By David Lassman, Micah Hartman, Benjamin Washington, Kimberly Andrews, and Aaron Catlin

By David Lassman, Micah Hartman, Benjamin Washington, Kimberly Andrews, and Aaron Catlin US Health Spending Trends By Age And Gender: Selected Years 2002 10 doi: 10.1377/hlthaff.2013.1224 HEALTH AFFAIRS 33, NO. 5 (2014): 815 822 2014 Project HOPE The People-to-People Health Foundation, Inc. ABSTRACT This article presents estimates of personal health care spending by age and gender in selected years during the period 2002 10 and an analysis of the variation in spending among children, working-age adults, and the elderly. Our research found that in this period, aggregate spending on children s health care increased at the slowest rate. However, per capita spending for children grew more rapidly than that for working-age adults and the elderly. Per capita spending for the elderly remained about five times higher than spending for children. Overall, females spent more per capita than males, but the gap had decreased by 2010. The implementation of Medicare Part D, the effects of the recent recession, and the aging of the baby boomers affected the spending trends and distributions during the period of this study. David Lassman (David.Lassman2@cms.hhs.gov) is a statistician in the National Health Statistics Group, Office of the Actuary, Centers for Medicare and Medicaid Services (CMS), in Baltimore, Maryland. Micah Hartman is a statistician in the National Health Statistics Group, CMS Office of the Actuary. Benjamin Washington is an economist in the National Health Statistics Group, CMS Office of the Actuary. This article provides an analysis and discussion of personal health care spending in the United States for selected years during the period 2002 10 by age group and gender. Our study shows that the variations in this spending by age and gender reflect numerous factors, including the changing age mix of the population, the combination of goods and services consumed, and the incidence of disease. Personal health care consists of all the medical goods and services used to treat or prevent a specific disease or condition in a specific person. Total personal health care spending in the United States reached $2.2 trillion, or $7,097, per person in 2010 (Exhibit 1). Females, who make up just over 50 percent of the US population, 1 accounted for 56 percent of the spending in 2010, or $7,860 per person. That is about 25 percent more than per capita spending for males on health care. The difference is down from 29 percent in 2004, when the disparity in spending between the sexes peaked. In general, our study found that average health care spending increased with age, which is consistent with numerous published studies. 2 In 2010 the elderly people ages sixty-five and older spent $18,424 per person on personal health care. Thus, the average elderly person spent about three times more than the average working-age adult and about five times more than the average child. Similar to the annual historical National Health Expenditure Accounts (NHEA), 3 spending estimates here are provided by service and major payer. Information about the NHEA sources, methods, and definitions can be found on the website of the Centers for Medicare and Medicaid Services (CMS). 4 National health spending estimates, disaggregated by age and gender, provide an important view of the distribution of health care spending over time as the age and gender mixture of the population changes and public policy evolves. This article highlights several factors that affected the spending trends and distributions during the time period of this study, including the implementation of the Medicare prescription drug benefit (Part D), the recession that lasted from December 2007 to June 2009, and the aging of the baby boomers to near-eligibility for Medicare. Kimberly Andrews is a statistician in the National Health Statistics Group, CMS Office of the Actuary. Aaron Catlin is a deputy director of the National Health Statistics Group, CMS Office of the Actuary. May 2014 33:5 Health Affairs 815

