624X14564319 42. Finkelhor D, Shattuck A, Turner HA, Hamby SL. The lifetime prevalence of child sexual abuse and sexual assault assessed in late adolescence. J Adolesc Health. 2014;55(3):329-333. doi:10.1016/j.jadohealth.2013.12.026 43. Hibbard R, Barlow J, Macmillan H; Committee on Child Abuse and Neglect; American Academy of Child and Adolescent Psychiatry; Child Maltreatment and Violence Committee. Psychological maltreatment. Pediatrics. 2012;130(2):372-378. doi:10.1542/peds.2012-1552 44. Kimber M, MacMillan HL. Child psychological abuse. Pediatr Rev. 2017;38(10):496-498. doi:10.1542/pir. 2016-0224 45. US Department of Health and Human Services. Child maltreatment 2016. https://www.acf.hhs.gov/sites/default/ files/cb/cm2016.pdf. Published 2018. Accessed May 9, 2019. 46. Henry J. Kaiser Family Foundation. Health insurance coverage of adults with dependent children. http://kff.org/ other/state-indicator/nonelderly-adults-with-dependents/?currentTimeframe=0&sortModel=%7B%22colId% 22:%22Location%22,%22sort%22:%22asc%22%7D. Published 2017. Accessed March 15, 2017. 47. US Census Bureau. Current Population Survey (CPS). https://www.census.gov/programs-surveys/cps.html. Accessed May 8, 2019. 48. US Census Bureau. Current Population Survey (CPS): data. https://www.census.gov/programs-surveys/cps/ data-detail.html. Revised March 6, 2018. Accessed May 8, 2019. 49. National Partnership for Women & Families. State paid family and leave insurance laws. http://www. nationalpartnership.org/research-library/work-family/paid-leave/state-paid-family-leave-laws.pdf. Published February 2017. Accessed March 29, 2017. 50. US Department of Labor, Bureau of Labor Statistics. Local area unemployment statistics. https://www.bls. gov/lau/. Updated 2017. Accessed March 29, 2017. 51. Centers for Disease Control and Prevention (CDC), National Center for Health Statistics. National Vital Statistics System. https://www.cdc.gov/nchs/nvss/. Reviewed March 14, 2019. Accessed March 29, 2017. 52. US Census Bureau. American Community Survey. https://www.census.gov/acs/www/data/data-tables-and- tools/index.php. Updated 2016. Accessed March 29, 2017. 53. National Women’s Law Center. Persistent gaps: state child care assistance policies. https://nwlc.org/resources/ persistent-gaps-state-child-care-assistance-policies-2017/. Updated 2017. Accessed September 9, 2017. 54. Wing C, Simon K, Bello-Gomez RA. Designing difference in difference studies: best practices for public health policy research. Annu Rev Public Health. 2018;39:453-469. doi:10.1146/annurev-publhealth-040617-013507 55. SAS Institute. Influence statistics. SAS/STAT(R) 9.22 User's Guide https://support.sas.com/documentation/cdl/ en/statug/63347/HTML/default/viewer.htm#statug_reg_sect040.htm. Accessed March 12, 2019. 56. US Department of Health and Human Services. Child Maltreatment 2015. https://www.acf.hhs.gov/sites/default/ files/cb/cm2015.pdf. Published 2017. Accessed May 9, 2019. 57. Rebbe R. What is neglect? state legal definitions in the United States. Child Maltreat. 2018;23(3):303-315. doi: 10.1177/1077559518767337 58. Dubowitz H, Christian CW, Hymel K, Kellogg ND. Forensic medical evaluations of child maltreatment: a proposed research agenda. Child Abuse Negl. 2014;38(11):1734-1746. doi:10.1016/j.chiabu.2014.07.012 59. Frean M, Gruber J, Sommers BD. Premium subsidies, the mandate, and Medicaid expansion: coverage effects of the Affordable Care Act. J Health Econ. 2017;53:72-86. doi:10.1016/j.jhealeco.2017.02.004 JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 12/13 JAMA Network Open | Health Policy Rates of Child Maltreatment in States With and Without Medicaid Expansion 60. Medicaid.gov. State waivers list. https://www.medicaid.gov/medicaid/section-1115-demo/demonstration-and- waiver-list/index.html. Updated 2018. Accessed December 10, 2018. 61. Henry J. Kaiser Family Foundation. Medicaid waiver tracker: approved and pending Section 1115 waivers by state. https://www.kff.org/medicaid/issue-brief/medicaid-waiver-tracker-approved-and-pending-section-1115- waivers-by-state/. Published December 7, 2018. Accessed December 10, 2018. 62. Snyder L, Rudowitz R. Medicaid financing: how does it work and what are the implications? https://www.kff. org/medicaid/issue-brief/medicaid-financing-how-does-it-work-and-what-are-the-implications/. Published May 20, 2015. Accessed January 22, 2019. 