2014 Diabetes Health Care Cost and Utilization Report June 2016 www.healthcostinstitute.org Copyright 2016 Health Care Cost Institute Inc. Unless explicitly noted, the content of this report is licensed under a Creative Commons Attribution Non-Commercial No Derivatives 4.0 License. 2014 Diabetes Health Care Cost and Utilization Report ii Executive Summary BY THE NUMBERS: 2014 The Health Care Cost Institute (HCCI) for more information). is pleased to present the 2014 Diabe- The American Diabetes Association tes Health Care Cost and Utilization Report, with a principal focus on (ADA) estimated that for the entire $16,021 population of Americans with diag- trends in health care spending and Health care spending per capita for nosed diabetes in 2012, per capita utilization for the national population people with diabetes spending on health care was younger than age 65 and covered by $13,741.3 For that year (the most re- employer-sponsored insurance (ESI) cent year that per capita health care with diagnosed diabetes. In this re- port, we also examine selected spend- spending estimates were available $4,396 from the ADA), they found that per ing and utilization trends for the ESI Health care spending per capita for capita spending for people with dia- population without diabetes (see box people without diabetes betes was 3.9 times higher than “Trends in Spending and Utilization of spending for people without diabetes. Health Care Services for People with- This per capita figure for all Ameri- out Diabetes”) and compare trends for the two populations – with and cans with diabetes was about $1,000 12.1% lower than the 2012 spending esti- without diabetes – on selected spend- Share of total health care spending mate for our study population ing and utilization measures. Howev- paid by out-of-pocket by individuals (younger than 65 with ESI) with dia- er, unless specifically stated, the dis- with diabetes betes presented in this report cussion in this report is for the popu- ($14,711). We found that in 2012, per lation with diabetes. capita spending for our study popula- This report covers the period 2012 to tion with diabetes was 3.56 times 17.1% 2014 (the “study period”), with the higher than spending for the study bulk of the analysis focused on 2014. population without diabetes—a Share of total health care spending Additionally, this report is a follow-up slightly lower figure than the ADA’s paid by out-of-pocket by individuals publication to Per Capita Health Care estimate. Much of the difference be- without diabetes Spending on Diabetes: 2009-2013, tween HCCI’s and the ADA’s spending which can be found on the HCCI Web- estimates is probably due to popula- site.1 Using the same general popula- tion and methodological differences 1.2% and –1.0% tion as the previous report on the ESI between the two samples. population with diabetes,2 this report Changes in utilization of outpatient We estimated that between 2012 and details trends in total per capita visits by people with and without dia- 2014, per capita spending for the in- spending, per capita out-of-pocket betes sureds with diabetes grew by $1,310 spending, and utilization of health (Table 1). Further, by 2014, per capita care services. These spending and use spending for this population was 3.64 trends were created at the national- level and broken down by age group times higher than spending for the 9.6% insureds without diabetes (Table 1 and gender. Additionally, we studied Share of total spending per capita for and Appendix Table A2). In addition trends for two broad types of ser- people with diabetes that was spent to per capita spending trends, our vices: medical (acute inpatient admis- on antidiabetic agents report examines per capita out-of- sions, outpatient services, and profes- pocket spending trends and the utili- sional services) and prescription zation rates of health care service use (brand and generic; see box “Service for the insureds with diabetes. This Categories Analyzed in this Report” www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report iii report proceeds in three sections: per same, there were declines in the out- Appendix Table A19). Compared to capita spending trends, out-of-pocket of-pocket spending on prescriptions. people without diabetes, utilization of spending trends, and utilization rates This decline in prescription spending services by people with diabetes of health services. The findings from out of pocket was most pronounced ranged from 2.4 times higher use of each of these sections and the similar- for the oldest age group. Per capita outpatient visits to 7.1 times higher ities and differences in trends be- out-of-pocket spending by pre- use of brand prescriptions (Table 3 tween the insured populations with Medicare adults fell between 2013 and Appendix Table A36). and without diabetes as well as high- and 2014, due to a decline in their out The per capita spending trends for lights for a few notable spending and -of-pocket spending on prescriptions, people with diabetes generally looked utilization trends observed in our specifically brand prescriptions. In quite different from the spending study populations are briefly noted 2012 and 2013, pre-Medicare adults trends for people without diabetes below. had the highest out-of-pocket spend- (Table 1 and Appendix Table A2). For ing of all of the age groups studied. Spending and Utilization Trends for children without diabetes, spending However, due to the decline in this People with Diabetes in this Report per capita was the lowest of any age spending between 2013 and 2014, the group without diabetes studied In 2014, national per capita spending highest out-of-pocket spending per ($2,613 in 2014). In comparison, for insureds with diabetes grew 5.9% capita in 2014 was for children. spending for children with diabetes to $16,021, an increase of $897 over In contrast to the year-on-year in- was the second highest for any age the previous year (Table 1). Per capita crease in spending per capita ob- group with diabetes ($17,380 in spending grew over the entire study served over the study period, the uti- 2014). This trend was similar for period for all medical service catego- lization of medical services and brand young adults. Young adults without ries, both men and women, and all age prescriptions generally declined diabetes had the second-lowest groups (see “HCCI Age Groups”). (Table 3). Small increases in the rates spending per capita of any age group Spending on prescriptions grew at the of use of outpatient visits and generic without diabetes ($2,664), while fastest average annual rate over the prescriptions occurred in every year young adults with diabetes had the study period (8.7%). The second fast- studied. The increase in use of generic third-highest spending of any age est average annual growth was for prescriptions more than offset the group with diabetes ($15,839). Over- outpatient services (5.0%), followed decline in the use of brand prescrip- all, for those without diabetes, the by professional services (2.2%), and tions, leading to a net increase in total younger age groups had low spending acute inpatient care (2.2%). prescription use (brand prescriptions compared to the older age groups, Over the study period, per capita plus generic prescriptions) each year while the opposite tended to be true spending was highest for pre- of the study period. In 2014, there for the younger age groups with dia- Medicare adults (ages 55-64), fol- were 1,242,571 filled days of pre- lowed by children (ages 0-18), young scriptions per 1,000 people with dia- HCCI AGE GROUPS adults (ages 19-25), and middle-age betes. This prescription use rate aver- Children adults (ages 45-54); intermediate ages to just over 3 filled days of pre- Ages 0 through 18. adults (ages 26-44) had the lowest scriptions per person with diabetes spending. per day. Young Adults Corresponding to the increase in total Comparison of Trends for People Ages 19 through 25. per capita spending were increases with and Without Diabetes Intermediate Adults over the study period in per capita In 2014, total per capita spending for Ages 26 through 44. out-of-pocket spending for insureds the study population with diabetes with diabetes. In 2014, per capita out- Middle-Age Adults was 3.6 times higher than spending of-pocket spending was $1,944, a $10 for those without diabetes (Table 1 Ages 45 through 54. increase over the previous year and Appendix Table A2). Out-of- (Table 2). In each year of the study Pre-Medicare Adults pocket per capita spending was 2.6 period, per capita out-of-pocket times higher for people with diabetes Ages 55 through 64. spending on each of the medical ser- than for those without (Table 2 and vice categories increased. At the www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report iv SERVICE CATEGORIES ANALYZED IN THIS REPORT In this report, HCCI analyzed medical and prescription claims for individuals (ages 0–64) covered by ESI. As in prior reports, HCCI grouped these claims into distinct service categories for analytic purposes. See the 2014 HCCI Analytic Methodology for a detailed description of HCCI’s claims categorization methods. 5 Medical service, subservice, and detailed service categories in this report Three medical service categories were identified: inpatient facility, outpatient facility, and professional procedures. These categories divide into three medical subservice categories: acute inpatient admissions, which included labor and delivery (LD), medical, mental health and substance use (MH/SU), newborn, and surgery admissions; outpatient visits; and outpatient other services. These further divide into “detailed service” categories.5 Prescription service, subservice, detailed service categories, and therapeutic subclasses in this report HCCI analyzed prescription claims from retail and mail order pharmacies. The prescription service category divides into brand and generic subservice categories. These subservices further divide into “detailed service” categories, which divide into therapeutic subclasses .3 (See the methodology document for more details.) In this report, HCCI also examined two therapeutic subclasses of prescriptions – as defined by the American Hospital Formulary Service (AHFS) – antidiabetic agents and contraceptives .5,6 betes. This trend for the younger age dollars on acute admissions and with $1 spent out of pocket (Appendix groups was more pronounced for out- 25.7% on outpatient services, while Tables A4 and A21). Higher total of-pocket per capita spending (Table people without diabetes spent 6.1% spending and out-of-pocket spending 2 and Appendix Table A19). In 2014, and 28.5% of their out-of-pocket dol- on cardiovascular agents and central out-of-pocket per capita spending for lars on those services. nervous system (CNS) agents for peo- children with diabetes was highest of ple with diabetes were also observed. However, people with diabetes spent the five age groups ($2,173), while For people with diabetes, spending proportionally more out-of-pocket on lowest for children without diabetes per capita was 7.7 times higher on prescription services per capita ($466). cardiovascular drugs and 2.4 times (32.4%) compared to people without higher on CNS agents than spending Shares of total per capita spending by diabetes (18.9%; Table 2 and Appen- for people without diabetes. The rates categories of services also differed dix Table A19). At same time, people of use of these prescriptions help ex- between the two populations. For without diabetes spent proportionally plain these spending trends. People people with diabetes, the share of more out of pocket on professional with diabetes used 7.5 times more total spending for each service cate- services (46.4% of their out-of-pocket filled days of cardiovascular drugs gory ranged between 24% and 26%, a spending) compared to people with and 2.6 times more CNS