Capacity planning efecctive forecasting

Most hospitals in this country have already attempted – on numerous occasions – to improve operating margins by reducing labor costs and average length of stay. Armed with annual budget ratios and months of retrospective data, they've tried to identify opportunities to tighten staffing by gazing backwards at the various causalities that resulted in increased overtime and poor patient flow. Now that predictive analytics has arrived, hospitals are eager to implement smart tools that will help them proactively set capacity, recruit staff and deploy resources without unnecessary expense.

While it's certainly true that a proactive approach to capacity management can save your hospital millions in labor costs, few facilities have any prior experience using predictive analytics for this purpose. Vendors are taking advantage of this lack of knowledge to misrepresent the accuracy and savings potential of their predictive tools. While a static, 30-day forecast based on census only can represent a step forward, hospitals need the full spectrum of forecasting power in order to sustainably reduce operating expenses.

To learn more download the white paper,
Predictive Analytics for Capacity Planning: All Forecasts Are Not Created Equal

A full-spectrum predictive analytics tool for capacity management should allow your facility to forecast patient demand far enough in advance that you can align resources without incurring agency and overtime costs. The tool should also be accurate enough to help drive your strategic, budgetary, scheduling, patient flow and staffing decisions, and flexible enough to provide for on-the-fly changes.

Many vendors only offer short-term demand forecasting. To truly impact operating costs, the tool should provide forecasts across multiple planning horizons:

  • Strategic (3–5 years): Strategic business plans include guidance around labor costs and capacity needs, including construction.
  • Budgetary (1–2 years): Establishing a realistic annual budget depends upon reliable capacity planning and accurate labor forecasts for the upcoming year.
  • Scheduling (1–4 months): The scheduling planning process should allow frontline leaders to address variation in planned vs. actual activity by adjusting staffing needs before schedules are created and posted.
  • Operational (1-2 weeks): Continuous planning systems that accept real-time patient flow data allow leaders to proactively adjust capacity and planning over the near term, rather than at the beginning of each shift.
  • Continuous planning: Incorporating unexpected, day-of adjustments into the capacity planning system refines the forecast for future planning periods.

Before your organization invests in a predictive analytics tool, make sure to evaluate the solution's ability to fully impact outcomes. In your OR, does the tool predict scheduled and emergent demand by procedure, allowing you to optimize your schedule and align resources downstream? Does it provide your nurse managers with real-time patient arrivals by the hour, with adjustments to the staffing plan already incorporated?

In addition to historical census data, does the tool also incorporate predictive and algorithmic modeling? What about pattern identification and scenario modeling? A dynamic capacity planning tool should incorporate multiple forecasting methods to establish a continuous forecast with a high degree of accuracy. And in order to truly understand how effective a particular forecasting tool will be in practice, you should ensure that estimations of its accuracy are based on an absolute daily comparison of the forecast to what actually happened. Some vendors rely on a comparison of the model's average monthly census to your average actual census, which is ultimately quite misleading.

With a full-spectrum predictive analytics tool, your hospital can engage in a cycle of real-time, continuous capacity planning that transforms the way you allocate resources. Predictive analytics can be a revolutionary force for your workflow and your bottom line – but first you must become a smart buyer who understands the terminology and knows what it really takes to achieve sustainable results. Our white paper, Predictive Analytics for Capacity Planning: All Forecasts Are Not Created Equal, will help you become better informed.

Download White Paper Now!

Billie Whitehurst

About the author

Billie Whitehurst, M.S., R.N., is vice president of Capacity Management for McKesson Connected Care and Analytics, where she leads the company's efforts to offer solutions that help health organizations achieve sustainable operating improvements by enabling a proactive approach to workforce optimization, capacity and throughput. Whitehurst has more than 15 years' experience leading clinical IT teams serving acute and homecare markets. Whitehurst has a master's degree in health care administration and a bachelor’s degree in nursing from the University of Colorado. She is board certified in nursing informatics from the American Nurses Association.