Assets in health savings accounts (HSAs) and health reimbursement arrangements (HRAs), two relatively new employment-based health benefit plan options , have grown in recent years and totaled $7.1 billion in 2009, up from $835.4 million three years earlier, according to a study by the nonpartisan Employee Benefit Research Institute (EBRI). In addition, the number of accounts in these plans, collectively known as consumer-driven health plans (CDHPs), also was up during the same period—totaling 5 million in 2009, up from 1.2 million in 2006.
ASSET LEVELS GROWING: In 2009, there was $7.1 billion in consumer-driven health plans (CDHPs), which include health savings accounts (or HSAs) and health reimbursement arrangements (or HRAs), spread across 5 million accounts. This is up from 2006, when there were 1.2 million accounts with $835.4 million in assets, and 2008, when 4.2 million accounts held $5.7 billion in assets.
AVERAGE ACCOUNT BALANCE LEVELING OFF: Increases in average account balances appear to have leveled off. In 2006, account balances averaged $696. They increased to $1,320 in 2007, a 90 percent increase. Account balances averaged $1,356 in 2008 and $1,419 in 2009, 3 percent and 5 percent increases, respectively.
TYPICAL ENROLLEE: The typical CDHP enrollee was more likely than traditional plan enrollees to be young, unmarried, higher-income, educated, and exhibit healthy behavior. No differences were found between CDHPs enrollees and traditional plan enrollees with respect to gender, race, and presence of children.
MORE ROLLOVERS: Overall, the number of people with a rollover, as well as the total level of assets being rolled over, have been increasing. The average rollover increased from $592 in 2006 to $1,295 in 2009.
DIFFERENCES IN ACCOUNT BALANCES: Men tend to have higher account balances than women, account balances increase with household income, education has a significant impact on account balances independent of income and other variables, and no statistically significant differences in account balances were found by smoking, obesity, or the presence of chronic health conditions. Individuals who developed a budget to manage their health care expenses had a higher account balance ($1,726) than those who did not ($1,428), but otherwise, no statistically significant differences in average account balances were found between individuals who exhibited various aspects of cost-conscious decision-making behaviors and those who did not.
DIFFERENCES IN ROLLOVER AMOUNTS: Men rolled over more money than women, whites have higher rollover amounts than minorities, and the youngest adults and the oldest adults had the largest rollover amounts in 2009. Rollover amounts increase with household income and education, and individuals with single coverage rolled over a slightly higher average amount than those with family coverage. There was no statistically significant difference in rollover amounts by health status, although individuals who smoke had higher rollover amounts than those who do not and obese individuals had lower average rollover amounts than nonobese individuals. Individuals who talked to their doctor about treatment options and costs, those who used an online cost-tracking tool provided by the health plan, and those who asked their doctor to recommend a less costly prescription drug had higher rollover amounts than those who did not take such actions.
Accounts(millions) Accounts (millions) Figure 1 Figure 7 Race It was CDHP enrollees are muc The employer Figure 5, Dist Whe than those wh sample n examin foun cont d t wia h ribution of Traditional Plan n ribution o di s divided into at in g dif d not divi fer of les ences part duals h mis o in ro one c who than re likely to ipate in of t talke llovers $1,000 such hre d to e h fo a En groups: t had a r men ve a co their a pro rollee n doctor g an av ram. llege h s d wo erage ose an Figure 5 Th a or d CDHP Enro men with b e post-gradu out tr accou avera a consum by a eatmen n gt e g acc bala e, amen a te educatio llees, by Ho o er-dr t options nce o unt balanc g iven fes $1 hea 21 ,4 and n. Nearl 03, - u e s 34 have highe lwas $ th pl costs, those w eh whi old Incom y an le t 1one-half (46 percent) ,61 (CDH hose 4 aP), those m e r w h , Adults ong those avera io use th ang em de an with ro ploy w on llo a h ver of o use line er high - d a b Paul Fronstin is director of the Health Research and Edu Figure 25 cation Program at EBRI. This Issue Brief was written with Health Savings Accounts and Health Reimbursement Figure 23 Figure 13 Figure 27 Figure 15 Figure 11 Figure 3 Length of Time With HRA or HSA, 2006 and 2009 Figure 35 Figure 29 Figure 19 a Figure 31 Figure 17 a Distribution of Traditional Plan Enr Figure 33 ollees and CDHP Enrollees, by Self- Minorities with HRAs or HSAs Distrib have ution hi of gher Traditional account a balanc Plan Enr Figure 9 bFigure 21 es that ollees whitand CDHP es with these ac Enrcounts. On ollees, average, minorities Ages 21–64, 2009 ........................................................................................................................................ 9 cos deductible contributio cost-tracking t amounts than CDHP enrollees had a coll t and qua n hea o li ftool provi women at l ly informatio th plan east $1,000 ha in t (HDH ded ege degr his nby the to choos P), a age gro n d a d those healt ee a Rollo n u e aver p. a n hd T d pl ver Amount, another h o with tr age o an, an ctor, and e aver faditio d th age $1, 21 percent h $ 717 rol ose 1 na , 413 b l l he over for in th who asked th y Gender amon aleir acco th a c d a men o g th ve gr and Ag ra u os aduate degr was $ eir doctor n gt (Fi e who e. In g 1 e d ure ,32 , di i vid 2009 to recommend a l 1 0 d u , ee (Figure 9 n al w ). os w hile t do s Similar efor reo a wome l( 6). In y, in ssi Fig gu n di re e contrast, 23 n it e 1 viduals d ss costly t6 o was ). t Th he C who $97 e ac D0 HP percent count (Fi ag nur d e assistance from the Institute’ Account Balances, s research and Avera Account Tg oAccount Balances, e HRA tal Rollo Rollo by Balances, ed Health Beha or HSA itorial staffs. Any ver ver Amount, Assets and Number b Account Balances, y b Education, vior y Race b views exp y and Health Status, Race , August A , 2009 ugust 2009 rof Adults essed 2006–2009 2009 in this report August ar 2009 e those of the author Rollover Amount, by Level of Annual Employer Contribution, 2009 Account Balances, Rollover Amount, b b yy Education, Level of Ann 2009 ual Arrangements: Assets, Account Balances, and Rollovers, 45% Account Rollo Balances, ver Amount, by V ba yrious Health Cost-Conscious Behavior and Health Status, Behaviors, August 2009 2009 Rated Rated Rollo H Health ver Amount, ealth Status Status b and and y Va H Health rious Cost ealth Beha Beha -Conscious vior vior, Adults AdultsBeha Ag Ages es vior 21 21s, –64 64 2009 , 2009 2009 References Account Balances, bAccount Balances, y Household Income by Gender by , Adults Ag Rollo and Ag ver Le es vel, e21–64, , A ugust August 2009 2009 2009 have an account balance of $1,729, while whites have an account balance of $1,444 (Figure 11). HDHP g balanc 2 contribut prescript of traditional 6). A sma es w roup ie on d le le le re if dr plan e r d ss than $ similar they ug iffhad hi ere nhad rolle nby c1 e be a gher ,0 es h wel 0 de 0 tw rollover ha lness pro ducti ae d a e d a n men colle b 42% n le o ave amou gfge ram a at rn age degre l d partic nts than t e w ast $1, account ome e an ipat n c 00 d h io a balance 0 ose 11 n n, fo a percent with who lr in so b dividual co of e no statist di se d $1, ha e not t n d a gradu 1 at 00, a i o verage or cally signific ke s lwhil der ag ue thos a ch te degr actio es$2,000 . e w an n ee. Tradit t di s h.o contr In for fferenc th family c e seri iional buted at l e. es of o pla verage. nquestio e enrollees ast $ To 1n ,00 be s, 0 and should not be ascribed to the offiAg cers, es 21 trustees, or ot –64 With a her spo Rollon ver sors of EBRI, , 2006–2009 EBRI-ERF, or their staffs. Neither EBRI Figure 6, Dist $1,400 $1,400ribution of Traditional Plan Enrollees and CDHP Enrollees, by Education, Adults Ages $3,000 $1,800 $1,400 $2,000 $1,600 Employer Contribution, August 2009 $2,500 70% 2006–2009 Fronstin, Paul. $1,400 "Findings from the 2009 EBRI/MGA Consumer Engagement in Health Care Survey." EBRI Issue Brief, no. these as were had a signed much n were av to th erage ba th mo e only e CDHP re like lance ones ly tha gr of oup w $ nh CDHP ,1 ere t they m ,890h enrol e (Fi di ug sffer ur t al le e e e s20). s o nce to have have a was n fo onl ac uny count, such d to a hibe gh statistically sign schoo as a h l degre ealth e, savin 35 ifica percent ng t (Figure s account and 8 pe 33 (H ). SA) or rcent, re hea sp lth ectively. nor EBRI-ERF lobbies or tak 64% es positions on specific policy proposals. EBRI invites comment on this research. $1,288 21–64, 2009 40% .................................................................................................................................................. 