agents than fairly even spending distribution diabetes (35.2%). The largest differ- people without diabetes (Appendix across types of services (Table 1). In ence in per capita spending and out-of Tables A37 and A38). contrast, per capita spending for peo- -pocket spending between the two ple without diabetes was highest for populations was for prescriptions. Notable Trends professional services (35.4% of their The largest component of this pre- Use of antidiabetic agents: For chil- total spending) and lowest for pre- scription spending difference was due dren and young adults with diabetes, scriptions (16.9%; Appendix Table to antidiabetic agents. In 2014, for the most commonly used prescription A2). For out-of-pocket per capita people with diabetes, per capita class was brand antidiabetic agents spending, the out-of-pocket dollars spending on antidiabetic agents (Appendix Tables A43 and A45; see spent on acute admissions and outpa- (brand and generic agents) was “Service Categories Analyzed in this tient visits were quite similar across $1,531, with $182 of that spent out of Report”). This class includes medica- the two populations (Table 2 and Ap- pocket (Appendix Tables A3 and tions commonly taken to manage pendix Table A19). People with diabe- A20), compared to per capita spend- blood glucose levels in people with tes spent 6.6% of their out-of-pocket ing for people without diabetes of $6, diabetes, including, insulin, metfor- www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report v THE PREVALENCE OF DIAGNOSED DIABETES IN THE HCCI POPULATION In 2014, we identified 5.0% of the individuals younger than age 65 and covered by ESI as having diagnosed diabetes. From 2012 to 2014, the percentage of those with diabetes remained constant at 5.0% (Appendix Table A1). The percentage of insureds with diabetes in 2012 and 2013 identified in this report is now slightly lower than the population HCCI reported in Per Capita Health Care Spending on Diabetes: 2009-2013.1 In that prior report, 5.3% of insureds were identi- fied as having diagnosed diabetes in both years. This downward adjustment in the membership trend of insureds identified as having diagnosed diabetes (about 0.3%) was the result of changes made to the analytic methodology and the diabetes flag.5 The newly reported number for 2012 and 2013 is closer to other estimates of the prevalence of diabetes among the privately insured population. For instance, the ADA estimated a rate of 4.6% in 2012 for the privately insured population. 3 This small difference in prevalence rates (0.4%) may be attributable to differences in population and study methodology. In this report, we also identified 6.6% of all adults and 0.3% of all children in the dataset as having diagnosed diabetes. The percentage of adults and children with diabetes remained constant over the study period. min, and oral hypoglycemics.4 In the most commonly used prescription was 1 LD admission per 1,000 chil- 2014, there were 255,481 filled days class was generic cardiovascular dren without diabetes (Appendix Ta- of brand antidiabetic agents per 1,000 agents (Appendix Tables A47, A49, ble A43), 18 LD admissions per 1,000 children and 172,369 filled days per A51). For middle-age adults and pre- young adults without diabetes 1,000 young adults. For the three old- Medicare adults, the second-most (Appendix Table A45), and 32 LD ad- est age groups, use of generic antidia- used prescription class was generic missions per 1,000 intermediate betic agents was more common than antidiabetic agents. However, for in- adults without diabetes (Appendix use of brand antidiabetic agents termediate adults, the second-most Table A47). For intermediate adults, (Appendix Tables A47, A49, A51). For used prescription class was generic there were more LD admissions per example, for intermediate adults, central nervous system (CNS) agents. 1,000 adults for the population with- there were 96,730 filled days of brand out diabetes (32) than for the popula- LD hospital admissions: For the study antidiabetic agents per 1,000 adults tion with diabetes (25). However, this population without diabetes, the use and 116,843 filled days of generic pattern is reversed for young adults. rate of labor and delivery (LD) hospi- antidiabetics per 1,000. For young adults with diabetes, there tal admissions increased with age was a higher use rate of LD (28 ad- Use of generic cardiovascular drugs by over the first three age groups missions per 1,000) than for the pop- oldest adults: For the three oldest age (children, young adults, and interme- ulation of young adults without diabe- groups (intermediate adults, middle- diate adults; Exhibit 1). In 2014, in the tes (18 admissions per 1,000) or for age adults, and pre-Medicare adults), population without diabetes, there Exhibit 1: Utilization of Labor and Delivery Hospital Admissions by Children, Young Adults, and Intermediate Adults with and without Diabetes (2014) Labor and Delivery Admissions per 1,000 Insureds in 2014 Age Group With Diabetes Without Diabetes Children (ages 0-18) 1 1 Young Adults (ages 19-25) 28 18 Intermediate Adults (ages 26-44) 25 32 Source: HCCI 2016. www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report vi the intermediate adults with diabetes rate of MH/SU use was for children, rate of use of outpatient services, spe- (25 admissions per 1,000). We do not who had 21 MH/SU admissions per cifically outpatient-other services; have an explanation for these differ- 1,000 children with diabetes and, after young adults, the second- ences. (Appendix Table A43). And for chil- highest use rate of acute hospital ad- dren, these admissions increased missions, specifically medical hospital MH/SU hospital admissions for young 23.5% over the study period. admissions (195 acute admissions per adults: In 2014, there were 37 mental 1,000 children in 2014). health and substance use (MH/SU) Spending and use trends for children: hospital admissions per 1,000 young Compared to most of the other age adults with diabetes (Appendix Table groups, children had high per capita A45). This this use rate increased spending, high out-of-pocket spend- from 22 MH/SU admits per 1,000 in ing, and high rates of use of health 2012, a 68.2% increase. MH/SU ad- services (Tables 2 and 3, Appendix missions had the second highest use Table A41). In all years studied, chil- rate of any type of admission for dren had the second highest spending young adults. Comparatively, this was per capita; the highest out-of-pocket the highest rate of MH/SU use of any spending per capita in the last year of age group studied. The next highest the study period (2014); the highest TRENDS IN SPENDING AND UTILIZATION OF HEALTH CARE SERVICES FOR PEOPLE WITHOUT DIABETES During the study period (2012-2014), per capita spending for people without diagnosed diabetes was over $10,000 lower than per capita spending for people with diagnosed diabetes (Table 1 and Appendix Table A2). For example, in 2014, spend- ing for people without diabetes was $4,396 per capita, while spending for people with diabetes was $16,021 per capita. In all years studied, spending for those without diabetes was about one quarter of the spending for people with diabetes. This dif- ference in spending was due largely to much lower rates of utilization of services by people without diabetes as compared to the service use of people with diabetes. For people without diabetes, per capita spending between 2012 and 2014 increased from $4,138 to $4,396 (Appendix Table A2). During this period, spending grew about 3% per year. In 2014, the largest share of per capita spending for people with- out diabetes (35.4%) was on professional services ($1,556). Per capita spending on outpatient services was 28.5% of total spending in 2014 ($1,255). Of that per capita amount on outpatient services, 63.7% was on outpatient visits ($799), while the rest was spending on outpatient-other services ($456). Per capita spending on acute inpatient services accounted for 18.9% of total spending ($832), leaving 16.9% of spending on prescriptions. Of the $745 per capita spent on prescriptions in 2014, $264 was spent on generic prescriptions, while $481 was spent on brand prescriptions. In 2014, out-of-pocket spending per capita for people without diabetes was $752 (Appendix Table A19), increasing by $43 per person over the study years. The largest share of out-of-pocket spending (46.4%) was on professional services ($349 per capita). Out-of-pocket spending on prescriptions ($142 per capita) and outpatient services ($214 per capita) accounted for 18.9% and 28.5% of total out-of-pocket spending, respectively. The smallest share of out-of-pocket spending (6.1%) was on acute inpatient services ($46 per capita). Between 2012 and 2014, for people without diabetes, the use of acute inpatient services, outpatient services, and brand pre- scriptions declined each year. During that period, the number of acute inpatient admissions declined from 50 per 1,000 peo- ple without diabetes in 2012 to 47 in 2014 (Appendix Table A36). Similarly, the use of outpatient services declined by about 1.5% per year. This decline was observed for both outpatient visits (a 8 visit per 1,000 person decline) and outpatient-other services (a 66 services per 1,000 person decline). The largest declines in use were for brand prescriptions, for which the number of filled days per 1,000 people declined by more than 15% per year during the study period – from 44,287 filled days per 1,000 people without diabetes in 2012 to 30,361 filled days in 2014. Use of professional services rose slightly between 2012 and 2013 (by 57 services per 1,000 people). The following year, use of these services declined by 175 services to 15,152 professional services per 1,000 people without diabetes. Only use of generic prescriptions increased during the study period: use of generic prescriptions increased from 189,305 filled days per 1,000 people to 202,808 filled days. www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 1 Spending Per Capita for Insureds with Diabetes In 2014, total per capita spending (payer spending plus out-of-pocket spending) was $16,021 on health care for insureds under 65 with diagnosed diabetes (Table 1). Over the study period (2012-2014), per capita spending increased every year, rising at average annual rate of 4.3%. Spending per capita increased from 2012 to 2013 by $413 and from 2013 to 2014 by an additional $897. Proportionally, per capita spending for insureds with diabetes was dis- tributed fairly evenly across the ma- jor service categories (see “Service Categories Analyzed in this Report”). Of total per capita spending in 2014, the largest share, 26.0%, was on out- patient care (Table 1 and Figure 1). The remaining shares were split spending on outpatient services was room (ER) visits ($715), and observa- across acute inpatient admissions on outpatient visits. For example, in tion visits ($143; Appendix Table A3). (24.8% of total spending), prescrip- 2014, per capita spending for in- The fastest growth rate in spending tions (24.8%), and professional ser- sureds with diabetes was $2,254 on during the study period among outpa- vices (24.0%). outpatient visits, compared with tient visits was for ER visits, which For each of the service categories, per spending on outpatient-other ser- grew at an average annual rate of capita spending rose every year of the vices, at $1,918. 