3.2 Quantitative Findings
Data from AAQ2, CD-RISC 10, and PSRS questionnaires met all necessary assumptions for ANOVA. Descriptive statistics are displayed in Table 2. JASP datafiles are provided on the Open Science Framework [14].
Table 2
Descriptive Statistics
|
|
|
|
Psychological Flexibility (AAQ2)
|
Total sample (N = 47)
|
Intervention
(n = 29)
|
Active Control
(n = 18)
|
T1 (M, SD, range)
|
24.98, 7.02, 11–42
|
25.83, 7.37, 13–42
|
23.61, 6.39, 11–37
|
T2 (M, SD, range)
|
21.47, 6.46, 8–34
|
21.76, 6.39, 9–34
|
21.00, 6.73, 8–34
|
T3 (M, SD, range)
|
20.77, 6.39, 7–39
|
19.52, 5.27, 9–30
|
22.78, 7.61, 7–39
|
Resilience (CD-RISC)
|
Total sample (N = 39)
|
Intervention
(n = 23)
|
Active Control
(n = 16)
|
T1 (M, SD, range)
|
27.82, 5.12, 17–39
|
28.26, 5.26, 18–39
|
27.19, 5.02, 17–35
|
T2 (M, SD, range)
|
30.54, 5.15, 18–40
|
31.17, 4.39, 23–40
|
29.63, 6.12, 18–40
|
T3 (M, SD, range)
|
29.08, 5.73, 14–39
|
30.30, 4.67, 20–39
|
27.31, 6.76, 14–36
|
Stress Reactivity (PSRS)
|
Total sample
(N = 47)
|
Intervention
(n = 29)
|
Active Control
(n = 18)
|
T1 (M, SD, range)
|
24.94, 6.45, 12–38
|
26.07, 6.78, 12–38
|
23.11, 5.57, 13–33
|
T2 (M, SD, range)
|
20.43, 7.12, 4–39
|
21.10, 7.46, 9–39
|
19.33, 6.61, 4–31
|
T3 (M, SD, range)
|
19.79, 6.67, 6–34
|
20.66, 6.72, 8–34
|
18.39, 6.52, 6–31
|
3.2.1 Primary Analyses
Psychological Flexibility (AAQ2)
ANOVA results revealed a significant interaction between time and group, F (2, 90) = 6.30, p = .003, n2p = 0.123, BF10 = 18.18. The BF10 value indicated strong evidence in support of this finding. For the ESRT group, psychological flexibility significantly improved by a mean of 4.07 points (15%) between T1 and T2, t (28) = 4.11, p = .001, d = 0.62, BF10 = 171.26. Between T2 and T3, scores improved by a further 2.24 points (10%), resulting in a total mean improvement of 6.31 points (24%) between T1 and T3, t (28) = 6.38, p = < .001, d = 0.96, BF10 = 4969.94. This was nearly one standard deviation from their baseline scores (SD = 7.73). Figure 2 illustrates mean scores across time, by comparison to active controls.
An ITT sensitivity analysis was performed by returning to the original sample of T1 participants (n = 73) and carrying the last observations forward. After carrying 16 T1 values and 24 T2 values forward, assumption checks were repeated. T1 data produced a significant Shapiro-Wilk value (p = .012), though both distribution and Q-Q plots indicated sufficient normality to continue analysis. The assumption of sphericity was violated and controlled for using Greenhouse-Geisser’s correction.
Congruent with the previous ANOVA, a significant interaction effect between time and group was observed, F (1.69, 119.98) = 4.86, p = .013, n2p = 0.064, BF10 = 6.12. Follow-up pairwise comparisons indicated a significant improvement in psychological flexibility among the ESRT group between T1 and T2, t (46) = 4.01, p = .001, d = 0.39, BF10 = 108.08, and between T1 and T3, t (46) = 5.87, p = < .001, d = 0.57, BF10 = 1838.00.
Resilience (CD-RISC 10)
ANOVA results revealed no significant interaction between time and group, F (2, 74) = 0.68, p = .51, n2p = 0.018, BF01 = 4.34. The Bayes factor indicated moderate evidence to support this finding.
Stress Reactivity (PSRS)
ANOVA results revealed no significant interaction between time and group, F (2, 90) = 0.24, p = .78, n2p = 0.005, BF01 = 6.57. The accompanying Bayes factor indicated moderate evidence in support of this finding.
