One of the objectives of the Integrated Health Model (IHMI) is to support the representation of Function, State, Goals, and Outcomes in the context of health. IHMI is defining Functional Status, Health-related State, Patient Goals and Social Determinants of Health to ensure consistent capture of relevant health information.
Status message
A request has been sent to connect to this user.Patients and clinicians are setting Health Goals together and individually, but often lack tools to measure progress using health data.
So my interest revolves around Goals of Care discussions with patients facing major interventions. We have shown that using functional status measures that we can trigger the provider and the patient to have a detailed and in depth discussion about future Health Goals. The way forward is to provide these triggers to physicians and patients so that a meaningful conversation can occur. Conversely, many patients are very clear about what they want at for their care and what they want to see, its just isn't routinely captured and documents.
I agree with Jason. Also, conversations with patients about their goals is a good way strengthen the patient-provider relationship because it promotes patients' perceptions that the provider understands them, wants to hear what they have to say, and encourage shared decision making that takes the patient's preferences and concerns into account. In this way, the patient feels special.
We are looking at goal setting as a method of client intake for online therapy. That way, we can provision either telehealth services in more complex cases, or coaching for those requiring more basic maintenance care (like fitness, nutrition, or stress management services). Ultimately this model will result in savings for the payor.
Allowing the client to set many of their own goals seems to achieve better "buy-in" to the health behavior change process.
To follow up on Adam's point, we've identified many different behavior therapy/counseling and assessment interventions mapped to specific conditions and/or symptoms. No matter how a patient's health goals are captured, I suggest that they also be mapped to specific interventions, along with conditions and patient-reported symptoms.
The IHMI definition of Health-related States can be viewed here: files.ama-ihmi.org/image/authe...
Everyone has focused on classic co-morbidities prior to operations yet we have shown that Functional Status and Health related state are predictive of outcomes after operations. We have used these variables to identify and counsel patients prior to operation to improve surgical outcomes amongst multiple institutions.
With funding from PCORI, we developed and are testing a web-based patient assessment that integrates traditional review of system (medical history) with behavioral risks (e.g., smoking, BMI) and functional assesmsent using validated patient-reported outcomes for patients with advanced knee and hip osteoarthritis. A real-time patient and clinician report is provided to summarize medical and behavioral risks as well as pain and functional status to guide treatment decisions. The longitudinal use of the system allows patients and clinicians to monitor treatment impact on function. Finally, the patient health and functional profile is compared to patients in a national cohort (FORCE-TJR) and individualized estimates of surgical treatment effectiveness and risks are generated.
Functional status and Health-related State is a broad area and can mean many things. We have shown that with the use of ubiquitous technologies like the smartphone, you can collect deeper insight into patient functional status and state that have not been available before. For instance, using sensors and machine learning, one can guide a patient through collection of certain functional measures like Range of Motion. Additionally, with the smartphone camera and deep learning, you can begin to address social determinants like the state of the home which can contribute meaningfully to a patient's overall outcome. Better still, these methods allow for a two-way dialogue with patients. They can more directly relate to the information and take steps to address issues.
We believe that the current confluence of technologies and healthcare innovation can enable a much more continuous and on-going approach to patient care, and provide physicians and care-teams with insights to drive better outcomes. We are realizing this by combining traditional healthcare data and genetic data with real-time data from wearables (that can indicate activity, heart rate, etc.) as well as an individual's own assessment of how they are feeling. This holistic approach and gives clinicians the ability to assess an individual in a comprehensive way, and allows for treatment that is on-going and focused, vs. episodic and based on sparse clinical data (primary vitals).
Concur that we can collect now insane amounts of information and data and use deep learning. The issue for the patient in front of me in clinic are whether the information collected is valid and does it impact my recommendation for future care. The concern I have with the technology and measurement moving forward is what the biomechanists I collaborate with face on a daily basis...how to take the knowledge gained through advanced analysis and make it clinically applicable for patient care.
