Where does healthcare cost the most?

In 2016, America's $679 billion Medicare expenditures provided health insurance coverage for 56.8 million beneficiaries. According to a 2017 Medicare Trustees Report, the program remains solvent through 2029. But after that, the effects of a rapidly aging population are expected to consume program income. In this lesson, you'll explore Medicare's costs by creating a national heat map reflecting regions experiencing high and low costs. Identifying areas of high cost is the first step in understanding and containing the factors threatening the program's longevity.

This lesson was last tested May 5th, 2021.

View final result

Requirements

• Publisher or Administrator role in an ArcGIS organization (get a free trial)

Map Medicare costs

Maps are important decision-making tools. They help determine problem areas and indicate where resources can be better spent. But maps don't always present a single truth. Sometimes you can find different views of the truth in the same set of data. In this case, you'll consider the per capita (per person) cost of the Medicare program in 2012. These costs vary appreciably from place to place. When you make a map, you need to make decisions about how to group these varying cost values. Which range of costs is high or low? Your decisions help create spatial patterns, and these patterns lead map users to draw conclusions. This raises concerns as to the best way to visualize data and find reliable patterns.

First, you'll compare some common techniques for classifying (grouping) data and see how your choices affect spatial patterns on the map. You'll work with Medicare cost data aggregated by hospital referral region. Hospital referral regions are regional health care markets defined by the Dartmouth Atlas of Health Care. Each region contains at least one city where both major cardiovascular surgical procedures and neurosurgery are performed. Medicare is a United States government health insurance program covering about 50 million people who are over age 65 or who meet certain medical conditions. Information about the Medicare program is available from the Centers for Medicare & Medicaid Services.

Open the map

In this section, you'll open a map, familiarize yourself with its features and attributes, and save your own version of the map for further work.

1. Go to the ArcGIS Online group, Where Does Healthcare Cost the Most. If necessary, sign in to your ArcGIS organizational account.

   Note:

   If you don't have an organizational account, you can sign up for an ArcGIS free trial.

2. Click the thumbnail of the Hospital Referral Regions map by Learn_ArcGIS to open it.

   

   The map opens to the United States. The country is divided into health care markets, called hospital referral regions.

   

3. At the top of the Legend pane, click the Content button.

   

   Note:

   Depending on the width of your Contents pane, you may or may not see text next to the button icon.

   The map contains three layers: State Boundaries (turned off), Hospital Referral Regions, and the Light Gray Canvas basemap.

4. In the Contents pane, check the box next to the State Boundaries layer to turn it on.

   Referral regions sometimes cross state lines. Also, they don't completely cover the country: you can see the gray basemap showing through in large, unpopulated areas.

5. Uncheck the box next to the State Boundaries layer to turn it off.
6. Click a hospital referral region on the map.

   

   A pop-up appears, showing the name of the referral region and its per capita Medicare costs in 2012. For example, costs in the Wichita, Kansas, region were $10,061.

7. Click a few other regions to see their pop-ups, and then close the open pop-up by clicking the X in the upper right corner.

   Pop-ups tell you about individual features, but they don't help you see spatial patterns. To see patterns, you need to symbolize the data. You'll do this in your own version of the map so you can save the changes.

8. On the ribbon above the map, click the Save button and choose Save As.
9. In the Save Map window, change the title to Medicare Costs per Capita in 2012.

   

10. Click Save Map.

    The map is saved to My Content.

    Note:

    The per capita cost data you'll work with reflects the standardized risk-adjusted cost. It differs from the actual cost in two ways. First, it's standardized to even out differences in wages and the cost of goods and services from one part of the country to another. Second, it's risk adjusted to even out differences in age, sex, wellness, and other relevant demographic factors. The standardized risk-adjusted value is the best estimate of what the actual costs would be if socioeconomic, demographic, and health conditions were uniform across the country. For a detailed explanation of how standardization and risk adjustment are calculated, see Medicare Data for the Geographic Variation Public Use File: A Methodological Overview (Latest Update).

Map costs by natural breaks

A typical way to present spatial patterns on a map is to associate ranges of data values with colors. There are a few common methods for specifying value ranges. In this section, you'll use the Natural Breaks method.

1. In the Contents pane, point to the Hospital Referral Regions layer and click the Change Style button.

   

2. In the Change Style pane, for Choose an attribute to show, choose Standardized Risk-Adjusted Per Capita Costs at the bottom of the list.

