Agile Planning for Healthcare Providers

Forces old and new are buffeting healthcare providers. Medicaid,

accounting for 17% of state budgets, is being cut. America is get-

ting grayer, and living longer. The reality for healthcare provid-

ers is that the average Joe and Jane will need more services than

before and will use those services for longer than any prior gen-

eration. The Affordable Care Act will bring some 32 million peo-

ple into ranks of the insured with unclear impacts on provider

capacities, payer mix, hospital-physician alignment and many

other variables.

These and other socio-economic trends are pressuring healthcare

provider finance teams to make planning systems more immediate

and responsive. The old way of thinking — you run the facility

through the once a year budget process — is on the way out in

favor of new concepts and system approaches characterized by

more forecasting and incorporation of “agile planning” tech-

niques into the forecasting and planning processes.

This white paper explores the concepts and practices of agile

planning for healthcare providers with a focus on defining soft-

ware application requirements for scenario analysis, driver-

based modeling and integrating actuals — the three foundations

of an agile planning system where the goals are insight, options

for action, and decisions, rather than budgetary control.

White paper

Agile Planning™ for

Healthcare Providers

by Rand Heer and Tom Hartman

“The old way of thinking

— you run the facility

through the once a year

budget process — is on

the way out in favor of

new concepts and system

approaches characterized

by more forecasting and

incorporation of “agile

planning” techniques

into the forecasting and

planning processes.”

Agile Planning for Healthcare Providers

2

Ray Wolfe, CFO of Pitts-burgh Mercy, strode into the conference room. Everyone was seated and ready to roll for the fore-cast review—Shelley and Dave from finance, five heads of community treatment centers, the service coordinator, and heads of acute case man-agement and emergency care—including Ray, eleven persons total.

Who attended this particular session of the review was carefully coordinated, though Ray often shuf-fled forecasting group membership based on com-parability of underlying data and key measures, and the corporate reporting hierarchy. From the finance side, Dave was the planning administrator who ran the planning application in real time during the forecast review session. The seven line managers were a subset of 100+ managers in the organization with revenue and cost management responsibility across 25 treatment programs.

Each quarter Ray conducts fifteen such planning sessions. Each session includes the same finance team and combinations of line managers whose responsibilities are functionally similar or overlap in service requirements. The agenda kicks off with Ray updating the room on global trends, overrid-ing budget issues and community dynamics. Then the fun begins: a review of prior meeting commit-ments including a focus on what worked and what didn’t; an open exchange of information with a focus on best and worst practices; and a real-time update of commitments and financial plans.

For the most part, the process works. In just a few hours each quarter, mental health practitioners without business or finance training have the op-portunity to participate in a planning and perform-

ance review process with a real-time understand-ing of how their own groups’ results impact overall results and a clarification of short term priorities. Then, back to the basics of running the clinics.

It wasn’t always so smooth. Just two years earlier when Ray was appointed CFO, the organization was in trouble. After a painful disposition of an under-performing inpatient facility, the remaining organi-zation—60 outpatient facilities providing mental health, retardation, drug/alcohol rehab and home-less services—was “operationally challenged”: ser-vice demands were increasing at a 20% annual rate; budgeting was spreadsheet-based with all the at-tendant problems—linking and formula errors, too much wasted time consolidating the pieces; and no efficient way to import actuals and compare re-sults. In addition, budgeting was a ritual without meaning—full year totals with no seasonality, no operational integration, no P&L visibility for the hundreds of revenue/cost centers, and worst of all, no real manager buy-in.

Now, a year and a half later, the forecast review meetings are working with remarkable results.

Ray choreographs the agenda and participa-tion which is based on relevancy and congru-ent interests. Groups and participation dynam-ics are reshuffled regularly to optimize sharing of common experiences.

Data is presented with an overhead projector in real time. The planning tool is also the pres-entation tool. Forecast assumption changes, such as forecasting collection rates based on latest averages, immediately ripple through the financial statements and cash flow.

The presented data, both historical actuals and forecast, incorporate operational activity driv-ers such as number of outpatient transactions, payer types, and collection rates. The availabil-ity of operational information integrated with financial results forms the basis of the line manager dialog for understanding and sharing of best practices.

Fundamental to the process is multiple scenar-ios. The team learning experience and forecast commitment process is based on creating and comparing working scenarios in real time.

