|Year : 2022 | Volume
| Issue : 2 | Page : 152-158
Child and Adolescent Service Intensity Instrument Development in Japan: Initial Psychometrics and Use
Yoshiro Ono1, Andres Julio Pumariega2, Akira Yamamoto3, Kazuhiro Yoshida4, Hiroshi Nakayama5, Daisuke Nakanishi6, Kanae Aihara3, Kayoko Ichikawa7, Masami Hanafusa8, Udema Millsaps9
1 Wakayama Prefecture Mental Health and Welfare Center, Wakayama, Japan
2 Department of Psychiatry, University of Florida, Gainesville, Florida, USA
3 Higashiosaka City Support Center for Persons with Disabilities, Higashiosaka, USA
4 Miyagi Psychiatric Center, Natori, USA
5 Kawasaki City Center for Child and Family Services, Kawasaki, USA
6 Mie Prefectural Medical Center for Growth, Development, and Disability, Tsu, USA
7 Kyoto University School of Public Health, Kyoto, Japan
8 Osaka Psychiatric Medical Center, Hirakata, Japan, Japan
9 The Reading Hospital and Medical Center, West Reading, Pennsylvania, USA
|Date of Submission||30-Jun-2022|
|Date of Acceptance||30-Jun-2022|
|Date of Web Publication||22-Aug-2022|
Wakayama Prefecture Mental Health and Welfare Center, 2-1-2 Tebira, Wakayama, 640-8319
Source of Support: None, Conflict of Interest: None
Objectives: Level of Care (LOC) determination is an important process of care planning for emotionally disturbed children and adolescents. As a tool of LOC determination, the Child and Adolescent Service Intensity Instrument (CASII) has been developed and tried to extend the usage out of the United States. As part of that effort, the CASII is currently being evaluated for its psychometric property in Japan. Methods: The CASII has been translated into Japanese, and then, 23 child psychiatrists and 70 nonpsychiatrist professionals who completed CASII training rated 7 vignettes to test inter-rater reliability. External validity was being evaluated by comparing the CASII ratings of 163 participants aged 6–17 against the Child Behavior Checklist (CBCL)/Youth Self-Report (YSR) and Child Global Assessment Scale (CGAS). Results: Inter-rater reliability among psychiatrists showed intra-class correlation coefficients ranging from 0.63 to 0.91 and those by nonpsychiatrists from 0.42 to 0.80. Cronbach's alpha was higher than 0.97 for both psychiatrists and nonpsychiatrists. The CASII ratings and LOC determination for 145 participants were significantly correlated with the Actual LOC and ratings of CBCL/YSR and CGAS except for the Internalizing Problem scores. There was no significant difference between LOC Recommended and Actual LOC. Conclusions: The Japanese version of the CASII showed fairly good inter-rater reliability and validity to use in the practice of mental health as well as child welfare in Japan.
Keywords: Child mental health services, child welfare, level of care
|How to cite this article:|
Ono Y, Pumariega AJ, Yamamoto A, Yoshida K, Nakayama H, Nakanishi D, Aihara K, Ichikawa K, Hanafusa M, Millsaps U. Child and Adolescent Service Intensity Instrument Development in Japan: Initial Psychometrics and Use. World Soc Psychiatry 2022;4:152-8
|How to cite this URL:|
Ono Y, Pumariega AJ, Yamamoto A, Yoshida K, Nakayama H, Nakanishi D, Aihara K, Ichikawa K, Hanafusa M, Millsaps U. Child and Adolescent Service Intensity Instrument Development in Japan: Initial Psychometrics and Use. World Soc Psychiatry [serial online] 2022 [cited 2022 Sep 26];4:152-8. Available from: https://www.worldsocpsychiatry.org/text.asp?2022/4/2/152/354186
| Introduction|| |
Children and adolescents have a range of mental health needs, which should be addressed by various agencies including child welfare, education, juvenile justice, pediatrics, public health, as well as mental health. Currently, it has been well known that an effective system of care which includes every agency and resource in the community would be necessary to ensure appropriate care for children and their families. However, in providing metal health care for each individual, without carefully developed individual care plan, any care would be resulted in desirable outcome.