3.2.2 Post-hoc Analyses
Median T1 resilience scores (Mdn = 28) were used to perform a median split on CD-RISC 10 survey responses. The median score converged with 25th percentile population scores (25th % = 29) [16]. Consequently, baseline resilience scores < 28 were coded as ‘low’ (n = 15) while values ≥ 28 were coded as ‘medium-high’ (n = 24). Baseline resilience scores were then added as a third independent variable of the ANOVA model.
Results for psychological flexibility indicated no significant three-way interaction between time, group, and baseline resilience, F (2, 70) = 3.21, p = .057, η²p = 0.084.
For resilience scores, a significant interaction between time and baseline resilience was observed, F (2, 70) = 6.70, p = .002, η²p = 0.16. Additionally, there was a significant three-way interaction between time, group and baseline resilience, F (2, 70) = 4.43, p = .015, η²p = 0.11. Subsequent t-tests revealed that participants with low baseline resilience scores who underwent ESRT exhibited no significant increase in resilience between T1 and T2, t (7) = 3.28, p = .11, d = 1.21, BF01 = 0.03. However, a significant increase was observed between T1 and T3, t (7) = 5.04, p = < .001, d = 1.86, BF10 = 12.72. By contrast, those with similarly low baseline resilience scores in the active control group showed no significant increase in resilience between T1 and T3, t (6) = 0.57, p = 1.00, d = 0.23, BF01 = 2.58. See Fig. 3 for a graphical comparison between low and medium-high baseline resilience groups.
Results for stress reactivity revealed a significant three-way interaction effect between time, group and baseline resilience, F (2, 70) = 5.06, p = .009, η²p = 0.126. Subsequent t-tests indicated that those with low baseline resilience scores who underwent ESRT displayed significant decreases in stress reactivity between T1 and T2, t (7) = -4.61, p = .001, d = 1.38, BF10 = 25.79. This effect was sustained between T1 and T3, t (7) = -5.26, p = < .001, d = 1.58 BF10 = 20.82. However, those with medium-high baseline resilience who underwent ESRT exhibited a non-significant decrease in stress reactivity between T1 and T2, t (14) = -2.75, p = .50, d = 0.60, BF01 = 0.33. A similar result was observed between T1 and T3, t (14) = -1.90, p = 1.00, d = 0.42, BF01 = 0.80. See Fig. 4 for a graphical comparison between the low and medium-high baseline resilience groups.
3.3 Qualitative Findings: Reflexive thematic Analysis
Six themes emerged through analysis of exit evaluations, interviews and focus group transcripts. Qualitative datafiles are provided on the Open Science Framework [14].
3.3.1 The “relentless” pressure of Medical Training
Students consistently reported distress among themselves and peers, which was attributed to individual and institutional factors, including the personality types of those studying medicine, the high demands of the curriculum, and a stigmatising culture within the medical profession. One student disclosed that her ongoing struggle culminated in “having to” contact her GP and begin antidepressant treatment. Upon reflection she noted, “it was quite clear that it was a mixture of different things. A big thing was my academic stress.” This sentiment was echoed throughout the sample, who admitted to struggling with the transition from undergraduate studies to graduate-level medical education. One student highlighted the pressure associated with the demand for constant knowledge acquisition:
I’m used to knowing lots [on] a subject but in medicine there's always [such a] vast amount, like you need to know so much. You always feel like you should be looking up something or checking something, like that feeling of not knowing is constant. (Student 3, focus group A1)
Beyond curricular pressures, several students agreed that the medical profession attracts a certain “type” of individual who is prone to competitiveness, self-criticism, and distress. One student noted the cohort’s competitive nature and concurrent maladaptive perfectionism: “Everyone is quite highly strung and quite competitive. There's … an undercurrent of competition, which is quite subtle, but it can be quite difficult to ignore.” Another student noted, “I don't think it's any secret that medics are quite… generally quite proud, quite competitive. I know a lot of people in my year just keep bashing out until they're completely burnt out.” Accompanying this competitiveness was the sense that they were just “scraping” a pass.