   

   Once you choose the attribute, available drawing styles are presented. The style that is typically most suitable is applied by default and indicated by a check mark.

   

   On the map, the referral regions are now drawn in shades of brown. Darker shades represent regions where costs are higher.

3. In the Change Style pane, on the Counts and Amounts (Color) drawing style, click Options.
4. In the lower part of the Change Style pane, check the Classify Data box. Change the number of classes to 5.

   

   In the map legend, the range of cost values is grouped into five classes by the default Natural Breaks classification method.

   

   Natural Breaks uses clusters and gaps in the value range to define classes—as you might do if you were grouping children by age and had some seven- and eight-year-olds and some ten- and eleven-year-olds, but no nine-year-olds.

   One characteristic of the Natural Breaks method is that value ranges may be different from class to class. Here, the value range of the lowest class ($7,173 to $8,390) is $1,217, while the range of the next class ($8,390 to $9,120) is just $730. Another characteristic is that classes may have different numbers of members. For example, one class may include 40 referral regions while another includes 50.

5. In the Change Style pane, click the Symbols button.

   

6. On the color palette, with the Fill tab selected, choose another monochromatic color ramp, such as purple. Click OK.

   

   The new color ramp is applied to the map.

   

7. At the bottom of the Change Style pane, click OK and click Done.
8. On the ribbon, click the Save button and choose Save.

Explore spatial patterns

You'll explore the spatial patterns on the map by looking at the legend of the Hospital Referral Regions layer, and then zooming to different geographic areas and opening pop-ups.

1. In the Contents pane, turn on the State Boundaries layer by checking its box.

   

   The map shows distinct patterns. Costs are high throughout the South, especially in Texas, Louisiana, Mississippi, and Florida. High costs fan northward through Oklahoma and Kansas and through western Tennessee and Kentucky. There are also isolated high cost areas around the map.

2. At the top of the Contents pane, click the Legend button.

   

   The legend shows the value range associated with each color. In any classification scheme, class breaks are important because they lead map users to form judgments: in one place costs seem to be high, while in another they seem to be very high. In fact, however, the difference between a given pair of values in different classes may be small.

   

3. At the top of the Legend pane, click the Content button.
4. On the ribbon, click the Bookmarks button and choose Southwest.

   

5. Click the Palm Springs/Rancho Mirage referral region and note that the cost is $9,680 per capita.

   

   This referral region falls in the middle class of the five classes.

6. Close the pop-up. Click the Las Vegas referral region, adjacent to the northeast.

   Here the cost is $9,853. The difference between the two regions is only $173, but it's enough to put them in different classes—at least with the Natural Breaks classification method.

7. Close the pop-up.
8. Zoom to the Midwest and Northeast bookmarks and compare some other referral regions.

   Regions that stand out from their neighbors are interesting. Waterloo, Iowa, is one example, and there are others. In some cases, you may feel that the cost differences are as striking as the symbology suggests; in other cases, possibly not.

9. When you're finished, zoom to the USA bookmark.
10. Turn off the State Boundaries layer.
11. Save the map.

Classify the data by other methods

Natural Breaks isn't the only available classification method. You'll see to what extent the spatial patterns change when you use the Equal Interval and Quantile methods.

1. In the Contents pane, point to the Hospital Referral Regions layer and click the Change Style button.
2. In the Change Style pane, on the Counts and Amounts (Color) drawing style, click Options.
3. For Classify Data, change the classification method from Natural Breaks to Equal Interval.

   

   These class breaks are different. The defining characteristic of the Equal Interval method is that value ranges are the same among all classes. In this case, the range is $903. A class can have any number of members, or even no members.

   

   Although a fairly similar pattern of high and low values is evident, a different impression is created. Fewer regions fall into the lowest and highest classes, giving the map a more homogeneous appearance.

4. Change the classification method to Quantile.

   

   The classes change again. The defining characteristic of the Quantile method is that all classes have the same number of members (in this case, either 60 or 61 referral regions). The value ranges among classes may be very different. Here, the value range of the lowest class is $1,411, while the range of the middle class is $375.

   

   In contrast to the previous map, this map gives a strong impression of diversity. The Quantile method tends to emphasize highs and lows and may exaggerate their importance.