10.0% (a $123 increase). study period. Spending on prescrip- In 2014, outpatient visits accounted The remaining 46.0% of per capita tions grew at the fastest average an- for 54.0% of spending on outpatient spending on outpatient services was nual rate over the study period services (Table 1). Per capita spend- on outpatient-other services ($1,918; (8.7%), followed by growth in spend- ing on outpatient visits grew from Table 1). From 2012 to 2013, per cap- ing for outpatient services (5.0%), professional services (2.2%), and For people with diabetes, spending per capita on outpatient an- acute inpatient care (2.2%). cillary services was 6.5 times higher than for people without dia- Outpatient Spending betes (Appendix Tables A3 and A4). Per capita spending for the study pop- ulation with diabetes was higher on 2012 to 2013 by $102, and from 2013 ita spending on outpatient-other ser- outpatient services than on any other to 2014 by an additional $157 (Table vices increased by $62 and then grew service category. In 2014, per capita 1). Of the dollars spent on outpatient from 2013 to 2014 by an additional spending on outpatient services was visits in 2014, insureds with diabetes $63. Of total spending per capita on $4,172 (Table 1). The majority of spent the most on outpatient surgery outpatient-other services, the largest ($1,396), followed by emergency per capita amount was for miscellane- www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 2 ous services ($820; Appendix Table For people without diabetes, spending on professional services A3). This was followed by spending made up 35.4% of their total per capita health care spending, on radiology services ($475), ancil- while it made up just 23.9% of the spending for people with dia- lary services ($417), and laboratory and pathology services ($207). betes (Table 1 and Appendix Table A2). Acute Inpatient Spending professional services ($655), and ad- just 9.8% of brand spending.8 Spend- In 2014, per capita spending for ministered drugs ($555; Appendix ing on brand anti-infective agents was individuals with diabetes was $3,975 Table A3). the third highest spending detailed on acute inpatient services (Table 1). class of brand prescriptions in 2014 Prescription Spending From 2012 to 2013, per capita spend- ($206). Spending on these prescrip- ing on acute inpatient admissions in- In 2014, per capita spending on all tions also rose over the study period – creased by $51 dollars – the smallest prescriptions was $3,970 (Table 1). from 2012 to 2013, by just $1, and by increase for a service category that Spending on prescriptions grew at the $99 from 2013 to 2014. year. From 2013 to 2014, spending on fastest average annual rate of any Spending on generic prescriptions acute inpatient admissions increased service category during the study pe- was $1,138 in 2014 (Table 1). Of that by an additional $121. Of the total riod (8.7%). Of the per capita spend- amount, the most dollars were spent dollars per capita spent in 2014 on ing on prescriptions in 2014, 71.1% on cardiovascular drugs ($238) and acute inpatient admissions, the larg- was on brand prescriptions and central nervous system (CNS) agents est share was on surgical admissions 28.6% was on generic prescriptions.7 ($215; Appendix Table A3). Generic (57.5%, $2,287 per capita) (Appendix From 2012 to 2013, per capita spend- antidiabetic agents made up just Table A3). Most of the rest that ing on brand prescriptions grew by 10.5% of total spending on generic spending was on medical admissions $77 and from 2013 to 2014, grew by prescriptions ($119), and fell by $19 (39.6%), $1,575 per capita. an additional $395. Over the study from the previous year. Between period, spending on generic prescrip- 2013 and 2014, spending on generic Per capita spending on mental health and substance use admis- prescriptions increased by $87. The sions was over 4.5 times higher for people with diabetes than for biggest contributor to this growth was an increase in spending on gener- those without diabetes (Appendix Tables A3 and A4). ic CNS agents, which rose $36 dollars. tions increased by a smaller amount Per Capita Spending by Gender Professional Services Spending than did spending on brand prescrip- Women with diabetes had higher In 2014, per capita spending on pro- tions. Between 2012 and 2014, spending per capita than men with fessional services was $3,832 (Table spending on generic prescriptions diabetes in every year of the study 1). This was the service category with increased by $125. period. For example, in 2014, per cap- the lowest per capita spending, alt- In 2014, spending on brand prescrip- ita spending for women was $16,409, hough spending on professional ser- tions was $2,829 per capita (Tables compared with $15,688 for men vices was just $340 per capita lower 1). Of that amount, 50.0% was for (Table 1). The difference between per than on the highest spending service brand antidiabetic agents ($1,412; capita spending for men and women category (outpatient services). From Appendix Table A3). Between 2012 narrowed over the study period— 2012 to 2013, per capita spending on and 2013, spending on brand antidia- from $951 to $721. This reduction in professional services increased by betic agents grew by $88, and by ad- the spending difference between men $85, and from 2013 to 2014, grew by ditional $272 the following year. Both and women was due to a larger in- an additional $76. This growth in of these increases were the largest crease in spending over the study pe- 2014 was the smallest increase in increases for detailed classes of brand riod for men ($1,420) than for women spending observed for any category in prescriptions in their respective ($1,190). that year. Of the total spending in years. The next highest spending on a 2014 on professional services, the In 2014, per capita spending on acute detailed class of brand prescriptions largest amount was on miscellaneous inpatient admissions and prescrip- was on cardiovascular drugs ($276), services ($982), followed by surgical tions for men was $376 higher than www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 3 SHARE OF PER CAPITA SPENDING ON MEDICAL SERVICE CATEGORIES FOR INSUREDS WITH AND WITHOUT DIABETES Compared to individuals with diagnosed diabetes, per capita spending for people that were not diagnosed with diabetes was much lower. Spending for people without diabetes in 2014 was $4,396 per capita (Appendix Table A2), compared to $16,021 per capita for people with diabetes (Table 1 and Figure 2). The distribution of per capita spending across health care service categories was different for the populations with and without diabetes. The largest share of the total per capita spending for individuals without diagnosed diabetes in 2014 was on professional services (35.4%, $1,556 per capita; Appendix Table A2). Comparatively, for those with diagnosed diabetes, professional ser- vices accounted for the smallest share of their spending per capita (24.0% or $3,832; Table 1). For both populations, the shares of spending on professional services were distributed similarly across the professional services detailed categories. However, we observed a difference in the share of dollars spent on visits to the doctor. For people with diabetes, 18.4% ($705) of their per capita spending on professional services was on visits to the doctor, compared to 22.0% ($342) for those without diabetes (Appendix Tables A3 and A4). The largest share of spending for people with diabetes was on outpatient services (26.0%, $4,172 per capita), while outpa- tient services was the second highest-spending share for people without diabetes (28.5%, $1,255 per capita); Table 1 and Appendix Table A2). People with diabetes spent a smaller share of their outpatient dollars on outpatient surgery and emer- gency room visits than people without diabetes. Those with diabetes spent 50.6% of their outpatient dollars on emergency room visits and outpatient surgery. Comparatively, those without diabetes spent a larger share, 60.8% on emergency room visits and outpatient surgery (Appendix Tables A3 and A4). People without diabetes had a larger share of their outpatient spending on outpatient surgery, while those with diabetes had higher spending in terms of dollars per capita on outpatient surgery ($1,396) compared to people without diabetes ($480). Figure 2 Spending Per Capita and Change in Spending for Insureds With and Without Diabetes: 2012-2014 $16,021 $14,711 $15,124 5.9% 3.3% 3.2% 3.0% 2.7% 2.8% $4,138 $4,262 $4,396 2012 2013 2014 Diabetes No Diabetes Source: HCCI, 2016. Notes: All data weighted to reflect the national, younger than 65 ESI population. Data from 2013 and 2014 adjusted using actuarial completion. www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 4 for women. (Appendix Table A5). Ad- capita spending for children in- acute inpatient admissions, as was ditionally, spending on inpatient ad- creased by a larger amount than did also true for pre-Medicare adults missions for men accounted for the spending for pre-Medicare adults: a (Figure 3). In 2014, young adults had largest share of spending for men $2,337 increase for children and a the highest spending per capita on (26.4%, or $4,149 per man) but the $1,374 increase for pre-Medicare acute admissions ($4,608 per young smallest share of spending for women adults. From 2012 to 2013, per capita adult) of any age group (Appendix (23%, or $3,773 per woman). Per cap- spending for children grew by $1,158, Table A7). For young adults, the high- ita spending for men on prescriptions and from 2013 to 2014, grew by an est spending for acute admissions in was $142 higher than prescription additional $1,179. The largest amount 2014 was on medical admissions spending for women. This higher for children was spent on prescrip- ($2,734 per capita), followed by surgi- spending for men was due to higher tions ($6,286 per child; Appendix Ta- cal admissions ($1,217), labor and spending on brand prescriptions. In ble A 7 and Figure 3). This spending deliver (LD) admissions ($305), and 2014, spending on brand prescrip- on prescriptions was higher than for mental health and substance use tions was $2,953 per man compared any other age group. Spending per (MH/SU) admissions ($352; Appendix to $2,686 per woman. However, capita for children on prescriptions Table A11). spending for women on generic pre- was $1,866 per capita higher than for The second highest spending for scriptions ($1,205) was higher than pre-Medicare adults, the next highest young adults was on prescriptions for men ($1,081). spending age group. Of the amount ($4,074; Appendix Table A7 and Fig- per capita for children, most was on In 2014, per capita spending for wom- ure 3). Of this amount, nearly 60% brand antidiabetic agents: $3,301 per en was higher than for men, largely was on brand antidiabetic agents child in 2014. due to higher spending on outpatient ($2,332 per capita; Appendix Table services ($417 more for women) and Conversely, compared to the other A11). While spending per capita on professional services ($808 more for age groups in 2014, children had the outpatient services was comparative- women; Appendix Table A5). For lowest per capita spending on profes- ly lower than spending on acute ad- women, the highest spending catego- sional services ($2,771; Appendix missions and prescriptions, a compar- ries of health care services was on Table A7 and Figure 3). The most pro- atively high amount was spent per outpatient services and professional fessional dollars were spent on visits capita on two types of outpatient ser- services. Outpatient services repre- to the doctor ($834 per child) and vices (Appendix Table A7). For young sented 26.8%, or $4,396, of per capita professional surgical services ($194 adults, $1,253 was spent per capita on spending for women, while profes- per child; Appendix Table A9). Chil- ER visits and $1,143 was spent on sional services made up 26.0%, or dren also had the lowest spending of ancillary services (Appendix Table $4,226, of their spending. any age group on outpatient visits A11). ($1,350 per child). Of this amount, Per Capita Spending by Age Group Intermediate Adults 57.4% was on ER visits ($775 per Over the study period, per capita child) and 33.7% were on outpatient In 2014, intermediate adults had the spending was highest for pre- surgical visits ($455 per child). lowest