Machine learning is a big idea, but hospitals need business plans first

http://www.healthcareitnews.com/news/machine-learning-big-idea-hospitals-need-business-plans-first

machine learning and AI in healthcare

Elizabeth Clements, business architect at Geisinger Health, will be hosting a session at HIMSS18 on March 7.

Don’t get lost in the complexity of large-scale use cases.

Machine learning has the potential to transform healthcare through new knowledge discovery and improved productivity, but many health systems do not have a business plan in place to support advanced analytics beyond research and development.

As health systems consider how best to leverage machine learning and artificial intelligence, it will require a shift in IT strategy to focus on not just data, but managing the model itself. This means, among other things, defining the value of machine learning and providing a framework for evaluation and application.

Health systems need to keep things simple when moving into machine learning, said Elizabeth Clements, business architect at Geisinger Health.

“When working with new technology and developing a service from scratch, it can be easy to get lost and slow progress down with the complexity of a large use-case,” Clements said. “If you keep your scope narrow and define near-term goals, you will find you are able to make more meaningful progress in a short amount of time.”

And healthcare professionals dealing with machine learning must themselves learn how to partner with the business.

“Understanding the current and future state use of the machine learning solution is critical,” Clements said. “If you don’t consciously determine how much you want or need human intervention with the model, it will make your solution much more difficult to implement and gain buy-in.”

Clements said a simple framework for thinking about machine learning in the context of the business is needed. That includes understanding its value and use-cases before embarking on this type of analytics advancement, as well as knowing the basic challenges and how to design a program that takes those into account.

“Machine learning is the next wave of advanced processing technology offering us new avenues for information discovery and productivity enhancement,” she said. “It has the potential to transform how we conduct business; however, it will require a shift in our IT strategy.”

It is not just about the data or the application, it is also about the model itself. IT leaders should consider how to complement their existing IT and data scientist teams with new skill sets and consider how machine learning can advance existing task execution, she added.

Clements will be speaking in the HIMSS18 session, “Managing Machine Learning: Insights and Strategy,” at 11:30 a.m. March 7 in the Venetian, Palazzo D.

Trump budget seeks savings through ObamaCare repeal

Trump budget seeks savings through ObamaCare repeal

Trump budget seeks savings through ObamaCare repeal

The White House budget for fiscal 2019 seeks major savings by repealing ObamaCare and endorsed a Senate GOP bill as the best way to do so.

“The Budget supports a two-part approach to repealing and replacing Obamacare, starting with enactment of legislation modeled closely after the Graham-Cassidy-Heller-Johnson (GCHJ) bill as soon as possible,” the White House said in its budget request.

The legislation from Sens. Lindsey Graham (R-S.C.), Bill Cassidy (R-La.), Ron Johnson (R-Wis.) and Dean Heller (R-Nev.) would replace ObamaCare with a series of block grants to states.

The budget proposes over $90 billion in savings over 10 years if the policies in the Graham-Cassidy bill were enacted. Combined with other provisions like Medicaid changes, the White House projects there would be nearly $675 billion in savings over a decade tied to repealing ObamaCare.

Advocacy groups were quick to denounce the proposal, which is unlikely to gain traction in Congress.

“By asking Congress to revive the deeply unpopular Graham-Cassidy repeal bill that ended protections for Americans with pre-existing conditions, gutted Medicaid, ripped away coverage from millions, and raised costs for millions more, while also proposing drastic cuts to Medicare, Trump has chosen to ignore the American public’s overwhelming preference for a bipartisan path forward on health care,” said Protect Our Care campaign director Brad Woodhouse.

Republican leaders have signaled that they are not interested in diving back into the contentious ObamaCare repeal fight this year. The Senate last year failed to pass a repeal bill, and there is no indication that the votes have shifted since then.

A number of Republicans have even discussed taking bipartisan actions to stabilize ObamaCare markets and try to bring down premiums through actions such as funding known as reinsurance.

Graham has said he will continue fighting for his bill and is not completely alone. Sen. Ted Cruz (R-Texas) is also calling for Congress to not give up on repeal this year.

 

Learn and apply these leadership traits in an afternoon

Learn and apply these leadership traits in an afternoon

7 Traits

 

 

The Medicaid Agency of the Future: What capabilities and leadership will it need?

https://www.mckinsey.com/industries/healthcare-systems-and-services/our-insights/the-medicaid-agency-of-the-future–what-capabilities-and-leadership-will-it-need?cid=other-eml-alt-mip-mck-oth-1802&hlkid=f3e96b6a484442438bbdc44b7744e1e9&hctky=9502524&hdpid=c2f3cd34-a3e9-46d4-bc69-f4a9cc3b0ac9

Image result for The Medicaid Agency of the Future: What capabilities and leadership will it need?

Medicaid’s scale and complexity are unprecedented. State Medicaid leaders will need to innovate if they are to develop the capabilities that will enable them to steer their agencies into the future.