In the process of care planning, every factor related to the child or adolescent problems, including not only the individual psychopathology but also psychosocial and environmental factors, should be taken into account to make decision on what kind of care and the situation where care will be provided (Winters and Pumariega, 2007). Especially, in the case of child welfare, typically with remarkable psychopathology and protective needs, clinicians are often face to make decision on the most appropriate placement where the child would be cared by considering both mental health needs of the child and protective needs form adverse family environment.,
The concept of Level of Care (LOC) consists of two dimensions, therapeutic milieu and service intensity, suggesting where to provide treatment and how much treatment to provide. LOC determination is an important part of decision-making in the clinical practice for children and adolescents with emotional disturbance to provide effective and safety but least restrictive array of mental health services. Because it is common that every relevant agency in the community including child welfare, education, juvenile justice, police as well as mental health collaborates each other in the practice of children and adolescent mental health care, the concept of the LOC may be a best candidate for the decision-making in the treatment planning in the multidisciplinary and multiagency system of care.
For this purpose, the American Academy of Child and Adolescent Psychiatry (AACAP) has developed the Child and Adolescent Service Intensity Instrument (CASII) with the assistance of the American Association of Community Psychiatrists, which is an open evidence-based standard for LOC determination in children's systems of care (recently the latest version has been renamed the CALOCUS/CASII). The CASII has been evaluated in three major studies conducted in the United States, in which the CASII showed a good inter-rater reliability and validity.,,,,
Currently, the CASII is increasing use in the United States across a number of state mental health systems (including Maine, Nevada, Arizona, and Hawaii) as well as a number of mental health providers, managed care entities, and child welfare and juvenile justice programs.,,, Among the characteristics of the CASII, it cares for cultural competence, which allows appropriate administration for culturally diversity population. Given the property of cultural competence, the CASII would be available and useful internationally. Then, in response to the efforts of the Work Group on Community-Based Systems of Care of the AACAP to extend the use of the CASII outside the United States, the CASII has been translated into Japanese and investigated its initial psychometrics for the Japanese children and adolescents.
| Methods|| |
Developing Japanese version of the Child and Adolescent Service Intensity Instrument
The Japanese version of the CASII was developed through the process of forward translation (from English to Japanese) and back translation (from Japanese to English again), with panel adjudication by linguists and clinicians. Then, the final translation had been approved by the AACAP in December 2007.
Study 1: Inter-rater reliability
Raters were recruited from the staff members at the facilities participating in this study consisting of mental health and/or child welfare professionals (including child psychiatrists, psychologists, social workers, care workers, and nurses/public health nurses) and trained on the CASII per the AACAP-sanctioned procedure adjusted to the Japanese culture by one of the authors (Y. O.) who had completed the CASII training course in 2006. After 4-h training course at each study sites, participants were asked to rate seven vignettes which were same as those used in the original study by Fallon et al. but adjusted to the Japanese culture. Of the participants, 23 child psychiatrists and 70 professionals other than psychiatrists submitted CASII ratings for 7 vignettes. In total, 651 ratings were analyzed for the inter-rater reliability.
Study 2: Validity
Participants were children and adolescents aged 6–17-year-old referred to one of eight local research sites for mental health evaluation or LOC determination, including two psychiatric hospitals for children and adolescents, two child welfare centers (CWCs), and one community mental health clinic for children and adolescents across Japan. All participants were Japanese. Written assent was obtained from all participants and their parents/guardians. The study was approved by the Institute Review Board of the Reading Hospital and the local ethical committees at each research site.
Participants were rated by the CASII as well as the Child Behavior Checklist (CBCL), and/or the Youth Self-Report (YSR), if the participants were older than 11 years old, and the Child Global Assessment Scale (CGAS). To test criterion-related validity, the CASII ratings were compared to the ratings of CBCL/YSR and CGAS. Furthermore, the LOC Recommended (LOC Recommended) obtained by the CASII rating was compared to the Actual LOC (Actual LOC) of each participant.