9 50% $2,000 $2,000 $1,275 $1,250 $2,000 $1,600 June 2010 • No. 343 $4,500 Yes No 7 Yes No Yes No $1,419 $1,778 By Paul Fronstin, Employee Benefit Research Institute 337 (Em $1,750 $1,800 ploye 59% e Benefit Research Institute, December 2009a). reimbursement arrangement (HRA) with a rollover provision that they can use to pay for medical expenses or the ability Household Income 36% $1,729 R Cos ace t-Con 60% $1,400 scious Behaviors 45%* $1,356 $1,803 $1,214 Figure 7, Distribution of Traditional Plan Enrollees and CDHP Enrollees, by Self-Rated Health Status $1,320 $1,163 Yes No $1,200 $1,200 $1,200 $1,161 45% It wil There l are be impo no $1,800 $1,800 statistically significant rtant to track this tren diff d over eren time. Curr ces in selfen -retl py, account orted health bal status betw ances are lo ee w n CDHP and ar an e th d tr ereaditional fore investe plan d in $1,140 $1,158 $1,726 54% 54% $2,474$3,975.7 $1,139 $1,200 35% $1,800 Where the world turns for the facts $1,374 on U.S. employee benefits. to take their account with them should they change jobs. Individuals with only a flex$1,717 ible spending a$2,017 ccount (FSA) were 2006 2009 Copyright Inf $2,500 $1,400 $4,000 ormation: This report is copyrighted by the Em 52% ployee Benefit Research Institute (EBRI). It may be Accordin Length of Ti g to Figure me with Ac 12, acco count unt balances for CDHPs increase with household $1,587 income. The average account balance was $1,200 $1,188 $1,113 In contrast to A number o and Health Behavior, Adult f c o the f st-conscious indings $1,106 thbehaviors at minoriti s Ages wer es with H e e21–6 xamine RAs or 4, 2009 d to HSAs have see ......................................................................................... 10 if individu higher acco als whount balances ex $1,102 hibit more cost- than whites, it conscious d wa es found cision $1,700 $1,626 $1,647 $1,096 $1,087 ___ _____. $2,000 $1,600 "What Do We Know About Enrollment in Consumer-Driven Health Plans?" EBRI Notes, no. 12 6 (Employee 49%* relatively enrollees. Ho safewever, vehicles suc CDHP h enrolle as money mar es are $1,066 less likely th ket funds. As an traditio account nabalan l planc enro es gro llees w, th to e report t potential hat th to $1,590 in ey ha vest in mor d at leas e ri t one of sky $1,056 $1,049 Traditional CDHP $1,570 not include 50% d in the CDHP group. $1,534 $1,544 $1,035 $1,574 40% $1,600 $1,200 $1,048 30% Traditional CDHP used without permission $1,019 but citation of th$1,508 e source is require 46%* 38% d. $1,505 $1,030 $1,020 $912 amo$1,600 ng individuals with less than $50,000 in household income; $1,452 among individuals with $50,000 -$99,999, $1,505 $1,000 The that w making len hites h g had hi th$1,600 of atime that gher d high account er rol anl i over amounts n balanc dividual es than has th had t those who an minor he accou ities. $1,495 did not. nt has a On avera The n im ex ge, pact on ro pe minorities ctation w llover a had s th amounts. a $ at those 1,000 $1,520 rollo who The analysis fo ver in exhibite 2009, d cost- un w d that hile $1,002 30% $1,476 $982 $1,488 Figure 8, Dist Benefit Resear ribution of Traditional Plan ch Institute, December Enro 2009b llee ): 11 s an -15. d CDHP Enro llees, by Firm Size, Working Adults investment ve eight differenthicles, such chronic con as dit mutual ions. Just ove funds rand stocks, w one-half (52 iperce ll grow nt) of tr . The $1,455 o apportu ditional nities plan fo enrollees r capital a report ppre $1,480 c ed t iation i hat th ncreas ey ha ed at , but $3,500 $962 $1,000 $1,000 $1,000 $1,429 Retirement $1,000 $1,200 $1,418 and health benefits are at the heart of w $1,453 $957 orkers’, employers’, and our nation’s $1,416 $1,140 $1,428 $1,650 Assets (millions) $933 $1,391 and $1,766 $1,400 among individuals with $100,000 or more in household income. While high $927 er-income households may Introduction $1,150 $1,105 persons with a conscious be whites had a h$1,2 anvior wou account for 75 rollove ld ha r (Figure one ve a -higher two 27 year ).a vs had an av erage accoun erage ro t balance t llover h of $830 an those (Fi who gure di d not 34). In ex compa hibit su rch ison, in behavior. dividu The als 26% 35% $1,400 $1,403 Ages 21–64, $1,620 2009 .............................................................................................................. $1,075 ........................ 10 $1,354 5 least one so do th Becausee o the of 40% p ei base sample portunities ght differefor c n (n $1,279 t cational sam h apital ronic co losses, even amo nditi ple o) i ns, whil ncluded onl e n46 g i n perce ydi 94 viduals at in nt of dividual CD th HPs report s in a e high CDHP, ened d o a at havin f the n oversampl employer g at lee ast one o and of inindividua divifduals the l with a economic security $1,400 . Founded in 1978, EBRI is the most authoritative and objective source of Recommended Citation $879 : Paul Fronstin, “Health Savings Accounts and Health Reimbursement Arrangements: Assets, $2,000 $1,029 $1,000 $1,400 $1,005 contribute25% higher amounts to their HSAs than lower-income households, the regression equation controls for ________. "The Use of Health Savi Acngs ac counts (mcounts for illions) Health Care in Retirement. EBRI Notes, no. 4 (Employee Benefit with questions regard an account for $3,000 ing three cost- 4 - conscious four years dec had ision-makin an avera $834 g ar ge rol e as follo lover of $1,378. ws: And those with the account at least five years Employers first started offering account-based health $1,246 plans in 2001, when a handful of plan sponsors began to offer $1,600 32% $953 contributio CDHP conditions was$1,000 (Figure a n di dded. stribution. T 7). he ove Sim rila sarly, those i mple includ ned CD 87 HPs w 9 i$1,165 ndivi ere dles uals s likely th with a CD an $2,764.8 HP. In individuals e addition nrol to led in tr being s aditio tratifie nadl pl , th ans to e base be information on these critical, complex issues. Figure 9, Account Balances, by $1,500 $1,200 $800 Gender and Age, August 2009 .................................................................... 31% ......... 11 Account Ba Health Savings Accounts $1,200 lances, and Rollovers, 2006–2009,” EBRI Issue Brief, and Health Reimbursement no. 343 (June 2010 $1,416 ). 30% $800 $1,397 $905 $1,386 contributio $1,100 n $800 $800 levels and educational attainment. The difference in account balances by household income may be due to $1,200 $1,366 had a Research n average rollover Institute, Apr of $1,774. il 1010): 2 -6. health reim 30% bursement arrangements (HRA 20% s), a type of employer-funded health plan that reimburses workers for 27% Household Income $1,200 obese. sample Near was al ly so wei one-thg ird (31 hted by gen perced ner, t) of tr age, a education, ditional plan region, enrollees wer income, e obese, and race/ethn compairced with ity to reflect 23 pe trcent he act 4 amon ual g CDHP 20% $800 $2,500 Figure 10, Account Balances, by Gen In the last 12 months or since you join der and Age ed your curr Combined, ent health August 2009 plan, did you do ......................................................... any of the following: 11 higher-income $1,550 households being less likely to take distributions from the account. 23%* (Th 24%*is is an empirical question that is $1,000 Arrangements: Assets, Account Balances, and Rollovers, 25% 2 qualified me $1,000 $800 dical expenses. In 2004, employers were able to start offering health plans with health savings accounts $696 21% Report availability: EBRI focuses solel This ry eport is on emplo availay ble on th ee benefits research — no lobb e Internet at www.ebri.org ying or advocacy enrolle proportions es. Furt in hermore, the population CDHP ages enro 21 llees –6 we 4 wre less l ith private ikely t hea han lth insura traditiona ncel c pla overage. n enrolle es to smoke ci The CDHP oversample garettes, an was d were According to Figure 28, rollover amounts increase with household income. In 2009, the average account balance was $1,000 $1,500 $600 Rollovers EBRI Employee Benefit Research Institute Issue Brief (ISSN 0887 ?137X) is published monthly by the Employee Benefit Research Institute, Sandman, David, Elisabeth Simantov, and Christina An. Out of Touch: American Men and t 18% he Health Care System $1,000 15% worth exp$1,050 l 20% oring as data become available.) Figure 11, Account Balances $600 , by Race, August 2009 ............................................................................................. 13 (HSAs), a type of tax-exempt trust or custodial account that an individual can use to pay for health care expenses. The 1100 13th S $600 EBRI stand t. NW, Suite 8s alone in em 78, Washington, Dpl Coyee ben , 20005-4051, efits rese at $300 pear r ye ch a ar or sis an inde included as pend part eof nt, nonprofit, and no a membership subscription. npa Pe rtisan riodi- Employer an $600d Individual Con $2,000 tributions 15% 15% wei more likely to ghted by gende $800exercise regul r, age, inc arly. ome, and race/ethnicity. More information can be found in (Fronstin, 2009a). $60 2006–2009 9 amo$1,500 ng individuals with less than $50,000 in household income; $1,026 among individuals with $5 30,000 -$99,999, 20%1. Checked whether my health plan would cover my care or medication? $600 13%* 13%* Like contri (Commonw $1,000 buti $800on leve ealth Fund, ls, rollo Marc ver amounts h 2000). have one of the largest impacts on average account balances. Individuals with cals postage rate paid in Washington, DC, and additional mailing offices. POSTMASTER: Send address changes to: EBRI Issue Brief, 1100 $ $600 17% 17% 3.1 organization. It analyzes and reports research data without spin or underlying agenda. All findings, $800 $800 theory behind these accounts is that when individuals are given more control over funds allocated for health care Figure 12, Account Balances, by Annual Household Income, August 2009 ........................................................... 