| The lack of treatment for depression patients usually brings about chronic, periodic, and negative effects including family and social dysfunction, physical illness, suicidal ideation and suicidal behavior. In addition to medical treatment and traditional psychotherapy, along with the development of computer science, the new modes of computerized treatment for depression have been explored by an increasing number of researchers, which renders the area to be a frontier and hot topic in psychotherapy research in the West.This study reviews the psychological theories of depression treatment and treatment-related empirical research at home and abroad, as well as Acceptance and Commitment Therapy (ACT) and computerized treatment. It then uses meta-analysis to systematically investigate the effects and influencing factors of computerized treatment for depression; and develops an ACT-based computerized treatment system, and examines the effectsof the treatment and latent class of the applicable clients; and then, with the latent growth curve modeling (LGM), the study investigates the mechanism of ACT-based computerized treatment for depression. The main body of this dissertation consists of three parts.Part I uses meta-analysis and meta-regression techniques to investigate the effectiveness and influential factors of computerized treatment for depression. Forty independent samples from30RCT researches in32literatures meets the criterion of meta-analysis (N=4795). What the result shows are as follows.(1) Computerized treatment for depression is effective and its overall effect is close to a medium effect size, and the effect size is small in the teenager group and large (i.e. up to the medium effect size) in the adult group. The effect size of3-month follow-up is0.2and that of6-month follow-up is0.15.(2) Subgroup analysis finds that after the treatment the measured value of the self-reported depression levels of all subgroups dropped significantly, but the aggregate effect size of the subgroups varies. There is marked difference in effect sizes in the analysis of such subgroups as severity of depression, recruitment settings, supporting systems and measurement scales. The effect size of the subgroup of major depression, community samples. Email support and using other measurement scales is significantly larger than that of the other subgroups. In addition, publication year, drop rate and the number of intervene units have a certain degree of effect on the effect size of treatment, but since there might exist interaction among the variables such conclusion ought to be considered cautiously.(3) Although there might be publication bias in this study, it is difficult to demolish current conclusions.Part II develops an ACT-based computerized treatment system and examines the effects of the treatment and latent class of the applicable clients. Participants are recruited via campus billboard and screened by their BDI scores, resulting182depressive participants entering the following procedure. They are randomly distributed into treatment group (N=95) and control group (N=87). The treatment group completes6self-help treatment units, with one unit each week, and then follow-up outcome after3month are assessed. What the result shows are as follows.(1) The ACT-based computerized treatment can effectively relieve depression. Compared with the waiting group, the ACT self-help group displays a more significant decline in depression level measured after treatment as well as a larger effect size which remains a medium effect size in the3-month follow-up survey.(2) The ACT-based computerized treatment can also enhance the clients’positive mental health level. The positive mental health level enhancement displayed by the ACT self-help group is greater than that of the waiting group and the former has a medium effect size. The overall positive mental health level in3-month follow-up shows little difference, but the emotional well-being (d=0.55) and the psychological well-being (d=0.54) remain to show medium effect size.(3) The latent profile analysis shows that the eight client features, which include depressionprevalence, extraversion, introversion, resistance traits, personality, biology, childhood and relationships, can be grouped into3latent classes, i.e. high psychological reaction group (47%), low psychological reaction group (37%) and endogenous depression group (16%).(4) The clinical change analysis shows that the high psychological reaction group displays the highest cure rate, the endogenous depression group displays the lowest one, and the endogenous depression group shows the highest rate in reliable change without full recovery. There is no significant difference in the rate of absence of reliable changes among the above three groups. And the result remains the same in the follow-up measurement after3months.Part Ⅲ applies LGM to the investigation of the mechanism of ACT-based computerized depression treatment, and mainly analyzes the mediating effect of experience acceptance and cognitive change. Participants are recruited via campus billboard and screened by their BDI scores, resulting124self-assessed depressive participants. A random controlled trail are conducted (treatment group N=63. control group N=61). The treatment group completes6self-help treatment units. What the result shows are as follows.(1) In the course of ACT-based computerized depression treatment, acceptance and action, cognitive defusion, and dysfunctional attitude all undergo significant changes, and yet the change in automatic thoughts is not noticeable. The mediating effect of experiential avoidance is significant, but that of cognitive change is partially significant.(2) Acceptance and action, avoidance and fusion, cognitive defusion, and dysfunctional attitude display a significant mediating effect between the treatment and the changes of depressive symptoms; that is to say, the mechanism of ACT-based computerized treatment lies in the changes in acceptance and action, avoidance and fusion, cognitive defusion, and dysfunctional attitude, and these changes therefore affect the depression level.(3) Acceptance and action, avoidance and fusion, and dysfunctional attitude display a significant mediating effect between the treatment and the positive mental health enhancement; that is to say, the mechanism of ACT-based computerized treatment also lies in the changes in acceptance and action, avoidance and fusion, and dysfunctional attitude, and these changes then affect the positive mental health level.The comprehensive discussion part emphasizes that the interactivity and intelligence of programming should be improved technologically, and that individualized treatment should be enhanced. In addition, researchers should continue to examine the treatment effects of this approach in other groups and to further explore the mechanism of ACT-based computerized treatment for depression. |
PDF Full Text Request