Issue Brief #10 May 2015 Per Capita Health Care KEY FINDINGS: 2013 Spending on Diabetes: $14,999 Per capita spending for indi- 2009-2013 viduals with diabetes $4,305 Per capita spending for indi- viduals without diabetes Diabetes is a costly chronic condition in the United States, medi- cal costs and productivity loss attributable to diabetes were es- $1,922 Out-of-pocket spending per timated to be $245 billion in 2012.1 In this issue brief, for indi- capita for individuals with diabetes viduals covered by employer-sponsored insurance (ESI) and younger than age 65, per capita spending for people with a di- $738 Out-of-pocket spending per cap- ita for individuals without diabetes agnosis of diabetes was compared to those without a diagnosis for the years 2009 through 2013. During that period, spending $15,456 Per capita spending for chil- for individuals with diabetes increased by roughly $1,000 to dren (ages 0–18) with diabetes about $15,000 per capita. The average per capita spending difference between people with and without diabetes was $16,889 Per capita spending for pre- Medicare adults (ages 55–64) with dia- $10,310 (Figure 1). Additionally, during this period, people with betes diabetes spent on average 2.5 times more out of pocket than people without diabetes. Among individuals with diabetes, chil- $1,361 2013 year-over-year increase dren (ages 0 through 18) and pre-Medicare adults (ages 55 in per capita spending for children with through 64) were the two groups with the highest per capita diabetes health care spending in every year of the study period. $604 2013 year-over-year increase in per capita spending for pre-Medicare adults with diabetes Diabetes is one of the most common chronic conditions and is the seventh leading cause of death in the United States. 2,3 Since 1980, the prevalence of diabetes has risen steadily. According to the Centers for Disease Control and Prevention (CDC), in 2012, 29.1 million Americans—or 9.3% of the population—had diagnosed and undiagnosed diabetes. From 1980 through 2012, the number of individu- als with diagnosed diabetes in the United States increased from 5.1 million to 21 mil- lion.3 During this period, the number of deaths from diabetes in the United States rose, while the death rate from diabetes de- clined.4 The American Diabetes Association (ADA) estimated that 4.6% of the privately insured www.healthcostinstitute.org 1 population in the United States had di- agnosed diabetes in 2012. In addition to the disease’s rising prevalence, the med- ical costs associated with diabetes are high. In 2012, the estimated direct medi- cal cost of diabetes in the United States was $176 billion.1 Although information on diabetes prev- alence and the economic costs of diabe- tes exists, less is known about per capita health care spending for individuals with diagnosed diabetes relative to spending for those without diabetes, either in terms of their total spending or their out-of-pocket costs. This is espe- cially true for the per capita spending for children with diagnosed diabetes. This issue brief begins to fill that re- search gap: It compares per capita health care spending for the two popula- Individuals with Diabetes in Claims Da- nosed diabetes in the United States in tions, those with and without a diabetes ta,” “Data and Methods,” and 2013.5 diagnosis. Further, this issue brief exam- ines health spending for individuals “Limitations”.) Per Capita Health Care Spending for with diabetes by age and gender. We examine the population identified as Individuals With and Without Diabetes HCCI’s ESI Population with Diabetes having diabetes as compared to the pop- In 2013, for individuals with diagnosed ulation who did not have diabetes, ra- diabetes, per capita health care spend- The analytic dataset used in this issue ther than comparing per capita spend- ing was $14,999 (Table 1; Figure 1). Be- brief includes nearly 40 million individ- ing for individuals with diabetes to the tween 2012 and 2013, per capita spend- uals from all 50 states and the District of national ESI population (see ing for individuals with