Best Practices in Physician Relations

by Lee Ann Lambdin

Interviews with: Mitzi Kent and Dustie Maguire


Originally shared as a webinar, on April 12, 2016, this interview with Stratasan and LifePoint Health discusses the organization of physician relations, goals and responsibilities, hiring and leadership, relationship building techniques, intelligence and information, and compliance. You can see presentation slides and hear the audio from the webinar presentation here.


Read More

Jackpot! Why Winning the Lottery Won’t Change Our Data

by Jason Haley


With the three big winners in last night’s lottery splitting such a large pot, the income-level demographic data for where they live will be effected. However, depending on the way you look at this effect, there will either be slight changes if any, or there will be a large change. The reason for this difference is that demographic datasets supply both mean and median household incomes.LottreryBalls_Clipped

The lottery winners provide us with a great example of why we run median household income rather than average household income. This is because using a mean generally works well for data with normal distributions while medians are generally used on data with skewed distributions. And as you can probably guess, income data is quite the skewed dataset. Since a mean is so heavily influenced by outliers, we use a median. The median value will provide the value in the middle of the data (when sorted in ascending order).


To illustrate this, let’s pretend that one of the lottery winner’s home ZIP code has 10,000 people. Let’s also pretend that, by some kind of freak chance, every one of these people has an income of exactly $50,000. That would mean that that ZIP code has an average and median household income of $50,000. But, the lottery winner has now changed that. With their income changing from $50,000 to somewhere around $500,000,000, the new mean household income would be $99,995 and the median income would remain $50,000.

Read More

Achieving More with Less: Working Smarter, Not Harder

It might seem like ages ago, but do you remember a time before e-mail?  There are likely some of you who can, and some of you who can’t.

When I first entered into the healthcare industry, computers were making their way into just about every healthcare setting, but emailing had yet to come.  You would type a MEMORANDUM interoffice envelopeusing a word processing program, print it, place it in an interoffice envelope and write the recipient’s name on the outside of the envelope.  You would then place the envelope in an outgoing mail box in your office.

Read More

Introducing: Hospital Market Share Dominance Maps

We are constantly searching for new, effective ways to present intelligence from data. In order to accomplish this, the Stratasan team uses tables, charts, graphs and maps instead of raw data tables alone. Pairing these visual tools with raw data increases the efficacy of our products and adds a necessary element of excitement to the black and white world of data for our customers. The most common requests we fulfill are those around market share. This report specifically benefits from converting the data into visuals. Our goal is to present market share data in a way that clearly displays a hospital’s (and their competitors’) presence in a market. We wanted it to be visually interesting while still providing a useful picture of a market area. So, Stratasan is pleased to introduce the Market Dominance Map.

Market Dominance Map

Map illustrating market share leader by ZIP; color gradient illustrates how strong their market share is within each ZIP.

The Market Dominance Map (above) shows hospital reach by color and the significance of that reach by color saturation. If a hospital has a majority market share in a ZIP code, that ZIP is filled with the hospital’s color according to the map legend. Depending on how high the hospital’s leading market share is, the color is shaded from dark (high market share) to light (lower market share). We refer to this visual representation of reach as “dominance” of a ZIP code.  The percentage number in each ZIP represents the leading hospital’s market share. As you can see in the example above, only the leading hospital per ZIP is represented. This is a quick and clear picture of the market’s major providers and what portions of the market these providers have cornered. “But what about our competition within ZIPs?”

Market Dominance, Highly Contested ZIPs

ZIPs with less than 10% market share separating the two leaders are cross-hatched

To increase the value of this project, including the top competitor is the obvious move. To maintain the clarity of the map, we put a limit on what constitutes “competition.” In each ZIP code, if the dominant hospital has market share less than ten percentage points higher than the next hospital; the second leading hospital’s color will cross-hatch through the ZIP.

Highly Contested ZIPs

Limiting map to only those ZIPs that are “highly contested”, defined as the top two market share leaders being within 10% of each other.

The final view for this project is to give the hospital strategic planning team a view of the ZIP codes in their service area where the market leader is in a “highly contested” market share race with another facility. The percentage listed in each ZIP code is the market share percentage of the second facility. Market share is an undeniable part of hospital strategy. For a clear, concise portrait of dominance in your market, contact us today.

Community Health Needs Assessment – Implementation, Monitoring and Evaluation

Your Community Health Needs Assessment is complete with lots of community involvement.  Your community has assisted in prioritizing the most important health issues.  Your hospital has selected and prioritized the health issues and goals that it will work on, and has justifications for the issues it will not work on.  There is still one more requirement to be met: an implementation plan with monitoring and evaluation of efforts.  Stratasan suggests creating a Health Issues dashboard containing as much of the following information as possible.

