How Do High-Deductible Health Plans Affect Use of Health Care Services and Spending Among Enrollees With Mental Health Disorders?

In this paper, we explore the impact of moving from a preferred provider organization (PPO) to a high-deductible health plan (HDHP) among people with mental health disorders. We focus on individuals with major depressive disorder (MDD), anxiety, and attention deficit hyperactivity disorder (ADHD) since these represent the majority of mental health patients. Furthermore, the prevalence of these conditions in the population with employment-based health insurance is especially high relative to other mental health conditions, and it has been increasing. In our analysis, we observe:

Moving from a PPO to an HDHP reduced the probability of using health care services.
The amount of health services used was also lower among patients with mental health disorders who moved from a PPO to an HDHP. Office visits, prescription drug fills, inpatient days, and emergency department visits all declined.
Utilization of preventive services including cancer screenings and some vaccinations was also negatively affected by the move from a PPO to an HDHP.
The reductions in use of health care services prompted declines in overall health care spending. Employer spending fell by a greater amount and percentage than overall spending. Employee spending increased because the move from the PPO to the HDHP shifted some pre-deductible spending onto users of health care.
Moving from a PPO to an HDHP had a mixed impact on use of out-of-network health care services. While the probability of using an out-of-network mental health provider fell among those with ADHD and MDD, neither the number of office visits nor overall spending on out-of-network care were affected by the plan change.

When employers raise deductibles, they do so to better manage their costs. They are trying to balance their efforts to reduce low-value health care services while incentivizing the use of high-value services. Our findings are limited by the fact that outcomes related to changes in health care use during the switch from PPO to HDHP are unobserved. Such outcomes would ultimately be impacted by the value of care received, for which data is unavailable for this analysis.

Our research can help employers make targeted benefit design decisions. They can also inform policymakers as they consider allowing employers to provide enhanced coverage for health care services that prevent the exacerbation of chronic conditions.

This study was conducted through the EBRI Center for Research on Health Benefits Innovation (EBRI CRHBI), with the funding support of the following organizations: Aon, Blue Cross Blue Shield Association, ICUBA, JP Morgan Chase, Pfizer, and PhRMA.

Figure 1 Figure 9 1 Figure 18 Figure 7 Impact of HDHP on Probability of Receiving Various Types of Prescription Drugs Annual Spending on Various Types of Health Care Services Among Individuals Probability of Using Out-of-Network Health Care Providers, Among Individuals Figure 20 Figure 14 Annual Use of Various Types of Health Care Services Among Individuals Ages Among Individuals Ages 18 –64 With Depression or Anxiety Ages 18 –64 With Depression or Anxiety and Individuals Ages 5 –24 With ADHD Ages 18 –64 With Depression or Anxiety and Individuals Ages 5 –24 With ADHD I Ma m Fi sele m S S int Da B Gundo m ood t g e e eud e ny ur n a nt ns nde t c pac dte a a w y rion) he nd l he r ug itivit in, Re 7, f, M d Va a t tw , A spe lt a o ria nnu h se a . o om ltts w K h ne ne rf nding y e a b e a H a t r e le n w e e l U A v t ll a e ,DHP D ic s a naly d " ., e se n it C m s b s s ut h M onsu e A e q of xis ndrea e a ua c ili su s o D Va t aza nt ing m e D us rn e rs ify e tio s, a e ir H H on ill nd - us ing of . D w ne e W ir a e Ty a