Ray choreographs the agenda and participation in forecast sessions based on relevancy and shared inter-ests... The planning tool is also the presentation tool.


3

Since abandoning spreadsheets for a database plan-ning application, Ray and Pittsburgh Mercy have successfully renovated its planning processes and culture by implementing Agile PlanningTM: line managers have a renewed sense of participation and energy; $600,000 in revenue enhancements and cost efficiency improvements have been redirected to improve services; and forecast accuracy at the bottom line is now 2% quarter to quarter.

Pittsburgh Mercy no longer budgets. All planning and decision making derive from the two year roll-

ing forecast process described here. When budget presentations are made to the Board, Ray reports on the latest and most credible forecast. With a forecast accuracy of 2%, nobody cares or questions that the process is not more structured and does not go to the lowest level of detail in the chart of accounts.

As Ray says, “We have a whole new culture for planning and analysis. It’s agile and truly real-time collaboration.”

The unit/rate/amount structure of the planning application lets the finance staff and community treatment team members understand underlying driver relationships and project forward based on historical rates and visible trends. For example in the screenshot, the average year to date actual Revenue and Allowance rates for each Treatment Team are the basis for forecasting forward into May and later timeframes. These rates are automatically computed from the underlying units and general ledger amounts.


4

Healthcare, an Industry Under Pressure

Forces old and new are buffeting healthcare pro-viders. General economic malaise, state financial disorders, demands for lower charges concurrent with higher quality—the list goes on.

At the state level, budgets are strained; cuts of dis-turbing proportion are being made across all spending areas. Medicaid, accounting for 17% of state budgets, will surely be a victim of some cost cutting. In 2010 states received some relief through Recovery Act (Stimulus) funds to make ends meet. That federal buffer will not be there in the future. To compound the situation, the faltering economy of 2008 and 2009 moved many formerly employed persons with commercial health insurance into the ranks of the unemployed enrolled in Medicaid. While there are signs that the economy is rebound-ing, unemployment remains at stubbornly high levels. State tax receipts should be on the rise with the economic recovery, but will it be enough?

Can you say “Baby Boomer”? America is getting grayer, and living longer. While that may be a good omen for the country (increased wisdom comes with increased age, right?), the reality for healthcare providers is that the average Joe and Jane will need more services than before, and will use those services for longer than any prior genera-tion. The irony here is that advances in healthcare are the primary cause of the increased longevity.

Numerous challenges and opportunities spring from the historic Affordable Care Act (The Act). The interpretation of purported benefits from The Act differ depending on whether you are Democrat or Republican, young or old, currently insured or uninsured, healthcare provider or healthcare in-surer, government entity or private industry. Where there is general consensus: some 32 million people are likely to be added to the ranks of the

insured, and significant change is happening in the industry with uncertain results for each provider.

One aspect of The Act is a reduction in government payments for medical care. Offsetting this, to some

extent, will be a reduction in uncompensated care, as the 32 million shift to insured status in 2014. Healthcare Insurance Exchanges also lead to uncer-tainty, as they will be open and available even to those currently insured.

The concept of Accountable Care Organizations (ACO’s), coupled with the mandate for value-based reimbursement for evidence-based quality service, will certainly lead to hospital-physician alignment. The ACO concept will likely lead to greater affiliation between acute care hospitals and community hospitals as well.

It will be critical for healthcare providers to analyze shifts in the Payer Mix (or erosion of the mix, as some have labeled it) to government programs, primarily Medicaid. Also key will be analyzing reductions in payments. As always, providers will need to con-stantly monitor their cost efficiency efforts in both staffing and supplies, and keep a strong focus on their Revenue Cycle improvements.

Can you say “Baby Boomer”? The reality for healthcare providers is that the average Joe and Jane will need more services than before, and will use those services for longer than any prior generation.


5

Requirements for Agile Planning

The above trends in healthcare are pressuring fi-nance teams in provider organizations to make planning systems more immediate and responsive. The old way of thinking—that you run the facility through the budget process—is on the way out in favor of alternate approaches such as Agile Plan-ningTM as practiced by Ray Wolfe and the Pitts-burgh Mercy organization.