Exhibit 1 US Personal Health Care Spending, By Gender And Age, 2002 10 Total spending (billions of dollars) Average annual growth (%) Per capita spending (dollars) Average annual growth (%) Age group (years) 2002 2010 2002 10 2002 2010 2002 10 Total 1,372 2,193 6.0 4,768 7,097 5.1 0 18 183 285 5.7 2,369 3,628 5.5 19 64 718 1,164 6.2 4,098 6,125 5.2 19 44 336 478 4.5 3,104 4,422 4.5 45 64 382 686 7.6 5,707 8,370 4.9 65+ 471 744 5.9 13,345 18,424 4.1 65 84 362 553 5.4 11,692 15,857 3.9 85+ 109 190 7.2 25,192 34,783 4.1 Females 780 1,231 5.9 5,343 7,860 4.9 0 18 87 137 5.9 2,308 3,572 5.6 19 64 408 657 6.1 4,654 6,892 5.0 19 44 208 299 4.6 3,887 5,579 4.6 45 64 200 358 7.6 5,859 8,577 4.9 65+ 285 437 5.5 13,831 19,110 4.1 65 84 205 303 5.0 11,673 15,805 3.9 85+ 80 134 6.6 26,237 36,296 4.1 Males 592 962 6.3 4,177 6,313 5.3 0 18 96 148 5.6 2,427 3,680 5.3 19 64 310 507 6.3 3,540 5,353 5.3 19 44 128 179 4.3 2,338 3,283 4.3 45 64 182 328 7.7 5,549 8,154 4.9 65+ 186 307 6.4 12,665 17,530 4.1 65 84 158 251 6.0 11,716 15,920 3.9 85+ 29 57 8.8 22,691 31,670 4.3 SOURCE Centers for Medicare and Medicaid Services, Office of the Actuary, National Health Statistics Group. NOTE Numbers may not sum to totals because of rounding. Study Data And Methods Data The CMS Office of the Actuary provides periodic updates of health spending by age 5 and, more recently, by gender. These data were last released in early 2011. 6 The updated data presented in this article represent the even years in the period 2002 10 7 and form the basis for a new biennial series of data on health spending by age and gender. We revised our previous estimates for 2002 and 2004 and created new estimates for 2006, 2008, and 2010. Therefore, previously released estimates are not directly comparable with these estimates. 8 In this study we divided personal health care spending by gender and five age groups: children (ages 0 18), younger working-age adults (ages 19 44), older working-age adults (ages 45 64), and two elderly groups (ages 65 84 and ages 85 and older).we present data for personal health care spending only, because detailed data on gender and age for non-personal health care spending are not available. 9 Methods Our estimates of health care spending by age and gender for the Medicare and Medicaid programs were developed using administrative data from Medicare s National Claims History Files, Prescription Drug Event Files, and the Medicaid Analytical Extract System. The estimates were adjusted to equal the total spending levels in the historical NHEA. 10 We used one of two general methods to develop estimates of health spending by age and gender for all other payers. The first method used spending per use data from the Medical Expenditure Panel Survey (MEPS) Household Component, multiplied by utilization counts from provider surveys conducted by the National Center for Health Statistics. MEPS does not include spending for the institutionalized population. However, the provider surveys from the National Center for Health Statistics do record utilization for this population. We created separate estimates for care provided in an institutional setting (such as a nursing home) based on population counts from numerous sources and per person spending data from the Medicare Current Beneficiary Survey and MEPS. The MEPS data were adjusted for the population residing in institutional settings. Because provider surveys were not available for all services in the NHEA, we used a second method to develop estimates for these categories: We calculated spending by gender and age using MEPS data for the noninstitutionalized population and adjusted those estimates to include spending by the institutionalized population using data from the Medicare Current Beneficiary Survey. Similar to our estimates for Medicare and Medicaid, estimates from both methods were adjusted to equal the total spending levels in the historical NHEA. When necessary, we averaged our estimates over several years to yield a more stable trend over time. In this study we provide descriptive information to explain the more notable findings. However, we have not constructed a model to derive empirical estimates of these explanatory factors. Furthermore, our study was limited to selected historical years and is lagged based on the availability of source data. A more detailed description of the methods and sources can be found on the CMS website. 11 Study Results Aggregate Findings SPENDING BY AGE: In the period 2002 10, average annual growth in per person spending among the major age groupings varied by 1.4 percentage points (Exhibit 1): Spending for people ages sixty-five and older increased at the slowest rate (4.1 percent annually), while spending for children grew at the fastest rate (5.5 per- 816 Health Affairs May 2014 33:5

