Maria survived the worst week of her life in the ICU. Three months later, she’s home — but she can’t get out of bed before noon. Her doctor says her depression score is high. But Maria doesn’t feel sad. She just feels exhausted.
She’s not alone. Millions of people recovering from severe COVID-19 face lasting fatigue, brain fog, and low mood. Many are told they’re depressed. But what if the test is getting it wrong?
Post-intensive care syndrome (PICS) affects up to half of ICU survivors. It brings physical, mental, and emotional challenges. Fatigue is one of the most common symptoms. So is low mood. But telling them apart is harder than it sounds.
Doctors may be mistaking tiredness for depression.
For years, clinics have used the PHQ-9 — a nine-question survey — to check for depression. Two of those questions ask about tiredness and sleep. But those same symptoms are hallmarks of fatigue.
It’s like using a thermometer to measure both fever and room temperature — you can’t tell which heat source you’re seeing.
The test that mixes up symptoms
When fatigue and depression share symptoms, the PHQ-9 can’t always tell the difference. A patient might score high not because they’re deeply sad, but because they’re worn out.
That could lead to antidepressants when what they really need is rest, rehab, or energy management.
But here’s the twist: researchers may have found a fix — and it doesn’t require a new test.
They looked at 82 people one year after surviving severe COVID-19. All had been in the hospital, many in the ICU. Each took two standard surveys — one for depression (PHQ-9), one for fatigue (FACIT-Fatigue).
Nearly two-thirds had severe fatigue. About 16% scored high for moderate to severe depression.
The numbers showed something striking: fatigue and depression scores were tightly linked. But when researchers removed the two fatigue-related questions — about tiredness and sleep — the link weakened.
A smarter way to screen
The team used a statistical model to see how fatigue was affecting the depression score. They found fatigue was directly boosting answers to the “tired” and “sleep” questions — not because patients were more depressed, but because they were more worn out.
When they recalculated depression scores without those two items (using the PHQ-7), the numbers dropped. For fatigued patients, the median score fell from 7 to 4.5 — a big difference in how care might be guided.
Even better: the shorter two-question version (PHQ-2), which skips fatigue-heavy items, showed a weaker link to fatigue. That means it may better reflect true emotional distress.
This doesn't mean this treatment is available yet.
It means doctors could start today using a different part of the same tool to avoid misreading symptoms.
Experts say this isn’t about dismissing patients’ feelings. It’s about listening more clearly.
“If we’re not careful, we risk treating the wrong problem,” said one researcher involved in similar work, noting that fatigue management looks very different from depression care.
For patients, that could mean fewer medications, more targeted rehab, and better support for what’s really going on.
Right now, many long COVID clinics already screen for both fatigue and mood. But few adjust how they interpret depression scores based on fatigue levels.
This study suggests they should.
But there's a catch.
The study had only 82 people. All were from one national cohort in Canada. Most were white and male. That means results might look different in other groups.
Also, this was a snapshot in time — one look at patients a year after hospitalization. We don’t yet know how this plays out over time.
Still, the message is clear: one size does not fit all when screening mental health after critical illness.
The next step? Larger studies across diverse groups. Researchers want to test whether using the PHQ-2 first — then adding more questions only if needed — leads to better outcomes.
Some clinics may start making changes now. Others will wait for more proof.
Either way, the goal is the same: care that sees the whole person — not just the symptoms on a checklist.