Brief comparison of Clinical Opiate Withdrawal Scale (COWS) and Subjective Opiate Withdrawal Scale (SOWS) for the type of opioid withdrawal we are most likely to encounter in palliative care—particularly when opioids are reduced relatively quickly because of clinical decompensation, altered mental status, sedation, or worsening weakness.

• COWS is generally the better primary tool when the patient can participate reliably. Although clinician-administered, it is not completely objective; several items still depend on patient report, cooperation, or attributing symptoms specifically to withdrawal.
• SOWS is entirely patient-reported. It can help describe subjective symptom burden in an alert, attentive patient, but it should not be used when delirium, aphasia, sedation, or unreliable responses prevent meaningful self-report.
  • SOWS is also somewhat less well matched to palliative patients with iatrogenic physical dependence because one of its 16-item asks about craving or feeling like using opioids.
• In a delirious, nonverbal, or profoundly weak patient, I would not force a COWS total by scoring unassessable items as zero. Profound weakness or hypoactive delirium may suppress restlessness, tremor, or other observable findings and create a falsely reassuring score.
• Neither scale has been validated specifically in clinically decompensating palliative patients. Tachycardia, sweating, agitation, nausea, diarrhea, and pain may be caused by sepsis, hypoxia, delirium, bowel regimens, or uncontrolled symptoms—or may be masked by medications such as beta-blockers, alpha-2 agonists, sedatives, antiemetics, and antidiarrheals.

MDCalc COWS calculator is a reasonable and convenient option:

https://www.mdcalc.com/calc/1985/cows-score-opiate-withdrawal
Note: It reproduces the published item anchors and avoids manual addition errors. The limitation is interpretation rather than calculation: much of the accompanying MDCalc guidance is oriented toward buprenorphine induction and opioid use disorder, so those treatment thresholds should not automatically be applied our patients receiving palliative care.

Because opioid withdrawal may coexist with clinical decompensation, and the withdrawal scales cannot distinguish etiology.  When documenting a COWS assessment, I would consider including the following:

• The prior opioid regimen and exact dose or route change.
• When the change occurred and whether delayed withdrawal is plausible.
• Whether the patient could participate reliably.
• Items that were unassessable or potentially confounded.
• The serial pattern rather than relying on one isolated score.