Major Depressive Disorder Podcast

'Primary Care First' Model to Screen and Treat Major Depressive Disorder

Madhukar H. Trivedi, MD; Manish K. Jha, MBBS


March 07, 2023

This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.

Madhukar H. Trivedi, MD: Hello. Welcome. I am Dr Madhukar Trivedi. I'm the founding director of the Center for Depression Research and Clinical Care here at UT Southwestern and the Peter O'Donnell Brain Institute. Welcome to season 2 of Medscape's InDiscussion series on major depressive disorder. Today I'm very excited to discuss the topic of the "primary care first" model to screen and treat depression. This is really going to be a very exciting and interesting conversation with one of my esteemed colleagues, Dr Manish Jha. Let me introduce you to him. He is an assistant professor of psychiatry at UT Southwestern. Listeners may also remember that Dr Jha was a guest on season 1 of this podcast, where he discussed novel treatments in major depressive disorder, especially ketamine, esketamine, transcranial magnetic stimulation, etc. So, welcome to InDiscussion. Let us jump into the discussion, Dr Jha, and maybe start with a little background on why we should talk about "primary care first" in terms of depression, major depressive disorder, and some idea of how common it is in primary care and why we should bother with this.

Manish K. Jha, MBBS: Thank you, Dr Trivedi. It's truly an honor to be here. Really, where it starts is that major depressive disorder, or depression commonly, is very prevalent. We know that 1 in 5 adults in the United States will suffer from it during their lifetime. It goes underrecognized, it goes underdiagnosed, it goes inadequately treated. Often the course is that of a chronic illness. Other chronic illnesses, like diabetes and hypertension, often get treated initially in primary care. And that was essentially the thinking by making a "primary care first" approach for identifying and treating depression.

Trivedi: Maybe a little more background on this. Obviously, one of the issues that we have run into traditionally with depression in primary care is that often the system is designed in such a way that we wait for a crisis. A patient comes in and has thoughts of death or suicide, who's having significant personal social problems. Or sometimes they really do report having significant depression, losing a job, etc. And it is done in a way where we are not routinely screening. Imagine if we did the same for high blood pressure. We would identify high blood pressure or hypertension when somebody has a heart attack or stroke. We don't do that. Same thing with diabetes. So say a little more about how we should be thinking about this in terms of not only primary care first, which is what we're talking about, but why we should even start thinking about routine universal screening like we do for a lot of common illnesses.

Jha: So we know that the US Preventive Services Task Force, or what we call the USPSTF, recommends routine screening for depression in both youths as well as in adults, and also in higher-risk populations like those with postpartum depression. But really, you're absolutely right. If we wait for a crisis situation to occur and then implement screening, we are too late. As the screening suggests in name, it should be applied to everyone who's coming through and at specific intervals. So maybe at least once a year administering simple rating scales or self-reported measures in clinical practice that can be done. And again, this is not new, but we know from data of maybe a decade ago that fewer than 1.4% of outpatients in US ambulatory practices were actually getting screened for depression. I think that number is slightly inching up but continues to be fairly low, which underscores the need to screen more effectively for depression in primary care or other settings.

Trivedi: And I think we showed in the STAR*D study 15 years back that if you use this model, "primary care first" and measurement-based care, then your outcomes, at least for the first two to three steps, are very comparable in primary care to what you see in specialty settings. Very similar, again, to other chronic medical diseases, like arthritis, hypertension, etc. And yet there still seems to be some remaining hesitation. I know you are involved with a lot of primary care networks. Speak on how you work with primary care settings to introduce this idea so that they are able to include screening and measurement-based care, and yet deliver the kind of care we would like for patients to get.

