Comparative Interrupted Time Series Research Articles

A causal machine-learning framework for studying policy impact on air pollution: A case-study in COVID-19 lockdowns.

When studying the impact of policy interventions or natural experiments on air pollution, such as new environmental policies and opening or closing an industrial facility, careful statistical analysis is needed to separate causal changes from other confounding factors. Using COVID-19 lockdowns as a case-study, we present a comprehensive framework for estimating and validating causal changes from such perturbations. We propose using flexible machine learning-based comparative interrupted time series (CITS) models for estimating such a causal effect. We outline the assumptions required to identify causal effects, showing that many common methods rely on strong assumptions that are relaxed by machine learning models. For empirical validation, we also propose a simple diagnostic criterion, guarding against false effects in baseline years when there was no intervention. The framework is applied to study the impact of COVID-19 lockdowns on NO2 in the eastern US. The machine learning approaches guard against false effects better than common methods and suggest decreases in NO2 in Boston, New York City, Baltimore, and Washington D.C. The study showcases the importance of our validation framework in selecting a suitable method and the utility of a machine learning based CITS model for studying causal changes in air pollution time series.

Infant Deaths After Texas’ 2021 Ban on Abortion in Early Pregnancy

Prior observational research has shown that infants born in states with more abortion restrictions are more likely to die during infancy. It is unclear how recent and more severe abortion bans in the US have impacted infant mortality. To examine whether Texas Senate Bill 8 (SB8), which banned abortions after embryonic cardiac activity and did not allow exemptions for congenital anomalies, is associated with infant mortality in the state of Texas. This population-based cohort study of all recorded infant deaths from the state of Texas and 28 comparison states used a comparative interrupted time series analysis with an augmented synthetic control approach and national birth certificate data from January 1, 2018, to December 31, 2022, to estimate the difference between the number of observed and expected infant and neonatal deaths and death rates among monthly cohorts exposed to Texas' SB8. Deaths in March 2022 were treated as the first cohort exposed to the Texas' SB8 abortion policy because these infants (if born full term) were approximately 10 to 14 weeks' gestation when SB8 went into effect on September 1, 2021. The exposure period was thus March through December 2022. Our outcomes were monthly counts and rates of infant (aged <1 year) and neonatal (aged <28 days) deaths in the exposure period in Texas. In secondary analyses, annual changes in cause-specific infant deaths between 2021 and 2022 in Texas and the rest of the US were examined. Between 2018 and 2022, there were 102 391 infant deaths in the US, with 10 351 of these deaths occurring in the state of Texas. Between 2021 and 2022, infant deaths in Texas increased from 1985 to 2240, or 255 additional deaths. This corresponds to a 12.9% increase, whereas the rest of the US experienced a comparatively lower 1.8% increase. On the basis of the counterfactual analysis that used data from Texas and eligible comparison states, an excess of 216 infant deaths (95% CI, -122 to 554) was observed from March to December 2022, or a 12.7% increase above expectation. At the monthly level, significantly greater-than-expected counts were observed for 4 months between March and December 2022: April, July, September, and October. An analysis of neonatal deaths found somewhat similar patterns, with significantly greater-than-expected neonatal deaths in April and October 2022. Descriptive statistics by cause of death showed that infant deaths attributable to congenital anomalies in 2022 increased more for Texas (22.9% increase) but not the rest of the US (3.1% decrease). This study found that Texas' 2021 ban on abortion in early pregnancy was associated with unexpected increases in infant and neonatal deaths in Texas between 2021 and 2022. Congenital anomalies, which are the leading cause of infant death, also increased in Texas but not the rest of the US. Although replication and further analyses are needed to understand the mechanisms behind these findings, the results suggest that restrictive abortion policies may have important unintended consequences in terms of trauma to families and medical cost as a result of increases in infant mortality. These findings are particularly relevant given the recent Dobbs v Jackson Women's Health Organization US Supreme Court decision and subsequent rollbacks of reproductive rights in many US states.

Impact of a Remote Primary Care Telehealth Staffing Model on Primary Care Access in the Veterans Health Administration.

