And so I'm very excited to introduce Dr. DeFilippis, and he is a licensed clinical psychologist with more than 30 years experience as a clinician, researcher, and educator, predominantly in the fields of addiction treatment. He earned his PhD in clinical psychology in 1992 from Hahnemann University in Philadelphia. He currently serves as the Deputy National Mental Health Director for Substance Use Disorders in the Office of Mental Health and Suicide Prevention in the United States Department of Veterans Affairs. His particular interests and expertise in expanding access to evidence-based substance use disorder treatment and measurement-based SUD care. He's a nationally recognized subject matter expert in contingency management, whose work in implementing contingency management has been published in peer-reviewed scientific journals and featured in media reports in the New York Times, Washington Post, Scientific American, and NPR. He's a trainer in motivational interviewing and cognitive behavioral therapy for substance use disorders. Dr. DeFilippis also is a member of the Motivational Interviewing Network for Trainees. And I'm grateful that he was able to take some time to join us for this lecture today. And thank you, Dr. DeFilippis. Thank you so much, Dr. Blaise. I wanna thank you and Dr. Stifler, Beth, and Nicholas for inviting me to give this talk on contingency management. I was mentioning earlier that I consider myself a bit of an evangelist for contingency management. And I'm delighted anytime I have an opportunity to share our experiences in VA with contingency management and encourage and hopefully build some passion to make contingency management available outside VA, where obviously it's so sorely needed both in and outside VA. What I'm hoping by the end of today's presentation that you'll be able to do is identify the challenges to recovery from substance use disorder, because that's the milieu or the landscape into which we're gonna be introducing contingency management. Then being able to differentiate both the rationale and methods of contingency management, the various applications of it to substance use disorder treatment. Number four, very important, being able to refute common critiques of CM that I think are born largely of good faith, but I think of a misunderstanding of CM, a concern out of issues like the moral hazard that CM might pose. And then of course, is it all worth the effort? Is the juice worth the squeeze? Describing the evidence supporting the effectiveness of CM in the treatment of substance use disorder. So let's begin with this milieu into which we'd be introducing contingency management, the challenges that our patients face in their recovery efforts from substance use disorder. Now, at one level, there's a neurophysiological challenge that patients face. We know that substances in particular stimulants for which CM has the strongest evidence of efficacy, act on the dopaminergic systems of the brain. And with chronic exposure, you get dysfunction in the dopaminergic areas of the brain. These PET scan images come from the work of Dr. Nora Volkow at the National Institute on Drug Abuse. And what they demonstrate is that indeed, there is evidence that chronic exposure to stimulants manifested in patients with stimulant use disorder causes a dysfunction in the dopaminergic areas of the brain. Now, as a psychologist who usually, especially a behaviorally oriented psychologist, I'm not quite scenarian in that I wouldn't say I ignore inside the black box as it were. But from my perspective, the question is, well, what does that mean experientially when you have dysfunction in the dopaminergic areas of the brain? Well, what it means experientially is that activities that would otherwise comprise a fulfilling life without substances, enjoying a good meal, spending time with loved ones, work, volunteering, hobbies, reading a book are not as appealing any longer because they are not associated with sufficient dopaminergic function to maintain their consistent performance. As a result, the pursuit of the substance displaces other pursuits in the patient's motivational hierarchy. Now, we add to that a time challenge. Now, the time challenge actually is a double-sided coin because on the one hand, we need to be able to establish sustained recovery, sustained abstinence in order to promote a recovery of dopaminergic function, which is the good news story. These PET scans indicate that with sustained abstinence, you get a return to near normal, near baseline functioning in the dopaminergic system. Of course, the challenge is, gosh, how do we help patients sustain ongoing abstinence? Well, if that wasn't enough of a challenge, add to that a treatment attrition challenge that anyone who's worked with patients struggling with substance use disorder is aware of. Now, we know that substance use disorders are chronic illnesses, and as such, they respond best to continuous treatment that, of course, may wax and wane in intensity, but is not episodic. Unfortunately, the experience of most patients with substance use disorder treatment is episodic. We see attrition rates as high as 70%, which clearly, after just four sessions of treatment, by the way, which clearly is insufficient to support that sustained recovery that can allow for the restoration of dopaminergic function so that a patient can experience as reinforcing, as satisfying again, activities that will be fulfilling in a sober lifestyle. Add to that the concern that in this episodic experience of treatment, patients often return in subsequent episodes in a clinical state more severe than their prior baseline and need higher cost services, more intense services that are more intrusive in their lives and more difficult for the patient to maintain engagement with. And to continue that rather daunting story, there's a cognitive behavioral challenge. And I think, though it's cliched, I think the image of the patient with the devil on one shoulder and the angel on another is very apropos in explaining this cognitive behavioral challenge. Because on the one hand, the patient with a substance use disorder is being whispered to by the devil of substance use disorder or substance use saying, I can make you feel real good right now. I could take away your pain right now. I could give you energy right now. Oh, by the way, down the road, you're more likely to die. Your finances will crater. Your relationships will suffer. But right now you're gonna feel real good. And then on the other hand, you have recovery. The angel saying, oh, hang in there. If you could stick with your recovery plan, life will likely get better. Your health will improve. You'll be less likely to die. Your finances will have an opportunity to improve as will your relationships. But oh, by the way, in the immediate aftermath of your decision to pursue recovery, you're probably gonna feel kind of awful. You're gonna have a clear-eyed lucid view of the devastation in your life brought on by substance use disorder. You're going to be likely to have to cease associations and relationships that were immediately reinforcing, albeit ultimately harmful due to their association with substance use. Well, CM embraces that challenge that immediacy of reinforcement is crucial to sustaining, to initiating and sustaining a behavior. So the challenge, therefore, is to make recovery immediately reinforcing so the patient has a fighting chance of choosing a recovery behavior as an alternative to substance consumption. Because we know that immediacy of consequences is a force multiplier of magnitude, the size of the consequence. Modest consequences delivered with immediacy are very influential on behavior. Don't take my word for it. Let's switch to a more common experience of substance use. People who have used or struggled with the use of tobacco. You'd be hard-pressed to find a person who's either used tobacco in the past or is currently using tobacco, who is unaware of the negative health consequences of tobacco. Heck, every single tobacco product carries a warning, attesting