alt te nd ss the e nxi e he cinbe p , m t e e h Care spe e inj cp d s of im onne tloy y ur rH . p g nding e a y e e H The a , c rd c, lt e tor t a g ion b h P June of lt e out m c S h C ne H ond la ov pe D r c ns e H a a om H e tit liz w . r: m ndi P e ion, K e D s on e e e im Se e d s o nt n ng w , lin r T i nc v MD fMe he e ov ric e om rluding a e e y e D lod r s A r e D a m P a xa ll use e y nd P m od liv Os t m W ong d e e p ine r u, lrugs rs ? o o of ,a d w I d " H ndivi nd . uc it se Re or D h g These tr H Sa iv sea v m P d ic ia tn s d e ua y m ed r d s a clos r m ic ls id h Synt inc o a a nd A ses G tnot lude ions g fa ae lspe m e s (a he a G ." ily 18 ff , inpa nding oe si e " – a A T c s Re t64 nd tr s our ze te sc ie nds l e . W log nt r p e or S t i e t st hospit th D in A a nin t lin udie l. udy sa Nk20 g o. e . nxi p s for a 18 r A 24 ha e e l a ll m ss t ) y c v . ( d p olor ion o e Rob m A le G m e is iv xa inc e on si e ec rrons m n th ttlg a ude ine l ed d Annual Spen Imp din act g o on f H Ou Dt H -of P - o N netw Use oro k f H Vealth ariou C s are TypServ es of ices Screen Amo in ng gs Individuals Paul Fronstin is Director of the Health Research and Education Program at the Employee Benefit Research Institute How Do High-Deductible Health Plans Affect Use of Health 18 –64 With Depression or Anxiety and Individuals Ages 5 –24 With ADHD Ages 18 –64 W® ith Depression ® or A De n pres F xi igure et sion y an 17 A d nx Iin etd yividuals Ages 5 –24 With ADHD Among Individuals De Ages pressi18 on –64 AnxW ieti yth D ADHD epression or Anxiety only m c a inc a nd or nc re e spe d e t a hose a rspe si .y ng c s, e A W ifi c A d ifi c ood a nxi w ult a m d cit t op e ion s in t out e h me rJohns tg tyion o s e ca nc om inc he nd ny ton a l e f HD ude Unit I d l he sndivi e Founda su pd a e H ac rd lt c P h d t ont h d m s a ua St ae s m a ltr s w nt is s A ion tols for e or s, 200 v eg e , d is de ll a e 2 ic is 5 tr01 s, p a s s t a ta 2 ion a 8 – gs a ) 24 – he e r. im 20 , ht g W tp a d 18 re tr e rhe p it nd e y nd : :// h - r c De e Ra e e A a w xis rtnc pres r, D ow p w erH tid e e w ing p s a D g for ihy .r on r I ion, d nc ................................ w c si highe ond jindiv rc f.or e A ia ra e nx n o ses it la g iidua et ion p r /c t yions ff d ont A ie c m ls d e A rhip e uc ong ior DHD w v nt is t it ible /da t t it h ch o o Y s, spe oung t p m s m he olic r/f onic ................................ a swit or c y rA ia hold m e d c lis /r c g ult oh e te n e n s p offic d fr r,e or , it " om rions c a Jour tom s/ lly e P rv , e P or is n (it p O Fr a ib or tis l o s, m idit ons tt o s/ cf P rH ie 20 it te in, Se ........................ D s, a ic sy n H 12 a tca P l for hia nd /r l he , p w w tulv y re je a f4 ic e m lt fin a e 02 Re h o rd p-d a loy fix 4 sea , ff 05 this e a ic ed nd .r r e 10 s c h (EBRI). M. Christopher Roebuck is President and CEO of RxEconomics, LLC. This Issue Brief was written with Thi The The s st d fin eudy ma c dlin ings in es in use ke this s us p of ea p of he erta he e ltx h ca a Im BM ine re se d Ma trhe v ric k e e im ts sc p aa a ss n coc t of C iaotm m ed ov m w e ing r itch th ia frl C om ela H im aD P H s a P PO nd trta o E ns a nc n HD laount ted H int eP rs D . o W re ae d ta uc a bls a tions o se e( xa C in he C m Aine E)a. d ltThe h ca the C im re C p A spe a Ec tnding of For each of the three mental health disorders examined Figure 4 Care Services and Spending Among Enrollees With Mental Depression Anxiety ADHD Impact of HDHP on Health Care Spending, by Type of Health Care Service (per 100 Plan Members) Roe v na to isrit unde rb s, a ow ef uc fe k nd r d cst 13 e t 20 s. fin ap 0: 441 nd 13 rit e In th ion o sc )t, he rg ipti – e e e 4 f th ne e ff on 46 ns e ric e c , d uin t 20 s t " r dsa ug