From the Editor
Studies show that measurement-based care improves patient outcomes, from better functioning to fewer readmissions. Not surprisingly, expert practice guidelines, including the CANMAT clinical guidelines for depression, recommend their use. Yet – despite years of enthusiasm – most of us don’t use scales or questionnaires to monitor symptoms and outcomes.
In a new systematic literature review, Christian Burr (of the Bern University of Applied Sciences) and his co-authors try to identify effective strategies by analyzing recent studies. In their Psychiatric Services paper, they focused on 16 studies in a systematic literature review. “Effective measurement-based care implementation requires context-sensitive strategies, interprofessional collaboration, and active involvement of patients, service users, and caregivers.” We consider the paper and its implications.
In the second selection from The British Journal of Psychiatry, Dr. Scott Monteith (of Michigan State University) and his co-authors write about the evolution of artificial intelligence. They focus on agentic AI (arguably the next big thing), seeing a clear role in education, yes, and in patient care – but they also note problems. “Agentic artificial intelligence systems offer a unique approach to supporting physicians in dealing with the challenges of clinical medicine but also pose validity, privacy, malpractice and security risks.”
And in this week’s final selection, Dr. Paniz Johari, a psychiatrist who works at Northwell Health, writes personally and thoughtfully about the loss of her patient to suicide. In JAMA, she discusses her grief and rumination. “The absence of anything remarkable made it harder to understand and harder to set aside.”
DG
Selection 1: “Implementation of Measurement-Based Care in Mental Health Services: A Systematic Literature Review”
Christian Burr, Sabine Hahn, Aaron M. Carr, and Andres R. Schneeberger
Psychiatric Services, 8 July 2026 Online First
Measurement-based care (MBC) is an evidence-based practice that uses validated rating scales and questionnaires to routinely and quantitatively monitor patients’ symptoms and outcomes over time. These data are then used collaboratively by clinicians and patients to guide ongoing treatment decisions… MBC improves mental health care at the patient, provider, and system levels… Successful MBC implementation requires alignment across clinicians, organizations, and health care systems because barriers at one level often interact with those at other levels. Despite providing benefits for patients, providers, and the health care system, MBC faces substantial challenges… Provider-level adoption of MBC into routine clinical practice has been slow and gradual. National surveys from several countries indicate that only 30%–50% of psychiatrists regularly use symptom rating scales or other standardized outcome measures… The reasons for the slow uptake of MBC are complex and involve barriers at the patient, provider, system, and policy levels. The need for leadership buy-in, training and coaching, health IT resources, and quality improvement efforts and the lack of an interprofessional conception also substantially affect implementation success. These multilevel challenges from implementation science research help explain the lack of MBC penetration into typical practice.
Recent studies have therefore focused specifically on implementation strategies to promote the sustained uptake and fidelity of MBC and its associated clinical benefits, with growing interest in tailoring its implementation to local contexts. In this systematic review, we synthesized evidence of interventions facilitating MBC implementation in adult psychiatric settings over the past 25 years. Given that research in this area is ongoing, our objectives were to describe implementation approaches and the personnel involved on the patient, provider, and system levels; determine the effectiveness of MBC at each level; and identify key barriers and facilitators on different levels. Although early MBC research emphasized clinical and system outcomes, recent literature increasingly prioritizes patient-reported experiences, aligning with the focus of the quintuple aims (i.e., of improving patient experience or population health) on health equity and patient-centered care.
So begins a paper by Burr et al.
Here’s what they did:
- The authors conducted a systematic literature review, following PRISMA guidelines, “aimed to synthesize evidence… on interventions supporting MBC implementation in adult mental health services.”
- They searched five databases, including PubMed.
- Studies examined measurement-based care (MBC) implementation in adult psychiatric inpatient or outpatient settings. Qualitative, quantitative, and mixed-methods studies were eligible.
- Exclusion criteria: studies involving primary care, children and adolescents, and non-psychiatric populations.
- Two pairs of authors independently screened abstracts and full texts. Data extraction included information on study design, participants, and interventions.
- Findings were synthesized thematically across system-, provider-, and patient-level factors.
Here’s what they found:
- An initial review yielded 1 120 records; after screening and review, 16 studies were included.
- Study characteristics. Most of the studies were from North America (n = 13); many involved outpatient care (12); depression was the most common diagnosis (5).