diabetes rose Columbia, younger than age 65 and hav- “Limitations”). Using the analytic da- $595, or 4.1%. Additionally, from 2009 ing employer-sponsored insurance taset, we estimated that 5.3% of the ESI to 2013, per capita spending increased (ESI). HCCI identified individuals with population had diagnosed diabetes in at an average annual rate of 1.8%. The diagnosed diabetes mellitus (type 1 or 2013 (see “Individuals with Diagnosed only time when spending declined was type 2) in the dataset to examine their between 2010 and 2011, when it fell Diabetes in the HCCI Population”). This per capita health care spending trends. 0.1%, or $15 per capita (Figure 2). HCCI suggests there were roughly 9 million (For more information see “Identifying did not examine the causes of this de- individuals covered by ESI with diag- Individuals with Diagnosed Diabetes in the HCCI Population For 2013, HCCI identified 5.3% of individuals younger than age 65 and having ESI as having diagnosed diabetes (Table 2). From 2009 through 2013, the percentage of those with diagnosed diabetes rose by 0.6%, from 4.7% of the population. HCCI’s 2013 estimated rate of those with diagnosed diabetes, younger than age 65, and covered by ESI is slightly higher than other estimates. For example, the ADA estimated a rate of 4.6% in 2012 for the privately insured population. 1 These rate differ- ences may be attributable to differences in population and study methodology. In 2013, in the HCCI dataset of ESI individuals by age group, the share of those with diagnosed diabetes ranged from 7.0% for young adults (ages 19–64) to 0.3% for children (ages 18 years and younger). www.healthcostinstitute.org 2 HCCI AGE GROUPS Children Ages 0 through 18. Young Adults Ages 19 through 25. Intermediate Adults Ages 26 through 44. Middle-Aged Adults Ages 45 through 54. Pre-Medicare Adults Ages 55 through 64. cline in spending; however, the figure is out diabetes; however, the ADA popula- betes spent $1,922 out of pocket on consistent with a national slowdown in tion included not only privately insured health care services, an increase of $75 health care spending.6 individuals but those with Medicare and (4.1%) over that of 2012 (Table 3; Fig- Medicaid.1 ure 3). From 2009 to 2013, out-of- In 2013, per capita spending for people pocket spending grew every year, at an without diagnosed diabetes was $4,305. Focusing on the HCCI data, the ratio of average annual rate of 3.4%.7 Between Per capita spending for individuals with health care spending for individuals 2010 and 2011, out-of-pocket spending diabetes was 3.5 times higher than per with diabetes to those without diabetes grew 1.6%, despite a small decrease in capita spending for those without diabe- dropped from 3.74 in 2009 to 3.5 in total per capita spending for those with tes. The difference in per capita spend- 2013. Spending for those without diabe- diabetes. ing for people with and without a diabe- tes grew during the study period at an tes diagnosis, as identified by HCCI, was average annual rate of 3.6% as com- A large difference was seen in per capita larger than other estimates of the differ- pared with a 1.8% growth for those with out-of-pocket health care spending be- ence. For instance, the ADA estimated diabetes. tween those with diagnosed diabetes that in 2012, individuals with diagnosed and those without; in 2013, out-of- Out-of-Pocket Spending diabetes had health expenditures that pocket spending for those with diabetes were 2.3 times higher than those with- In 2013, individuals with diagnosed dia- was $1,184 higher than for those with- Identifying Individuals with Diagnosed Diabetes Using Claims Data HCCI determined a diagnosis of diabetes using the Dictionary of Disease Management Terminology (DDMT) methodology for identifying health care activity associated with diabetes.8 Individuals with a diagnosis of diabetes for at least one inpatient ad- mission, one outpatient visit, or two office visits within the same calendar year were identified in a year of data as having diabe- tes (see the HCCI Methodology document for a list of codes included in this categorization). 