A multiple-baseline design evaluation of the feasibility of a brief RNT-focused ACT intervention in health professionals experiencing burnout

Full reference: Criollo AB, Bernal González PA, Odriozola González P, & Ruiz FJ (2026). A multiple-baseline design evaluation of the feasibility of a brief RNT-focused ACT intervention in health professionals experiencing burnout. International Journal of Psychology & Psychological Therapy, 26, 2, 177-192.

Study type: Pilot feasibility study using a randomized, nonconcurrent, multiple-baseline single-case experimental design across participants (SCED).

Background and objectives

Burnout is a significant occupational health problem among healthcare workers, recognized by the WHO (2019) in the ICD-11 as a phenomenon arising from unsuccessfully managed chronic workplace stress. Following the classic conceptualization by Maslach and Jackson (1981), the syndrome comprises three dimensions: emotional exhaustion, cynicism or depersonalization, and reduced professional efficacy. Its consequences extend beyond individual suffering to affect the quality of patient care, absenteeism, and the retention of professionals. Prevalence is particularly high in Latin America, and the COVID-19 pandemic worsened the situation, with marked increases in emotional exhaustion, depersonalization, and even suicidal ideation and behaviors among healthcare staff.

The study positions Repetitive Negative Thinking (RNT) as one of the psychological processes that sustain burnout. RNT—the tendency to ruminate about the past and worry about future threats—is conceptualized as a transdiagnostic process linked to anxiety, depression, and stress, and in healthcare professionals it is associated with greater emotional exhaustion and difficulty recovering after demanding workdays. From a contextual-behavioral perspective, RNT is understood as a form of psychological inflexibility, and Acceptance and Commitment Therapy (ACT) emerges as an approach for promoting psychological flexibility, conceptualized from relational frame theory (RFT) as the ability to respond in hierarchy with a deictic "I" rather than in coordination with the immediate functions of private events.

The work is framed within RNT-focused ACT (Ruiz et al., 2016, 2020a, 2020b), an approach that explicitly integrates RNT and experiential avoidance, conceiving RNT (worry and rumination) as a predominant experiential avoidance strategy with a paradoxical effect: it prolongs and amplifies negative affect. Because most ACT protocols developed to date have been group-based and lengthy—limiting their applicability in high-workload settings—the aim of this pilot study was to evaluate the feasibility and preliminary effects of a brief, three-session RNT-focused ACT protocol, delivered individually via videoconference, in healthcare professionals experiencing burnout symptoms.

Method

Participants

Participants were recruited through social media advertisements. Inclusion criteria were: (a) at least six months in the current job without significant role changes, (b) age between 20 and 50 years, (c) at least three months of work-related emotional distress, (d) working at least 30 hours per week, and (e) clinically significant MBI-GS scores (Emotional Exhaustion ≥13, Cynicism ≥10, Professional Efficacy ≤26). Exclusion criteria included ongoing psychological or psychiatric treatment, scheduled contract termination within three months, planned vacations during the intervention or follow-up, high suicide risk, substance abuse, and chronic pain or fatigue.

Five female healthcare professionals met the inclusion criteria (P1 to P5). During baseline, one participant (P4) experienced an unexpected change in her work situation and, although she received the intervention for ethical reasons, her results are not reported; another participant (P5) withdrew during baseline due to increased work demands. Consequently, three participants (aged 22-26 years) completed the study: P1 (26 years, clinical psychologist, 40 h/week; GAD, agoraphobia, panic disorder), P2 (22 years, nursing assistant, 52 h/week; depression, panic disorder, social phobia, GAD), and P3 (24 years, psychologist, 72 h/week; GAD). All three met criteria for at least one anxiety or mood disorder according to the MINI interview.

Design

A randomized, nonconcurrent, multiple-baseline single-case experimental design across participants was implemented. The independent variable was the brief RNT-focused ACT protocol (three individual 60-minute videoconference sessions). Primary dependent variables were burnout symptoms; secondary variables were emotional symptoms, RNT, work-related psychological flexibility, and valued actions. Baselines lasted 4 to 7 weeks, randomly assigned (randomizer.org), and a four-week follow-up was included. Daily assessments were administered only on working days via the m-path app.

Intervention

The protocol consisted of three 60-minute RNT-focused ACT sessions, adapted from previous studies (Bautista et al., 2023; Ruiz et al., 2016, 2020a, 2020b) for occupational settings. It focused on identifying work-related RNT triggers, practicing defusion exercises, clarifying values, and promoting value-consistent actions. Session 1 established the intervention context, conducted a functional analysis using the "car metaphor," identified RNT triggers and experiential avoidance strategies and their consequences, and introduced a defusion exercise ("Concert Exercise"). Session 2 reviewed progress, practiced multiple defusion exercises ("Balloon" and "Transmilenio" exercises), and clarified values with the "Garden Metaphor," establishing committed actions. Session 3 reinforced RNT discrimination and perspective-taking ("Report" and "Where's Wally?" exercises), anticipated obstacles, and consolidated committed actions. The psychologist who delivered the intervention was trained through two sessions of protocol review and role-play, in addition to prior ACT training.