The Veterans Health Administration (VHA) implemented the Clinical Resource Hub (CRH) program to fill staffing gaps in primary care (PC) clinics via telemedicine and maintain veterans' healthcare access. To evaluate PC wait times before and after CRH implementation. Comparative interrupted time series analysis among a retrospective observational cohort of PC clinics who did and did not use CRH during pre-implementation (October 2018-September 2019) and post-implementation (October 2019-February 2020) periods. Clinics completing ≥10 CRH visits per month for 2 consecutive months and propensity matched control clinics. Two measures of patient access (i.e., established, and new patient wait times) and one measure of clinic capacity (i.e., third next available appointment) were assessed. Clinics using CRH were 1:1 propensity score matched across clinical and demographic characteristics. Comparative interrupted time series models used linear mixed effects regression with random clinic-level intercepts and triple interaction (i.e., CRH use, pre- vs. post-implementation, and time) for trend and point estimations. PC clinics using CRH (N = 79) were matched to clinics not using CRH (N = 79). In the 12-month pre-implementation, third next available time increased in CRH clinics (0.16 days/month; 95% CI = [0.07, 0.25]), and decreased in the 5 months post-implementation (-0.58 days/month; 95% CI = [-0.90, -0.27]). Post-implementation third next available time also decreased in control clinics (-0.48 days/month; 95% CI = [-0.81, -0.17]). Comparative differences remained non-significant. There were no statistical differences in established or new patient wait times by CRH user status, CRH implementation, or over time. In a national VHA telemedicine program developed to provide gap coverage for PC clinics, no wait time differences were observed between clinics using and not using CRH services. This hub-and-spoke telemedicine service is an effective model to provide gap coverage while maintaining access. Further investigation of quality and long-term access remains necessary.

Comparative Interrupted Time Series Analysis of Medical Expenses in Patients with Intertrochanteric Fracture Who Underwent Internal Fixation and Hemiarthroplasty.

The objective of this study was to assess postoperative direct medical expenses and medical utilization of elderly patients who underwent either hemiarthroplasty (HA) or internal fixation (IF) for treatment of a femoral intertrochanteric fracture and to analyze differences according to surgical methods and age groups. Data from the 2011 to 2018 Korean National Health Insurance Review & Assessment Service database were used. Risk-set matching was performed for selection of controls representing patients with the same sex, age, and year of surgery. A comparative interrupted time series analysis was performed for evaluation of differences in medical expenses and utilization between the two groups. A total of 10,405 patients who underwent IF surgery and 10,405 control patients who underwent HA surgery were included. Medical expenses were 18% lower in the IF group compared to the HA group during the first year after the fracture (difference-in-difference [DID] estimate ratio 0.82, 95% confidence interval [CI] 0.77-0.87, P<0.001), and 9% lower in the second year (DID estimate ratio 0.91, 95% CI 0.85-0.99, P=0.018). Length of hospital stay was significantly shorter in the IF group compared to the HA group during the first two years after time zero in the age ≥80 group. A noticeable increase in medical expenses was observed for patients who underwent HA for treatment of intertrochanteric fractures compared to those who underwent IF over a two-year period after surgery. Therefore, consideration of such findings is critical when designing healthcare policy support for management of intertrochanteric fractures.

Mandated Rest Breaks and Occupational Injuries and Illnesses in Dallas County, Texas Construction Workers: A Quasi-Experimental, Comparative Interrupted Time Series Study.

We estimated associations of a rest break ordinance, implemented for construction workers in Dallas, Texas in 2016, with workplace injuries and illnesses. We used workers' compensation claims data to compare changes in rates of injuries and illnesses among Dallas County, Texas construction (ie, "treated") workers with changes in untreated workers, before (2013-2015) and after (2016-2018) a rest break ordinance was implemented. Immediately after the ordinance was implemented, rates of injuries/illnesses among treated workers were modestly lower than in comparison workers (rate ratio comparing postmandate vs premandate rates, treated vs comparison workers: 0.89, 95% confidence interval: 0.72-1.11). Postordinance versus preordinance slope trends were similar in the treated versus the comparison group. Ten-minute rest breaks were associated with modestly lower rates of workplace injury/illnesses. More comprehensive standards may be needed for protection.

