Compiling and analyzing raw research study data into findings

¶ … Counseling

This study explored the efficacy of the client-directed outcome-informed therapeutic approach to counseling children and youth. Subjects were children and youth between the ages of eight and 16 who were assigned to a therapy as usual (TUA) group or to a CDIO group. Seven counselors trained in the CDIO approach and engaged in therapeutic with clients for two six-month periods. A suite of formal assessment tools was used to measure clients' satisfaction with therapeutic sessions and perceptions of goal attainment. Satisfaction levels and therapeutic outcomes were significantly better on all measures for clients in the CDIO group.

A therapeutic relationship is a little bit science and a little bit art. The rest is akin to the proverbial black box. The literature in the field of psychotherapy indicates that different therapeutic techniques account for only about eight to ten percent of the variance in client outcomes (Boeree, 2006; Brann, et al., 2001; Cooper, et al., 2010). Research on psychotherapeutic practices increasingly points to the substantive impact that the common factors of therapy have on the therapeutic relationship and on the therapy outcomes of clients (Duncan, et al., 2004; Rogers, 1951; Sachse & Elliott, 2002). Common factors of therapy include the dynamics of client and therapist expectations about therapy outcomes, shared understanding of therapy goals and objectives, and the personal characteristics of clients, which are indicated by their optimism, persistence, and positive attribution toward therapy (Duncan, et al., 2004; Zins, et al., 2000).

The work of Duncan, Miller, and Sparks (2004) argues for the importance of the common factors in therapy. Duncan et al., (2004) emphasize those common factors indicated by client feedback about how well they believe the therapy sessions are working for them, and about what clients believe about their own framework for change. A therapist's obligation in client-directed and outcome-informed (CDOI) therapy is more to monitor the therapeutic outcomes, attend to the therapeutic alliance established between the therapist and the client, and to make changes in the therapeutic approach based on feedback from the client (Duncan, et al., 2004; Hubble, et al., 1999).

The idea of a therapeutic alliance underscores the use of self in therapy (Fusco, 2012). In a relational approach to therapy with children and youth, use of self enables the construction of a "developmentally responsive practice" (Eccles, 2001; Fusco, 2012; Galaif, et al., 2001). The success of a CDOI approach may be related to that fact that it is "heavily dependent upon charismatic individuals who are a lot about the kids" (Eccles, 2001, p. 3). or, as Duncan et al. (2004) attest, the successful formula may rest in the responsive, iterative nature of the therapy that fosters within the client a measure of control over the therapeutic process. By design, the CDOI approach is a dynamic, responsive, and relational methodology. In its conventions, CDOI employs outcome measurement strategies that do not deal in static inputs and outputs, but rather are established in ecologies or dynamic systems (Fusco, 2012; Riggs & Greenberg, 2004; Riggs, et al., 2006).

Research problem. Counselors who are training in client-centered, outcome-informed therapy may go about their work differently than counselors who do not have this training. Potentially, client outcomes can differ as a result of the way that counselors in each of these two groups -- those with training in CDIO and those without that training -- conduct and monitor therapy. This study explores the relation between CDIO training of counselors and client outcomes.

Research questions. A number of assessment and feedback tools were used with clients in this study. Differences in the absolute scores and patterns of scores are of interest as they may indicate clients' responses to and perceptions of the CDOI approach as they experience it in their therapy sessions. To that end, the following research questions were posed:

Research Question #1 to what extent do Outcome Rating Scale (ORS) scores differ for clients in the TAU group and the CDOI group?

Research Question #2 to what extent do Session Rating Scale (SRS) scores differ for clients in the TAU group and the CDOI group?

Research Question #3 to what extent do Strengths and Difficulties (SDQ) scores differ for clients in the TAU group and the CDOI group?

Research Question #4 to what extent do Family Apgar (F-APGAR) scores differ for clients in the TAU group and the CDOI group?

Research Question #5 to what extent do Health of the Nation Scales for Children and Adolescents (HoNOSCA) scores differ for clients in the TAU group and the CDOI group?

Research Question #6 to what extent do SRS scores differ for clients in the TAU group and the CDOI group?

Hypothesis. The alternative hypothesis is that adopting the theory of a client-directed outcome-informed therapeutic framework with clients leads to meaningful differences in outcomes for these clients. The null hypothesis is that adopting the theory of a client-directed outcome-informed therapeutic framework with clients does not lead to any meaningful differences in outcomes for these clients.

Methodology

The setting for the research study was a public child and adolescent mental health service. Instruments used in this study are those routinely used by the counselors who provide treatment in the public child and adolescent mental health services facility and program.

Participants. All of the subjects in the research study were clients who had undergone treatment for a 12-month period of time. Two cohort groups were formed based on when they were registered for treatment over the course of the study. The first cohort group consisted of 35 families who were registered for treatment during the first three months. The second cohort group consisted of 42 families who were registered for treatment from the sixth month of the study.

The gender ratio of subjects was roughly 50-50 in the two cohort groups with slightly more males than females in each group. The age range of the subjects was from eight years of age to 16 years of age, with females tending to fall into the middle to upper age ranges and males tending to fall into the lower to middle age ranges.