The curricular demands and accompanying stress led to a de-prioritisation of self-care activities and a “pushing onto the background all those things that you enjoy and all those things that are good for your mental health.” One student admitted, “I just wasn't looking after myself at all.” This de-prioritisation was seemingly perpetuated by feelings of guilt when performing leisure activities, particularly during weekdays. One student asserted that, as adult learners, “we know when to spend our time learning and when to spend our time, you know, revising… and we know we should spend our time on our mental health, or on our physical health.” Nevertheless, students ruminated over how ill-equipped they were to effectively manage their stress under such high demands.
3.3.2 A Particularly Difficult Time
Students continually acknowledged record low staffing as negatively impacting on their placement experience and as a contributing factor to anxiety around their impending career. There were claims that during placement, students were used to support service provision due to a disparity between the numbers of patients and the availability of staff. One student said, “it’s just a bit terrifying when you get to the point of actually being a doctor, because you know that”. Students alluded to the mistreatment of medical students on clinical placement. They reported conflict with senior staff and a fear of requesting time off for mental health concerns:
It's not necessarily an easy thing to do, because, you know, you don't know who you're talking to. You don't know if the person that you're delivering that information to is going to be receptive [.. .] you could just be called unprofessional. (Student 5, focus group A1)
One student claimed that the cultural taboo surrounding burnout and wellbeing was “perpetuated by this bitterness that [qualified doctors] don't have the tools.” Students also acknowledged the role of the COVID-19 pandemic in exacerbating distress:
I found last year so difficult, with COVID as well, I found I was getting quite upset quite easily and so I thought a bit of extra help might be good … because well you never know when things could crop up again. (Student B, interview)
Additionally, there was agreement among the sample that their curriculum was unable to perceive them as anything other than medical students, failing to allocate sufficient time to extracurricular commitments. Several students reported having to undertake paid work to finance their studies and felt this was not factored into placement allocations. Other students pointed to a generational anxiety relating to an economic downturn and cost-of-living crisis.
3.3.3 From Scepticism to Confidence in ESRT
In the face of this distress, pre-intervention scepticism of ESRT was widely reported. Students reflected on the ineffectiveness of their curriculum’s current ‘Preparation for Clinical Practice’ sessions, designed to manage burnout and prioritise wellbeing:
They're kind of like, ‘oh, yeah, we need more people to attend. This is gonna’ support you in how you deal with burnout’. But people aren't attending because it's not gonna’ be examined and everyone's burnt out trying to do all the stuff that they're trying to do at the moment. [.. .] And you're like, ‘okay, probably just better to stay at home, I'd be less burnt out’ [.. .]. (Student 6, focus group A1)
This sentiment echoed throughout the focus group, who expressed cynicism regarding wellbeing interventions. Reflecting further, they pointed to a lack of understanding and negative preconceptions about mindfulness, highlighting a potential source of the recruitment difficulties we faced in our study. Critically, however, students reported dramatic shifts in their appraisals of ESRT before and after the intervention. One student said, “I was sceptical at first, but it really makes a difference.” Another explained, “maybe I was a little bit sceptical before. But now that, like I've been through it, I think that I've taken more from it than I thought I was going to.” For several students, it was “seeing the benefit of how [they] felt after class” and “knowing that practicing can bring [them] clarity and stress-relief” that encouraged continuing practice.
3.3.4 ESRT Enhanced Stress Management
Students reported improvements to their stress management, which centred around a change in their relationship to stress. One student said, “I learnt a lot about myself and how best to deal with stress. I feel more able to deal with stress and be in touch with my emotions.” Some described a “mental frame shift”, while others reported a better “understanding” of stress. This ability to recognise stress allowed them to deal with it “before it becomes overwhelming”. Beyond learning to better identify their stress, students reported an ability to “step away from it” or, “put the stress more in proportion … with a more realistic view.” One student noted a lowered level of stress after just two weeks:
I found it starting to take effect a lot sooner than I thought it would have. Er, so it was a bit hard initially to get into the routine of the meditation, but I think it was about two and a half weeks in and I just felt like my baseline level of stress was a lot lower. (Student B, interview)
Others noted an improved ability to manage their physiological arousal: “I found it helps me to decrease my heart rate a little in stressful situations.” They also noted a decrease in maladaptive tendencies, with one student noting that “[getting] questions wrong in clinics and things doesn't play on my mind quite as much as it would have.”