The inter-rater reliability was estimated by using the reliability analysis for SPSS that matches to intra-class correlation coefficient (ICC) (2, k) described in the “Intraclass Correlations: Uses in Assessing Rater Reliability” by Shrout and Fleiss. The SPSS Inc. Released 2007. SPSS for Windows, Verson 16, SPSS, Inc., Chicago, Illinois, US match to the formulas is described by Yaffee. The method used on SPSS was a Reliability Analysis for Intra-Class Correlations with the options chosen of Two-Way Random, Absolute Agreement, and Single Measure.
Cronbach's alpha is equal to the stepped-up consistency version of the ICC, which is commonly used in observational studies. This can be viewed as another application of generalizability theory, where the items are replaced by raters or observers who are randomly drawn from a population. Cronbach's alpha will then estimate how strongly the score obtained from the actual panel of raters correlates with the score that would have been obtained by another random sample of raters.
For validity tests, CASII ratings (Subscale Scores, Total Scores, and LOC Recommended) were compared to CBCL/YSR (Internalizing Problems, Externalizing Problems, and Total Problems), CGAS, and the Actual LOC by the Pearson's correlation coefficients. The statistical significance was adopted by the level of 0.05.
Differences of the LOC Recommended and the Actual LOC were by two-tailed t-test with the significance level of 0.05.
| Results|| |
Study 1: Inter-rater reliability
ICCs of 7 vignettes ratings by 23 child psychiatrists were very strong, with absolute agreement single measures scores ranging from 0.63 to 0.85 for dimension scores, 0.91 for CASII Total Scores, and 0.84 for LOC Recommended [Table 1].
The ICCs for the ratings by nonpsychiatrists were relatively low compared to those by child psychiatrists, ranging from 0.42 to 0.799 for dimension scores and 0.73 and 0.62 for Total Scores and LOC Recommended, respectively [Table 1]. When compared the ICCs of ratings by nonpsychiatrists by clinical experience and education level, raters with clinical experience more than 10 years were somewhat higher than those with clinical experience below 10 years, but there was little difference between bachelor level raters and master level raters [Table 2].
|Table 2: Intra-class correlation coefficients by clinical experience and education of nonpsychiatrist raters|
Click here to view
Cronbach's alpha coefficients were all above 0.9 for both child psychiatrists and nonpsychiatrists raters.
Study 2: External validity
The participants of validity study were 111 males and 52 females, aged 6–17 years. Eighty-six were living their homes and 15 and 86 were placed in temporary residential care and residential facilities, respectively, and 47 were hospitalized. Among 163 participants, 145 (89.0%) had any psychiatric diagnosis of Diagnostic and Statistical Manual of Mental Disorders-IV (American Psychiatric Association, 2002), of which learning disorders or pervasive developmental disorders were most prevalent (57 participants). The demographic data of the participants are shown in [Table 3].
Although male participants were more than twice as many as females, there was no statistically significant difference in the LOC Recommended between males and females (χ2 = 7.2, df = 5). Furthermore, there was no significant difference in LOC Recommended between the participants aged 12 years and younger and those over 12 years (=5.35, df = 5).
Correlations of youth Self-Report, Child Behavior Checklist, and Child Global Assessment Scale
Actual LOC significantly correlated with the score of the CBCL Externalizing Problems. CASII Total Scores significantly correlated with CBCL Externalizing and Total Problems scores. The CASII LOC Recommended significantly correlated with the YSR Externalizing score as well as CBCL Externalizing and Total Problems. CGAS scores were significantly correlated with the Actual LOC, CASII Total Score, and LOC Recommended (CGAS correlations are negative since scores increase with higher level of function). Correlations between the CASII Total Scores, LOC Recommended, and the CGAS were higher than the correlations between the CGAS and the Actual LOC [Table 4].
|Table 4: Correlations between Actual Level of Care, Child and Adolescent Service Intensity Instrument, Youth Self-Report, Child Behavior Checklist, and Child Global Assessment Scale|
Click here to view
Correlations between the Child and Adolescent Service Intensity Instrument Scores and Actual Level of Care and Child Global Assessment Scale
All CASII Subscale Scores, Total Score, and LOC Recommended significantly correlated with the Actual LOC. The CASII Subscale Scores except for the Comorbidity, Environmental Support, and the Family Acceptance negatively correlated with the CGAS significantly [Table 4].