Jha: Absolutely. At the end of the day, there are three very simple things. There's screening. And then what happens after a screening is coming to a diagnosis and treatment. And that's where it starts with providers — just talking about how often the barrier exists, that if I screen for depression and someone screens positive for depression, what's the next step after that? And because of the limited access to mental health services, if the goal is that we just refer them out, then no one ever gets seen. I've heard you say that if you refer 100 people, 120 of them will never get seen or something like that, right? The barriers are often there. So, starting with screening and using a simple measure like the nine-item Patient Health Questionnaire (PHQ-9), or the first two items of that PHQ-9 (or PHQ-2), are really powerful and simple tools that can be used by people as they're coming in and getting checked in. And the program that you have led, called VitalSign6, really is making depression and anxiety screening a sixth vital sign. And then starting from there, once a person screens positive, how can we use simple tools to make an effective diagnosis? That piece does require some education and an ongoing collaboration so that the primary care providers do not feel that they just got 2 hours of lecture and now they have to manage, but rather that they have an expert who's available to consult with them. That piece about diagnosis comes, and then it leads to treatment, which we will touch on in a few minutes.

Trivedi: I think the whole idea of measurement-based care has been talked about, but it's not often practiced in primary care. Is it hard to do? What are the kinds of tools that primary care providers would need to do measurement-based care? And give us some sense of how you educate primary care providers in the North Texas area, where you've developed a lot of these networks. How do you educate them and partner with them?

Jha: So, first thing after a person screens is to get to a diagnosis. And there, what we really find helpful is a very simple thing called DSM-5 checklists. It's just a checklist of symptoms. And then also educating providers about some common differential diagnoses. So, making sure that bipolar disorder is ruled out before we make a diagnosis of major depressive disorder, keeping a person's circumstances in mind so that there are no medical issues that may account for those symptoms. The simple diagnostic checklist is the first step. And when a treatment is going to be initiated, that's where measurement-based care becomes very important. As I said, if you started with the PHQ-9, then we have an index of symptom severity at baseline that we recommend be checked at every visit that person comes to for their treatment. Symptoms are not the only measurements. If you're using medications, we also have to measure the side-effect burden and the adherence because they are really critical in making the next-step decision. If someone is having burdensome side effects, then we would want to address them instead of increasing the dose of medication. So really, the three legs of measurement-based care are monitoring symptoms, side effects and adherence, and consistently making changes to treatment or optimizing treatment based on that.

Trivedi: And so there are rating instruments on both adherence questionnaires for the side effects. One of them is called the patient adherence questionnaire. It's nothing fancy — a two-item questionnaire — and we've used it in many practices. The side-effect rating is to observe frequency, intensity, and burden of side effects. And again, those can easily be done routinely. They are self-rated, patients can do it, but you can then monitor progress like we do with hemoglobin A1c, with blood pressure, etc. I think the field is indeed rapidly changing.

I want to take the next 10 minutes to talk about two things. One is the idea that the answer to treatment for depression is not always medication. It could be therapy, it could be other options, especially as we go to youth and pediatric practices. The second part we'll come to in a minute, and that is the idea that Medicare now is paying for collaborative care. Texas is paying for collaborative care for Medicaid and therefore we have to bring behaviorists like licensed clinicians, licensed clinical social workers, into practice.

Jha: So the key step is that after a diagnosis has been made, we are going to initiate treatment. Which specific treatments to select? Often it may be a factor clinically; it could be a factor of what is the severity, or what are the available resources in the community? In people who have very severe depression, combining psychotherapy and medication seems to work better. For people with mild to moderate depression, either of those would be appropriate — pharmacotherapy or psychotherapy. And an active surveillance approach, which is consistently monitoring, may even be appropriate in someone who has mild depressive symptoms when they are being screened for depression. Obviously, as you mentioned, in pediatric age groups, the choice of medications that are available is fairly limited. So we have to be more careful about that. Side effects could be more burdensome and we have to be careful about that. There is a funnel of care involved. In people for whom the treatment is relatively straightforward, they can be managed by the primary care providers themselves. What we have done in our program is to layer the additional pieces on top of that. For example, we are bringing behaviorists to do behavioral activation. Sometimes we have teletherapies, so we can now combine evidence-based psychotherapy with medication management, which leads to far more improved care. So, I do not see primary care provider–first or collaborative-care models as in competition with each other; rather, they are fairly complementary. And really the key thing that gets missed is that screening is often undervalued. If you start with screening and then you start with effective diagnosis at the primary care level, that will solve a lot of access-related care issues later on.