- Outcomes. MBC was associated with better clinical outcomes. Four of the five studies examining outcomes reported positive patient-level effects, including reductions in depressive symptoms, improved medication adherence, and better weight-related outcomes.
- Engagement. MBC increased patient participation in self-reporting – for example, from 10.8% to 57.1% in one study; a digital MBC system achieved 86% weekly check-in completion.
- Patient and family satisfaction. They appreciated it especially when MBC supported shared decision-making and communication.
- Clinician perspective. Providers generally viewed MBC positively and increasingly used measures in clinical decisions – in one study, psychiatrists rated PHQ-9 scores as helpful in 93% of patient visits.
- Barriers. Major barriers were time, training, organizational culture, clinician attitudes, and technology.
A few thoughts:
1. This is a good paper on a relevant topic and published in a solid journal.
2. The main findings: “Patients, service users, and their relatives must be systematically involved in the implementation and application of MBC interventions. Such involvement should span multiple levels, including system and instrument selection, definitions of the MBC process during the introductory phase, and the application of shared decision making.”
3. The secret sauce for MBC doesn’t seem so complicated.
4. But is it, say, a lack of engagement with patients and families that has undermined widespread adoption? Is the burden of paperwork simply too great? Or is part of the problem that MBC introduces greater accountability to providers and health systems?
5. Like all studies, there are limitations. The authors note several, including “insufficient diversity of patient populations and heterogeneity of the MBC interventions.”
The full Psychiatric Services paper can be found here:
https://psychiatryonline.org/doi/10.1176/appi.ps.20250065
Selection 2: “Artificial intelligence evolution in medicine”
Scott Monteith, Tasha Glenn, John Richard Geddes, et al.
The British Journal of Psychiatry, 8 July 2026 Online First
Artificial intelligence is a transformative technology included in the range of applications found in everyday life that are routinely used by the general public, as well as in products designed for physicians. In medicine, applications include artificial intelligence decision support tools, wearables and monitoring devices for physicians, patients and caregivers… A new type of generative artificial intelligence is agentic artificial intelligence, which extends the capabilities of generative artificial intelligence by making decisions and acting automatically in a human manner based on perception and reasoning. Agentic artificial intelligence is autonomous, can specialise in specific tasks, can adapt and learn from experience and can respond to natural language prompts. Agentic artificial intelligence can proactively initiate and reconfigure processes based on changing conditions.
So begins a paper by Monteith et al.
What are agentic AI agents? “Although there is no standardised definition, agentic artificial intelligence agents typically use LLMs to perform a wide range of tasks and actions, including decision-making, problem-solving and interacting with the product environment. Although artificial intelligence agents act independently, their actions are based on the predefined rules and goals defined by humans during training, and by the user who provides the specific goals and available tools.” They provide an example: “an agentic artificial intelligence agent will automatically ask questions of a user, look up the responses internally and determine whether it can respond or whether the question needs to be passed to a human.”
They consider potential uses in healthcare “including both physician- and patient-facing applications.” That said, they offer a note of caution: “The need for patient safety and privacy, the potential for serious errors and the tight regulatory environment pose major and unique challenges that, to date, have limited the adoption of agentic artificial intelligence in healthcare. Expansion of the use of agentic artificial intelligence in patient care will require ongoing research and testing. Unlike general-purpose artificial intelligence applications that interact with users, agentic artificial intelligence will need to be carefully and repeatedly validated, and safely integrated within existing healthcare systems. It is important that the physician remains ‘in the loop’ overseeing the agentic training, process implementation and outcomes. The need for rigorous oversight and validation includes the use of agentic artificial intelligence in medical situations that are currently automated with LLMs, such as laboratory medicine.”
“It is important to recognise the challenges related to implementing agentic artificial intelligence systems, which have the same limitations, risks, biases and privacy and cybersecurity concerns as standard artificial intelligence algorithms.”
With regard to implementation, they focus on several areas:
- Data. “Some of the critical challenges regarding agentic artificial intelligence in healthcare include obtaining high-quality, representative data for training the model while preventing bias.”
- Security. “Agentic artificial intelligence systems may introduce a variety of security risks, which may be magnified in poorly designed systems.”