cent). Per person spending for individuals ages 19 64 increased 5.2 percent annually. A recent study using Health Care Cost Institute data supports the idea that spending for children s health care increased more rapidly than that for all other age groups. The study found that in the period 2002 10, costs covered by private health insurance for hospitals, professional services, and prescription drugs rose more rapidly for people ages 8 20 than for any other age group. 12 This can be attributed to increased insurance coverage for children, since enrollment increases in Medicaid and the Children s Health Insurance Program offset losses of private insurance during the recession. Slower growth in the spending for the population ages sixty-five and older was partly due to sluggish per person growth in spending on nursing care facilities and continuing care retirement communities a 3.3 percent average increase in the study period. This relatively slow growth was mainly the result of states efforts through the Medicaid program to keep the elderly out of costly institutional care settings by using lower-cost home and personal care services. 13 Modest growth in per person spending for hospital care for the elderly also contributed to slower overall growth for this age group increasing 3.7 percent on average in 2002 10 and was primarily due to slower growth in Medicare Advantage payments. 14 Per person spending for home health care for the elderly increased 7.6 percent annually in the period 2002 10, offsetting some of the slower growth in spending from nursing care facilities and continuing care retirement communities and hospitals. The relatively high growth in spending on home and community-based care in the latter years of the study period contributed to these trends. SPENDING BY GENDER: Growth in spending was faster for males than for females in the study period (Exhibit 1). This was the case across most payers and services, but the difference was most dramatic for prescription drugs (data not shown). In 2010 per capita prescription drug spending for females was $919 25 percent more than the spending for males ($734). In 2002 females spent 38 percent more than males. The change may be because during this period, males closed the gap in the number of prescriptions filled. One study found that females filled an average of 4.6 more prescriptions per capita than males in 2011, compared to 5.1 more in 2003. 15 Additionally, during the study period large increases in prescription drug spending were reported for drugs used to treat HIV, disorders of lipoid metabolism, hypertension, and hyperactivity. 16 All of these drugs have higher rates of use in males than in females. Prescription drug spending for males ages sixty-five and older also increased more rapidly, on average, than for females in that age group. This was partly due to faster increases in life expectancy for males 17 and more rapid growth in drug spending for heart disease, prostate care, and leukemia. 16 Part D, Recession, And Baby Boomers In the period 2002 10 three factors had major impacts on the age and gender distributions of health care services and payers. These factors were the implementation of Medicare Part D prescription drug plans, the recession in 2007 09, and the aging of the baby boomers. PART D: Medicare Part D drug expenditures represented relatively minor transitional assistance benefits in 2004 and 2005. However, the retail prescription drug market experienced major changes in 2006 as full implementation of the Medicare Part D drug benefit changed the mix of payers. Spending for Medicare prescription drugs increased from $3.4 billion in 2004 to $39.6 billion in 2006. Before the implementation of Part D, Medicare paid for some drugs. However, this was a small percentage of overall drug spending for people ages sixty-five and older and accounted for only 5 percent of retail prescription drug spending for that age group in 2004 (Exhibit 2). Following full implementation of Medicare Part D, that share jumped to 45 percent in 2006 and 53 percent in 2010. At the same time, Medicaid, out-of-pocket spending, and private health in- Exhibit 2 Shares Of US Retail Prescription Drug Spending For Adults, By Payer, 2004, 2006, And 2010 Percent 100 80 60 40 20 0 2004 2006 2010 Ages 65 and older 2004 2006 2010 Ages 19 64 Other payers and programs Out of pocket Private health insurance Medicaid Medicare SOURCE Centers for Medicare and Medicaid Services, Office of the Actuary, National Health Statistics Group. May 2014 33:5 Health Affairs 817

surance together declined as a share of spending on prescription drugs for the elderly from 90 percent in 2004 to 43 percent in 2010. The share of retail prescription drug spending paid by Medicare for the working-age population also increased, from 1 percent in 2004 to 12 percent in 2010 (Exhibit 2). This change was because drug coverage for working-age disabled beneficiaries who were dually eligible for Medicaid and Medicare was transferred from Medicaid to Medicare. Medicaid s share of drug spending for the working-age population fell from 16 percent in 2004 to about 8 percent in 2010. RECESSION AND MODEST RECOVERY: The recession had a powerful impact on most sectors of the economy, including health care. Its effects were observed across all age groups and for both males and females. Per capita spending on personal health care increased, on average, 5.7 percent in the period 2002 08, compared with an average