3.3.5 ESRT Enhanced Known Resilience Factors
The majority of students looked forward to ESRT sessions and experienced positive changes in affect and motivation to practice ESRT skills. Positive changes included greater feelings of autonomy and confidence in their situation. One student who struggled with imposter syndrome and a sense of uncertainty over their place in medicine showed a renewed sense of confidence and commitment by the end of the intervention, stating, “I feel more confident in myself, I know what I’m doing, I know why I’m here. [.. .] I know now what I need to do if things get like that again.”
Students also reported an increased ability to stay present and engage with and manage their feelings, thoughts and emotions (i.e., psychological flexibility). One student explained:
I was able to separate my thoughts from my actions, and was able to appreciate what I was thinking and why I was thinking it, and I wasn’t scared when negative thoughts came, I was able to deal with them [.. .]. (Student D, interview).
Another reported:
I feel much more able to ground myself in the present and to relax after even stressful days on placement. I feel more able to be fully present in tutorials and placement learning opportunities such as clinics, which helps me to get more out of them and learn more. I also feel more resilient and am able to reflect on and act on criticism in a constructive way, and even when criticised in an unprofessional way I have been able to respond calmly with professionalism and move the conversation forward. (Anonymous exit evaluation)
One student described their increased sense of presence during their clinical skills practice: “I’m just taking it step by step and being more mindful, and not worrying about what's going to happen next, or what’s happened before.” This helped them to consciously decide on “the weight [they] give to thoughts”. Students’ descriptions hinted at an increased capacity for cognitive reappraisal, perceiving previously stressful circumstances as momentary challenges: “I've been far more just accepting of [things], and just, you know, deal with what comes up. And also, I just feel a bit stronger to deal with it, stronger in myself.” Additional improvements were identified in sleep, cognitive functioning, academic performance, general wellbeing, and interpersonal relationships. This included handling loss in the context of a relationship breakdown, and enhancements in active listening, communication, and candour regarding internal emotional states.
3.3.6 A Call for Integration
Despite the effectiveness of ESRT, students overwhelmingly voiced concern relating to their careers. While students appreciated the institutional efforts to raise awareness of wellbeing, the perceived lack of practical action left them feeling frustrated:
[.. .] they like to talk about self-care, because you see all these posters up in the break rooms and in the toilets and so, hopefully, that's a start that people are becoming aware of it, but I don't think we're actually seeing it yet. (Student 9, focus group B2)
One student called for a better “understanding of the limitations of enhanced resiliency, compared to dealing with [the] problems in the system”. Another argued that improving the wellbeing of students demands more systemic changes:
Probably on an individual level people can appreciate the benefit. But then it's just practically implementing that, and I think you kind of need the organisational backing to do that really, which isn't there at the minute from the NHS [National Health Service] and the med school. (Student 7, focus group B2)
One student pointed to the culture of medicine, which they believed to be at odds with prioritising self-care:
I think the whole culture around the NHS is based around you giving yourself up to help patients and that is reflected in the course and in everyone’s attitude generally. And that’s a culture shift that I don't think is going to change anytime soon. The NHS is built on the goodwill of people at their own detriment and I think that having self-care inside that narrative doesn't really match up. So I'm not surprised that organisations within medicine aren’t huge on promoting self-care as a priority. (Student 10, focus group B2)
Though ESRT was seen as a viable option, students suspected that recruitment challenges were due to a perceived lack of time. They acknowledged that stress management interventions are “definitely something that should be maintained” but that implementation creates a “burden” forcing them to “give up some of that valuable time in order to do something that they don't know is necessarily going to actually work for them.” One student explained, “the time to be able to do it is just really not there” because many students have already “dropped [extracurricular activities] because they can't keep up with extra things alongside their workload”. Consequently, a consensus emerged that ESRT should be integrated into their academic curriculum:
It would be good to actually see integration of this in course time, where the [university] understands that it actually is part of what makes you a better doctor rather than ‘oh, this is cool, but do it in your own time’. (Student 8, focus group B1)
The most prominent feature of student evaluations was a feeling that “if it was integrated into the course, [it] would be a lot more accessible.” Additionally, the abbreviated nature of ESRT was considered appropriate for integration, making it “quite suitable, because it is manageable.” As one student explained, “if people get benefit just from that little bit then I definitely think it’s useful… you know, something obviously needs to be done because people are struggling right now. That goes without saying.” Students generally agreed that this relatively small time-commitment to the development of stress management could be practically implemented into the curriculum, significantly improving their capacity to cope with the demands of their training.