  • Your hospital’s prioritized health issues
  • Action items related to health issues
  • Measurable goal statements
  • Pre-CHNA health indicator score
  • Your hospital’s goal for the health indicator
  • National or state median for comparison
  • Best practice or national benchmark for the health indicator
  • Healthy People 2020 goal
  • Source of the data
  • Timeframe for update of the indicator

For example, if the hospital selects diabetes as a community issue to improve, it may have an action of increasing exercise in the population through community exercise programs, a fitness center, and physicians’ offices prescribing exercise for their patients.  The health indicator score may be physical inactivity defined as the percent of adults aged 20 and over reporting no leisure time physical activity.

The pre-CHNA score was 26% and the current year’s available data may indicate 27% physical inactivity.  This is in comparison to national and state benchmarks of 21% and 30% respectively.  The Healthy People 2020 goal is a 10% improvement, so the goal would be set to reduce down to 24%.  The source of the data is the CDC’s Behavioral Risk Factor Surveillance System (BRFSS) survey.  The current year available is 2010.

A concern relative to the implementation plan is the efficacy of initiatives that hospitals and other community organizations are undertaking.  Are they actions proven to result in health improvement? contains interventions and resources to assist hospitals select evidence-based actions.

The drawback to many of the health indicators it he lag time in measurements.  Unless the hospital is performing its own survey of the population, many of the national surveys, BRFSS and National Health Interview Survey or American Community Survey, will lag behind at least one year.

According to a report form the CDC regarding Best Practices for Community Health Needs Assessment and Implementation Strategy, “A key concern in the community benefit arena is the relative capacity of hospitals to monitor and evaluate the effectiveness of programs and activities.  This becomes more important in the context of health reform, with the anticipation of a reduced demand for charity medical services, and increased emphasis on addressing the underlying causes of persistent health problems.”

The key is to select actions that are evidence-based and will result in actual health improvement and to measure the results even if the measure lags behind the efforts. The IRS allows a hospital to keep the same implementation plan for three years.  The measures will eventually reflect the actions of the hospital and the community.  If the hospital finds its actions are not improving the measure, they may make adjustments to the implementation plan.

For help with your Community Health Needs Assessment and Implementation Plan, give us a call: 866-628-5051 or visit

2012 Medicare Statistics

Medicare accounts for 38.4% of the total U.S. discharges based on the latest Stratasan projection of 38,919,778 inpatient discharges. The ability to understand who, and where your Medicare patients are coming from is critical to all providers who accept Medicare as a payor. Below are some key statistics from the 2012 Medicare Provider Analysis and Review (MedPAR) claims file.

Table 1: Top 10 DRGs sorted by total number of discharges for Medicare inpatients

Table 1: Top 10 DRGs sorted by total number of discharges for Medicare inpatients

The top DRG by volume in 2012 was 470, Major joint replacement or reattachment of lower extremity w/o MCC (Major Complications or Comorbidities), passing Psychoses by nearly 100,000 discharges. The average length of stay for DRG 470 is three-days, which is two days less than the overall average.

Figure 1: Medicare rate per 1,000 persons by age group

Figure 1: Medicare rate per 1,000 persons by age group

Figure 1 (above) illustrates an increase of Medicare use rates as the age group increases. As expected, the 85+ age group has the highest use rate, with 514 Medicare discharges for every 1,000 people. This is more than 80% higher than the rate of the total population 65 years and older. Table 2 (below) shows the number of discharges and total population for each age group over 65.

Table 2

Table 2

The map below illustrates Medicare use rates by ZIP code per 1,000 people, age 65 and older. Due to this being a rate per 1,000 people, there is a natural adjustment for low population.

There are concentrated areas of high Medicare rates throughout the southeast. As you move west across the country, Medicare use rates decrease significantly. A few exceptions exists in areas such as Tucson and Albuquerque, where Medicare use rates are high compared to the region.



Table 3

Table 3

Table 4

Table 4

Table 3 (above) lists the top 5 states by Medicare discharges. California is leading the country by volume, with 8% of total Medicare discharges. However, because of such a high population, California’s use rate is not as high in comparison to other states such as Kentucky or West Virginia, as you can see in table 4 (above).

At Stratasan, we work with our clients to help them gain a better understanding of the patients they serve. Contact us today for more information.