Answering the call: An evaluation of the Educate to Innovate campaign’s efforts to increase and diversify science, technology, engineering, and mathematics college graduates

Increasing and diversifying STEM college graduates, and consequently the STEM workforce, remains a pressing national priority. As such, the purpose of this quantitative study was to evaluate the Educate to Innovate campaign within higher education. Using preexisting public survey data together with interrupted and comparative interrupted time series techniques, this inquiry aimed to determine whether STEM education has been rejuvenated and broadened or if there remains work to be done. Results ultimately showed that the campaign exerted a statistically significant positive effect on the trend of STEM college graduates. However, no significant difference in the trend between STEM college graduates who are women and STEM college graduates who are men or between minority and nonminority STEM college graduates was observed. Such comprehension is both timely and useful as individuals and organizations along with federal, state, and local policymakers continue to refine and tailor the development of future STEM education interventions.

The decrease in alcohol consumption and suicide rate during the COVID-19 pandemic and their association

Despite the considerable change in alcohol consumption during the COVID-19 pandemic, the impact of the pandemic on the suicide rate in terms of alcohol consumption was not studied. This study was performed to examine whether the change in the suicide rate during the COVID-19 pandemic was related to alcohol consumption and whether the relation was specific to suicides when compared to mortality due to other causes. We performed a comparative interrupted time series (CITS) analysis for the suicide rate of people aged 19 to 60 with three comparison groups (the suicide rate of people aged 19 and under, the cancer death rate of people aged 19 to 60, and alcohol induced death rates). The suicide rate of people aged 19 to 60 and alcohol consumption per capita, along with alcohol induced death rates, continued to decrease during the pandemic in 2020 and 2021 while the suicide rate of people aged 19 and under and the cancer death rate showed an increase. In the comparative interrupted time series model, alcohol consumption had an increasing effect on the adult suicide rate compared to comparison groups when time trends and changes associated with COVID-19 were adjusted. This study shows that the decrease in adult suicide rate in Korea during the pandemic was associated with the decrease in alcohol use among the adult population. Considering that means restriction is the most effective way of controlling suicide and that alcohol can be the most potent and final trigger to suicide, the decrease in suicides during the pandemic and its association with alcohol consumption should be understood as a call for further efforts to decrease alcohol consumption.

Expanding access to medication treatment for opioid use disorders: Findings from the Washington State hub and spoke effort

BackgroundOpioid use disorder (OUD) is a leading cause of preventable death and injury nationwide. Efforts to increase the use of medication for opioid use disorder (MOUD) are needed. In 2017, Washington State implemented a Hub and Spoke (HS) model of care with the primary goal of expanding access to MOUD. We examined changes in MOUD utilization among Washington State Medicaid beneficiaries before and after HS implementation. MethodsWe used Medicaid claims data to examine longitudinal changes in MOUD use for beneficiaries with OUD. We conducted a comparative interrupted time series analysis to examine the association between HS policy implementation and rates of MOUD utilization, overall and by type of medication. ResultsBetween 2016 and 2019, a period of increasing OUD prevalence, rates of MOUD utilization among Washington Medicaid beneficiaries increased overall from 39.7 to 50.5. Following HS implementation, rates of MOUD use grew at a significantly greater rate in the HS cohort than in the non-HS cohort (β=0.54, SE=0.02, p< 0.0001, 95% CI 0.49, 0.59). Analyses by medication type show that this rate increase was primarily due to buprenorphine use (β= 0.61, SE= 0.02, p< 0.0001, 95% CI 0.57, 0.65). ConclusionImproved systems of care are needed to make MOUD accessible to all patients in need. The Washington HS model is one strategy that may facilitate and expand MOUD use, particularly buprenorphine. Over the study period, Washington State saw increased use of buprenorphine, which was an emphasis of their HS model.

