The sickest patients got the therapy, not because it failed them
In a health system reaching toward wholeness, researchers discovered that the patients most drawn to music therapy were also the most burdened by illness—a finding that reframes what it means to offer care and to whom. A study from University Hospitals Connor Whole Health found that surgical patients with mental health or substance use disorders who received music therapy stayed 41.8% longer in the hospital, not because the therapy harmed them, but because the sickest patients were the ones being referred. The work raises a quiet but consequential question: when a healing intervention gravitates toward those already most fragile, how do we distinguish recognition from remedy?
- Surgical patients with mental health and substance use disorders already face higher pain, worse mortality, and elevated addiction risk—making every treatment decision carry unusual weight.
- Music therapy patients turned out to be measurably sicker than those who received standard care, carrying more anxiety, trauma, heart failure, and prior opioid exposure before therapy even began.
- The 41.8% longer hospital stays alarmed at first glance, but the data revealed they reflected pre-existing medical complexity, not a failure of the intervention itself.
- Patients received only a median of three sessions covering 18% of their hospital days—a dosage too sparse to produce the hoped-for reduction in opioid use.
- Researchers now argue that music therapy referrals must become more precise, targeting specific diagnoses rather than being offered broadly across all surgical patients with mental health conditions.
Researchers at University Hospitals Connor Whole Health set out to examine whether music therapy could ease the surgical experience for patients with mental health or substance use disorders—a population already known to face worse pain, longer recoveries, and higher addiction risk. What they found complicated any simple answer.
Patients who received music therapy stayed in the hospital 41.8% longer than those who did not. But the study, led by Sam Rodgers-Melnick, revealed that this gap was not evidence of failure. The music therapy group was simply sicker from the start—more likely to have anxiety disorders, trauma-related conditions, heart failure, and prior opioid exposure. Many were receiving palliative care. The therapy had drifted, as if by instinct, toward the most medically complex patients in the room.
This reframing matters. The longer stays reflected the patients' underlying burden, not the therapy's effect. The study is the first to document this gravitational pattern within a health system—music therapy being offered disproportionately to those whose complexity would independently extend their hospitalization.
A second concern emerged alongside the first. Patients received a median of three sessions across only 18% of their hospital days. That frequency was not enough to meaningfully reduce opioid use—one of the primary reasons health systems have turned to music therapy in the first place. Kristi Artz, Vice President of Connor Whole Health, acknowledged the tension, noting that understanding which patient factors shape outcomes is essential before any intervention can deliver its intended impact.
The researchers stopped short of dismissing music therapy. Instead, they called for more precise referral decisions—targeting patients with specific diagnoses like anxiety or trauma-related disorders, and asking whether more frequent sessions might finally move the needle on opioid use. The study does not close the door on music therapy's promise. It asks, with care, that the door be opened more deliberately.
Researchers at University Hospitals Connor Whole Health set out to understand whether music therapy could help surgical patients with mental health or substance use disorders—a vulnerable population known to suffer worse pain, longer recoveries, and higher risks of addiction. What they found was more complicated than a simple endorsement: patients who received music therapy did stay in the hospital 41.8% longer than those who did not, but not because the therapy failed them. Rather, the patients who got music therapy were sicker to begin with.
The study, led by Sam Rodgers-Melnick and supported by the National Center for Complementary & Integrative Health, examined electronic health records from University Hospitals to compare surgical patients with mental health or substance use disorders. Some received between two and ten music therapy sessions during their stay; others received standard care. After accounting for basic demographic differences, the researchers found that the music therapy group carried a heavier disease burden. They were more likely to have anxiety disorders, trauma-related conditions, and heart failure. They had already been exposed to opioids within the first 48 hours of admission at higher rates. Many were receiving palliative care. In other words, they were the sickest patients in the room.
This matters because it reframes what the data actually shows. The longer hospital stays were not a sign that music therapy failed—they reflected the reality that music therapy was being given to patients whose medical complexity would naturally extend their time in the hospital. The study is the first to document this pattern: music therapy, as currently deployed in this health system, gravitates toward patients with greater comorbidity burden, a factor that independently drives prolonged stays.
But there was another finding that gave the researchers pause. Patients received a median of three music therapy sessions spread across only 18% of their hospital days. That sparse dosage did not produce a meaningful reduction in opioid use. This is significant because one of the main reasons health systems have turned to music therapy is to reduce reliance on opioids for pain management—a critical goal given the addiction risks these patients already face. The therapy, as delivered, was not frequent or sustained enough to move that needle.
Kristi Artz, Vice President of University Hospitals Connor Whole Health, acknowledged the tension in the findings. "Improving patient outcomes including hospital length of stay and opioid exposure for chronic pain are important targets for whole health interventions," she said. "A greater understanding of patient factors which may influence outcomes is necessary for these interventions to have the desired impact." In plainer terms: knowing who benefits most from music therapy matters as much as offering it.
The researchers were careful to note that their findings should not be read as a verdict against music therapy itself. Rather, they suggest that referral decisions need to be more precise. Instead of offering music therapy broadly to all surgical patients with mental health conditions, clinicians might target it toward those with specific diagnoses—anxiety, trauma-related disorders, heart failure—where the evidence suggests it could make a difference. The study also hints at a dosage question: if three sessions across three weeks does not reduce opioid use, perhaps more frequent or sustained intervention would. Or perhaps music therapy works best as one tool among many, not as a standalone solution.
For patients with mental health and substance use disorders, the stakes are real. They experience higher postoperative pain, longer recoveries, worse mortality rates, and elevated risk of developing opioid use disorder. Any intervention that could ease their path through surgery and recovery deserves serious study. This research does not diminish that need. It simply clarifies that good intentions and available interventions do not always align with the patients who need them most—and that sometimes the sickest patients end up in the treatment group not because they are being helped more, but because they are being recognized as sicker.
Notable Quotes
A greater understanding of patient factors which may influence outcomes is necessary for these interventions to have the desired impact— Kristi Artz, Vice President of University Hospitals Connor Whole Health
The Hearth Conversation Another angle on the story
So the study found that music therapy patients stayed longer in the hospital. Doesn't that suggest the therapy didn't work?
Not quite. The patients who got music therapy were already more medically complex—more anxiety, more trauma, more heart failure. Those conditions alone make hospital stays longer. The therapy didn't cause the longer stay; sicker patients got the therapy.
Then why give them music therapy at all if it didn't reduce opioid use?
That's the real question the study raises. They got an average of three sessions spread across less than a fifth of their hospital days. That's a thin dose. We don't know if more frequent therapy would have worked.
So this is a dosage problem, not a therapy problem?
Partly. But it's also a targeting problem. The therapy went to the sickest patients, which makes sense clinically, but it means we can't easily separate what music therapy did from what their underlying conditions did.
What should hospitals do differently?
The researchers suggest being more selective. Target music therapy toward specific conditions—anxiety, trauma, heart failure—where there's evidence it helps. And maybe increase the frequency. Three sessions isn't much of a commitment.
And if they do that?
Then we might actually see whether music therapy can reduce opioid use in this population. Right now, we just don't know.