Russian Medical Students Are Burning Out Faster Than They Did in 2020
- In a St. Petersburg medical institute cohort (n=152, 76.3% women), OLBI disengagement screened positive in 87.5% in 2024, up from 73.3% in 2020 (p=0.002).
- OLBI exhaustion screened positive in 90.1% in 2024, up from 80.0% in 2020 (p=0.012).
- GHQ-12 positive for minor mental disorders in 88.2% (vs 84.8% in 2020; p=0.389) — minor disorder rate stable, burnout rate is what moved.
- Self-reported stressors: education-related factors (83.6%), uncertainty about the future (72.4%), the broader geopolitical situation (73.0%); CAGE-positive alcohol risk stable at 20.4% (vs 20.0% in 2020).
This is one of very few methodologically comparable burnout repeat-cross-sections from a Russian medical training site, and it lands at an uncomfortable angle: the GHQ-12 rate of minor mental disorders barely moved between 2020 and 2024, but Oldenburg Burnout Inventory positivity jumped on both subscales with significance. Burnout in this cohort is decoupling from generic distress and tracking the work itself — training load, uncertainty about the future, and contextual stressors.
For clinicians who supervise residents, train psychotherapists, or take referrals from physician colleagues, the practical reading is that the people entering helping roles from Russian medical schools are arriving with a much heavier baseline of disengagement and exhaustion than the cohort five years ago. That is the population your future supervisees and your future medical referral partners are drawn from.
What the data shows
Same institute, same instruments (OLBI, GHQ-12, CAGE), same recruitment channel — June-July 2024 versus 2020. The OLBI shift is not a small drift: disengagement +14.2 percentage points, exhaustion +10.1 percentage points, both with p-values that survive any reasonable correction.
The CAGE result is informative in the other direction: problematic alcohol use risk did not increase. So the cohort is not coping by drinking more. They are coping by disengaging. Disengagement on the OLBI captures cognitive and emotional distance from the work itself — going through the motions, losing meaning, mentally pulling out. That is the precursor to leaving the profession, not the precursor to a drinking problem.
The stressor profile is also informative. Respondents named education load (83.6%), uncertainty about their own future (72.4%), and the broader geopolitical context (73.0%) as their top sources of stress, in roughly equal measure. None of those will be resolved by a wellness app.
For your practice
If you take supervisees from Russian medical programs or run a clinic that absorbs medical graduates as referral partners, expect higher baseline burnout in incoming cohorts and adjust how you onboard them. Specifically: ask about disengagement, not just exhaustion. The OLBI disengagement subscale is the more sensitive signal here, and clinicians tend to ask about tiredness while missing the loss-of-meaning marker.
For your own practice as a psychotherapist working with helping-profession clients: medical students and early-career physicians presenting now are likely to test positive on burnout screens even when their GHQ-12 is unremarkable. Treating the affective profile alone (sleep, mood, anxiety) without naming the burnout will miss the load-bearing problem. If you use a screening tool with this population, the OLBI is reasonable, brief, and has Russian-language validity in this setting.
For yourself, finally: the same trend almost certainly applies to practicing clinicians. The mechanism — chronic uncertainty, opaque future, stressors that have no negotiated solution — is identical. Notice the disengagement signal in yourself before you notice the exhaustion: dreading specific clients you used to like, going through documentation without reading it, ending sessions slightly early without a clinical reason. Those are the cheap indicators.
Burnout in this Russian medical-student cohort is decoupling from general distress and tracking the work itself — and your incoming supervisees are coming from this pool.
Single-institute sample (n=152), self-report screening rather than clinical diagnosis, and the 2020 comparison sits inside the early COVID-19 period — so the "baseline" was already elevated. The effect could be larger against a true pre-pandemic baseline.