increase of 3.4 percent in 2008 10. Spending growth in each age group slowed by at least 2 percentage points. Differences in the rate of health spending growth were also observed during and just after the recession for both genders. Spending decelerated by 2.6 percentage points for females and by 1.9 percentage points for males, when average annual spending trends in the period 2002 08 are compared with those in 2008 10. In 2008 10 the largest difference in average spending growth between males and females was for the working-age group. In this period per capita spending growth in this group was 4.0 percent for males but 2.6 percent for females. A 3.7 percent decline in the birth rate in the period 18 may be one of the causes: Growth in spending for females ages 19 44 slowed as they spent relatively less on maternity care. According to a Pew Research Center survey, one in five females ages 18 34 said that they postponed having a baby because of economic conditions. 19 The recession also had a dramatic impact on the payers of health care. With major job losses and high unemployment, many people became eligible for Medicaid, and enrollment increased by 7.5 million from 2007 to 2010. Medicaid spending increased 7.3 percent on average for the period 2008 10. However, per enrollee annual growth was only 1.2 percent because most of the new enrollees were children and nondisabled adults, whose spending was relatively low. During this period, Medicaid spending increased 8.0 percent on average for people younger than age sixty-five and 5.2 percent for the elderly. Eleven million fewer people had private health insurance in 2010 than in 2007. 20 Enrollment in such insurance for people younger than age fiftyfive declined, while enrollment for those ages fifty-five and older increased modestly. Spending growth from private health insurance for the working-age population slowed, from an annual average of 6.9 percent in the period 2002 08 to 3.8 percent in 2008 10. The impact of the recession on the elderly is less clear. However, per capita spending growth in 2008 10 was just 2.4 percent annually, which was lower than growth for the other age groups. In this period private health insurance spending per enrollee for those ages sixty-five and older grew slowly, at 3.0 percent annually the slowest growth rate among the major age groups. Out-ofpocket spending per person for the elderly declined 0.4 percent annually. Medicare spending growth was also slow in 2008 10, as discussed above. This was largely because of spending trends for hospital and overall nursing care facilities and continuing care retirement communities. Slightly offsetting this slower growth was growth in Medicaid spending: Per enrollee spending for the elderly increased 2.3 percent annually the fastest growth rate for any of the age groups. AGING OF THE BABY BOOMERS: Baby boomers, defined as people born in the period 1946 64, accounted for about one-quarter of the US population in 2010. 21 Simply because of its size, this population has a large impact on spending trends. In 2010 all of the baby boomers were included in the category of older working-age adults (people ages 45 64). Not surprisingly, as the baby boomers aged, expenditures for this category increased. The annual growth in spending for them was 7.6 percent, on average (7.6 percent for females and 7.7 percent for males), during 2002 10, the fastest growth rate for any of the age groups (Exhibit 1). The only age-gender category with faster spending growth was males ages eighty-five and older, whose spending increased 8.8 percent in part because of longer male life expectancy, as discussed above. In the period 2008 10 spending increased 5.0 percent, on average, for older working-age adults (data not shown). This was 1.6 times as fast as the spending growth rate for younger working-age adults. Children, Working-Age Adults, And The Elderly CHILDREN: Health spending on children amounted to $285 billion in 2010, or $3,628 per child (Exhibit 1). Per capita spending for males was $3,680, slightly more than the $3,572 spent on females. Children s share of overall personal health care spending was 13 percent in 2010 (Exhibit 3), virtually unchanged from 2002. However, their share of the US population fell from 27 percent to 25 percent an all-time low. 22 818 Health Affairs May 2014 33:5

