Children's use of mental health services in different ... - Springer Link

Brief Reports

Children's Use of Mental Health Services in Different Medicaid Insurance Plans David S. Mandell, ScD Roger A. Boothroyd, PhD Paul G. Stiles, JD, PhD Abstract This study examined the effect of different Medicaid insurance plans on children's mental health service use through survey, claims, and encounter data collected between February 1998 and February 1999. Participants were assigned to I of 3 insurance plans:fee-for-service, a Health Maintenance Organization and a prepaid carve-out. Logistic and stratified logistic regression were used to examine the effect of plan on service utilization, adjusting for caregiver report of need for services and psychosocial functioning. There was no difference in service use by plan controlling for demographic characteristics; however, when psychopathology and caregiver report of need for services were included in the model, the odds of using services in the Health Maintenance Organization was half of and the odds in the carve-out 29% less than the odds of using services in fee-for-service. Characteristics of the interaction between need, psychopathology, and insurance plan that may be associated with the reduction in service use are discussed. An increasing number of children are enrolled in managed care plans, especially among the Medicaid population. The proportion of Medicaid-eligible children in managed care increased from 9.5% in 1991 to 47.8% in 1997.1 A major concern that has been expressed about this change is that managed care arrangements will limit children's access to needed medical services, 2 especially among children with special health care needs 3,4 and poor children. 5 Evidence about the effect o f prepaid plans, carve-outs, and managed care techniques on access and utilization of mental health services is mixed. A number of studies have found no differences in access between Health Maintenance Organization (HMO) and fee-for-service (FFS) plans. For example, in a recent study of low-income children, access among those enrolled in Medicaid FFS and those in private H M O s was comparable. 6 Another study with random assignment to different insurance plans found no differences in checkup and emergency room visits, but did not examine mental

Address Correspondenceto David S. Mandetl, ScD, Assistant Professor,Center for Mental Health Policy and Services Research, Universityof Pennsylvania,3535 Market Street, 3rd Floor, Philadelphia, PA 19104-3309. Telephone (215) 662-2504, e-mall: [email protected]. Roger A. Boothroyd, PhD, is Associate Professor in the Department of Mental Health Law & Policy in Louis de la Parte Florida Mental Health Institute at University of South Florida, Tampa, Florida. Paul G. Stiles, JD, PhD, is Assistant Professor and Co-Director at Policy and Services Research Data Center in Department of Mental Health Law & Policy, Louis de la Parte Florida Mental Health Institute at University of South Florida, Tampa Florida.

Journal of Behavioral Health Services & Research, 2003, 30(2), 228-237. (~) 2003 National Council for Community Behavioral Healthcare.

228

The Journal of Behavioral Health Services & Research

30:2 April~June2003

health service utilization. 7 Although their study did not include children, Norquist and Wells, 8 using data from the Epidemiologic Catchment Area Study, reported that among adults with a psychiatric disorder, there was no difference in access to specialty mental health care between those enrolled in HMOs and those enrolled in FFS, findings similar to those of Sturm and colleagues. 9 Manning and colleagues 1° report similar results, although they did find that costs among those who did use mental health services were higher in the FFS plan than in the HMO. There is also some evidence to suggest that managed care arrangements decrease costs and increased the probability of using outpatient rather than inpatient services. 11-13 In 3 studies with prepost designs, the implementation of mental health managed care carve-outs appeared to reduce costs by reducing inpatient stays, the average number of outpatient visits, and per service reimbursements. These studies stand in contrast to others suggesting that youth with serious emotional disorders are having more difficulty obtaining necessary services under managed c a r e . 14,15 Parents of children in HMOs are more likely to report an unmet health need and difficulty in scheduling an appointment16; caregivers are also more likely to report worse access to and less satisfaction with services. 17 Hurley and colleagues 18found that children with special health care needs experience a 30% to 50% reduction in specialty service utilization under managed care, although the authors do not discuss the effect of this decreased utilization on outcomes. Dickey and colleagues, 11 who reported that the managed care carve-out in Massachusetts increased the number of children using mental health services, also report that continuity of care for children with disabilities appeared to be reduced under managed care.

