Initial publication: 2019.12.9
The Office of the Actuary, Centers for Medicare and Medicaid Services (CMS) published their 2018 estimates of US healthcare spending last week. The big news was that healthcare spending as a fraction of the total economy shrunk to 17.7% in 2018, down from 17.9% in 2017. Here are few excerpts from the publication:
| Funding source | 2012 | 2014 | 2016 | 2018 |
|---|---|---|---|---|
| National Health Expenditures per capita | $8,908 | $9,518 | $10,379 | $11,172 |
| Gross Domestic Product per capita | $51,695 | $55,143 | $58,025 | $63,004 |
| National Health Expenditures, billions of inflation-adjusted dollars | $2,791 | $2,937 | $3,182 | $3,354 |
| Gross Domestic Product, billions of inflation-adjusted dollars | $16,197 | $16,912 | $17,689 | $18,638 |
| National Health Expenditure as a percent of Gross Domestic Product | 17.2 | 17.3 | 17.9 | 17.7 |
| Spending category | 2012 | 2014 | 2016 | 2018 |
|---|---|---|---|---|
| National Health Expenditures, billions of 2012 dollars | $2,791 | $2,937 | $3,182 | $3,354 |
| Hospital care, billions of 2012 dollars (percent increase from 2012) | $902.5 | $944.9 (4.7) | $1,031 (14.2) | $1,082.6 (20.0) |
| Physician and clinical services, billions of 2012 dollars (percent increase from 2012) | $557.1 | $592.0 (6.3) | $667.4 (19.8) | $ 719.2 (29.1) |
| Home health care, billions of 2012 dollars (percent increase from 2012) | $78.3 | $84.1 (7.4) | $89.8 (14.7) | $95.5 (22.0) |
| Nursing care facilities and continuing care retirement communities, billions of 2012 dollars (percent increase from 2012) | $147.4 | $149.5 (1.4) | $152.7 (3.6) | $148.8 (0.9) |
| Prescription drugs, billions of 2012 dollars (percent increase from 2012) | $253.0 | $276.0 (9.1) | $296.8 (17.3) | $308.5 (21.9) |
| Durable medical equipment, billions of 2012 dollars by chain-weigthed National Health Expenditure deflator (percent increase from 2012) | $43.7 | $46.2 (5.7) | $50.2 (14.9) | $53.5 (22.4) |
| Net cost of health insurance, billions of 2012 dollars (percent increase from 2012) | $165.2 | $186.9 (13.1) | $194.0 (17.4) | $206.6 (25.1) |
Data from the National Health Expenditure Accounts. The net cost of health insurance is “the difference between calendar year-incurred premiums earned and benefits paid for private health insurance. This includes administrative costs, and in some cases, additions to reserves, rate credits and dividends, premium taxes, and plan profits or losses. Also included in this category is the difference between premiums earned and benefits paid for the private health insurance companies that insure the enrollees of the following programs: Medicare, Medicaid, Children’s Health Insurance Program, and workers’ compensation (health portion only).”
The tables above suggest that although we may be spending a slightly smaller percentage of our Gross Domestic Product on healthcare, we seem to be spending a lot more on the physician & clnical services, net cost of health insurance, home health care and prescription drugs compared to 2012.
Factors accounting for growth in per capita national health expenditures (NHE), selected calendar years 2004–18. SOURCE Centers for Medicare and Medicaid Services, Office of the Actuary, National Health Statistics Group. NOTES Medical price growth, which includes economywide and excess medical-specific price growth (or changes in medical-specific prices in excess of economywide inflation), is calculated using the chain-weighted NHE price deflator. “Residual use and intensity” is calculated by removing the effects of population, age and sex factors, and price growth from the nominal expenditure level.