to those risks. So why isn't that terribly influential? Because those consequences, though high magnitude, death, emphysema, chronic lung disease, they're potent, but they are neither certain nor are they immediate. Therefore, they are not terribly influential on behavior. What is influential? The immediate consequence of using the tobacco, likewise with other substances. Now, a good exemplar of this challenge is methamphetamine and the crisis that we're seeing throughout the world, in the United States, certainly, with the methamphetamine use. Now, this slide comes from Knight and I think it really spells out nicely this challenge between immediate appealing consequences versus high potency, but not terribly influential longer-term consequences. So with methamphetamine, we're talking about immediate reinforcement, reduced fatigue, euphoria, a rush, increased attention. Down the road, addiction, changes in brain structure as evidenced by those PET scans from Dr. Volkow's work, mood disturbances, dental problems, and on and on and on. But the immediacy of the reinforcing effects of methamphetamine use make that behavior very, very appealing, very seductive. Well, contingency management says that's the landscape we're dealing with. There's the challenge, the set of challenges, the set of challenges. How do we go about meeting those challenges? Well, CM does so by bringing immediate reliable reinforcement for engaging in recovery behaviors, most notably abstinence verified by immediately available drug test results. But CM can also be used to reinforce other recovery activities like treatment attendance, medication adherence, even recovery activities like completion of therapy homework assignments. The challenge with some of those behaviors, and I'll get into more detail later, is it's more difficult to objectively monitor those behaviors than it is abstinence. With abstinence, we're lucky. We have a very reliable and valid means of determining its presence or absence, toxicology testing. With something like homework assignments, how do we define completion of the homework assignment? Say it's a CBT, a Cognitive Behavioral Therapy for Substance Use Disorder homework assignment on tracking one's craving episodes. Well, is it the number of episodes tracked? Is it the details provided in the homework? Is it the pithiness of the response? It becomes challenging to identify what constitutes completion versus non-completion. But regardless, when we provide reinforcement for engaging in a recovery behavior, we are in essence providing de facto reinforcement for retention and treatment, because in order for the patient to secure the reinforcement for the abstinence, or for the medication adherence, or for the homework assignments, they have to stay engaged in treatment to do so. Now, the root, the foundation of this entire discussion harkens back to the work of B.F. Skinner. That's a picture of B.F. Skinner in his younger years. And his work in studying operant behaviors, behaviors that operate on the environment, cued by stimuli, and followed by consequences that either make them more or less likely to recur. And the study of operant behavior that Skinner and other researchers following him used is the operant chamber, with either a rat or typically a pigeon in an operant chamber where the environment was exquisitely controlled. There was one behavior, either pressing a lever or pecking a key, if we're talking about a pigeon, in order to secure reinforcement. Now you might say, Dom, you lost me. How in the world can concepts and principles elucidated by the study of animals like rats and pigeons in an exquisitely controlled environment like an operant chamber, that by the way, came to bear B.F. Skinner's name, it became known as the Skinner box. How could that be applied to the human behavioral experience? That's a great question. And I would posit that obviously humans don't live in such exquisitely controlled environments. So what we do to recreate that control in CM is assiduous adherence to the procedures of contingency management. That crafts the virtual Skinner box, if you will, for the patient where we establish the lever that the patient will press for reinforcement, abstinence, completion of homework, medication adherence, resulting in the delivery of reinforcement. Now that might sound easy conceptually. Okay, the patient performs a behavior that we've agreed on. I provide reinforcement. Pretty straightforward. And I agree with you. Conceptually, contingency management, operant conditioning is very straightforward. But the devil is in the procedural details. In fact, as Skinner himself stated, the way you do positive reinforcement is more important than the amount. Now that is a profound statement. What he's saying there is the procedures of CM are more important than the size of the reward you provide. It almost sounds hard to believe that procedures somehow are more important than size of reward. But thank goodness, BF Skinner was absolutely right. Because if the alternative was true, if it was magnitude alone, that was the key consideration, I can't even imagine what level of reinforcement we would have to provide patients to displace cocaine or methamphetamine use. In fact, methamphetamine use, there were two researchers out of Italy, and of course, I'm blocking on their names in the publication, that looked at various orders of magnitude of reinforcement of various activities. Methamphetamine was orders of magnitude higher than any other activity, even cocaine. If we had to reinforce patients at a magnitude equivalent to methamphetamines reinforcing properties, I can't imagine tens of thousands of dollars it would take, it would make CM practically impossible. But because Skinner is right, and procedures are more important than magnitude, modest amounts of reinforcement delivered with immediacy, the force multiplier, thank you, it's Dicchiari, thank you, Chris. That procedural approach of providing modest reinforcement with immediacy can indeed displace potently reinforced behaviors like methamphetamine consumption. This is the most important slide I'm gonna present you today. In this slide are the active ingredients, the operant conditioning ingredients of contingency management. Now, our first order of business, of course, if we're going to apply contingency management to a patient's recovery support, our first order of business is, what recovery behavior do we wanna reinforce? And we need to be able to specify that behavior specifically and objectively. Thankfully, in the treatment of substance use disorder, we have a behavior that is, for the most part, quintessentially associated with recovery, abstinence. Now you might say, hold on a second, Dom, what about harm reduction? Hold that thought if I could ask you to. I will make the case later that CM, even CM reinforcing abstinence is wholly consistent with a harm reduction approach to recovery. We'll get to that later. So if we decide on abstinence as our target behavior, our next order of business is, how do we know if it occurred or not? Again, as I mentioned a short while ago, we have a very reliable and valid means of doing so with urine toxicology testing. The key is the toxicology testing regime we use must yield results immediately. Lab testing is not going to be adequate for contingency management because by the time the sample is sent to the lab and the lab processes the sample for testing and the results are entered and made available for the clinical encounter, that's an eternity in CM and delays in CM are like kryptonite to Superman. Delays destroy contingency management effectiveness. But when we can determine with immediacy whether that abstinence occurred, that is the patient tests negative for the target substance, typically stimulants. And when I say stimulants, I'm really talking about the three major stimulants, cocaine, methamphetamine, and amphetamine. The patient must test negative for all three of those stimulants. Really no need to test for the more esoteric stimulants, your methylphenidates and other