13 9% Annual and $1, contri 192 a $800 m bution amou ong individu nts from als withindividu $100,000 als have or mor a stro e in house ng imh pact old i on ncome. overall rol lover amounts, whereas employer 13th St. NW, Suite 878, Washington, DC, 20005-408% 51. Copyright 2010 by Employee Benefit Research Institute. All rights reserved. No. 343. $400 10% $993 www.commonwealthfund.org/~/media/Files/Publications/Fund%20Report/2000/Mar/Out%20of%20Touch%20%20 less than a $ $600 1,000 rollover had an avera $1,444 ge account balance of $939 in 2009 (Figure 21). In comparison, individuals $1,000 wheth $1,500er on financial data, options, or trends, are reve 9% aling and reliable — the re 9% ason EBRI information is 15% By Paul Fronstin, Employee Benefit Research Institute 10% 2.8 Whil With Education services, they e resp pane ect t $1,450 l I no will ternet surv firm spen size, t d teys are hh e e d money more ifferenc nonran es do betw rm, studies esponsibly, een CDH hav es P an pecial e de d tra monstrated that ly onc ditional e thplan ey become enrol such le surveys, when c es are more educated a not large areful but bout t ther ly design he a e isc one tual ed, 2. Checked8% the price of a$939 doctor’s visit, medication, or other health care service before I received care? $1,000 $400 contribution $600 $400 s do not. Individuals with an employer contribution of less than $1,000 had an average rollover of $993, Figure 13, Account Balances, by $600 $400 Education, August 2009 $1,163.7 ..................................................................................... 14 Table of Contents $400 2 $400 the gold sta 11% ndard for private analysts and decision makers, government policymakers, the media, and with American a rollover%20 of at M least en%20an $1,000 ha d%20td a he%2 n av 0Health% erage accou 20Cnare t bala %20 nce System/s of $2,474 andman (see n _outof ext section touch_3 ). 74%20pdf.pdf $600 10%* 10% 10% obtain dif price fere of n results ce healt wor hcomparabl t services. h mentionin Furt e to g: CDHP hermore, t random enrol -digit hes le -e accou dial t es are mor elephon nts ca e le survey in be kely used as than s. Tayl tradi tax-a or (2003), tional dvantaged veh plan for enr exampl ollees to icles to e, be prov save for em ides th ployed in hea e rel fir sult th m care s from s $400 Education has an impact on account balances independent of income and other variables, as they are controlled for in 0% while those 10% with an employer contribution of at least $1,000 had an average rollover of $1,188, but the difference was $1,000 $200 the public. Introduc Education tion .................................................................................................................. $1,400 .............................................. 4 (accesse 5%d March 2010). Figure 14, Account Balances, $950 by Plan Type, August 2009 ..................................................................................... 14 3% $500 3. Checked the quality rating of a doctor or hospital before I received care from them? expenses in with 2 a num - ber 49 of s wo $400 $400 rerkers u tirement. rveys that Excellent (Fig /Very ure 8). were co Good Althou nduct gh l ed at th arge employ Fair/Poore same At Least er ti s me usin are One muc g th h Health m e osame qu re likely tha Obese estion n small em naires Smokes both vi ployers to of a tele No Regular phon fer a e a CDHP, nd 2% the regression equation that these es EXEC The Em timatplo es are yee UTIV Benefi based t Research Institut on. In E SUMMAR dividuals with e (EB R a I) was founded in 1978. Its high scYhool degr ee or less have mission is to an not statisticall $200 $200 y significant (Figure 35). In contrast, individuals who contributed less than $1,000 had an average rollover About the 2009 EBRI/MGA Consumer Engage $200 $400 ment in Health Care Survey ............................................................ 4 $200 Good Chronic Health Problem*** Cigarettes 1% Exercise 1 Education has 5% $200 an impact on rollover amounts independent of income and other variables, as they are controlled for in $200 1.2 3%* Figure 15, Account Balances, by Health Behavior and Health Status, August 2009 .............................................. 16 this online. fin He ding m foua ny be d t $500 h ex at th plai e us ned e of by the demogra contribute to fact tph hic at w we , to encourag hen ighti lar Condition** ng alone ge em e, an ployers o was d to enhanc sufffer ficient a C e the developmen to DHP t brinh g almost all ey typica t of sound empl lly offer it of the resu as a c oylts fro ee ben hoic m eefit the $500 average of $1,350 $1,279 in their ac$302.9 count, while those with a college degree have $1,587, and those with a graduate degree of $750, while 0% those who contributed at least $1,000 had an average rollover of $1,350, a difference that is statistically Asset Taylor, Hum EBRI explo s and A phrey. “Does I ccount res the breadth of emplo Balances nternet ................................................................................................... Research ‘Work’? yee ben Comparin efits and related issues g Online Survey Results With Teleph ................................ 5 one Surveys.” $200 $0 Who we are $200 4. Talked to my doctor about the prescription options and costs? $900 By 2 the regression equation that 009, 15 percent of empl these es oyers with programs and so timat 10–49 es are 9 workers und public policy based on. In and 20 dividuals with p through objective ercent of th a ose w high sc resear ith 50 hool degr ch and 0 or educati more ee or on. EBR workers of less have I is the only fere an d $0 along with online survey othclose to th eLes r he s Talth han 6 Months pla e rn es, wher plies 6 Months from th eas w to Lese sh Ten a paralle han small l te 1–2 Year leph emsployer o one survey. ffe 3–4 Year rs a He also CDHP s foun it is 5 or typically th d t Mor he at in Years some c e onD ly a on't Know ses propensit available optio y n. $200 $0 Rollovers Checked Checked Price Checked Talked to Used Online Talked to Developed Asked for Asked Doctor have $ Figure 16, Accou 1,778 0% (Figure nt Balances, by Use of Co 13). It is possible 0.5 that more st and Qu highly e ality Information and ducated individuals understand t Use of Wellne he va ss Progra lue of saving mon m, ey in The Typical Account-B $- EBRI studie ased s the worl Health Pl d of health and retirement ben an Enrollee ................................................................................ efits — issues such as 401(k)s, IRAs, retire ..................... 5 ment signific Internation ant (Figure 36 al Journal o ). f Market Research, Vol. 42, no. 1 (August 2003). 1 privat 1 Year e, nonprofit, nonpartisan, Washington, DC-based organization committed exclusively to Checked Checked Price Checked Talked to Used Online Talked to Developed Asked for Asked Doctor Source: Source: $0 EBRI/MGA EBRI/MGA Consumer Consumer Engagement Engagement in in Health Health Care Care Surv Survey, ey, 2009. 2009. 0 either an H average rollov $1,300 RA $0eor HSA-eligible plan. r of Whether $879, Plan while t of Servh icose e As a r Qwith uality Rating a coll esult, these eg Doce tordeg About prlee h ans covere Cost T ad rac an king average d Doc 15–1 tor About 9 mi rollover llion Budget to p of $ eople 1,1 G in ener 61, a 2 ic009, Drn ug d those repr to Recommend esentin with a g ASSET LEVELS GROWING: Smokes In 2 Cigarettes 009, there was $7 Ex.1 ercbil iselion in consumer-drive Obesen health plans (CDH Health Ps), whic Problem* h include Less Than $30,000 2006 $30,000–$49,999 2007 $50,000–$99,999 2008 $100,000–$149,999 $150,000 2009 or more weighting (me $0 $0aning the propensity for a certain type of person to be online) reduced the remaining gaps, but in other aincom Whether e ade Plan quacy, co of Service nsu Quality mer-drive Rating Doc n b torenefits, Socia About Cost Tracking l Security, tax treatment of both retireme Doctor About Budget to Generic Drug to Recommend nt and health the account a August 20 $0 $0nd are th 09 ............................................................................................................................... erefore less likely to take distributions from the account. ................. 16 Account Balanc $0 Consumer e Variation Would Cov -driven er health Befor ..................................................................................................... plan. e Getting of Treatment Tool Provided Prescription Manage Health Instead of ................................... 8 Less Costly There is no use-it-or-lose-it rule associate 2006 public poli d with cy r HSAs, as es 2007 earch and any mon education on ey l 2008 eft in econ the omi accou c securi ntt at t y 2009 and em he end ploy of the ee benefi year t is sues. 5. Talked to my doctor about other treatment options and costs? White, Non- White, non-H His ispanic panic M Minor inority ity $0 $850 Would Cover Before Getting of Treatment Tool Provided Prescription Manage Health iIstead of Less Costly health gradu 9–11 perce a savings te degr nt of the ee accounts haCare d pr Smok an average ro Male ivatel (or es Cigar y insure HSAs) an Care ettes llover dDoc mar F d h emale tor e o /Hos k alth re et (Fronstin, f $ pital 1,1 E imburs O 39 xptions erc(Figu ise 21–34 and e 200 m re 2 by ent arra 9b). As t Health 9). Plan nge h35–44 e num O m ptions ents (or Obes and ber e Car of HR e Ex people As), sprea pens 45–54 es w Briand Name th acc Hd ealth acPr ross 5 millio ooblem* unt-based pl 55–64 Prescription n ans * Difference between traditional and CDHP is statistically significant at p = 0.05 or better. $0 Source: High Sc EBRI/Commonwe Male hool Graduate alth orF Les und Cons F semale Some umerCollege, ism in Health Trade or 21–34 Car e Bus Sur ines vey s, 2006. EBRI/MG College G 35–44raduate A Cons or umer Some Engagement 45–54 in Health Graduate Degr Care 55–64 ee cases it did no benefits, cost t reduce t High School heG r manage raduate emainin or Les g ment sgaps Some , worker a . Perha College, p Ts th rade or nd e employe Bus mo ines st 6sstriking dif r attitude College Grf aduate ere s, policy reform p nce in demographics betwee or Some rG opo raduate Degr sals, and p ee n te ele ns ph ion asset one s EBRI’s membership includes a cross-section of pension funds; businesses; trade associations; Gender and Age .................................................................................................................................................... 