A few studies have focused on the experiences of individuals with specific disorders, rather than general enrolled population experiences. Sturm and colleagues 19 found that people with depression used less services in prepaid versus FFS plans, adjusting for illness severity. The authors, however, were unable to determine if the difference in service intensity affected outcomes. A pre-post study of individuals with schizophrenia found that those in a mental health carve-out experienced worse outcomes than those in a FFS plan, and that the difference was greatest for those whose condition was most severe. 2° At least one other study has focused on the effects of a specific mechanism for managing care. Liu and colleagues 21 found that preauthorization requirements have a powerful effect on the volume of mental health services utilized. On average, individuals in plans that preauthorize units in smaller increments use less services. Part of the reason for contradictory findings about the effects of managed care arrangements on mental health service utilization may have to do with the populations under study. Studies that focused on the general population of eligible individuals often found no differences or found that the probability of use was increased in the managed care plan. Studies that focused on populations with special needs often found that those in FFS plans had greater service utilization and satisfaction, and experienced better outcomes. A related reason for the different findings may relate to the data available for research purposes. Often, the data available are health care claims, which provide accurate estimates of service use, but do not give any indication of pathology or perceived need relative to those who remain untreated. Other studies have relied on self-reports of service use. 22,23 While this type of study can provide more accurate assessments of the presence of health conditions and the need for treatment, the accuracy of parents' reports of children's mental health service utilization is less clear, 24 and certainly different than children's self-report. 25 The state of Florida has been engaging in an evaluation of the relationship among risk arrangements, benefit structure, and enrollees' access to and satisfaction with care. This evaluation is made possible by recent changes in Florida's health and mental health care financing systems. Under a waiver from the Health Care Financing Administration, the Florida Medicaid Authority implemented a Medicaid prepaid Mental Health Plan (PMHP) in the Tampa Bay area in March 1996. In August 1996, HMOs in the Tampa Bay area were required to provide the same comprehensive array of mental health services

Effect of Insurance Plans on Children's Mental Health Service Use

MANDELL et al.

229

to their Temporary Assistance for Needy Families (TANF) and supplemental security income (SSI) enrollees that are provided by the PMHP. In other parts of Florida, comprehensive mental health services are reimbursed through a FFS mechanism. The details of these different financing and management mechanisms are discussed in more detail in the "Methods" section of this paper, and also by Ridgely and colleagues 26 elsewhere. Data collected as part of this evaluation offer a unique opportunity to examine the effects of managed care on children's utilization of specialty mental health services. Because this evaluation included the collection of identified survey data that could be concatenated with Medicaid claims and encounter data, it was possible to link service records with self-reported pathology, and need for and use of services. The purpose of this study is to compare the utilization of mental health services among children and youth enrolled in an HMO, a mental health prepaid carve-out, and a FFS plan, and to examine the relationship between perceived need, psychopathology, and service use in the different plans.

Methods Data collection Between February 1998 and February 1999, surveys were mailed to the caregivers of 4781 randomly selected Medicaid-enrolled children (ages 5-21) living in Tampa and Jacksonville, Fla. Mail survey techniques have been used with varying results in differing settings. This evaluation used a highly systematic and structured approach to survey design and follow-up similar to those recommended by Dillman 27 and Salant and Dillman. 28 In total, 5 separate mailings were conducted. The first mailing consisted of a prenotification postcard informing sampled Medicaid enrollees that a study was being conducted to examine their health care services and that they would receive a questionnaire in the mail in about a week. One week later a second mailing was conducted, which included a personalized cover letter and questionnaire in both English and Spanish, an explanation of the purpose of the study, a statement that respondents would be paid $7.00 for returning a completed questionnaire, and information about the days and hours of operation of the toll-free telephone number. A preaddressed stamped return envelope was also included in the mailing. One week later, a postcard reminder was sent to each person who had not yet responded. This reminder emphasized the importance of the study and again included information on the toll-free telephone number they could call. Two weeks after the postcard reminder was mailed, a fourth mailing containing a cover letter, questionnaire, and return envelope was mailed to each nonrespondent. Finally, 4 weeks later, a fifth mailing was sent via certified mail to individuals who still had not responded. As with the first and fourth mailing, enrollees received a personalized cover letter, questionnaire, and a preaddressed, stamped return envelope. These mailing procedures were based on the findings of a feasibility study conducted to assess the validity of using mail survey procedures with a Medicaid population. The findings from this feasibility study are summarized in Boothroyd and Shern. 29 The collected survey data were then merged with all Medicaid mental health claims (from the FFS plan) and encounters (from the prepaid plans) of survey recipients for 1 year previous to and 1 year after the date of the survey return.

