19 Aug 2021
0

Predictive Analytics Assist with Chronic Disease Prevention

August 19, 2021 | 0 Comments | , ,

The healthcare system has an immense wealth of information at its digital fingertips. Big data is constantly expanding from sources such as digitized patient records, patient wearables, medical apps, genome datasets, monitoring devices, and more. A critical challenge facing hospitals and health systems today is in effectively identifying strategies and personnel to utilize big data in a way that influences clinical care. Those that succeed in this task will find themselves in a much better position to advance care and improve patient outcomes.

One developing strategy to convert big data sets into improved patient outcomes is the use of predictive analytics, an approach that differs from what many hospital quality improvement departments are currently utilizing. For example, the Michigan Value Collaborative (MVC) Coordinating Center has been helping hospitals identify opportunities for quality improvement since 2013 by aggregating and analyzing payor claims data and presenting the results on the registry and in analytics reports. The goal of these efforts is to help hospitals compare utilization against peers and draw important insights across a range of medical and surgical procedures. This retrospective approach helps MVC members to learn from their past performance in order to pursue meaningful, observable improvements within their buildings. It is one piece of the big data puzzle. Predictive analytics, on the other hand, allows clinicians to utilize big data before their patient experiences significant healthcare services or treatments. As its name denotes, this approach identifies prevention opportunities before the incidence of disease by predicting a patient’s risk. This is especially important for diseases that require early detection for optimal treatment and survival.

Unlike Robotic Process Automation (RPA), which is also on the rise in health systems across the country, predictive analytics is performed by Artificial Intelligence (AI). This means that computer systems will perform tasks typically requiring human intelligence, including analyses and decision-making. In some ways, this strategy mimics what physicians have long been doing at a patient’s bedside: collecting a patient’s medical history and risk factors in order to tailor their treatment and advice. This process is essential in evaluating a patient’s risk of developing chronic diseases, which often run in their family or are more likely due to socioeconomic factors. An article from the University of Illinois Chicago posits that predictive analytics represent a significant potential for cost savings if they help clinicians and their patients prevent the onset of chronic diseases, one of healthcare’s costliest areas.

“On a population-wide level, predictive analytics can help greatly cut costs by predicting which patients are at higher risk for disease and arrange early intervention, before problems develop,” the article stated. “This involves aggregating data that are related to a variety of factors. These include medical history, demographic or socioeconomic profile, and comorbidities.”

The Centers for Disease Control and Prevention (CDC) states that, “90% of the nation’s $3.8 trillion in annual health care expenditures are for people with chronic and mental health conditions.” So the potential cost savings from reducing chronic disease treatment are significant.

Using predictive analytics in a clinical setting can leverage both patient records and socioeconomic factors. Medical records will often include family history of chronic diseases such as cancer, diabetes, and heart disease, which would make a patient more likely to develop the condition themselves. In addition to family history, a patient’s socioeconomic factors (e.g., education, employment, and environment) and lifestyle choices are significant predictors of chronic disease. A study in the American Journal of Preventive Medicine outlines how researchers used predictive analytics to screen for cardiovascular disease risk from social determinants of health, and ultimately guide clinician treatment options. The researchers also suggest that large databases about social determinants of health variables, especially environmental ones, are not as readily available as they should be, and are an important area of opportunity for future data collection efforts.

A similar application of this technology was used in a study published by Cancer Immunology Research to predict lung cancer immunotherapy success. In the study, researchers used an AI algorithm to identify changes in patterns from CT scans that were previously not detected by clinicians, which ultimately predicted how well a patient would respond to immunotherapy. This suggests that predictive analytics can help improve the accuracy of diagnoses and treatment.

Of course, the applications for predictive analytics extend beyond chronic disease prevention and treatment. In the past year, researchers have also used predictive analytics to forecast outcomes for patients positive for COVID-19. In The American Journal of Emergency Department Medicine, a published study validated a tool that helps physicians predict adverse events among patients presenting with suspected COVID-19. The study suggests that the algorithm and scores can help physicians decide when to hospitalize or discharge patients during the pandemic. Therefore, predictive analytics appear to also provide insights that enhance treatment.