8 Care Care Doctor/Hospital Options and by Health Plan Options and Care Expenses Brand Name Prescription automatically rolls over and is available in the following year. But with HRAs, employers have a tremendous amount of Account Balance Variation ** Arthritis; asthma, emphysema Les Les or lung ss T T han $1,000 han $1,000 disease; cancer; depression; diabetes; Costs heart attack or Costsother heart disease; $1,000 or $1,000 or Mor Mor high choles ee Drug terol; or Drug Figure 17, Acc SurSo o vey unt Balances urce , 2009. : EBRI/Common, by Various wealth Fund Consum Co erism st Sc -Conscious in hool Health Care Surv Behaviors ey, 2006–2007 Graduate Wor ., EBRI/MGA Augus k Co t 2009 nsumer Engage ............................................. ment in Health Care 17 School Graduate Work Source: EBRI/MGA Consumer Engagement in Health Care Survey, 2009. Source: EBRI/MGA ConsumerLes Engagement s Than $1,000 in Health Care Survey, 2009. $1,000 or more grows, total assets in these G plans ender will grow as well. Cos ts Costs Age Age Drug Drug accounts. This is u and fundi a p from 2 ng. There is wi 006, Gender whe labor un n th deer spread ions; h e were ealth re 1.2 cog car mil e prov nition that if employee be lion acco iders and insur unts with ers; $8 government org 35.4 n m efits data exist, EBRI kno illion a in nizations; assets, and 2 and service firms. 008w , w s it. hen hypertension, high blood pressure or stroke. Health reimbursement arrangement. and online surv aeys was the under-representation of minorities in online samples. Race .......................................................................................................................... 6. Surv Us Sour ed an eyc, e: 2008–2009. EBRI/MG onlin Ae Cons cos umer t-trackin Engagement g tool in Health provi Car de Sur ed by my h vey, 2009. ealth plan to manage my .......................................... 12 health expenses? * Sour Health Co ce: nsum EBRI/MG problem er-driv defined Ae n Cons healt umer as h plan. fair Engagement or poor health in Health or one of ei Care Sur ght chr vey onic , 2009. health condition: arthritis; asthma, emphysema, flexibility in plan design: Leftover funds at the end of each year can be carried over to the following year or not, at the Type of Cov berage This section e *** Health problem x Health s amines var Sour avic ngs e: EBRI/Commonwe ac defined as icat ount. ion in fair or accou alpoor heal th Fund Cons nt bala th or one umer nces is of eight chronic m in Health by a num Car health conditions. e Sur ber o vey, 2006 f different –2007; EBRI/MG variaA b les, such as Consumer Engagement demograph in Health ics, income, a Figure 18, Account Balances, by Length of * Health He Sour alt pr ch e: oblem re EBRI/MG imburse defined m A e Cons nt as account. fair umer or poor Engagement health or Time With Acco in Health one of eight Carc e Sur hronic vey health , 2009. unt, August 2009 condition: arthritis; asthma, emphy ...................................................... sema, or lung disease; . 17 4.2 mi llion acc oro lung unts hel disease; c d anc $5. er7 billion ; depression; in diabetes assets. ; heart attack or other heart disease; high cholesterol; or hypertension, Conc lusion * Sour Difference ce: EBRI/MG between A Cons traditio umer nal Engagement and CDHPis in Health statistically Car significant e Survey, 2009. at p = 0.05 or better. Source: EBRI/MGA Consumer Engagement in Health Care Survey, 2009. Type of Household Inc Cov Sour erage Sour ce: o EBRI/MG cme e: EBRI/MG .............................................................................................................. A Cons A Cons umer umer Engagement Engagement in Health in Health Care Sur Care Sur vey, 2009. vey, 2009. ................................ 12 Sour Sour Sour ce: ce: EBRI/MG Car cEBRI/MG e: e Sur EBRI/MG v A ey A Cons , 2008 Cons A Cons umer umer –2009. umer Engagement Engagement Engagement in Health in Health in Health Car Car e Sur Car e Sur e Sur vey vey , 2009. v, 2009. ey, 2009. cSour anc b er ce: ; depr EBRI/MG ession; A diabetes Consumer ; hear Engagement t attack or other in Health hear Car t dis e Sur ease; vey high c , 2009. holesterol; or hypertension, high blood pressure, or stroke. employer’s dis high cretion, a blood press n ur d restrictions can e, or stroke. be placed on the amount that can be carried over. SourHe ce: alt EBRI/MG h savingsA account. Consumer Engagement in Health Care Survey, 2009. health HRAs and HSA status, health s are r be elatively haviors, and var new, but a io gr us measures o owing percen f cost-conscio tage of the po us dec pulation ision m has ahking ad th and he em for n alth en umerous year gagements. In . The In analyzing account balances for persons with individual coverage and those with family coverage, a number of factors EBRI delivers a steady stream of invaluable research and anal EBRI’s work advances knowledge and understanding of emplo ysis yee benefits and their Endnotes 7. Developed a budget to manage my health care expenses? Employers firs Figure 19, Accou Education ..................................................................................................................... t started o nt Balances, by Level of ffering HRAs in 2001, a Annual Employer Contri nd they were able to start offerin bution, Augu g h st 2009 ealth plans ........................................ 19 ....................................... 12 with HSAs in 2004. By While it was found that individuals with family coverage had higher account balances than those with individual importance to the nation’s economy among policymakers, the news media, and the public. It estimates wer 2006, 15 percent of e generat theed fr population om a regressi with an on equation HRA or HSA thhad at also hel control d the alccount fo ed for how r l lo ess than ng an i six mo ndividu nths, 42 al has had an percent HRA or for AVERAGE AC COUNT BALAN EBRI publications CE LEVELI inN clu G de in-de OFF: Increase pth cove s in aver rage of key issues a age account balanc nd tren es appear ds; summ to ha arie ve leve s of rese led ofarch f. In need to be considered. While individuals with family coverage often have higher deductibles than those with single Type of Coverage ................................................................................................................................................ 12 2009, 15 percent of employers with 10–499 workers and 20 percent of employers with 500 or more workers offered Overall, the percentage of individuals with a rollover has increased. In 2006, 23 percent of individuals with an HRA or coverage, the opposite was true with respect to rollover does this by conducting aand p mounts. Indivi ublishing policy re duals with sear sin ch, analysis, gle coverage and had a special reports on n average Figure 20, Account Balances, by Level of Annual Individual Contribution, August 2009 ........................................ 19 What we do 1 findings and policy developments; timely factsheets on hot topics; regular updates on legislative and 200 HSA, employ Assets an six months to 6, account e d Account Balances r bala a year, a contri nces ave butions nd 30 perc r to the aged $ ent accou 69for o 6. Th nn tey incr , individua e -twoe years ased l co t(Fi o ntributio $1, gur32 e 1 0 in n ). s to By 20200 the acco 07, a 9, 9 8 0perce unt, a perce n n n t ha d t incr unuse d h ease. A eld an d bal account ccount ance rol balanc l for over lees ss coverage, an8. d they a Asked for re statu a ge torily allo neric dru wed g ins to make h tead of a br ighan er co d nam ntrie butions drug?, they may also make more use of health care 8 See www.mercer.com/summary.htm?idContent=1364345 Health Behaviors and Health Status ................................................................................................................... 12 either an HRA or HSA-eligible plan.emplo As a r yee benef esult, these pl its issues; holding educational br ans covered 15 -19 mill iefings for EBRI memb ion people in 2009, ers, congressional and representing HSA did not roll over any money (Figure 22). The number not rolling over any money fell to 15 percent in 2007, and rollover of 5 $1,142, whereas those with family coverage had a $1,011 average rollover (Figure 30). Furthermore, unlike regulatory developments; comprehensive reference resources on benefit programs and workforce amounts. than six mont Figure 21, Acc hs, 20 percent ount Balances for si , byx mont Rollov he sr Level, to a year, Augus and 36 t 2009 perc................................................................... ent for one -two years. And as the length of........... 20 time averaged $1,356 in 2008 and $ 1,419 in 2009, 3 percent and 5 percent increases, respectively. Accordin services g to and tfihndin erefor gs from e have high the 200 er 9 EB costs. This RI/MGA Consum makes it dif er En ficult ga to geme det nt ermine in He wh altheth Car er fam e Surv ilies ey, $ will 7.1 have billion higher was i onrv low ested er 2 Use of Health Information and Pr federal agen ograms ........................................................................................ cy staff, and the news media; and sponsoring public opinion surveys on emplo .................... 12 yee 9 -11 percent of the privately insured market (Fronstin, 2009b). was 1 In th 6 eory, a perce r nt in andom sam 2008. By ple of 2009, 2,007 yi 10 perc elent ds a statistical did not have a rol precision lover. of pl Some of us or min the us 2. decrease 2 percent in tahge e perc points enta (wit ge o h 9 f 5 the difference in contri issuebutio s; and major survey n amounts, whic sh w of pu ereblic at not sttitudes. atistically significant, differences in rollover amounts by type 9. Asked my doctor to recommend a less costly prescription drug? benefit issues. EBRI’s Education and Research Fund (EBRI-ERF) performs the charitable, individuals have these plans increases, average account balances should increase as well. account balances than individuals. It was found that individuals with family coverage had $1,526 in their account in in HSAs a Cost-Consci nd HRAs in ous Behaviors ...................................................................................................... 2009, spread across 5 million accounts (Figure 2). In 2006, there were 1. 2 million ............................ 15 accounts with Figure 22, Rollover Amounts, Among Individuals With Account for More than One Year, 2006–2009 .................. 20 percent confidence) of what the results would be if the entire population ages 21–64 with private health insurance EBRI meetings present and explore issues with thought leaders from all sectors. individuals without a rollover between 2006 and 2007 may be due to the fact that the percentage of individuals who did of coverage were statistically significant. TYPICAL ENROLLEE: The typical CDHP educa en tiona rollee was mo l, and scientific func re likely tha tions of the Instit n traditiona ute. EBRI l plan enroll -ERF is a tax-ex ees to be youn empt organizatio g, unmarried, n 3 Length of Time With Account .............................................................................................................................. 15 $83 August 5.4 millio 2009, nwhile in assets, and by those with in20 dividual 08, 4.2 c o millio verage n accoun had $1 ts held $ ,444 (Fig 5.7 ur bi e 14); llion i ho n assets. wever, the difference was not statistically Gende Accordin r and g to fiAge ndin gs from the 2009 EBRI/MGA Consumer Engagement in Health Care Survey, there was $7.1 billion in coverage were surveyed with complete accuracy. There are also other possible sources of error in all surveys that may EBRI regularly provides congressional testimony, and briefs policymakers, member organizations, Figure 23, Total Rollover Assets an supported b d Number of Adult y contributions s and g Ages rants. 21–64 With a Rollover, 2006–2009 ....................... 21 Other tha not know w n ih ne questions ther they ha 1, d 3, and 7, a rollover ther or t e hwere no stat e amount de iscr tically signific eased from 19 ant di perc fferenc ent toe s 13 in avera percent. I ge acc n contrast, the ount balances This Issue Brief examines HSA and HRA assets, account bala Figure 16 nces, and rollover amounts. It examines the types of higher-income, educated, and exhibit healthy behavior. Figure 36 No differences were found between CDHPs enrollees and Figure 22 Figure 24 Figure 20 Figure 30 Figure 14 Figure 12 Figure 26 Figure 32 Figure 6 Figure 4 Figure 10 Figure 34 Employer and Individual Contributions ................................................................................................................ 15 signific be morant. e serio us than theoretical calculations of sampling error. These include refusals to be interviewed and other and the media on employer benefits. Figure 28 HSAs and HRAs in 2009, spread across five million acco Figure 18 unts. In 2006, there were 1.2 million accounts with $835.4 Men have higher account balances than women. By August 2009, men had an average of $1,620 in their HRA or HSA decrease in the percentage of individuals without a rollover between 2008 and 200a 9 can be attr a ibuted to an increase in between individuals who exh Account Balances, ibited the cost-consc b ious y Use deciof Cost sion-makin and Quality g behaviorsInf ano d t rmation hose who did not. With respect Rollover Amount, by Level of Annual Individual Contribution, 2009 traditional Figure 24, Average HRA or H individuals pla like nl enro y to have llees a Distrib Account Balances, wit cRollo o hnsumer-drive respect to Rollo Distrib SA Rollover Amoun A uvera tion ver Amount, Account Balances, Rollo ver Amount, u g Account Balances, of Rollo tion e gende ver Amounts, T HSA n hea raditional of ver Amount, b rT or HRA Rollo , race, y lb r th Ann aditional b y Use of Cost and Quality plan y t, 2006–2009 Gender and Plan Enr a u Figure 2 Figure 8 n al Household (C b Among d yDHP Plan presen b Plan Enr b y y ver Amount, Plan ) Le . It ollees Tce o Individuals ...................................................................... y the vel of Ann Ag pe To y fn , chi e Combined, and CDHP llees A pe ex Income ugust ,lamines 2009 dre 2006 and CDHP n u W Inf .al ,2009 iA di th o –ugust 2009 rmation f Enr f 2009 erenc ollees, 2009 es in account balance ....... 21 s by Account Balances, Rollover Amount, by Gender by Length and Ag of Time e Combined, With Account, August 2009 2009 Health Be Whil Rollovers ..................................................................................................................... e total assets in t haviors and He he acco alth Sta unts have tus incr eased each year, increases in average account balanc ........................................ 18 es appear to have forms of nonresponse, the effe Rollo cts of qu ver eAmount, stion wor b din yg and qu Annual Household estion order, aIncome nd screenin , 2009 g. While attempts are made to EBRI issues press releases on newsworthy developments, and is among the most widely quoted Account Balances, by Length of Time With Account, August 2009 million in assets, and by 2008, 4.2 million accounts held $5.7 billion in assets. While total assets in the accounts have EBRI Issue Briefs are periodicals providing expert evaluations of employee benefit issues and while women had $1,366 (Figure 9). This may be due to the fact that men use less ahealth care than women, which and Use of Wellness Program, August 2009 to wh the percentage with a ether individuals c rollov hecke er of between d their heaT lth otal Assets and Number $500 plan to see i and $999, as f the plan well wou as a of Adults ln d incr cover ease am their cong those wit are or medicat h a ro ion an llover o d whet f at her Distribution Enr b of Individual Contrib y o and llees, T Education, raditional Plan Enr Use of by Ag W Adults Ag ee , Adults Ag llness Pr utioollees and CDHP n,es 21–64, ogram, Aes ugust 21–64, 2009 2009 2009 2009Enrollees, demographics, income, contribution Account levels, and for More engagem than One ent in an in Year div,i dual 2006 ’s ow –2009 n health care, using a regression minimize these factors, it is impossible to quantify the errors that may result from them. levele Rollovers ..................................................................................................................... d off. In 200 sou 6, acco rces o unt n bala employee be nces averaged $ nefits by all media. 696 (Figure 3). They rose to $1,320 in ............................................. 18 2007, a 90 percent increase. $1,600 $1,200 $1,600 $2,000 $1,400 Figure 25, Rollover Amount $2,000 $3,000 , by Gender and Age, 2009 ......................................................................................... 23 Individuals who smoke had higher rol trends, as well lover amounts th as critical analy an nonsmokers, and ses of employee benefit po the differ licies an ence is d proposals. statistically EB sign RI Notes ificanis a t increased ea Our ch year, increases in average account balances appear to have leveled off. In 2006, account balances Health Behaviors and Health Status allows them to maintain a higher account balance (Sandman, Sima a ntov, and An, b 2000). lMORE ROLLO east $2,000. VERS: Overall, the num by Firm Siz ber of peopl e, W e orking Adults Ag with a rollover, as well es 21–64, as the tot 2009 al level of assets being rolled over, they checked the quality rating ofAg a doctor es 21 or hos –64 With pital an beHRA fore reor HSA, ceiving care from 2006–2009 them, it was found that those $1,295 3 equatio Gender an n. Rol lo d Age ver amounts EBRI direct ................................................................................................................ are th s membe en exam rs ained. nd other constit uencies to the information they need, an .................................. 