Results Summary
What was the research about?
Children with disabilities have complex health needs that require care from many different doctors, nurses, and healthcare providers. These providers may be in different offices and locations. As a result, coordinating care for these patients can be challenging.
Accountable care organizations, or ACOs, are groups of healthcare providers, such as doctors and hospitals, who work together to provide patient care. In this study, the research team compared the use of healthcare services by children with disabilities who did and didn’t take part in an ACO.
What were the results?
Compared with children not in an ACO, children in an ACO had more
- Adolescent well-child visits
- Follow-up doctor visits after starting medicines for attention-deficit/hyperactivity disorder, or ADHD
Compared with children not in an ACO, children in an ACO had the same use of
- Primary care doctor visits
- Hospital visits
- Emergency room, or ER, visits
- Primary care and hospital follow-up visits after being in the hospital
- Antidepressant, anticonvulsant, antianxiety, antipsychotic, and asthma medicines
- Office-based physical, occupational, and speech language therapies
- Home-based physical, occupational, and speech language therapies
Compared with children not in an ACO, children in an ACO had less use of
- ADHD medicines
- Specialty mental health services
- Home health services
What did the research team do?
The research team looked at Ohio Medicaid claims records from 2011 to 2016, for 64,382 children with a range of disabilities. These children had health insurance through Aged, Blind, and Disability Medicaid. The team compared children’s use of healthcare services for up to three years among children who took part in an ACO and those who didn’t. This analysis was part of a larger study that looked at the experiences of children with disabilities and their families with care coordination before and after taking part in an ACO.
Among children in an ACO, 65 percent were white, and 29 percent were African American. The average age was 12, and 66 percent were boys. For children not in an ACO, 49 percent were white, and 46 percent were African American. The average age was 13, and 65 percent were boys.
Families of children with disabilities, health organizations, and doctors gave input on the study.
What were the limits of the study?
The three-year follow-up may not capture the long-term impact of being in an ACO. The study looked only at services covered by Medicaid. Children may have used other health services not covered by Medicaid.
Future research could look at healthcare use for children with disabilities in ACOs located in other regions and states. Research could also explore the reasons for some of these differences, like why children in an ACO used fewer home health services than those who weren’t.
How can people use the results?
States and policy makers can use these results to learn more about the impact of ACO programs on children with disabilities.
Professional Abstract
Objective
To assess healthcare use among children with disabilities enrolled in an accountable care organization (ACO) compared with children enrolled in a non-ACO plan
Study Design
| Design Elements | Description |
|---|---|
| Design | Observational: case-control study |
| Population | Medicaid claims data for 64,382 children aged 18 and younger with disabilities and who were ABD eligible enrolled in either an ACO or a non-ACO managed care plan in Ohio |
| Interventions/ Comparators |
|
| Outcomes | Healthcare use as measured by primary care visits, ED use, inpatient hospitalization, mental health services, home health services, medication use and adherence |
| Timeframe | Up to 3-year follow-up for study outcomes |
This pre- and post-enrollment case versus control study used Ohio’s Medicaid claims data for children with disabilities to compare healthcare use by enrollees in an ACO versus those in non-ACO managed care plans, before and after enrolling in an ACO.
ACO enrollees included 17,356 children ages 18 and younger who were eligible for Medicaid’s aged, blind, and disabled (ABD) category and automatically enrolled in an ACO. Of these, 65% were white, 29% were African American, and 6% were other races. The average age was 12, and 66% were male.
A patient advisory panel, including caregivers and representatives from statewide advocacy and health organizations, helped select study outcomes.
Results
Compared with non-ACO enrollees,
- Primary care services. ACO enrollees age 12 and older were more likely to have well-child visits over time (beta=0.05; 95% confidence interval [CI]: 0.03, 0.07). Researchers found no significant differences in other primary care services.
- ED services. ACO enrollees did not differ significantly in emergency department use over time.
- Inpatient services. ACO enrollees did not differ significantly in inpatient hospitalization, readmissions, or follow-up over time.
- Mental health services. ACO enrollees showed a decrease in specialty mental health procedures (beta=-0.02; 95% CI: -0.03, -0.01) but no significant change in other mental health services.
- Home health services. ACO enrollees were less likely to access home health services (beta=-0.05; 95% CI: -0.05, -0.04). ACO enrollees did not differ significantly in the use of office-based or home-based physical, occupational, and speech language therapy services.
- Medication use and adherence. ACO enrollees had a decreased likelihood of attention-deficit/hyperactivity disorder (ADHD) medication use (beta=-0.01; 95% CI: -0.02, -0.00) and showed a greater probability of follow-up after starting ADHD medication (beta=0.07; 95% CI: 0.04, 0.10). Antidepressant, anticonvulsant, antianxiety, antipsychotic, and asthma medication use was not significantly different.
Limitations
Three years of follow-up precludes assessments of healthcare utilization associated with long-term ACO program enrollment. The claims data used do not reflect potential use of therapies and services not covered by Medicaid. Findings are based on Medicaid claims data from one state, which may limit generalizability to other states.
Conclusions and Relevance
ACO enrollees had increased use of adolescent well-child services. Use of ADHD medications as well as home health services declined for children in the ACO, while use of a number of other services was no different.
Future Research Needs
Future research on the effect of ACO enrollment could focus on enrollees from more than one state and for longer duration. Researchers could examine reasons for differences in service use, such as lower rates of home health service use, for children with disabilities enrolled versus not enrolled in ACOs.
Final Research Report
View this project's final research report.
Journal Citations
Results of This Project
Related Journal Citations
Peer-Review Summary
Peer review of PCORI-funded research helps make sure the report presents complete, balanced, and useful information about the research. It also assesses how the project addressed PCORI’s Methodology Standards. During peer review, experts read a draft report of the research and provide comments about the report. These experts may include a scientist focused on the research topic, a specialist in research methods, a patient or caregiver, and a healthcare professional. These reviewers cannot have conflicts of interest with the study.
The peer reviewers point out where the draft report may need revision. For example, they may suggest ways to improve descriptions of the conduct of the study or to clarify the connection between results and conclusions. Sometimes, awardees revise their draft reports twice or more to address all of the reviewers’ comments.
Peer reviewers commented, and the researchers made changes or provided responses. The comments and responses included the following:
- Reviewers wondered how the analyses took into account the transition period from fee-for-service into managed-care plans, since the effects of the transitions were likely to change during the three years that the researchers observed. The researchers said they did not expect the transition to managed care to be disruptive since the provider networks stayed the same. Even so, the researchers ran new analyses that compared the three years after transition separately to see if there were differences in patterns of healthcare use. These analyses suggested differences in service use for three variables between the first year, and the second and third years after the transition. The researchers described this in their report.
- Reviewers noted that for aim 3, the report presented 23 adjusted comparisons, even though the multiple comparisons are more likely to introduce false-positive findings. The reviewers suggested focusing on the magnitude of differences rather than statistical significance and asked the researchers to define what they would consider clinically important effect sizes. The researchers said that there are no established metrics for clinical significance in the use of services as there are for clinical outcomes. They did agree that it would be more clinically relevant if the presentation of differences in service use was in proportion to baseline service use.
- Reviewers suggested that measuring service in a binary way—whether patients used services or not, rather than frequency of use—would limit the study’s ability to measure the differences between groups in the frequency of service use. The researchers disagreed that this should be considered a limitation. They said that their data do allow them to count service use episodes. However, in general they considered any use of the service to be a more important outcome, since any use indicates that patients had access to the service. The researchers declined to make any changes to the report on this score.