Instrumentation. Six instruments were used to measure outcomes in this study. Counselors who approach therapy from the client-directed outcome-informed perspective (Duncan, et al., 2004) routinely use two of the instruments, the Outcome Rating Scale (ORS) and the Session Rating Scale (SRS). All of the instrumentation in this research have been established in the field of child and adolescent therapy, including those that have previously been adapted for use with children. Brief descriptions of the instruments follow:

The Outcome Rating Scale (ORS) provides a four-scale measure of personal well-being, overall well-being, and satisfaction with life and personal relationships. Scores on the four interval-based scales are totaled to indicate a score between 0 and 40. This is one of the primary tools used to assess outcomes in CDOI therapy (Duncan, et al., 2004).

The Session Rating Scale (SRS) is similar to the Outcome Rating Scale (ORS), using four visual analog scales to assess the therapeutic sessions in which the client has just participated (Duncan et al., 2004). The four areas evaluated are the goals and topics of the therapy, the counselor's approach to therapy, the therapeutic relationship, and the overall satisfaction with the session.

The Strengths and Difficulties (SDQ) is a screening instrument with items that pertain to negative and positive attributes evenly divided across five scales (Goodman, 1999). Each scale is focused on a single area, such as emotional symptoms, conduct problems, hyperactivity / inattention, peer relationship problems, and prosocial behavior (Goodman, 1999). The Prosocial score is reported separately based on the single positively oriented scale and the scores from the other scales are combined for a Difficulties score (Goodman, 1999).

The Family Apgar (F-APGAR) is a five-item questionnaire developed to assess the levels of satisfaction for family adaptation, partnership, growth, affection, and resolve (Smilkstein, et al., 1982). In this study, the mothers who presented with the clients were the respondents on this measure. In previous studies, the score of the mother has been shown to be a reliable indicator for the average family APGAR score, where the measure was administered to all family members (Smilkstein, et al., 1982).

The Health of the Nation Scales for Children and Adolescents (HoNOSCA) was developed to measure the functioning and health of children and youth who experience mental illness (Gerralda & Yates, 2000). The rating scale provides an assessment of symptoms, behaviors, impairments, and social functioning of patients or clients (Gerralda & Yates, 2000). A total score is derived from the clinician scored items that reflects the symptoms the client is experiencing and the social / family functioning levels of the client's immediate family (Gerralda & Yates, 2000).

Research design. The study uses a quantitative approach to research and is grounded in positivist theory. The research design is quasi-experimental approach that uses a nonequivalent control group (DePoy & Gitlin, 1998). Since the background and descriptive information about the study does not indicate that the subjects were randomly assigned to groups, the research design must be considered quasi-experimental and not experimental (DePoy & Gitlin, 1998).

Procedures. All patients, regardless of whether they were participating in the study or not, received treatment as usual (TAU) for the first six months of the study. Measurement for this initial six-month period followed this sequence: A standard suite of measurements was administered at session one, session 6 and session 12; ORS and SRS assessments occurred at every treatment session for identified patients (IP) only. During this initial six-month period, counselors only received training in the use of the ORS and SRS as instruments to be added to the standard suite of outcome measures.

In the second six-month period, training in the client-directed outcome-informed approach to therapy was provided to all the counselors. The training components included the following: (1) 16 hours of formal introduction to theory of change according to the Duncan and Miller framework; (2) in-depth training on the use of ORS and SRS for obtaining client feedback and monitoring progress in therapy; (3) assessment of counselors' levels of understanding of CDOI therapy; (4) mastery of CDOI implementation by counselors; and (5) weekly meetings of the research study coordinators to discuss, workshop and resolve any emerging issues related to CDOI. Six counselors were involved in the entire study, and one new counselor was added for treatment of the second cohort.

Results

Data were analyzed through repeated measures using an Analysis of Variance (ANOVA). Outcome measures of behaviors, impairments, symptoms, and social functioning of the children and adolescents participating in the study were obtained through administration of the HoNOSCA. For the HoNOSCA, the means of the treatment groups (Treatment as Usual or TAU and Client-Directed Outcome-Informed or CDOI) were compared for three sources of variation: Treatment, interaction, and time. A significant effect was found for Treatment (F (1,201) = 206, p

Two scores were reported for the SDQ: Difficulties and Prosocial Skills. Assessment of behavioral attributes occurred at session 1, 6, and review. ANOVA analyses of the mean scores of the treatment groups (Treatment as Usual or TAU and Client-Directed Outcome-Informed or CDOI) were compared for three sources of variation: Treatment, interaction, and time.

ANOVA analysis of the mean scores at review points for the SDQ Difficulties generated the following results. A significant effect was found for Treatment (F (1,201) = 23.3, p

ANOVA analysis of the mean scores at review points for the SDQ Prosocial Skills measure gave the following results. The effect for Interaction was not significant. Of the three sources of variance, Time showed the most significant effect (F (2,201) = 84.8, p