per capita spending of all of Medicare adults (ages 55-64) fol- the age groups: $13,141 (Table 1). Young Adults lowed by children (ages 0-18), young Similar to middle-age adults, interme- adults (ages 19-25), and middle-age In 2014, per capita spending for diate adults spent the most on outpa- adults (ages 45-54); intermediate young adults was $15,839 (Table 1). tient services ($3,670 per capita; Ap- adults (ages 26-44) had the lowest Between 2012 and 2013, spending for pendix Table A7 and Figure 3). Outpa- spending (Table 1). young adults increased by $517 per tient surgical visits ($1,005 per capi- capita. The following year, spending ta) and ER visits ($971 per capita) Children for young adults increased by an addi- accounted for the most of these Children had the second highest per tional $1,563. This growth was the adults’ spending on outpatient ser- capita spending in 2014 ($17,380), largest increase in spending observed vices (Appendix Table A13). Interest- which was $462 per capita lower than for that year for any age group. The ingly, intermediate adults had the spending for pre-Medicare adults largest portion of the per capita lowest spending of any age group on (Table 1). Over the study period, per spending for young adults was on outpatient-other services. One factor www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 5 young adults ($1,143; Appendix Table A11). Middle-age adults were the only age group where acute inpatient admis- sions was the lowest spending catego- ry of services ($3,492 per capita; Ap- pendix Table A7 and Figure 3). One factor driving this appears to be the comparatively low spending on medi- cal admissions ($1,364 per capita) compared to the other age groups (Appendix Table A15). Pre-Medicare Adults As with pre-Medicare adults general- ly,9 pre-Medicare adults with diabetes had the highest per capita spending in every year of the study period. In 2014, per capita spending for pre- Medicare adults was $17,842 (Table driving this was the comparatively for every age group) was lower 1). From 2012 to 2013, spending for low per capita spending on outpatient ($1,286 per intermediate adult; Ap- pre-Medicare adults rose $429 per lab/path services ($178 per interme- pendix Table A13). capita; from 2013 to 2014, spending diate adult). increased by an additional $945. Of Middle-Age Adults The second most per capita spending the total per capita spending for pre- In 2014, per capita spending for mid- for intermediate adults was on pro- Medicare adults in 2014, the largest dle-age adults ($14,755) was the sec- fessional services ($3,259), followed amount was spent on acute inpatient ond lowest of any age group (Table by acute inpatient admissions admissions ($4,563; Appendix Table 1). Middle-age adults also had the ($3,127; Appendix Table A7). The A7 and Figure 3). Pre-Medicare adults lowest per capita spending increase most acute inpatient dollars were and young adults were the only two ($1,145) over the study period. Be- spent on surgical admissions ($1,468 age groups that had higher spending tween 2012 and 2013, per capita per capita) followed by medical ad- on acute admissions than any other spending for these adults grew by missions ($1,308; Appendix Table service category. Pre-Medicare adults $397, and spending grew by an addi- A13). Compared to spending for inter- also had the highest per capita spend- tional $748 the following year. As mediate adults, spending for young ing on surgical admissions of any age with intermediate adults, for middle- adults on LD admissions and MH/SU group ($2,799; Appendix Table A17). age adults, the service category with admissions was 11.3% and 363.0% Interestingly, this group had the low- the highest spending in 2014 was out- higher, respectively (Appendix Tables est per capita spending on MH/SU patient services ($3,951 per capita; A11 and A13). admissions ($35). After acute admis- Appendix Table A7 and Figure 3). sions, for pre-Medicare adults, the The lowest spending per capita was Within outpatient services, the high- second largest amount per capita was on prescriptions ($3,060; Appendix est per capita spending was on outpa- spent on outpatient services ($4,501), Table A7). The intermediate adult tient surgical services ($1,323) and followed by prescriptions ($4,420), group was the only age group for ER visits ($734; Appendix Table A15). and professional services ($4,251; which prescriptions was the lowest Compared to the other age groups, Appendix Table A7). spending service category. Compared middle-age adults had the lowest per to the other age groups, spending for capita spending on ancillary services intermediate adults on brand antidia- ($349), a comparatively higher betic agents (the prescription class spending category for children with the highest per capita spending ($1,932; Appendix Table A9) and www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 6 www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 7 Out-of-Pocket Spending Per Capita for Insureds with Diabetes In 2014, people with diabetes spent ing per capita was smaller from 2013 to vices. Out-of-pocket per capita spending $1,944 out of pocket per capita (Table 2014 than from 2012 to 2013. The on outpatient services rose by $40 from 2). Of total per capita spending on health smaller increase was due largely to de- 2012 to 2013, and then grew by an addi- care in 2014 ($16,021), 12.1% was clining per capita out-of-pocket spend- tional $34 from 2013 to 2014. Out-of- spent out of pocket by people with dia- ing on prescriptions and acute inpatient pocket per capita spending on profes- betes. Overall, per capita out-of-pocket admissions, which offset increasing out- sional services also rose during the spending grew more from 2012 to 2013 of-pocket spending on outpatient ser- study period, growing by $41 from 2012 ($66, or 3.6%) than it did from 2013 to vices. Per capita spending out-of-pocket to 2013, and by an additional $19 from 2014 ($10, or 0.5%; Figure 4). This is on prescriptions declined both years of 2013 to 2014. opposite of the trend for total per capita the study period, first by 8.9%, or $22, Professional Services Out-of-Pocket spending, which rose more between from 2012 to 2013, and then by an addi- 2013 and 2014 than between 2012 and tional 6.8%, or $37 from 2013 to 2014.In 2014, among the service categories, 2013. Out-of-pocket spending per capita on out-of-pocket spending per capita was acute inpatient admissions rose by $9 highest for professional services ($684; Of the per capita out-of-pocket dollars Table 2). Of the out-of-pocket spending dollars from 2012 to 2013, but fell by $6 spent in 2014, insureds with diabetes from 2013 to 2014. on professional services in 2014, the spent the most on professional services largest share (31.7%) was spent on doc- ($684; Table 2). After professional ser- Overall, total out-of-pocket spending tor visits ($201 per capita; Appendix vices, out-of-pocket spending per capita rose in all years studied (Table 2). The Table A20). This was followed by out-of- was highest on prescriptions ($629), increases in out-of-pocket spending pocket spending on professional surgi- followed by spending on outpatient ser- were due largely to increases in per cap- cal services ($90 per capita) and profes- vices ($500). The smallest amount of per ita spending on outpatient and profes- sional laboratory and pathology (lab/ capita out-of-pocket spending was on sional services. The increase in out-of- path) services ($85 per capita). acute inpatient services ($128). The pocket spending was larger for outpa- overall increase in out-of-pocket spend- tient services than for professional ser- Between 2012 and 2013, out-of-pocket spending on doctor visits rose from $201 to $217 (Appendix Table A20). Between 2013 and 2014, however, there was no increase in out-of-pocket per capita spending on doctor visits. This flat growth led to a smaller increase in total out-of-pocket spending on all pro- fessional services in 2014 ($19) than occurred in the previous year ($41; Ta- ble 2). Prescriptions Out-of-Pocket From 2012 to 2013, per capita out-of- pocket spending on prescriptions fell by $22 dollars, and fell by an additional $37 dollars between 2013 and 2014 (Table 2). In 2014, out-of-pocket spending on all prescriptions was $629 per capita. Of that amount, $297 was spent on brand prescriptions and $331 was spent on www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 8 generic prescriptions. Brand prescrip- diabetic agents ($39; Appendix Table pocket spending on ER visits grew by tion out-of-pocket spending per capi- A20). Total out-of-pocket spending on $13 per capita (59.1% of the total ta declined each year studied, leading generic prescriptions rose between increase in outpatient visit spending). to the decline in total prescription out 2012 and 2013 due to a small in- And from 2013 to 2014, it grew by an -of-pocket spending. Conversely, per crease in out-of-pocket spending on additional $14 dollars per capita capita spending on generic prescrip- anti-diabetic agents ($5 per capita (82.4% of the total increase in outpa- tions increased from 2012 to 2013 by increase). However, from 2013 to tient visit spending). $9, but fell by $16 between 2013 and 2014, a decline in out-of-pocket In 2014, outpatient-other services 2014. spending on generic cardiovascular accounted for 49.6% ($248 per capi- Per capita out-of-pocket spending on brand prescriptions was 6 ta) of out-of-pocket spending on all outpatient services (Table 2). Of these times higher for people with diabetes than for those without dia- dollars, out-of-pocket spending was betes (Table 2 and Appendix Table A19). highest for ancillary services ($77 per capita), followed by outpatient radiol- drugs ($5 per capita decrease) and In 2014, out-of-pocket spending on ogy services ($65; Appendix Table anti-diabetic agents ($6 per capita brand prescriptions was $297 per A20). Out-of-pocket spending on an- decrease) drove declining total out-of capita or 47.2% of out-of-pocket cillary services grew the most of any -pocket spending on generic prescrip- spending per capita on all prescrip- outpatient-other detailed category tions. tions (Table 2). Of the out-of-pocket over the study period. From 2012 to spending on brand prescriptions, the Outpatient Out-of-Pocket 2013, out-of-pocket spending on an- largest share (22.7%) was on brand cillary services grew by $5 per capita, In 2014, out-of-pocket spending per anti-diabetic agents ($143 per capita; and from 2013 to 2014 it grew by an capita on outpatient services was Appendix Table A20). The only other additional $5. $500 (Table 2). These dollars were detailed category that had higher out- almost evenly divided between outpa- Acute Inpatient Out-of-Pocket of-pocket spending was visits to the tient visits ($253 per capita) and out- doctor. After brand anti-diabetic Out-of-pocket spending per capita on patient-other services ($248). Overall agents, insureds with diabetes spent acute inpatient admissions was the out-of-pocket spending on all health the most brand prescription out-of- lowest spending for any service cate- services grew from 2013 to 2014 be- pocket dollars on brand cardiovascu- gory over the study period. Out-of- cause of an increase in out-of-pocket lar drugs ($55 per capita). The decline pocket spending on acute inpatient spending on outpatient services. Be- in out-of-pocket spending over the admissions was $128 per capita in tween 2013 and 2014, out-of-pocket study period on brand prescriptions 2014 (Table 2). Of the dollars spent spending on outpatient visits rose was driven largely by declining out-of out-of-pocket on inpatient admissions $17, and rose by $18 on outpatient- -pocket spending on cardiovascular in 2014, the most were spent on med- other services. drugs. Out-of-pocket spending on ical admissions ($71 per capita; Ap- brand cardiovascular drugs fell every In 2014, outpatient