stimulants, not to suggest that those aren't problematic. There's a problematic use, but by and large, when you target the three major classes of stimulants, you can effectively implement CM. We provide, when the patient is abstinent, immediate, tangible, desirable reinforcement. But we don't stop there, because what you see with that third bullet point is the application, the very straightforward application of positive reinforcement. The patient engages in the behavior, receives a reinforcing consequence, a reward. But we're not gonna stop there. We also want to promote consistent performance of that behavior, because remember the lesson from Dr. Volkow's work. We need consistent abstinence to promote the brain healing necessary for restoration of dopaminergic function. We do that. We differentially and simultaneously reinforce consistency of abstinence by escalating the size of the reward for consistent performance of the behavior. What that means, in essence, is for the first urine sample the patient submits that tests negative for stimulants, the patient gets X amount of reward. For the second row, say 2X. For the third row, 3X, and so forth. So that there's a differential value from the patient's perspective of being consistently abstinent. But we don't stop at positive reinforcement. In fact, we don't stop at escalating positive reinforcement. We add a second ingredient from operant conditioning, extinction. When the patient is not abstinent, that is when he or she does not test negative, we withhold reinforcement. That's the classic textbook definition of extinction. Why do we use extinction? Because extinction is the best learning contingency for eliminating a behavior. Again, don't take my word for it. We've all experienced extinction in a variety of different behaviors. Have you ever gone, the one we've all, I'm sure, experienced, have you ever gone to a vending machine, unbeknownst to you that vending machine was out of order, you put money in, you hit the button for a snack or a drink, and the machine is out of order. You might hit the change return a couple of times. You might even engage in what we might call an extinction burst. You'll slam the side of the machine a couple of times, but you abandon the behavior relatively quickly because you know reinforcement is not forthcoming. That's the power of extinction. So we've got positive reinforcement, strengthening the abstinence behavior. We've got extinction, weakening the drug taking behavior, but we don't stop there. I'm beginning to sound like one of those infomercials, but there's more. With four easy payments, no, we'll get to that later. What we also do with contingency management is we introduce a mild punisher. That is when the patient is not abstinent, when the patient does not test negative for stimulants, not only do they not receive anything, that's the extinction, but they also have the size of the reinforcement that they had escalated up to with consistent abstinence reset back down to the starting amount. That's a penalty. It's a punisher, textbook definition of punishment. So now we have positive reinforcement, strengthening the abstinence. Extinction and punishment, weakening the drug taking behavior, and by introducing punishment, we actually leverage a fourth active ingredient from operant conditioning, negative reinforcement. Think of positive and negative reinforcement not as good and bad, and certainly negative reinforcement is not the same as punishment. When we talk about positive and negative reinforcement, think of them arithmetically, mathematically. Positive reinforcement is reinforcement by addition. I add something to the patient, I give them a tangible reward. Negative reinforcement is reinforcement by avoidance. In other words, I avoid or escape from an unpleasant situation by engaging in a behavior. And negative reinforcement is a potent contingency for learning. Certainly you see this manifest very palpably with the struggles of patients with opioid use disorder, especially for whom the first order of business, the chief motivator every day when one is physically dependent on opioids is to avoid getting sick. And the pursuit of opioids first and foremost is a negatively reinforced behavior. Well, in this case with CM, the negative reinforcement is I avoid the reset. So we've got four active ingredients in CM, all foundational operant conditioning principles, two strengthening recovery, positive reinforcement and negative reinforcement, two weakening drug taking, extinction and punishment. So what behaviors do we choose to reinforce? Now, technically this is operant conditioning. Theoretically, any volitional behavior can be brought under contingency management control. But as I mentioned, the devil is in the procedural details. How do we define and monitor the behavior? Our key considerations, even something like attendance. You might say, Dom, wait a minute. Attendance is self-evident. The patient shows up, you reinforce them. If they don't show up, you don't reinforce them. How about if the patient shows up and promptly falls asleep in group? Did they attend? Should we reinforce them? What if they were disruptive in group? Should we reinforce? They did attend, but not exactly in a manner that we would want to strengthen. So even a seemingly straightforward behavior like attendance can get complicated in a contingency management paradigm. Now, another important consideration, of course, is the target behavior has to be something the patient is capable of performing. Otherwise, it becomes an exercise in learned helplessness. Whereas we keep dangling the carrot, but the patient's unable to achieve the carrot. So it has to be something the patient can do. And we know that while some patients struggle to maintain consistent abstinence, at least an episodic period of abstinence is by and large achievable for a wide array of patients. We're talking about literally two to three days of abstinence from stimulants in order to test negative at least once to receive reinforcement. Now, medication adherence is another behavior that seemingly is straightforward. The patient consumed the medication or did not. But the problem is monitoring medication adherence, especially with self-administered medications. How do you do that? Do you use blister packs? Well, they're not foolproof. Do you use medication event monitoring caps, MEMS caps? Well, they're not foolproof either. A patient could open and close a cap and record those openings with the cap, but not take the medication. We could try biomarkers, but in that case, the patient still has to appear for treatment. We get the sample, either urine or blood to verify presence of the medication. With medication adherence, the gold standard now is to have the patient video record themselves consuming the medication and sending the video into a provider who verifies it and provides immediate reinforcement. Obviously that becomes a logistic challenge. What are some implementation concerns when it comes to making CM available? Well, some of them are procedural. First, a procedural concern is which model of contingency management should we use? Now, the earlier version of CM, the one that was studied, in fact, the one born of the work of Nate Azrin and the early work that essentially evolved into community reinforcement and later explicated even further by Steve Higgins at University of Vermont is voucher CM, where the patient earns prescribed and escalating dollar values of reinforcement for consistent performance of the behavior. So in other words, for the first sample negative, $5. For the second sample negative, it increases by a set amount, say a 250, so $7.50, then $10, then $12.50 and so on and so forth. So the patient and provider both know exactly how much the patient will earn for each sample. Now that model of CM, voucher CM, extremely effective. In fact, both the voucher and price systems are comparably effective. CM even developed by the late great Nancy Petrie because the studies of voucher CM, especially the early studies, indicated that a 12-week course of