Winning the Hardware Software Game Winning the Hardware-Software Game

Using Game Theory to Optimize the Pace of New Technology Adoption

Innovators of new technology systems requiring users to combine components from several different providers (e.g., hardware, software or content, accessories) often face delays in adoption of their new systems. Users will not buy the hardware until enough software or content is available, while at the same time software or content providers will not provide content until enough users have adopted the new system. Winning the Hardware-Software Game examines the dynamics of this adoption process and provides methods for optimizing the pace of adoption of new technology systems.     Read more...


Underlying Issue

Trends in Total US Healthcare Expenditures

Trends in Personal Healthcare Expenditures

Trends in Healthcare Expenditures by Condition

In Sum



Underlying Issue

The total annual costs of healthcare paid by each individual is the sum of the healthcare premiums he pays and the out-of-pocket costs he incurs:

Total Cost of Healthcare = Insurance Premiums + Out-of-pocket Costs

Roughly speaking, the annual insurance premium an individual pays is the average of the total annual costs paid by his insurance company for the healthcare costs incurred by all individuals in his (age) group. What this means is that if the healthcare costs of one individual rise, then that individual does not bear the full burden of the costs increase, but rather, the burden is shared by all members of the group. This is the very nature of risk-pooling, and it works fine when all the members in the group face the same risks.

Out-of-pocket costs for healthcare depend on the type of coverage an individual has, plus the amount of healthcare individuals use.

Moving on, the amount of healthcare an individual will use/need during the year depends on several factors:

  • Genes: People will end up using more healthcare services to the extent that they have “bad” genes.
  • Luck: People will end up using more healthcare services to the extent that they have bad luck or are otherwise accident prone.
  • Lifestyle: People will end up using more healthcare services to the extent that they have an unhealthy diet, don’t exercise, smoke, don’t take safety precautions (e.g., wear seatbelts), or otherwise lead more risky lifestyles.
  • Compliance: People will end up using more healthcare services to the extent that they don’t comply with their doctors’ recommendations (e.g., take medication, lose weight, stop smoking, etc.)

Obviously, people can’t control whether they have bad genes or bad luck. However, they can control the type of lifestyle they live and whether they comply with their doctors’ recommendations.

This begs the following question: To what extent are healthcare costs attributable to factors that people cannot control (bad genes and bad luck), as opposed to factors that they can control (lifestyle and compliance)?

Most people would probably agree to have society (government) subsidize healthcare costs associated with factors people cannot control. However, to the extent that people choose to not control those factors over which they do have power, then to what extent should society be responsible for subsidizing those people’s higher healthcare costs?

Clearly, the issue becomes more important as the costs of healthcare have increased so dramatically over the years.

Trends in Total US Healthcare Expenditures

I undertook an analysis of historical US expenditures on healthcare to answer the question

To what extent are healthcare costs attributable by factors that people cannot control (bad genes and bad luck) as opposed to factors that they can control (lifestyle and compliance)?

Figure 1 displays total national and per-capita healthcare expenditures for the 1960 through 2010 period. To account for inflation, expenditures have been converted to real 2009 dollars, and to account for population growth, expenditures have also been presented in per-capita terms.

Total national healthcare expenditures are comprised of government spending on public health activities, costs associated with public and private administration of healthcare, expenditures on investment (research, structures, and equipment) in healthcare, and expenditures on personal healthcare.

Figure 1

Averages and growth rates for per-capita healthcare expenditures (the black line in Figure 1) are provided in Figure 2.

Figure 2

During the first half of the 1960s, $1,193 was spent per person annually on personal healthcare , and by the 2010, that number had grown to $8,266. While growth rates in per-capita spending have decreased over time, the absolute levels have increased almost seven times during the period.

Figure 3 shows the distribution of total national expenditures across the subcategories of personal healthcare expenditures, administration costs, investment, and public health activities.

Figure 3

As seen in Figure 3,

  • Public and Private Expenditures for Personal Health is the largest category, ranging from about 84 to 86 percent of total healthcare expenditures during the period
  • Public and Private Investments in Research, Structures, and Equipment has decreased substantially during the period as a percentage of total expenditures from 11 percent in the 1960s to six percent in 2010.
  • Public and Private Administration Costs have grown during the period from four percent of total costs in the 1960s to seven percent during 2010.
  • Public Expenditures for Health Activity have increased during the period from one percent of total costs in the 1960s to three percent during 2010.

So while the portion of total healthcare expenditures that actually go toward personal health have remained roughly constant over the past five decades, expenditures on healthcare investments have been crowded out to a greater extent by administrative costs and to a lesser extent by expenditures on public health activities.


Trends in Personal Healthcare Expenditures

Let’s focus now on personal healthcare expenditures. Figures 4, 5, and 6 decompose personal healthcare expenditures into categories based on venue or type of treatment. Note that the figures still include the non-personal-care-expenditures categories of total healthcare spending -- investment, administrative, and public health activities -- to keep the subcategories being discussed in proportion to total healthcare spending.