Trivedi: In some ways it is a very exciting time because I think the landscape is changing. There are now, as I mentioned, collaborative care billing codes available. Behaviorists like a clinical social worker could join a primary care practice or a group. There are now many vendors available so that if a primary care practice wants to partner with a vendor who can do psychotherapy or behavioral therapy online, that is possible. It is becoming much more like how we treat other medical illnesses, where it starts in primary care but there is collaboration and back-and-forth referrals with specialists. That, I think, is the next really big advancement. Finally, as you know very well and people are beginning to hear about it, there are now very exciting digital therapeutics coming online that may be able to assist the primary care provider. We used to have a system of healthcare for depression where they were separate, so that mental health care was always done in a different arena and medical care in primary care. I think we are hoping that this will change.

Jha: Absolutely. I completely agree with you. I want to emphasize that for most people, depression is as chronic and debilitating an illness as high blood pressure, diabetes. We do not need to reinvent the wheel. We can adopt the practices that are there for all of these other things. Management of diabetes need not just be a prescription but may involve behavioral intervention and digital tools. Similarly, we do not need to restrict to one modality for management of depression, and by having a team-based approach, that allows us to get more effective care for people who suffer from depression.

Trivedi: There remains, unfortunately, after all these years, some skepticism about measurements and whether they really help us monitor. So maybe take a few minutes to try to use both PHQ-9, nine items that map depressive symptoms and anxiety, with GAD-7, the seven-item generalized anxiety disorder screener, and use those two rating instruments, if you can, to give some concrete numbers. And then how do you use that to monitor — and how accurate is it?

Jha: I want to emphasize that PHQ-9 and GAD-7 are really widely clinically used. And the important thing to emphasize is that their interpretation is fairly similar. So a score of < 5 on both PHQ-9 and GAD-7 is indicative of no or minimal symptoms, what we would use on PHQ-9 as the criterion for remission. That is, symptoms are either almost absent or minimal. A score of 5 to 9 on both of these scales is indicative of mild symptom severity; 10 to 15 is indicative of moderate symptom severity. GAD-7, because it has seven items, maxes out at 21. So scores over 21 on GAD-7 are considered severe. For PHQ-9 we have an additional category, so 15 to 19 is considered moderately severe and 20-plus is considered very severe. Again, these groups allow us to keep in mind the symptoms' severity in these different domains. The goal of care really is to get patients to the no-to-minimal range, which is 0 to 4.

So for example, I start working with someone who comes in with a PHQ-9 severity of 18 and a GAD-7 severity of 12, and I start them on an initial treatment. If they have the willingness, we engage with a behavioral interventionist. And then we check the symptom severity again in 4 weeks; 4 weeks is typical because that's how long conventional antidepressants take sometimes to be effective. And then if the PHQ-9 is down to 7 or 8, that means we are moving in the right direction. If the GAD-7 went up or has stayed there, that again is valuable information for decision-making about what should be the next step in treatment. Are there things we need to change? Are there updates to our treatment approach that need to be made? This iterative process begins and hopefully over a period of 6-8 weeks we are able to get symptoms down into the range of < 5, which is indicative of remission.

Trivedi: One thing to remind people of is that there is a myth that it may not be easy for us to recognize the change. I can safely say that somebody who walks into your office with a PHQ-9 score of 15 and gets down to 8 or 7, you can easily tell; their neighbors can tell, their family can tell. So these are very reflective of severity. And the goal should be to continue to monitor. These are self-rated. Therefore, any practice can set it up such that the patient coming in can fill it out in the waiting area or in the exam room before seeing the doctor. That's the other advantage of this. Increasingly there is software like VitalSign6 and others that can help you monitor the progress graphically.

We've really had quite a bit of, at least, awareness in adult primary care and family practice. More needs to be done; more treatment needs to happen so that people don't get ignored, because half the people with depression in the United States never get diagnosed. So this would help. In pediatric practices the uptake has been slow. It is beginning to get there. Hopefully these tools that we have been talking about will help quite a bit.