CM, the typical length of a CM course, costs anywhere from 800 to $1,200 in incentives per patient. Now you might say, Tom, you're making that sound like a lot of money. $1,200 to arrest the stimulant use disorder? That's cheap. Let's do it. And as a fellow clinician, I'd agree with you, but I'd also say good luck implementing that, especially in publicly funded programs that don't have $1,200 per patient to set aside for incentives. So voucher CM was this terrifically effective treatment that remained largely unavailable except in clinical trials. Well, those of us, and I was both a supervisor and provider of voucher CM, our response to the lamentations of this is too expensive to make happen. Our response to that was, well, in essence, too bad. It's the Jerry Maguire principle. If you want the patient's behavior, you got to show the patient the money. And the lamentations were much like what Scotty would say to Captain Kirk when Captain Kirk said, I need more war power. Well, I cannot change the laws of physics, Captain. Well, we would say, you can't change the law of effect. You got to pay for the behavior. Well, we thought we were all so clever until a true genius by the name of Nancy Petrie came along who said, hold on a second. I have a way that we can reduce the costs markedly without sacrificing efficacy. Now, that statement was borderline heresy to a behaviorist. Wait a minute, you're going to get the same behavior for lower magnitude reinforcement? Nah, can't happen. Well, like any true genius, Nancy Petrie saw the answer hiding in plain sight. So with Pri-CM, came to be known as Fishbowl-CM, the magic is that instead of the patient earning prescribed and escalating dollar values of reinforcement, the patient earns prescribed and escalating opportunities for higher value reinforcement. Those opportunities coming in the form of draws of prize slips, typically from a container like a plastic fishbowl, which contains prize slips of various denominations. By introducing probability in the determination of magnitude, Nancy Petrie was able to reduce the cost of CM without sacrificing efficacy. All right, so we could choose either Prize or Voucher-CM, they're comparably effective. Wait, shouldn't we all choose Prize? It's lower cost? Nope, it's not that easy. There are some circumstances where you may want to use Voucher instead of Pri-CM, because it's frankly easier to implement. Instead of having to bring out a fishbowl every time and have the patient make draws, you know by prescribed escalation exactly what the patient's going to earn. Target drug. Hey, you mentioned, Don, that it was stimulants. Why not total abstinence? Isn't that the healthiest goal for patients is to abstain from all substances? Yeah, for the most part, I'd agree that that is the healthiest goal, but total abstinence is an extraordinarily complex behavior. And what do we know about extraordinarily complex behaviors? The best way of helping build them is through successive approximations by shaping those target behaviors. So instead of reinforcing a patient for total abstinence, we reinforce an approximation of it, abstinence from the target, in this case, stimulants. The other reason we don't target total abstinence is imagine this scenario. A patient has an excellent weekend, didn't use any methamphetamine, cocaine, or amphetamine. They've abstained successfully, but they indulged in some cannabis use over that weekend. That patient would know, come Monday at the CM session, I'm not gonna get anything. My effective successful abstinence from methamphetamine and cocaine is not gonna earn me any reward. In essence, what we would be doing is putting those successful abstinence behaviors through extinction. The patient would have engaged in them and received nothing for it. So both theoretically and empirically, we know that targeting a drug class rather than total abstinence yields better results in CM. So what are the most common targets? Stimulants, the biggest body of literature is supportive. In fact, both AAAPs and VAs practice guidelines for treatment of substance use disorder, stimulant use disorder. Identify CM as the most effective treatment for stimulant use disorder. Cannabis can also be targeted. It's a little more challenging because of the lengthy detection window and how residual molecules of THC can cause positive testing even among patients who abstain. They require some procedural adjustments that are more complicated to implement than when one targets stimulants. Alcohol, shout out to Dr. Michael McDonald at Washington State University. He's doing some terrific work on using ethyl glucuronide as a metabolite for determining alcohol abstinence. The problem with clinical implementation is Dr. McDonald uses analyzers, desktop analyzers to determine ETG presence. Those analyzers are kind of expensive, upwards of tens of thousands of dollars, not really practical for wide-scale clinical use. But however, as soon as a point-of-care test cup or dipstick is available for ETG that is effective, the ones that had been on market in the past were not terribly accurate, we can target alcohol. Well, what about opioids? One of the great elephants in the room. We're in the midst of an overdose crisis largely driven by opioids. Well, the truth of the matter is that CM, there's a signal for CM effectiveness for targeting opioids, but it's not nearly as effective as the frontline life-saving medications for opioid use disorder, methadone, buprenorphine, buprenorphine combination products, and extended release naltrexone. Those are the frontline treatments. And bear in mind, the challenge with using CM for opioid abstinence is a successful patient would actually raise their risk of fatal overdose. Why? If I am in CM for opioids, I'm abstaining from them and earning reinforcement. If I'm not receiving those medications, my tolerance is dropping, and were I to experience a recurrence of use, my reduced tolerance puts me at risk of an overdose. Well, why not do the CM with the medications on board? Well, it gets complicated. How do you differentiate medicinal from illicit methadone, medicinal from illicit buprenorphine? So things get complicated there. The other complication, even if we were able to overcome those challenges, is fentanyl. And frankly, nowadays, the fentanyl analogs and the nidazines, we don't have tests, effective rapid tests for fentanyl, fentanyl analogs and nidazines. And if you cannot surveil fentanyl in a CM targeting opioids, don't even try. It's obviously so prevalent that if you're blind to it, you can't do CM. We use toxicology testing with immediate results, hence the value, the importance of point-of-care tests. And who administers CM? This is a good news story. CM is not discipline-specific. Medically trained providers, psychologists, social workers, addiction therapists, nurses, all can administer CM, as long as the provider can assiduously adhere to the procedures of CM, they can effectively make CM available. Type of reward in terms of procedural challenge. Well, in VA, we're very fortunate. In fact, I like to say that VA is the best setting of all to make CM available because we have an internal legal tender called the Veterans Canteen Service Coupon System. These coupons are not what you would think a coupon, you would clip out of a flyer or a Sunday newspaper to get a discount on a product. These coupons are legal tender for the purchase of goods from any VA canteen or retail store, coffee shop or cafeteria throughout the VA enterprise. And those canteens, those retail stores are like mini big box stores. They merchandise everything from sticks of gum up to flat screen televisions and tablets, computer tablets. So they're terrific variety of reinforcement that the veteran can use his or her coupons to purchase. Now, outside of VA, it gets more challenging because you have to resource the incentives. And currently there is no billing code for CM. So the type of reward typically is gonna be merchandise or gift cards that you have to resource through grants or other mechanisms. For example, even