Historically, budgeting has been the command and control foundation for managing organizations—that once-a-year Uber-process that produces every-thing you need for reporting and control: revenue targets, headcount and spending details, capital budgets, and more. The budgets may or may not be supplemented monthly or quarterly by a fore-cast process whereby managers theoretically have the opportunity to re-allocate resources based on changing conditions and new information. Fore-casting, though, more often turns into a politically charged exercise for testing target achievement—focusing on the sensitive question “will individual organizations make their top and/or bottom line financial objectives for the fiscal year”? How and when individual organizational units answer the question, of course, is the foundation for many po-litical games.

The table below lays out a comparison of tradi-tional budgeting and forecasting processes com-pared to Agile Planning as illustrated by the Ray Wolfe/Pittsburgh Mercy case study. Here are the highlights comparing budgeting, forecasting and

agile planning in terms of basic goals, process areas and common features:

Goals—Basic resource management is an im-portant goal of all planning processes. For budgeting and forecasting there is the added “control” element of setting standards for measuring performance which is presumably achieved through precision of the data in budgets and monitoring the accuracy of projec-tions in the forecasting process. In contrast to budgeting and forecasting, however, the goals of Agile Planning are to achieve higher level management effectiveness by gaining insights into the current business situation resulting in greater clarity about what actions can be taken, including the financial impacts of each, and more timely decision making as a result.

Key Process Areas—In their quest for preci-sion and accountability, all line managers par-ticipate in budgeting if they have any level of revenue or spending responsibility regardless of materiality. The process is one of meticulous submission and approval of versions which causes budget development to span many months; thank goodness it’s done only once a year. By contrast, forecasts must be put to bed within days or just a couple of weeks because of the month/quarter end reporting cycle. Thus, forecasting tends to be more finance-driven with a less formal, more ad hoc partici-pation of line management.


6

By contrast again Agile Planning, which is more often driven by business events (e.g. shifting payer mix) or strategic initiatives (e.g. a major capital equipment purchase), is a real-time collaborative activity or series of activities requiring the relevant managers, meaning those who are players, either information providers or decision makers, in the business event or strategic initiative. Unlike budgeting and fore-casting, the cycle time for planning sessions and decision making is typically hours, or at most, days.

Common features—Both budgeting and fore-casting are financial-data driven at the lowest level of detail of natural class accounts (e.g. all Travel sub-accounts rather than Travel in total) and departments. This level of focus results from the passionate need to produce variance reports and analyses—actual versus budget and forecast to forecast—which is presumed not possible if the lowest level GL structure is not adhered to. The result of this low-level fi-nancial focus is that for budgeting and fore-casting, there are a lot of input cells that need to be filled in. The focus is data entry with minimal modeling of interrelationships. By contrast, Agile Planning is frequently done at a higher level of detail than budgeting and fore-casting—e.g. a summary level of accounts such as Travel rather than Travel Sub-accounts.

Despite the problems, few healthcare finance or-ganizations are ready to dump budgets per se. However, many are looking critically at their spreadsheets or canned budgeting applications

asking how they can deliver more immediate and responsive financial plans including incorporating agile planning elements into forecasting processes.

The accompanying challenge for finance is to find ways to make forecasting less political. In health-care, the evolution to agile planning has been slower, principally due to the lack of flexible plan-ning tools.

In the materials that follow, we walk through three software feature/function areas that are central to all planning applications, but especially critical for moving up in planning maturity to achieve the Ag-ile Planning goal of real-time insights leading to actionable knowledge with financial impacts which, in turn, lead to faster, better decisions.

Robust Scenario Analysis—the issue is not whether an application supports versions and scenarios; they all do. The issue is maintaining the scenarios, visibility into the underlying ac-tivity drivers of each scenario, and response times.

Driver Based Modeling—the issue is not whether an application includes modeling tools; all of them do. The issues are how flexi-ble are the tools for building complex driver-based models and how easy is it for the super user to build and maintain the models.

Integrating Actuals—the issue is not whether an application can import actuals data; all of them do. The issue is how well imported actu-als can be integrated into the financial plan.


7

Robust Scenario Analysis

Budgeting is about managing versions, not scenar-ios. As the budget is developed, finance keeps track of versions to understand who changed what and to make sure the right amounts are approved. Once a version is superseded, there is rarely a need to look back at the old numbers or change them.