   None of the classification methods you've looked at is right or wrong. The Quantile and Equal Interval methods give good results when data is continuously and evenly distributed throughout the value range. This is often not the case, however. When there are gaps and clusters in the data, the Natural Breaks method is better.

   In this situation, the data has a fairly normal, or bell-shaped, distribution, as shown by the gray bar chart adjacent to the color ramp. With this distribution, the Quantile method is a good choice.

   

5. At the bottom of the Change Style pane, click OK and click Done.
6. In the Contents pane, point to the Hospital Referral Regions layer. Click the More Options button and choose Rename.

   

7. In the Rename window, change the layer name to Medicare Costs by Hospital Referral Region, and click OK.
8. Save the map.

Decisions about how to classify data are at least partly subjective. You might like the way a map looks or you might want to convey a certain message. No classification method is wrong, and each may help emphasize an aspect of the data that's not apparent from the others. But you may wonder if it's possible to get a better sense of which spatial patterns are stable and reliable, to know which places definitely stand apart from the others.

The answer is yes: there are analysis techniques that help you group and visualize data in less subjective ways. Next, you'll explore hot spot analysis and see how statistical evaluation can find spatial clusters of significantly high and low values in your data.

Analyze Medicare cost hot spots

Previously, you saw that spatial patterns on a map change depending on how data is classified. Next, you'll run a hot spot analysis on the data in order to draw more definite conclusions about patterns. Hot spot analysis applies statistical tests to find areas where values are significantly different from the norm.

As in the last lesson, you'll work with a cost measure representing average per capita spending on Medicare beneficiaries in 2012. This measure is not the actual cost, but a standardized risk-adjusted measure that equalizes regional differences in the cost of goods and services as well as differences in health risk among local populations.

Find hot spots

You've seen that there are fairly pronounced cost differences around the country. In general, costs are high in Texas, Florida, and the Gulf Coast. They're low in New England, along the West Coast, and in some other parts of the country. But are these differences statistically significant? Hot spot analysis will answer that question and help you target areas for further study into what drives costs.

1. If necessary, open your Medicare Costs per Capita in 2012 map from My Content.
2. In the Contents pane, point to the Medicare Costs by Hospital Referral Region layer and click the Perform Analysis button.

   

3. In the Perform Analysis pane, click Analyze Patterns and click Find Hot Spots.

   

   The Find Hot Spot pane opens. There are four settings to make. The first setting, Choose layer for which hot spots will be calculated, is already set correctly to the Medicare Costs by Hospital Referral Region layer. The third setting, Divide by (optional), is also set correctly to None.

4. For the second setting, Find clusters of high and low, choose Standardized Risk-Adjusted Per Capita Costs.

   

   This is the cost attribute you want to analyze.

5. For the fourth setting, Result layer name, replace the default name with Medicare Cost Hot Spots and add your name or initials.

   

   Tip:

   Personalizing the result layer name will prevent it from conflicting with somebody else's in the organization.

6. If necessary, zoom out so all the states (not counting Hawaii and Alaska) are visible.

   Tip:

   By default, analysis operations are applied only to the visible part of the map. Unchecking the Use current map extent box at the bottom of the pane is a way to ensure that all your data is analyzed.

7. Click Run Analysis.

   When the operation is finished, the new hot spots layer is added to the map.

   

8. Rename the new layer by removing your name or initials.
9. Turn off the Medicare Costs by Hospital Referral Region layer.

   On the map, red and blue areas represent statistically significant clusters of high and low costs, respectively. In regions symbolized in gray, costs do not stand out as significantly high or low. By and large, the hot spot map probably confirms your earlier impressions—it also sharpens them and focuses your attention on the places that matter most.

10. Save the map.

Change layer symbology

Users of your map may find it more helpful to see the hot and cold spots displayed with state boundaries than with hospital referral regions.

1. In the Contents pane, point to the Medicare Cost Hot Spots layer and click the Change Style button.
2. In the Change Style pane, on the Counts and Amounts (Color) drawing style, click Options.
3. In the Change Style pane, click the Symbols button.

   

4. In the symbol window, click the Outline tab and change Line Width to 0 px (pixels).

   

5. Click OK.

   

   The labels on the layer legend explain the symbols. For example, a hot spot with 99 percent confidence means there is only a 1 percent chance that a cluster of high costs occurred randomly. Where confidence levels are very high, or very low, there is almost certainly an underlying reason (or multiple reasons).