visits accounted pendix Table A20). After medical ad- year of the study period, declining by for 50.6% ($253 per capita) of the out missions, out-of-pocket spending was a total of $37 dollars between 2012 -of-pocket spending on outpatient highest on surgical admissions ($47 and 2014. services (Table 2). Of the out-of- per capita). Out-of-pocket per capita pocket dollars spent on outpatient spending on acute admissions fell by In 2014, $331 per capita was spent visits, spending was highest on outpa- $6 between 2013 and 2014. This de- out of pocket on generic prescrip- tient surgical visits ($132 per capita), cline was largely the result of falling tions, which made up 52.6% of out-of- followed by ER visits ($111), and ob- out-of-pocket spending on medical pocket spending on all prescriptions servation ($9; Appendix Table A20). admissions (fell by $3 per capita) and (Table 2). Of this amount, the most The $39 growth in out-of-pocket surgical admissions (fell by $3). was spent on generic cardiovascular spending on outpatient visits over the drugs ($96 per capita), followed by Out-of-Pocket Spending by Gender study period was largely the result of generic central nervous system (CNS) As with the total health care spending rising out-of-pocket spending on ER agents ($58), and generic anti- per capita trends, women spent more visits. From 2012 to 2013, out-of- www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 9 OUT-OF-POCKET SPENDING ACCOUNTED FOR A SMALLER SHARE OF TOTAL SPENDING FOR THOSE WITH DIA- BETES Each year of the study period, insureds with diabetes spent, per capita, two and half times more out-of-pocket than insureds without diabetes (Table 2 and Appendix Table A19). For example, in 2014, insureds with diabetes spent $1,944 per capita out of pocket, compared to $752 per capita spent out of pocket by insureds without diabetes (Figure 5). However, the share of spending paid out of pocket by those with diabetes was smaller than the share of total spending by those without diabetes. In 2014, for those with diabetes, out-of-pocket spending accounted for 12.1% of total per capita spending (payer spending plus out-of-pocket spending), compared with 17.1% for those without diabetes (Appendix Table A19), but of a much larger total dollar amount (Table 2). These trends were also observed in 2012 and 2013. Figure 5 Out-of-Pocket Spending Per Capita for Insureds With and Without Diabetes: 2012-2014 $1,934 $1,944 $1,868 $709 $733 $752 2012 2013 2014 Diabetes No Diabetes Source: HCCI, 2016. Notes: All data weighted to reflect the national, younger than 65 ESI population. Data from 2012 and 2013 actuarially completed. For both groups, those with and without diabetes, the largest share of their out-of-pocket spending was on professional ser- vices. For people with diabetes, in 2014, 35.2% of their out-of-pocket spending was on professional services, while for people without diabetes this share was 46.4%. Of the detailed categories of professional services, the most out-of-pocket spending for both populations was on visits to the doctor: $217 per person with diabetes and $108 per person without diabetes (Appendix Tables A20 and A21). Proportionally, this spending was nearly even across the two groups. Visits to the doctor made up 31.7% of the spending of professional services for people with diabetes, compared with 30.9% of the professional spending for people without diabetes. As a share of total spending, out-of-pocket spending on outpatient services for people with and without diabetes was similar (25.7% and 28.5%, respectively; Appendix Tables A2 and A19). However, the types of outpatient services on which they spent their dollars differed between the two populations. Nearly 60% of out-of-pocket outpatient spending for people with- out diabetes was on ER visits and outpatient surgical visits. In contrast, less than 50% of the outpatient spending for people with diabetes was on ER visits and outpatient surgical visits. Also, per capita out-of-pocket spending for people with diabetes on outpatient ancillary services was 15.4%, compared with 6.5% for people without diabetes. www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 10 out of pocket per capita than did men for men on brand prescriptions was pocket spending was higher for chil- in every year of the study period. For higher than this spending for women. dren than for pre-Medicare adults example, in 2014 women spent In 2014, men spent $311 per capita because the spending for children $2,060 out of pocket, while men spent out of pocket on brand prescriptions, increased by 5.7%, while the spend- $1,845 out-of-pocket (Table 2). Over while women spent $280. For both ing for pre-Medicare adults declined the study period, out-of-pocket men and women, out-of-pocket by 1.0%. Out-of-pocket spending per spending for women grew by more spending on brand prescriptions de- capita for children grew by the most dollars (rose by $81 per woman) than clined across the study period. Be- dollars over the entire study period out-of-pocket spending for men (rose tween 2012 and 2014, spending out ($256 per child), while out-of-pocket by $75 per man). of pocket for men fell by $54 per man spending grew the least for pre- and by $52 per woman. Conversely, Medicare adults ($38 per adult). Women had higher out-of-pocket per out-of-pocket per capita spending on capita spending on every service cate- In every year of the study period, the generic prescriptions rose for both gory than did men. In 2014, the larg- lowest per capita out-of-pocket men and women rose between 2012 est differences between men and spending was by intermediate adults and 2013, by $10 per man and $8 per women’s out-of-pocket spending (Table 2). In 2014, intermediate woman. However, this out-of-pocket were observed for professional ser- adults spent $1,791 per capita out of spending fell the following year by vices (a $122 per capita difference) pocket. This was followed by out-of- $15 per man and $17 per woman. and outpatient services (a $85 differ- pocket per capita spending for young ence; Appendix Table A22). For both Additionally, women’s out-of-pocket adults ($1,850) and middle-age adults men and women, more out-of-pocket spending on acute inpatient admis- ($1,869). As with spending for chil- per capita dollars went to profession- sions was quite similar to that by dren, per capita out-of-pocket spend- al services than any other service cat- men. In 2014, women spent $131 per ing for these three age groups in- egory. Per capita out-of-pocket spend- capita on acute admissions while men creased in every year studied. ing on professional services was $750 spent $126, $5 difference (Appendix Children per woman and $628 per man, a $122 Table A22). Similar to the trends for difference. Over the study period, out- out-of-pocket spending on generic Out-of-pocket spending per capita for of-pocket spending per capita on both prescriptions, spending on acute ad- children in 2014 was $2,173, the professional and outpatient services missions rose for both men and wom- highest of any age group (Table 2). In rose for both men and women. For en between 2012 and 2013. Out-of- 2012 and 2013, children had the sec- men, out-of-pocket spending rose by pocket spending on acute admissions ond highest per capita spending after $58 per man on professional services rose by $10 per man and by $9 per pre-Medicare adults. Out-of-pocket and $70 on outpatient services. For woman. The following year, out-of- spending grew over the study period women, out-of-pocket spending rose pocket spending on acute admissions for children, rising $139 from 2012 to by $63 per woman on professional fell for both genders: by $4 per man 2013 and by an additional $117 from services and $78 on outpatient ser- and by $8 per woman. 2013 to 2014. vices. Of the dollars spent out of pocket for Out-of-Pocket Spending by Age Out-of-pocket per capita spending on Group children in 2014, the most were spent prescriptions was nearly equal for on prescriptions ($751 per child; Ap- In 2014, out-of-pocket per capita men and women. Spending out of pendix Table A24). This was the high- spending for children was the highest pocket on prescriptions was $630 per est per capita out-of-pocket spending for any age group ($2,173; Table 2). woman and $627 per man in 2014 on any service category by any age After children, out-of-pocket per capi- (Appendix Table A22). Women only group. Children were the only age ta spending was highest for pre- out spent men by $3 per capita on group for whom out-of-pocket per Medicare adults ($2,046). However, prescriptions. This was also true for capita spending on prescriptions in- 2014 was the only year in the study out-of-pocket spending on generic creased in 2014. For all of the other period for which spending per capita prescriptions, where women out age groups, this out-of-pocket spend- out of pocket for children was higher spent men by $33 per capita. Howev- ing declined by at least 3%. The in- than spending for pre-Medicare er, per capita out-of-pocket spending crease for children was due an in- adults. In 2014, per capita out-of- crease in out-of-pocket per capita www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 11 spending on brand prescriptions, young adult; Appendix Table A28). professional services ($648 in 2014; which increased by $38 per capita to After outpatient services, the most Appendix Table A24 and Figure 6). $433. The most prescription out-of- out-of-pocket per capita dollars were The most out-of-pocket dollars on professional services were spent on Out-of-pocket spending per child was over twice as high on doctor visits to the doctor: $203 per interme- visits and ER visits for children with diabetes as compared to chil- diate adult in 2014 (Appendix Table A30). dren without diabetes (Appendix Tables A26 and A27). After professional services, the most out-of-pocket dollars were for outpa- pocket dollars for children were spent spent on professional services ($539), tient services ($532 per capita; Ap- on brand antidiabetic agents ($350 prescriptions ($478), and acute inpa- pendix Table A24 and Figure 6). Un- per child; Appendix Table A26). After tient admissions ($216; Appendix like all of the other age groups, the prescriptions, out-of-pocket spending Table A24). most outpatient out-of-pocket dollars per capita for children was highest on Unlike all of the other age groups, the spent by intermediate adults were on outpatient services ($703), followed per capita out-out-pocket spending by ER visits. In 2014, out-of-pocket by professional services ($522), and young adults on brand and generic spending on ER visits was $172 per acute inpatient admissions ($196; prescriptions declined each year stud- intermediate adult (Appendix Table Appendix Table A24). After spending ied (Appendix Table A24). Total out- A30). Out-of-pocket spending on out- on brand antidiabetic agents, the of-pocket spending on prescriptions patient services generally, and ER most out-of-pocket dollars for chil- declined from $506 per young adult visits specifically, saw high growth dren were spent on outpatient ancil- in 2012 to $478 in 2014. The most out over the study period. Between 2012 lary services ($395 per child) and -of-pocket dollars on prescriptions and 2014, out-of-pocket spending on visits to the doctor ($228 per child; were spent on brand antidiabetic outpatient services grew $91 per cap- Appendix Table A26). agents ($230 per capita in 2014), a ita and grew $44 on ER visits Young Adults slight increase over the study period (Appendix Tables A24 and A30). This (by $7 per capita; Appendix Table growth in out-of-pocket spending was Out-of-pocket per capita spending for A28). Compared to the other age an important driver in total out-of- young adults in 2014 was $1,850 groups, young adults had higher per pocket spending growth for interme- (Table 2). This was the second lowest