rg 2 ug he fe 0 rf. e ill se us ha su s. e p rlt b la s, w For (o Fr n d r"onst a e m ell of si p ag in a rye ns sent ha the nd m vse, e a Roe ty n he De e ha w g pres b e D a v uc iD te aiv s n ik on on e a e l20 ly st y m ze im a 14 Aff ed nx a m e ) t, itc et or b he the yed es a r a s w e lt out xt A s m h ca DHD e itnt c h me a or rm tg eo in ep w nt s d a ric l e hic aue e l he ff h p sh e tc a o op ta lt s c fe th condit ip ew ing a nt lc es us rula (offic Btre ions e ot d de - a G av . ny told is tThu it he of b s a e m s, r trnd he ge e a se elow tns sea aser l. of er r 20 cv h a ic ll 15 on es, ), d ( m Fi aov tgaur ing a bss a ese is 16 frt om c a)ont nc . Tot e aa fr ins PP a om l O m a tnnu t e o he m ab a n Ie l ns H rsp R teie nr t Aut nding . ollm eI’ts r ce a e p nt n b sea e rinf m e rc or e a h a m rm gb ue a nd e tion, rd e fe td ha ll itaor b ts w yp ia a $ l st e t1,13 ll a iea nt ff s a s m 7 s. for dA jov udic ny p ing avta ie ie tfr e w nt d om s e s w inp xpr Pit a Ph M tOs t e iess nt D e o D d aH nd ; in t RA $9 out his s m 84p for ra a ey tp ie or r pnt e atspond t ie a m rnt e e d s w tic hose daiff it l h a a end rof enxi nt the ly e t tyha ; n in this paper, weS p ed ra etisent ve/Ben m zoe dia az nep s ifor ne F ia lls set of member Hypnotic/Anxiolytic Fills Antidepressant Fills Demographics, by Mental Health Disorder Figure 8, Annual Spending on Health Care Services Among Individuals Ages 18–64 With Depression or Anxiety and Among Individuals Ages 18 –64 With Depression or Anxiety and Individuals Ages Sim B He unt ila in, M a rlylt , e som h lissa Di e B va es c ec uw ina okr te ion r d s, C ehe ar tes rs y? l D als a o mdbeecrlin g, eA dm for elia those Havila wnd, ho m Ka ov nik ed a int Kao pur an , N Hic Dole HP, Linc urie luding , Rola flu nd sh Mc ot Ds for evitt , pa and tient M. s with 0% Depression Anxiety $3,500 p q how as w ua resc lit re p e ry ll a ipti la g 9% (rn d Fr e s t on ss onst e he or si dr s. gsp iug n a n ch e St fil nding nd aa ls nda nge .Roe ron s a db e uc rtre his rk oim r 20 s i cp a14 a n a re c)t . , ing ll m u Use se p od a of of teie ls spe tnt he w s w e c se r ia eit lt se h me c ylus rm vic tee e n d rs is tic ead a l he t b ions fa yir a p lt ly e h d ( rFr high son. ons isor for tdin a e rp s c e nd op an he Roe le wb lp ituc h me ek m 20 pnt loy 18 ae l he )r , s m aa nd ltah d k de e is la ta or yrs in b g de etresd rw ehe astn author and should not be ascribed to the officers, trustees, or other sponsors of EBRI, Employee Benefit Research t p ahose nd har$8 mm a 68 cov y for icng la im pfr as om t. ie nt I P t s w P inc Os t it lude h ADHD o H s D dH atP . as on Iw n itp ov h er e H crS e 20 A nt sa . m g D e ill iff t ion p eerrm es, n eto spe p re le sponse nding with e s m fe mll b pa loy yy e m 9 p xis ent e t rfor -cbeant sed the for follow a nd botot h M ing heD rr D he e aasons a nd lth p a:nxi lan cov ety and erag be y 15 in characteristics, including age; gender; relationship t5o –24 With ADHD Individuals Ages 5–24 With ADHD .......................................................................................................... 10 80 March 10, 2022 • No. 555 H MD avD ila and, ndSus a Anxi m ae n M e lia ty a , M., rtq euis tN anus e , e"rC a v onsu j aSood ccina me , trio Rola -D ns ir e nd for cte Mc p da D H tie e ent v ait lts w th C , ait nd ah a reM. : nxi Ea Sus e rly ty a , En M a vnd ide anc H rq Puis eV va A,b "out c H cow ina E ff tD io eo c ns tC s onsu for On w C m om os erte - D A n w nd ireit cQua th M ed lit D Hy D e,a " ( ltFi H h P e ga ur llt ae h ns 76 b c the a enc ne y e $3 fit w r,50 e sc d r0 e e r c ee is ennin ions rolle g. d ( W Iin t ha ca arn a a PPO mls e . o t inf A ab l. or out 20 m19 one p) olic . - tE y hir ff m ed a c k tof s of ers those ’ e dff ed or uc w ts it tible h M by us D leD ing v eor l, soun p ar nxi esenc d e tinf y e ha or of d m a a an HSA t n e ion. m e ror ge he ncy a lt dh r epe aim rtm be ur nt seme visitnt ; c lose Breast Cancer Screening Cervical Cancer Screening Colorectal Cancer Screening Prostate-Specific AD nti ege prn essLi iopi nd TesA t (M nxal ieets y, Age>=3 ADHD 5 p any erc g I ens nt ive t it for n ye ut e p- a aE rt.d ie uc nt Da a s w ttion a a it frh AD om nd 2016 H Re Dsea . thr rcoug h Fund h 2019 (EB RI we-rEeRF) us, ed or for th etir his st st aff udy. s. N eW ithe e lim r EiB te RI d no our r E aB na RI ly-si Es t RF lob o indiv bieidua s or ls ta w ke ho s By Paul Fronstin, Ph.D., Employee Benefit Research Institute, and M. Christopher Roebuck, policyholder (self, spouse, child/dependent); geographic 8% Depression Anxiety ADHD (Females, Age>=50) (Females, Age>=21) (Age>=50) Test (Males, Age>=55) or Females, Age>=45) I 15) n 2019 . LTher im 8% , itA A a tehe ff ff t ions w a e ir ca Its RS s no Vu w – 51 g lne Aa im v 6 kre – e p a y w b a H c 53 le is D tsu H P 0 on P , o es t p 20 p in o ula ne he 06 tum ur ions .a b st onia ili? udy is t t" y For tv o aa c um d cha ina op for tt t b ions a eH c m e a or a us for lte e h Ec flexib p indiv atonomi ient idua les w de cls s a si it sele h gn o nd cong cff P te olic ee d ring st tyhe iv 14 e m ir he or ow (2) e a r n he : p t 1 rfa ot –12 ilu a elt cr , th p e ion for 2 ( 01 C laH 1 n, .F) unm c , ecrhr ta eonic in m asu re ed dic al Figur •e 9, HA RA nnu s of al Sp ten ha ending ve a o high n Var dious edu cTy tible pe, s of butH som ealte h C tim ae rs t e Se hery v ic de o s A not m . ong HDH IP ndivi s wit dh HS uals A As m ges 18 ust –m 64 e eW t a it h D stae tut pror ess y ion or Gender t ao rr9 in ange10 m evnt is it( $2,908 e Hd RA a )p , ra im nd arty y p ce a rof e p H hy DH siP cia ha n; a vend alsne o b ae rly en ae ll xplor had e ad t le (H aa st v ila one nd perte a sc l. ripti 2011 on ). d rA ug com fill ( m Fi on gur fin e d 6) ing . Only in this positions on specific policy proposals. EBRI invites comment on this research. $3,136 were $3 cont ,000inuously enrolled in their health plan within each calendar year. Those enrolled in capitated health plans Ph.D., 1 RxEconomics, LLC region; and Charlson Comorbidity Index and its chronic 70 -2% Figure 2 ob serst vic r ruc eeas t tsons ivhr e oug p for ulm h a tona ha VB t rc yI hoic D d is pe e la a n tse ha st( tr C uc aOP re tur D ae ) ls, o vor ia c or d Nira ot ebla ic ete te e d 20 s. w 19 ith he -45. a Thi lth se s w rv aic s a es us n ext e rw em ould ely lik pop ely ula lera d g uid to a bnc iase e — d etst hrim eea-tqeua s of rters Primary Care Physician Male Mental Heal28% th Office 32% 65% Annual em m Aini p nxi loy m eu e ty m r spe a d nd ending dIuc ndivi tible p de .ua r m ls e Am ge bs 5 er d –e 24 clin W eit dh bA yD aH n e D ................................ ven greater amount: $1 ................................ ,463 for patients wit........................ h MDD, $1,305 for 11 lit beetrw ae tur en e 39 is tp ha ertc H ent DH aPnd s le 56 ad p te o rc re ent duc ha tions d a in meus nte a l he of he altah lth p 0.2 rser ovide vicre s. vis it. w ere excluded from the analysis. $3,000 condition indicators; as well as flags for anxiety, $2,632 63 7% Inpatient Hospital Emergency Department Office Visits Specialist Office Visits Visits Prescription Drug Fills L Cu, haC rls hr on, istine M. Y E., ., P A. lis Paom P Ber . pB e ce uc i, K ntage sh . , L. Fa Ang leof s, a Zh P nd aopulation ng, C. Je R. 62 anne Mac K M. Under enzi Ma ed , d " A A en, ge Ne M w a 65 tMe the Dia tw hod X. gnosed of CaC lla laha ssify n, Wit ing Rob h Pe V rrog ar t F. Le nost ious ic Ca C te om s, Ja orb m idit ie y in Female 72% 68% 35% of em the ploy