SAMHSA's State Opioid Response Grants and Tribal Opioid Response Grants set a cap on the amount of the grant that could be spent on an individual patient CM, $75. It's an insufficient amount for contingency management. The studies on CM in terms of magnitude that are at least a decade old now say for voucher CM, a minimum reinforcement over 12 weeks should be $500. It's likely far greater than that now. And with Price CM, it was $200 to $300. In fact, we've increased the prices in VA recently to accommodate for our inflationary period that we're in. Clearly $75 is insufficient for abstinence reinforcement. In fact, Nancy Petrie herself tested years ago, 20, oh, I think it was 2014, comparing, in fact, the paper was called CM, how low can you go? $200 CM versus an $80 CM. The $80 CM, no results, equivalent to treatment as usual, no effect of the CM whatsoever. So clearly $75 would be insufficient. Frequency of sessions. Yes, David. Yes. I don't think, I have a question that's related to that. And I don't think you said it if I missed it, is the frequency of, if you're going in the fishbowl, the frequency of either probably getting one of the top value prizes, right? Like I actually don't play the slots. And the reason being, I think that they're programmed at some level, some ideal level that if it's below the threshold of winning, nobody's going to play them. But I would imagine if you're in Vegas or something and everybody says, oh, go to this one because you tend to win a little bit, like it seems like the signal, and I don't, that might have to do with the environment of if you actually want the people in the treatment talking about how they won. I don't know. I mean, you, I don't know if that threshold has been identified. Yeah. The, the, at least in terms of that study by Nancy Petrie, you want to keep average. Well, and again, this is dated average or average of expected earnings at about two to $300 in price CM versus $500 in voucher CM. Those figures are likely higher now. In fact, Carla Rash, who was a colleague of, who is a colleague of the late Nancy Petrie at the university of Connecticut recently published a paper suggesting increased adjustments to the magnitudes in voucher and price CM. In terms of the typical fishbowl, there are 500 price slips in the fishbowl. Half of them have no monetary value. They're just affirmations, words of encouragement, good job, congratulations, so forth. The other half of the slips, the other 250 are, are as follows. 209 of the, of the 250 or 209 of the 500 are small slips. They used to be worth a dollar in VA's implementation. They're now worth $3. Large, there are 40 of the large slips in the bowl. They are worth $20. And one of the 500 slips is the jumbo worth $100. The average value of each draw is $3. Doesn't mean that there's a necessarily a, each slip will give you $3, but on average you will earn $3 per slip. Some patients will earn more, some will learn less based on their own luck with the, with the system and of course their own success. Key thing, because you alluded to gambling behavior. One of the, one of the critiques of price CM is, oh, that looks like gambling to me. It is absolutely not gambling. The key difference, although both gambling and price CM leverage probability, leverage chance. Remember gambling requires that the person make a wager, take a risk. In price CM the patient's taking no such risk, making no such wager. The patient who never produces a negative sample in CM is no worse off than the patient who wasn't in CM at all in terms of, of losses. They may indeed though be far better off in terms of potential gains if they were to abstain. So in terms of, of the, the threshold for optimal magnitude, we actually don't know. At this point it's largely speculative. We suspect that with voucher CM at this point, you're looking at $750 to $1,000 over a course of 12 weeks. With price CM, $300 to $500 over a course of, of five weeks. Karl Oresh's paper gets into the, the details on those determinations. Does that answer your question? Yes. Thanks. Yep. Happy to. Platform. Oh, frequency of sessions. The frequency of CM sessions is yoked to the detection window of stimulants in urine. So since stimulants in urine can be detected 24 to 72 hours on average, you separate your two testing days per week by, by roughly 72 hours. So you have a, either a Monday, Thursday schedule, a Tuesday, Friday schedule, or a Monday, Friday schedule. And then you could be confident that with those two days, you're surveilling a full week's worth of behavior insofar as if a patient were to use on any of the days, at least one of those two tests will detect the use. And that if both tests are negative, you can surmise that the patient has been abstinent all seven days. So frequency is typically twice a week, can also be three times a week, which increases the opportunity for reinforcement, but increases the demand on the implementer and the patient by obviously an extra session, 33%. What programs can you introduce CM into? Just about any treatment setting, outpatient, intensive outpatient, residential, inpatient settings can all be platforms into which CM can be introduced. OTPs are another typical setting for CM programs. Contraindications. Well, if a patient's on a medication that can produce false positives for the target drug for stimulants, then it's going to be difficult because you will have difficulty differentiating the positive result due to the medication versus a positive result due to illicit use. So for those patients, sometimes the medication can be altered, a dosing schedule can be altered, but the CM may not be the optimal strategy with those patients. By the way, two of the typical suspects among medications to produce false positives for amphetamine and methamphetamine are trazodone and bupropion, unfortunately too widely used medications. When test results can be used punitively, that's not an optimal scenario for CM because of course then CM's high frequency surveillance is more of a risk to the patient than a benefit because if the test result is positive, it can be used punitively, the patient's going to feel threatened by that. And then we in VA out of an abundance of caution decided that we were not going to repeat CM courses on a rapid schedule because we didn't want the chance of the opportunity to get back into CM, incentivize a recurrence of use to do so. And also we didn't want to incentivize patients doing CM on what I call the installment plan. Come in, do three weeks of CM, have a return to use, come back, do another three weeks, have a return to use, because that would contravene the purpose of the CM, which is to promote sustained consistent abstinence to promote that brain healing. So we introduce a condition that the patients who are considered for a second course of CM must not have had their course in the past 12 months. Some organizational concerns being able to implement point-of-care testing. I can tell you that inside and outside VA, there is reluctance in many quarters to support point-of-care testing, despite the fact that the myriad studies of CM supported by point- of-care testing attest to its accuracy. Now, to say point-of-care testing is not accurate is to say that automobiles are unreliable. Well, some automobiles are unreliable and some are quite reliable, likewise with point-of-care tests. Changing the clinic culture, getting clinicians to buy into the idea of focusing on the encouragement incentives of healthy behavior rather than the punishment and persecution of unhealthy behavior. Training and coaching, I can't overstate how important training and coaching are. Even long-term implementers of CM will have stray in their practice of CM from the empirically derived protocol. You need training and coaching. I like to say even the greatest athletes in the world, your LeBron Jameses and so forth, need coaching. So do CM implementers. Telehealth-administered CM introduces some challenges. Obviously, urine drug testing is not going to be terribly feasible in a telehealth encounter, but oral fluid