Since its inception in 1965, Medicaid has grown to have an expanded role in state governance—it is usually the first- or second-largest state program. Nationwide, Medicaid agencies manage about $574 billion in annual spending.1In an average-sized state, Medicaid directors are the single largest purchaser in the health sector, overseeing about $10 billion each year in payments to providers (roughly 17% of the state’s economy).2The agencies typically serve nearly one-quarter of their state’s population3and, in our experience, procure the largest IT infrastructure projects in state government.

Recently, demands on state Medicaid agencies have grown. Increasingly, Medicaid agencies are playing a multi-part role, fulfilling their traditional responsibilities (i.e., payer-purchaser, operator) and taking on new ones (i.e., market shaper and innovator). And while they are doing this, the agencies must address an array of competing priorities, as well as uncertainty about future funding levels. To perform well in these new roles, the agencies must develop new, next-generation capabilities so they can address the needs of their state’s citizens—even those not enrolled in Medicaid.

We believe the roles played by state Medicaid agencies will continue to evolve, but the agencies—in conjunction with the state government leaders they work with—can choose their strategic path forward (primarily, the extent to which they want to be market shapers). Those that pursue this path aggressively will be the first of their peers to evolve into a Medicaid Agency of the Future.

As we discuss below, all state Medicaid agencies will need to strengthen their capabilities if they are to meet the demands of the future—an issue that has important implications not only for the agencies themselves but also for state government leaders, managed care organizations (MCOs), local providers, and others. However, agencies focused on becoming market shapers will need to double down on capability building if they want to succeed on their chosen path.4

Trends shaping Medicaid

A number of trends are shaping what Medicaid agencies need to do to prioritize the direction of their efforts.

Spending growth is putting pressure on state budgets. Medicaid is putting continued cost pressure on state budgets. Program spending (including federal and state funding) increased from 20.5% of state budgets in 2008 to 29.0% in 2016.5The continued increase in Medicaid spending could have funding implications for other state programs, such as elementary and higher education, public assistance, and transportation. The cost pressures resulting from Medicaid spending are expected to continue regardless of what, if any, changes are made at the federal level.

Medicaid programs can lead to payment inno­vation. As program costs have risen, Medicaid directors have increasingly tried to slow the medical cost trend. One lever available to them is transitioning from fee-for-service reimbursement to payment innovations that include meaningful levels of provider risk-sharing. New payment models that reward providers for delivering high-quality care at lower cost have been shown to improve care quality and reduce costs by 5% to 10% when rolled out across the full spending base.6In several states, Medicaid programs have led multi-payer efforts to achieve payment inno­vation across the state.7

Medicaid plays an important role as both a payer and a convener. In the aggregate, Medicaid is the country’s largest payer in terms of covered lives and, in many states, is the largest purchaser of healthcare ser­vices.8Thus, Medicaid agencies are uniquely positioned to facilitate change. In addition, the agencies can often bring together multiple stakeholders to help align on improvements that would affect not only Medicaid but also the entire healthcare market.

Recognition of Medicaid members as consumers is increasing. Some Medicaid agencies are beginning to approach Medicaid members as consumers. For example, they are offering members technological tools, such as apps and patient portals, that empower greater decision making (e.g., about choice of provider, care setting, or treatment). If well utilized, these tools can improve member experience and encourage higher-value care.

Awareness of social determinants is rising. Increasingly, states are turning their attention to non-health factors, such as housing, education, and transportation, that influence Medicaid members’ ability to maintain their health and adhere to treatment. Some programs are beginning to address these determinants head on (e.g., by providing housing or transportation vouchers). Experimentation in special needs care is underway. Integrated models typically deliver better quality and cost outcomes. For example, integrated behavioral and physical healthcare approaches for high-needs patients have been shown to reduce emergency department and inpatient visit spending by 10% to 25%.9The successes to date are paving the way for further innovation in other special needs areas.

Analytics is playing an increasing role. Advanced analytics and big data can help Medicaid and other public health officials better understand state needs, design programs, and target interventions to maximize the impact of limited funds. The emergence of new national data-sets, such as the Transformed Medicaid Statistical Information System (T-MSIS, which includes states’ comprehensive claims and enrollment data) and CMS’ online database of state waivers and state plan amendments, may enable states to draw on experiences elsewhere when designing new programs.10

In short, state Medicaid agencies are facing an increasingly complex and difficult set of challenges at a time when the expectations of multiple stakeholders—members, families, and advocates; providers and MCOs; the federal government, state leaders, and other state agencies—are rising. If Medicaid agencies are to address these challenges successfully, the role they play must evolve.

Introducing the Medicaid Agency of the Future

In the future, some state Medicaid agencies may opt to follow the path set by previous state leaders. In other cases, they may want to respond to the trends just discussed by taking the lead in transforming healthcare delivery in their programs and their states. These Agencies of the Future will have to be able to chart a strong strategic direction and execute the activities that follow both efficiently and effectively. To accomplish those goals, they will need to use a data-driven approach to program management, build new capabilities, and improve their organizational health (Exhibit 1).

Agencies that opt to follow a more traditional path would also benefit from strengthening their operational performance and organi­zational health, but the level of improvement needed is lower for them than for the Agencies of the Future. Both sets of agencies, however, will want to prioritize their strategic investments once they have chosen their path forward.