   The legend heading above the symbols is generated by a field name alias in the layer table. You'll change this heading to something meaningful in the next section.

6. At the bottom of the Change Style pane, click OK and click Done.
7. In the Contents pane, turn on the State Boundaries layer.
8. Point to the State Boundaries layer to display three vertical dots to the left of the layer name.

   

   The pointer becomes a four-headed arrow. You can now drag the layer to a new position. As you drag it, its position will be represented by a dotted horizontal line.

9. Drag the State Boundaries layer above the Medicare Cost Hot Spots layer.

   

   Plainly, the Gulf Coast region is a huge Medicare cost hot spot. The hot spot rises through Tennessee, Kentucky, and Indiana into Michigan and western Ohio. There are major cold spots along the West Coast, the upper Midwest, and New England.

Configure pop-ups

Users of your map will probably want to see the costs associated with the various hot and cold spots. You'll make this information available through a pop-up.

1. Click a feature on the map to open a pop-up for the Medicare Cost Hot Spots layer.

   

   The pop-up displays an ID value, the cost attribute you analyzed, and its statistical significance. (You have to scroll to see the values because the cost field name is so long.) The ID isn't needed. You should also alias the name of the cost attribute to make it shorter.

2. Close the pop-up.
3. In the Contents pane, point to the Medicare Cost Hot Spots layer. Click the More Options button and choose Configure Pop-up.
4. In the Configure Pop-up pane, in the Pop-up Title box, delete the existing title and type Medicare Cost Hot Spots.

   

5. In the Pop-up Contents section, click Configure Attributes.
6. In the Configure Attributes window, check the box next to the Display column header to check all fields, then uncheck it to uncheck all fields.

   

7. Check the box next to the {STANDARDIZED_RISK_ADJUSTED_PER_} field, and then click its Field Alias to edit it. Delete the existing alias and type 2012 Average Cost per Person.

   Note:

   Field names have restrictions on length and valid characters. (For example, spaces are not allowed.) An alias lets you describe a field name more informatively.

8. On the right side of the window, change Format to 0 decimal places.
9. Check the box next to the {Gi_Text} field, and then click its Field Alias to edit it. Change the alias to Significance. Press Enter.
10. Click in the Field Alias column for the {Gi_Bin} field and change the alias to Significance. Press Enter.

    

    Although the {Gi_Bin} field is unchecked and does not display in the pop-up, its field alias is used as a heading in the map legend. You saw this earlier in the Change Style pane. Changing the alias here will fix it automatically in the legend.

11. Click OK in the Configure Attributes window. At the bottom of the Configure Pop-up pane, click OK.
12. Click a feature on the map to open its pop-up.

    

13. Click a few other features to see their pop-ups, and then close the open pop-up.
14. At the top of the Contents pane, click the Legend button.

    

    The legend heading reflects the change you made to the field alias.

15. Save the map.

The hot spot region along the Gulf Coast can't be explained by a unique demographic profile or differences in the cost of living. These factors have already been accounted for in the standardized, risk-adjusted cost data that you used. Therefore, the high costs in this area must in some way be associated with the Medicare program itself. But how exactly? Is the quality of care different? Less effective care might raise costs by making patients return more often for treatment. Are drugs prescribed differently—could there be regional patterns in the prescription of name-brand versus cheaper generic drugs? Is facility usage different? Maybe in some parts of the country hospice and home health care are more common alternatives to expensive in-patient hospitalization.

At this point, you don't know. Hot spot analysis tells you where to look, and that's an important first step. The next logical step is to find out what it is about the Gulf Coast region that leads to higher costs. The search for answers leads to regression analysis. Regression analysis is a statistical technique that helps you understand relationships among variables in your data. It involves trying to explain a variable of interest—in this case, cost—using a set of potentially contributing variables. For instance, you might try to understand costs using such variables as hospital readmission rates (as a measure of quality of care), hospice use, and the use of diagnostic imaging (such as CT scans and MRIs). You determine which combination of variables is best based on statistical tests that show how well they predict the variable of interest.

To see what you can learn by applying regression analysis to this Medicare cost data, watch the free Esri training seminar Beyond Where: Using Regression Analysis to Explore Why.

You can find more lessons in the Learn ArcGIS Lesson Gallery.