Sample Data were collected on 1797 children. The response rate for the survey was 43% after adjusting for wrong addresses. The average age of children on whom data were obtained was slightly less than of those whose caregivers did not return the survey (means = 12.21 and 12.49 years respectively, P = .015). Differences in gender were not statistically significant. Significant racial differences were found between the groups (X 2 = 47.44, df = 3, P < .001). Survey respondents were more likely to be white (39.9% vs 33.3%) and less likely to be African American (38.2% vs 46.1%). Finally,

230


30:2 April~June2003

children receiving SSI were more likely to be survey respondents (53.7% vs 47.8%) and children receiving TANF were less likely to respond (46.3% vs 52.2%) (X 2 = 15.81, d f = 1, P = .001).

Measures

Demographic characteristics Demographic characteristics were identified from the Medicaid eligibility files, and included age, gender, and race.

Presence o f significant psychosocial problems Presence of significant psychosocial problems was measured by caregivers' responses to questions on the Pediatric Symptom Checklist (PSC). 3° This 35-item questionnaire measuring psychosocial dysfunction has been validated on poor, minority, and urban samples such as the one in this study. 31 Using the clinical cutoff score of 28 or higher to indicate dysfunction, comparison of the PSC with clinician ratings on the Child Global Assessment Scale yielded a kappa of 0.82 with a sensitivity of 88% and a specificity of 100%. For the purposes of this study, children were identified as having a significant psychosocial problem if their score on the PSC was 28 or above.

Insurance plan Respondents in this study were enrolled in 1 of 3 Medicaid health care conditions: the Medipass/ prepaid Mental Health Plan (mental health carve-out), a Health Maintenance Organization (HMO), or the Medipass (FFS) program. These 3 plans are described below. Carve-out. In the carve-out, a monthly, prospective payment is made to a specialty behavioral health organization to provide a comprehensive array of mental health services to TANF and SSI Medicaid enrollees. The carve-out is fully at risk for mental health services. It involves a partnership of 5 community mental health centers that serve the Tampa Bay area and a private, for-profit behavioral health care organization. 26 Carve-out enrollees receive their general health care through MediPass, a case management program in which primary care physicians are paid a nominal monthly fee to coordinate and authorize various components of care. While primary care providers manage care for both mental and general health, there is no gatekeeping system. Enrollees access mental health services directly. 26 In the MediPass program, general health services are paid on a FFS basis with no risk to providers. HMO. The HMOs receive an augmented premium to assume risk for the cost of comprehensive mental health services in addition to the general health services for which they were already at risk. All of the HMOs purchase their mental health services from specialty behavioral health care organizations (BHOs) that are also at risk for the provision of services. These BHOs contract with vendors who provide services on a FFS basis. In general, the BHOs engage in more management utilization strategies such as preauthorization and service limits. 26 Some HMOs also offer an enhanced dental service benefit to their members, which is not available for individuals who are enrolled in the MediPass program. All TANF and SSI Medicaid recipients are required to join either MediPass/PMHP or an HMO. Interestingly, the BHOs contract with the same 5 community mental health centers that are partnered in the carve-out. Fee for service. In other parts of Florida, comprehensive mental health services are reimbursed through a FFS mechanism with little utilization management oversight with the exception of inpatient services. All Medicaid enrollees are eligible to participate in this program for their specialty mental health care. Either HMOs or MediPass may be used for general health services outside of Tampa Bay. The major features of the 3 financing arrangements are summarized in Table 1.


MANDELL et al.

231

Table 1 Summary of risk, payment and benefit structures in the 3 financing conditions

Financing condition Tampa Bay Service type Medical Mental health

Dental Substance abuse

(MediPass/PMHP) No risk, FFS At risk, capitated, enrollees access services directly No risk, minimal benefit No risk, FFS

Tampa Bay (HMOs) At risk, capitated At risk, capitated, all services must be preauthorized by HMO At risk, enhanced benefit No risk, FFS

Jacksonville (MediPass) No risk, FFS No risk, FFS enrollees access services directly No risk, minimal benefit No risk, FFS

Note. FFS: fee-for-service.

