Browsing by Author "Osorio Thomas, Marcela"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Content validity & initial reliability of the rating instrument for therapy contract setting revised at UC (RITCS-UC)
    (2019) Osorio Thomas, Marcela; Behn Berliner, Alex Joseph; Pontificia Universidad Católica de Chile. Escuela de Psicología
    Personality Disorders (PDs) are serious, weakening and complex clinical presentations, with patients having substantially reduced life expectancy and elevated mortality of four times compared to the general population. This occurs especially in young individuals, who have 10-fold increased mortality. Often, PD patients are known as poly-consultant users of the public mental health system, increasing mental health costs and leaving patients without relief. Thus, it is important to study treatments for PDs to provide patients with appropriate care. Among PDs, Borderline Personality Disorder (BPD) has received the most scientific attention because it can be considered as a prototype for PDs in general, expressing a general factor that explains the phenotypic covariance among all them. The unique characteristic of BPD would be that it represents the impairment to maintain coherent images of self and others, a core dimension underlying the ten PDs, which include impairments in self and interpersonal functioning. Even though evidence-based BPD treatments stem from divergent theoretical backgrounds, all of them share a common characteristic, namely they are all very structured treatments, including explicit therapeutic contracts and therapeutic frames. This is especially important for Transference Focused Psychotherapy (TFP), which hypothesis is that active work around ‘treatment contract’ from the beginning of treatment is a key active ingredient driving patient improvement. However, there is no empirical evidence to demonstrate it. The main reason for this gap in the literature is a deficit in instrumentation: there are no psychometric valid instruments to measure the establishment of ‘treatment contract’. So, this study aimed at developing a content & psychometric valid instrument to evaluate contracting: the Rating Instrument for Therapy Contract Setting (RITCS) revised at UC (RITCS-UC). Based on the TFP clinical guide and on an early version of the RITCS instrument, the construction of RITCS-UC included content validation through an iterative process of expert consensus among expert panels from the U.S. and Chile. The RITCS-UC includes three content dimensions of contracting: therapist`s adherence to TFP clinical guide, therapist’s competence establishing a ‘treatment contract’ and also, a measure of the agreement between therapist and patient. To test confiability, four psychologists were trained to evaluate the establishment of four contracting videotaped sessions using the RITCS-UC. Then, data was analyzed using Generalizability Theory to establish inter-rater reliability estimates. A Generalizability Study (G_Study) and a Decision Study (D_Study) for each dimension were used to disentangle different sources of error and to determine the most efficient measurement design. Results of G_Studies showed relative generalizability coefficients (G_coef) of .64 for dimension adherence, with a relative standard error (SE) of 0.32. For dimension competence G_Studies showed a relative G coef of .87, with a relative SE of .25. And, for dimension agreement G_Studies showed a relative G coef of .81, with a relative SE of .25. The principal source of error in all dimensions was the interaction between dyads and raters (D x R), anexpected result which confirms that the source of error didn´t come from the items but from the subjectivity of D and R. Results of D_Studies showed possible optimizations in the number of items, raters, and dyads, and showed the items for future revision. Finally, and following the Standards for Educational and Psychological Testing, the development of the RITCS-UC has been completed and it is ready to be used in group and research contexts accounting the observations made.