9 After individuals have been identi- fied as having received a diagnosis of diabetes, they retain this designation in all subsequent years of the dataset. We excluded radiology and laboratory claims from this methodology, as these can be used for screening purposes and may not reflect a diag- nosis of diabetes. www.healthcostinstitute.org 3 children with diabetes was $15,456 (Table 4). While per capita spending declined slightly for children between 2009 and 2011, large spending increas- es occurred in 2012 and 2013. Between 2011 and 2013, children with diabetes had the fastest per capita spending growth as compared to growth for the other age groups with diabetes: 7.0% between 2011 and 2012 and 9.6% between 2012 and 2013 (Table 5). The $1,361 increase experienced between 2012 and 2013 was the largest increase in dollars spent per capita for any age group during that period, nearly double the next largest dollar increase, which was for young adults ($753). In 2009 and 2010, in the population of children with diabetes, spending was higher for boys as compared with girls out diabetes ($738). However, out-of- spending trends for individuals diag- ($515 in 2010; Figure 5). However, this pocket spending was a larger share of nosed with diabetes by age group and trend switched in 2011, and spending total per capita spending for people gender. was higher for girls between 2011 and without diabetes: 17.1% as compared to 2013; in 2013, spending for girls as Children (Ages 0–18) with Diabetes 12.8% for people with diabetes. Addi- compared with boys was $557 more per tionally, from 2009 to 2013, per capita From 2009 through 2013, children with capita. out-of-pocket spending for people with- diagnosed diabetes had the second high- est per capita spending of any age out diabetes grew faster than spending group. In 2013, per capita spending for for those with diabetes: at average an- nual rates of 5.0% and 3.4%, respective- ly. Per Capita Health Care Spending for Individuals with Diabetes by Age and Gender In 2013, the highest per capita spending for individuals with diabetes was for children and pre-Medicare adults (Figure 4). Between 2011 and 2013, per capita spending grew faster for children than for those in any other age group. Additionally, in each age group, per cap- ita spending was higher for women than for men, with the exception of the pre- Medicare adults group, in which spend- ing was higher for men. The following sections discuss the total health care www.healthcostinstitute.org 4 Middle Three Age Groups (Ages 19–25, spending growth between 2011 and diabetes had the highest per capita 26–44, and 45–54) with Diabetes 2013. health care spending for any age group in all years of the study period. In 2013, Middle age adults with diagnosed diabe- As shown in Figure 4, per capita health spending for pre-Medicare adults was tes (young adults, intermediate adults, care spending was higher for women $16,889 per capita, 3.7% higher than in and middle-aged adults) had the lowest than for men in each of the three middle 2012. In most years of the study, spend- per capita spending. In each year stud- age groups. The spending difference ing growth was relatively slower, and ied, intermediate adults had the lowest between men and women was largest the per capita dollar increases were per capita spending ($11,946 in 2013) for intermediate adults; in 2013, spend- lower for pre-Medicare adults than for followed by that for young adults ing was $3,300 higher for intermediate the other age groups (Table 5). ($13,524 in 2013; Table 4). While per adult women than spending for interme- capita spending levels were low for diate adult men (Table 4). This differ- Spending for pre-Medicare adult men these groups, they generally had the ence between men and women was with diabetes was higher than that for second (young adults) and third compared to spending differences of pre-Medicare adult women with diabe- (intermediate adults) fastest spending $3,139 for young adults and $1,309 for tes in every year of the study; in 2013, growth between 2011 and 2013. In