testing, patient self-administered to boot, might be the answer. And then also, you need a means of remotely and immediately reinforcing the patient. And one mechanism there that shows promise is the use of apps to remotely and immediately reinforce. Criticisms of CM. All right, Dom, you've been putting a fancy spin on CM, but at the end of the day, what you're really talking about is you're bribing patients. That's all you're doing. You're bribing them. And I don't like the idea of bribing patients to recover. Well, excuse me, I would make the case that CM doesn't even meet the textbook definition of bribery. When I bribe someone, I'm providing them with material or financial incentives to perform a behavior that is unethical or illegal, that serves my interest as the payer and puts both of us in legal or ethical jeopardy. With CM, there's no such application of that definition. In fact, if we want to grapple with the moral hazard concern, well, if you incentivize recovery, aren't you incentivizing patients to use substances in order to get in CM? I would say that I get that concern at a superficial level, but the notion that a patient would engage in stimulant use to the point that they now are struggling with a stimulant use disorder in order to access a treatment that provides $500 to $1,000 in reinforcement for abstinence strains credulity. And also we know that stimulant use disorder is a life-threatening condition. What about the moral hazard of withholding an effective treatment for a life-threatening condition? All right, well, maybe it isn't bribery, Dom, but you know what? I don't like the idea that you're paying people. You're paying them to be abstinent. We shouldn't have to pay people to do the right thing. They should do the right thing on their own. Well, again, I think definitionally that critique doesn't make sense. When I pay someone, I am compensating them for time, labor, and material to serve my interest. I pay the auto mechanic to fix my car. I pay the plumber to unclog a pipe. I'm not paying the patient to be abstinent. I'm reinforcing that behavior that is in the patient's best interest. And let's boil it down to first principles. Managed reinforcement contingencies are how the substance use disorder developed. Why would we not apply the same effective learning contingencies to promote recovery behaviors? All right, maybe you're not paying them, and maybe you're not bribing them, but I got you with this one. You start giving people rewards, and you're going to reduce their internal motivation to recover. In other words, you stop rewarding them, they're going to go right back to using, and I know that happens, so I got you, DeFilippis. That critique has some teeth. We know from, and if you're harking back to, if you took a class in social psychology, there's a famous historic experiment that elucidated a concept known as the over-justification effect. What these researchers, Leper and colleagues, did is they looked at children who enjoyed coloring in coloring books. Now, what kid doesn't enjoy coloring in coloring books? And those kids would just, when given the opportunity to color, they went to it very, very readily. Well, then they separated those kids into two groups. One group, they provided external incentives to color. When the kids colored, they got candy. The other group, they just let them color at will. What they found was when they withdrew the incentives from the incentive group, those kids did indeed reduce their coloring behavior. Oh boy, that's scary. Could it be true that by providing external rewards, we reduce internal motivation? Well, the problem is we're talking about apples and bowling balls, not even apples and oranges. Coloring for children is automatically reinforcing. The behavior itself is immediately rewarding. The engaging in coloring, like riding a bike or whatever automatically reinforcing behavior you want to choose, doesn't require external reinforcement. Recovery, on the other hand, is not automatically reinforcing, certainly not in the immediate consequences. In fact, if anything, the immediate consequences of recovery behaviors are punitive. Withdrawal, clear-eyed view of one's, the devastation in one's life. So, we're not talking about the same category. It's a category error to apply the over-justification effect to recovery behaviors. Indeed, paper after paper that looked at reduced internal motivation did not find any evidence of it. The most potent disconfirming evidence came from a meta-analysis by Meredith Ginley and colleagues, published in the Journal of Consulting and Clinical Psychology in 2021. I think it's the most important paper in CM science in the past decade. Ginley and colleagues did a meta-analysis of CM studies that followed patients for up to a year after reinforcement was discontinued, and also did toxicology testing. And what Ginley and colleagues in their meta-analysis showed is that CM effects on abstinence can endure up to a year after reinforcement is continued. Up to a year. Clearly, if we diminished internal motivation, you wouldn't see those sustained effects for up to a year after reinforcement is discontinued. And it's no surprise why that occurs. Because Nora Volkow was right. Her PET scan studies were right. You get enduring abstinence. You get brain healing. The brain healing helps support enduring abstinence. Well, then how do you explain, Dom, that some patients, when you stop the CM, they do have a recurrence of use? Well, I would say, you know, it's an interesting observation. Have you gone to your internist and said, challenged your internist by saying, you know, those anti-hypertensives don't do anything. When I stop taking them, my blood pressure goes back up. No, in fact, that recurrence of symptoms upon the discontinuation of the treatment is prima facie evidence of the effectiveness of the treatment. Yet, because of stigma, we say to patients with substance use disorder treatment, when you stop your treatment and the condition comes back, it's because the treatment failed. Nonsense. Why does the treatment, why does the effect recur? Because for some patients, the course of CM was insufficient to promote sufficient healing of the brain to sustain recovery long-term. Perhaps those patients need longer courses of CM. Another factor is that with sustained success in CM, the patient's self-efficacy improves. They learn that I can be abstinent and it's worth being abstinent. And their self-efficacy in concert with their improved brain dopaminergic function helps sustain their abstinence moving forward. I promised you at the outset, we'd tackle this harm reduction issue. Abstinence CM is wholly consistent with a harm reduction approach. How can I say that? Because CM does not say to the patient, you must make a commitment to abstinence to be in CM. It says to the patient, hey, give it a try. If you're not abstinent, you're not going to be any worse off as far as the CM is concerned than any other patient. But if over the course of these 12 weeks, you give abstinence a try, we're going to reinforce you immediately. And that experience may tip the motivation you have for abstinence moving forward. Okay. Well, maybe all that's true, but do you realize, DeFilippis, what you're proposing? You're proposing that we give items of value, monetary value, to patients with a substance use disorder. I know from my own clinical experience that patients have sold off belongings, sold off the belongings of family members to support their substance use disorder. What makes you think they won't divert the incentives you provide them? Well, there is a non-zero probability that can occur. Of course, patients could divert their incentives. But that risk is not as pronounced as one might think. In fact, the late David Festinger, we're talking about figures in substance use disorder treatment research who passed away, David Festinger not too long ago, passed away. He had that research question under examination. He thought, well, what is the easiest incentive to divert? Cash, of course. You can trade cash off for