The COWS Score: How Helpful Is It?


COWS stands for Clinical Opioid Withdrawal Scale, and it’s probably the most commonly used tool to determine the degree of opioid withdrawal experienced by the patient. The scale has eleven items related to opioid withdrawal. Some are subjective, like the question about the degree of anxiety or irritability the patient is feeling. Some items are strictly objective, such as pupil size and pulse rate. And some are sort of a combination of objective and subjective, like the question asking about both nausea and vomiting. The patient may report nausea and score points on the scale, and if the patient vomits, this scores more points.

I’ve worked in clinics that used the COWS for each dose increase, and I’ve worked in clinics that didn’t use the COWS at all.

I think it’s a good tool, but has some drawbacks. I use it during dose induction, particularly on a patient new to medication-assisted treatment. Sometimes patients aren’t sure how they’re “supposed” to feel on replacement medication, and a COWS score gives me a better idea of how much withdrawal they are in.

For example, I had a patient who felt much fatigue in the evenings. He’s been on the program about a month, and had been dosing at 70mg for about a week. He worked at a strenuous job, and got off work around 5pm. One day, he told the nurses that he needed an increase, since it felt like his methadone “gave out” as soon as he got home, and he had to take a nap before his evening meal because he was so sleepy. When the nurses heard him say “sleepy,” they correctly became worried he was on too much methadone, and sent him to see me. When I checked him just before dosing the next morning, his pupils were a wide 8 mm and reacted briskly to the bright light I shone in his eyes. He was in withdrawal and he felt better after a few dose increases. His use of the word sleepy was confusing, since to us, we worry “sleepy” means “headed towards a methadone overdose.”

Sometimes, a patient reports severe withdrawal but doesn’t score very high on the COWS. I don’t assume the patient is lying, because some patients don’t tolerate withdrawal symptoms easily. More commonly, I see patients, mostly long-term users, who are in what I would consider to be moderate or severe withdrawal by their COWS score, but they experience it as “not so bad, I’ve felt worse”

In another example, I had a patient on 110mg who reported terrible withdrawal, to the point she couldn’t function during the day. She was restless, anxious, jittery, and felt like her heart was racing. She wasn’t sleeping well. This was puzzling, since a month ago she’d been fine on that same dose. There were no new medications, no change in activities, and she wasn’t drinking alcohol (a common reason for drop in methadone blood level). On the COWS, she scored an 8, but when I looked at the actual COWS, she scored very high on the more subjective items, yet her pupils were pinpoint and her pulse rate in the 60’s

The more we talked, the more I suspected anxiety as the cause of her symptoms. She had a terribly stressful living situation. She was saving money to move out on her own, but felt like she had to endure the circumstances in the short term. In this case, she appeared to be blaming opioid withdrawal for her symptoms of anxiety, and anxiety was a normal response for what she was experiencing. She didn’t need a higher dose of methadone; she needed someone to help her think of better immediate options for safe housing.

I do not think a COWS score is helpful for fine-tuning a patient’s dose of methadone. Many times the COWS score doesn’t pick up subtle withdrawal, so I don’t tend to use it for higher dose changes.

COWS scores are helpful when defending one’s self from regulatory bodies. About five years ago, a state investigator took me to task for authorizing dose increases. “You just believe them when they say they’re in withdrawal?” she asked sarcastically. The investigator didn’t think I should increase the doses of those patients, and yet the studies clearly show methadone patients have better outcomes if they are on an adequate dose. By doing a COWS score, the patient’s signs and symptoms are recorded in the chart for an investigator to see.

In summary, the COWS scale is a useful tool, though probably more useful at lower doses. Like all tools, it’s helpful in some situations, but it’s not perfect. It should be used alongside our other tools, like talking and listening to our patients both before and after dosing, using blood levels in rare cases, and always asking about other medications or new medical problems.

12 responses to this post.

  1. Posted by kevin on May 25, 2014 at 1:25 pm

    I’ve been told that peek and trough is not a perfect tool for adjusting a patient’s methadone too. when my medical director opens the chart when someone needs an increase, without looking at the peek and trough or cow score, when he looks at the current dose and notices the patient is already on or above 120 (obviously from the last medical director) he has already formed an opinion and becomes very judgmental about the rest of everything he has to look at. Kinda like each milligram comes off his paycheck. My counselor actually had to appeal my last turn down and beg him to increase me. This was almost 6 months ago. I still feal that my dose wears off too soon. I use receive 50mg more than what I am on now at the last clinic I was in. I’m not saying I want that dose again. I just want to be on the right amount that covers me through the day. But why would a clinic hire a dr that has never prescribed methadone and had to go get certified b4 he/she could start. I wish I could afford to come to your practice Dr. Burson. You are so kind and caring and you seem to always do the right thing even if it’s not doing what the patient wants. I see you being the senior director over a company of clinics one day


  2. Posted by Gareth Joseph on May 25, 2014 at 1:35 pm

    Here in the UK even when on a treatment programme you are very lucky to have access to increases/decreases for example after messing up my treatment plan whilst reducing from 10mg to 6 mg I went back to my prescriber a psychiatric Dr Who teaches other junior Drs & med students I was prescribed 8mg daily supervised by the pharmacy and sent away with a two week prescription and a nurse appointment for 14 days later yet everything I read on any bup forum seems to be completely at odds with this practice sounds to me like the Uk treatment providers are not using the drugs as they are meant and care little for patient discomfort oh heres a magic pill now go away and I will see you when I can fit you in surely this is not how it should work no wonder the failure rate is so high fo relapsing back to old behaviours if those charged with helping are failing you also addiction is hard enough to beat as it is!