middle-aged adults. the per capita spending difference was 2013, at $13,886 per capita, middle- $36. Between 2012 and 2013, pre- Pre-Medicare Adults (Ages 55–64) with aged adults had higher health care Medicare adult men’s spending in- Diabetes spending than either young adults or creased 4.2% as compared to 3.1% for intermediate adults but had the slowest Pre-Medicare adults diagnosed with women. Per Capita Spending on Anti-Diabetic Agents To analyze prescription drug spending related to diabetes, HCCI calculated per capita spending for individuals with diabetes for their use of anti-diabetic agents (defined using the American Hospital Formulary System definition; see the HCCI Methodology document for more information).9,10 This class of prescription drugs, including, for example, insulin and metformin, are general- ly used to treat and manage diabetes. Spending on branded anti-diabetics and spending on generic anti-diabetics differed by age group. For example, per capita spending on branded anti-diabetic agents was higher for children and young adults than for individuals in the other age groups. In contrast, spending on generic anti-diabetic agents was higher for middle-aged adults and pre-Medicare adults than for those in other age groups. In 2013, for children and young adults, per capita spending on branded anti-diabetic agents made up most of the spending on anti-diabetic agents (99% and 98%, respectively; Table 6). For middle-aged adults and pre-Medicare adults the majority of spending on anti-diabetic agents was on branded drugs (88% and 87%, respectively), but this was a lower share of the spending than for children and young adults. From 2009 to 2013, per capita spending on branded anti-diabetic agents increased 70% for children and 38% for pre-Medicare adults. Over the same period, per capita spending on generic drugs fell 87% for children as compared with a 10% drop for pre- Medicare adults. One factor that might influence these spending differences is higher spending on insulins for children and young adults than for those in other age groups (Table 7). In 2013, spending on insulins constituted 99% and 95%, respectively, of spending on anti- diabetic agents for children and young adults as compared with 52% and 48%, respectively, spent for middle-aged adults and pre-Medicare adults. www.healthcostinstitute.org 5 Conclusion Data and Methods tes—are similar but not methodological- ly identical. Per capita spending trends In 2012, in the United States, diabetes This issue brief used an analytic dataset for these populations should be treated cost $245 billion in direct medical costs that consisted of weighted and aggregat- as estimates. and reduced productivity.1 In the HCCI ed claims data for people younger than data in 2013, spending per capita for age 65 and covered by ESI for calendar Limitations individuals diagnosed with diabetes was years 2009 to 2013.9 The analytic da- Our study has several limitations that $14,999, more than 3 times higher than taset was derived from health care can affect the interpretation of the find- spending for individuals without diabe- claims for 40 million Americans per year ings. For this reason, HCCI considers its tes ($4,305). Furthermore, individuals contributed by three national insurers work a starting point for analysis and with diabetes had out-of-pocket costs and was used for the 2013 Health Care research on individuals younger than that were 2.5 times higher than out-of- Cost and Utilization Report.5 All data age 65 covered by ESI and diagnosed pocket costs for individuals without dia- used for our study were de-identified with diabetes, rather than a complete betes. and compliant with the Health Insur- analysis of this population’s effect on ance Portability and Accountability Act. Among individuals diagnosed with dia- health care in the United States. betes, the two age groups that had the A diagnosis of diabetes was determined First, our findings are estimates for the highest per capita spending were chil- using the DDMT methodology for identi- United States ESI population based on a dren and