63. Fallon B, Trocmé N, Fluke J, MacLaurin B, Tonmyr L, Yuan YY. Methodological challenges in measuring child maltreatment. Child Abuse Negl. 2010;34(1):70-79. doi:10.1016/j.chiabu.2009.08.008 64. Sumner SA, Mercy JA, Dahlberg LL, Hillis SD, Klevens J, Houry D. Violence in the United States: status, challenges, and opportunities. JAMA. 2015;314(5):478-488. doi:10.1001/jama.2015.8371 SUPPLEMENT. eMethods. Data Source and Main Outcome Models eTable 1. Change in Parental Medicaid Coverage and Medicaid Eligibility Criteria From 2013 to 2016 as a Function of Medicaid Expansion Status eTable 2. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Including West Virginia eTable 3. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Excluding Possible Outlier States eTable 4. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Coverage eTable 5. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Eligibility Criteria JAMA Network Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 (Reprinted) June 14, 2019 13/13 Supplementary Online Content Brown ECB, Garrison MM, Bao H, Qu P, Jenny C, Rowhani-Rahbar A. Assessment of rates of child maltreatment in states with Medicaid expansion vs states without Medicaid expansion. JAMA Netw Open. 2019;2(6):e195529. doi:10.1001/jamanetworkopen.2019.5529 eMethods. Data Source and Main Outcome Models eTable 1. Change in Parental Medicaid Coverage and Medicaid Eligibility Criteria From 2013 to 2016 as a Function of Medicaid Expansion Status eTable 2. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Including West Virginia eTable 3. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Excluding Possible Outlier States eTable 4. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Coverage eTable 5. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Eligibility Criteria This supplementary material has been provided by the authors to give readers additional information about their work. ©2019 Brown ECB et al. JAMA Netw Open. eMethods. Data Source and Main Outcome Models The analyses presented in this publication were based on data from the National Child Abuse and Neglect Data System (NCANDS) Child File, FFYs 2010v5, 2011v5, 2012v4, 2013v3, 2014v2, 2015v1, and 2016v1. These data were provided by the National Data Archive on Child Abuse and Neglect at Cornell University, and have been used with permission. The data were originally ureau. Funding was provided by Children, Youth and Families, Administration for Children and Families, U.S. Department of Health and Human Services. The collector of the original data, the funding agency, NDACAN, Cornell University, and the agents or employees of these institutions bear no responsibility for the analyses or interpretations presented here. The information and opinions expressed reflect solely the opinions of the authors. Main outcome models Unadjusted: Y expansion + + it 0 1 it i t it Adjusted: Y = + expansion + leave + unemp + teen + fpl + wait + + + it 0 1 it 2 it 3 it 4 it 5 it 6 it i t it In these models, Y denotes the aggregated outcome rate calculated from individual-level data as the number of it screened-in referrals for physical abuse or neglect per 100,000 children under the of 6 years for each state and each year, expansion denotes whether or not a state had a Medicaid expansion at that year (0, 1, or a fraction for the it states that expanded Medicaid in the middle of a year), and and respectively denote the state and year fixed i t effects. In addition, the vector of covariates for each state in each year included leave denoting paid family leave policy status, unemp denoting the unemployment proportion, teen denoting the teen pregnancy rate, fpl denoting the percentage of families with children under the age of 5 living below the FPL, wait denoting whether there was a child care wait list, and denoting the error term. Based on this model, 1 is the effect of Medicaid expansion after accounting for covariates as well as state and year effects. ©2019 Brown ECB et al. JAMA Netw Open. eTable 1. Change in Parental Medicaid Coverage and Medicaid Eligibility Criteria From 2013 to 2016 as a Function of Medicaid Expansion Status Medicaid Medicaid coverage eligibility Medicaid 1.56 (0.10, 48.60 expansion 3.02)* (22.50, estimate (95% 74.60)* CI ) Medicaid coverage is expressed as a percentage and Medicaid eligibility represents the percent of federal poverty level CI = confidence interval *p-value < 0.05 ©2019 Brown ECB et al. JAMA Netw Open. eTable 2. