Healthcare costs

Archive for Healthcare costs

Machine Learning in Medicine

As a new intern, I remember frequently making my way to the Emergency Department for a new admission; “Chest pain,” the attending would tell me before sending me to my next patient. Like any good intern I would head directly to the paper chart where I knew the EKG was supposed to be waiting for me, already signed off on by the ER physician. Printed in standard block print, “Normal Sinus Rhythm, No significant ST segment changes” I would read and place the EKG back on the chart. It would be later in the year before I learned to ignore that pre-emptive diagnosis or even give a thought to about how it got there. This is one of many examples how machine learning has started to be integrated into our everyday life in medicine. It can be helpful as a diagnostic tool, or it can be a red herring.

Example of machine-learning EKG interpretation.

Machine learning is the scientific discipline that focuses on how computers learn from data and if there is one thing we have an abundance of in medicine, data fits the bill. Data has been used to teach computers how to play poker, learn laws of physics, become video game experts, and provide substantial data analysis in a variety of fields. Currently in medicine, the analytical power of machine learning has been applied to EKG interpretation, radiograph interpretation, and pathology specimen identification, just to name a few. But this scope seems limited. What other instances could we be using this technology in successfully? What are some of the barriers that could prevent its utilization?

Diagnostic tools are utilized in the inpatient and outpatient setting on a regular basis. We routinely pull out our phones or Google to risk stratify patients with ASCVD scoring, or maybe MELD scoring in the cirrhotic that just got admitted. Through machine learning, these scoring systems could be applied when the EMR identifies the correct patient to apply it to, make those calculations for the physician, and present it in our results before we even have to think about making the calculation ourselves. Imagine a patient with cirrhosis who is a frequent visitor to the hospital. As a patient known to the system, a physician has at some point keyed in the diagnosis of “cirrhosis.” Now, on their next admission, this prompts this EMR to automatically calculated and provide a MELD Score, a Maddrey Discriminant Function (if a diagnosis of “alcoholic hepatitis” is included in the medical history). The physician can clinically determine relevance of the provided scores; maybe they are helpful in management, or maybe they are of little consequence depending on the reason for admission. You can imagine similar settings for many of our other risk calculators that could be provided through the EMR. While machine learning has potential far beyond this, it is a practical example where it could easily be helpful in every day workflow. However, there are some drawbacks to machine learning.

Some consequences of machine learning in medicine include reducing the skills of physician, the lack of machine learning to take data within context, and intrinsic uncertainties in medicine. One study includes that when internal medicine residents were presented with EKGs that had computer-annotated diagnoses, similar to the scenario I mentioned at the beginning of this post, diagnostic accuracy was actually reduced from 57% to 48% went compared to a control group without that assistance (Cabitza, JAMA 2017). An example that Cabitza brings up regarding taking data in context is regarding pneumonia patients with and without asthma and in-hospital mortality. The machine-learning algorithms used in this scenario identified that patients with pneumonia and asthma had a lower mortality, and drew the conclusion that asthma was protective against pneumonia. The contextual data that was missing from the machine learning algorithm was that the patient with asthma who were admitted with pneumonia were more frequently admitted to intensive care units as a precaution. Intrinsic uncertainties in medicine are present in modern medicine as physician who have different opinions regarding diagnosis and management of the same patient based on their evaluation. In a way, this seems like machine-learning could be both an advantage and disadvantage. An advantage this offers is removing physician bias. On the same line of thought, it removes the physician’s intuition.

At Wellpepper, with the Amazon Alexa challenge, machine learning was used to train a scale and camera device (named “Sugarpod“) in recognizing early changes in skin breakdown to help detect diabetic foot ulcers. Given the complications that come with diabetic foot ulcers, including infections and amputations, tools like this can be utilized by the provider to catch foot wounds earlier and provide appropriate treatment, ideally leading to less severe infections, less hospitalizations, less amputations, and lower burden on healthcare system as a whole. I believe these goals can be projected across medicine and machine learning can help assist us with them. With healthcare cost rising (3.3 Trillion dollars in 2016), most people can agree that any tools which can be used to decrease that cost should be utilized to the best of its ability. Machine learning, even in some of its simplest forms, can certainly be made to do this.