Difference in the Child and Adolescent Service Intensity Instrument Level of Care Recommended and the Actual LOC
[Figure 1] shows the distribution of ratings by the difference of LOC Recommended and the Actual LOC. The LOC Recommended was same to that of the Actual LOC in 68 ratings (41.7%), and there was −1 and +1 difference in 37 (22.7%) and 34 (20.9%), respectively. Thus, 139 (85.3%) were within plus–minus 1 LOC between LOC Recommended and Actual LOC.
[Table 5] shows differences in the LOC Recommended and the Actual LOC placement. The negative means indicate that several children/adolescents would have been better served at a lower LOC than actually placed. The mean difference between the LOC Recommended and the Actual LOC (LOC Recommended minus Actual LOC) was very low (−0.19). This was very close to zero but still showed that the average LOC Recommended was less than the average Actual LOC. The mean difference was not statistically different from zero.
| Discussion|| |
Although children and adolescents with emotional and behavioral problems draw increasing attention in Japan as well as other countries all over the world, Japanese society has been suffered from serious lack of child and adolescent mental health service resource. For example, there are only 401 specialty child and adolescent psychiatrists who are certified by the Japanese Society for Child and Adolescent Psychiatry for 20.6 million children and adolescents under the age of 19 years in 2020 (one child psychiatrist/500,000 children and adolescents).
To compensate for the lack of specialty mental health services, child welfare system has been addressing emotional and behavioral problems in childhood and adolescence. The CWCs which have been established by the Child Welfare Act in 1947 have played an important role in community mental health systems of care for children and adolescents. The CWC is a public child welfare agency responsible for supporting child rearing, providing services for children with physical or intellectual impairments, guidance for juvenile delinquents, and for ensuring protection of vulnerable children. Each CWC has a range of professionals as a staff that allows them to address a wide range of emotional and behavioral problems in children, including social workers, psychologists, and psychiatrists., However, because the CWCs are the primary agency for the reports, investigation, and intervention of child abuse and neglect, recent rapid increase of reports of child abuse and neglect has the CWC more responsible for mental health care for those children and adolescents involved in the child welfare system. As such, because the child welfare systems have been playing the role of de facto mental health system in Japan, the Japanese version of the CASII has been development not only in the clinical practice of specialty mental health services but also in the practice of the CWC.
In this study, the CASII showed inter-rater reliability as good as those tested in the process of development of the original version. The ICCs of 7 vignettes ratings by child psychiatrists were very strong ranging from 0.63 to 0.85 on dimension scores, 0.91 for Total Score, while those in the field test of the original version were 0.73 to 0.93 and 0.89, respectively. Although the ICC of the LOC Recommended was not appeared in the original study, it was 0.084 for the ratings by the Japanese psychiatrists which was also an evidence of good inter-rater reliability.
The ICCs for the ratings by nonpsychiatrists were relatively low compared to those by child psychiatrists: The ICCs for each subscale ranged from 0.42 to 0.80 and those for Total Score and LOC Recommended were 0.73 and 0.62, respectively. The ICCs of nonpsychiatrists in Japan were also lower than those of comparable raters in the United States (0.57 to 0.95 for Subscale Scores and 0.93 for Total Score). Nonetheless, the ICCs for the Japanese version by nonpsychiatrist raters were above acceptable level (0.40― 0.60), suggesting practical availability.
The lower ICCs of the Japanese professionals other than child psychiatrists might be explained by their difficulties in assessing impairment or symptoms of children and adolescents because they are typically not trained in the assessment using formal rating scales. Thus, the ICC on the subscale of Functional Status which reflects the assessment of impairment or symptoms was the lowest for the ratings by the Japanese nonpsychiatrists. In general, the ICCs for nonpsychiatrists with more than 11 years of clinical experience were higher than those for less experienced raters, while the ICCs were not different by the education level (master level vs. bachelor level). Therefore, it was considered that the CASII had substantial inter-rater reliability for both child psychiatrists and nonpsychiatrist professionals, and the ratings by more experienced raters were more reliable.
The external validity of the CASII for Japanese children and adolescents were tested by means of comparing the CASII scores and LOC Recommended to the scores of the CBCL/YSR and CGAS in the clinical samples both from specialty child and adolescent mental health service and child welfare.
