Psychological Assessment as a Protective Factor against Treatment Dropout in Psychoanalytic Child Psychotherapy of Children : Empirical Data

Psychological assessment can provide important information on patient’s symptoms, personality, diagnosis and prognosis assisting in the development of therapeutic strategies. This retrospective cohort study aimed to determine whether there is association between performing a psychological assessment prior to psychotherapy and keeping the treatment. Medical records of 2,106 children seen in two outpatient mental health clinics in Porto Alegre were analyzed. Children who had undergone psychological assessment before starting psychotherapy were 65% more likely to adhere to the treatment and 44% less likely to drop it out than were children who had not been assessed. Psychological assessment helps parents understand their children’s treatment, making it a less coercive process and provides a more concrete foundation for the otherwise subjective psychotherapeutic process.

Assessment has been a historical strength of psychology, with sophisticated traditions of measurement, psychometrics, and theoretical underpinnings (Youngstrom, 2013).A Psychological Assessment1 is a process which investigates a patient's symptoms and enables a better understanding and description of a patient's personality.This information helps the professional to reach a prognosis and to decide on a strategy or approach for treatment.Thus, psychological assessment may well be seen as a scientifi c process using psychological evaluations and techniques to understand psychological problems in light of theoretical assumptions (Cunha, 2000;Kamphaus, Rowe, Dowdy, & Hendry, 2006), within a limited timeframe.
In order to conduct this assessment, the child psychologist must have sound knowledge of normal child development, while also having a clear understanding of the developmental stage and context the patient is in.Subsequently, the psychologist will be able to assess the extent of the pathology and any possible stagnations or regressions in the child's development (Dian, 2007).Trinca (1984) contends argues that psychological assessment should include a variety of tools instruments so as to reach a global psychological understanding of the patient: interviews, tests, psychological tactics, contact with other professionals connected to related to the case, visitation to the child's school and medical exams.
Several authors point out that making evaluation, whether therapeutic, preventive or health-promoting in nature, is part of a psychologist's job description (Almeida, 1996;Casullo, 1999;Quelho, Munhoz, Damião, & Gomes, 1999).For a patient to be referred to proper psychotherapy, it is highly advisable that there be a moment in which the patient will be psychologically evaluated.In these evaluation appointments, the professional makes use of tools and techniques to better understand the person who is seeking assistance (Nunes, Silva, Deakin, Dian, & Campezatto, 2006).
According to França (1999) and Lilja (1997), a psychological assessment should be the starting point for any psychotherapeutic treatment, as it enables the therapist to decide what to do with his/her patient, how to do it and why do it in this or that way.In this sense, it can be a valuable means help to guide treatment, assisting in establishing an accurate indication.Data from more than 125 meta-analyses on test validity and 800 samples examining multimethod assessment about psychological testing and psychological assessment suggest that clinicians who rely exclusively on interviews are prone to incomplete understandings (Meyer et al., 2001).
Besides, it is possible that performing a psychological assessment can help to enhance therapeutic alliance during later psychotherapy.Hilsenroth, Peters and Ackerman (2004) examined a sample of outpatients and their therapists at the end of the assessment feedback session, early, and late in psychotherapy, aiming at verifying the impact of patient-and-therapist-rated alliance developed during psychological assessment on the subsequent alliance measured early and late in formal psychotherapy.They found that a working alliance developed during psychological assessment between the patient and therapist would carry into formal psychotherapy.More studies are needed to verify if this fi nding remains true in cases where psychological assessment and psychotherapy are not conducted by the same professional.