im ep ra s csa t of y tswit hat c the hin yg h fr aom ve e a nha PPnc O e to d a cn HD overaH gP e (aes a .g., re sel sueltc t(ion d Fronst ue in a to nd antFe icipa ndr tion o ick 20 f i 21 nc )r . e a Ased s allow or d ee dc unde reased r t he 0 patients with anxiety, and $1,149 for patients with ADHD. These represent 13 percent, 14 percent, and 25 percent Ho$1 w 00 Do High-Deductible Health Plans Affect Use of Health Figure 12 Suggested Citation: Fronstin, Paul, and M. Christopher Roebuck, “How Do High-Deductible Health Plans Affect Fi de gp ur re $2 ess ,50 10 ion, 0, Im scphiz acop t of hrH eD nia H/bip P on P olarrob dis ab or ilidte yr of , Receiving Various Types of Health Care Services Among Individuals Ages 60L W ong allaitcudina e, et l Stu al., "d Im ies: D pacte of velop High m-eD nt Mental e d auc nd tible Valida Health Hetaion lth P ," Disorders, la Jour ns n oa n M l of C edic hr 2019 aonic tion U Dis se ea Ase m ong 40 ( I 5) ndivi : 373 dua –83 ls , W 19 ith B 87.ipo lar $2,341 • HRAs can prov -0.2 ide pre-deductible coverage for any service the plan sponsors choose, including inpatient care Notic dee, m ea m nd ploy for e rh s a ead ltd h ca ed p re re ser -devd ic uc es) tible . coverage for services related to diabetes, hypertension, and heart disease. Generosity of Employment-Based Health Insurance, 2013 – 6% 2019 $2,286 reductions, respective Imp ly. act As a o f re Hsu DH lt P of otn he U highe se of rV d ari edo uc uts ible Ty p in t es he of H H D ealth HP, b C ut are despit Serv e itces he d A emo crea nsed g use of health care Introduction $2,478 Figure 6 -0.5 Use 6% of Health Care Services and Spending Among Enrollees With Mental Health Disorders?,” EBRI Issue Brief, no. hy Ca perr te -1 e ns Service ion, and dyslipids e ma ian . Me d Sp an value en s for ding the Amo Age, Yn earg s Enrollees 3W 9.9ith M 39.e 6 ntal 13.3 -4% 18–64 With Depression or Anxiety and Individuals Ages 5–24 With ADHD ................................................. 12 $2,500 $14 6% $2,381 Disorder." Psychiatric Services 72 (8): 926–934, 2021. doi:10.1176/appi.ps.202000362. and primary care. HDHPs with HSAs may only provide pre-deductible coverage for services specified by law. H owever, medicatA ions nxiet yt D o In isp d orde rie vv ird s ent u 51 al ts he A e gxa es ce 18 rb – a64 tion Wof ithm D eep nta rl he ession alth o dr is A orn dxi eret s y w e an red no Intd inc ivilude duald s in the Not8.1% ice. services, annual employee spending Prob ab (i.e il., ity out of -of Rece -poc iv ke in t g c o Vst ari s)o in uc s re Ta ysed pes a obf out Health $300 C p are er person, depending on the Methods -1.1* Deduc 55t5 $0 ible (Es ha mploy vee b ee B com eneefit a Re prom seaine rch I ntns pa tit rtut of e, tMa he rhe ch 10, alth b20 ene 22) fit <1.s la 8 ndscape. Perhaps one of the strongest trends 76% in Chernew, Michael E., Allison B. Rosen, and A. Mark Fendrick, "Value-Based Insurance Design," Health Affairs Web demographic variables are shown in Figure 4, and 5% 50 -1.3* 5% Health Disorders? $2,000 Ages 5 –24 With ADHD (per 100 Plan Members) Despite increases in deductibles, trends in actuarial value (AV) — or the relative generosity of health plans — in Services Among Individuals Ages 18 –64 With Depression or Figure 11, Impact of HDHP on Probability of Receiving Various Types of Prescription Drugs Among Individuals Ages -2 mental he 5% alth disorder cohort, representing 17 per 45 cent to 23 percent increases in out-of-pocket spending when patients 18–24 16% 14% 24% Major Depressive Disorder -$16 5% employment-based health benefits has been their increased adoption, particularly 5.3% in the context of high-deductible Quan, H., Exc P. lu H si av lfon Sunda e: w195–w ra 20 ra3 ja , n, 20A 07 . .Fong, B. Burnand, J. C. Luthi, L. D. Saunders, C. A. Beck, T. E. Feasby, and