18 d undertakes new monthly periodical providing current information on a variety $1,526 of employee benefit topics. Account balan 35% $1,600ces averaged $1,356 in 2008 and $1,419 in 2009, increasing 3 percent and 5 percent, respectively. $2,000 50% 30% (Figure 31). Those who exercise had higher rollover amounts than those who did not, but the difference was not The term $1,700 assets is used loosely as they relate to health reimbursement arrangements (HRAs). HRAs are typically set averaged $696. This increased to $1,320 in 2007, and then averaged $1,356 in 2008 and $1,419 in 2009. Figure 26, Rollover Amount $2,500 $1,600 , by Gender and Age Combined, 2009 ........................................................................ 23 $1,766 $1,890 $1,142 $1,774 Indivi have bee duals n$1,800 in wh cro smoke have easing. 28% The a28% vmore money erage rollove ir incre n their acco ased 28% fro unm ts $5 than i 92 in ndivi 20 28% d 0u 6als to w $1, ho do 295 in no 20 t, but 09. the difference is not exhibiting such behavior research on a had lower n ongoi EBRI’s avera ng ba g Pe ac ension Inv sis. count bal estme ances nt Re tha port nprovides detailed fin those who did no at exh ncial iin bit formation on th such $1,385 behavior. Als e universe of o, $1,150 46%* Race .......................................................................................................................... $1,800 .......................................... 22 publications $1,500 60% $1,400 up as notional arrangements and exist only on paper. Employees may view the account as if mone $2,563 y was actually being $8,000 27% 7 statistically significant. $1,400 Obese individuals had lower average rollover amounts than $1,660 nonobese in31% dividuals, a statistically With respect to age, there is no difference in account balances by age for those below age 55. Individuals ages 55 -64 The number o $1,200 f people with a rollover as well as the total level of assets being rolled over has been increasing. In 2006, 45% defined b $1,444 enefit, defined contribution, and 401(k) plans. EBRI Fundamentals of Employee EBRI maintains and analyzes the most comprehensive database of 401(k)-type programs in the statistically significant Figure 27, Rollover Amount $1,800 (Figure 1 , by Race, 2009 5). Similarly, obese ..................................................................................... individuals have less money in their account th $1,606 an nono ...................... 24 bese indivi Hous duals ehold Inc who deve ome lope .............................................................................................................. d a budget to manage their hea Yes lth care ex Nopenses had a higher account ................................ 22 balance ($1,726) than $2,500 deposited into an account, but employers do not incur expenses associated with the arrangement until an employee $1,400 $1,650 $1,350 The number o 30% f people with a rollover as well as the total level of assets being rolled over has been increasing. In 2006, significant $1,600 difference. There was no statistically significant difference in account balances Trby healt aditional hCDHP status. had higher $1,400 account balances than you Benefit Programs nger individuals offers a straightforward, b (an average of $2,0 asic exp 17 in their lanation of account, emplo com yee b pare enef d it programs in with roughly Traditional CDHP $7,094.8 $1,340 500, DIFF00 ERENCES I 0 individuals rol Nworl ACCO d. Its co led UNT ovBALANCES: er $ mputer 302.9 m simul Men te illion (Fi ation analy nd to hav gure 2s 3e ) in H e s hi on So gher SAs an acco cial Secu d unt HRAs. By 2 ba rity reform lances t 009 h, 3. an and retireme 1 mil women, lion a in ccount divi nt income a duals rol balanc le d es d ove equarcy $1,600 $1,100 Male Male Fe Fem ma alle e $1,320 individuals, 25% but the difference is not statistically significant. There was a statistically si $1,192 gnificant difference in account those who Education ..................................................................................................................... did not develop such a budget ($1,428) (Figure 17). This was the o $1,205 nly statistically sign ....................................... 22 $2,040 ificant difference that incurs a claim. $1,199 $1,614 $1,400 Figure 28, Rollover Amount, by Annual Househol 2006 2007 $1,453 d Income, 2009 2008 2009 ......................................................................... 24 $1,200 40% $7,000 the private and public sectors. The EBRI Databook on Employee Benefits $1,988is a statistical 23 percent of individuals with an HRA or HSA did not roll over any money. By 2009, 10 percent did not have a rollover. 23% $1,600 $1,200 6 $1,400 for individuals under age 55 are unique. ). This occurred despite the fact that o $1,000 lder individuals use more health care on About the 2009 EB increase with $2,000 household incRI/MGA Consumer En ome, education has a signifi Yes cangagement in Health Care Survey t im Nopact on account balances independent of income and nearly $4 bi 50% 50% llion. The average rollover increased from $592 in 2006 to $1,295 (Figure 24). 48% 48% 48% 48% The Typical Account-Based Health Plan Enrollee $1,221 Type of Coverage ................................................................................................................................................ 22 balance by level of exercise: Individuals who exercise have an average of $1,647 in their account, while those who do 4 was identifie $1,600 $1,000 d where individuals exhibiting the cost-conscious behavior had a higher account balance than those not $1,400 reference work on employee benefit programs and work force-related issues. www.ebri.org $948 $1,378 $1,200 35% $2,028 The conditions are arthritis; asthma, emphysema, or l Traditionalung di CDHP sease; cancer; depression; diabetes; heart attack or other In 2006, 5$1,050 $1,200 00,000 individuals rolled over $302.9 million. Furthermore, by 2009, 3.1 million individuals rolled over nearly Figure 29, Rollover Amount 25% $1,400 , by Education, 2009 ................................................................................ 21% .................... 25 average than younger ones. It is possible that older i $1,026ndividuals are saving the money in the account to use to pay for Use o other varia f He 35% b alth Informatio les, and no statin and Progr stically signific ams ant differences in account balances were found by smoking, obesity, or the The Employee Benefit Research Institute (EBRI) and Mathew Greenwald & Associates (MGA) created the Consumer Health Behaviors $2,000 23% and Health Status ................................................................................................................... 22 not have The typical $1 CDHP enrol ,246. lee was more likely than traditional plan enrollees to be young, unmarried, higher-income, exhibiting suc $1,300 h behavior. Assets (millions) $1,400 $6,000 $1,679 $993 heart disease; $1,000 high cholesterol; or hypertension, high blood pressure $5,697.1 or stroke. $1,011 $4 billion. The average rollover increased from $ 31%592 in 2006 to $1,295 in 2009. The remain EBRI make 20% der of t s information freel his section examines vari y av ation in ailable to al rollover amounts. l The estimates in this section were also generated health care $1,000 expenses in retirement, but it was not possible to determine that from the survey. Prior research 5 has found presence of $1,200 chronic health conditions. Individuals $944 who developed a budget to manage their health care expenses had a Figure 30, Ro Use of Health Inform $1,550llover Amount ation and Pr , by Plan Typ ograms ........................................................................................ e, 2009 .................................................................................................... 25 .................... 22 Engagement in Health Care Survey (CEHCS) to examine issues surrounding consumer-directed health care, including Rollover amounts were also examinContact EBRI ed by wheth ePublications, (2 r individuals used 02) 659-0670; cost and qua fax publication lity information to orders to choose a (202) 775-6312. doctor and educated, and exhibit healthy behavior. It is not known if the health be $990 havior differences are due to the impact of the $1,200 $1,000 30% $970 20% 5 $1,000 40% 5.0 EBRI assumes a publi Accounts c service re (millions)$960 sponsibility to make its findings completely accessible at www.ebri.org $1,000 $1,487 $800 from a A regression regression e equatio quatio n is n a s thtat also co atistical mode ntrolled for l that allows how l r oe ng a searchers to n individual has determi had a ne thn e effect o HRA or HSA, f an i employ n19% depend er ent variable that, while 20% HSAs can be used to save for health care expenses in retirement, they are far from sufficient because of $1,200 Subscriptions to EBRI Issue Briefs are included as part of EBRI membership, or as part of a To measur high Cost-Consci er acco $1,500 e h une t bala alth status, s ous nce Behaviors ...................................................................................................... ($1,e 726 lf-r) t eport han t ed h hoe se who alth status did $911 no wa t ($ s combin 1,428)ed , bu $1,419 with t other whw ether ise, n an in o statistically s dividual hai............................ 22 d on gnificant e of e di igh ffet rences the cost of insurance, the cost of care, satisfaction with health care, satisfaction $1,412 with the health care plan, reasons for $1,200 wheth healthe pla r th ney or partici wheth $5,000 peated in r healt a hy peo wellness pr ple $1,100 disproportionate ogram. It was ly foun gravitate d that i to C ndividuals w DHPs. No di ho ffused cost a erence 18% wasn d fou quality nd bet inform ween C ation to DHP Figure 31, Rollover Amount, by Health Behavior and Health Status, 2009 ........................................................ 