At a time of unprecedented change across the industry, the need for healthcare organizations to look and plan ahead has never been greater. Senior leadership faces increasingly complex decisions laden with risk and uncertainty.
Yet most healthcare forecasting still relies on single-point methods that don’t hold up to mathematical scrutiny. This traditional approach to forecasting is based on static cases for “best, worst, and most likely” scenarios, requiring executives to make complex decisions based on only a handful of observations. The flaw is in the false premise that the underlying assumptions can be known with such precision. The real world doesn’t work that way.
The science of statistics tells us that much larger sample sizes are essential to analyzing variability, uncertainty and risk. With the exponential growth in data and the complexity of healthcare, using single-point forecasting methods to answer strategic business questions can actually add more risk by giving false confidence.
Making informed strategic decisions in a fast-changing environment requires an analytical approach that considers all the various combinations of a range of performance drivers, or assumptions (e.g., growth, utilization, payer mix, reimbursement, workforce planning, etc.). In our work with leading healthcare organizations around the country, we often recommend simulation forecasting as the best way to look ahead.
Simulation forecasting overcomes the dangers of flawed single-point forecasts, providing the most accurate representation of the likelihood of an outcome and improving the overall reliability of projections.
A couple of false assumptions have slowed the industry’s adoption of simulation forecasting. Some organizations assume the process will be overly complex. But today’s software makes simulations intuitive and more user friendly. Second, finance and planning professionals often think simulation depends on being certain about your range of values for assumptions. In fact, the opposite is true: The less certain you are about your assumptions, the more applicable—and valuable—simulation is as a forecasting tool.
As you consider the value of simulation forecasting for your organization, here are a few lessons learned that will help ensure success.
Think in ranges. Rather than defining assumptions as single-point estimates, it’s critical to delineate them as distributions, or ranges, of possible values. These ranges are included as inputs into the forecast model. When the simulation runs, the ranges are sampled thousands of times—each time producing a single output. The output of thousands of observations is then summarized in an overall probability distribution, which provides a more accurate representation of the likelihood of outcomes. You gain the information to better assess risk, explore options, and make decisions that put the odds in your favor.
Pay attention to probabilities. The probability of an outcome offers more insight than one outcome alone. But it takes a different mindset to dwell in the realm of probabilities. Since simulation results don’t yield one answer, they must be interpreted differently than single-point forecasting. For example, without simulation, we might arrive at a conclusion such as “an investment is forecasted to lose $1.3 million over a three-year period.” Simulations allow us to add the dimension of likelihood: “There is a 20 percent chance of breaking even with the investment,” or “There is a 30 percent chance of losing more than $3 million.” Weighing probability is critical to making informed decisions.
Based on such information from a simulation model, one of our clients was able to see that a proposed ACO agreement had a much larger potential downside than was originally forecast by more traditional, static analyses. This equipped our client to negotiate with new insight.
Treat forecasting as a team sport. Traditional forecasting is often delivered to senior leadership as a product, without their involvement in developing process or methods. In order for simulation forecasting to work effectively, senior leadership needs to be involved in defining and testing the ranges for the most sensitive performance drivers (e.g., assumptions). Allowing time for iterative review and discussion will help ensure a higher degree of buy-in with the predictive results.
Simulation is the best method to incorporate uncertainty and improve the accuracy of strategic forecasting. By avoiding single-point forecasting and, instead, focusing on the likelihood of a range of potential outcomes, you’ll increase the overall quality of your predictions. When the future is clearer, you can move forward more quickly—with greater confidence.
Register or login for access to this item and much more
All Health Data Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access