In 2010 the Centers for Disease Control and Prevention reported that 4.0 million births were registered in the United States 3 percent less than in 2009 and the lowest number of births since 1999. The birth rate per 1,000 population was 13.0 in 2010, down from a recent peak of 14.3 in 2007 and compared with 14.0 in 2002 and 14.2 in 1999. 18 Fertility rates tend to fall when overall economic conditions deteriorate, as they did during the Great Depression and in the recent recession. 23 In the period 2000 10 the number of children grew by 1.9 million. This compares with an increase of 8.7 million children in 1990 2000; an increase of 17.5 million in 1950 60, the majority of the baby boom; and a decline of 2.6 million in the period 1930 40, due largely to the Great Depression. 22 Children s health care is primarily financed through private health insurance and Medicaid (Exhibit 4). Together these sources paid for 72 percent of children s health care in 2010, compared with 71 percent in 2002. However, because of the impact of the recent recession and lingering high unemployment rates, the composition of these payers changed. Medicaid s share of children s spending increased from 31 percent in 2008 to 34 percent in 2010, and the share of private health insurance decreased from 41 percent to 39 percent. The out-of-pocket share of children s health spending decreased from 12 percent to 11 percent, which we also attribute to the recession. Hospital care accounted for the largest share of children s spending in 2010 (Exhibit 5): At 42 percent, it was up from around 38 percent in 2002. Physician and clinical services accounted for the second largest share in 2010, at 27 percent down from 30 percent in 2002. This follows the national trend in the study period, during which average annual growth in spending for hospital services (6.6 percent) increased more rapidly than that for physician and clinical services (5.4 percent). Dental care was the third-largest share of spending for children in 2010, at just over 10 percent. This share was down almost 1 percentage point from 2008, most likely as a result of recessionary pressures. Income is highly positively correlated with dental visits: One study showed that in 2004, 62 percent of children from highincome families saw the dentist at least once, compared to 34 percent of children from lowincome families. 24 WORKING-AGE ADULTS: Working-age adults accounted for 62 percent of the US population in 2010 but only 53 percent ($1.2 trillion) of personal health care spending (Exhibit 3). In comparison, the group s shares of population and Exhibit 3 Shares Of The US Population And Personal Health Care Spending, By Gender And Age Group, 2010 Percent 100 80 60 40 20 0 Population Spending Total Population Spending Males spending in 2002 were 61 percent and 52 percent, respectively. Per capita spending for working-age people in 2010 averaged $6,125 (Exhibit 1). Per capita spending for females in this age group was $6,892 in 2010, which was 29 percent higher than the $5,353 for males. This difference was down from a recent high of 34 percent in 2004. The difference in per capita spending between males and females ages 19 44 was much greater than that for people ages 45 64, largely due to the high costs associated with maternity care. In the younger working-age group in 2010, females spent 70 percent more than males, compared to just 5 percent more in the older working-age group. Not surprisingly, the consumption of personal health care goods and services by the workingage population is financed largely by private health insurance (Exhibit 4) 47 percent of total spending in 2010. This share stayed relatively constant in the period 2002 10. Medicare s share increased from 5 percent in 2002 to 7 percent in 2010 because of the shift of spending from Medicaid to Medicare Part D for prescription drugs for dually eligible disabled adults. Medicaid s share was 16 percent in both 2002 and 2010. However, it decreased a percentage point from 2004 to 2006 (the first year of Part D) and then increased a percentage point from 2008 to 2010 as a result of the recession. The recession also contributed to the decreased share of out-of-pocket spending from 17 percent in 2002 to 15 percent in 2010. Population Spending Females 85+ 65 84 45 64 19 44 0 18 SOURCES Centers for Medicare and Medicaid Services, Office of the Actuary, National Health Statistics Group; and US Department of Commerce, US Census Bureau. May 2014 33:5 Health Affairs 819