Although the Actual LOC was correlated with only the Externalizing Problems of the CBCL/YSR and CGAS, the CASII Total Scores, and LOC Recommended were significantly correlated with both Externalizing Problems and Total Problems of the CBCL/YSR as well as CGAS. However, the Internalizing Problems of the CBCL/YSR failed to show significant correlation with the Actual LOC and the CASII Total Scores and the LOC Recommended. While those results suggested that the CASII Total Scores and LOC Recommended correlated well with the present problems and functioning level of the child or adolescent, the absence of significant correlations with the Internalizing Problems of the CBCL/YSR suggested that the Internalizing Problems contributed less to the LOC determination by the CASII in the current sample.
Because the original CASII have been demonstrated significant correlation with all scales of the CBCL/YSR (AACAP, 2004), the lack of significant correlation with the Internalizing Problems in the Japanese sample should be discussed further. First of all, in the practice of child welfare as well as mental health care, professionals are likely to pay attention to more Externalizing Problems, which may resulted in the inconsistency of the CASII ratings and the Internalizing Problem score of the CBCL/YSR. In fact, the Internalizing Problems such as depression are less frequently diagnosed in the Japanese child and adolescent mental health clinics. Furthermore, there are relatively few mental health services of higher LOC for children and adolescents with predominantly Internalizing Problems in Japan, which might be a possible cause of the lack of correlation between the Actual LOC and the Internalizing Problems of the CBCL/YSR. Second, the oversampling of developmental disorders (57 out of 145 participants; 39.3%) may attribute to inconsistent assessment of the Internal Problems in the CBCL. Contrary to the Externalizing Problems which are typically rather easy to be recognized by others, the Internalizing Problems such as anxiety or depression may be underestimated by the caregivers who administered the CBCL because of the limitation of communication in the individuals with developmental disorders.
The CASII ratings (Subscale Scores, Total Scores, and LOC Recommended) were all significantly correlated with the Actual LOC. Moreover, 85.3% of LOC determination by the CASII fell within one LOC difference, and showed no significant difference in the mean difference between the LOC Recommended and the Actual LOC. Those results all indicate good predictive validity of the CASII.
The results in this study suggested the utility of the CASII in the practice of both child and adolescent mental health service and child welfare in Japan. However, some caution should be noted in the application of the Japanese version of the CASII. First, as the AACAP strongly encourages trainings on the CASII for every rater to maintain the level of reliability and validity, the contents of trainings should be tailored to the culture and clinical situation where the CASII would be used. The current study revealed some weakness of the Japanese professionals such as difficulties in rating clinical symptoms and impairments, especially Internalizing Problems. The training course for the Japanese professionals should focus on clinical assessment of a range of emotional and behavior problems in children and adolescents. Second, to maximize the utilities of the CASII in the mental health practice, development of full range mental health service would be critical. Although most of the community in Japan has been suffering from lack of mental health service for children and adolescents, service programs of higher LOC are especially scarce. It is expected that the introduction of LOC determination by the CASII stimulate the public and policy-makers to develop full range system of care for emotionally disturbed children and adolescents in Japan.
| Conclusion|| |
The CASII has been translated into Japanese which showed fairly good psychometric properties, promising utility in the practice of child and adolescent mental health as well as child welfare in Japan.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Daleiden EL, Chorpita BF, Donkervoet C, Arensdorf AM, Brogan M. Getting better at getting them better: Health outcomes and evidence-based practice within a system of care. J Am Acad Child Adolesc Psychiatry 2006;45:749-56.
Stroul B. Issue Brief-System of Care: A Framework for System of Reform in Children's Mental Health. Washington, DC: Georgetown University Child Development Center, National Technical Assistance Center for Children's Mental Health; 2002.
Hurlburt MS, Leslie LK, Landsverk J, Barth RP, Burns BJ, Gibbons RD, et al.
Contextual predictors of mental health service use among children open to child welfare. Arch Gen Psychiatry 2004;61:1217-24.
Bai Y, Wells R, Hillemeier MM. Coordination between child welfare agencies and mental health service providers, children's service use, and outcomes. Child Abuse Negl 2009;33:372-81.