The third area, which is becoming much more exciting in some ways, is the postpartum area. In the remaining minutes, give us some information about these exciting new treatments that are coming along for postpartum depression, like brexanolone. I think that landscape is also likely to change quite rapidly because we now have good new treatments.

Jha: Absolutely. One of the challenges often had been, why screen if we cannot diagnose, and why diagnose if we cannot treat? And that is rapidly shifting. As you said, brexanolone has an FDA approval specifically for postpartum depression. There are also other medications from that company that are orally active. For brexanolone, one of the limitations is that it's a 60-hour–long infusion. But there are now drugs that are orally bioavailable which are in phase 3 trials. Hopefully we'll have more options there for the management of postpartum depression and maybe even for the management of major depressive disorder. These along with other newer medications are also paradigm-shifting, whereby the onset of effect may be much more rapid. And that has often been a challenge in engaging people in care: Take this medication and trust that it will lead to some side effects early on, but the improvement will happen in 6-8 or 10-12 weeks. But now with faster-acting options, that will likely improve adherence and get people more involved and engaged in care. These are really promising and exciting times in this field.

Trivedi: I think there are three domains of really fascinating excitement coming up. One is what you just mentioned: rapidly acting antidepressants. Brexanolone is one example. Ketamine, or esketamine, is another. There is a lot of interest in psychedelics. There are also other molecules being looked at for rapid onset of action because that reduces concerns for the provider, but more importantly, for patients. The second big excitement, in my view, is this whole idea that we may be coming to a threshold where, beyond that, we will start identifying treatments that can be used for short courses of treatment, as episodes of treatment, like we do with transcranial magnetic stimulation. For example, in electroconvulsive therapy, we treat patients for a period of time — some for several weeks, 4-6 weeks — at the end of which that episode of care has led patients to feel better. They may have to return to it in the future, but they're not continuing the treatment forever. And the same thing is happening pharmacologically. Brexanolone is leading in that, and esketamine, but more treatments are being studied for that. And the third biggest thing, which we talked about last season several times on several episodes, is that we are really getting much closer to being able to identify biomarkers: blood-based, proteomic, metabolomic, RNA sequencing, as well as in stool samples to look at gut microbiome imaging. EEG is something we talked about a lot in the first season. That allows you to match patients very rapidly to the right treatment.

I am so pleased that you were able to join today, Dr Jha. Any final thoughts on our topic of primary care first? Should I, if I am listening and I'm a primary care physician, really figure out how to do this?

Jha: Start with screening. Follow that and reach out to providers, a psychiatrist in your community, to establish some connections so that you can effectively start managing patients yourself. And then seek out opinions in challenging or more complex cases.

Trivedi: I think everybody talks about this as an access problem. There aren't enough psychiatrists and psychologists, and I think I am convinced that it is a triage problem. So if we do what we do with hypertension, diabetes, and other common illnesses, and have a fully emphasized triage through primary care and then to specialty care, I think we will have much better outcomes for the population.

Thank you for joining. Today we have talked to Dr Manish Jha about the "primary care first" model, emphasizing measurement-based care in primary care settings so that care can extend to a larger population and then triage on to specialty care when needed. Thank you for tuning in. If you have not done so already, take a moment to download the Medscape app to listen and subscribe to this podcast series on major depressive disorder. This is Dr Madhukar Trivedi for InDiscussion. Thank you.

Listen to additional seasons of this podcast.



USPSTF Screening for Depression in Adults

Depression Screening Patterns, Predictors, and Trends Among Adults Without a Depression Diagnosis in Ambulatory Settings in the United States

Primary versus Specialty Care Outcomes for Depressed Outpatients Managed With Measurement-Based Care: Results From STAR*D

The PHQ-9: Validity of a Brief Depression Severity Measure

VitalSign6: A Primary Care First (PCP-First) Model for Universal Screening and Measurement-Based Care for Depression

A Brief Measure for Assessing Generalized Anxiety Disorder: The GAD-7

Prevalence and Impact of Diagnosed and Undiagnosed Depression in the United States

FDA Approves First Treatment for Post-partum Depression

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.