pre-Medicare adults. Between fying health care activity associated with sample of approximately 25% of ESI 2011 and 2013, children with diabetes diabetes.8 Individuals with a diagnosis of insureds younger than age 65. The esti- also had the largest increases in per cap- diabetes for at least one inpatient ad- mates for numbers of insured individu- ita health care spending of any age mission, one outpatient visit, or two of- als by each plan type were weighted to group. This increase in spending for fice visits in the same calendar year account for any demographic differ- children with diabetes was driven in were flagged in a year of data as having ences between the HCCI sample and part by increases in spending on brand- been diagnosed with diabetes (see the population estimates based on the Unit- ed anti-diabetic agents, specifically HCCI Methodology document for a list of ed States Census, making the dataset branded insulin. These findings demon- codes included in this categorization).9 representative of the national, ESI popu- strate the need for additional research Once individuals have been flagged as lation younger than age 65.9 on the health care spending trends of having received a diagnosis of diabetes, individuals with diabetes in the United they retain this flag in all subsequent Second, because HCCI’s claims holdings States. years of the dataset. This methodology reflect only explicit health care activity, excluded radiology and laboratory HCCI could not identify individuals with claims, as these can be used for screen- diabetes who (1) did not seek medical ing purposes and may not reflect a diag- care between 2009 and 2013, (2) did nosis of diabetes. not meet our criteria for study inclusion, (3) did not file with their health insurer To be flagged in the HCCI dataset as hav- a claim that indicated a diagnosis of dia- ing a diagnosis of diabetes, individuals betes, or (4) had undiagnosed diabetes. must have had at least one medical Moreover, claims data have a mixed rec- claim filed with their insurer in one of ord of utility for population health stud- the years of the study period. The popu- ies.11 Work is ongoing to improve the lation of individuals without diabetes is methods used to determine health sta- composed of all members in the HCCI tus from administrative claims. To that analytic dataset who were not flagged as end, HCCI invites readers to review the having received a diabetes diagnosis. methodology for this report and com- This population without diabetes includ- ment on how to better identify the ed individuals who never had a medical chronically ill from claims data.7 Sugges- claim filed with their insurer during the tions and other inquiries should be di- study period. Therefore, these two pop- rected to the contact form on the HCCI ulations—individuals diagnosed with Website. Third, this is a descriptive diabetes and individuals without diabe- study, and findings are not causal. The www.healthcostinstitute.org 6 tables and figures presented are limited Endnotes between 2009 and 2013, to $1,135. to descriptive statistics for individuals 1. American Diabetes Association. 8. Duncan, I, ed. Dictionary of Disease covered by ESI and younger than age "Economic Costs of Diabetes in the US Management Terminology. 65. In this brief, we presented per capi- in 2012."Diabetes Care 36.4 (2013): Washington, DC: Disease Management ta spending trends for individuals 1033-1046. Association of America, 2004. flagged as having diagnosed diabetes and those not flagged as having diabe- 2. Centers for Disease Control and 9. Health Care Cost Institute. 2013 tes. For more information about the Prevention (CDC). Risk Factor Health Care Cost and Utilization calculation of per capita spending Surveillance System Survey Data. U.S. Report Analytic Methodology v. 3.2. trends, see the HCCI Methodology docu- Department of Health and Human Health Care Cost Institute, Oct. 2013. ment.9 Services, 2013. Web. Web. Fourth, diabetes may onset owing to 3. Centers for Disease Control and 10. McEvoy, GK., ed. AHFS Drug genetic factors (type 1 diabetes) or ow- Prevention (CDC). National Diabetes Information. Bethesda, MD: American ing to environmental or lifestyle factors Statistics Report, 2014. CDC, 2014. Society of Health-System Pharmacists, (type 2 diabetes). HCCI did not distin- Web. 2014. PEPID. Web. guish between type 1 and type 2 diabe- tes in this report. 