By contrast, rolling forecasts and agile planning should be about scenarios, lots of them. If you can’t predict the future, the next best thing is to set up scenarios that let you explore how you might be-have (or decide) if things are better or worse or just different. Unlike budgeting where you care about who changed what number, scenario analysis is about understanding what’s behind the numbers—the most critical assumptions, volume and rate im-pacts, and especially what’s driving material changes to the Statement of Activities and Cash Flow.

The goals of agile planning— a) insights about the business; b) actionable knowledge and c) decision making—are achieved through scenario analysis: by analyzing a specific scenario and comparing it to a baseline case or other scenarios, the manage-ment team is better able to understand what’s go-ing on and then evaluate best courses of action. Where there is an immediacy to the issues—e.g. to proceed with a capital project or change suppli-ers—the deliverable is decision making. Because it is decision and action focused, robust scenario analysis is the most critical underpinning of agile planning.

The functionality you need for effective scenario analysis goes beyond simple budget versioning. Here are our criteria:

Real time feedback. Whether you’re a finan-cial analyst working through the numbers late at night or the CFO answering questions live in an operations review, scenario analysis should be delivered by the planning tool in real time. That is, when you change a value, all elements of the financial model—the statement of activi-ties, balance sheet, cash flow, financial ratios, performance metrics—should update in sec-onds, not minutes or hours. In short, scenario analysis must satisfy the need for speed we’re used to with Excel. Scenario analysis must be an interactive process responsive to questions and testing of assumptions on-the-fly. Agile planning needs tight feedback loops on the numbers.

Unlike budgeting applications, in Alight you can change values of line items across multiple scenar-ios. Clicking the Scenario button lets you choose which Scenarios you want to apply the new values to. Updates to financials are immediate


8

Maintenance across scenarios. Budget ver-sions don’t require ongoing maintenance be-cause old versions are superseded by new ones. Scenarios do. The planning application should support adding, modifying and delet-ing line items across selected scenarios in a sin-gle operation. Calculation and update of finan-cials after structure changes should take only a minute or two, at most.

Robust comparison at the line item level. Budgeting and to a considerable extent, rolling forecasts focus on amounts in accounts. Agile planning is about in-depth comparison of sce-narios and differences in values at any level of detail, especially at the line item level where the most significant driver inputs and model-ing occur.

Why the Variance?

Alight lets you compare scenarios at any level of detail including analysis of underlying units, rates and amounts. Below is a Variance Analysis comparing the Affordable Care Act Impact to the Base Case Scenario. Unlike standard variance reports that show only dollar amounts, this Volume-Rate Variance report automatically computes the volume and rate components by line item — that is, the unfavorable impacts of a payer mix shift to Medicaid and a declining rate of government payments. The con-

clusion is obvious: the shift to Medicaid is the primary contributor to the Variance.


9

Driver-Based Modeling

Budgeting focuses on gathering static inputs from users, principally for headcount and expenses. Lit-tle modeling is involved— typically just calculation of payroll taxes and benefits.

Such a process is not workable for agile planning. The cycle time for completing a forecast after month-end close is too tight to accommodate broad-based user involvement on the scale done for budgeting. As well, user-based planning involves high volumes of static inputs in order to achieve precision. By contrast, forecasting that incorporates agile planning reduces the volumes of data, focus-ing instead on fewer but truly material forecast elements.

The objective of reducing data volume is achieved in substantial part through “driver-based plan-ning”, a type of financial modeling where the most material items in a financial plan are linked to op-erational drivers or quantifiable activities of the business —e.g. volume measures such as patient admissions, length of stay, patient days, number of procedures, number of tests, and the like.

Such driver-based modeling has three focuses: 1) revenue forecasting based on the volume measures and driver relationships between revenue areas—e.g. contractual rates by payer type, 2) variable and semi-variable headcount and expenses driven by the volume measures or identified underlying ac-tivity levels—e.g. admissions, patient days, etc.; and 3) for cash planning, balance sheet items such as accounts receivable and accounts payable, each based on their respective income and expense driv-ers at a relevant level of detail.

Healthcare organizations moving to an agile plan-ning discipline should closely evaluate the model-ing environment of the planning application they are buying into. Modeling and model maintenance should be fast, flexible, and (unlike Excel) provide clear visibility into linking relationships.