whatever easily. Merchandise would be a little bit tough. You've got to sell the merchandise or find a willing barter for the merchandise. So he thought, let's look at CM studies where the patient got cash versus CM studies where the patient got merchandise. And I would hypothesize, or Dr. Fessinger hypothesized, that cash patients would show more higher levels of drug use because they're diverting more often. Didn't find a difference. Now, that's not to say that diversion of incentives cannot occur. Of course it can occur. But bear in mind that CM has a built-in fail-safe for that. If a patient were to divert incentives, what's going to happen? They're going to return, test, not negative, not receive reinforcement, have their reinforcement reset. And also, to think that a patient with fulminant stimulant use disorder will say, all right, I've got this now. I'm going to abstain for 12 weeks so I can get a lot of reinforcement for a big planned recurrence of use in 12 weeks. Phenomenologically, we just don't see that type of behavior occurring. That's not to say we shouldn't be wary of it and maintain the integrity of a urine testing regime and the accountability of our incentives. But diversion in and of itself is an insufficient reason to not make CM available. OK, well, I've talked about the rationale. I've talked about how CM embraces the challenge of substance use disorder recovery. The methods, the critiques, and responses to the critiques. But at the end of the day, it all boils down to this. Is implementing CM, the juice of implementing CM worth the squeeze? No question about it. Meta-analysis after meta-analysis. Decades of meta-analyses published in the most prestigious journals, JAMA, Journal of Consulting in Clinical Psychology, and so forth, have demonstrated the replicated findings that CM is effective in the treatment of substance use disorder, stimulant use disorder, in particular, with a moderate to large effect size. Some of the quotes from these meta-analyses. It's among the more effective approaches to promoting abstinence. The strongest effect was found for contingency management. Brown and DiFulio, in a literature review on CM for methamphetamine, broadly effective in reducing methamphetamine use. The last meta-analysis by Bensley and colleagues, published in 21 in JAMA, 157 studies. This was a remarkably potent meta-analysis looking at treatment of cocaine use disorder. Only CM, I repeat, only CM programs were significantly associated with an increased likelihood of having a negative test result. OK, that's in the empirical studies, Dom, where those studies typically exclude pregnant patients, psychotic patients, patients with suicidality. It's sort of kind of a cherry-picked population. Where the rubber meets the road in the clinic, does it really work? It sure does. We've been making CM available in VA since 2011. Our outcomes, in terms of attendance and abstinence, are comparable to those observed and reported in the empirical randomized clinical trials of CM. What do I mean? Well, we reported on our outcomes in the first, roughly the first 55 months of the implementation. By that time, we had treated over 2,000 veterans and over 94 programs across the enterprise. 50% of CM patients completed 14 or more sessions. Remember that attrition statistic I mentioned earlier? 70% attrition after just four sessions. We're getting 14 or more CM sessions over a 12-week period. Compared to patients that didn't get CM, only 42% of those patients get more than two sessions of treatment in a year. Yeah, but are they abstaining? 92% of the nearly 28,000 samples that we collected from veterans tested negative. All right, but what about since then? Well, through FY23, we've now implemented CM in 120 VA stations. Over 6,400 veterans have received it. 92% of the more than 83,000 samples that we've collected have tested negative, and the average retention is 6 1⁄2 weeks. All right, that's veterans. You're talking about an integrated healthcare system. What about outside of VA? I present you this slide. CM has demonstrable effectiveness with a variety of patient populations. The homeless, patients with serious mental illness, patients with PTSD, patients with HIV disease, justice-involved patients. Yes, veterans. Patients on medication for opioid use disorder. Across ages, across races, across sexes, with pregnant women in the LGBT community. Yes, and even across income levels, CM has demonstrable efficacy. So why implement it? It's needed. It works. I could stop there, and that would be sufficient. It's not limited by discipline. It's brief. A CM session is under 15 minutes in length. It's low cost. It could be combined with any treatment. Medication, psychotherapy, combination of medication and psychotherapy. And last but not least, it's a lot of fun to make CM available to patients. I am a trainer and practitioner of motivational interviewing, motivational enhancement therapy, cognitive behavioral therapy for SUD, noble effective treatments all. I practice them, and I train others in them. But I've never had a patient with whom I was conducting a functional analysis of behavior jump out of their chair and do a happy dance. But that happens in CM when patients earn reinforcement. So with that, I'll close, and thank you so much for allowing me to join you this evening and talk about CM. Well, thank you so much, Dr. DeFilippis. That was really a truly remarkable talk. I really feel smarter for having been a part of this. I've seen your lectures before, but I continue to learn so much about it. While we're waiting for people to come up with questions, I'm going to ask one myself. Please. So, you know, like just what you were talking about with the smiles, the shouts, and the happy dances, you know, it's obvious that the structure of the CM is important, but also there's some nuance and the flavor. So I'm thinking that, like, if I was implementing the CM with kind of my, you know, austere, bordering on disagreeable demeanor was doing it versus you doing it, I suspect the outcomes would be much better if you were doing it. Can you comment on that? That's an outstanding question and an excellent observation. Yes, competent delivery of CM. In fact, Nancy Petrie developed a competency scale for delivery of CM, and the provider's enthusiasm and surrogate hopefulness are key features of effective CM provision, especially in price CM, where by design, a patient who is abstinent may still not get tangible reinforcement that day if they draw the slips that don't have any monetary value. So it's crucial to maintain that positive outlook, optimism, even with a patient who did not test negative, not test negative, staying focused on today. However, when you come in next time and you test negative, you'll be back to earning. And I always recommend taking a motivational interviewing approach to all work in psychotherapeutics, including in CM sessions. So when a patient appears for that CM session, praise and affirm and acknowledge that the patient showed up because the truth of the matter is, that patient could have engaged in countless alternative behaviors other than meeting with us, countless alternative behaviors. And I could tell you, that's especially true for the veterans that my colleagues and I treat who think about it from their perspective. Here they are, they come to what's often a very imposing, brutalist design edifice, surrounded by law enforcement, VA police, to meet with agents of the federal government to discuss a behavior that is fraught with shame and guilt and remorse. And yet they managed to harness enough motivation to fight through traffic, deal with parking, find my office and come see me. As far as I'm concerned, that's nuclear fuel for recovery. And I wanna know more about what's driving that patient. And I am humbled, humbled by their presence under those circumstances. I mean, even if they're not getting one of the rewards, right, like, I mean, dopamine signaling starts with like the anticipatory, like preparatory phase anyway, right, so I mean, they're getting something that feels rewarding even before they maybe don't get it. Vani, you should mention that. When Nancy Petrie did focus groups of patients following their participation in her CM studies, asking the patients to rank the most important features of CM and the reinforcement was typically a middling ranked feature. It was meeting with the provider, the provider being optimistic, the urine testing regime as just the informational feedback from the urine testing regime. In fact, the recasting or reframing of urine testing, not as a means of persecuting a patient, but as an opportunity to catch the patient doing something good. In fact, reminds me of a story that was shared with me by an implementer in California, a VA implementer in California. At the end of the course of CM, one of his CM patients handed him a handwritten letter and the patient disclosed in the letter, he said, you know, when I first met with you and you talked about CM, I felt humiliated. He said, here I am an adult, I'm a veteran and I'm gonna use the language the patient used. I have to pee in a cup for another adult, that's humiliating. But he said, you know, then I started to earn rewards and I found myself racing through traffic, running down the halls to find you so I could provide a urine sample and earn reinforcement. And he said, my only regret as I came to my last session was I never got the jumbo prize. My hand to the almighty. At the patient's final session, he drew the jumbo prize slip. Hollywood couldn't write a better end to that story. So absolutely, anticipation, the reframing of urine testing, the optimism in the sessions, the affirmations that the provider offers all key to competent delivery of CM. So one other logistical question. So I assume everybody uses the same fishbowl. And so what, like if the jumbo prize is taken, then you put it right back in for the next person. Great question, yes. This is draws with replacement. So at every CM session, the bowl has the 500 slips in it because if it was without replacement is just as you pointed out, let's say at my very first CM session, I drew the jumbo slip. I might do my happy dance for a moment until I realize, oh gosh, I've got 23 more sessions of this over the next 11 and a half weeks and I can never get the jumbo again. So it's draws with replacement. In fact, good CM practice mandates that the implementer keep an eye on those slips to make sure they don't develop markings, creases, tears, because I assure you the patients will learn to associate those markings with the value on the slip. So Carla was at, Carla Marienfeld was asking is, first of all, she said, this was a powerhouse talk and I appreciate all that you've presented. Do you happen to have a one pager countering all those common critiques? Oh, that's a good suggestion. I don't have a one pager, but the slides are available. I'm happy to share them. So feel free. You could distribute them to attendees and make them available. So I myself have been involved with using CM for stimulants, of course. I've used it for smoking cessation. We're using it in Portland VA for people who, if they show up for their long-acting injectable naltrexone shot, can you maybe review like some of the other situations that we should get excited about where there's been shown efficacy and? Yeah, absolutely. You mentioned some of the target behaviors we're applying CM to in VA, medication adherence. Now, I made the case at the outset that medication adherence can be challenging because for self-administered medications, the monitoring is the challenge. In VA, we made a decision to make a foray into medication adherence CM with medications, the adherence of which is self-evident, which is the provider administered long-acting injectable medications for substance use disorder. Extended release naltrexone and extended release buprenorphine, both the sublocade version and now the Bruxade version. And the reason why is it's obvious that the patient adhered in that the injection was administered by a provider. The length of course is different, however. For adherence CM, the length of the course is 12 months rather than 12 weeks. So we're getting typically 12 administrations of the medication over a one-year period. And the escalation schedule is different. It starts at four draws for the first injection and escalates by fours to a cap of 16. So second injection gets you eight draws, third injection gets you 12, and the fourth and subsequent injections in a row will get you 16 draws per injection. The other behavior that we target extensively in VA is attendance, attendance in group especially. And there we define attendance as physically being present and how we address the issue of falling asleep or being late or leaving early is if you leave early or you fall asleep, you've earned that day's reinforcement but you're reset for the next session to the starting amount. So we introduce that disincentive for engaging in those behaviors that contravene optimal attendance in group. Are we missing anything else, Dave, that I? I don't think so. I had a question. I know there'd be like ethical considerations around this, but I wondered if there, and certainly for people that are very under-resourced, this wouldn't work, but if there was a way to have people essentially fund their own contingency management, like you have to put in $500 and then essentially you're just winning, winning is clearly not the right word, but it's your own money that you're getting back. There's some applications where there's a match, like if the patient or patient's family puts up an amount, a third-party payer would put up an amount. This diet bet system that's out there for weight management is along the same lines, although it's true gambling, where the patient puts up, makes a wager, and then depending on if they reach weight loss goals, they can either earn more or lose their wager. So I think there's more of an acceptance of the idea that incentives matter and resourcing incentives is key. I can tell you that even the third-party payers, the insurers, are starting to warm up to the idea that, hey, it might cost us $1,000 a patient, but if we could save the cost of the residential treatment and the hospitalizations and so on and so forth, heck, one hospitalization avoided would fund the course of CM for dozens of patients. One other thing that I often will talk to patients about and talk to fellows about is that if life in recovery is not better than life while they're using or drinking, well, then logical people will go back to using or drinking. And the fact that early recovery is such a painful state for most people, it just seems like this is such a logical way to help bridge that gap until they're able to kind of appreciate a walk in a park or have a normal dopaminergic tone. And so I really appreciate the perspective that you brought about relating to contingency management on that. Absolutely. The choice to pursue recovery is an extraordinarily difficult choice, considering that the consequences of that choice, the immediate consequences are often so profoundly aversive compared to the immediate seductive costs, seductive consequences of continuing to use. Well, I have to say that was just such a remarkable lecture. I'm so grateful that you took the time out of your evening. I know this took a few minutes and we appreciate you being here and everybody. Next month, we have Shelly Greenfield talking about psychotherapy in women. And remember, these are all archived for future viewing. There's already comments that people are gonna wanna share this with others. This was truly remarkable lecture. And thank you again, Dr. DeFilippis. My pleasure, Dr. Blazes. Thank you so much for this opportunity and please do share the slides widely. They are yours to share with colleagues. We absolutely will. Thanks so much. Bye everybody. Bye everyone.

addiction treatment

contingency management

substance use disorders

reward systems

recovery behaviors

positive reinforcement

extinction

punishment

negative reinforcement

implementation concerns

patient outcomes