Those who follow Altarum Institute’s monthly health sector briefs and trend reports are well aware that the five-year run of record low growth rates in national health spending (from 2009 through 2013) has come to an end, or at least been interrupted. According to data just released by the Centers for Medicare & Medicaid Services (CMS), health spending grew by 5.3 percent in 2014, compared to 2.9 percent in 2013 and roughly 4 percent from 2009 through 2012. Our estimates for the first eight months of 2015 show growth of 6.2 percent, though on a downward path, indicating that the year could finish at around 6 percent growth.
Many analysts had predicted well in advance an increased health spending growth rate for 2014 due to the implementation of the Patient Protection and Affordable Care Act (ACA). The expanded coverage provisions of the ACA were expected to increase the number of persons with health insurance and hence health spending (Note 1).
However, the huge jump in prescription drug spending that occurred in 2014 was unforeseen. In 2013, spending on prescription drugs grew by only 2.4 percent, but in 2014, it skyrocketed to 12.2 percent. Some of this increase in spending on prescription drugs is due to expanded coverage, and some is due to the acceleration in prescription drug prices.
A major source of this growth was the introduction of Sovaldi in December 2013, and Harvoni in October 2014. Both are very expensive breakthrough drugs for the treatment of hepatitis C (a disease of the liver, affecting approximately 3 million Americans). The new hepatitis C drugs have had a significant impact on growth rates in spending on prescription drugs and on national health spending.
Company Sales Of New Hepatitis C Drugs
I use company reports to track quarterly U.S. sales of the four bestselling new prescription drugs for the treatment of hepatitis C: Harvoni and Sovaldi (Gilead Sciences, Inc.), Viekira Pak (AbbVie), and Olysio (Janssen Pharmaceutical Companies of Johnson & Johnson). While these sales figures are not identical to the final sales captured in health spending, they provide a basic sense of the impact of hepatitis C drugs on overall health spending growth rates (Note 2).
The quarterly sales are shown in Figure 1. Combined sales were $100 million in Q4 2013, jumping to $2.4 billion and $3.8 billion in Q1 and Q2 2014. In Q3, sales of the existing drugs fell to $2.9 billion, presumably due to the expected release of a more appealing drug, Harvoni, in October. Starting in Q4 2014, sales moved back to $3.5 billion and have stayed in that range during the first three quarters of 2015.
Figure 1: Quarterly Company Sales Of New Drugs To Treat Hepatitis C
Source: Altarum Center for Sustainable Health Spending, computed from Gilead, AbbVie, and Johnson & Johnson Company reports.
Impact On 2014 Growth In Health Spending
National health spending in 2013 was about $2.9 trillion. In 2014, company sales of the new hepatitis C drugs were $12.3 billion higher than in 2013. This represents 0.4 percent of total health expenditures in 2013, and suggests that the new hepatitis C drugs added 0.4 percentage points to the 2014 health spending growth rate. In the absence of these new drugs, national health spending would have grown by about 4.9 percent rather than 5.3 percent.
Prescription drug spending in 2013 was $265 billion. Similar math suggests that the new hepatitis C drugs added about 4.6 percentage points to the growth in prescription drug spending in 2014. The total growth rate in prescription drug spending in 2014 was 12.2 percent, so we estimate that it would have been 7.6 percent in the absence of the new hepatitis C drugs. That latter would still have been a large increase relative to the 2.6 percent growth seen in 2013 for prescription drugs and is likely due to a combination of expanded coverage and accelerated price growth.
Likely Impact On 2015 Growth In Health Spending
While spending on hepatitis C drugs in 2015 remains very high ($14.4 billion is a reasonable estimate, assuming $3.5 billion in Q4), it represents a much smaller increase ($1.9 billion) than experienced in 2014 ($12.3 billion). This means that in 2015, hepatitis C drugs will have added less than 0.1 percentage points to the growth rate in national health expenditures and 0.6 percentage points to the growth rate in prescription drug spending.
To put it another way, the relatively small increase in hepatitis C spending in 2015 should reduce the overall rate of growth in national health expenditures by more than 0.3 percentage points compared to the growth rate in 2014. For prescription drugs, it should bring the growth rate down by about 4 percentage points (Note 3).
Our November health sector spending brief shows that the growth rate in spending on prescription drugs has indeed been falling in 2015 and, as of September, was at 8.4 percent, year over year. This is 3.8 percentage points below the growth rate in 2014, and we attribute much of the decline to the leveling off in spending on hepatitis C (Note 4).
The decline in the prescription drug spending growth rate in 2015 has been more than offset by a higher rate of growth in spending on health care services so that, in September, the health spending growth rate of 5.5 percent was actually higher than the rate for 2014. In the absence of the prescription drug spending slowdown, the September growth rate would have been about 5.9 percent.
Implications For 2016 And Beyond
I conclude with speculation about the future using rough estimates that I believe nonetheless provide some useful information. U.S. sales of hepatitis C drugs seem to have stabilized at about $14 billion per year, which equates to about 150,000 persons treated per year (at a rate of $90,000 per person treated).
The number of new cases per year is about 30,000 so our current rate of spending should be sufficient to reduce prevalence by about 120,000 per year. There are about 3 million people with hepatitis C in the U.S., so our current rate of spending could continue for many years before we run out of cases to treat. (Because hepatitis C is spread by those already infected, the number of new cases should dwindle toward zero as the overall prevalence falls toward zero.)
Of course, the amount we actually spend per year over the next few years could be higher or lower than what we spent in 2015. Recent recommendations argue for treatment of nearly all diagnosed cases, yet many states and other payers continue to prioritize candidates for treatment based on severity of illness and other factors. This means that the near-term rate of spending is quite unpredictable and depends upon how much effort is put toward identifying and treating existing cases, how much resistance there is to approving treatment, and what competition might do to reduce drug prices. In the longer term, the curative powers of these drugs should drive hepatitis C prevalence, and therefore spending, to very low levels.
Estimates from the American Community Survey show that the percentage of the population with health insurance grew by 2.8 percentage points between 2013 and 2014.
One difference is the lag between when the sale (to an intermediary) is recorded by the company and when the final sale is made to the consumer. Another is that retail markups are included in health spending but not in company sales. Finally, prescription drug sales in the national health accounts refer only to retail sales, while company sales include sales to hospitals and physicians for drugs administered during the health encounter and whose costs are embedded in the hospital or physician bill. The vast majority of hepatitis C drugs are sold at retail.
For further analysis of prescription drug spending, see: R. Kamal and G. Claxton, October 30, 2015, “Recent trends in prescription drug spending and what to look out for in coming years,” Peterson-Kaiser Health System Tracker.
The leveling off we observe in company sales in 2015 runs counter to a recent report stating that Medicare spending on these drugs grew by $4.5 billion in 2015. Some of this may be due to lags between company sales to intermediaries and the subsequent sales to Medicare patients. We have not built lags into our analysis, and this could mean we have somewhat overstated the 2014 impact and understated the 2015 impact.