Temporal Relationships of Child and Family Outcomes Across In-home Family Therapy[1]

Jennifer R. Lannigan, Lindsey R. Shorts, and Jeanne M. Slattery[2] (Clarion University)

 

In-home family therapy is an intensive treatment option, requiring multiple contacts with the child, family, and various sub-systems per week.  Therapists maintain frequent, on-going communication with Child Protective Services (CPS), the school, doctors, and other involved agencies, so that all agencies are working in concert.  Because this treatment is offered in a less restrictive environment (the home and community) than hospital or residential treatments, it is also more humane (Slattery & Knapp, 2003).

 

Mental health professionals can take several steps to determine whether the services they provide are beneficial.  Ideally, children and families will be assessed at intake to and termination from therapy to measure improvement.  We will discuss four of the six measures that we use: the Child and Adolescent Functional Assessment Scales (CAFAS), the Family Assessment Device (FAD), the Global Assessment of Functioning (GAF), and the Global Assessment of Relational Functioning (GARF).  The CAFAS is administered to parents in an interview format and is used to assess children’s behavior in eight realms of home, school, and the community (Hodges, 1999; Hodges, Doucette-Gates, & Liao, 1999; Hodges & Kim, 2000).  The FAD is a self-report completed by the family and used to assess family functioning and interactions (Kabacoff, Miller, Bishop, Epstein, & Keitner, 1990).  The GAF and GARF, respectively, assess individual and relational functioning and are based on therapists’ interactions with the family.

 

Child functioning is believed to both result from and contribute to family functioning (Minuchin, 1974).  One way that we can assess our work is to look at the relationships among family and individual outcomes across time.  While we expect that correlations across realms (family and child) would be more modest than relationships within a single realm (either child or family), we did expect to see statistically significant relationships, and that these would decrease across our measurement period.  That is, we expect to see stronger correlations among family and child outcomes at intake than between intake and exit.

 

Methods

We examined 240 cases of children and their families who were seen at Family Links, since its opening in May of 1994.  Clients considered for the study were those who had terminated treatment by the last data collection point in August 2003 and had given consent to have their information used in the study.  In thirty-two of the cases, the identified patient (IP) had gone through the program multiple times.  Throughout the course of treatment clients were evaluated with four assessment scales, the CAFAS, FAD, GAF, and GARF.

 

Client and family assessments were to be administered upon entry into the program and just prior to termination from the treatment.  Because this population is less “organized” than other groups, many do not complete outcome assessments at each point (Slattery et al., 2000).  With the exception of the CAFAS, which was first administered in 1997, more than 50 % of the clients examined were administered each assessment within one month of entry and also upon exit.  Only 99 of the 240 clients (41%) seen had completed both the entry and exit CAFAS.  The GAF and GARF measures were also distributed during the third month of treatment, however, only 5% of the clients had completed a mid-GAF assessment, whereas 59% of clients had a mid-GARF.

 

All identified patients were between the ages of two and eighteen, 90 females and 150 males.  The mean age for females entering the program was 12.8 years (SD = 3.38) with the youngest client at 3.7 years and the oldest 17.4.  The mean age for males entering the program was 11.6 (SD = 3.47) with the youngest client being 2.7 and the oldest 18.4.  The average length in treatment for clients was 156.4 days (SD = 4.29) with a minimum stay of 11 days and the longest stay 383 days.

 

All data were entered into StatView.  We conducted a correlation analysis with StatView, investigating relationships among the four assessment scales over three periods: within one month of intake, within one month of exit, and at the third month of treatment (GAF and GARF only).

 

Results

CAFAS entry scores and the difference between CAFAS scores at entry and exit (change score) were positively correlated (r(101)= .629, p<.0001), while exit CAFAS data were negatively correlated (r(99)=

-.284, p=.0042).  A similar pattern was found with FAD scores (entry-change: r(126)=.434, p<.0001; exit-change: r(125)= -.418, p<.0001).

 

We also compared the FAD and CAFAS measures to each other.  Looking at the entry scores of both measures, a small but significant positive correlation was found (r(172)= .275, p=.0002).  Similar relationships were observed in the exit data of these assessments (r(102)=.223, p=.0243), and their change scores (r(98)=.293, p=.0033).

 

GAF and GARF scales were compared to themselves and to each other.  In both cases we found that intake data are positively correlated with exit data, but the strength of these correlations decreased over time (intake to mid-GARF: r(138)=.438, p<.0001; intake to exit GARF: r(154)=.347, p<.0001).  GAF scores were compared in the same fashion, although the mid-GAF assessments were only administered to 14 clients.  This number was too small to make generalizations.  However, the entry GAF was positively correlated with the exit-GAF (r(153)=.222, p=.0056). 

 

Conclusions

Measures were correlated with themselves across all points and in general, tend to be less correlated at later than more recent points in time.  Furthermore, clients displaying more problems at entry changed more by exit on all assessment measures (CAFAS, FAD, GAF, GARF).  We suspect that this is because people with more problems initially were able to make more change across treatment. We also found positive relationships between the functioning of the child and the functioning of the family.  Such correlations suggest that child and family functioning are related to each other. 

 

These outcome data are, by necessity, limited (Slattery et al., 2000).  Although we have not compared outcomes with other treatments, most of the clients were reported “failures” of traditional outpatient therapy.  In effect, the families serve as their own comparison group.  We are also aware that families who complete all assessments may be different than those who do not and that this may limit generalizations made about the overall effectiveness of the program. 

 

Finally, we have not yet begun to pull apart the various treatment elements that might lead to the effectiveness of this program and cannot identify with confidence the cause of the changes we have observed.  We can only note that most of the clients had been seemingly unsuccessful in other programs, whereas many succeeded with in-home family therapy.   

 

References

Hodges, K.  (1999).  Child and adolescent functional assessment scale (CAFAS).  In M. E. Maruish (Ed.), Use of psychological testing for treatment planning and outcome assessment (2nd ed., pp.631-664). Mahwah, NJ: Lawrence Erlbaum.

Hodges, K., Doucette-Gates, A., & Liao, Q.  (1999).  The relationship between the Child and Adolescent Functional Scale (CAFAS) and indicators of functioning. Journal of Child and Family Studies, 8,109-122.

Hodges, K., & Kim, C. (2002).  Psychometric study of the Child and Adolescent Functional Assessment Scale: Prediction of contact with the law and poor school attendance. Journal of Abnormal Clinical Psychology, 28, 287-297.

Kabacoff, R. I., Miller, I. W., Bishop, D. S., Epstein, N. B., & Keitner, G. (1990).  Psychometric study of the McMaster Family Assessment Device in psychiatric, medical, and non-clinical samples.  Journal of Family Psychology, 3, 431-439.

Minuchin, S. (1974). Families and family therapy.  Cambridge, Mass: Harvard University Press.

Slattery, J. M., Buchenauer, E. L., & Bean, T. (2000). When real and ideal collide: Outcome assessment of family therapy. Poster presented at the annual meeting of the American Psychological Association, Washington, D.C.

Slattery, J.M., & Knapp, S.  (2003).  In-home family therapy and wraparound services for working with seriously at-risk children and adolescents.  In L. VandeCreek & T. L. Jackson (Eds.), Innovations in clinical practice: Focus on children and adolescents (pp. 135-149).  Sarasota, FL: Professional Resource Press.



[1] Presented at the 2004 annual meetings of the Eastern Psychological Association, Washington, D.C.

[2] Contact Jeanne Slattery by e-mail (jslattery@clarion.edu) or at Clarion University, Clarion, PA 16214