Need f o r mental health services Need for mental health services was measured by caregivers' response to the question, "for each of the services listed below, please indicate if your child needed that service in the past year." The service categories to which caregivers could respond were medical services, dental services, eye or vision services, mental health services, alcohol or drug use services, special services in school, and juvenile justice services. Children were identified as needing services if caregivers responded "yes" to mental health services.

Use of mental health services Originally, use of mental health services was to be measured by the presence of a mental health claim or encounter record; however, validity checks provided evidence that claims data from the FFS plan were more likely than encounter data from the prepaid plans to be an accurate representation of service use, probably due to financial incentives for submitting claims. 32 A contingency table of claims/encounters versus self-report provided evidence that in 18% of cases, the administrative data indicated that services had been used, even though the caregiver report indicated that no services had been used or caregiver report was missing (data not shown). To address these issues of misreporting and to increase statistical power, a dichotomous variable was created indicating whether a child had any claims for mental health service use in the year prior to the survey or the caregiver reported that their child had received mental health services in the last year. This combination of claims data and caregiver report was used to indicate utilization of mental health services. Tests using general linear models indicated that the relative difference in service use among the insurance plans was not significantly changed by combining self-report and claims data (data not shown), although the overall rates of service use increased in all 3 plans.

Analyses Demographic characteristics, use of services, perceived need, and psychopathology were compared among children in the 3 different insurance plans using one-way ANOVAs.

232


30:2 April~June 2003

Three logistic regression models were conducted. The first was used to determine the effect of HMO and/or carve-out enrollment on mental health service utilization compared to FFS after controlling for demographic characteristics. In the second model, caregiver report of need and psychosocial dysfunction were added. In the third model, the sample was stratified into 1 of 4 conditions (no need, low PSC score; yes need, low PSC score; no need, high PSC score; yes need, high PSC score) to determine the effects of insurance plan on service use in each group.

Results There were 378 (21%) children who had at least one claim for mental health services in the year prior to survey return, and an additional 242 (13%) whose caregiver reported that their child had received mental health services in the past year but that child had no service claims (data not shown). Conversely, there were 413 (23 %) caregivers who reported that their child had received mental health services, and an additional 207 (l 1%) who had a claim indicating mental health service use but no caregiver report of service utilization. Therefore, a total of 620 (34.5%) children were identified as having received mental health services. Table 2 provides a comparison of children in the 3 groups. Of the 1797 children in the sample, 610 were enrolled in the FFS group, 651 were enrolled in the prepaid carve-out, and 536 were enrolled in the HMO. One-way ANOVAs revealed no statistically significant differences among the groups on use of mental health services, gender, age, or caregiver report of need for services. There were significant differences among plans by race. Those in the FFS plan were more likely to be white, those in the carve-out were slightly more likely to be Hispanic, and those in the HMO were more likely to be African American. Finally, children in the HMO were more likely to have PSC scores above the clinical cutoff, indicating a higher degree of psychopathology. Plan enrollment and use of mental health services

Table 3 provides the results of 2 logistic regression models predicting the odds of mental health service use. In model 1, service use was regressed onto demographic characteristics and plan membership. The results of this model suggest that there are no statistically significant differences in service use by insurance plan, although African American and Latino children are less likely than white children to use mental health services. In model 2, psychosocial dysfunction, need for service,

Table 2 Comparison of children in the 3 insurance plans

Average age (in years) Male White Black Hispanic Other Above PSC cutoff Caregiver report need for mental health services Percentage who used services

FFS

Carve-out

HMO

P value

12.3 62% 48% 34% 3% 15% 45% 26%

12.2 55% 45% 30% 8% 17% 42% 24%

12.1 62% 40% 40% 7% 13% 52% 29%

.485 .060 .010 .002 .005 .005 .002 .141

37%

33%

33%

.259

Note. FFS: fee-for-service, HMO: Health Maintenance Oraganization.