4. Murphy SL, Xu JQ, Kochanek KD. 11. Burton, B, Jesilow, P. "How "Deaths: Final Data 2013". National Healthcare Studies Use Claims Data." vital statistics reports; vol 64 no 2. The Open Health Services and Policy Hyattsville, MD: National Center for Journal 4.1 (2011): 26–29. Health Statistics. 2014. 5. Estimate is based on Community Population Survey estimate of individuals covered by employer- based insurance (~169 million). 6. Health Care Cost Institute. 2013 Health Care Cost and Utilization Report. HCCI, Oct. 2014. Web. 7. The rise in out-of-pocket spending over the study period parallels rising deductibles for individuals with ESI. According to the Kaiser Family Foundation the average deductible for an individual with ESI rose 40% Authors Copyright 2015 Patrick Shakiba, Amanda Frost Health Care Cost Institute, Inc. Unless explicitly noted, the content of Contact this report is licensed under a Creative Amanda Frost Commons Attribution Non-Commercial Health Care Cost Institute, Inc. No Derivatives 4.0 License 1100 G Street NW, Suite 600 Washington, DC 20005 This HCCI research product was inde- 202-803-5200 pendently initiated by HCCI and is part of the HCCI research agenda. www.healthcostinstitute.org 7 Table 1: Per Capita Expenditures for Insureds With and Without Diabetes (2009-2013) 2009 2010 2011 2012 2013 Insureds with Diabetes Per Capita $13,981 $14,108 $14,093 $14,404 $14,999 Percent Change - 0.9% -0.1% 2.2% 4.1% Insureds without Diabetes Per Capita $3,742 $3,844 $4,000 $4,146 $4,305 Percent Change - 2.7% 4.1% 3.6% 3.8% Source: HCCI, 2015. Notes: Data represents the population of insureds 0-64 covered by ESI. Actuarial completion was performed on data from 2012 and 2013. All per capita dollars calculated from allowed amounts. All figures rounded. Table 2: Percentage of Insureds With and Without Diabetes (2009-2013) 2009 2010 2011 2012 2013 Insureds with Diabetes Total Insureds 4.7% 4.9% 5.2% 5.3% 5.3% Adults (19-64) 6.2% 6.5% 6.9% 7.0% 7.0% Children (0-18) 0.3% 0.3% 0.3% 0.3% 0.3% Insureds without Diabetes Total Insureds 95.3% 95.1% 94.8% 94.7% 94.7% Adults (19-64) 93.8% 93.5% 93.1% 93.0% 93.0% Children (0-18) 99.7% 99.7% 99.7% 99.7% 99.7% Source: HCCI, 2015. Notes: Data represents the population of insureds 0-64 covered by ESI. All per capita dollars calculated from allowed amounts. All figures rounded. Table 3: Out-of-Pocket Per Capita Expenditures For Insureds With and Without Diabetes (2009-2013) 2009 2010 2011 2012 2013 Insureds with Diabetes Per Capita $1,684 $1,768 $1,796 $1,847 $1,922 Percent Change - 5.0% 1.6% 2.8% 4.1% Insureds without Diabetes Per Capita $607 $648 $678 $710 $738 Percent Change - 6.7% 4.6% 4.8% 4.0% Source: HCCI, 2015. Notes: Data represents the population of insureds 0-64 covered by ESI. Actuarial completion was performed on data from 2012 and 2013. All per capita dollars calculated from allowed amounts. All figures rounded. www.healthcostinstitute.org 8 Table 4: Per Capita Expenditures Spending for Insureds with Diabetes, by Gender and Age Group (2009-2013) 2009 2010 2011 2012 2013 Children (0-18) All $13,688 $13,556 $13,178 $14,095 $15,456 Boys $14,004 $13,813 $12,931 $13,902 $15,181 Girls $13,365 $13,298 $13,441 $14,294 $15,738 Young Adults (19-25) All $11,591 $12,643 $12,043 $12,771 $13,524 Men $10,486 $11,276 $10,522 $11,683 $11,851 Women $12,562 $13,874 $13,369 $13,716 $14,990 Intermediate Adults (26-44) All $11,287 $11,222 $11,142 $11,458 $11,946 Men $9,789 $9,597 $9,653 $9,856 $10,282 Women $12,841 $12,893 $12,631 $13,045 $13,582 Middle-Aged Adults (45-54) All $13,120 $13,123 $13,130 $13,359 $13,886 Men $12,413 $12,424 $12,544 $12,737 $13,290 Women $13,964 $13,965 $13,829 $14,099 $14,599 Pre-Medicare Adults (55-64) All $15,633 $15,887 $15,928 $16,285 $16,889 Men $15,682 $16,019 $16,091 $16,371 $17,055 Women $15,575 $15,732 $15,738 $16,186 $16,694 Source: HCCI, 2015. Notes: Data represents the population of insureds 0-64 covered by ESI. Actuarial completion was performed on data from 2012 and 2013. All per capita dollars calculated from allowed amounts. All figures rounded. www.healthcostinstitute.org 9 Table 5: Change in Per Capita Expenditures for Insureds with Dia- betes, by Gender and Age Group (2009-2013) 2009/2010 2010/2011 2011/2012 2012/2013 Children (0-18) All -1.0% -2.8% 7.0% 9.6% Boys -1.4% -6.4% 7.5% 9.2% Girls -0.5% 1.1% 6.3% 10.1% Young Adults (19-25) All 9.1% -4.7% 6.0% 5.9% Men 7.5% -6.7% 11.0% 1.4% Women 10.4% -3.6% 2.6% 9.3% Intermediate Adults (26-44) All -0.6% -0.7% 2.8% 4.3% Men -2.0% 0.6% 2.1% 4.3% Women 0.4% -2.0% 3.3% 4.1% Middle-Aged Adults (45-54) All 0.0% 0.1% 1.7% 3.9% Men 0.1% 1.0% 1.5% 4.3% Women 0.0% -1.0% 2.0% 3.5% Pre-Medicare Adults (55-64) All 1.6% 0.3% 2.2% 3.7% Men 2.1% 0.5% 1.7% 4.2% Women 1.0% 0.0% 2.8% 3.1% Source: HCCI, 2015. Notes: Data represents the population of insureds 0-64 covered by ESI. Actuarial completion was performed on data from 2012 and 2013. All per capita dollars calculated from allowed amounts. All figures rounded. www.healthcostinstitute.org 10 Table 6: Per Capita Expenditures on Anti-Diabetic Agents for In- sureds with Diabetes, by Age Group (2009-2013) 2009 2010 2011 2012 2013 Branded Children (0-18) $1,488 $1,819 $1,910 $2,103 $2,525 Young Adults (19-25) $1,125 $1,373 $1,323 $1,428 $1,712 Intermediate Adults (26-44) $674 $784 $799 $853 $1,000 Middle-Aged Adults (45-54) $745 $861 $898 $941 $1,041 Pre-Medicare Adults (55-62) $816 $950 $1,003 $1,048 $1,124 Branded Percentage Change Children (0-18) - 22.2% 5.0% 10.1% 20.1% Young Adults (19-25) - 22.0% -3.7% 8.0% 19.9% Intermediate Adults (26-44) - 16.4% 1.9% 6.8% 17.1% Middle-Aged Adults (45-54) - 15.5% 4.3% 4.8% 10.5% Pre-Medicare Adults (55-62) - 16.5% 5.5% 4.5% 7.2% Generic Children (0-18) $199 $32 $28 $28 $27 Young Adults (19-25) $144 $28 $25 $24 $27 Intermediate Adults (26-44) $123 $56 $51 $67 $82 Middle-Aged Adults (45-54) $153 $79 $73 $99 $129 Pre-Medicare Adults (55-62) $180 $98 $91 $126 $162 Generic Percentage Change Children (0-18) - -83.9% -11.2% -2.3% -4.1% Young Adults (19-25) - -80.5% -10.8% -2.6% 9.3% Intermediate Adults (26-44) - -54.7% -8.5% 30.5% 23.7% Middle-Aged Adults (45-54) - -48.1% -8.3% 36.6% 29.4% Pre-Medicare Adults (55-62) - -45.6% -7.3% 39.4% 28.0% Source: HCCI, 2015. Notes: Data represents the population of insureds 0-64 covered by ESI. Actuarial completion was performed on data from 2012 and 2013. All per capita dollars calculated from allowed amounts. All figures rounded. www.healthcostinstitute.org 11 Table 7: Per Capita Expenditures on Insulin for Insureds with Dia- betes, by Age Group (2009-2013) 2009 2010 2011 2012 2013 Branded Children (0-18) $1,478 $1,808 $1,899 $2,090 $2,511 Young Adults (19-25) $1,072 $1,315 $1,267 $1,372 $1,644 Intermediate Adults (26-44) $406 $489 $504 $560 $695 Middle-Aged Adults (45-54) $310 $377 $407 $475 $589 Pre-Medicare Adults (55-62) $298 $375 $412 $494 $617 Branded Percentage Change Children (0-18) - 22.3% 5.0% 10.0% 20.1% Young Adults (19-25) - 22.7% -3.7% 8.3% 19.9% Intermediate Adults (26-44) - 20.6% 3.0% 11.1% 24.2% Middle-Aged Adults (45-54) - 21.4% 7.9% 16.6% 24.1% Pre-Medicare Adults (55-62) - 25.5% 9.9% 20.0% 24.8% Generic Children (0-18) $193 $28 $25 $23 $21 Young Adults (19-25) $130 $18 $17 $14 $14 Intermediate Adults (26-44) $56 $15 $14 $15 $15 Middle-Aged Adults (45-54) $44 $15 $14 $14 $16 Pre-Medicare Adults (55-62) $46 $18 $17 $17 $19 Generic Percentage Change Children (0-18) - -85.6% -11.4% -5.5% -11.9% Young Adults (19-25) - -85.9% -10.0% -14.8% -3.5% Intermediate Adults (26-44) - -72.9% -6.6% 1.6% 5.9% Middle-Aged Adults (45-54) - -66.8% -4.6% 2.1% 12.0% Pre-Medicare Adults (55-62) - -62.2% -2.5% 2.0% 9.1% Source: HCCI, 2015. Notes: Data represents the population of insureds 0-64 covered by ESI. Actuarial completion was performed on data from 2012 and 2013. All per capita dollars calculated from allowed amounts. All figures rounded. www.healthcostinstitute.org 12