The figure implicates a rise in medical prices could be contributing as much (if not more) to the rise in healthcare spending compared to medical service utilization independent of increased utilization due to an older population that skews more female. But the numbers beg the question, “How much health did we purchase for the dollars spent?” The Global Health Data Exchange did not have 2018 DALYs. Here are the leading causes of disability-adjusted life years (DALYs) for 2012-2016 (2018 data not available):
| Rank | 2012 | 2014 | 2016 |
|---|---|---|---|
| All causes | 91,556,386 | 94,437,732 | 98,883,549 |
| #1 (DALYs) | Ischemic heart disease (7,379,876.6) | Ischemic heart disease (7,502,420.7) | Ischemic heart disease (7,854,095.4) |
| #2 (DALYs) | Chronic obstructive pulmonary disease (4,241,091.0) | Chronic obstructive pulmonary disease (4,426,265.7) | Drug use disorders (5,291,175.9) |
| #3 (DALYs) | Low back pain (4,097,214.7) | Drug use disorders (4,399,849.9) |
Chronic obstructive pulmonary disease (4,613,469.9) |
| #4 (DALYs) | Drug use disorders (3,812,334.9) | Low back pain (4,195,134.6) | Low back pain (4,293,148.7) |
| #5 (DALYs) | Diabetes mellitus (3,578,524.3) | Diabetes mellitus (3,667,309.1) | Diabetes mellitus (3,814,559.2) |
| #6 (DALYs) | Tracheal, bronchus and lung cancer (3,428,387.4) | Tracheal, bronchus and lung cancer (3,464,731.2) | Tracheal, bronchus, and lung cancer (3,658,700.8) |
| #7 (DALYs) | Stroke (3,132,808.3) | Stroke (3,269,244.0) | Stroke (3,477,077.0) |
| #8 (DALYs) | Headache disorders (2,676,205.2) | Headache disorders (2,712,082.4) | Headache disorders (2,741,922.1) |
| #9 (DALYs) | Depressive disorders (2,451,998.2) | Depressive disorders (2,484,319.5) | Road injuries (2,547,020.0) |
| #10 (DALYs) | Road injuries (2,427,283.6) | Road injuries (2,406,687.9) | Depressive disorders (2,515,493.6) |
There are at least three notable findings when listing these DALYs side-by-side. First, the goal of every health system should be to minimize DALYs. Using this definition, we in America are spending more money for less “health” year after year. Second, the number of DALYs attributed to each condition increased every year. Third, the rank order of diseases contributing to DALYs remained mostly stable with two exceptions:
- Drug use disorders rose in rank every two years (drug use disorders in bold). It may be a few years before drug use disorders overtakes ischemic heart disease, but without a set of effective interventions, it seems inevitable, and
- Road injuries overtook depressive disorders in 2016.
The American healthcare system is designed to address diseases after they manifest rather than address risk factors and root causes. Smoking prevention and avoiding weight gain do not receive nearly the attention that a new cardiac stent or inhaler to treat chronic obstructive pulmonary disease might receive. The long intervals between starting behaviors that increase DALYs and the onset of those DALYs reduces the likelihood that any indvidual or healhcare payer would appreciate the full costs of an unhealthy behavior until after an advanced diagnosis. From a commerical payer’s perspective, the likelihood of reducing spend on members with an advanced diagnsosis for these top 10 conditions within 18-24 months seems low. With longer retention periods, Medicare (traditional and Medicare Advantage) seems to be in the best position to consider interventions to address behaviors with longer payoff times.
The lists do suggest a few interventions that might be useful in the short-term. An intervention (e.g., cardiac rehabilitation) withn even a small reduction in DALYs for ischemic heart disease might be magnified over the eligible population who could receive that intervention. Delivering tools to members with mobile technology for low back pain and headache disorders might be a way for payers to make a meaningful difference in members’ lives without redesigning the entire healthcare system. New approaches to chronic care disease management could address the DALY burden from many conditions on these lists.
Rather than focusing on the total number of dollars spent in a given year or within a specific set of services, it may be more helpful to link services to a set of conditions that could be meaningfully addressed by the healthcare system (e.g., I do not have any cogent recommendations from a traditional healthcare perspective about the rise in drug use disorders). As we increase our confidence in addressing conditions that have a disproportionate influence on DALYs, we can consider addressing additional conditions that affect a person’s functional status. The approach might help Americans feel like dollars spent in healthcare are dollars well-spent.
*This post was updated on 2019.12.10 after I obtained access to the CMS tables with more specific inflation numbers to help adjust each category’s spend over time.