MANDELL et aL

233

Table 3 Logistic regression predicting odds of mental health service use*

Model 1: Plan and demographics

HMO Mental health carve-out Caregiver reported need Above PSC cutoff score Interaction of need and PSC score Age (in years) Gender (male) African American Hispanic

Model 2: Model 1 plus perceived need and psychopathology

OR

95 % CI

OR

95 % CI

0.87 0.91 * * * 1.00 1.84 0.65 0.27

0.69-1.14 0.72-1.15 * * * 0.98-1.03 1.49-2.26 0.53-0.81 0.12-0.41

0.50 0.71 35.06 1.74 0.54 0.96 1.36 0.59 0.42

0.34-0.73 0.50-0.99 19.99-61.48 1.18-2.58 0.27-1.05 0.92-0.99 1.00-1.85 0.43-0.81 0.18-0.94

Note. HMO: Health Maintenance organization, PSC: Pediatric Sympton Checklist.

*Reference group consists of 6-year-old white female children in the fee-for-service plan.

and the interaction between these 2 variables were added to the model. The results of this model suggests that, controlling for need and psychopathology, the odds of using services in the HMO are reduced by 50% and the odds of using services in the carve-out reduced by 29% compared with children in FFS. Other factors were also significant in predicting service use. Caregiver report of need for services increased the odds of receiving services 35 times. Presence of psychosocial dysfunction increased the odds of receiving mental health services by 74%. Being male increased the odds of receiving services by 36%, while being African American reduced the odds of receiving services by 41% and being Latino reduced the odds of receiving services by 58% compared with the odds of receiving services by whites. Figure 1 provides the results of the stratified logistic regression. Adjusting for demographic characteristics, there was no difference in service utilization among children whose caregivers reported low psychopathology, regardless of perceived need. Among children whose parents reported that their children had high psychopathology, the odds of using services were 58% lower among children enrolled in the HMO compared with those in FFS. There were no differences between those in the carve-out and those in FFS in this stratified analysis.

Discussion This study found that, adjusting for demographic characteristics, there were no differences in mental health service utilization among children in the FFS carve-out, or HMO. This finding is similar to those reported in a number of other studies, 7-1°,16 although it is by no means a universal finding. Adjusting for need and psychosocial dysfunction reduced the odds that children in the HMO would use services by 50% and in the carve-out by 29% compared with the odds of service use of those in FFS. Examination of the distribution of service utilization, stratified by caregivers' report of need and psychosocial dysfunction, suggests that the difference in service use among plans is particularly pronounced among children whose PSC scores indicate a high level of psychosocial dysfunction. This finding is similar to those reported by a number of studies that focused on specific psychiatric or special needs populations. 18-20

234



Figure 1 Adjusted odds of using mental health services stratified by need and pychopathology* 10-

NS p = 0.03

p = 0.Ol

NS NS

• Carve Out I-IHMO ] NS

NS

NS

No need low PSC n = 1029

Yes need low PSC n = 96

No need high PSC n = 300

Yes need high PSC n = 372

0.1 *Adjusted for age, gender and ethnicity. Reference group consists of those in FFS

Limitations A number of limitations of this study should be acknowledged. The primary limitation is the differential validity of claims and encounter data for the different insurance plans, and the dependence on self-report to correct for this difference. If caregivers uniformly underreport their children's use of mental health services, then the results would be skewed to suggest that those in the FFS plan receive more services than those in the prepaid plans. However, while there is evidence suggesting a lack of reliability of parental reporting, there is no evidence suggesting underreporting. 24,25 In addition, the lack of difference in the unadjusted rates of service use among the 3 plans suggests that the combination of parental report and claims data to identify service utilization did not bias the sample. Another limitation is that since all FFS enrollees lived in Jacksonville, while all carve-out and HMO enrollees lived in Tampa, there was no way to control for potential geographic differences. It may be that practice variations in these 2 places accounted for some of the difference in service use, independent of payment plan. A third limitation is that caregivers may have misinterpreted the question "did your child need services" to mean "did your child use services," which would have artificially strengthened the observed association between need and utilization in this study; however, there is no reason to think that this would be different by insurance plan. Finally, because this study used data from 1 managed care evaluation in 1 state, the generalizability of these findings is unclear.

Effect of lnsurance Plans on Children's Mental Health Service Use

MANDELL et al.