Exhibit 4 US Personal Health Care Spending, By Gender, Age, And Payer, 2010 Payer/age group (years) Total spending (billions of dollars) Per enrollee/per capita spending (dollars) Total Females Males Total Females Males All personal health care 2,193 1,231 962 7,097 7,860 6,313 0 18 285 137 148 3,628 3,572 3,680 19 64 1,164 657 507 6,125 6,892 5,353 65+ 744 437 307 18,424 19,110 17,530 Out of pocket 306 178 128 989 1,136 838 0 18 32 16 16 409 409 409 19 64 175 100 75 920 1,048 792 65+ 99 62 36 2,440 2,721 2,073 Private health insurance 753 425 328 4,044 4,454 3,613 0 18 110 55 55 2,458 2,506 2,412 19 64 544 317 227 4,554 5,164 3,909 65+ 99 53 46 4,506 4,382 4,660 Medicare 489 273 216 10,505 10,660 10,440 0 18 0 0 0 31,597 32,800 31,288 19 64 85 43 42 9,348 9,711 9,081 65+ 404 231 174 10,784 10,855 10,780 Medicaid 367 211 156 6,899 6,812 7,020 0 18 96 44 52 3,573 3,318 3,819 19 64 188 108 80 9,079 7,744 11,830 65+ 83 59 23 14,979 15,584 13,651 Other payers and programs a 278 144 134 900 918 881 0 18 46 22 24 590 586 593 19 64 172 90 82 906 941 870 65+ 60 32 28 1,479 1,384 1,602 SOURCE Centers for Medicare and Medicaid Services, Office of the Actuary, National Health Statistics Group. NOTES Numbers may not sum to totals because of rounding. Data for private health insurance, Medicare, and Medicaid are per enrollee. Data for all other payers are per capita. a Includes health-related spending for Children s Health Insurance Program Titles XIX and XXI, Department of Defense, Department of Veterans Affairs, worksite health care, other private revenues, Indian Health Services, workers compensation, general assistance, maternal and child health, vocational rehabilitation, other federal programs, Substance Abuse and Mental Health Services Administration, other state and local programs, and school health. A closer examination revealed that total Medicaid spending in 2010 for the working-age group was 35 percent higher for females than for males. Medicaid is a primary payer for women s reproductive health services, paying for 48 percent of all births in the United States. 25 Conversely, when we considered Medicaid spending on a per enrollee basis, working-age males were 53 percent more expensive than females in 2010. This percentage has been declining since 2002, when it was 65 percent. The decline was partly due to a shift of spending for prescription drugs from Medicaid to Medicare for beneficiaries enrolled in both programs. This shift has disproportionately affected male enrollees in this age group because most of them are dually eligible (based on disability). In contrast, most female enrollees in this age group are pregnant women. In 2010 hospital care (37 percent), physician and clinical services (26 percent), and retail prescription drugs (14 percent) constituted just over three-quarters of the personal health care spending for the working-age population. Males and females spent about the same amount of their respective total spending on drugs. However, males spent a larger portion on hospital care (39 percent, versus 35 percent for females), and females spent a larger percentage on physician and clinical services (29 percent, versus 23 percent for males). On a per capita basis, females spent more than males for hospital and physician and clinical services (Exhibit 5), primarily because of pregnancies among females ages 19 44. For retail prescription drugs, females in the same age group spent 47 percent more than males, partly because of spending on drugs for depressive disorders, multiple sclerosis, mood disorders, migraine headaches, and oral contraceptives. 16 Conversely, per capita hospital spending for males ages 45 64 was 20 percent greater than that for females of the same age, in part due to procedures that treat heart disease. 16 THE ELDERLY: People ages sixty-five and older accounted for 34 percent ($744 billion) of all personal health care spending and 13 percent of the US population in 2010 (Exhibit 3), compared with 34 percent and 12 percent in 2002, respectively. On a per capita basis in 2010, they spent more on health care than the other groups (Exhibit 1) because demand for health care increases with age. The largest payer for seniors in 2010 was Medicare (Exhibit 4), at 54 percent (compared to 47 percent before the implementation of Part D), followed by private health insurance and out-of-pocket expenditures, each of which was 13 percent. In the period 2002 10 Medicaid s share declined 2 percentage points, to 11 percent. This was because of Part D and cost containment strategies by states to keep patients out of nursing homes. Medicaid spending for nursing care facilities and continuing care retirement communities for the elderly increased, on average, only 2.5 percent for each year in the same period. In 2010, 35 percent of health care spending for the elderly was for hospital services, 19 percent was for physician and clinical services, and 16 percent was for nursing care facilities and continuing care retirement communities. In 2010 females ages sixty-five and older devoted 19 percent of their health care spending to nursing care facilities and continuing care retirement communities, compared with 21 percent in 2002. The percentage for males remained stable, at 12 percent. 820 Health Affairs May 2014 33:5

