Fig. 5: RESPOND-RCT Spain: Design.
The trial sample (N = 232) was predominantly female (n = 200, 86%), with an average 37.5 years of age (SD = 10.3) at baseline. Most participants had a university degree (n = 192, 82%) and were nurses (n = 130, 56%), physicians (n = 50, 22%), or nursing technicians (n = 29, 13%). The intervention group (top panel, n = 115) took part in a stepped-care program consisting of Doing What Matters (DWM) and, if distress continued to be present (score ≥16 on the Kessler Psychological Distress Scale (K10)86 five to 7 days after DWM), in Problem Management + (PM + ; n = 84 or 75% of participants in the intervention group). The control group (n = 117) received enhanced care as usual in the form of psychological first aid. Empty boxes illustrate no intervention. Stressor exposure (E), mental health problems (P), and positive appraisal style (PAS) were assessed in all participants at four time points (T0 to T3). Stressor assessment used instruments adapted to the specific population and context in prior qualitative work36,78, one list featuring three major life events and one list featuring six general, five pandemic-related, and four population-specific stressors (Supplementary Tables 11 and 12). Mental health assessment used the Patient Health Questionnaire-Anxiety and Depression Scale (PHQ-ADS)49, a composite measure of anxiety (GAD-7) and depression (PHQ-9) symptoms. PAS assessment in the RESPOND trial was restricted to the PASS-content instrument, which had shown significant SR associations in MARP and LORA. This scale also has the advantage that it directly targets the element in appraisal (appraisal contents, or outcomes) that is hypothesized to eventually determine stress responses, rather than antecedent cognitive processes leading to these contents (as in PASS-process)12. Also, social support and stress recovery were not assessed. Figure adjusted from ref. 41.
