PodMed Double T is a weekly podcast from Texas Tech. In it, Elizabeth Tracey, director of electronic media for Johns Hopkins Medicine, and Rick Lange, MD, president of the Texas Tech University Health Sciences Center in El Paso, look at the top medical stories of the week. A transcript of the podcast is below the summary.
This week’s topics include arthritis and depression, AI and physician decisions in hospitalized patients, five ways for doctors to practice presence with patients, and healthcare hotspotting.
Program notes:
0:48 Healthcare hotspotting
1:45 95% of individuals did receive prescribed care
2:56 Five strategies to foster physician presence
3:58 Prepare with intention
4:56 Physicians would concur
5:58 Many barriers
6:19 Hospital-based AI system
7:15 6500 patients
8:13 Guarded about such systems
9:10 Arthritis and depression
10:20 Lots in the southern states in the U.S.
11:22 Results in poor compliance
12:22 End
Transcript:
Elizabeth Tracey: Practices to help physicians practice presence and connection with their patients.
Rick Lange, MD: Can hospital-based computers help improve decision and patient outcomes?
Elizabeth: When depression and arthritis coexist, does looking at state-level data help?
Rick: And does healthcare hotspotting really work?
Elizabeth: That’s what we’re talking about this week on PodMed TT, your weekly look at the medical headlines from Texas Tech University Health Sciences Center in El Paso. I’m Elizabeth Tracey, a medical journalist at Johns Hopkins, and this will be posted on January 10th, 2020.
Rick: I’m Rick Lange, President of the Texas Tech University Health Sciences Center in El Paso, where I’m also Dean of the Paul L. Foster School of Medicine.
Elizabeth: Rick, how do you feel about starting with the New England Journal of Medicine, this issue of healthcare hotspotting? What in the world is that?
Rick: This is a term that many of our listeners may not be familiar with. About 5% of the population accounts for about 50% of annual spending, and about 1% accounts for almost a quarter of annual spending. There’s been a lot of interest in trying to identify those “superutilizers” of healthcare to see if we can intervene and reduce healthcare costs.
What happened is there were 800 hospitalized patients that were considered “superutilizers,” which is they’d been hospitalized within the last 6 months. They were randomized during their hospital stay to either usual care after discharge or to have a group of nurses, social workers, community health workers, and physicians visits scheduled very early to see if we could reduce hospital readmission rates. These were medically and socially complex patients.
What they determined is the readmission rate was virtually similar over the next 6 months, about 62%. This despite the fact that 95% of the individuals that were enrolled in the hotspotting actually received requisite care that was prescribed. This is really disappointing.
Elizabeth: Let me ask you. I’m going to throw this back to your clinical expertise then and say, “What do you think might help if this comprehensive approach doesn’t?”
Rick: That’s a great question. I’m not sure that I have the answer to it. What I can say is that even though 95% of patients had at least three encounters with the program staff, and the patients received an intensive intervention averaging about 7.6 home visits, they were less likely after discharge to be seen within 5 days after their hospital discharge or to see their primary care physician within 7 days.
Elizabeth: I’m just going to interject that there’s certainly an awful lot of promise that’s being realized with telemedicine and maybe those kinds of strategies and employing technology might help with regard to this strategy.
Rick: I’m a little concerned if the face-to-face didn’t work that doing telemedicine would be any more helpful. I think all of us agree that having a 62% readmission rate in the next 6 months is not acceptable. It’s not good for the patient and it’s not good for the healthcare system, so anything else that we can do I think needs to be tested rigorously.
Elizabeth: Let’s turn to the Journal of the American Medical Association, speaking of telemedicine and then flipping entirely to another topic, which is these practices to foster physician presence in connection with patients and the clinical encounter. There are lots and lots of discussions, of course, about burnout and about what is going to bring joy back to the practice of medicine for many physicians.
They did a systematic literature review and then a subsequent additional review to identify other studies looking at effective interpersonal interventions. They also observed primary care encounters in three diverse clinics. They had qualitative interviews with physicians and with non-medical professionals whose occupations involve intense interpersonal interactions, including firefighters, social workers, and chaplains, I’m happy to say.
After they finished putting all of this stuff together, they came up with a final set of five recommendations and I’m going to identify all of them. The first is they say to physicians, “Prepare with intention.” That is take a moment to prepare and focus before you greet your patient. Listen intently and completely. Sit down, lean forward, and do not interrupt.
They cite that famous study that shows that physicians, in general, wait about 11 seconds before they interrupt a patient. The third is to agree on what matters most. Find out what the patient cares about and incorporate these priorities into the visit agenda.
The fourth is connect with the patient’s story, considering their life circumstances that might be influencing their health, and acknowledge their positive efforts like they’ve stayed on their anti-hypertensive meds. Finally, the fifth one is to explore emotional cues — all of the things that the patient is manifesting — both with facial expression and emotion that are telling you, “Hey, here’s how this is all going.”
Rick: The things you’ve mentioned are things that foster the connection between patients and physicians. I can tell you that physicians, in general, would concur. That’s really desirable.
The trouble is there are a lot of barriers that prevent this from happening. For example, electronic medical records, which gets in the way of accomplishing these things. The time crunch that physicians have and having medically complex patients, so that seeing them in 10 and 15 or 20 minutes doesn’t allow you to do all these things.
Elizabeth: Of course, in this study, they note that an average visit has actually increased in length by about 5 minutes over the last several years, so five minutes is pretty significant when you’re taking a look at a 15- or a 20-minute time window. I think some of the time crunch, from my perspective, is really in the stuff that they would like physicians to do before they come in the room like review the patient’s story. Who is this person? Review some of their previous data. Those things are hard to do, I think, as you’re going from one room to the next.