Beyond Prescriptive Reforms: An Examination of North Carolina’s Flexible School Restart Program

Although multiple studies have examined the impact of school turnaround, less is known about reforms under the Every Student Succeeds Act (ESSA). To advance this literature, we examine North Carolina’s Restart (NCR) model. NCR aligns with ESSA by giving school leaders increased flexibility. Also, NCR differs from previous turnaround models by repackaging a traditionally sanction-based approach to instead motivate school leaders with increased autonomy. Using comparative interrupted time series models, we find positive NCR effects in math, but not in English Language Arts or on nontest-based student outcomes. Also, nearly a quarter of the positive NCR effect can be explained by decreased teacher and principal turnover. These results provide evidence to support current shifts toward reform models featuring local autonomy.

Trends in private caregiving cost after implementing a comprehensive nursing service covered by national health insurance: Interrupted time series

BackgroundThe Korean government has implemented a comprehensive nursing care service system (CNS) to mitigate the stress faced by caregivers. ObjectiveThis study aimed to assess trends in the estimated average costs of private caregiving and determine the difference in costs between those using CNS and those not using it. DesignA comparative interrupted time series analysis with a 2-year lag period verified total private caregiving cost trends; biannual differences in costs were evaluated based on using CNS. ParticipantsThe main unit of analysis was episode. We extracted a total of 6418 episodes of hospitalization in acute care settings that included the use of caregiving services (formal, informal caregiving and CNS). MethodsWe conducted segmented regression to assess the impact of CNS on total private caregiving costs using data from 2012 to 2018, excluding the years 2015 and 2016 of the Korean Health Panel dataset. ResultsWe presented that the immediate mean difference in total private caregiving costs between CNS users and non-users was −444.7 USD two years after the implementation of the CNS policy (95 % CI −714.5 to −174.5, p-value 0.001). Among individuals living in rural areas, two years after the implementation of the CNS policy, there was a significant immediate mean cost difference of −476.9 USD in total private caregiving costs between CNS users and non-users (p-value 0.011). Similarly, for episodes with a Charlson Comorbidity Index (CCI) score of 0 to 1, there was a substantial immediate mean cost difference in total private caregiving costs between CNS users and non-users, amounting to −399.9 USD two years after the CNS policy (p-value 0.008). ConclusionsThis study evaluated the trend of total private caregiving costs between groups using and not using CNS. After two years of being covered by CNS health insurance, those who utilized CNS paid $433 less for their total private caregiving cost over a 6-month period, compared to those who did not use CNS. The adoption of CNS may be an effective system for relieving the financial burden on inpatients in need of private caregiving services. Tweetable abstractKorean Comprehensive Nursing Service reduces private caregiving costs

Comparative Interrupted Time Series Analysis of Direct Medical Expense and Length of Stay in Elderly Patients with Femoral Neck Fractures Who Underwent Total Hip Arthroplasty and Hemiarthroplasty: A Real World Nationwide Database Study.

The objective of our study was to analyze the postoperative direct medical expenses and hospital lengths of stay (LOS) of elderly patients who had undergone either hemiarthroplasty (HA) or total hip arthroplasty (THA) for femoral neck fractures and to determine the indication of THA by comparing those variables between the 2 groups by time. In this comparative large-sample cohort study, we analyzed data from the 2011 to 2018 Korean National Health Insurance Review and Assessment Service database. The included patients were defined as elderly individuals aged 60 years or older who underwent HA or THA for a femoral neck fracture. A 1:1 risk-set matching was performed on the propensity score, using a nearest-neighbor matching algorithm with a maximum caliper of 0.01 of the hazard components. In comparative interrupted time series analysis, time series were constructed using the time unit of one-quarter before and after 3 years from time zero. For the segmented regression analysis, we utilized a generalized linear model with a gamma distribution and logarithmic link function. A total of 4,246 patients who received THA were matched and included with 4,246 control patients who underwent HA. Although there was no statistically significant difference in direct medical expense and hospital LOS for the first 6 months after surgery, direct medical expenses and hospital LOS in THA were relatively reduced compared to the HA up to 24 months after surgery (p < 0.05). In the subgroup analysis, the THA group's hospital LOS decreased significantly compared to that of the HA group during the 7 to 36 months postoperative period in the 65 ≤ age < 80 age group (p < 0.05). Direct medical expenses of the THA group significantly decreased compared to those of the HA group during the period from 7 to 24 months after surgery in the men group (p < 0.05). When performing THA in elderly patients with femoral neck fractures, the possibility of survival for at least 2 years should be considered from the perspective of medical expense and medical utilization. Additionally, in healthy and active male femoral neck fracture patients under the age of 80 years, THA may be more recommended than HA.