Figure 4

As seen in Figure 4, the largest categories of personal healthcare spending are expenditures on hospital inpatients; expenditures in outpatient settings, that is, in clinics or physicians’ offices; expenditures on prescription medications, and expenditures in nursing care facilities and continuing care retirement communities.

The Other Personal Healthcare Expenditures category includes in decreasing order of 2010 expenditures:

  • Other Health, Residential, and Personal Care;
  • Dental Services;
  • Home Health Care Services;
  • Other Professional Services;
  • Non-Durable Medical Products; and
  • Durable Medical Equipment.

Figures 5 and 6 present the distribution of personal healthcare expenditures over the period.

Figure 5

Figure 6

Figures 5 illustrates that

  • Total Personal Healthcare Expenditures, which include spending in/for hospital services, physicians’ (OP) services, prescription medication (Rx), and other personal services, together accounted for 84 to 85 percent of total national healthcare expenditures during the period (this is what we already saw in Table 3, but I’m repeating here for perspective).
  • Expenditures for the Four Largest Categories of Personal Healthcare Expenditures, Hospital Services, Physician Services, Prescription Mediation, and Nursing Care, ranged from about 66 to 69 percent of total healthcare expenditures during the period. The top four categories combined reached a peak portion of total expenditures during the late 1980s, decreased slightly through the 1990s, and has remained roughly constant at a slightly lower level (than its peak) since then.

Figures 4 through 6 indicate that there seems to have been a shift in the trend of healthcare spending between the 1980 – 1984 period and the 1985 – 1989 period. More specifically,

  • Shift in Annual Growth Rate of Expenditures: Figure 4 shows that the trend of increasing annual spending on healthcare expenditures seems to have increased between the 1960 – 1984 period and the 1985 – 2009 period.
  • Shift in Composition of Expenditures: Figures 5 and 6 show that the growth rate in annual expenditures is concentrated in expenditures in doctors offices and on prescriptions, while relative spending in hospitals decreased.

These patterns in expenditures are consistent with a relative shift in expenditures away from acute health problems towards expenditures to manage chronic problems.


Trends in Healthcare Expenditures by Condition

In 1996, the US Department of Health and Human Services (DHHS) began its MEPS Survey, which collects survey information on healthcare expenditures for specific medical conditions.

The Medical Expenditure Panel Survey, which began in 1996, is a set of large-scale surveys of families and individuals, their medical providers (doctors, hospitals, pharmacies, etc.), and employers across the United States. MEPS collects data on the specific health services that Americans use, how frequently they use them, the cost of these services, and how they are paid for, as well as data on the cost, scope, and breadth of health insurance held by and available to U.S. workers.

The MEPS survey reports expenditures for different medical conditions separately, depending on whether the expenditures were incurred for hospital inpatients (HIP) or in emergency rooms (ER), as outpatient services in physicians offices (OP), or for prescription medication (Rx).

I compared the MEPS Survey data for these three venues, HIP, OP, and Rx, with the national expenditure data to see how completely the MEPS Survey is capturing total national expenditures on healthcare services.

Figure 7

Figure 7 shows that the Survey data are relatively complete for prescription data, but that the surveys fail to capture about one quarter of healthcare expenditures in physician offices and almost half of expenditures in hospitals. Nevertheless, the portion of data captured in the MEPS Surveys is large enough so as to give us a good indication of what’s going on.

The MEPS Surveys include healthcare expenditures on patients that fall into 56 different medical conditions. However, as seen in Figure 8, the top 20 conditions account three-fourths of all MEPS expenditures.

Figure 8

Figures 9 and display the total historical expenditures captured by the MEPS Surveys for the top 20 medical conditions. The numbers in parentheses in the figure legends indicates the 1996 – 2010 growth rate in total expenditures for each condition. Of the top 20 conditions,

  • Largest Expenditures: The largest expenditures were for treatment of patients with heart conditions, which grew 33 percent during the period, from $79 billion in 1996 to $105 billion in 2010.
  • Fastest Rate of Growth: The fastest rate of growth by far occurred in expenditures on patients being treated for hyperlipidemia (abnormally high concentration of fats or lipids in the blood), which grew almost six times during the period from $5 billion in 1996 to $37 billion in 2010. 

Figure 9

Figure 10


As an aside, according to F Xavier Pi-Sunyer, "Health Implications of Obesity," the following diseases are associated with overweight and obese individuals:

  • Hypertension
  • Hypercholesterolemia
  • Hyperlipidemia
  • Impaired Glucose Tolerance (Diabetes)
  • Cardiovascular Disease
  • Gallbladder Disease
  • Arthritis
  • Gout
  • Pulmonary Function
  • Cancer

Other sources (e.g., the American Diabetes Association) note that people with diabetes tend to have higher rates of depression.  Taken together, this means that at least 7 of the top 10 conditions may be associated with being overweight/obese. (This issue will be discussed further in another blog post.)

The increases in expenditures over the period for each medical condition may be decomposed into two factors: (i) increases in expenditures due to increases in the number of patients being treated for the condition, and (ii) increases in expenditures due to increases in the costs of treating the condition.