W. A. Figure 5 contains information on comorbidities. We -1.9 • HRAs are funded only with employer mone Depres y. B siot onh eA m nxp iet loy y erA s a DHD nd employees can contribute to HSA$1,708 s. According to the guidance, the list of Anp xi re et ve ynt an ivd e ser Indv iv icie ds t ual ha s tA cg aes n b5 e – -c $36** 24 ove W *re ith d p A re D-H de Dductible will be reviewed on a the employment-based health coverage ma -$4 rk 2* e*t* have not fallen comm-e $3 ns 9*ur ** ately. Both mean and median Impac $2,000t of HDHP on Use of Health Care Services Prevalen 18 ce – 64 of M Wie th D nta el pr H ee ss aion o lth C r o And nxiit etiyon ................................ s -$50***-$49***25–34................................................................ 21% 23% .......... 13 moved from a PPO to 1 an HDHP. -$54*** Copyright Information: This report is copyrighted by the Employee Benefit Re -$58* sea * rch Institute (EBRI). You may By Paul Fronstin, Ph.D., Employee Benefit Research Institute, and M. Christopher Roebuck, health p -6%la Ans tte n( tiH on De DHP fics) it Hy . pe Irn 20 activity 2 -$6 0, 6*89 ** .2 percent of worke -6%** rs Prw ima *ith ry Ca hera elth insurance had a deductible, up from 70.7 percent Ghali, "Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data." Medical found -$1 4 t0 ha 0 0 t individuals with mental health disorders are -3 -2.7** periodic basis. In fa-c 6%** t, t*he guidance specifically states tha2.8% t the periodic review is expected to occur approximately ac4% tuarial value — the percenta -$8 ge 6* *of * covered De he pres alth ca sion reA e nx 35–44 xpe iety nse As t DHD hat is paid by the 23% plan — wer26% e about 83 Moving $1,50 fr0om Inpa a ti Di en P so t Hos P rd O er ( t pi o A tal Da HD) n HD InpatiH en Pt Hos had pi tal a subst Eme ant rge ia nc l im y pacP t hy on sicius an O e ffi of ceheaSlt pec h ca ialisrte O se fficr evic Me es nta . l He Firal st th , O the ffice pP la re n sw scriptiit oc n Drug h reduced 7 copy 2, print, or download this report solely for personal and noncommercial use, provided that all ha 35 rd copies retain Ph.D., Copeland, R x CEco raig,nomics "2021 EB , RI LLC Fina ncial Wellbeing Employer Survey: Focus on COVID-19 and -$1D 04iv **ersity Goa -3.0 ls*,*" * EBRI 34 in 200 We sea • 8r .cE he m Bd y p loy 20 me e 20 d rs c ic , a ne a l c n p ala rly im ut 60 s for re st pe r r tic he cteion nt p rs on of esenc hehow ae lt h p of HIla RA Cn Ds a e9/ nr r10 e oll us e d e eia s w dg . nosi e Fo rer s c in p ins od tla ae nc ns s for e w , tit he e h w ay c h o ha matf 13 me y w ra es c quir onsi nt e a tha d l he e -t r 3.2 et ad he *lt * *h condit a H high RA be dion e d spe uc s, tnt ible Fi gure 12 C, aIrm e p 42 ac t ( 11 of )H : D 11 H30 P – o9 n U , 20 se 05 of . Various Types of Health Care Services Among Individuals Ages 18–64 With often diagnosed with more than one condition. For Admissions Days Department Visits Visits Visits Visits Fills every fiv Oe the to r B10 eha vyio era alr /M s. en tal For Hea pla thtients and employers alike, 10 y45–54 ears may be a long time to w 22% ait for such cov 22% erage percent in each year from 2013 to 2019. Despite rigorous requirements around what can be excluded from Because there was an across-the-board decrease in use of health care as a result of the HDHP, reductions in spending 30 the e10 xt0% ent of health care use. Specifically, the proba2.2% bility of having a mental health office visit fell 14 percent for those $1,50 -4 0 96% any 20 a Iss nd ue a ll c Brie op f, yr no. ight 54 a4 (E nd ot mhe ploy r ae pe p lic Beane ble fit not Reic sea es c rch I ontns ain tit eut d e the , 20 re21 in, ).a nd you may cite or quote sma 96% ll portions of includingon anxi heeatlt y Dih ca , so m rda ers rje or se dr evpic