What’s Going on inside the Heart of Cardiac Services?

Recently, we performed an analysis for a customer on cardiac services trends.  Based on previous studies, we knew cardiac inpatient services were declining overall.  However, we discovered that inside cardiac services many of the sub-service lines behave quite differently; many declining but some actually increasing.  We analyzed all-payer data from Florida, Texas and California alongside full U.S. Medicare data and discovered a deeper understanding than “inpatient cardiac services are declining.”

The first step was to define sub-categories for Cardiology and Cardiovascular/Thoracic Surgery (CVT surgery) service lines for a clearer picture of where declines are occurring.

Cardiology Product Line

Cardiovascular/Thoracic Surgery Product Line

Total Medicare Cardiac Services inpatient cases declined, -8.6% from 2009 to 2011.  Cardiology inpatient cases declined, -7.6% from 2009 to 2011.  Cardiovascular/Thoracic Surgery inpatient cases also declined but at a much larger rate, -11.2% from 2009 to 2011.

Screen Shot 2013-03-12 at 8.32.55 AM

Looking inside Cardiology, we can see that the largest decline in Electrophysiology (EP)/Devices (pacemaker replacements and revisions).  This decline was followed by Cardiac Catheterizations, Other Cardiology and Acute Myocardial Infarctions (Acute MI).  Acute MIs declined the least, -6.5% from 2009-2011.

Screen Shot 2013-03-12 at 8.33.09 AM

Although Cardiovascular/Thoracic Surgery declined more than Cardiology, some subservice lines actually increased in volume. Heart and lung transplant MS-DRGs increased 32% from 2009 to 2011.  Valve surgeries increased 3.9% from 2009 to 2011.  Major chest trauma increased 6.4% from 2009-2011.  The largest declines in the CVT Surgery service line were Cardiac defibrillator implants with cardiac catheterization, declining 23% from 2009 to 2011.  This decline was followed by EP/Devices declining 18%.  Cardiac bypass surgery is down 13% and Cardiac interventions are also down 13%.

Screen Shot 2013-03-12 at 8.34.54 AM

Looking even further inside the Transplants sub-service line (MS-DRGs 001, 002, and 007).  MS-DRG 001, Heart transplant or implant of heart assist system with major complications or comorbidities increased 51.7% from 2009 to 2011.  The proliferation of heart assist systems has propelled this tremendous growth, not necessarily heart transplants.

State Data Comparisons

In California and Florida we analyzed the same time period, 2009 to 2011 for trends.

In California, the Cardiology trends are the same as for Medicare, declining 5.8%.  The highest decline was in EP/Devices (-17%), and the lowest decline in Acute MI (-1.5%).  In CVT Surgery, again the same trends, total volume down 6.2% with increases in transplants (21.9%), trauma (11.6%) and valve surgery (9.5%) while EP/Devices with Cath (-13.2%) and EP/Devices (-11.8%) declined the most.

In Florida, Cardiology declined 5.4% while CVT surgery declined 8.5%.  In Cardiology, Cardiac Caths (-8.2%) edged out EP/Devices (-7.2%) for the highest declines in Cardiology.  Other Cardiology (-4.2%) declined less than Acute M.I (-6.2%).  In CVT Surgery, the same trends appeared, transplants (21.6%), trauma (13.5%) and valve surgery (13.5%) increased while the EP/Devices with Cath (-23%) led the decliners followed by EP/Devices (-14.8%).

In Texas, we analyzed 2010 and 2011 data.  All of the Cardiology sub service lines declined  for a total 2.7% decline, and EP/Devices (-9.3%) led the way.  CVT surgery declined 3.5%, but the trends within CVT surgery shifted slightly, transplants were down (-17.8%), but Other CVT surgeries increased 4% from 2010 to 2011, trauma (8.6%) and valve surgery (4.1%) also increased.  The decliners were led by EP/Devices with Cath (-27.1%) and EP/Devices (-10.2%).

If you would like a cardiac services analysis for your market, give us a call. Phone: 615-500-3497, Fax: 615-208-9657, Email:,

Sources:  Center for Medicare & Medicaid Services, Medicare Provider Analysis and Review (MedPar) hospital discharge data, 2009-2011. State of California Office of Statewide Health Planning and Growth, 2009-2011 hospital inpatient discharge data.  State of Florida Agency for Health Care Administration 2009-2011 hospital inpatient discharge data.  Texas Department of State Health Services, Texas Health Care Information Collection 2010-2011 hospital inpatient discharge data.