16%* .... 27 Orders/ It is expected — so that all deci that by giving indivi siod ns that uals mo relate to emp re control ove loyee ben r funds allocat efits, wheth 7 ed for health er made in Cong care services, ress or bo they will s ard room pend the s or Length of Ti $1,000 $800 me With Account $851 on a dependent variable while holdi $199 annual sub ng the effect o scription f a 24%* ll ot to her i EBR nde I Notes pende and nt varia EBRI Issue Brie bles constant. For fs. Individual thi cop s analysis, the ies are availabl e contributions to the account, $912 individual contributions $1,301 to the account. $1,500 $1,500 statutory constraints pla Length of Tim 25% $800 e with Account ced on contri ................................................................................................... bution levels relative to expected health care spending in retir............................ 26 ement (Fronstin chronic co in averagendit accio ount balanc ns. It was fo es und that t were foun hd be ere was tween no statistica individuals who lly signific exhibi a 23% nt di ted vario fferenc ue s as in accou pects of nt bala cost-c nces by onscious dec 4 healthi sion- 16% choosing a $1,000 pla $950n, and sources of health information. The 200$1,356 9 EBRI/MGA Consumer Engagement in Health Care Survey choose a doctor had a higher rollover amount than those who did not participate in such a program. The average enrollees an 15% d families’ ho traditional pla men s, are enrollees based wit on the highe h respect to g st qualit ender, race, y, most depe and presen nda ce o ble informatio f children. n. EBRI’s Web site posts money more responsibly, $800 especially once they become more educated about the actual price of health care services. amount of$1,000 mo $800ney an 15% individual has with prepa in$750 an HRA or HSA yment for $25 each is determine (for printed d by a num copies). ber Change of Address: of factors. The re EBRI, 1100 13th St. gression equation Figure 32, Rollover Amount, by Use of Cost and Quality Information and Use of Welln 4.2 ess Program, 21%* $592 Not surprisingly, the length of time that an individual has had th 14% e account has a major impact on the amount of money 201 Employer an 0). Subscriptions d Individual Contributions ................................................................................................................ 26 14% 15% 30% $609 status. making behavi $600 ors and those who did not. is compara $1,100 ble $800 with $4,000findings from the $830 2005, 2006, and 2007 EBRI/Commonwealth Fund Consumerism in Health Care all research findings, publications, and news alerts. EBRI also extends its education and public service rollover was $1,199 among th $1,050 ose who used cost an NW, Suite 878, Washington, DC d quality informati , 20005-4051 on to choose , (202) 659-0670; f a doctor, a and x number, (202 $944 amo ) 775-6312; ng those 20% 13% 13% allows researchers to determine the strength of each factor independently. More information about the regression However, limit $600ed evidence was found to support this. Individuals who developed a budget to manage their health care 2009 .......................................................................................................................... $1,450 .................................. 27 $1,435 in th Conc e accou lusion .................................................................................................................... nt. The analysis found that persons with an account for less than six months had an ........................................... 26 average of $842 in $800 $800 $900 $900 12% 12% The Gende data r and for $1$600 $600 000 a role to improving Ameri Age ge dif ferences betwee e-m n C c aan il: DH s’ finan subscriptions@ebri.o P enrocial kn llees an owle d tra r dge gd itional thro Membe pl ugh its a an en rship Information: rolle ward es c -win an be se ning p Inquiries reg en i ubli nc se Figure rvice arding EBRI 4. O cam ne-fifth paign $1$800 ,000 Survey and the 2008 EBRI/MGA Consumer Engagement in Health Care 11% Surv 11% ey. 3 who equatio did not n is availabl use cost and qu e upon reality in quest from t formati ho e aut n (Fih gor. ure 32). The difference in rollover amounts was statistically significant. $1,413$3,037.0 expenses ha $600 $600 d a higher account balance than those who did not develop such a budget. Otherwise, there were no ® $1,000 10% 10% 10% 10% 10% 10% The higher ac $600 count balance among older indivi 21% 21%duals * may also be due to the fact that they are allowed to mak 10% e “catch- D their ac Referenc IFFERENCES I count es .................................................................................................................... and t N ROLLOVER AMOUN hose with the accou TS: nt Men roll at least six mo $875 ed over more nths but mone less t y than han a year wome ha n, whit d $87 es have .......................................... 30 5 (Figur high e 18e ). r rol In com loverp arison, Figure 33, Rollover Amount 15% , by Various membership and/or contributions to EBRI-ERF Cost Conscious Behaviors, 2009 should be directed to EBRI President/ASEC ............................................................ 28 10%Ch $ $3,000 oosetoSave and the companion site www.choosetosave.org $1,000 $842 9% 9% 9% 9% (21 perce10% nt) of traditional plan enrollees were ages 55 -64, while 16 percent of CDHP enrollees were ages 55 -64. The Men 6 rolled over more money than women. In 2009, men had an average rollover of $1,140 while women had $957 Use o There f w He as alth Informatio no statistically significant n and Progr differe ams nce by participation in a wellness program. 20% $1,400 $400 8% 8% 8% statistically significant $600 $400 differences in average account balances between individuals who exhibit 11% ed various aspects of Individuals $850 are also able to roll over Chairman Dallas funds from o Salisbur ne HSy A into a at the above nother addr HSA witho ess, (202) 659-0670; ut subjecting the e-mail: salisbur distributio y@ebri.org n to up” contri Endnotes ...................................................................................................................... buti $600 ons that individuals under age 55 are not allowed to make. However, the re ............................................ 30 gression equation controls for indivi amounts than duals wit h minorities, a an accountn fd th or one e youn -twges o ye t adu ars ha lts an d an d th a 8%verage of $1, e oldest adults 356. had the Those lar wit gh e an st rollover am account thounts in ree -four y 20e 09. ars had The 2009 survey was condu 7% cted within the United States between August 8 and Au 7% gust 20, 2009, through a 14-minute 7% difference was $400 statistically significant. Figure 34, Rollover Amount, by Length 15%of Time With Account, 2009 ..................................................................... 28 (Figure 2510% ). 6% 6% 2 $400 $400 2.3 income and penalty ta $400 xes as long as the 8%* rollover does not exceed 60 days. Rollover contributions from Archer medical Account cost-cons balan ciousc dec es wer ision-maki e examn ined by g beha two viors variables to and those wh me o di asure d not an e in xhi dibit s vidual uch ’s en begagement havior. Howev in his o er, irt her was h foun ealth d th care: at individual contributions, and thus would not be a factor in the observed differences. 5% an avera EBRI is sup ge of $2,208. A $2,000ported b nd those wit y organizations from all industries and sectors that appreciate the value of h the account at least five years had an average account balance of $2,563. Rollover amounts increase with household income and educatio12% n, and individuals with single coverage rolled over a Internet surve $900y. The national or base sample was drawn from Synovate’s online panel of Internet users who have $1,350 $500 Editorial Bo 5% ard: Dallas L. Salisbur 1.2y, publisher; Stephen Blakely, editor 11% . Any views expressed in this publication and those of the authors should $400 $400 savings accounts ( $500 $800 MSAs) are also permitted. Figure 35, Rollover Amount , by Level of Annual Employer Contribution, 2009 10% ...................................................... 29 (1) whether a 5% n individual used cost and quality information to choose a doctor, and (2) whether he or she participated Cos indivi td -Con uals sci wh $200ous Be o talked to t havio heris r doctor about treatment options and costs, those who used an online cost-tracking tool 5% unbiased, reliable information on employee benefits. Visit www.ebri.org/about/join/ for more. slightly In terms o not be higher ascribed to the officers, 10% f i $200 nc average amo ome, CDHP enrollees unt tha trustees, n thos wer mem eb e ers, or more likely th with fam other sponsors of the E ily co an trad verage. itional Thm ere ploye plan was e Benefit Research enrol no statistically signific lees to beInstitute, the EBRI Educ in houseahnolds t diffwith erenc ation and $5 e i 0n ,000 agreed to part $200 icipate in research surveys. Over 2,000 adults (n=2,007) ages 21 -64 who have health insurance through With respect to age, the youngest adults and the oldest adults had the largest rollover amounts in 2009. Individuals $200 7% 7%* $200 7 $200 1 There were differences in account $835.4 balances between men and women when examined by age: Men ages 21–34 had a Figures in a provide wellness d by th program, e health such plan, as one and th design ose who as ed to ked t direh ctly eir doctor to r improve hea eco ltmmend h, like a a l wei esg sh cos t loss, nutritio tly prescripn tion dr , stress ug had higher More i Resear nforma ch Fund, t $1,000 ion abo or their staffs. ut th 5% Nothin e regr g her ession equation ein is to be constr is ued as an attem available upo pt to aid or n request from hinder the adoption the author. of any pending le gislation, regulation, Figure 36, Rollover Amount, by Level of Annual Individual Contribution, 2009 ...................................................... 