In our experience, there are two make-or-break functionalities in the modeling interface needed to make driver-based planning work:

Visibility into driver relationships. This re-quires object based linking where the modeler establishes data relationships based on the names of things—e.g. Patients Days are linked to Admissions and the Average Length of Stay—versus cell-based linking in Excel with formu-las like = Stats! C45 * $L$15. Object-based link-ing makes auditing and visibility into activity driver relationships significantly easier to track, thus eliminating many errors, speeding up development and reducing maintenance time.

Modeling across dimensions. Planning appli-cations should support custom dimensions—e.g. department, payer, stat, facility, job type, grade level, etc. The modeling environment should then allow tapping into the custom di-mensions for building specialized activity-driver models—for example, aggregating in-patient and out-patient data across depart-ments and payers. Applications without mod-eling based on dimensions do not have the flexibility required for driver-based planning.

Alight’s modeling interface is based on object-based linking where you create data relationships by linking to the names of other line items or totals, e.g. drug expense for ICU is based on average drug cost per stat X ICU stat (patient days)


10

Integrating Actuals

Virtually all budgeting applications support im-porting actuals from the general ledger, typically based on rigid chart of accounts structures. As well, you can usually import headcount and sala-ries from the HR system. Once imported, budget applications also support comparing actuals to budget based on levels in the chart of accounts with computation of amount and percentage vari-ances.

Outside the safety of the accounting structures, however, lining up actual and plan data in a con-sistent format for a rolling forecast is typically a zoo, principally because it is difficult to get actuals data apples-to-apples with plan data at the line item level.

For example, for basic revenue planning:

Patient Days

* Reimbursement Rate

= Gross Revenue

250 * $1000 = $250,000

For actuals, however, the data is typically amount and units. Therefore, for an apples-to-apples com-parison, the actuals data must be modeled to back-calculate the reimbursement rate:

Gross Revenue

/ Patient Days

= Reimbursement Rate

$250,000 / 250 = $1000

Alight includes a section for Key Measures with custom dashboards for displaying actual and plan measures of the organization such as admissions, patient days, denial and collections rates, payer mix, headcount analysis, balance

sheet ratios and more.


11

As the examples illustrate, actuals data frequently need modeling different from plan data to get valid comparisons and trends. For maximum flexibility, the planning application should allow modeling of actuals with algorithms and linking separate from modeling of plan data.

In addition, actuals data need to be imported from any source at any level of detail, not just the general ledger. The actual revenue amount is imported from the general ledger, but units for back-calculating average charge rate are typically im-ported from a data warehouse or an ERP or EMR system.

Finally, for robust agile planning, you must have the capability of truly integrating actuals into the financial plan. That means: 1) being able to model actuals as an activity driver and 2) spreading actu-als data into plan time periods—e.g. the historical average cost per stat into future months, and fur-ther, being able to modify those costs going for-ward to reflect, for example, inflation or cost con-trol initiatives.

Summary

Agile planning requires special features that extend beyond the capabilities of spreadsheets and many budget applications. Scenario analysis should be real time with maintenance across scenarios. Mod-eling tools must be robust and provide transpar-ency into underlying activity drivers. Actuals data must be available for driving forecast values.

In an agile planning environment, the return on investment is not from reducing time spent on planning. The ROI comes from better financial per-formance as a result of a more responsive planning environment and improved decision making.

Agile Planning is a pending trademark of Alight, LLC.

Alight includes separate tabs for modeling actuals and plan at the line item level. In the example, plan units * rate = amount. For actuals, amount / units =

rate.

Rand Heer is CEO of Alight and the creative force behind Alight Planning. He was also the founder of Pillar and designer of Hyperion Pillar, the first enterprise software for budgets and forecasting.

Tom Hartman, a twenty year veteran of the planning software space, is Director of Sales at Alight, driving the company’s Healthcare sales and marketing programs.

Web: AlightPlanning.com Tel: (530) 622-5485.

Special Thanks to Ray Wolfe, CFO of Pitts-burgh Mercy Health System , Don Koenes, VP of Services at Alight and Linda Hull, Director of Sales Support at Alight.

Agile Planning for Healthcare Providers

Education

Transcript of Agile Planning for Healthcare Providers