The Art (and Science) of Defining Service Areas – Part 2

This post is a follow-up case study to illustrate the detail outlined in Part 1 found here.

Case Study – Memorial Hospital, Anytown, USA. 

In the last blog post, we covered how to define a service area.  In this post, we will cover a case study demonstrating the pros and cons of different service area definitions focusing on competitor identification.  Below is a case study based on a real hospital and real service areas blinded to protect the innocent.

Anytown is a city of 33,000 located in Mine County population 159,000, 20 miles north of a metropolitan city with population of 752,000.  Below is a table of market share by hospital by the different service area definitions discussed in-depth in the previous blog post.  Eighteen ZIP Codes represent 90 percent patient origin, 7 ZIP Codes represent 75 percent patient origin (all contiguous ZIP Codes meeting Stark service area definition) and 2 ZIP Codes represent 50 percent patient origin.   The 7 ZIP Code area is essentially the primary service area (PSA) and secondary service area (SSA) defined by PSA being 50 percent patient origin (the 2 ZIP Code area referenced above) and the SSA the next 25 percent of patient origin combined.


The table above demonstrates how the competitors change based on the definition of service area.  Using the two larger service area definitions, the primary competitor is Neighbor Hospital.  However, when only looking at the two ZIP Codes that make up 50 percent of patient origin, Large Medical Center becomes the primary competitor.  If market share by service line is available, look at outmigration by service line to Large Medical Center.  The only services out-migrating may be tertiary services Memorial Hospital doesn’t provide.  In that case, our focus would shift to Nearby Town Medical Center as our primary competition if they are seeing patients in service lines where we offer the services.  In the 7 ZIP service area, Neighbor Hospital is our primary competitor, followed by Nearby Town Hospital.

Also, notice the how the market share percentages change as the service area changes.  Using the 18 ZIP Code definition, our hospital has a market share of 13.1%.  When using the 7 ZIP Code service area our market share is 27.7%.  Using the 2 ZIP Code definition, our market share is 56.9%.  I am arguing the primary service area of our hospital is the 2 ZIP Code definition.  Those are the people we primarily serve.

This exercise demonstrates the importance of a thoughtful, deliberate service area definition.  A well-defined service area assists with targeted marketing and physician business development.  It also assists in understanding who your true competitors are by service line.  The service area definition is both science and art involving the use of analytics and judgment.



The Art (and Science) of Defining Service Areas – Part 1

Part 1 – How to Define a Service Area

The first step in any planning or marketing project for a hospital/provider is to define a service area.  Before demographics or market share can be run, an appropriate service area must be applied.  Defining a service area is both art and science.

There are several ways to define a service area.  Which one is best?  The answer is: it depends.

Regional Strategy Approach

With the regional strategy approach the facility defines a service area as large as possible.  A facility may define its primary service area as 75 percent of its inpatients (and/or outpatients) and its secondary service area as the next 15 percent for a total of 90 percent patient origin.  This is the “no ZIP Code left behind” definition.  This service area definition will most likely incorporate a very large area and ZIP Codes where the hospital has very low market share.

Service Area defined by 75% and 90% of patient origin results in an over-expanded view


Two of the key areas where an overly large service areas cause issues are marketing and physician recruitment. Marketing departments with limited budget and staff struggle to make a meaningful impact over such a large geography.  The large area also causes a lack of focus by your physician liaison potentially leading to missed opportunities with great partner physicians close to your facility. However, if a facility is looking at a regional outreach strategy or growing a specialty service line, a larger service area may be appropriate.

Stark-Influenced Definition

The 2004 Stark regulations relative to physician recruitment define a hospital’s geographic service area as the lowest number of contiguous ZIP Codes from which the hospital draws 75 percent of its inpatients.  Some hospitals prefer to define their service area in this manner, essentially killing two birds with one stone – the service area for marketing/planning etc. matches the physician recruitment service area.  However, what is ideal for physician recruiting may not be ideal for service line marketing and development.  If 50 percent of Cardiology patients come from 2 ZIP Codes, but 10 ZIP Codes make up 75 percent of patients in contiguous ZIP codes, cardiology development loses geographic focus.

The Stark defined service area for a hospital is an easy solution, as you should already have these defined. Be careful when using tools out of convenience over selecting the right tool for the job.

Stratasan’s ZIP Code Recommendation

When using ZIP Codes, we prefer a more focused primary service area representing 50 percent patient origin with a secondary service area for the next 25 percent of patients.  If a tertiary service area is warranted, then the next 15 percent of patients is sufficient.  Sounds simple, right?  As football analyst Lee Corso says, “not so fast my friend.”  The percentages are science, now for the art.