29 A number of cost-conscious behaviors were examined to see if individuals who exhibit more cost-conscious decision or more in$1,300 i$800 $200 ncome. Forty-five percent of CDHP enrollees had household income of $50,000 0% -$99,999, compared with rollover amou $200nts by health status, although individuals who smoke had higher rollover amounts than those who do not an employer or purchased directly from a carrier were drawn randomly from the Synovate sample for this base sample. 55 Employer an -64 had an $750 average rol d Individual Con lover of $1, tribution 288, s compared wit h $1,163 for individuals 21 -34, $933 for individuals 35 -44, 3% 0% or interpretative 0% rule, or as legal, accounting, actuarial, or other such professional advice. high managem er acco ent, un $0 $0 t bala or smoking c nce than e wome ssation progra n of them, or a same a healt ge. Th he avera risk assessment-type ge account balanc progr e fa om that r r men was equir $1 es partic ,679, whil ipants e for to fill 8 rollover amounts than those who did not take such actions. Future research should examine differences between Figure 1, Length of Time With HRA or HSA, 2006 and 2009 .................................................................................... 6 making had higher rollover amounts than those who did not exhibit such a decision-making process. The expectation and o 38 See perc b www.m ese in ent ofdi etraditional viduals rcer.com/sum High School had lo pla Graduate nwer av mary enro or Les .htm?idC llees erage r s (Fi Some g o ontent=1 ure llover amou College, 5). S Trade or 36 imila 43 n Bus trly, 24 45 s than ines s percent no College G nobe of C sre aduate inD divi HP or duals Some enro.llees h Individua aG d hous raduate Degr ls wh ehold incom o talke ee d to t e hof eir This sampl e was stratifie $0 Les d by gens Than der, $100–$299 age, region, i $300–$499 ncome$500–$999 , and race. $1,000–$1,499 The resp$1,500–$1,999 onse rate was $2,000 27. or Mor 4 epercent Don't Know (21 percent and $927 for i 0% $0ndividuals 45 -54. $- $0 $0 $0 Annual contribution amou Les Les ss T T han nts, wh han $50,000 $50,000 ether they come $50,000–$99,999 $50,000–$99,999 from the employer$100,000 (i $100,000 n the or or case o Mor Mor ee f both HRAs an Dec Dec lined d lined HSAs), or from 0% $0 women it was $0$1,0 $050 (Figur$100 e 10). The difference Sc betw hool een men and w Go raduate Wor men can k also be seen at ages 35 0 -44, but it is out a individquestion uals in HSAs and HRA naire and thens in co , and sho nsultatio uld also exami n with a me ne t dical pr he imo pact t fession hat al examine th account bala eir h nces have on ealth histou ry to identify se of health c any are $1,250 $0 $700 $700 was that those who e< 6 Months xhi21–34 bited cost-conscio 6 Months us be to 1 Year 35–44 havior would 1–2 Year have s a hig45–54 her rollover 3–4 Year a smounts than 5 55–64 or t Mor he ose w Years ho did not. $100,000 -$149,999, compared with 21–34 21–34 Used Cost or 1 Q7 uality percent Information of 35–44 35–44 traditional plan enrollees. Ther 45–54 45–54 Used Wellnes e was s no di Program ffer 55–64 55–64 ence in the doctor a Figure 2, Total Asset bout treatment o s and Number of Adults pti 2006 ons and costs, those Ages 21–64 With an HRA who 2007 used an online cost-tracking to 2008 or HSA, 2006–2009................................. ol provided 2009 by the health plan, and 6 for the base s EBRI Issue Brief ample or nat Self-employed With is registered in the U. ional sampl No S. Patent and T e, a2–49 nd 38 radem perc ark ent Office. for the oversa I50–199 SSN: 0887 ?13 mple). The m 7X/90 0887 200–499 ?13argin 7X/90 $ . of 50+. error 50 500 or for the natio More nal Less Than $1,000 $1,000 or More 2006 Less Than $1,000 2007 2008 $1,000 or Mor2009 e individuals (as they apply to HSAs only) have a strong im Annual Hous pact on ehold Inc overal omel account balances. Not surprisingly, the more 1–2 Year Used Cos s t or Quality Info 3–4 Years Used Wellness5 Pr or ogr Mor am e Years Indiv Indiviidual Cov dual Cover erage age F Family amily Cov Cover erage age conditions smaller, as me they may hav n have an ave e or might rage account be at risk of balanc dev e ofe $ lo1ping, ,487 w inhile order woto deve men hav lop a e anprogram average accou for earn ly interv t balance ention of $1,3 . In 01. services as individuals co Employees Employees ntinue to accumulate funds in their accounts. The questions regarding cost-conscious decision making were described above. thos percentages e who as o Sour kfe those with in d t ce: EBRI/MG heir docto A Cons r come of to recomm umer Engagement at le ea n in Health sd a t $15 les Car 0,0 s cos e Sur 00. vt ey ly, 2009. prescription drug had higher rollover amounts than those who sample was ±2.2 percent. th Source: EBRI/Commonwealth Fund Consumerism in Health Care Survey, 2006–2007; EBRI/MGA Consumer Engagement in Health Figure 3, Average Source: Sour EBRI/Commonwe HRA or HSA ce: EBRI/MGA al Cons th FAccount umer und Cons Engagement umer Balanc ism in Health in Health es, 2006–2009 Car Care Sur e Surv vey ey, 2009. , 2006–...................................................................... 2007; EBRI/MGA Consumer Engagement in Health Care ......... 7 money that is contributed to the account, the high1100 13 er the average Street NW · Suite 878 account balance. For instance, individuals with an a Source: EBRI/Commonwealth Fund Consumerism in Health Care Survey, 2006–2007; EBRI/MGA Consumer Engagement in Health The difference Co in nsum account er-driven he ba altlances h plan. between men and women disappears for those ages 45 -54 and 55 -64. both cases, it was fou a nd that individuals who participated in a wellness program had a higher average account balance Source: EBRI/MGA Consumer Engagement in Health Care Survey, 2009. Sour Sour ce: Cons c EBRI/MG e: EBRI/MG umer-A dr Cons ivA en health plan. Cons umer umer Engagement Engagement in Health in Health CarCar e Sur e Sur veyv , 2009. ey, 2009. Sur Source: EBRI/MGA Sour vey Sour , 2008 cCare e: ce: EBRI/MG EBRI/MG –Surv 2009. ey,A Consumer 2008–2009. Cons A Cons umer umer Engagement Engagement Engagement in in Health Health Care Survey, in Health Car Car e Sur e Sur vey v2009. ey , 2009. , 2009. did not take su Sour ch actio ce: EBRI/MG ns. A Consumer Engagement in Health Care Survey, 2009. Sourc Sour e: EBRI/MG ce: EBRI/MG A Cons A Cons umerumer Engagement Engagement in Health in Health Care Sur Car Washington, DC e Sur vey, 2009. vey, 2009. 20005 Sour Carc e Sur e: EBRI/MG vey, 2008 A – Cons 2009. umer Engagement in Health Care Survey, 2009. * Differenc Source:e EBRI/MGA between traditional and Consumer Engage CDHP meis nt s in tatis Hetic alt ah lly Care signific Surv ant eyat p , 2009. = 0.05 or better. Source: EBRI/MGA Consumer Engagement in Health Care Survey, 2009. Sour a c* e: Differenc aEBRI/MG e between traditional and A Consumer Engagement CDHP in Health is statis Car tic e Sur ally svignific ey, 2009. ant at p = 0.05 or better. Note: Consumer 2006 data Healt -driven h w reere im health burse capped at m plan. ent $1,500 or arrangemen mor t.e. Figure 4, Distribution of Traditional Pl © 2010, Employee Bean Enrollees and CDHP Enrollees, by nefit Research Institute ?Education and Rese Age, Adults Ages arch Fund. All rights re 21–64, 2009 served. ......... 7 (202) 659-0670 b * Difference between traditional and CDHP is statistically significant at p = 0.05 or better. Health savings account. www.ebri.org www.choosetosave.org ebri.org ebri.org ebri.org A research rep ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org Issue Bri Issue Bri Issue Bri Issue Bri Issue Bri Issue Bri Issue Bri Issue Bri Issue Bri Issue Bri Issue Bri Issue Issue Issue Issue Issue Issue Issue Issue Issue Issue Issue Issue Issue Issue Issue Issue Issue Issue orte e e e e e e e e e e from the EBRI fffffffffff • June 2010 • June 2010 • June 2010 • June 2010 • June 2010 • June 2010 • June 2010 • June 2010 • June 2010 • June 2010 • June 2010 Brief Brief Brief Brief Brief Brief Brief Brief Brief Brief Brief Brief Brief Brief Brief Brief Brief Brief • • • • • • • • • • • • • • • • • • June June June June June June June June June June June June June June June June June June 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 Education and R • No. 343 • No. 343 • No. 343 • No. 343 • No. 343 • No. 343 • No. 343 • No. 343 • No. 343 • No. 343 • No. 343 • • • • • • • • • • • • • • • • • • No. No. No. No. No. No. No. No. No. No. No. No. No. No. No. No. No. No. 343 343 343 343 343 343 343 343 343 343 343 343 343 343 343 343 343 343 esearch Fund © 2010 Employee Benefit Research Institute 27 20 23 13 25 29 10 28 14 21 19 17 16 11 24 9 6 7 18 15 30 22 26 12 4 2 8 3 5 $ (millions) $ (millions)