Many additional articles (such as one article from Health IT Analytics) and published studies recommend predictive analytics for its potential benefits. As with any technology, however, it is not without its risks. The use of AI brings about concerns for privacy, especially since hospitals must properly steward patient data and comply with HIPAA regulations. But there are several other considerations identified in a recent Deloitte analysis (see Figure 1), not the least of which is ensuring the algorithm doesn’t introduce bias that disproportionately harms minorities and communities of color. Predictive analytics may also present evaluation challenges. Once algorithms are validated, their widespread use in clinical settings should be confirmed for their efficacy, which requires measuring the absence of disease.

Figure 1.

The potential benefits of predictive analytics are variable and significant; however, as healthcare learns to integrate AI technologies, it will be important to keep its risks in mind and address them accordingly. The MVC Coordinating Center endeavors to assist its members through their data analytics journey by providing insights into specific data sets. When pursuing additional technologies or analytic tools, the Coordinating Center encourages members to volunteer as a sounding board and resource for other members. If your hospital or physician organization is currently utilizing AI or considering it with your patient data, we encourage you to reach out so MVC can share your experience with others. You can reach the MVC Coordinating Center at michiganvaluecollaborative@gmail.com.

12 Aug 2021
0

Growth On the Horizon for Robotic Process Automation in Healthcare

August 12, 2021 | 0 Comments | ,

Hospital quality improvement teams have an increasingly difficult task ahead. Their efforts to improve quality of care across a wide range of medical services must be balanced with the need to expand their facility’s capacity, ensure proper handling of sensitive data, adhere to strict procedures, cut costs, and adapt to the limitations of a pandemic. This work poses challenges both organizational and operational. Even though patient care is the primary focus for hospital staff, they must maintain a constant stream of paperwork and other administrative tasks such as data entry, scheduling appointments, billing, and managing claims paperwork. Robotic process automation (RPA) presents an opportunity to decrease these administrative costs and streamline some operations.

RPA is defined as software that can automate repeatable, rule-based processes. RPA interacts with the assigned applications in the same way that a human does, logging into a given system and following a defined set of keystrokes and rules. It is not the same as artificial intelligence (AI)—there is no decision-making capacity. RPA can only offload manual, high-volume computer processes. The primary benefit of RPA, therefore, is its ability to free up time for humans to complete more complex tasks, such as interfacing with patients or interpreting data.

RPA is a burgeoning field recommended by consulting groups such as Deloitte, McKinsey, and Bain & Company. Although RPA hasn’t had sufficient time to make its way into academic literature, it is spreading quickly in all types of industries. For example, according to Deloitte’s Global RPA Survey, more than half of their 400 respondents from multiple industries were already pursuing automation with as many as 72% looking to add RPA in the next two years.

“RPA exceeds adopters’ expectations not only when it comes to the rapid rate of ROI increase, but also when it comes to facilitating compliance (92%), improved quality and accuracy (90%), or improved productivity (86%),” the report read. The report also suggests that the benefits of utilizing RPA may include cost reductions, boosts to productivity, more stable workflows, and fewer human errors, among others (see Figure 1).

Figure 1.

Source: Summit Healthcare

Healthcare as an industry has the potential to significantly benefit from offloading administrative tasks to bots. According to McKinsey, the healthcare sector has the potential to automate around 36% of tasks. They suggest that the greatest potential for healthcare payers is in areas such as claims processing, customer service, and billing activities (see Figure 2).

Figure 2.

Hospitals and health systems have pursued RPA in these areas as well and found success. One example written about in Forbes recently described the efforts of Baylor Scott & White Health (BSWHealth), an academic medical system with 52 different hospitals and the largest not-for-profit provider in Texas. BSWHealth uses RPA to automate “claim statusing” in its insurance collections department. The bot helps to check the status of outstanding insurance claims that, previously, a human employee would have to do by logging into multiple payer websites or placing phone calls. The RPA bot uses screen-scraping technology that mimics keystrokes the employee would enter to obtain claim statuses from payers. As a result, an abundance of claims—those that are accepted and scheduled to be paid—never clutter the employee’s desk. Instead, the employee only sees those that are denied and require human attention, resulting in outstanding claims being addressed faster. BSWHealth is pursuing a variety of RPA projects like this one across all of its revenue cycle departments. They reduced their total FTEs by over 20% while simultaneously reducing payer denials by 20%.