Reducing your definition to 50% and 75% tightens your ability to know your patient base


When selecting ZIP Codes for a Primary Service Area (PSA) and Secondary Service Area (SSA) as defined above, gather five pieces of information: patient origin by ZIP Code, market share of top 75 percent of patient origin ZIP Codes (if available), population by ZIP Code, and the physical location of the ZIP Code on a map.

Select ZIP Codes for the PSA representing approximately 50% of patient origin (blue area), where the facility has high market share, and that make sense geographically.  Select ZIP Codes for the SSA that represent approximately the next 25 percent of patient origin, where the facility has decent market share, and that make sense geographically.

What does “make sense geographically” mean?  It means eliminating ZIP Codes on an “island” away from the rest of service area or adding “bullet holes” back in to fill in gaps inside the newly defined service area.

Removing the ZIP (37210) in the northeast on an “island” allows for even further tightening of your Service Area


An example of this is the removal of ZIP 37210 in the map above. The floating ZIP Code is usually a highly populated ZIP Code in a nearby larger city where your facility receives some patients placing the ZIP code in the top 75% of patient origin.  However, your market share of that ZIP Code may be very low, 2% for example.  You may want to omit this ZIP Code in favor of ZIP Codes closer to the facility where the facility has higher market share.  An example is 65432 above.  It is a very large ZIP Code in a metropolitan area south of our facility.  There are facilities that designate dream ZIP Codes in their service, meaning those they dream of serving.  This is fine as long as you don’t actually include these in your real service area, defined as those people you currently serve.

Moving Away from ZIP Codes

But that is not the end of the story. When patient data is available, a custom polygon shape file built from your patient address level geo-coded data can serve as the best solution for all parties involved.

Building a custom Market Service Area based on your patient data tells the most accurate story


The service area above is based on an algorithm between the hospital location and the actual addresses of their current patients. ZIP Codes are built to deliver mail. They have no real influence on where a patient decides to receive healthcare services (when they have a choice).

Large populations are included in your service area by using standard geographic boundaries like ZIP Codes


Overlaying this new approach with the tighter ZIP Code approach helps you visualize the difference.

Tighter Area + Market Share + Population

Using a tighter service area is a great first step into improving your service area definitions. But many times this is not the end of your journey. Adding a quick study of market share and population by ZIP Code keeps a facility from “majoring in the minors.” What we mean by this is focusing too many efforts on the wrong ZIP Codes.


For example, our fictitious Brentwood Medical Center facility gets a mere 3.8% market share of ZIP Code 67891.  The other ZIP Codes in the primary service area warrant more attention because we already have better traction but have opportunity to grow.  One could even argue that if a facility only gets 4% of a market, then it’s NOT your service area.  You may WANT it to be your service area, but it isn’t.  In ZIP 87654, our facility gets 27% market share.  However, there are only 2,400 people in the ZIP Code.  I would recommend focusing on another ZIP Code with more population first.  In fact, I am arguing for focusing on two ZIP Codes highlighted in yellow first, then the remaining list in the PSA, other than 67891.

Arriving at Your Destination Definition

Based on all five pieces information, selecting your primary and secondary service areas (whether ZIP Code-based or Market Service Area) is much more precise.

Select ZIP Codes for the PSA that represent approximately 50% of patient origin, where the facility has high market share, and that make sense geographically.  Select ZIP Codes for the SSA that represent approximately the next 25 percent of patient origin, where the facility has decent market share, and that make sense geographically.

What does “make sense geographically” mean?  It means eliminating ZIP Codes on an “island” away from the rest of service area or adding “bullet holes” back in to fill in gaps inside the newly defined service area.

A floating ZIP Code is usually a highly populated ZIP Code in a nearby larger city where your facility receives some patients placing the ZIP code in the top 75% of patient origin.  However, your market share of that ZIP Code may be very low, 2% for example.  You may want to omit this ZIP Code in favor of ZIP Codes closer to the facility where the facility has higher market share.  An example is 65432 above.  It is a very large ZIP Code in a metropolitan area south of our facility.  There are facilities that designate dream ZIP Codes in their service, meaning those they dream of serving.  This is fine as long as you don’t actually include these in your real service area, defined as those people you currently serve.

The benefits of a well-defined service area are: better use of constrained resources, more focused resources in marketing and physician liaison budgets and personnel, focused development of service lines, and better defined competitors.

Page 1 of 212