re ess s b ive efor dis e or wd or ek r e(rMD s re Da)c , h th attee nt irion d deduc efic tib itle h.y p Mor erae cov tive itry, d the isoyr d m eight rs (A not DH D a) llow , bipo un la us r e ad nd fun md as t nic o (Fronstin and Fendric -k $160 2021). Further, deductible levels have be-e 7%** n inc * reasing more generally regardless of being 30 Depression or Anxiety and Individuals Ages 5–24 With ADHD (per 100 Plan Members) ............................... -7%*** 14 examp3% le, 43 percent of patients with MDD also had 55–64 17% 15% 3% decis-ions $200 to be made given the pace of research on plan design and medical innovation. There is alread -$1 y 72 an 93% ** appetite 3% wered e re dflect uctible ed s, a in ne vearraly g ea ll of AV atls he o inc hea re lta h se sedr v in HSA ices e- xa eligib mine led h(e Fi ag ltur h p e la 17 ns ).. IThe npatm ieov nt ehospit ment a of liza wtor ion ke r r-e s fr $1 p81 re om *sent ** PP eOs t d the o -7%*** with anxiety, 15 percent for those with ADHD, and 18 percent for those with MDD (Figure 10). The probability of the -8% report provided that you do so verbatim and with proper citation. Any use beyond the scope of the foregoing Vahra $1 tia ,00n, 0 Anjel , Stephen J. Blum3.9 berg, Emily P. Terlizzi, a 89% nd Jeannine S. Schiller, "Symptoms of Anxiety or Depressive 2% c dla isss orifi de erd sr , oll a Bp s ipo ost ov “lhigh ar- e t an rr a d o ” Ma um r in a bna a icla tn Di icnc H sst oD e rde rs t H erss P so . d g The is row ord a a e vb re,ov rp ae hob g e a d c ia e es, d rtuc a ain ut tible is lem v e inc , l.ob rN esess a o sed su iv c e fr h lim -om com i$4 tp s c uls 46 ain b v to e e d $1 is p,9 or la45 c de e d rfr , on om eatH ing 20 SA 02 s. dis tor o d 20 er20 s, among 0.7% anxiety, and 29 percent of patients with anxiety also 90% -5 for adding more servic $772 es, as evidenced by the Chronic Disease Management Act, which was reintroduced in the U.S. HRAs and HSA-eligible health plans would be exp 85% ected to lower overall mean and median AV. Yet, despite Fi lag rg ur est e 13 de, cI lin me p a in spe ct of nding HDHP a om n U ong se indiv of Va idua rious ls w Ty it 85% p h M es of DDP ( re $5 sc 13 ripti pe on r m De rugs mbeA r)m aong nd aInxi ndivi ety d ua ($3 ls53 Ag p ee s 18 r m– e64 mb e W ri)t. h Among Deyo, R. A., D. C. Cherkin, and M. A. Ciol, "Adapting a Clinical Comorbidity Index for $723 Use w$730 ith ICD-9-CM Administrative having 0a primary care physician office visit fell b A yT 5– 6 p Aer ceG nt, L and Ait N fell 3 C –E 7 p ercent for specialist office visits. requir Dis es E ordBeRI r a ’s p nd rUse ior e of xpress Ment p ae l rH m eis alt sih C on. aFor re A p m eong rmis si Aons, dult Rels D a p tile our na sse ing hip c tt o ont he Po aC lic c OVI ty h Eo BD ld RI - e19 r at Ppaend rm eis msi ic ons@e — Unit brei.or d St g.a tes, $1,000 schizop 2% hrenia, dissociative disorders, delusional disorders, and other behavioral/mental health disorders. Individuals those w 20ith employee-only coverage$622 and from $958 to $3,722 among those with family coverage (Figure 1). -0.2 had MDD. In addition, adults w$597 ith mental health Congress Pos at-s r Trae um -c 1.4 e ant ti *c ** S lytr a ess Ma s Disorde y 20 r 21. This bipartisan, bicameral legis 80% lation would provide additional flexibility to allow 0.5% 16 ADHhighe D pat-r ie 2.1* d nt e *s *d uc for tible whom s, e m inpa ploy tie ent rs m spe aynding have w ea nha s not -3.2 nce si dg he nifa ic ltahnt bly e ne imfit pa s j ctus edt, epnoug rescrh to se iption d e rA ug V inc spernding ease if t exhi heb yit w ed e rtehe -$300 Depression or Anxiety and Individuals Ages 5–24 With AD $514 HD 78% (per 100 Plan Members) ............................... 