Posted in: Healthcare costs, Healthcare Technology, Healthcare transformation

Leave a Comment (0) →

Investing in primary care

The US healthcare system is an underperformer (highest healthcare spending for the lowest health system performance) compared to the other ten economically advantaged countries primarily due to differences in access, administrative inefficiency, disparities in healthcare delivery, and also due to the illogical underinvestment in primary care. Despite evidence by the Dartmouth Atlas of Health that the regions in which a higher percentage of Medicare beneficiaries receive majority of their care from a primary care physician lends to overall lower costs, higher quality of care, and lower rates of avoidable hospitalizations, the US continues to underinvest in primary care relative to other nations. Because of perverse incentives and overall fragmentation that is rampant in American healthcare, conscious and deliberate effort is needed to keep primary care at the forefront of clinical practice and population health improvement, including:

  • Implementation of quality improvement practices that have a theoretical basis
    According to Harvard Medical School’s Center for Primary Care established in 2011, there are five components necessary in improving primary care including evidence-based change concepts and tools, fostering strong relationships within and across practices, simple systems for reflection and feedback, structured time for team discussion and planning, and regular and meaningful engagement of leaders. The general theme is that quality improvement processes that have been validated (e.g. PDSA cycle) and implementation of driver diagrams that break up larger processes into smaller chunks/concepts have value and are worth the time to problem solve.
  • Prioritizing patient-centered care
    Care should be collaborative with patients’ preferences and values in the context of their socioeconomic conditions being respected. If there is less information asymmetry in clinical practice, then patients can be more active participants in their healthcare. Overall quality would improve with cost savings, as patient engagement research has demonstrated. Truly understanding a patient’s capacity and health literacy will improve a primary care physician’s ability to be effective in delivering patient-centric care.
  • Payer reimbursement for provider innovation in preventive and multidisciplinary care
    Primary care prioritization with the US healthcare system depends on heavy investment from payers because of the nature of reimbursement for clinicians’ time and services. In addition to a value-based compensation model that payers like Blue Cross Blue Shield reward providers with, more creative and interdisciplinary measures could be more payer driven. Humana’s Bold Goal program is a partnership between an influential payer and San Antonio Health Advisory board to partner with HEB grocery stores, community clinicians, and the YMCA to increase patients with diabetes’ better nutritional understanding of their choices. Because of the cost savings involved with more investment in primary care, it would make sense that payers would be incentivized towards this trend.
  • Leveraging of non-clinical members of a team to deliver comprehensive, value-based care
    Substantial evidence suggests that patients do not receive all of the preventive and chronic disease care that the U.S. Preventive Services Task Force advises on the basis of its best evidence because clinicians simply don’t have the time. Oak Street Health is a Chicago based network of value-based primary care centers that developed a clinical informatics specialist program 2014 where technical scribes were able to provide evidence-based recommendations and data support which resulted in improved effectiveness metrics, overall operational efficiency, and physician joy of practice.

Investment in primary care is necessary for the US healthcare system to have improved outcomes. Efforts at the community level, reinforced by theoretical models and financially backed by payers, are necessary in making changes that can yield significant population health improvements.

Posted in: Healthcare costs, Healthcare Policy, patient engagement

Leave a Comment (0) →

Pointing Fingers at Healthcare Problems

I’m only halfway through Elizabeth Rosenthal’s “An American Sickness: How Healthcare Became Big Business and How You Can Take It Back” which means that I haven’t gotten to the “what you can do about the problem” part. It’s a slow read, not because it’s not compelling but because it’s too compelling, and if like the current President, you were surprised at how complicated healthcare is, this book will do nothing to dissuade you. It’s really really complicated.

So far, I have two main takeaways from the book, that are easily illustrated through my recent experience of breaking and dislocating my finger: a simple, non-life-threatening problem, that unearthed a couple of key dysfunctions and unintended consequences.

My first takeaway is that everyone is complicit, and yet seem to manage to finger point at everyone else. Rosenthal spares no punches in unearthing decisions that are not made with the best interest in of the patient at heart. Providers, healthcare organizations, payers, pharma, and employers all are complicit in the mess that is our current healthcare system.

This past fall, I broke and dislocated my finger. It wasn’t a big deal, but because it happened on a Saturday night, my only option for care was at the ER. Last week I received a letter in the mail from my insurance company, that according to the envelope required my urgent reply. In the letter, the insurance company suggested that perhaps someone other than them may be on the hook for my ER bill. While I understand they wanted to make sure this wasn’t a worker’s compensation claim, the form was basically for me to tell them whose fault my injury was so that they could go after another insurance company to pay. This was a sports injury in a game of Ultimate Frisbee, a game so granola-like that there are no referees: players call fouls on themselves. . No one was at fault, and even if they were, I would never have considered suing. However, the form didn’t give me that option: only gave me the option of saying whether I had settled my claim. I created a new box that said “NA” and checked it.