Exhibit 5 US Personal Health Care Spending, By Gender, Age, And Service, 2010 Total spending (billions of dollars) Per capita spending (dollars) Service/age group (years) Total Females Males Total Females Males All personal health care 2,193 1,231 962 7,097 7,860 6,313 0 18 285 137 148 3,628 3,572 3,680 19 64 1,164 657 507 6,125 6,892 5,353 65+ 744 437 307 18,424 19,110 17,530 Hospital care 813 433 380 2,630 2,763 2,493 0 18 121 59 61 1,538 1,548 1,528 19 64 429 232 197 2,259 2,434 2,083 65+ 262 141 121 6,500 6,179 6,918 Physician and clinical services 519 299 220 1,680 1,911 1,441 0 18 76 36 40 972 937 1,005 19 64 304 188 116 1,601 1,977 1,223 65+ 138 75 63 3,425 3,277 3,618 Nursing care facilities and continuing care retirement communities 143 94 49 463 602 320 0 18 1 0 1 11 9 14 19 64 21 10 12 113 104 122 65+ 121 84 37 2,990 3,677 2,095 Retail prescription drugs 256 144 112 827 919 734 0 18 18 8 10 229 199 257 19 64 161 92 70 849 962 734 65+ 76 44 32 1,892 1,946 1,822 All other services a 463 261 202 1,497 1,655 1,325 0 18 69 34 35 878 879 877 19 64 248 135 113 1,303 1,415 1,191 65+ 146 92 54 3,617 4,031 3,076 SOURCE Centers for Medicare and Medicaid Services, Office of the Actuary, National Health Statistics Group. NOTE Numbers may not sum to totals because of rounding. a Includes dental services; other professional services; home health care; retail outlet sales of durable and nondurable medical equipment; and other health, residential, and personal care. On a per capita basis, females ages sixtyfive and older spent $19,110 on personal health care in 2010, $1,580 more than males spent (Exhibit 1). This difference was largely due to spending for nursing care facilities: Females spent 1.8 times more than males did for this long-term care service, in part because of longer relative life expectancies. Life expectancy in 2010 was 78.7 years overall, 81.0 years for females, and 76.2 years for males. In comparison, in 2002 life expectancy was 77.0 years overall, 79.6 years for females, and 74.4 years for males. 17 The narrowing of the difference between males and females in life expectancy from 5.2 years in 2002 to 4.8 years in 2010 follows a longer-term trend. The difference in life expectancy was 7.6 years in 1970, 7.4 years in 1980, and 7.0 years in 1990. 17 These changing demographics have an impact on Medicare spending trends, both historically and in the future. A discussion of how historical demographic trends are accounted for when analyzing Medicare spending is provided in the online Appendix. 26 Conclusion In the period 2002 10 the proportion of personal health care spending by age remained relatively stable. During this period, however, the rate of increase in per capita spending was most rapid for children, reflecting an increase in their insurance coverage. The gap in per capita spending between males and females narrowed slightly, mainly due to trends in prescription drug use. Future releases of this biennial series of health spending data by age and gender will provide a first look at the impact of the baby boomers entering the elderly age group and becoming eligible for Medicare. Additionally, the estimates will show the impact of Affordable Care Act provisions such as the requirement for insurers to cover dependents of enrollees up to age twentysix, expanded Medicaid coverage for workingage adults, and the implementation of health care Marketplaces. May 2014 33:5 Health Affairs 821

TheauthorsthankmembersoftheCMS Office of the Actuary and anonymous peer reviewers for their helpful comments. The opinions expressed here are the authors and not necessarily those of the Centers for Medicare and Medicaid Services. NOTES 1 US Census Bureau. Population estimates [Internet]. Washington (DC): The Bureau [cited 2014 Mar 31]. Available from: http://www.census.gov/popest/data/intercensal/ national/nat2010.html 2 Alemayehu B, Warner KE. The lifetime distribution of health care costs. Health Serv Res. 2004;39(3): 627 42. 3 Martin AB, Hartman H, Whittle L, Catlin A, National Health Accounts Team. National health spending in 2012: rate of health spending growth remained low for the fourth consecutive year. Health Aff (Millwood). 2014;33(1):67 77. 4 Centers for Medicare and Medicaid Services. National Health Expenditures Accounts: methodology paper, 2012: definitions, sources, and methods [Internet]. Baltimore (MD): CMS; [cited 2014 Mar 24]. Available from: http://www.cms.gov/research-statistics-data-and- Systems/Statistics-Trends-and- Reports/NationalHealthExpend Data/Downloads/dsm-12.pdf 5 Hartman M, Catlin A, Lassman D, Cylus J, Heffler S. US health spending by age, selected years through 2004. Health Aff (Millwood). 2008; 27(1):w1 12. DOI: 10.1377/hlthaff.27.1.w1. 6 Cylus J, Hartman M, Washington B, Andrews K, Catlin A. Pronounced gender and age differences are evident in personal health care spending per person. Health Aff (Millwood). 2011;30(1):153 60. 