Therapeutic alliance is also one of the key components of the psychotherapeutic process in all psychotherapeutic approaches.A poor therapeutic alliance often contributes to psychotherapy dropout and can be associated with unilateral termination (Hill & Knox, 2009).Despite the relative lack of research in the fi eld of child psychotherapy -especially therapies based on a psychoanalytic approach -understanding the factors that contribute to child psychotherapy dropout seems very important, not only to identify which cases are at risk of dropout, but also to intervene in a way to best engage the family in the treatment (Kazdin, 1996).Defi ning psychotherapy dropout, though, is not an easy task for researchers.To defi ne psychotherapy dropout grounded purely in the non-attendance to therapy criteria fails to cover cases in which the patient simply does not schedule a next meeting (Wierzbicki & Perakik, 1993).
Specifi cally for psychoanalytic psychotherapies, because the treatment modality does not defi ne the number of sessions to achieve treatment goals, it is not possible to defi ne dropout based on the criteria of a cut-off point for treatment duration.Therefore, defi nitions of psychotherapy dropout based on the therapist's judgment turn out to be the most reliable, since they respect the uniqueness of each case and can be applied to any form of treatment (Deakin, Gastaud, & Nunes, 2012).However, studies using this defi nition are often criticized because of the subjectivity of the evaluation of each therapist, who often uses different criteria to classify a patient as a dropout case.Based on a consideration of the above factors, Gastaud and Nunes (2010) proposed the following categories for completion (or non-completion) of treatment in psychoanalytic psychotherapy: 1. Non-adherence: treatment is interrupted in the psychotherapy evaluation phase, in other words, before the treatment goals are made clear to the participants or in situations in which there is no indication for treatment.2. Dropout: psychotherapy is ended before the goals set on the contract have been met, regardless of the reasons that led the patient or therapist to interrupt treatment and whether the decision was unilateral or bilateral.3. Discharge: psychotherapy is ended when the goals set on contract have been met.Even though there are methodological diffi culties in researching the phenomenon, the study of the variables that may be related to treatment dropout is crucial to the clinic, since the literature indicates dropout as the most prevalent outcome of psychotherapy (Luk et al., 2001;Midgley & Navridi, 2006).Deakin and Nunes (2009) suggested that families who lack motivation to engage their children in child psychotherapy in the initial sessions terminate treatment prematurely.Nock and Kazdin (2001) examined the relation of parent expectancies and participation in therapy.The results indicated that parents' expectations predicted subsequent barriers to treatment participation, treatment attendance and premature termination from therapy.Parents' expectations in child psychotherapy are very important because they are the ones who usually decide whether a child will attend psychotherapy.Performing a psychological assessment before starting psychotherapy might help parents to adjust their expectancies regarding their child and regarding therapeutic goals.Specifi cally for psychoanalytic psychotherapy, in which the process can often be perceived by parents as too subjective, the initial psychological assessment may be useful to check suitability to treatment and to concretely support treatment indication.
Within the psychoanalytic theoretical approach, psychological assessment is recognized as an important tool in evaluating children for psychotherapy (Zavaschi, Bassols, Bergmann, & Costa, 2005;Zavaschi, Iankilevich, Recondo, & Rohde, 1998).Still, some authors do not mention its relevance to psychotherapists of children in the main textbooks read by local professionals (Aberastury, 1982;Lowenkron & Frankenthal, 2001;Zimerman, 2004); for them, psychoanalytic psychotherapists of children evaluate patients using playful interviews that leave the child free to play with unstructured materials.In order to leverage this discussion, it seems necessary to test through empirical studies some of the potential harms and benefi ts of psychological assessment related to psychoanalytic psychotherapy.
Based on studies at outpatient facilities using records of children who sought psychotherapy and based on the above mentioned points, our working hypothesis is that there is an association between the performance of a psychological assessment, considered an important process for the patient, and the patient's permanence in treatment, given that a complete evaluation using these tests as a resource can provide important information on the functioning of the patient and can direct each treatment towards a more specifi c course of action.Aims 1.To examine whether the implementation of a psychological assessment (psychodiagnostic process) prior to psychotherapy of children is positively associated with greater adherence to psychoanalytic psychotherapy.2. To examine whether the implementation of a psychological assessment (psychodiagnostic process) prior to psychotherapy of children is a protective factor against dropout of psychoanalytic psychotherapy.