capita spending on acute inpatient diate adults over the study period. per capita out-of-pocket spending for admissions in each year of the study any age group. In terms of dollars, out The least out-of-pocket dollars for period (Appendix Table A24). -of-pocket spending by young adults intermediate adults were on prescrip- grew the second most after out-of- Intermediate Adults tions ($464 per capita), and acute pocket spending for children over the inpatient admissions ($146; Appendix Out-of-pocket per capita spending for study period. From 2012 to 2013 out- Table A24). As with the other adult intermediate adults was the lowest of of-pocket spending by young adults age groups, out-of-pocket spending any age group in all years studied. In grew by $103 per capita, and rose by on prescriptions fell over the study 2014, out-of-pocket spending was an additional $66 from 2013 to 2014. period, driven by declining out-of- $1,791 per intermediate adult (Table Of the dollars spent out-of-pocket, the pocket spending on brand prescrip- 2). Out-of-pocket spending rose over most were spent on outpatient ser- tions. the study period, increasing more vices ($615 per young adult; Appen- from 2012 to 2013 ($85 per capita), Middle-Age Adults dix Table A24 and Figure 6). The most than from 2013 to 2014 ($61). Unlike outpatient out-of-pocket dollars were In 2014, middle-age adults spent the younger age groups, intermediate spent on ancillary services ($228 per $1,869 out of pocket per capita (Table adults spent the most per capita on young adult) and ER visits ($202 per 2). Compared to out-of-pocket spend- ing per capita trends for the other age Out-of-pocket spending per young adult on acute inpatient admis- groups, spending out of pocket by sions was nearly five times higher for young adults with diabetes middle-age adults was in the middle. as those without (Appendix Tables A24 and A25). For middle-age adults, out-of-pocket spending grew more from 2012 to www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 12 2013 ($61 per capita), than from spending on all prescriptions for mid- Pre-Medicare adults spent the most 2013 to 2014 ($20). dle-age adults over the study period out-of-pocket per capita on profes- sional services ($720 in 2014), fol- The ratio of out-of-pocket spending for people with diabetes to lowed by prescriptions ($712), outpa- those without diabetes declined with age. The difference in spend- tient services ($487), and acute inpa- ing between people with and without diabetes was largest for chil- tient admissions ($122; Appendix Table A24 and Figure 6). Compared to dren (4.7) and smallest for pre-Medicare adults (1.7; Table 2 and the other age groups, pre-Medicare Appendix Table A21). adults had the highest per capita out- of-pocket spending on professional In 2014, middle-aged adults spent the (Appendix Table A24). services. In contrast, pre-Medicare most of their per capita out-of-pocket adults had the lowest out-of-pocket dollars on professional services Pre-Medicare Adults spending per capita on outpatient ($666), followed by prescriptions In 2014, after children, pre-Medicare services, and comparatively low ($594), outpatient services ($488), adults spent the most out of pocket spending on acute admissions. For and acute inpatient admissions ($119; per capita ($2,046; Table 2). In the pre-Medicare adults, the most out-of- Appendix Table A24 and Figure two previous years, pre-Medicare pocket per capita dollars were spent 6).Compared to the other age groups, adults had the highest spending of on visits to the doctor ($227 per pre- middle-age adults had the lowest out- any age group. Pre-Medicare adults Medicare adult), outpatient surgical of-pocket spending on acute admis- had just the second highest out-of- visits ($147), and brand antidiabetic sion during the study period. Within pocket per capita spending in 2014 agents ($143; Appendix Table A33). these service categories, the most out- because it fell by $20 from 2013 to of-pocket dollars were spent on visits 2014. Over the entire study period, to the doctor ($211 per capita), brand out-of-pocket spending for Pre- antidiabetic agents ($133 per capita), Medicare adults grew by the least and outpatient surgical visits ($129 dollars of any age group. Between per capita; Appendix Table A32). As 2012 and 2014, out-of-pocket spend- was observed for intermediate and ing increased by $38 per pre- pre-Medicare adults, falling spending Medicare adult. out of pocket on brand prescriptions drove the decline in out-of-pocket www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 13 www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 14 Utilization of Health Services for Insureds with Diabetes For people with diabetes, utilization of most medical subservice categories declined over the 2012-2014 study period (Table 3 and Figure 7). The number of acute inpatient admissions, outpatient-other services, and profes- sional services declined every year studied. Only use of outpatient visits increased slightly between 2012 and 2014 (by 9 visits per 1,000 people). Additionally, the use of filled days of brand prescriptions declined over the study period (Table 3). At the same time, the use of generics increased each year. This increase was larger than the decline in use of brand pre- scriptions, leading to a net increase in the use of prescriptions between 2012 and 2014. Use of Professional Services 2012 and 2013 (by 122 services per declined over the study period, from In each year of the study period, the 1,000), before declining slightly the 9,423 outpatient services per 1,000 most services used by people with following year (by 63 services per people 2 to 9,085 services per 1,000 diabetes were professional services. 1,000; Appendix Table A37). Another people. In 2014, there were 36,927 profes- frequently used service for people Of these outpatient services, the vast sional services per 1,000 people with diabetes was visits to the doctor. majority were outpatient-other ser- (Table 3). This number declined In 2014, there were 6,692 doctor vis- vices. In 2014, there were 8,373 out- slightly each year studied. Between its per 1,000 people. The number of patient-other services per 1,000 peo- 2012 and 2014, the number of profes- visits increased slightly in 2013 (by ple (Table 3). The use of these ser- sional services declined by 607 ser- 84 visits per 1,000), and declined vices fell slightly each year studied. vices per 1,000 people. slightly the following year (by 84 vis- The most used type of outpatient- its per 1,000). other service was outpatient labora- Individuals without diabetes had half the number of doctor visits tory and pathology services. In 2014, and half as many ER visits compared to people with diabetes there were 3,458 outpatient lab/path services per 1,000 people, falling (Appendix Tables A37 and A38). slightly from 3,634 services per 1,000 Of the types of professional services, people in 2012 (Appendix Table A37). the one with the highest rate of use Use of Outpatient Unlike the use of outpatient-other was professional laboratory and pa- services, use of outpatient visits in- The second-highest medical service thology services (lab/path). In 2014, creased slightly over the study period. category used was outpatient ser- there were 13,193 lab/path services Use increased by 1 visit per 1,000 vices. In 2014, there were 9,085 out- per 1,000 people. The use of lab/path people in 2013 and by an additional 8 patient services per 1,000 people services increased slightly between visits the following year (to 712 visits (Table 3). Use of outpatient services www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 15 study period by about 5% per year, to 368,607 filled days per 1,000 people in 2014 (Appendix Table A37 and Figure 8). The second most used class of prescriptions was generic antidia- betic agents (171,919 filled days per 1,000 people in 2014), followed by brand antidiabetic agents (107,570 filled days). Combined, brand and generic antidiabetic agents accounted for 22.5% of total prescription use by people with diabetes. Use of generic antidiabetic agents increased each year studied, by 17,013 filled days between 2012 and 2014. Use of brand antidiabetic agents, however, de- clined between 2012 and 2013 (by 8,705 filled days), before increasing in 2014 (by 2,170 filled days). Use by Gender per 1,000 people; Table 3). Use of out- patient visits was split fairly evenly For all the medical subservice catego- Use of Prescriptions between ER visits and outpatient sur- ries, use of services per 1,000 women gery visits. In 2014, there were 329 In 2014, people with diabetes used was higher than use by men ER visits per 1,000 people and 314 1,242,571 filled days of prescriptions (Appendix Table A39). This difference outpatient surgery visits (Appendix per 1,000 people (Table 3). This pre- in service use, expressed as a ratio of Table A37). In comparison, there scription use rate averages to just services used by women to use by were just 69 observation visits per over 3 drugs per person with diabetes men, was largest for outpatient visits 1,000 people. per day. Over 80% of these filled days (1.3) and smallest for acute inpatient of prescriptions were generic. The use admissions (1.1). Use of filled days of Use of Acute Inpatient of generic prescriptions increased generic prescriptions per 1,000 peo- Of the medical subservice categories each year studied. Use increased by ple was also slightly higher for wom- in 2014, acute inpatient admissions 94,244 filled days per 1,000 people, en than men (1.06). Only for filled had the lowest use rate (160 acute from 933,587 days in 2012 to days of brand prescriptions was use admissions per 1,000 people; Table 1,027,831 in 2014. At the same time, higher for men than for women 3). There was a small decline in the the use of brand prescriptions de- (0.92). use of acute admissions over the study period. The number of admis- Compared to people without diabetes, people with diabetes used 5 sions declined by 7 admissions per times more filled days of generic prescriptions and 7 times more 1,000 people between 2012 and filled days of brand prescriptions (Table 3 and Appendix Table 2013, and by an additional 3 admis- A36). sions the following year. Medical ad- missions were the most common for clined by 54,537 filled days per 1,000 For women with diabetes, the medical people with diabetes. In 2014, there people, from 268,738 filled days in subservice category with the highest were 95 medical admissions per 2012 to 214,201 filled days in 2014. use was professional services. In 1,000 people, nearly twice as many as The most used therapeutic class of 2014, there were 41,038 professional the number of surgical admissions prescriptions in 2014 was generic services per 1,000 women (Appendix (55 admissions per 1,000; Appendix cardiovascular drugs. Use of this class Table A39). Between 2012 and 2014, Table A37). of prescriptions increased over the a small decline occurred in the rate of www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 16 use of these services: by 833 services 1,000 women in 2012 to 8,882 in was a decline of 52,942 filled days of per 1,000 women. A small year-on- 2014. While use of outpatient visits brand prescriptions over the study year decline also occurred in the use declined slightly between 2012 and period. However, the use of filled days of acute inpatient admissions. 