The Effects of Bookworms Literacy Curriculum on Student Achievement in Grades 2-5

ABSTRACT Purpose In this study, we investigated the effects of a schoolwide program, Bookworms K-5 Reading and Writing, on student achievement. Method The study included seven cohorts of students (N = 8,806) in grades 2–5 in 17 elementary schools across three school years. We used a comparative interrupted time-series design, conducting multilevel growth curve models of Measures of Academic Progress reading scores with up to 10 data points per student. By modeling each student’s growth curve, including a time by treatment interaction term, we were able to estimate the change in students’ achievement trajectories corresponding to the implementation of Bookworms. Results Results confirm a significant positive impact of Bookworms on achievement, with gains compounding over time and producing an overall standardized effect size of .26 by the end of 5th grade. Students who began third grade with relatively weaker achievement experienced more growth than those with average achievement, and those with average achievement experienced more growth than those with the highest achievement. Conclusion This study provides evidence that a comprehensive literacy curriculum that emphasizes high-volume reading of grade-level texts and the use of evidence-based instructional practices produces positive effects on student achievement for students with a range of initial reading achievement.

Impact evaluation of the turnaround arts pilot program: an interrupted time series analysis

The purpose of this study was to conduct an evaluation of the pilot program of the Turnaround Arts reform using a comparative interrupted time series design. Because the only existing evaluation of the Turnaround Arts pilot program lacks clarity and transparency, reanalyzing this program is important to understand the effects of the initiative. I created a unique dataset with nine variables to specifically evaluate the achievement outcomes of the pilot schools, which I paired with similarly matched control schools. Achievement measures included average math and reading language arts (RLA) percent proficient scores for each of the schools from 2010 to 2014 and were collected from the Stanford Education Data Archive. My results indicated that the original Turnaround Arts reform had no statistically significant impacts on the pilot schools achievement measures, an important link in the initiatives theory of action. These results support previous music education literature that found no positive causal effects between music education and non-music achievement scores. School music and arts programs are important for providing children with a well-rounded education; however, justifying the use of arts education for non-arts gains can undermine high-quality arts program development.

Changes in the Place of Death of Patients With Cancer After the Introduction of Insurance-Covered, Home-Based Hospice Care in Korea

Although hospice care has been covered by health insurance for the purpose of improving the quality of life of patients with terminal cancer as well as their caregivers, few studies have evaluated the outcomes of the policy to cover home-based hospice care services. To investigate the changes in the place of death of patients with cancer after the introduction of insurance-covered, home-based hospice care services in Korea. This cohort study used data from February 1, 2018, to December 31, 2021, from the Causes of Death Statistics database, released annually by Statistics Korea, which contains information on all deaths in the country. Individuals who died of cancer, a representative hospice-eligible disease, were assigned to the case group, and those who died of dementia, a non-hospice-eligible disease, were assigned to the control group. A total of 218 522 individuals constituted the study population. Because the Korean Health Insurance Service had begun covering home-based hospice care services on September 1, 2020, and the last follow-up date was December 31, 2021, the follow-up periods for before and after intervention were 31 months and 16 months, respectively (preintervention period: February 1, 2018, to August 31, 2020; postintervention period: September 1, 2020 to December 31, 2021). The place of death was categorized as a binary variable according to whether it was the person's own home or not. Comparative interrupted time-series models with segmented regression were applied to analyze the time trend and its change in outcomes. Of the 218 522 deaths eligible for the analysis (mean [SD] age at death, 78.6 [8.8] years; 130 435 men [59.7%]), 207 459 were due to cancer, and 11 063 were due to dementia. Immediately after the introduction of home-based hospice care, the rate of home deaths was 24.5% higher for patients with cancer than for those with dementia (estimate, 1.245 [95% CI, 1.030-1.504]; P = .02). The difference in the level change between cancer deaths and dementia deaths, on intervention, was more pronounced for those living in rural areas (estimate, 1.320 [95% CI, 1.118-1.558]; P = .001). In addition, a higher educational level was associated with a larger difference in the immediate effect size due to home-based hospice care (low educational level: estimate, 1.205 [95% CI, 1.025-1.416]; P = .02; middle educational level: estimate, 1.307 [95% CI, 0.987-1.730], P = .06; high educational level: estimate, 1.716 [95% CI, 0.932-3.159]; P = .08). In this cohort study exploring the changes in the place of death for patients with cancer after the insurance mandates for home-based hospice care in Korea, the probability of patients with cancer dying in their own homes increased after the intervention. This finding suggests the need to broaden the extent of home-based hospice care to honor the autonomy of individuals with terminal illness and improve their quality of death.