235

Implications for Behavioral Health Services Despite these limitations, there are important implications related to the findings of this study. Major strengths of this study are the measures of caregiver-reported need and psychosocial dysfunction as measured by the PSC. The presence of these measures allowed us to assess the independent effect of payment plan on children's use of mental health services. To our knowledge, no other study has addressed this issue in mental health. These results suggest that children enrolled in HMOs are less likely than children in FFS to receive mental health treatment when the need for treatment--as measured either by caregivers' perceptions or measures of psychosocial dysfunction is present. Perhaps more interestingly, the stratified analyses provide evidence that the effect of payment plans on service use is to a certain extent concentrated among those children whose PSC scores indicate psychosocial dysfunction. It is also important to note that African American and Hispanic children were much less likely to receive mental health treatment, controlling for perceived need and psychosocial dysfunction. These racial/ethnic differences have been found in other research, 33'34 but bear further mention here because of the addition of caregivers' report of need and psychosocial dysfunction to the model. In interpreting the results of this study, it is important to consider the different financing and management mechanisms used by the 3 plans. As discussed earlier, in the FFS plan, providers bear no risk and enrollees have direct access to mental health service providers. In the carve-out, the managed care organization bears full risk for mental health services. The primary care provider is the gatekeeper for services and the community mental health centers manage mental health services. In the HMOs, the subcontracted BHOs acted as the gatekeepers for all mental health services, require preauthorization for services, and have service limits. Although the HMOs pay a capitated rate to the BHOs, the BHOs usually reimburse providers on a FFS basis. The differences between the FFS and HMO plans suggest that it is the utilization management techniques rather than the payment mechanism that is responsible for the differences in service utilization. This result is consistent with the finding of Liu and colleagues 21 regarding the effect of preauthorization on service use. Given that this difference appears to be concentrated among children with high psychopathology, it would be difficult to argue that this mechanism is reducing inappropriate utilization; rather, it appears that children who require services are not getting them. The differences in service utilization between the FFS and carve-out plans are smaller, as one might expect given that their financing and management mechanisms are similar. It may be, however, that having a primary care provider act as a gatekeeper does reduce utilization. It may be that this reduction is related to a reduction in inappropriate utilization, in that primary care providers may be addressing mental health concerns in general pediatric practice. Further studies are needed to determine if the reduction in service use is appropriate.

References 1. McCormick M, Kass B, Elixhauser A, et al. Annual report on access to and utilization of health care for children and youth in the United States--1999. Pediatrics. 2000;105(1):219-230. 2. Szilagyi P. Managed care for children: effect on access to care and utilization of health services. Future of Children. 1998;8(2):39-59. 3. Fox H, Wicks L, Newacheck P. Health maintenance organizations and children with special needs. A suitable match. American Journal of Diseases of Children. 1993;147:546-552. 4. Horowitz S, Stein R. Health maintenance organizations versus indemnity insurance for children with chronic illnesses. American Journal of Diseases of Children. 1990;144:581-586. 5. Szilagyi P, Schor E. The health of children. Health Services Research. 1998;33(4): 1001-1039. 6. Dubay L, Kenney G. Health care access and use among low-income children: who fairs best. Health Affairs. 2001;20(1):112-121. 7. Mauldon J, Leibowitz A, Buchanan J, et al. Rationing or rationalizing children's medical care: comparison of a Medicaid HMO with fee-for-service care. American Journal of Public Health. 1994;84(6):899-904. 8. Norquist G, Wells K. How do HMOs reduce outpatient mental health care costs. American Journal of Psychiatry. 1991;148(1):96-101. 9. Sturm R. How does risk sharing between employers and a managed behavioral health organization affect mental health care. Health Services Research. 2000;35(4):761-776.