2013 – from 805 visits per 1,000 of generic prescriptions increased women to 801 visits, a small increase every year studied. This increase in The number of admissions declined occurred in use of visits in 2014 (to generic filled days of 91,918 filled by 12 admits per 1,000 women, from 812 visits per 1,000). days of generic prescriptions per 182 admissions in 2012 to 170 in 1,000 women led to a net increase in 2014. Similarly, use of outpatient- In 2013, the use of filled days of brand the total filled days of prescriptions other services also declined each year prescriptions declined by 13.8% and over the study period. Total filled studied, from 9,297 services per by an additional 7.9% in 2014. This COMPARISON OF THE RATES OF SERVICE USE FOR PEOPLE WITH AND WITHOUT DIABETES In every year of the study period, people with diabetes had higher rates of use for all of the service categories compared to people without diabetes (Table 3 and Appendix Table A36). Across the service categories, the difference in use between the two groups was larger for some categories of services than for others. When examining the ratio of service use for people with diabetes compared to those without, the ratio is largest for brand prescriptions (7.1) and generic prescriptions (5.1) and smallest for professional services (2.4; Figure 9). In other words there is a larger difference in use of prescriptions than of professional services. In 2014, people with diabetes used 1,242,571 filled days of prescriptions, compared to 233,208 filled days for people without diabetes (Table 3 and Appendix Table A36). For prescriptions, this difference in use was due largely to the use of antidiabetic agents by people with diabetes (279,489 filled days of antidiabetic agents per 1,000 people with diabetes). The difference was also influenced by the comparatively higher use of cardiovascular agents by people with diabetes. In 2014, there were 414,524 filled days of cardiovascular agents per 1,000 people with diabetes compared to 55,291 filled days of cardiovascular agents per 1,000 people without diabetes. After brand and generic prescriptions, the next highest use ratio – the second-largest difference in rate of service use – in 2014 was observed for outpatient other services (3.8). Within the outpatient services category, the biggest difference in use between people with diabetes and those without was for ancillary services (6.0). In 2014, there were 1,980 ancillary services per 1,000 people with diabetes compared to 328 ancillary services per 1,000 people without diabetes. www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 17 days of prescriptions increased from (Appendix Table A41). This high use fessional services (7,943 services per 1,227,220 filled days per 1,000 wom- of outpatient services was driven by 1,000 children; Appendix Table A43). en in 2012 to 1,266,384 filled days in the use of outpatient-other services Use of acute admissions per 1,000 children declined every year studied, There were 3 times as many acute admissions for women with dia- from 200 admissions to 195 admis- betes compared to women without diabetes, and 4 times as many sions (Appendix Table A41). This de- acute admissions for men with diabetes compared to men without cline was due largely to a small de- cline over the study period in the use diabetes (Appendix Tables A39 and A40). of medical admissions. The number of 2014. (10,561 per 1,000 children), which medical admissions declined from accounted for 94.8% of outpatient 168 admissions per 1,000 children in As with women, the medical sub- service use. In contrast, children had 2012 to 160 admissions in 2014 service category with highest use by the lowest rate of outpatient visit use (Appendix Table A43). men was professional services. In (Figure 10). In 2014, there were 577 2014, there were 33,387 professional Children were the only age group for outpatient visits per 1,000 children, services per 1,000 men (Appendix whom increases occurred in the num- which was 56% lower than the use of Table A39). The number of profes- ber of filled days of brand prescrip- outpatient visits by young adults (the sional services increased slightly be- tions during the study period. Be- age group with the highest use). Of tween 2012 and 2013 (by 80 services tween 2012 and 2014, the number of these outpatient visits for children, per 1,000 men), before a small decline filled days of brand prescriptions rose 75.9% were ER visits (438 visits per in use the following year (by 442 ser- by 12,022 filled days per 1,000 chil- 1,000 children; Appendix Table A43). vices per 1,000 men). The use of acute dren (Appendix Table A41). Due to Children also had the lowest use of inpatient admissions declined in eve- this increase in use, by 2014, children professional services. In 2014, there ry year studied, from 159 admissions had higher rates of use of brand pre- were 25,646 professional services per per 1,000 men in 2012 to 152 in scriptions than any other age groups. 1,000 children a slight decline over 2014. A year-on-year decline of 1.5% Brand prescriptions also constituted a the previous year (Appendix Table also occurred in use of outpatient- larger proportion of total prescrip- A41). Of these professional services, other services. Use of outpatient- tions (40.0%) than for any other age 24.4% were visits to the doctor other services declined about 1.5% group. This high use of brand pre- (6,249 doctor visits per 1,000 chil- each year, from 8,215 services per scriptions was driven largely by the dren) and 31.0% were lab/path pro- 1,000 men in 2012 to 7,935 in 2014. use of brand antidiabetic agents. Chil- Conversely, use of outpatient visits increased by about 1.0% per year across the study period, from 11 visits to 626 visits. Similarly, use of filled days of generic prescriptions in- creased each year studied, by 96,544 filled days per 1,000 men between 2012 and 2014. This increase more than offset the decline of 56,008 in filled days of brand prescriptions dur- ing the same period. Use by Age Group Children Children (ages 0-18) with diabetes had a higher use rate of outpatient services compared to the other age groups. In 2014, there were 11,138 outpatient services per 1,000 children www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 18 dren used more filled days of brand groups. As with the other age groups, services. In 2014, there were 6,643 antidiabetic agents per 1,000 people young adults’ most used subservice outpatient-other services per 1,000 then did any other age group category was professional services working age adults (Appendix Table (Appendix Table A43 and Figure 11). (30,289 professional services per A41). And this use increased each year 1,000 young adults; Appendix Table However, working adults had the sec- studied. In 2014, there were 255,481 A41). Of these professional services, ond highest (after young adults) use filled days of brand antidiabetic 17.5% were visits to the doctor of outpatient visits: 740 visits per agents per 1,000 children. In compari- (5,287 doctor visits per 1,000 young 1,000 intermediate adults (Figure 10). son, children used 1.5 times more of these prescriptions than did young Young adults with diabetes had 4 times more mental health and adults, who had the second highest use (172,369 filled days of brand anti- substance use admissions and nearly 3.5 times more ER visits than diabetic agents per 1,000 young young adults without diabetes (Appendix Tables A45 and A46). adults; Appendix Table A45). adults), while 41.9% were lab/path Young Adults Of those visits, 65.3% were ER visits professional services (12,686 per (483 ER visits per 1,000 intermediate Young adults (19-25) with diabetes 1,000 young adults; Appendix Table adults) and 26.8% were outpatient were the only age group to see an in- A45). At the same time, compared to surgical visits (198 visits; Appendix crease in service use for all the medi- the other age groups, young adults Table A47). Between 2013 and 2014, cal subservice categories in each had lower use of filled days of pre- the use of professional services by study year. Young adults also had scriptions (592,734 filled days per intermediate adults increased slight- more acute hospital admissions per 1,000 young adults; Appendix Table ly, to 32,948 professional services per 1,000 people than the other age A41). 1,000 adults (Appendix Table A41). groups (Appendix Table A41). In Intermediate Adults The most used type of professional 2014, there were 259 acute admis- service was lab/path services (12,883 sions per 1,000 young adults. Acute For intermediate adults (ages 26-44), professional lab/path services per admissions rose from 247 admits per the rates of use of the medical sub- 1,000 intermediate adults) followed 1,000 in 2012 to 259 in 2014, a 5% service categories declined between by visits to the doctor (6,023 visits increase. In comparison, the next 2012 and 2013. The following year, per 1,000; Appendix Table A47). highest use of acute admissions was use of acute inpatient admissions, for children, who had 195 admissions outpatient visits, and professional Compared to the other age groups, per 1,000 children in 2014. For young services increased slightly (Appendix intermediate adults had the lowest adults, the most admissions in 2014 Table A41). Compared to the other use of brand prescriptions. In 2014, were medical (172 admissions per age groups, intermediate adults had a there were 152,127 filled days of 1,000), followed by MH/SU admis- low rate of use of acute inpatient ad- brand prescriptions per 1,000 inter- sions (37 admissions), and LD admis- missions. In 2014, there were 154 mediate adults (Appendix Table A41). sions (28 admissions; Appendix Table acute admissions per 1,000 interme- As with the other age groups, the larg- A45). diate adults. The majority of these est proportion of brand prescriptions admissions were medical admissions was filled days of brand antidiabetic Young adults also had a higher rate of (82 admissions per 1,000 intermedi- agents. However, intermediate adults outpatient visit use compared to the ate adults) followed by LD admissions had the lowest rate of use of filled other age groups (Appendix Table (25 admissions per 1,000 intermedi- days of brand antidiabetic agents A41 and Figure 10). In 2014, there ate adults; Appendix Table A47). (96,730 filled days per 1,000 interme- were 902 outpatient visits per 1,000 Compared to the other age groups, diate adults) of the five age groups young adults, an increase of 24 visits intermediate adults had the lowest (Appendix Table A47 and Figure 11). over 2012. Of these outpatient visits, rate of use for outpatient services Brand prescriptions made up 18.6% 80.4% were ER visits (725 visits per (7,383 services per 1,000 intermedi- of total prescription use by intermedi- 1,000 young adults in 2014; Appendix ate adults) (Appendix Table A41). ate adults (Appendix Table A41). The Table A45), which was the highest This low use was due largely to com- total number of filled days of pre- rate of ER visit use among the five age paratively low use of outpatient-other scriptions increased each year during www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 19 path services per 1,000 pre-Medicare adults (Appendix Table A51). There were also 7,163 visits to the doctor per 1,000 adults. Similar to middle- age adults, there were small declines in the use of acute inpatient admis- sions and outpatient-other services over the study period (Appendix Ta- ble A41). The number of acute admis- sions declined from 181 admits per 1,000 pre-Medicare adults in 2012 to 171 in 2014. The number of outpatient-other ser- vices declined from 9,890 per 1,000 pre-Medicare adults in 2012 to 9,383 in 2014 (Appendix Table A41). Con- versely, pre-Medicare adults’ use of outpatient visits increased every year studied. Between 2012 and 2014, out- patient visits per 1,000 pre-Medicare the study period. This was the only study period. This was due to an in- adults rose from 710 visits to 719 adult age group from which a year-on crease in the number of filled days of visits. Unlike the other age groups, -year increase occurred in total pre- generic prescriptions between 2012 outpatient surgical visits were the scription use. and 2014, which rose from 879,274 most common type of outpatient visit filled days of generic prescriptions to Middle-Age Adults for pre-Medicare adults (Appendix 974,063 filled days. Table A51 and Figure 11). In 2014, We observed general declines in the Pre-Medicare Adults there were 381 outpatient surgical rates of use of the medical subservice visits per 1,000 and 258 ER visits per categories by middle-age adults. The The rates of use of the medical sub- 1,000. Pre-Medicare adults used more only increase in use of services for service categories by pre-Medicare filled days of prescriptions per 1,000 this age group occurred between adults (ages 55-64) generally de- people than did the other age groups. 