Improving Student Access to and Teacher Training for Advanced Placement

The Advanced Placement Teacher Investment Program in Indiana (AP-TIP IN) program trains teachers to prepare students for college with rigorous math, science, and English coursework using the College Board Advanced Placement Program®. The program's goals are to increase enrollment in math, science, and English Advanced Placement® courses, and increase the number of qualifying scores on AP® exams; subsequently, improved performance in these exams should lead to better postsecondary outcomes. We use comparative interrupted time series models to compare program schools and matched comparison schools and find that AP-TIP IN schools increased the proportion of students taking and passing AP exams, while post-secondary outcomes were similar to matched comparison schools.

Differences in medicaid expansion effects on buprenorphine treatment utilization by county rurality and income: A pharmacy data claims analysis from 2009–2018

BackgroundAlthough use of buprenorphine for treating opioid use disorder increased over the past decade, buprenorphine utilization remains limited in lower-income and rural areas. We examine how the Affordable Care Act Medicaid expansion influenced buprenorphine initiation rates by county income and evaluate how associations differ by county rural-urban status. MethodsThis study used nationwide 2009–2018 IQVIA retail pharmacy data and a comparative interrupted time series framework–a hybrid framework combining regression discontinuity and difference-in-difference approaches. We used piecewise linear estimation to quantify changes in buprenorphine initiation rates before and after Medicaid expansion. ResultsThe sample included observations from 376,704 county-months. We identified 5,227,340 new buprenorphine treatment episodes, with an average of 9.2 new buprenorphine episodes per month per 100,000 county residents. Among urban counties, those with the lowest median incomes experienced significantly larger increases in buprenorphine initiation rates associated with Medicaid expansion than counties with higher median incomes (5-year rates difference est=3525.3, se=1695.3, p = 0.04). However, among rural counties, there was no significant association between buprenorphine initiation rates and county median income after Medicaid expansion (5-year rates difference est=979.0, se=915.8, p = 0.29). ConclusionsMedicaid expansion was associated with a reduction in income-related buprenorphine disparities in urban counties, but not in rural counties. To achieve more equitable buprenorphine access, future policies should target low-income rural areas.

The efficacy of Facebook's vaccine misinformation policies and architecture during the COVID-19 pandemic.

Online misinformation promotes distrust in science, undermines public health, and may drive civil unrest. During the coronavirus disease 2019 pandemic, Facebook-the world's largest social media company-began to remove vaccine misinformation as a matter of policy. We evaluated the efficacy of these policies using a comparative interrupted time-series design. We found that Facebook removed some antivaccine content, but we did not observe decreases in overall engagement with antivaccine content. Provaccine content was also removed, and antivaccine content became more misinformative, more politically polarized, and more likely to be seen in users' newsfeeds. We explain these findings as a consequence of Facebook's system architecture, which provides substantial flexibility to motivated users who wish to disseminate misinformation through multiple channels. Facebook's architecture may therefore afford antivaccine content producers several means to circumvent the intent of misinformation removal policies.