I completely agree. I think the data tsunami is inevitable and upon us to some extent. There is already a lot of data in the EMR (granted, not all of it is useful) that is not being leveraged. Wearables, mobile health, remote patient monitoring, genetics, etc. are only going to add to the data overload. I think there is an incredible opportunity to harness this data and turn it into information that results in better care - both from a clinician and patient perspective. It will take strong partnerships between tech and clinicians to strike the right balance and to establish trust and to ensure the solutions are providing useful and timely information.
Several have commented about the data deluge, which is clearly a concern, although I think it is actually more of a technical issue than a conceptual one. Improving patient outcomes, however, requires incorporation not just of more of data, or accurate data, or processed data. I would argue that the biggest missing piece here is patient generated data - experiences, concerns, functionality - that the patients deem important, even if the data don't "fit" into a standard clinical classification. This is not exactly the same as sensor data, which is really just more granularity on other measures like BMI or lab values. Incorporating patient reported outcomes directly into the medical record (and decision making process) allows patients' experience to actually drive the outcome measures!
I think there are two different elements here as you suggest. The first is the better availability and curation of sensor data - so that physicians can make good use of potentially valuable data. The second is information that is more subjective; how a patient is feeling, what is the environment they are in, how functional are they etc. Our approach to this is to enable individuals to journal and capture "how they are feeling" with a simple rating scale. I completely agree that this more subjective data is really important in the context of the outcome. As an example, a patient's sensor measurements may be in-range based on the prescription of some medication, but perhaps the meds make them feel nauseous. This sort of holistic approach, and bidirectional information flow and guidance is what we are focused on enabling.
Our approach is to collect and analyze biopsychosocial data that combines biomedical/objective data with subjective patient generated data about key life situations and the patient's psychological state (thoughts, emotions, goals, coping skills, etc.) related to those patient-prioritized situations.
In collecting information about social determinants of health and as IHMI considers how to incorporate such data into the model, how many of you are using Z codes (Z55-65) for individual and population assessments. If so, what is your opinion of the codes - enough to start with? what's missing? Are there better options to consider?
What do you think about T74s and T76s?
Kathleen. I agree that the Z codes represent a place to code for social determinants and other "factors influencing health status." What we have found when looking at large administrative and clinical datasets is that they are almost never coded. That's completely understandable, as ICD codes are primarily for billing. We have all got to figure out a way to capture these data within the current office workflow and without adding any more data entry burden to already overburdened Docs.
Hi Kathleen, great question. We agree that this area is important to focus on. The wide variety of methods and approaches to assessing and documenting this information across the continuum of care makes it exceptionally difficult to share information and compare populations/results. A consensus-driven standard for both use cases and coding is key to driving progress in this area.
I agree with William that overburdening clinicians with more data entry is bad. I wonder, however, if the ICD codes we're discussing are often absent due to the extra workload, due to failure of patient disclosure, and/or due to provider discomfort in discussing the underlying issues with the patient?
Patients, devices, and clinicians are all generating data to describe an individual’s Health-related State, but much of it is uncaptured or inaccessible.
The key question of which device shold be driven by what metrics matter. And in the literature perhaps the two most important are grip strength and walk speed. These two metrics have been shown to be predictive of multiple outcomes across multiple specialties. Therefore we advocate measurement of gait speed and grip strength due to ease of capture and standards which differentiate normal and abnormal.
We are using validated disease specific PROs that focus on arthritis pain and function, as well as a global assessment of functional status.
The IHMI definition of Functional Status can be viewed here: files.ama-ihmi.org/image/authe...
The definitions provide a solid framework for functional status and I believe ADL and IADL are a good place to start since these measure are validated and cross multiple boundaries in medicine. I think you can further explore functional status assessment with objective phenotypic measures such as walking speed and grip strength. While difficult to measure but important is the concept of life space which when combined with ADLS may be the best representation of true functional status.
ADL and IADL tasks are included in the validated PROs that we are using.