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Including West Virginia Variable Variable estimate (95% CI ) Physical abuse Neglect Medicaid expansion 140 (-103 , 382) -399 (-724, -74)* Family policy 91 (-789 , 971) -1070 (-2250, 110) Unemployment rate 76 (-20, 173) -69 (-199, 61) Below FPL 46 (17, 74)* 35 (-3, 73) Teen births -41 (-81, 0)* -13 (-67, 41) Waitlist 29 (-296, 353) 145 (-291, 581) The unadjusted changes for physical abuse and neglect were 119 (-123, 360) and -407 (-728 , -87)* per 100,000 children respectively. The adjusted model presented above controlled for the following state-level covariables: paid family leave policy, unemployment, teen birth rate, proportion of families living in poverty, and presence of a child care wait list. CI = confidence interval FPL = federal poverty level *p-value < 0.05 ©2019 Brown ECB et al. JAMA Netw Open. eTable 3. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years in States That Expanded Medicaid Relative to States That Did Not Expand Medicaid Excluding Possible Outlier States Variable Variable estimate (95% CI ) Physical abuse Neglect Medicaid expansion 26 (-78, 130) -354 (-673, -35)* Family policy -170 (-535, 196) -1303 (-2427, -180)* Unemployment rate -63 (-107, -20)* -163 (-297, -29)* Below FPL 1 (-11, 14) 36 (-2, 74) Teen births -4 (-21, 14) 5 (-48, 58) Waitlist 99 (-36, 233) 170 (-243, 583) The states excluded were Michigan, Oklahoma, Tennessee, and West Virginia. The unadjusted changes for physical abuse and neglect were 32 (-71, 136) and -347 (-667, -27)* per 100,000 children respectively. The adjusted model presented above controlled for the following state-level covariables: paid family leave policy, unemployment, teen birth rate, proportion of families living in poverty, and presence of a child care wait list. CI = confidence interval FPL = federal poverty level *p-value < 0.05 ©2019 Brown ECB et al. JAMA Netw Open. eTable 4. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Coverage Variable Variable estimate (95% CI ) Model 1 Model 2 Physical abuse Neglect Physical abuse Neglect Medicaid -51 (-104, 2) 16 (-65, 97) -46 (-100, 9) 25 (-58, 108) coverage Family policy 134 (-1470, 1737) -1461 (-3913, 167 (-1444, -1398 (-3854, 992) 1779) 1057) Unemployment -32 (-242, 178) 45 (-277, 366) -50 (-264, 165) 12 (-315, 340) rate Below FPL 14 (-28, 56) 24 (-40, 88) 15 (-27, 58) 27 (-37, 92) Teen births -6 (-109, 97) -11 (-169, 147) 3 (-103, 109) 5 (-156, 166) Waitlist 230 (-650, 1110) 430 (-916, 1776) 231 (-653, 1114) 431 (-915, 1777) Medicaid -196 (-663, 271) -364 (-1076, expansion 347) The unadjusted changes for physical abuse and neglect were -48 (-97, 1) and 16 (-60, 92) per 100,000 children respectively. Model 1 controlled for paid family leave, unemployment, teen birth rate, proportion of families with children under age 5 living below the federal poverty level, and presence of a child care wait list and model 2 controlled for the same covariates with the addition of Medicaid expansion. The analyses above exclude West Virginia as its outcome data was found to be unreliable. Medicaid coverage is expressed as a percentage and Medicaid eligibility represents the percent of federal poverty level CI = confidence interval FPL = federal poverty level ©2019 Brown ECB et al. JAMA Netw Open. eTable 5. Changes in Annual Child Abuse and Neglect Rates per 100 000 Children Younger Than 6 Years From 2013 to 2016 as a Function of Parental Medicaid Eligibility Criteria Variable Variable estimate (95% CI ) Model 1 Model 2 Physical abuse Neglect Physical abuse Neglect Medicaid -4 (-8, 1) -3 (-10, 4) -3 (-8, 3) -1 (-10, 7) eligibility Family policy -297 (-1951, 1356) -1598 (-4075, -211 (-1920, -1435 (-3991, 880) 1498) 1121) Unemployment -56 (-270, 159) 35 (-286, 356) -64 (-283, 155) 20 (-307, 348) rate Below FPL 11 (-32, 54) 20 (-44, 85) 13 (-31, 57) 24 (-43, 90) Teen births -2 (-107, 102) -27 (-184, 130) 5 (-106, 116) -13 (-178, 153) Waitlist 191.92 (-702, 595 (-744, 1934) 175 (-731, 1080) 563 (-791, 1917) 1086) Medicaid -132 (-686, 423) -248 (-1077, 581) expansion The unadjusted changes for physical abuse and neglect were -3 (-7, 1) and -1 (-8, 5) per 100,000 children respectively. Model 1 controlled for paid family leave, unemployment, teen birth rate, proportion of families with children under age 5 living below the federal poverty level, and presence of a child care wait list and model 2 controlled for the same covariates with the addition of Medicaid expansion. The analyses above exclude West Virginia as its outcome data was found to be unreliable. Medicaid coverage is expressed as a percentage and Medicaid eligibility represents the percent of federal poverty level CI = confidence interval FPL = federal poverty level ©2019 Brown ECB et al. JAMA Netw Open.

Journal

JAMA Network Open – American Medical Association

Published: Jun 14, 2019

There are no references for this article.