Success stories like this one are particularly exciting for hospitals struggling to manage their case load amidst the pandemic. Daily operations and procedures have been severely impacted financially and operationally by coronavirus. A recent survey conducted by the World Health Organization identified that almost half of the countries surveyed (49%) reported strains on their ability to treat diabetes, with 42% reporting the same for cancer and 31% struggling to properly manage cardiovascular emergencies. As a result, companies are pursuing automation opportunities more than ever before (see Figure 3), with Bain & Company reporting as many as 81% of hospitals pursuing RPA initiatives.

Figure 3.

Still, according to a 2019 white paper by The Economist, “extensive” use of automation is only used by half of healthcare organizations, and healthcare in general is among the most resistant to adopting it. Some healthcare organizations remain cautious for a variety of reasons, including concerns about initial investments, maintenance costs, and the possibility of failure. The same white paper also proposes that data privacy and security concerns might be a significant hinderance to RPA efforts, as well as a deficit in the skill sets needed to develop the bots.

Plus, any discussion of RPA sometimes begets fears about job replacement. In some scenarios, health systems have seen an overall decrease in FTEs after putting RPA initiatives in place. However, the overall goal is usually to reallocate effort toward more high-level, cognitive projects in a way that increases productivity without replacing people. If an administrative task requires no higher-level thinking, then giving it to an RPA bot will free up time for clinical staff to attend to patient care rather than paperwork. In fact, according to Harvard Business Review, most new adopters of RPA have promised their employees that it won’t result in layoffs.

Despite hesitations, health systems are likely to test out RPA projects in the coming years in response to the current state of affairs. Hospitals have been forced over the past year to find efficiencies where they can. RPA bots appear to have the potential for a variety of benefits, not the least of which is flexibility to redeploy personnel to areas in need of increased staffing. As RPA begins to make its way into the literature, it will be important to consider research findings about best practices going forward.

It will also be helpful going forward to share lessons learned with peer institutions. One of the goals of the MVC Coordinating Center is to support collaboration and idea sharing across its membership. If any member is implementing RPA projects and would be interested in sharing their experience with others, please contact the MVC Coordinating Center team at michiganvaluecollaborative@gmail.com.

05 Aug 2021
0

MVC Registry Expands with Addition of Medicaid Episodes

August 5, 2021 | 0 Comments |

The Michigan Value Collaborative (MVC) Coordinating Center recently added Medicaid data to its registry. This update reflects the culmination of many months of work to acquire, process, clean, and add the data, which became available on July 16 to MVC registry users. The current data set is from 1/1/15 through 9/30/19, which amounts to claims from 1/1/15 through 12/31/19. MVC data sources now comprise over 80% of Michigan’s insured population. This represents an additional 1.8 million covered lives (see Figure 1). MVC’s data sources now include Medicare FFS, Commercial Blue Cross Blue Shield of Michigan (BCBSM) PPO, Medicare Advantage BCBSM PPO, Commercial Blue Care Network (BCN) HMO, Medicare Advantage BCN, and Michigan Medicaid.

Figure 1.

The addition of Medicaid data will impact, among other things, the distribution of MVC episodes across its portfolio of payers. Medicare is still the dominant payer within MVC data with more than 641,747 episodes. However, the new distribution of MVC episodes by payer (Figure 2) showcases that Medicaid is now the third-largest payer in MVC data, accounting for 18% of total episodes.

Figure 2.

MVC currently serves 97 participating hospitals, including critical access members, and 40 physician organizations in Michigan. The proportion of Medicaid episodes in MVC data by facility (Figure 3) varies significantly across MVC’s membership, with some members attributing less than 5% of their episodes to Medicaid and some near 60%. For the bulk of MVC’s membership, between 10% and 30% of their episodes are in Medicaid, which represents a significant increase in the total episodes they can now utilize. For some MVC hospitals, the number of episodes they have in MVC data may double if they have a large share of Medicaid patients.

Figure 3.

MVC currently provides data on 40 defined conditions. The addition of Medicaid data is likely to impact certain conditions more than others in keeping with the types of procedures and conditions most prevalent with Medicaid-eligible populations. The top five Medicaid conditions include sepsis, C-section, vaginal delivery, cholecystectomy, and chronic obstructive pulmonary disease (COPD), so members are more likely to see changes to their utilization data for those conditions. The number of episodes being added for each condition is outlined in Figure 4.