- Workflow. “Successful implementation will require the challenging process of implementing agentic artificial intelligence into the interdisciplinary clinical workflow.”
“Physician oversight and involvement are required for successful agentic artificial intelligence implementation, which would require physician training and commitment.”
A few thoughts:
1. This is a good, concise paper on AI.
2. You should read this paper.
3. The authors are enthusiastic – and emphasize the role of physicians as AI continues to evolve. They also note the importance of training. “With the expansion of agentic artificial intelligence agents throughout healthcare, physicians need to understand its basic features. Increasingly, physicians will be working in environments that include agentic artificial intelligence models that may be acting autonomously.” Well said.
The full British Journal of Psychiatry paper can be found here:
Selection 3: “On Losing a Patient”
Paniz Johari
JAMA, 6 August 2026 Online First
The news arrived abruptly, almost clinically; a few words spoken in a fluorescent-lit room told me he was gone. I had lost a patient to suicide. I remember the weight of the chart in my hand, my grip tightening, and the indifferent hum of the printer. Still, there were patients to see, notes to write, and decisions to make.
When the team gently offered to let me go home, I hesitated. I didn’t need to stay, but I also dreaded the silence waiting for me outside. Staying at work felt safer. It allowed me to act as if nothing had changed, at least for a little while.
So begins a paper by Dr. Johari.
“There is a version of me that sits with patients, asks the right questions, waits before responding, and knows when to be gentle or direct. She lets silence stand. Her steadiness is rooted in repetition and training, and she was still there that day.”
She writes about her decision to push forward. “Moving through the afternoon guided entirely by habit, I listened, inquired about sleep, tracked intrusive thoughts, and methodically gauged risk. My voice betrayed nothing of what I had just learned.” She speaks of the end of the day. “I was not performing steadiness for anyone – no patients to protect, no role to maintain. Grief broke through, and I cried.”
She notes her realization: “even when everything is done right, you can still lose someone.” She adds: “That realization settled in gradually alongside the work, narrowing the reach of what I had been taught. I still moved through the same assessments, and listened for the same cues, but with less certainty about what they could or could not predict. I felt a growing awareness of the vast distance between identifying risk and changing outcomes. What once felt like safeguards now felt more like approximations – necessary, but not definitive.”
She speaks about our field. “In psychiatry, we are trained to be thorough, to spot patterns, and to intervene. We learn to formulate what we hear into something actionable, moving from narrative to a plan… Most days, this structure holds. It creates a sense of forward movement – the belief that something can always be done, and that the work is leading somewhere. But some moments resist that sense of resolution. They refuse to settle neatly into clinical narratives, remaining unmoved by the quiet reassurance that every possible action was taken.”
She discusses the patient. “I remembered his responses, his steady gaze, and the cadence of his speech. Even the way he rose from his chair was predictable. No hesitation. No lingering at the door. Later, I replayed that exit over and over again, searching for anything I might have missed. There was nothing.” She adds: “What unsettled me most was how ordinary the encounter felt.” There was “no rupture, no warning, no moment signaling what would follow.”
“I wanted the story to organize itself around a clear cause, something that would make the tragic outcome feel less arbitrary. After a loss like this, it is tempting to believe that if something can be explained, it can also be contained. But the details did not offer that kind of distance.”
She closes by mulling our work. “I also hold [the work] differently, with a sharper awareness of where it ends, even when that is hard to accept. This has not made the work feel smaller, only more precise in what it can and cannot hold. I don’t think that’s a loss. If anything, it feels closer to humility.”
A few thoughts:
1. This is a moving paper.
2. Most of us have lost a patient at some point in our work. The experience is painful and humbling – feelings that are well captured in this essay.
3. I can particularly relate to the search for an explanation. After the death of a young patient, I remember going over his chart dozens of times, trying to find my mistake. I never found one – in some ways, more horrifying than finding a mistake I could learn from.
4. Suicide has been discussed in past Readings. A recently selected paper from The British Journal of Psychiatry considered the Papageno effect – that is, whether widely disseminated stories of mastery of suicide crises and survival are followed by fewer suicides. That Reading can be found here:
The full JAMA paper can be found here:
https://jamanetwork.com/journals/jama/fullarticle/2852546
Reading of the Week. Every week I pick articles and papers from the world of Psychiatry.
Recent Comments