236



10. Manning W, Wells K, Benjamin B. Use of outpatient mental health services over time in a health maintenance organization and fee-forservice plans. American Journal of Psychiatry. 1987;144(3):283-287. 11. Dickey B, Normand S, Norton E, et al. Managed care and children's behavioral health services in Massachusetts. Psychiatric Services. 2001 ;52(2): 183-188. 12. Goldman W, McCulloch J, Stnrm R. Costs and use of mental health services before and after managed care. Health Affairs. 1998; 17(2):4052. 13. Grazier K, Eselius L, Hu T, et al. Effects of a mental health carve-out on use, costs and payers: a four-year study. Journal of Behavioral Health Services & Research. 1999;26(4):381-389. 14. Stroul B, Pires S, Armstrong M, et al. The impact of managed care on mental health services for children and their families. The Future of Children. 1998;8(2):119-133. 15. Pires S, Stroul B, Armstrong M. Health Care Reform Tracking Project: Tracking State Managed Care Reforms as They Affect Children and Adolescents With Behavioral Health Disorders and Their Families. Tampa, Fla: Louis de la Parte Mental Health Institute, University of South Florida; 2000. 16. Mark T, Mueller C. Access to care in HMOs and traditional insurance plans. Health Affairs. 1996;15(4):81-87. 17. Pina D. Medicaid beneficiaries' experiences in HMO and fee-for-service health care. Journal of Health Care for the Poor and Underserved. 1998;9(4):433-448. 18. Hurley R, Freund D, Paul J. Managed Care in Medicaid." Lessons for Policy and Program Design. Ann Arbor, Mich: Health Administration Press; 1993. 19. Sturm R, Jackson C, Meredith L, et al. Mental health care utilization in prepaid and fee-for-service plans among depressed patients in the Medical Outcomes Study. Health Services Research. 1995;30(2):319-340. 20. Manning W, Liu C, Stoner T, et al. Outcomes for Medicaid beneficiaries with schizophrenia under a prepaid mental health carve-out. Journal of Behavioral Health Services & Research. 1999;26(4):442-450. 21. Liu X, Sturm R, Cuffel B. The impact of prior authorization on outpatient utilization. Medical Care Research and Review. 2000;57(2): 182195. 22. Wu P, Hoven C, Bird H, et al. Depressive and disruptive disorders and mental health service utilization in children and adolescents. Journal of the American Academy of Chill and Adolescent Psychiatry. 1999;38(9): 1081-1090. 23. Verhulst F, van der Ende J. Factors associated with child mental health service use in the community. Journal of the American Academy of Chill and Adolescent Psychiatry. 1997;36(7):901-909. 24. Fendrich M, Johnson T, Wislar J, et al. Accuracy of parent mental health service reporting: results from a reverse record-check study. Journal of the American Academy of Chill and Adolescent Psychiatry. 1999;38(2): 147-155. 25. Leaf P, Alegria M, Cohen P, et al. Mental health service use in the community and schools: results from the four-community MECA Study: Methods for the Epidemiology of Child and Adolescent Mental Disorders Study. Journal of the American Academy of Child and Adolescent Psychiatry. 1996;35(7):889-897. 26. Ridgely S, Giard J, Shern D. Florida's Medicaid mental health carveout: lessons from the first years of implementation. Journal of Behavioral Health Services & Research. 1999;26(4):400-415. 27. Dillman D. Mail and Telephone Surveys: The Total Design Method. New York: John Wiley & Sons Inc; 1978. 28. Salant P, Dillman D. How to Conduct Your Own Survey, New York: John Wiley & Sons Inc; 1992. 29. Boothroyd R, Shern D. Assessing the Feasibility of Using Mail Survey Methodology With Medicaid Enrollees: A Summary of a Pilot Study for the member survey. Tampa, Fla: Louis de la Parte Florida Mental Health Institute, University of South Florida; 1998. 30. Jellinek M, Murphy J. Screening for psychosocialdisorders in pediatric practice. American Journal of Diseases of Children. 1988; 112:11531157.

31. Murphy J, Reede J, Jellinek M, et al. Screening for psychosocial dysfunction in inner-city children: further validation of the pediatric symptom checklist. Journal of the American Academy of Chill and Adolescent Psychiatry. 1992;31(6): 1105-1111. 32. Shem D, Giard J, Robinson E et al. Evaluation of Florida's Prepaid Mental Health Plan: Year 4 Report. Tampa, Fla: University of South Horida, Louis de la Parte Florida Mental Health Institute; 2001. 33. Maynard C, Ehreth J, Cox G, et al. Racial differences in the utilization of public mental health services in Washington State. Administration and Policy in Mental Health. 1997;24(5):411-424. 34. Diala C, Muntaner C, Walrath C, et al. Racial differences in attitudes toward professional mental health care and in the use of services. American Journal of Orthopsychiatry. 2000;70(4):455-464.

E f f e c t o f I n s u r a n c e P l a n s o n C h i l d r e n ' s M e n t a l H e a l t h S e r v i c e Use

MANDELL

e t al.

237