Method
A retrospective quantitative descriptive study was carried out using psychotherapy psychological records of children seen at two institutions located in Porto Alegre, Southern Brazil.
Age range was defi ned based on the Brazilian Statute of Child and Adolescent (Estatuto da Criança e do Adolescente, 1990Adolescente, /2000)), which defi nes a child as being a person under twelve years of age.All psychological records of children who sought psychotherapy between 1979 and 2007 were analyzed.If the same patient happened to seek assistance twice at different moments in time, the second psychological records were excluded.When seeking care at these two clinics, patients are told that they are educational institutions and are asked to sign a consent form authorizing their data to be used for teaching and research purposes.The consent form was kept in the psychological record if the patient authorized and signed it.We included in this research only medical records pertaining to patients whose guardians signed the informed consent form when starting treatment.
The database was set up and analyzed using SPSS 17.Data was collected based on the socio-demographic and clinical variables taken from the medical records which were fi lled in by the child's guardian, therapist and interviewer who performed the intake interview.
End of treatment (outcome of this research) was defi ned based on the therapists' notes on the medical charts, using the defi nition proposed by Gastaud and Nunes (2010), presented earlier.
The analysis of this research, being a retrospective study, took into account the records of the intake interviewer to determine whether or not the child had undergone psychological assessment before reaching the interview.Parents (or guardians present at the interview) were asked about the child´s previous assessments or treatments.Children whose parents (or guardians) reported psychological assessment prior to the intake interview were compared with children whose parents did not report the performance of a previous assessment.Thus, the notes made by the interviewer concerning any assessments carried out prior to arrival were taken into account.
So as to characterize the sample, the frequency and percentages of all variables of interest were collected.To examine the relation between factors and outcomes both the Chi-square test (x 2 ) and the Odds Ratio (OR) were used.To calculate the independent effect that a factor could play on the outcome, a logistic regression was performed.All variables presenting p≤ .10 when associated to outcome were included in the logistic regression.Results were considered signifi cant when p was equal to or below .05.

Results
The medical charts of 2,106 children were analyzed, of which 1,750 had been referred to psychotherapy after intake interview.
Given that the aim of this study was to check whether the rates of non-adherence/adherence and dropout/discharge differed for children who underwent psychological assessment, the fi nal sample of this study only comprised patients referred to psychotherapy at intake interview.
At intake interview, it was possible to obtain information concerning whether or not 90.8% of the patients had undergone psychological assessment.At least 187 children had undergone psychological assessment before being sent to psychotherapy (Table 1).2, where it can be observed that 13.3% of the children did not adhere to psychotherapy and 39.2% abandoned their treatment.Children who did not adhere to treatment left within the fi rst month, whereas 7 months (SD= 8.12 months) was the average duration of psychotherapy among those who dropped out and 18 months (SD=11.93)was the average among those who were discharged.

Associations with Non-Adherence versus Adherence
The fi rst analysis aimed at verifying whether previous psychological assessment would in fact associate with psychotherapy adherence rates.Thus, all dropout and discharged patients were placed together to make up a category named "adherence", given that all these patients went beyond the evaluation phase by starting the actual psychotherapy process.Those patients whose treatment was ongoing and those whose charts did not specify this information were excluded from the sample in this analysis as it would not be feasible to accurately determine whether they belonged to the "adherence" or the "non-adherence" category.
An association was found (x 2 =9.819; df=1; p=.002) between adherence/non-adherence and the fact that a child had already undergone a psychological assessment prior to receiving the indication to undergo psychotherapy.Children who had undergone the psychological assessment before being referred to psychotherapy were more likely to adhere to treatment than children who had not (OR=.341;p=.003).
Besides psychological assessment prior to psychotherapy, there were other variables in this sample which also associated positively with adherence to treatment: having undergone a psychiatric assessment (x 2 =5.404; df=2; p=.067), having undergone a neurological assessment (x 2 =5.430; df=2; p=.066) and source of referral (x 2 =30.804; df=7; p<.001).Having undergone psychological assessment was also associated with psychiatric (x 2 =43.186; df=2; p<.001) and neurological assessments (x 2 =72.556; df=2; p<.001) prior to intake interview and source of referral (x 2 =16.311; df=7; p=.022).It became necessary to control the individual effect of psychological assessment prior to triage on adherence to treatment.Logistic regression demonstrated that the association between psycholog-ical assessment and non-adherence to treatment withstood analysis even after controlling for such variables (OR=.353;p=.008).Children who underwent a psychological assessment prior to commencing psychotherapy were found to be 65% more likely to adhere to treatment than were those who did not.