When I received the letter, I couldn’t help but think back to Rosenthal’s book, and also consider the amount of effort and cost that was going into finding someone else to blame and pay. Just imagine what this effort and cost would have been if there were legal action….

The second takeaway is that the original intention of a decision always has much farther reaching implications than anyone who agreed on what seemed like a reasonable decision though. Again with the finger, I was asked a number of times if I wanted a prescription for OxyContin. I did not. As has been well publicized we have an opioid addiction problem in North America. While my finger hurt, aside from morphine during inpatient for an appendectomy, I hadn’t had opioids, and really didn’t think that it was necessary, which I explained to the physician. It wasn’t. Tylenol worked fine—however, it seemed that it was very important that I be the one to make this call, not the physician.

One of the unintended consequences of patient satisfaction scores may be the over prescription of pain medication, as many of the questions on the HCAHPS are about whether the patient’s pain was well managed. In Rosenthal’s book, I was also surprised to learn that a finger fracture where an opioid is prescribed has a different billing code than if it is not prescribed, and that with the fracture plus opioid billing code, hospitals get paid more. Now, if you are wondering how this may be the case, if you think about it, a fracture that requires an opioid must be more severe than one that doesn’t and therefore the billing code reflects the severity. This is exactly where the unintended consequences of billing codes can result in exactly the wrong behavior for patient care and safety.

It’s quite possible that the physicians on duty were not aware of either of these two drivers for prescribing, especially the billing code one. They may have just been told “this is our standard of care” and were following guidelines.

If a simple finger fracture and dislocation can shine a light on two key problems in our healthcare system, just imagine what else is out there. Actually, you don’t have to, just get a copy of Elizabeth’s book yourself, and let’s compare notes when I get to the part about what the fix is. It’s going to take all of us.

Posted in: Health Regulations, Healthcare costs, Healthcare Disruption, Healthcare Legislation, Healthcare Policy, Healthcare transformation, Opioids

Leave a Comment (0) →

EvergreenHealth: Evolving Care Outside The Clinic for Better Outcomes

In 2016 we formally announced our collaboration with EvergreenHealth to deliver interactive care plans for Total Joint Replacement.

“Across our organization, we strive to be a trusted source for innovative care solutions for our patients and families, and our partnership with Wellpepper helps us deliver on that commitment,” said EvergreenHealth CEO Bob Malte. “Since we began using Wellpepper in 2014, we’ve seen how the solution enhances the interaction between patients and providers and ultimately leads to optimal recovery and the best possible outcomes for our patients.”

EvergreenHealth is an integrated health care system that serves nearly 1 million residents in King and Snohomish counties in Washington State, and offers a breadth of services and programs that is among the most comprehensive in the region. More than 1,300 physicians provide clinical excellence in over 80 specialties, including heart and vascular care, oncology, surgical care, orthopedics, neurosciences, women’s and children’s services, pulmonary care and home care and hospice services. With expansion into more rural areas, and a catchment area that serves Seattle’s ‘eastside’ home to Microsoft and other major technology companies, delivering virtual care is both an imperative for an an expectation of EvergreenHealth patients.

Since our initial announcement, we’ve seen thousands of patients complete care plans and outcome surveys, and expanded within the musculoskeletal service line to include preventive care, spine surgery, and general rehabilitation.

User Experience

EvergreenHealth has a white labeled version of the Wellpepper patient application called MyEvergreen and available in Android and Apple App Stores. Clinicians use the Wellpepper clinic portal, and receive alerts to their email inbox if patients report any issues or unexpected outcomes.

EvergreenHealth has deployed care plans based on their own clinical best practices. 

Outcomes

  • Thousands of patients have used Wellpepper interactive care plans at EvergreenHealth
  • Interactive care plan users show higher scores on standardized outcome reports than those tracking outcomes without an interactive care plan
  • EvergreenHealth patients show a higher engagement level than Wellpepper’s overall 70% engagement

I would not want to have another knee surgery without the app. I was 81 and it wasn’t hard for me at all!

Total Knee Replacement Patient at EvergreenHealth

Technology

This deployment used a white labeled Android and iOS application for patients, and a clinic portal for clinicians. Patient invitation is synched with the Cerner medical records software using an ADT feed. Clinicians are notified of patients requiring additional help with an email alert. Wellpepper’s entire HIPAA secure platform was leveraged for this implementation, and EvergreenHealth deployed custom care plans based on their own best practices. They continue to add innovative features as they are added to the Wellpepper platform.

Posted in: Exercise Physiology, Healthcare costs, Healthcare Technology, HIPAA, Interoperability, M-health, Outcomes, patient engagement, Prehabilitation, Seattle

Leave a Comment (0) →
Google+