Figures 11 and 12 show the numbers of patients being treated for each condition during the period, with growth rates in number of patients during the period shown in parentheses. The figures show substantial increases in the numbers of patients being treated in 18 of the 20 conditions, with the largest growth by far in the number of patients being treated for hyperlipidemia.

Figure 11

Figure 12

Figures 13 and 14 show the expenditures per patient for patients being treated for each condition during the period, with growth rates in expenditures per patient during the period shown in parentheses. Figures 13 and 14 indicate that the growth in expenditures per patient for most of the top 20 conditions was relatively modest, especially compared with the growth in number of patients being treated for each condition.

Figure 13


Figure 14


In Sum

The information in Figures 11, 12, 13, and 14 are summarized in Figure 15.

Figure 15


In other words, the large increases in expenditures for the top 20 medical conditions is mostly explained by increases, often substantially, in the numbers of patients being treated for the conditions, rather than increases in the costs of treating each patient with the conditions.

Also, most of the increases in healthcare expenditures are going for patients who have healthcare problems that are at least somewhat under their control (lifestyle and compliance issues), as opposed to healthcare issues not under their direct control (genes or accidents).

To the extent that relatively healthy people are forced to subsidize the costs of high-intensity healthcare users, the healthy people appear to be paying a lot of money for healthcare conditions suffered by people receiving the subsidies that are largely prevantable, if the recipients were only to take the appropriate measures.

My next blog post examines the distribution  of healthcare dollars across the population: Why Are Healthcare Costs So High? - Part 2

More Blogs

Playing the Restaurant Delivery Game

07-11-2017 - Hits:314 - Ruth Fisher - avatar Ruth Fisher

Remember when restaurant delivery options included Domino’s, Pizza Hut, or Papa John’s? No longer! These pizza restaurants still have a majority share of the restaurant delivery market. However, GrubHub, Eat24, DoorDash, UberEats, and other restaurant delivery platforms are expanding users’ choices beyond just pizza. This analysis examines the Restaurant Delivery Game:...

Read more

Playing the Platform Game

08-10-2017 - Hits:491 - Ruth Fisher - avatar Ruth Fisher

The purpose of this analysis is to better understand the dynamics of internet platforms. The analysis considers the three basic types of platforms: Vendors (WalMart, Apple, Pandora, etc.) Social Media (Facebook, LinkedIn, YouTube, etc.) Matchmakers (eBay, Uber, etc.) And will seek to address such issues as Who are the different players in each type of...

Read more

Overview of Technology System Dynamics

25-09-2017 - Hits:505 - Ruth Fisher - avatar Ruth Fisher

This is a presentation I’m preparing for “Tech Startup Conference: Artificial Intelligence” being held on September 26, 2017. 1. Issues Covered Adoption of New Technology Systems: What does it take for new technologies to become successfully adopted in the marketplace? Why do some technologies become adopted while others do not? Value Creation: How...

Read more

Persuasion Technologies in the Digital Age: The Good, The Bad, and the Ugly

05-09-2017 - Hits:860 - Ruth Fisher - avatar Ruth Fisher

Persuasion technologies include methods and techniques derived from behavioral psychology and behavioral economics used to shape the choices people make. The favorable environment for using such methods, enabled by people’s increasing use of computers and smartphones, has led to the proliferation of their use by software developers. Like any technology, persuasion...

Read more

How Have Retail Sales Changed Over the Last 25 Years?

18-08-2017 - Hits:284 - Ruth Fisher - avatar Ruth Fisher

This brief analysis examines how retail sales have changed over the past 25 years, both in size and distribution. As seen in Figure 1, total retail sales have grown by 2.7% - 2.8% per year, excluding the short drop in spending during the Financial Crisis (2007 – 2009). Figure 1  

Read more

Playing the Amazon Monopoly Game

28-07-2017 - Hits:501 - Ruth Fisher - avatar Ruth Fisher

Amazon Ecosystem Components Amazon Segments Amazon Sales, Income, and Margins Amazon Methods of Competition Anti-Trust Issues   Amazon is leaving a large pile of battered companies in its wake as it increasingly steals sales away from traditional bricks and mortar companies and decimates their market shares. Some of the better-known victims include: Barnes & Noble (books)...

Read more

Do We Shape Technology, or Does Technology Shape Us?

25-07-2017 - Hits:493 - Ruth Fisher - avatar Ruth Fisher

I have always thought about technology as a tool that we develop and improve and customize and adapt to use to help us do whatever it is we want to do faster and easier and better. In other words, we shape and control technology. However, recently I’ve come to realize that...

Read more

Considerations for a Single-Payer Healthcare System in the US

06-07-2017 - Hits:451 - Ruth Fisher - avatar Ruth Fisher

This analysis considers how the US healthcare system would change if we were to transition from the current multi-payer system to a single-payer system. The analysis first presents facts that will be important in considering what a single-payer system might look like. Then, given these facts, the analysis considers specific...

Read more