Lyons JS, Abraham ME. Designing level of care criteria. In: Kiser LJ, Leftkovitz PM, Kennedy LL, editors. The Integrated Behavioral Health Continuum: Theory and Practice. Arlington: American Psychiatric Publishing; 2001. p. 123-42.
Winters NC, Pumariga A, Work Group on Community Child and Adolescent Psychiatry, Work Group on Quality Issues. Practice parameter on child and adolescent mental health care in community systems of care. J Am Acad Child Adolesc Psychiatry 2007;46:284-99.
American Academy of Child and Adolescent Psychiatry. CASII User's Manual: Child and Adolescent Service Intensity Instrument. Washington, DC: American Academy of Child and Adolescent Psychiatry; 2004.
American Association of Community Psychiatry and American Academy of Child and Adolescent Psychiatry. Child and Adolescent Level of Care/Service Intensity Instrument (CALOCUS-CASII). Edition 1.2 (b). Washington, DC: American Association of Community Psychiatry and American Academy of Child and Adolescent Psychiatry; 2020.
Sowers W, Pumariega A, Huffine C, Fallon T. Level-of-care decision making in behavioral health services: The LOCUS and the CALOCUS. Psychiatr Serv 2003;54:1461-3.
Daleiden D. Child Status Measurement: Operating Characteristics of the CALOCUS and CAFAS. Honolulu: State of Hawaii, Child and Adolescent Mental Health Division; 2004.
Fallon T Jr., Pumariega A, Sowers W, Klaehn R, Huffine C, Vaughan T Jr., et al
. A level of care instrument or children's systems of care: Construction, reliability and validity. J Child Fam Stud 2006;15:140-52.
Pumariega A, Millsaps U, Moser M, Wade P. Matching intervention to need in juvenile justice: The CASII level of care determination. Adolesc Psychiatry 2014;4:305-13.
Pumariega AJ, French W, Millsaps U, Moser M, Wade P. Service intensity/level of care determination in a child welfare population. J Child Fam Stud 2019;28:1502-11.
Nakamura BJ, Daleiden EL, Mueller CW. Validity of treatment target progress ratings as indicators of youth improvement. J Child Fam Stud 2007;16:729-41.
Achenbach TM. Manual for the Child Behabior Checklist and 1991 Profile. Burington, VT: University of Vermont, Department of Psychiatry; 1991a.
Itani T, Kanbayashi Y, Nakata Y, Kita Y, Fujii H, Kuramoto H, et al
. Standardization of the Japanese version of the child behavior checklist/4-18. Psychiatr Neurol Paediatr Jpn 2001;41:243-52.
Achenbach TM. Manual for the Youth Self Report and 1991 Profile. Burington, VT: University of Vermont, Department of Psychiatry; 1991b.
Kuramoto H, Kanbayashi Y, Nakata Y, Fukui T, Mukai T, Negishi Y. Standardization of the Japanese version of the Youth Self Report (YSR). Jpn J Child Adolesc Psychiatry 1999;40:329-44.
Shaffer D, Gould MS, Brasic J, Ambrosini P, Fisher P, Bird H, et al.
A children's global assessment scale (CGAS). Arch Gen Psychiatry 1983;40:1228-31.
Shrout PE, Fleiss JL. Intraclass correlations: Uses in assessing rater reliability. Psychol Bull 1979;86:420-8.
Yaffee R. Enhancement of Reliability Analysis: Application of Intraclass Correlations with SPSS/Windows v. 8. New York: New York University; 1998. Available from: http://www.nyu.edu/its/socsci/docs/intracls.html
. [Last accessed on 2022 May 15].
Ono Y, Ishida Y, Ide H, Okamoto M, Kataoka J, Kameoka S, et al
. Psychiatric involvement in Child Welfare Centers in Japan. Jpn J Child Adolesc Psychiatry 2004;45 Suppl:35-50.
Ono Y. Juvenile delinquency and challenge of child and adolescent mental health and juvenile justice in Japan. Adolesc Psychiatry 2014;4:270-7.
Ono Y. Suicide in children and adolescents: A Japanese perspective. In: Garralda ME, Raynaud JP, editors. Culture, Diversity and Child and Adolescent Psychiatry. Lanham, MD: Jason Aronson; 2008. p. 171-89.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]