(July 2026) It’s no secret that involuntary unemployment is not good for one’s mental health. It can have obvious financial impacts, plus psychological ones such as how one feels about their social status, and the large gaps it can leave in people’s daily routines.
For some, this kind of distress ends up being treated as a biomedical issue. Some take issue with pharmacological interventions for this kind of thing, given that unemployment is typically a social problem borne of structural issues. The concern is that framing it as a medical problem risks pathologising what is, in many cases, a normal and understandable response to a genuinely difficult situation.
Not everyone experiencing unemployment-related distress is prescribed medication for it. Who does is somewhat determined by cultural attitudes, economic incentives and the institutions people happen to pass through, such as their doctor, their country’s healthcare system, and what treatment is available and affordable to them.
People with less education consistently face higher unemployment rates and remain unemployed for longer periods of time. Education may also affect how people emotionally process being unemployed.
The age group forming the focus of SHARE, individuals aged 50 and above, is used to a settled career, where losing a job can carry different consequences than it would earlier in life. Older unemployed individuals often find it harder to get back into work – instances of age discrimination are well-documented, skills can become outdated as technology moves on, and health is not what it was. All of this can work to make unemployment-related distress worse.
Sociologist Lisa Colman and colleagues from Ghent University in Belgium used SHARE data to study which unemployed respondents aged 50 to 67 were using medication for depression or anxiety. They tracked over 14,000 people across 27 European countries, using six waves of SHARE data collected every two years. They looked at people's employment status at one interview, then checked two years later whether they'd started taking medication at least once a week for depression or anxiety.
Rather than just asking "are unemployed people more likely to take this medication," the researchers wanted to know something more specific: does losing your job predict a jump in medication use beyond what you'd expect just from people feeling worse? To check this, they built people’s reported health complaints and depression symptom scores into the model, then looked at whether unemployment still predicted extra medication over and above that.
Losing a job predicts more medication, even after accounting for how people say they feel
The main result: those who moved from employment into unemployment were substantially more likely to start medication for depression or anxiety two years later, even when the researchers accounted for how much distress or ill health people actually reported at the earlier interview. People stuck in ongoing unemployment also showed higher medication use, though that link weakened once existing symptoms were factored in.
One detail stood out: going back into work did not produce an immediate drop in medication use. In other words, the effect of unemployment on medication was easier to switch on than switch off.
The pattern looked different for the most educated
Here’s where it gets interesting. On average, people with a university-level education used less medication for depression or anxiety than people with less schooling. That part fits the usual pattern. But when the researchers looked specifically at people experiencing ongoing unemployment, that advantage disappeared. In fact, highly educated people who stayed unemployed were the group most likely to be taking medication, more so than similarly unemployed people with less education.
It's tempting to read this as "losing a job later in life is more distressing for highly educated people", but that's not actually what the data shows. Medication use isn't the same thing as distress. What the researchers found is a gap in who ends up medicated, not a measured gap in who's suffering more.
Two different explanations are floating around for why that gap in treatment might exist, and this study can gesture toward both without cleanly testing either.
The first is about expectations, not biology. For someone who spent decades in education and built a career on the promise of stable, secure work, staying unemployed in your late 50s or early 60s may be a bigger break from what life was "supposed" to look like — more disorienting, more likely to feel like personal failure rather than bad luck. The findings do not indicate that highly educated people experience more distress; rather, similar distress may be more readily interpreted as an individual issue, making biomedical treatment more likely.
This fits with earlier research findings that people who lost jobs through clear-cut causes, like a factory closing, actually used less medication than other unemployed people despite similar distress — suggesting that when there's an obvious outside reason to blame, people are less likely to turn the distress inward.
The second is more mundane: access. More educated people tend to be more comfortable navigating healthcare symptoms: they're often better at describing distress in the kind of language a doctor recognises as clinical, and they tend to have easier access to professional care in the first place. The study findings offer a clue here: people who saw a doctor more often were substantially more likely to be on medication, regardless of unemployment status. So part of what's going on may simply be that more educated people are more likely to end up in a doctor's office talking about how they're feeling, and once they're there, distress is more likely to get turned into a prescription.
Bear in mind this is not a proven cause-and-effect story
The researchers are upfront that their findings are correlational only. They can't fully rule out whether some underlying trait (like a general tendency to seek medical care, or a pre-existing but unmeasured vulnerability) explains both a person's job loss and their medication use. The medication measure is also fairly blunt: it captures self-reported weekly use, not diagnosis, dosage, or prescribing context. And unemployment itself is measured only every two years, so it cannot distinguish recent job loss from longer unemployment spells that happened to fall inside that window.
In all this is a well-powered, cross-national, longitudinal association study. It raises a good, well-evidence question: is later life unemployment being treated as a medical problem when it’s really a social one? Ultimately, answering that properly would need data that can tell the difference, like medical records that shed a light on physician decision-making itself.
Interested in conducting your own research on unemployment in later life using SHARE data? CLICK HERE.
Article: Colman, L., Bracke, P., & Delaruelle, K. (2026). Transitioning to unemployment and medication use for depression or anxiety among older adults: educational stratification in medicalisation processes. Aging & Mental Health. https://doi.org/10.1080/13607863.2026.2680262
Picture via Pixabay