Modeling the Impact of COVID-19 Interventions on Interstate Crash Rates Using Comparative Interrupted Time Series

In the early stage of the severe acute respiratory syndrome (SARS)-CoV-2 coronavirus (COVID-19) outbreak in March 2020, a series of restrictions were released to minimize the exposure and spread of the communicable virus. These restrictions had significant impacts on the mobility and safety of transportation networks. The objective of this study is to assess the causal effects of the COVID-19 restrictions on traffic crashes on interstate highways. Specifically, compared with previous years, the study is to understand whether the COVID-19 outbreak has increased the rate of traffic crashes on Interstate 80 in Nebraska. A comparative interrupted time series (CITS) model was established to examine the week-to-week differences in traffic crashes before, during, and after the implementation of the COVID restrictive measures in Nebraska. Traffic crash data and traffic count data on Interstate 80 for 33 weeks, from Day 1 to Day 183, of 2020 were used. Meanwhile, the same periods of data in 2018 and 2019 were also included as a control group in the models. Results show that the COVID-19 restrictive measures were associated with significant, immediate, positive effects on crashes (i.e., crash reduction or mitigation) with no or minor injuries, crashes with single or two vehicle(s), and crashes with one injury involved. The findings contribute to the understanding of the impact of COVID-19 interventions on interstate highway safety when future outbreaks of COVID-19 or other pandemics become possible.

Association of Arizona's Implementation of a Behavioral Health Crisis Response System With Suicide Hospitalizations.

In July 2022, the 988 Suicide and Crisis Lifeline went live. The Lifeline is part of larger federal and state efforts to build comprehensive behavioral health crisis response systems that include mobile crisis units and crisis diversion and stabilization centers. Comprehensive response systems are anticipated to reduce hospitalizations for suicide and other behavioral health crises; however, research testing this assumption has been limited. The authors used Arizona-a state known for its comprehensive crisis system-to determine the association between state implementation of a comprehensive behavioral health crisis response system and suicide-related hospitalizations. A comparative interrupted time-series (CITS) design was used to compare changes in suicide-related hospitalizations after the 2015 implementation of Arizona's crisis response system (N=215,063). Data were from the 2010-2019 Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases (SID). Nevada (N=84,091 hospitalizations) was used as a comparison state because it is a western state that had not yet implemented a comprehensive crisis system and had available HCUP SID data. The CITS model included controls for time-varying differences in state demographic composition. From 2010 to 2014 to 2019, annual suicide-related hospitalizations in Arizona increased from 122.0 to 324.2 to 584.5, respectively, per 100,000 people, and in Nevada, hospitalizations increased from 94.7 to 263.2 to 595.5, respectively, per 100,000 people. Arizona's crisis response system was associated with a significant relative decrease in the quarterly trend of 2.57 suicide-related hospitalizations per 100,000 people (p=0.033). More research is needed to understand how the implementation of a comprehensive crisis response system may affect suicide-related hospitalizations.

North Carolina's Medicaid Transformation: the Early Enrollee Experience.

On July 1, 2021, North Carolina's Medicaid Transformation mandatorily switched 1.6 million Medicaid beneficiaries from fee-for-service to managed care plans. We examined the early enrollee experience in terms of engagement in plan selection, provider continuity, use of primary care visits, and assistance with social needs. Using electronic health records (EHR) covering pre- and post-transition periods (1/1/2019-5/31/2022) from the largest provider network in western North Carolina, we identified all children and adults under age 65 with continuous Medicaid or private coverage. We conducted primary surveys of a random sample of Medicaid-covered enrollees and obtained self-reported rates of engagement in plan selection, continuity of provider access, and receipt of social need assistance. We used comparative interrupted time series models to estimate the relative change in primary care visits associated with the transition. Our EHR-based study cohorts included 4859 Medicaid and 5137 privately insured enrollees, with 398 Medicaid enrollees in the primary surveys. We found that 77.3% of survey participants reported that the managed care plan they were on was not chosen but automatically assigned to them, 13.1% reported insufficient information about the transition, and 19.2% reported lacking assistance with plan choice. We found that 5.9% were assigned to a different primary care provider. Over 29% reported not receiving any additional social need assistance. The transition was associated with a 7.1% reduction (95% CI, -11.5 to -2.7%) in the volume of primary care visits among Medicaid enrollees relative to privately insured enrollees. Medicaid enrollees in North Carolina may have had limited awareness and engagement in the transition process and experienced a reduction in primary care visits. As the state's transition process gains a foothold, future policy needs to improve enrollee engagement and develop evidence on healthcare utilization and patient outcomes.