2013 and 2014 in the rate of use of clined across the study period. Only In 2014, there were 1,486,253 filled outpatient visits per 1,000 (from 677 use of outpatient visits saw small in- days of prescriptions per 1,000 pre- visits to 687 visits; Appendix Table creases in the years studied Medicare adults (Appendix Table A41). This increase was due to an in- (Appendix Table A41). Similar to A41). Of these filled days, 83.4% were crease in the rate of use of ER visits young adults, total prescription use generics (1,238,539 filled days) and per 1,000 between 2013 and 2014 – declined slightly between 2012 and 16.6% were brand (247,031) – the from 316 visits to 330 visits 2013 before increasing the following smallest proportion of brand filled (Appendix Table A49 and Figure 11). year. Use of professional services by days for any age group. For pre- The rates of use of acute inpatient pre-Medicare adults declined over the Medicare adults in 2014, the most admissions, outpatient-other services, study period. At the same time, use of filled days were for generic cardiovas- and professional services all declined professional services, in general, in- cular agents (474,413 filled days per in each year studied (Appendix Table creased with age. Pre-Medicare adults 1,000 pre-Medicare adults), followed A41). Interestingly, in 2014, middle- used more professional services per by generic antidiabetic agents age adults had the lowest number of 1,000 people than the other age (201,236 filled days; Appendix Table acute admissions: 141 per 1,000 mid- groups. In 2014, there were 39,917 A51 and Figure 10). dle-age adults. As with intermediate professional services per 1,000 pre- adults, the number of filled days of Medicare adults. The most used type prescriptions per 1,000 middle-age of professional service was lab/path adults increased each year of the services: 13,632 professional lab/ www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 20 www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 21 About the Data and Methodology Data adjudicated. HCCI used the weighted, associated with diabetes. Individuals actuarially completed dataset to esti- with a diagnosis of diabetes for at HCCI’s dataset contains several billion mate per capita health expenditures, least one inpatient admission, one de-identified commercial health in- average prices, and utilization of ser- outpatient visit, or two office visits in surance claims for the years 2010 vices for 2012 through 2014. HCCI did the same calendar year were flagged through 2014. Three major health not correct dollars for inflation; thus, in a year of data as having been diag- insurers contributed data to HCCI for all reported expenditures and prices nosed with diabetes (see the HCCI the purposes of producing a national, were in nominal dollars. Methodology document for a list of multipayer, commercial health care codes included in this categoriza- claims database. These data include HCCI analyzed four major categories tion).5 Once individuals were flagged claims for individuals covered by of services, several subservice catego- as having received a diagnosis of dia- group insurance (fully insured and ries, and detailed service categories betes, they retained this flag in all administrative services only), individ- (see “Service Categories Analyzed in subsequent years of the dataset. This ual insurance, and Medicare Ad- this Report”). Inpatient facility claims methodology excluded radiology and vantage plans. The claims data in- were from hospitals, skilled nursing laboratory claims, as these can be clude prices paid to providers by both facilities (SNFs), and hospices where used for screening purposes and may insurers and insureds and details detail was sufficient to identify an not reflect a diagnosis of diabetes. about the services used. Furthermore, overnight stay by an insured. Outpa- HCCI’s claims data are compliant with tient facility claims did not entail an To be flagged in the HCCI dataset as the Health Insurance Portability and overnight stay and included observa- having a diagnosis of diabetes, indi- Accountability Act (HIPAA). tion and ER services. Both outpatient viduals must have had at least one and inpatient claims consisted of only medical claim filed with their insurer For 2014 Diabetes Health Care Cost the facility charges associated with in one of the years of the study peri- and Utilization Report, HCCI per- such claims. Professional services od. The population of individuals formed analysis on a subset of data (procedures) included claims billed without diabetes is composed of all for approximately 40 million insureds by physicians and non-physicians members in the HCCI analytic dataset per year (2012–2014). This analytic according to the industry’s standard who were not flagged as having re- subset consisted of all claims for in- procedure coding practices. Prescrip- ceived a diabetes diagnosis. This pop- sureds ages 0 through 64 and covered tion data are prescriptions filled at ulation without diabetes included by ESI. The data set used for this re- both retail and mail order pharma- individuals who never had a medical port represented about 26% of the cies. claim filed with their insurer during privately insured people in the United the study period. Therefore, these two For a more detailed description of States. populations—individuals diagnosed HCCI’s methodology and dataset, see with diabetes and individuals without Methods the Analytic Methodology on HCCI’s diabetes—are similar but not meth- Website.5 For a discussion of the cost HCCI weighed the analytic subset us- odologically identical. Per capita and utilization trends for the national ing United States Census Bureau age- spending trends and rates of service younger than 65 ESI population, see gender geographic-based estimates of utilization for these populations the 2014 Health Care Cost and Utiliza- the ESI population to make the analyt- should be treated as estimates. tion Report. Both documents are avail- ic subset representative of the nation- able on HCCI’s Website, for details on Limitations al ESI population. Claims in the ana- these changes. lytic subset from 2013 and 2014 were This report, like all research, has sev- actuarially completed to account for A diagnosis of diabetes was deter- eral limitations that affect the gener- claims that had been incurred but not mined using the DDMT methodology alizability and interpretation of the for identifying health care activity findings. For this reason, HCCI consid- www.healthcostinstitute.org 2014 Diabetes Health Care Cost and Utilization Report 22 ers the work a starting point for anal- Washington, DC: Health Care Cost Disease Management Association ysis and research on individuals cov- Institute. 2015. of America, 2004. ered by ESI rather than as a conclu- 2. The population and spending 11. Burton, B, Jesilow, P. "How sive analysis of the ESI population’s trends detailed in this report Healthcare Studies Use Claims effect on health care in the United differ slightly due to differences Data." The Open Health Services States. in the two data sets and in the and Policy Journal 4.1 (2011): 26 methodology used to develop the –29. First, our findings were estimates for diabetes diagnosis flag, see the United States ESI population ages “About the Data and 0 through 64 based on a sample of Methodology” for more approximately 26% of these insured information. Americans. Second, the analysis and 3. American Diabetes Association. results were descriptive, and the find- “Economic Costs of Diabetes in ings were not used to determine caus- the U.S. in 2012.” Diabetes Care al relationships. The tables and fig- 2013 Apr; 36(4): 1033-1046. ures presented are limited to descrip- 4. Diabetes.co.uk. “Anti-diabetic tive statistics for individuals covered drugs.” http:// by ESI and younger than age 65. In www.diabetes.co.uk/diabetes- this brief, we presented spending and medication/antidiabetic- utilization trends for individuals drugs.html. flagged as having diagnosed diabetes 5. Health Care Cost Institute. 2014 Health Care Cost and Utilization and those not flagged as having diabe- Report Analytic Methodology tes. V4.0. Washington, DC: Health Third, because HCCI’s claims holdings Care Cost Institute. 2015. reflect only explicit health care activi- 6. McEvoy, Gerald K., Ed. AHFS Drug ty, HCCI could not identify individuals Information 2010. Bethesda, MD: with diabetes who (1) did not seek American Society of Health- medical care between 2012 and 2014, System Pharmacists, 2010. Print. (2) did not meet our criteria for study 7. There was a small amount of inclusion, (3) did not file with their spending per capita each year on health insurer a claim that indicated a “unknown” prescriptions that diagnosis of diabetes, or (4) had undi- could not be identified as either brand or generic prescriptions. In agnosed diabetes. Moreover, claims 2014, the spending on unknown data have a mixed record of utility for prescriptions was $12 per capita. population health studies. Work is 8. In the tables there is a class of ongoing to improve the methods used brand and generic prescriptions to determine health status from ad- labeled “other therapeutic ministrative claims. classes”. This is a combination of Suggested Citation for This Report: 22 individual classes of prescriptions that have been “2014 Diabetes Health Care Cost and combined due to low spending Utilization Report.” Health Care Cost and utilization. Institute, Inc., June 2016. Web. 9. Health Care Cost Institute. 2014 Health Care Cost and Utilization Report. Washington, DC: Health Endnotes Care Cost Institute. 2015. 1. Health Care Cost Institute. Per 10. Duncan, I, Ed. Dictionary of Capita Health Care Spending on Disease Management Diabetes: 2009-2013. Terminology. Washington, DC: www.healthcostinstitute.org The Impact of Diabetes: Costs and Use of Health Care by Children in 2014 In this report, we estimated that 0.3% of children (ages 0-18) covered by employer-sponsored insurance (ESI) have been diagnosed with diabetes. While this might be a relatively small percentage of children, for the kids diagnosed with diabetes and their families, this condition can have a big impact. Below we compare the health care costs and service use for the average child with diabetes and the average child without diabetes. In 2014, compared to the average child without diabetes, the average child with diabetes used more than twice as many health services, and their parents paid more than 4 times the dollars out of pocket. In 2014, the average child (aged 0-18) with diabetes, compared to the average child without diabetes… Incurred health care costs that were 6.7 times higher Had 2 times as many doctor visits Had nearly 5 times more acute inpatient admissions Used 4.6 times as many professional laboratory/pathology services Had 2.5 times as many ER visits Used 10.5 times as many filled days of prescriptions Their parents spent: 4.7 times more out of pocket 2.4 times more out of pocket on doctor visits 6.8 times more out of pocket on acute inpatient admissions 1100 G Street NW, Suite 600 Washington, DC 20005 202-803-5200 www.healthcostinstitute.org Copyright 2016 Health Care Cost Institute Inc. Unless explicitly noted, the content of this report is licensed under a Creative Commons Attribution Non-Commercial No Derivatives 4.0 License