Rick: Right, because you have to have time in between patients to do that. Ideally, many of us in practice come in early in the morning to go through all the medical records, and these things all take time. I agree that these five things — there’s no doubt they improve connectedness. However, just putting them out there without removing some of the barriers we talked about won’t be successful in the implementation.
Elizabeth: It’s good, though, that we’re paying attention. Let’s move to your next one.
Rick: This is a study that looked at the effectiveness of what’s called a hospital-based computer decision support system on both clinical recommendations and patient outcomes.
Elizabeth: That’s in JAMA Network Open.
Rick: We’ve talked about electronic medical records and how they can be used. There is a thought that using evidence-based medicine, combining it with the power of a computer to alert physicians about how they ought to practice, would inform their decisions and improve patient outcomes.
Let’s say the computer can look at the different medications, what the patient’s diagnosis is, and make recommendations to the physician about what medications they ought to be on, or what dose they ought to be on, or are there drug-drug interactions. The computer would automatically do that, relieve the physician of doing that, and that could provide a suggestion to the physician that he or she would take, and that could improve patient outcomes.
They used a computerized clinical decision support system and they compared that to a group of patients receiving the same care in which the computer decision support system was not available.
They looked at almost 6,500 patients who were admitted to internal medicine. Half of those received reminders to the physician based upon the computer decision, and that resulted in over 28,000 physician reminders — about three reminders per patient per hospital stay. What that resulted in was the physician changing their decision in about 4 out of every 100 patients. However, it didn’t improve patient outcome.
Elizabeth: How do you explain that?
Rick: Elizabeth, there are a lot of barriers to making this work. Keeping up with what all the evidence is, is very difficult. Incorporating that into a computer decision analysis is very difficult because it’s very complex, so this is what we call a long run for a short slide — a lot of effort and you have to update all this clinical-based evidence. You have to upgrade the computer system and it didn’t really benefit the overall patient outcome.
Elizabeth: Would you say that you’re optimistic, though, about the future development of these kinds of tools in order to assist clinical decision-making?
Rick: I’d be guarded. People talk a lot about artificial intelligence and how that’s the answer. Intelligence is not artificial. It’s intentional.
Elizabeth: I guess there are just two other things I would add about this. One of them is this adage that I hear again and again which is that AI is only as intelligent as what you put in there. These are algorithms that are created by people with a bunch of factors and a bunch of data population that people select.
There’s already a bias that’s inherent in it, and I’m not sure we’re all that aware of all the biases. Then the other thing is, I’m wondering if other outcomes might be different among the patients that we’re not necessarily measuring here, but that could still be worth achieving, for example, polypharmacy and reducing that.
Rick: There may be benefits. Some of them will be softer. Some of them will be harder. In the overall, I kind of want to say, “Is the patient better?”
Elizabeth: Let’s finally then turn to a Morbidity and Mortality Weekly Report from the CDC and that’s the association between depression and arthritis. They cite at the beginning of the paper that about 23% of U.S. adults have provider-diagnosed arthritis. That number is projected to rise to considerably more people by 2040. These chronic pain conditions, of course, are associated with poorer mental health and especially with anxiety and depression.
The CDC took a look at, on a state level, among a cohort who were diagnosed by a provider with arthritis the incidence of depression and anxiety. They were able to generate some rather interesting data that shows that in Hawaii, that’s where we have the lowest rate of depression among adults with arthritis — I think if I lived in Hawaii with arthritis I’d still be really happy — to a high of 32% in Kentucky.
There was a disproportionate amount of this, if you will, that took this coincidence of depression, anxiety, and arthritis that seems to be down there in the southern states, and that’s the same place where we seem to have a lot of other trouble with other things. A lot of chronic health conditions and obesity. It points to a need for at least assessment and awareness of this.
Rick: The background of this is they used a surveillance system — what’s called the 2017 Behavioral Risk Factor Surveillance System — in which they conducted surveys of hundreds of thousands of individuals. They looked at the incidence of mental distress and depression just across the United States and that was 11% and 19%. They looked at those with arthritis and those numbers were 17% and 32%.
We know that people that are depressed or have mental distress are less likely to adhere with their medications. They’re more likely to have a poorer outcome. Your admonition to screen these patients is very good. Identifying them early and getting them on treatment will be really important to help in controlling their pain and to treating their arthritis.
Elizabeth: The other thing I would say about this that is a question clearly not answered by this data is what’s the chicken and what’s the egg? Does a patient become depressed because they have arthritis, it’s painful, and it limits their activities? Or alternatively do depression and anxiety lead to pro-inflammatory conditions that can result in arthritis?
Rick: Some of that may be as a result of inflammation. There’s no question about it. Some of it is the result of the chronic pain and their circumstances as well. There certainly is an interrelationship. People with arthritis are more likely to have depression, mental anxiety, and less likely to be, again, compliant and adherent with not only medications, but physical therapy as well, and that’s going to affect their disease process. There’s no question there’s a relationship between the two.
Elizabeth: If you have arthritis, think about being screened for depression?
Rick: I think so. More importantly, the primary care providers should be doing routine screening in these patients.
Elizabeth: On that note, then, that’s a look at this week’s medical headlines from Texas Tech. I’m Elizabeth Tracey.
Rick: And I’m Rick Lange. Y’all listen up and make healthy choices.
Source: MedicalNewsToday.com