Associations with Dropout versus Discharge
In the analysis of the second outcome, which aimed at checking whether previous psychological assessment was associated with psychotherapy dropout and discharge rates, any patients who had not adhered to their treatments were excluded from this sample as they had not effectively undergone any therapeutic intervention.Patients whose treatment was still ongoing and those whose medical charts failed to provide information concerning end of treatment were also excluded from this sample.
An association was found between dropout/discharge and the fact that a child had already undergone a psychological assessment prior to receiving the indication to undergo psychotherapy (x 2 =10.191; df=1; p<.001).
Children who had undergone a psychological assessment before being referred to psychotherapy were less likely to abandon their treatments than were children who had not undergone a psychological assessment (OR=.480;CI95% .304-.759; p=.002).
The variables of this study sample also associated with treatment dropout were: having undergone a psychiatric assessment (x 2 =5.606; df=2; p=.061), having undergone a neurological assessment (x 2 =15.7373; df=2; p<.001), source of referral (x 2 =16.348; df=7; p=.022) and people the child lives with (x 2 =6.858; df=3; p=.077).It then became necessary to control the individual effect of psychological assessment prior to intake interview on treatment dropout.Logistic regression demonstrated that the association between previous psychological assessment and treatment dropout rates withstood analysis even after controlling for such variables (OR=.552;p=.024).
Children who had undergone a psychological assessment prior to commencing psychotherapy were found to be 44% less likely to drop out of their treatments than were those who had not.

Discussion
The main fi nding of this study is that children who underwent psychological assessment prior to psychotherapy were approximately three times more likely to adhere to treatment and were approximately two times less likely to abandon their treatments.
Given these fi ndings, it can be said that the information provided by the psychological assessment might be a factor that favors patient adherence.Performing an effi cient referral is very important for the success of any psycho-therapy treatment.More concrete and tangible results, from which parents can understand and visualize their children's symptoms and diagnoses on tables and graphs, help parents make decisions themselves about treatments rather than be convinced to make them.It is important to point out that in children psychotherapy the motivation comes from the parents as it is they who bring the patient to treatment.Consequently, the clearer the path to be taken is to the parents, the better the chances are that the children will remain in treatment.Moreover, as pointed out by Meyer at al. (2001), a multimethod assessment battery provides a structured means for skilled clinicians to maximize the validity of individualized assessments.It is possible to hypothesize that both patients and therapists benefi t from performing a complete psychological assessment before starting a psychotherapy process.
Besides, Poston and Hanson (2010) highlight that psychological assessment itself can have a therapeutic intervention.Using meta-analytic techniques, they concluded that psychological assessment procedures -when combined with personalized, collaborative, and highly involving test feedback -have positive, clinically meaningful effects on treatment, especially regarding treatment processes.
The research presented here did not aim at evaluating the therapeutic action of psychological assessments, yet it found that psychological assessment may be of great importance for the psychotherapeutic process.The decision to prescribe psychotherapy for a child should include not only patient´s and therapist´s preferences, it should also integrate assessment data.An eclectic integration of evidence-based treatments and evidence-based assessment generates a powerful hybrid that is likely to have broad applicability within clinical psychology and enhance the utility of psychological assessments (Youngstrom, 2013).
The main limitations of this research concern the lack of standardized data on the psychological records.Two factors that may skew the results of this study in this sense are the infl uence of the interpretation of the professional who recorded the data and the fact that some information was left out of the psychological records.There are also limitations on parents' recollection regarding prior assessments and treatments.The possibility exists that some children have been misclassifi ed as not having undergone previous psychological assessment due to parents´ forgetfulness or ignorance.Possibly, parents who reported at the intake interview that their child had undergone psychological assessment are parents for whom the psychological assessment was important and constructive.This is an inherent problem of documental studies, in which research is carried out using secondary data.Yet due to the large number of subjects evaluated in this study, these limitations might not be interfering in the actual reliability of results.
This study also has retrospective and naturalistic methodologies own limitations.Designs like these do not allow a strict control of confounding factors.However, there are clinical advantages in adopting these research methodolo-gies, as they are conducted in the natural environment of psychotherapy, not in an environment deliberately created for research.This avoids the perceived artifi ciality of the fi ndings in prospective studies with strict control of bias, but it also constitutes an important limitation.

Conclusion
This research found that, in our sample, children who had undergone a psychological assessment before being referred to psychotherapy were more likely to adhere to their treatment and less likely to abandon it.
Campaigns aiming to disseminate to clinicians the importance of including psychological assessment of children before starting psychotherapy should be stimulated, for this practice would benefi t the patients as it reduces dropout rates.In line with several other countries in the world, professionals in Brazil should incorporate the practice of psychological assessment in the early stages of the psychotherapy of children.

Table 2
Distribution of the Sample as to How Psychotherapy Ended