7 Centers for Medicare and Medicaid Services. National health expenditure data: age and gender [Internet]. Baltimore (MD): CMS; [last modified 2013 May 9; cited 2014 Mar 18]. Available from: http://www.cms.gov/research-statistics-data-and- Systems/Statistics-Trends-and- Reports/NationalHealthExpend Data/Age-and-Gender.html 8 The years that have not been revised are 1987, 1996, and 1999. 9 Non personal health care spending includes government administration and the net cost of private health insurance, noncommercial research, investment in structures and equipment, and government public health activities. 10 These estimates were adjusted to equal the total spending levels in the 2012 National Health Expenditure Accounts, data that were released in January 2014. See Note 3. 11 Centers for Medicare and Medicaid Services. Age and gender estimates in the National Health Expenditure Accounts: definitions, sources, and methods [Internet]. Baltimore (MD): CMS; [cited 2014 May 5]. Available from: http://www.cms.gov/research-statistics-data-and- Systems/Statistics-Trends-and- Reports/NationalHealthExpend- Data/downloads/2010Genderand AgeMethodology.pdf 12 Yamamoto DM. Health care costs from birth to death [Internet]. Washington (DC): Health Care Cost Institute; 2013 Jun [cited 2014 Mar 18]. (Report No. 2013-1). Available from: http://www.health costinstitute.org/files/age-curve- Study_0.pdf 13 Young K, Garfield R, Clemans-Cope L, Lawton E, Holahan J. Enrollmentdriven expenditure growth: Medicaid spending during the economic downturn, FY 2007 2011 [Internet]. Washington (DC): Kaiser Commission on Medicaid and the Uninsured; 2013 Apr [cited 2014 Mar 18]. Available from: http:// kaiserfamilyfoundation.files.word press.com/2013/05/8309-02.pdf 14 In 2010 a risk score coding intensity adjustment was made to Medicare Advantage payment rates in order to compensate for excess growth in Medicare Advantage risk scores relative to those of fee-for-service beneficiaries. 15 Kaiser Family Foundation. Retail prescription drugs filled at pharmacies (annual per capita by gender) [Internet]. Menlo Park (CA): KFF; [cited 2014 Mar 18]. Available from: http://kff.org/other/stateindicator/retail-rx-drugs-by-gender/ 16 Data on medical conditions were collected from the household respondents sampled in the Medical Expenditure Panel Survey (MEPS). Agency for Healthcare Research and Quality. Medical Expenditure Panel Survey: Household Component [Internet]. Rockville (MD): AHRQ; [last revised 2010 Oct 25; cited 2014 Mar 18]. Available from: http:// meps.ahrq.gov/mepsweb/survey_ comp/household.jsp 17 Murphy SL, Xu J, Kochanek KD. Deaths: final data for 2010. National Vital Statistics Reports [serial on the Internet]. 2013 May 8 [cited 2014 Mar 18]. Available from: http://www.cdc.gov/nchs/data/nvsr/nvsr61/ nvsr61_04.pdf 18 Martin JA, Hamilton BE, Ventura SJ, Osterman MJK, Wilson EC, Mathews TJ. Births: final data for 2010. National Vital Statistics Reports [serial on the Internet]. 2012 Aug 28 [cited 2014 Mar 18]. Available from: http://www.cdc.gov/nchs/data/ nvsr/nvsr61/nvsr61_01.pdf 19 Pew Research Social and Demographic Trends. Young, underemployed, and optimistic [Internet]. Washington (DC): Pew Research Center; 2012 Feb 9 [cited 2014 Mar 18]. Available from: http:// www.pewsocialtrends.org/2012/02/ 09/young-underemployed-andoptimistic 20 DeNavas-Walt C, Proctor BD, Smith JC. Income, poverty, and health insurance coverage in the United States: 2012 [Internet]. Washington (DC): Census Bureau; 2013 Sep [cited 2014 Mar 18]. Available from: http://www.census.gov/prod/ 2013pubs/p60-245.pdf 21 Census Bureau. Population estimates: current estimates data [Internet]. Washington (DC): Census Bureau; [cited 2014 Mar 18]. Available from: http://www.census.gov/ popest/data/index.html 22 O Hare W. The changing child population of the United States: analysis of data from the 2010 census [Internet]. Baltimore (MD): Annie E. Casey Foundation; 2011 Nov [cited 2014 Mar 18]. (KIDS COUNT Working Paper). Available from: http://www.aecf.org/~/media/ Pubs/Initiatives/KIDS%20COUNT/ T/TheChangingChildPopulationof theunitedstates/aecfchanging ChildPopulationv8web.pdf 23 Payne C. Fertility forecast: baby bust is over; births will rise. USA Today. 2013 August 1. 24 Briody B. Kids and dental health: rising costs and struggling state programs a dangerous mix. Kaiser Health News [serial on the Internet]. 2010 Jul 20 [cited 2014 Mar 18]. Available from: http://www.kaiser healthnews.org/stories/2010/july/ 21/ft-dental-care-costs.aspx 25 Markus AR, Andres E, West KD, Garro N, Pellegrini C. Medicaid covered births, 2008 through 2010, in the context of the implementation of health reform. Womens Health Issues. 2013;23(5):e273 80. 26 To access the Appendix, click on the Appendix link in the box to the right of the article online. 822 Health Affairs May 2014 33:5