  3. Posted by Jessica Bielecki on May 25, 2014 at 2:19 pm

    Great article!!!!


  4. Posted by Alan Wartenberg on May 25, 2014 at 6:51 pm

    The COWS is meant for ACUTE withdrawal only. Once someone is on longer term methadone or suboxone, ALL their withdrawal is only subjective, and they will NEVER show any objective signs, so it is not useful. I am now retired, but requiring 4 hour post dose assessments would at least tell me that the patient is not being over sedated and looking for that specifically in their requests for increased doses.


    • Thank you Dr. Wartenberg! World-renowned in this field, you should know. Yes, I often like for the patient to return to the OTP in 3-5 hrs post-dose so I can assess for sedation.
      I’ve long suspected COWS isn’t useful for established patients…but regulatory agencies say it’s essential for assessment. For example, I’d like to write an order for a patient I’m evaluating who is at 85mg, and still describing evening and nighttime withdrawal: “May increase by 5 mg q 4 d until 105mg then return to see M.D.” But officials say I allow the nurses to practice medicine unless I add something like: “…for COWS greater than 7″… Do you have advice for rebuttal of regulatory agencies’ regulations?
      And thank you for reading.


      • Posted by Alan Wartenberg MD on August 15, 2015 at 3:30 pm

        Sorry, I never checked to see if there was a reply!! The COWS can actually work, as long as you don’t get hung up on the discordance between the subjective and objective findings, since most of the time there will be adequate subjective complaints to bring the score to greater than 7 (or whatever cut-off one chooses).

  5. Posted by Carlos on May 27, 2014 at 7:50 pm

    Fairly sure that is better than nothing. But too many of this so called tools and test for assessment too frequently have never been tested for validity and reliability.

    Although I agree that there should be some sort of tool to measure the level of Opiate Dependence. Too frequently we deny we are playing God with patients, but any time that we are guessing and do not let the patient know that this is a guess or a hunch and not necessarily has been measure for accuracy, I say we are playing God. Some researchers have called it Authoritative Pronouncements compared to valid and reliably tested instrument. This is a tool because I say it is and nothing more. Because I am good and I should know better.

    I very alarmed by the “NEW and IMPROVED” Diagnostics and Statistical Manual V (five) of the American Psychiatric Association. They seem to be getting worse rather than improving on a system. After reviewing specially the area of Substance Dependence. The inconsistencies and reliability is getting worse. If you are honest and really read for example the Diagnostic Criteria and or impairments for which there are 11 of them. Neither a patient that has been clean for a few weeks, patients on methadone or suboxone do not meat any of the 11 criteria. For patients are no longer impaired by the conditions stated there. Then they use a term like Remission as it the word will fixes the situation. The word itself is meaningless. It implies that the patient do not need treatment because the impairments are not longer existing. But any diagnosis always implied that there still something going wrong with the patient.

    I have never seen a dentist for example use the diagnosis the patient has a cavity on remission. Given that there is always a good chance that most of us will get a cavity in the future.

    If it was up to me and psychiatrist would be honest the DSM book should take the word STATISTIC from its title. The DSM system has never ever been very much dependent on statistic or even science to deserve a dominant place in the title. It give the user and the public the impression that the DSM has a strong foundation in science when it does not.

    When the first DSM book was created published I believe there were only 7 (seven) diagnosis in the book. With the DSM 5 I believe it has gone beyond 396 diagnosis. Are we as a society getting sicker? or is it that clinicians in the field need patients and need to create doubt in the community that we are a bunch of wimps that required constant psychiatric care. Anyone including the psychiatrist that write those diagnosis can and have been diagnosed. Any of the390+ diagnosis will fit almost everyone in the human race. This implies that there is no one healthy in this planet.

    And people are believing that this is so. I do not know how many times I have heard people claim that “we are all a little crazy” or some similar statement. Some of these diagnosis are just definitions or characteristics and part of being human. I think that is a very dangerous precedence.

    Although I agree that with better measuring devices are needed to treat patients. I also agree that we need to be a lot more ethical and honest with ourselves and patients. That perhaps this is the best that we can come up with, but stop giving the impression. If I was to search for any research on this instruments I would most like find nothing, but a lot of discussion.

    Too frequently I have found that when the studies are done, the outcomes usually show that the opposite is true. I see at least that there is some discussion on the matter. Measuring instruments are suppose to be foremost free from bias and personal opinions and preferences.

    Nevertheless, it is a good article and as I say before we are least discussing it. I suppose in the long run and if we continue in this direction it should litimedice medical assisted treatment. And may create less “controversy” from people who criticize it but have not the faintest idea of its application and science. Just because I do not like it for myself, it does not mean it is a good source for recovery.


  6. Posted by William Taylor, MD on May 27, 2014 at 11:46 pm

    Love the picture; I agree that COWS is about as discriminating as the pictured bovine.


  7. Posted by Diana on October 30, 2015 at 11:05 pm

    What does it mean to score 1 on the Cows chart?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: