Suicide Prevention

Editor’s note: Edwin Boudreaux, PhD, is a Professor of Emergency Medicine, Psychiatry, and Quantitative Health Sciences and the Vice Chair of Research for the Department of Emergency Medicine at the University of Massachusetts Chan Medical School.

Sarah Mossburg: Dr. Boudreaux, would you tell us a little bit about yourself?

Edwin Boudreaux: I am a clinical psychologist by training. I work at UMass Chan Medical School, and I run the Center for Accelerating Practices to End Suicide (CAPES), which is focused primarily on accelerating the adoption of best practices and evidence-based practices related to suicide care in healthcare settings like emergency departments, inpatient units, and outpatient care.

Sarah Mossburg: Before we dive into the topic of suicide screening and prevention, could you give us some background on the prevalence of suicide and suicidal behavior among people in healthcare settings and immediately after discharge?

Edwin Boudreaux: If we look at people who have died by suicide and work backward, we find that up to 85% of them have been seen in some healthcare setting in the year before they died.¹ And these are often not behavioral health settings; they are regular medical settings, emergency departments, primary care, outpatient specialty clinics, etc. If you look at the month before death, up to 40% of them have seen a healthcare provider.¹ What is most striking about this to me, besides the fact that the healthcare settings they visit are often not behavioral health settings, is that many of these people have not been screened for mental health disorders or for suicide risk in those settings. Even if they are being screened for mental health disorders, like depression, they are not being screened for suicide risk. While identifying mental health disorders could potentially help to improve our identification of suicide risk, suicidal ideation and behaviors can occur in people who do not have a psychiatric condition or do not share that they have been having mental health problems. Therefore, we should also specifically focus on screening for suicidal ideation and behaviors.

In the Emergency Department (ED), you can reliably screen up to 95% of people coming in. Our research has found that between 3-8% of all ED patients have some level of suicide risk that should require attention.^2,3 They do not all have the same level of suicide risk, of course. It is a spectrum. For example, people may admit to having had thoughts of killing themselves in the past two weeks, or that they have tried to kill themselves in the past six months. Anyone who reports this can be considered as some risk that requires at least some level of attention. This should include more screening and assessment to figure out exactly how serious this person's risk is.

Sarah Mossburg: So temporal proximity, indicates that healthcare providers might be in a unique position to play a critical role in suicide prevention. Is that accurate and are there other reasons why they might be uniquely positioned?

Edwin Boudreaux: Yes, the temporal component is important. If we are going to try to prevent suicide, we must ask, “Where are the opportunities to identify and intervene?” Healthcare settings can play a particularly important role, not only because of this time proximity we just talked about but also because the role of healthcare settings is to identify when people have problems that are different from what they are presenting for yet require attention. Many healthcare settings, especially primary care healthcare settings, OBGYN settings, even emergency departments, have a preventive health role. We try to identify other problems that require attention in addition to the primary reason they are presenting. For example, tobacco and blood pressure screenings are routine in many healthcare settings to catch hidden risks that lead to serious health issues. Even when a patient's visit is for unrelated reasons, these screenings are performed to identify potential dangers early. The same could be said for screening for mental health disorders and, of course, screening for suicide risk. Suicide is obviously a fatal issue; you do not want to ignore it because it is potentially life-threatening. I think healthcare settings have a role in screening and detecting suicide risk for that reason.

Sarah Mossburg: Suicide prevention is one The Joint Commission’s National Patient Safety Goals. When patients die by suicide in healthcare settings or shortly after discharge, it is considered a Sentinel Event. Could you speak briefly about both of those and their significance?

Edwin Boudreaux: A healthcare system that is accredited by The Joint Commission needs to do everything that they can to prevent suicide or serious intentional self-injury during the person's visit. If they are in an emergency department, an inpatient unit, or, of course, a behavioral health unit, it is the health system’s responsibility to make sure that the person is safe, does not hurt themselves, and does not die by suicide. Not only during the visit, but they are also responsible for reporting if the person dies by suicide within seven days of discharge. It used to be three days, and it was just recently increased to seven days. This is important because it emphasizes the healthcare system's role in preventing suicide, not just while the person is under observation or care within the facility, but they are also responsible for helping to transition the person to receive services in the community. This is being emphasized more now than it used to be in the past. In the past, health systems thought that they only needed to keep people protected while the person was within the health system. They were not taking responsibility for what happens after the person is discharged. Therefore, our transition plans, our capacity to transition people from an acute care setting to an outpatient setting, were not getting as much attention as they required, despite the fact that we know that there is a spike in suicides right after discharge from acute care. We were not doing as much as we should. The Joint Commission is trying to encourage and require health systems to think about what happens to patients when they leave the facility and make sure that they have appropriate resources to help bridge that gap.

Sarah Mossburg: Let us talk a little bit about the screening process. I am curious how the process of suicide risk screening differs across various settings, such as outpatient versus inpatient care.

Edwin Boudreaux: Most of the principles remain similar, although the tools that the different settings might use are often different. Let’s start with the emergency department and inpatient settings. When people present to care at an emergency department or get admitted to an inpatient facility, they typically have assessments and screenings that happen during the first part of that visit, and then they will have continued monitoring while they are there. In emergency departments, several different tools are used to screen for suicide risk. One tool that we have created and validated is called the Patient Safety Screener. There are other tools, like the Ask Suicide Questionnaire (ASQ), and then there is the Columbia Suicide Severity Rating Scale, which has a screening version. These have been validated for this purpose in emergency departments and are used throughout the country. Different emergency departments can decide which tool they want to use based on a host of factors, like the kind of patients that they treat, whether they treat children or adults, and their general preference for a tool. Once they choose which tool to use, they choose whether they want to use it only for people who present with behavioral health complaints or if they want to do broader universal screening. As I mentioned earlier, you could screen everyone who is capable of being screened when they come in. Those are two different models.

The same kind of decision-making would occur in an outpatient setting, let us say, primary care. They would have to ask themselves the same questions: Which tool are we going to use? How often or frequently are we going to screen? If a person is negative in the initial screen, when should the next screening be? If they are positive, how often do we reassess their risk to monitor them over time? In each of those settings, regardless of whether you are in the ED or outpatient setting, the organization needs to decide what to do after the screening. If a person is positive, what is the next step? Is the next step additional questions or additional assessment to further stratify that person’s risk and determine if their risk is imminent? In other words, do we need to act today, or is the level of risk lower, and we might be able to take some less intense action? One difference to note between primary care and EDs is that EDs will use the same tool and process every time that person comes in. If they come in for an ED visit today and they come in for another ED visit six months from now, that same process is going to be used at both of those visits. Whereas in outpatient care, since you see patients longitudinally, the frequency with which you screen a person might be different. If they see a primary care provider for an annual wellness visit, they may be screened then. If they are negative—in other words, they do not screen positive for concerning risk—then they may not be screened again until next year, even though they might come in to see the primary care provider for other problems during that year.

Sarah Mossburg: You talked about using clinically indicated screening versus universal screening. In your opinion, is one of those better or worse? Is there stronger evidence to support either approach?

Edwin Boudreaux: I think of it as a floor and a ceiling, the minimum standard and the aspirational standard. The minimum standard, as defined by The Joint Commission, is to use an evidence-based screening process for anyone presenting with a mental health or behavioral health complaint. If I were advising an emergency department on how to improve their practice, and they were not using an evidence-based screener with anyone presenting with a behavioral health complaint, that would be where we started. We should be using tools that have evidence, and we should be screening everyone in the same or a similar manner. A standardized approach ensures that when a person screens positive, and a clinician communicates that to the rest of the team, everyone understands what that means. If you are not doing that, you are not meeting the minimum standard. Many health systems have already been doing that. They have been using evidence-based screenings, they have trained their personnel on how to do a good screening. If they were trying to do something better than that, to try to improve upon that performance and provide the best possible care, they should consider universal screening. The health system I work in, as well as other organizations, have published data showing that when you expand your screening to universal screening, you identify people who are at risk. They are presenting with a medical complaint, not a primary behavioral health complaint, but they do have suicide risk.^1,2 It is what we call incidental risk, or sometimes people call it occult risk, meaning it was hidden. When we do universal screening, some estimates show that we almost double our detection of suicide risk. ^1,2 That is how I would think about it. Not that there is more evidence for one or the other, but that we should think of it as a minimum and aspirational standard.

Sarah Mossburg: That is a helpful outline for folks in the field who are thinking about where they are in this process and where they want to go. You mentioned screening tools multiple times at this point, and I am curious what factors someone should consider when they are selecting a suicide screening tool for their organization.

Edwin Boudreaux: Interestingly, there are more and more tools available, so it gets more complicated because there is not one single tool. By comparison, when people screen for hypertension, the method is the same, no matter where you are. Everyone uses the same type of device; the numbers mean the same thing. It is universally standardized across healthcare settings, which is very nice. There is no such thing for suicide risk, unfortunately. An organization should look at the available screeners, and there are lists of screeners that can be easily found through literature searches.⁴ The Joint Commission itself has a link to some suggested screeners that they have looked at.⁵ It is a bit dated, but it has most of the major screeners on it. What I would recommend is having a team of people look at those screeners and identify which ones they think are the best fit. That might depend on things like the length of the screener. Obviously, the shorter the screener, the more acceptability it is going to have in clinical practice. Clinicians are busy; they have a lot of tasks, including other screening tasks, that are all important. If we can make the screening as streamlined as possible, then that helps the providers conduct it with fidelity. The age of the patient population is also important. For example, the screener that has been most well-studied in children between 12 and 18 years old is the Ask Suicide Questionnaire (ASQ). It was developed with children, validated with children, and has been used in a lot of children's hospitals. If you are a pediatrics practice or you are caring for children, the ASQ is the go-to screening instrument because of that history. Other instruments have not been validated in children, so pediatric settings should avoid using them.

Another factor is how useful the screening results can be for understanding what is going on with the patient. For example, the Patient Safety Screener, which is the screening tool that we have created and that many of the health systems use, has branching logic. There are three mandatory questions you ask of every patient, which makes it short. If the person is positive on “thoughts of killing themselves” or “a recent suicide attempt,” there are six additional questions about evidence-based risk factors that are associated with higher risk for future suicide attempts or death by suicide. You get through it quickly, and if they have any risk, it stratifies them based on three levels that can be aligned with clinical action. It is not just positive or negative, it is not just a binary outcome. The people at high risk are going to get the highest safety precautions, such as an observer who is watching them to make sure they do not hurt themselves. They may have restrictions on what they can have access to, like their own personal belongings, to make sure that the person does not have something in their control that they could use to hurt themselves. We want to prevent the sentinel event, and some of those precautions are quite intense. You don’t want to place those most intense precautions on someone unless you are very concerned about them. You are removing people's clothing and putting them in hospital gowns and austere safe rooms, having a person watching them to make sure they do not hurt themselves. If you tried to do that for everyone with any suicide risk, it would be very costly, and it would lead to a lot of dissatisfaction. So the stratification component of a screener is super important because it identifies the people who need that higher level of care, and then people who are lower on the strata do not need it. For example, with a person who has mild risk, we cannot ignore the risk, but they probably do not need to have an observer watching them.

Sarah Mossburg: Could you tell us more about the Patient Safety Screener?

Edwin Boudreaux: Sure. There is a primary and a secondary part. The primary is what I mentioned earlier, the three questions that are asked of every person screened. If they screen above a certain threshold, there are asked some additional items. For example, if they have intent to act on their suicidal thoughts. A person can have thoughts of killing themselves but say that they never intended to act on them. If they do have an intention to act, they are asked if they have plans to act. We are trying to figure out where the person is on the ideation-to-action spectrum. Obviously, those people with intent and with a plan are people you are more concerned about. They would be in the high-risk group. There are a couple of other items on it that are not suicide-related. For example, if the person has had a psychiatric hospitalization or substance use issues. This all helps provide more information about what is going on with the person. It is useful from a clinical perspective to better understand the risk factors, and it can be scored to stratify, as I mentioned earlier, to match the intensity of the intervention to the severity of the risk.

Sarah Mossburg: Earlier you mentioned the challenge of how busy clinicians are in their day-to-day work. In these busy healthcare environments, what are some ways that providers can integrate suicide risk screening without compromising time for other tasks?

Edwin Boudreaux: Selecting the right screening instrument for the setting is important. Finding something that's evidence-based and short can really help with this challenge as well as determining whether you are going to do clinically indicated screening or universal screening. Clinically indicated screening obviously reduces the burden on the clinician because they are only going to be screening people presenting with a behavioral health problem, but the consequence is missing people who are at risk. There is always a decisional balance there.

It is also important to develop the protocol, or care pathway, ahead of time. This will clearly delineate how to do the screening, when to do the screening, what screener you are using, and then what to do if the person screens positive. The protocol should be written down, very clearly laid out, and standardized. Everyone on the team has to know about it. The clearer that is, the less cognitive burden there is. What often contributes to inefficiencies in workflow is not having written out expectations and not having everyone on the same page. There is confusion, and people are not able to apply the protocol with fidelity. Make sure everyone on the team is trained and practiced in how to do it. We can do all this stuff in a hypothetical environment, but in a real situation, when you identify someone who is suicidal, it can be stressful for everybody. It is obviously stressful for the person who is suicidal, but it can also be stressful for the clinicians, especially if it is a care location that does not see a high volume of suicidal patients. The ED sees a lot of patients who are suicidal, so the clinicians tend to be more comfortable with it. But a primary care clinic may not see that many suicidal patients, and if they do identify one, they could panic and not know what to do. A clearly laid out protocol and making the environment of care conducive to following the protocol can help avoid that. For example, you can develop tip sheets or build tools into the electronic health record that can quickly remind clinicians what they should do. If clinicians are not practicing the protocol regularly, they are not going to remember what they need to do next. Make things available to them to refresh their memory quickly when they need it. That will also improve communication and reduce the impact on a busy schedule.

Sarah Mossburg: That leads to my next line of questioning around post-screening assessments. When an inpatient screens positive for suicide risk, what are the immediate steps that providers should take to ensure their safety?

Edwin Boudreaux: When we talk about screening, we are talking about something very quick, and screeners are usually on the conservative side of trying not to miss anyone with risk. That bar might be low deliberately. However, the screeners are not going to get a lot of information about what is going on with the person, such as what is driving their suicide risk and what is the nature of the environment to which they are returning. All these factors, the risk factors, the protective factors, the level and intensity and controllability of their ideation, the severity of any past behavior, all those things should be done during the comprehensive suicide risk assessment. Screening is just the beginning.

Immediately after screening, you would take whatever action the protocol dictates based on their strata, which is usually geared more toward making sure a person is safe until a trained provider does a full assessment. That full assessment is going to determine what happens next. That full assessment, for example, might suggest that the person is not at imminent risk, they do not need to be on safety precautions, and they can be discharged from the hospital. Or it says that they are at high risk, safety precautions need to be maintained, and they need to be admitted to a psychiatric unit for further treatment. That would be the top end of the spectrum. Then there are, of course, levels between those two. That is what that suicide risk assessment is supposed to do—produce a much more accurate understanding of the suicide risk so that further care can be planned, and what that further care will look like.

In the outpatient setting, the process is similar. If you identify someone with suicide risk on a screener, an assessment should be completed. Then the decision is, do we need to send this person directly to a crisis center or get a behavioral health evaluation? What do we need to do today? They are coming in for their primary care appointment, but now we need to act. If they are at the highest level of risk, then the clinic should assure that they are not left alone in the clinic, a patient care attendant or someone should sit with the person while the clinician decides what to do. They might have a psychologist or a social worker or someone who is trained to see the person during their visit, that would be ideal, but it can’t happen everywhere. I encourage outpatient practices to figure out some alternative to sending the person to the emergency department. Only the most imminent and the most severe people should be sent directly to the emergency department from a primary care clinic. Instead, clinics should learn about what crisis services there are in the area. Many communities have crisis centers that are not EDs, but are specially designed to assess and understand suicide risk and determine what to do with that person. Most people who go to a crisis center do not get sent to an emergency department. Instead, they get care there, or they get connected to other care, and they can be diverted from the ED and cared for in the community, which is our goal. In many cases, you can call the crisis center from the primary care clinic and have a consultation right there to determine what is next. Many crisis centers will do that as a service they provide to the community. The easiest way to do that now across the country is to dial 988. Even if you do not know the specific crisis center, 988 is a nationwide, universally accessible phone number, just like 911, but for mental health crises. So, for clinics that do not have mental health professionals on-site, taking advantage of 988 and known crisis resources in the area is important.

Sarah Mossburg: Thinking back to the incidental appearance of suicide risk for people who are being admitted for medical-surgical reasons, why is it important to have a plan and a process in these medical-surgical units for suicide screening and intervention?

Edwin Boudreaux: In a medical unit, most beds or rooms have a lot of items that are unsafe for a person who is thinking about hurting themselves. Ligature points, medical equipment, etc. If a person is very suicidal, you want to try to prevent them from having access to the means with which to do that. That is why it is so important to understand what level of risk that person has. The medical center can choose to send the person to a bed or an area that has reduced risk. Some places will have that, like safer rooms. They could also potentially cohort people who do not need as much medical equipment into areas where there is less of that. But the other option is to make the room as safe as possible, removing all the equipment that can be removed from the room. That takes time and effort, so you would never want to do that unless you really had to. Most hospitals will have protocols for that, such as an environmental safety checklist. A nurse or a patient care attendant will have to check off that they have removed all these materials to prove that they made the room safe. Most often, when a person dies or gets seriously injured in a unit, they use something that is available in the room. Sometimes they will use something that they brought with them or a visitor brought in with them. So, in addition to making the room safe, hospitals should also remove the person's personal belongings to make sure there is nothing that they can hurt themselves with. Many hospitals will keep the person in hospital clothes rather than giving them back their clothes, just to make sure that the person does not have anything they could hurt themselves with. And if they have visitors, the visitors’ belongings need to be kept out of the room or must be searched to make sure that they are not bringing in anything that the person could harm themselves with. Often, the visitors do not bring in things for the person to harm themselves, the person just uses them for that purpose. For example, they bring the patient a pair of shoes with shoelaces that could be used as a ligature. They think they are helping their family members, but then the patient uses it to try to hurt themselves. Those are the precautions you would take if a person were highly suicidal in a medical unit; do your best to reduce their access to means with which they could hurt themselves and put an observer there.

Sarah Mossburg: You mentioned earlier the challenge with transitions. What does the discharge and referral process look like for a patient who requires mental health services after they have screened positive for suicide risk, and how can providers ensure a smooth transition as they are discharged from care.

Edwin Boudreaux: This is the area of suicide-related care that health systems often fail at presently. We have done a lot to improve screening, assessment, and management while a person is in the healthcare setting. But we have done the least work on what to do once they leave, which is why the Joint Commission recently pushed out their Sentinel Event window, to encourage health systems to think more carefully about this. Again, I like to think of it as the minimum expectation and then the aspiration. The Joint Commission says, at a minimum, if you have a person who has a positive screen for suicide risk, you need to carefully consider the environment to which the person is going to be discharged. If the risk assessment suggests that the person is at high risk when they return back to their environment, then you may send that person to an inpatient unit or an intensive outpatient program. They also state that if the person is being discharged home and not being admitted to an inpatient facility, at a minimum, you need to provide behavioral health discharge referral resources to the person, as well as crisis management information. Those crisis resources and crisis telephone and text lines are super important to provide to patients because sometimes they may be the only safety net the patient has if they cannot access behavioral healthcare easily. The good news is that 988 is now available for the whole country. Anyone can call 988 regardless of whether they have insurance; it is completely free. If they do not want to talk, they can text. In many communities there are also other crisis management places that are not traditional behavioral health facilities where someone having a mental health crisis can turn for help. The bad news is that many providers and patients do not know about these resources. This new emphasis from The Joint Commission will hopefully make sure that more people are aware and have access.

So that is the minimum expectation. The aspirational ceiling, which most health systems aren’t doing, is making additional post-visit follow-up telephone calls for patients. We know that these work from evidence-based interventions, where the health system reaches out to the patient usually within a few days of their discharge.⁶ The closer to discharge the better, because the days after discharge are the most vulnerable window for suicides. The call is usually semi-structured, and the primary goal is to check on the person and make sure that they are doing okay, that they are not thinking about killing themselves or not acting or preparing to kill themselves—it’s a safety check.

This call is also a good opportunity to review the discharge plan with the person. Often, when we make a discharge plan with a patient in the facility, they are not in a good condition to remember everything that they are supposed to do. Many of these patients are emotionally exhausted. They are distraught, they may be sleep deprived or have cognitive processing problems as part of a mental health condition. And the environment of care is often chaotic. Those conditions are not very good for absorbing and understanding information. When people leave, we wonder why they don’t follow up with anything. It is often because they never really retained the information to begin with. So this post-visit follow-up call can help with that retention by reviewing the discharge plan, walking through the steps, and making sure that the person is following up with all their discharge instructions, including seeing their primary care provider, for example, or making an appointment with a behavioral health provider.

The clinician can also review the person's safety plan, which is a plan for what the patient might do if they encounter a triggering situation or notice some warning signs that they are having a crisis and are thinking about hurting themselves. It is designed by the patient and provider together so that patients have, in their own words, things that they feel are feasible and realistic to address that suicide risk early, before it escalates too high. It is a prevention strategy that has been used in many different countries with good evidence behind it.⁷ Like I said, to develop a good safety plan before the person is discharged is an aspiration. When you call the patient, you reinforce the use of the safety plan so that it's not just a piece of paper or an exercise they did in the inpatient unit. What are your triggering warning signs? It should be on their safety plan; they should have it with them. If they do not have it with them, then the caller can build the safety plan with them over the phone. Some health systems have said this is important enough that we are going to adopt this practice. Most health systems do not, though.

Sarah Mossburg: Why don't most systems do it?

Edwin Boudreaux: It is costly. They need a designated person to provide that telephone follow-up. Previously, the biggest barrier was that you could not get compensated for any of this work. But recently, starting in February of this year or mid-January, there are two new accepted billing codes for the Centers for Medicare and Medicaid Services that allow clinicians to bill for safety planning and for post-discharge follow-up intervention.⁸ We are hoping that it helps to counteract the cost barrier that has prevented health systems from adopting this practice. Health systems are businesses; they must make money, and this is an incentive to make money off of these practices. If you combine this incentive for gain with the incentive to avoid losses due to The Joint Commission’s expansion of the Sentinel Event window after discharge, then health systems might change their practices. Hopefully, we will see health systems adopt this practice more frequently.

Sarah Mossburg: This has been great. Is there anything that we did not discuss that you would like to cover before we close?

Edwin Boudreaux: When we are thinking of high-quality suicide-related care, we often get bogged down in the protocols. We talked about specific things like screeners, what is the best screener and how to choose a screener. Then we talked about things like post-visit calls. Often in health care, we get deep into the details of what that protocol looks like. That is a good thing, and we should be doing it, but it often loses sight of the patient in the process. We often forget that the most important thing we can do with a suicidal patient is to listen to them, respond with empathy, not rush them, to treat them like human beings. One of the major causes of suicide is a sense of disconnectedness, a sense of hopelessness, a sense of not belonging to society, family, or community. If we get too bogged down in these nitty-gritty details and we lose the person who is there, then we haven’t done much better than before. We must do both. We must make sure we know exactly what we need to do and are well-trained in how to do it. We also must remember the most important thing is treating a person with compassion, humanity, respect, dignity, and helping that person to feel like they are not being rejected because of their suicidality. It is a stigmatized condition, and people often feel rejected. Rather than feeling welcomed into a place where they can get care and be helped, they feel like they are being judged and that they are being passed through this machine rather than treated like a person.

Sarah Mossburg: I love that reminder to return to relationship-based patient-centered care as the fundamental core of what we are doing. Thank you so much.

Edwin Boudreaux: You are welcome.

References

1. Ahmedani BK, Simon GE, Stewart C, et al. Health care contacts in the year before suicide death. J Gen Intern Med. 2014;29(6):870-877. [Free full text]
2. Boudreaux ED, Camargo CA Jr, Arias SA, et al. Improving suicide risk screening and detection in the emergency department. Am J Prev Med. 2016;50(4):445-453. [Available at]
3. Boudreaux ED, Jaques ML, Brady KM, et al. The patient safety screener: validation of a brief suicide risk screener for emergency department settings. Arch Suicide Res. 2015;19(2):151-160. [Available at]
4. Screening for suicide risk. Zero Suicide Institute. Accessed February 27, 2025. [Available at]
5. Suicide prevention resources to support Joint Commission accredited organizations implementation of NPSG 15.01.01. The Joint Commission. Accessed February 27, 2025. [Available at]
6. Miller IW, Camargo CA Jr, Arias SA, et al. Suicide prevention in an emergency department population: the ED-SAFE study. JAMA Psychiatry. 2017;74(6):563-570. [Free full text]
7. Fleischmann A, Bertolote JM, Wasserman D, et al. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bull World Health Organ. 2008;86(9):703-709. [Free full text]
8. Medicare changes in 2025. American Psychological Association. Published January 31, 2025. Accessed February 25, 2025. [Available at]