Outcome measures

The following were administered: the MINI (structured diagnostic interview); the B-EMA (Burnout-Ecological Momentary Assessment; 9 items, daily; Emotional Exhaustion, Cynicism, and Professional Efficacy subscales); the daily version of the PHQ-4 (4 items; depression and anxiety); the WRNT-EMA (Workplace Repetitive Negative Thinking-EMA; 3 items, daily); the MBI-GS (Maslach Burnout Inventory-General Survey; 16 items, weekly); the DASS-21 (21 items, weekly; emotional symptoms); and the PTQ (Perseverative Thinking Questionnaire; 15 items, weekly; RNT). The primary dependent variable was burnout symptoms assessed daily (B-EMA) and weekly (MBI-GS).

Data analysis

Analysis combined visual inspection with the nonparametric Tau-U statistic (Tau-U A vs. B, "b" approach; R package scan), which provides an effect size and a p-value resistant to autocorrelation. For between-case analysis, the design-comparable standardized mean difference (DC-SMD) by Hedges, Pustejovsky, and Shadish (2013) was used, analogous to Hedges' g, with an autoregressive factor (AR1) for serial dependence (R package scdhlm).

Results

For daily measures (Tau-U A vs. B), results were consistent with the visual analysis. P1 showed a statistically significant change only in emotional symptoms (PHQ-4 Tau-U = -0.51, p < .05). P2 showed significant reductions in exhaustion (-0.78), cynicism (-0.58), emotional symptoms (-0.85), and RNT (-0.68) (all p < .001), with no significant change in efficacy. P3 showed significant changes in all variables: exhaustion (-0.77), cynicism (-0.75), efficacy (+0.74), emotional symptoms (-0.80), and RNT (-0.69) (all p < .001).

For weekly measures, P1 showed significant improvements in all variables (exhaustion -0.87, cynicism -0.87, efficacy +0.78, DASS-Total -0.79, PTQ -0.94; all p < .05). P2 improved in exhaustion (-0.84), cynicism (-0.84), emotional symptoms (-0.81), and RNT (-0.89) (p < .05), and showed a marginally significant decrease in efficacy (-0.79, p = .05). P3 improved significantly in exhaustion (-0.87), cynicism (-0.94), DASS-Total (-0.80), and PTQ (-0.76), with no significant change in efficacy (+0.53, p = .12). Daily and weekly measures showed almost perfect agreement for P2 and P3; in P1, the change in emotional symptoms was significant only in daily measures, whereas weekly measures indicated improvements across all variables.

In the between-case analysis (DC-SMD), for daily measures the intervention produced large and significant effect sizes for exhaustion (g = -1.20; 95% CI [-1.88, -0.53]), cynicism (g = -0.91; [-1.47, -0.35]), and emotional symptoms (g = -0.89; [-1.78, -0.01]); a medium-to-large effect for RNT (g = -0.78; [-1.46, -0.10]); and a smaller but significant effect for efficacy (g = 0.48; [0.01, 0.94]). For weekly measures, effect sizes were larger for exhaustion (g = -1.98), emotional symptoms (g = -1.31), and RNT (g = -1.61), and smaller for efficacy (g = 0.12); the effect for cynicism (g = -1.15; [-2.53, 0.23]) did not reach statistical significance due to the lower measurement intensity of weekly assessments.

Discussion and conclusions

The authors conclude that the brief three-session intervention produced improvements in burnout symptoms, emotional distress, and RNT across the three participants, suggesting that brief interventions targeting RNT may be a viable option for healthcare professionals experiencing high occupational stress with limited time for traditional treatments. The reduction in RNT was one of the most consistent findings and likely contributed to the decreases in burnout and emotional distress, consistent with previous research on brief RNT-focused ACT protocols. Changes in psychological flexibility and valued actions were more heterogeneous and smaller in magnitude, and professional efficacy was the least responsive variable, possibly because it depends on broader workplace contextual factors.

Regarding practical implications, the fact that all three participants completed the protocol and showed clinically meaningful improvements suggests that brief, process-based interventions may help overcome common barriers to psychological care in this population (limited availability, emotional fatigue, demanding schedules), which is especially relevant in Latin American contexts.

The authors acknowledge several limitations: the small number of participants limits generalizability, compounded by the loss of two participants during baseline; the absence of an active control condition precludes separating the specific effects of RNT-focused ACT from nonspecific therapeutic factors; and repeated daily assessments may have produced reactive effects. Proposed future directions include expanding the sample, incorporating active control conditions, examining differences across healthcare professions, including diagnostic interviews at post-treatment and follow-up, and assessing work-related variables such as job satisfaction.

Relevance to RNT-focused ACT

This study extends the evidence base of RNT-focused ACT to a new problem and context—burnout in healthcare professionals—and to an especially parsimonious format: three individual sessions delivered via videoconference. It also contributes an intensive assessment methodology that combines daily ecological momentary measures with weekly instruments within a randomized multiple-baseline design, allowing close monitoring of change and examination of the consistency between measurement modalities. By showing that such a brief protocol can reduce burnout and emotional distress through the reduction of RNT, the work reinforces the role of RNT as a central therapeutic target and supports the continued development of brief, accessible, process-based interventions adapted to the demands of healthcare settings.

---