Determinants of recovery: a cross-sectional analysis 91 Clinical recovery Clinical recovery was operationalized as the score on the client-rated Brief Symptom Inventory (BSI), whose 53 items are rated on a scale of 0–4. Total mean scores were used, with higher scores indicating that a client’s symptoms are more severe. The BSI has nine dimensions: psychoticism, depression, somatization, phobic anxiety, obsessive compulsiveness, interpersonal sensitivity, anxiety, hostility, and paranoid ideation (39–41). To reflect the degree of recovery, the scores were reversed: i.e. a higher total score represents a higher degree of clinical recovery. Functional recovery Functional recovery was operationalized as the score on the Social Functioning Scale (SFS), a self-administered questionnaire that consists of 76 items with varying response formats. A higher score indicates more, or a higher frequency of, competent behavior. All items are assigned to seven subscales: social engagement/withdrawal; interpersonal behavior; pro-social activities; recreation; independence competence; independence performance; employment/occupation. Each subscale score is the sum of all item values of that subscale, and all subscale values are standardized and normalized to a scaled score (Mean = 100, SD = 15). The full SFS scale score is computed as the mean of the scaled scores of the seven subscales (42–43). Personal recovery Personal recovery was operationalized as the score on the Mental Health Recovery Measure (MHRM), whose 30 items are rated on a scale of 0–4. Total mean scores were used, with higher scores indicating better personal recovery. The questionnaire consists of three subscales: self-empowerment, learning and new potentials, and spirituality (44–47). Statistical analyses First, Pearson correlations were computed for the five constituents of illness management we measured: insight, coping, social support, medication adherence, and problems with alcohol and drugs; and also for clinical, functional, and personal recovery. Three categories were used in our interpretation of correlations: weak (0.1, 0.3), moderate (0.3, 0.5) and strong (0.5) (48). To test our recovery-path model, we used SEM. The major advantages of this analysis are the ability to identify direct and indirect pathways and corresponding errors and to examine the associations among multiple independent and dependent variables simultaneously (29).
RkJQdWJsaXNoZXIy MjY0ODMw