Figure 4.

The Medicaid data will also allow for the creation of new data visualizations and reports that capture information not previously available. For example, MVC analysts recently generated two new Medicaid-based maps (Figures 5 and 6) that help visualize utilization and location information for the Medicaid population. Figure 5 represents the patient Zip codes that can be attributed to Medicaid episodes in MVC data, with Zip codes appearing darker if a larger percentage of Medicaid patients reside there. This allows members to see those communities near their own facilities that are likely home to the Medicaid patients they serve.

Figure 6 also represents the percentage of episodes attributed to Medicaid patients, with darker colors representative of higher percentages; however, Figure 6 connects these Medicaid episodes to MVC member facilities rather than Zip codes and visualizes the total number of episodes in addition to the percentage. Together, these two figures provide MVC members with more information about their Medicaid populations as well as the extent to which utilization varies between peer facilities in the same region.

Figure 5.

Figure 6.

These maps are the first example of new outputs that are possible with the addition of Medicaid data. The MVC Coordinating Center plans to produce additional reports for members that leverage the new data set. One area of interest is the social determinants of health. Since Medicaid provides medical assistance to disabled and low-income individuals, statistical analysis using this data often reflects trends tied to low socioeconomic status populations. Ideally, this data set will allow MVC and its members to invest more attention and resources into equity-based quality improvement projects.

The MVC Coordinating Center is eager to learn which topics are of greatest interest to members that integrate Medicaid claims. If your team has specific ideas that could help guide this work, please contact MVC at michiganvaluecollaborative@gmail.com.

22 Jul 2021
0

Condition Selection Process Announced for MVC Component of BCBSM P4P Program

July 22, 2021 | 0 Comments

This week the Michigan Value Collaborative (MVC) Coordinating Center announced the condition selection process for program year (PY) 2022 and PY 2023 of the MVC Component of the Blue Cross Blue Shield of Michigan (BCBSM) Pay-for-Performance (P4P) program. The timeline for each program year’s stages are detailed in Figure 1.

Figure 1.

In the announcement, hospitals were tasked with selecting two conditions for which they will be evaluated and returning their condition selection form to the Coordinating Center by Friday, August 13, 2021. The announcement also outlined changes to the scoring methodology, cohort assignments, and bonus points available.

The Coordinating Center’s recent announcement included condition selection reports with targets for each condition option that may help inform hospitals’ selection decisions. Each participating hospital will choose two of the seven available conditions for PY22 and PY23: spine surgery, joint replacement, chronic obstructive pulmonary disease (COPD), coronary artery bypass grafting (CABG), congestive heart failure (CHF), colectomy (non-cancer), and pneumonia. When selecting conditions, the Coordinating Center recommends reviewing your data in the registry and considering several factors for each condition, including case counts and identifiable areas with the greatest cost opportunities. The Coordinating Center also recommends considering where resources are currently being directed in your facility and potentially aligning with those efforts.

One notable change from prior program years is the methodology by which hospitals earn achievement and improvement points. Hospital scores will continue to be based on a hospital’s risk-adjusted, price-standardized total episode payments for two selected conditions, and they can still earn a maximum score of 10 points. However, the improvement and achievement scores will become more similar in order to be placed on the same scale. As such, the achievement equation will change from being based on rank within MVC cohort at performance year to being based on distance from MVC cohort mean at baseline year. Similarly, the improvement equation will utilize the distance from the hospital’s mean at baseline. These new equations (see Figure 2) as well as complete descriptions of the updated methodologies are reviewed at length with examples in the technical document.

Figure 2.

P4P cohorts have also been reassigned for PY22 and PY23. These changes are also detailed in the technical document, and the new cohort assignments can be found on the MVC website. The cohorts are not intended to group hospitals that are exactly alike; rather, they create a reasonably-comparable grouping from which MVC can complete statistical analysis.

The final change is to the awarding of bonus points. In place of the previous 5% cohort reduction bonus, participants can instead earn bonus points by completing two questionnaires (one per selected condition) and submitting these to the Coordinating Center by November 1st of each program year. The purpose of this is to gather examples of quality improvement initiatives in operation at MVC member hospitals to share with the Collaborative. Moving forward, this will help support members in reducing costs through collaboration.

Each of the changes mentioned above are designed to deliver a more transparent, intuitive, flexible, and fairer P4P program. The Coordinating Center will offer an explainer webinar to answer questions and walk through the details of these changes in more detail. The webinar will be offered on two dates: the first is scheduled for Thursday, July 29 from 11:00-12:00 pm, and the second is on Tuesday, August 3 from 1:00-2:00 pm. Both webinars can be accessed using the following Zoom link: https://umich.zoom.us/j/95502303999. Participants can also call +1 301 715 8592 (meeting ID #955 0230 3999). For those interested in the explainer webinar who are unavailable on both dates, a recording of the first webinar will be available. If you are interested in receiving a link to this recording, please email the MVC team at michiganvaluecollaborative@gmail.com.

17 Jun 2021
0

Custom Hospital Analytics Result in Case Study for Collaborative

June 17, 2021 | 0 Comments | , ,

The Michigan Value Collaborative (MVC) Coordinating Center encourages its members to seek out custom analytics to inform and support ongoing quality improvement activities. These requests can help hospitals and physician organizations dig deeper into specific aspects of their administrative claims data and, as a result, better understand areas for improvement.

As custom analytics have been prepared and shared with respective members, the Coordinating Center has endeavored to learn the extent to which these analytics have been utilized. The resulting feedback has enriched MVC’s understanding of its members’ quality initiatives, and presents a great opportunity for MVC to educate its members about the successes and lessons learned of their peers.

In that spirit, the Coordinating Center has sought the permission of various hospitals to generate case studies based on this collaborative work. One such case study featuring McLaren Port Huron Hospital was created this past year and shared with the entire Collaborative via the MVC Newsletter (Figure 1). It features a custom analytics request about the rates and adherence of follow-up visits in their congestive heart failure (CHF) population as well as readmission rates for chronic obstructive pulmonary disease (COPD). The resulting custom analytics reports prepared by the Coordinating Center were also accompanied by best practice sharing sourced from other Collaborative members.

Figure 1.

The Coordinating Center plans to continue to generate shareable case studies about similar requests if those facilities have provided their permission. Similarly, MVC will continue to identify such opportunities for information sharing and networking across facilities in order to support its members.

If any members of the Collaborative are interested in pursuing custom analytics in the future or have ideas to share across hospitals, please contact the Coordinating Center at michiganvaluecollaborative@gmail.com.

27 May 2021
0
View Post
Introducing Jana Stewart, MPH, MS Communication Specialist for MVC

Introducing Jana Stewart, MPH, MS Communication Specialist for MVC

May 27, 2021 | 0 Comments

I am excited to join the Michigan Value Collaborative (MVC) as Communications Specialist. This new position will help highlight MVC’s success stories, support MVC member events and engagement activities, and promote MVC services throughout the state of Michigan. I look forward to getting to know MVC members and hearing their feedback in the coming months.

I have worked in communications in various capacities over the past 10 years, and as a result my writing has been published in academic journals, newspapers, magazines, and K-12 curricula. I started out in journalism as a sports writer, copy editor, and then managing editor of a regional newspaper, during which I earned writing awards from the Michigan Press Association and the Society of Professional Journalists. My time as a journalist allowed me to write about a wide range of topics, from city government and local business to high school sports and crime.

When I left journalism for a position at the University of Michigan, I continued to provide broad communications support to administrative offices and research labs. Through this work I developed a strong interest in programs that sought to improve the long-term health of people and places. As a result, I also enrolled as a dual-degree master’s student in public health and environmental science. I have three degrees from the University of Michigan in total, including a Bachelor of Arts in Kinesiology, a Master of Science in Environmental Psychology, and a Master of Public Health (MPH).

Following the completion of my MPH program, I spent several years in the field working for primary prevention programs. I worked for a hospital-based farm in the St. Joseph Mercy/Trinity Health network, and then with Michigan Medicine’s Project Healthy Schools program. Implementing interventions in the field helped me to see first-hand the impact that such programs can have on a population or institution. Now I am truly excited to support the mission and vision of the MVC as I return to communications full-time. I know there are tremendous untold stories about the impact of MVC’s efforts on Michigan hospitals and patients. If you have a story to tell or a question to ask, please reach out to me at janaemil@med.umich.edu. I would love to hear from you!