14 Databases," Journal of Clinical Epidemiology 45 (6): 613–9, 1992. Some or all of these differences between HRAs and HSA-eligib Sle el fHDHPs may affect the wa 60% y patients w 62% ith mental h 1% ealth 80% 14 -5.9*** $457 -6 August 2020–February 2021," MMWR and Morbidity and Mortality Weekly Repor $443 t 70: 490–494, 2021. -8.1*** -6.4***$417 were classified as having a -8.8 m **ental health condition if $402 the$398 y had claims for at least one inpatient admission or two In this$5 p00 aper, we explore the impact of moving from a preferred provider organization (PPO) to a high-deductible disorders have often also been diagnosed with The enhaod ncd es of d cov ha erva ing ge a for n ese mre vric ge es t ncy ha dte p prae $325 rv tm ent ent the vis eitxa w ce erre b a low tion o er b f c y hr 6 onic perc e c$326 nt ond am itions ong. those with ADHD, 8 percent among doing so to be competitive in a tight labor market. More information can be found in Fronstin et al. (2021). greaR t-e 10 -ep 20 st % o dr ro t Availabili p in spe 10 nding ty: (Thi $325 s r e pp eor r m t is em ab va eila r).b le on the interS ne potu a se t www.ebri.org 25% 25% 0% $511 $504 -$325*** 1% disorders respond Phobic A w nxhe ietyn th Disorder ey m s ove fr 0.4% om a PPO to a new Figure high-d1 eductible plan. However, estimated effects from the doi:10.15585/mmwr.mm7013e2. $443 Fi out gur paet 1 ie 14 0nt , I vm isit ps acon t of diff HD erH eP nt o d n U ate se s w of ithin Var ious a giv Ty en ye pes of ar w Sc itr h a ee nin candid gs Aa m te ong dia Ig ndivi nosid s c ua od ls eA in t ges 18 he p –r64 im a W ry it h D or seconda epression o ry r $408 $418 health p $500 lan (HDHP) among people with mental health disorders. We focus on individuals with major depressive hypertension and dyslipidemia. -$353*** -21.7*** Fronst 70 in, P -7% aul, and A. Mark Fendrick, "Employer Uptake of PreC -h D ile d/d Duc epe tible nden tCoverage for Pr14% event 6 ive Se13% rvices in HSA 99% - those with MDD, and 9 percent among those with anxiety. Similar to the decline in mental health office visits, the -10%*** 5 4 4 -27.0*** HRA -a $4 na 00lysis were very sim Av ila erage r to t $259 hose An nu of tal he Emp HDHP lo in t yee er- m Onl s of yd an irecd tion Fa , m m4 a ily gnit Ded udeuc , and tibst 4 le, atistical significance. 3 3 3 -29.3*** 63% 3 position. A Finxi nae lly ty , w (pee rc a 10 lcula 0 Ptla ed n M pre em va ble er nc s)e ................................ rates for each of the ................................ 13 mental health cate................................ gories as the number ............. of memb e15 rs Sm disor ard te err d (MD educ D) t, ible anxi A s a utie sc t tiy c com , Dia snd ord mod er ata te ting ntion d 0.3% serveic fic es p it hryep ve ernt acing tivitty he d is exa ord ce er rb (a AtD ion o HD)f c sihr nconic e the cse ond re itp ions resent migh the t bm ea ajor na ittyur of al Although spending on primary care physician and specialist office visits fell between $36 and $54 per member due to In this st $0 E udy, ligibw lee H inv ealt eh P stiga latns e ,t" he E B im RIp a Iss ctue of B HrD ieH f, Pno. s on 54 us 2 (E e of mhe ploy alte h ec B ae re ne se fitr vRe ice se s a arnd ch I spe nstnding itute, a 20 m21 ong ). individuals with likelih-ood 40 0% of being admitted to a hospital declined 16 percent a-nd 36.6* 19 ** percent among those -34w .1* it*h * anxiety and MDD, -36.1*** Wharam, J. Frank, Alisa B. Bus Ac mo h, Je ng anne Workers Madden, in FaPr ng iv Zh ate ang, -Sector Matthew Establishm Callahan, Rob ent ers t F. LeCates, Phyllis Foxworth, Table of Contents Thus, these results are not presented. 60 -8% Inpatient Hospital Emergency Primary Care Phy-s 37 ic.0* ian**Specialist Office Visits Mental Health Office Prescription Drug Fills identifi - ed with e Pria ma ch condit ry Care Phy io sin d cianiv Oide ffice d V ib siy ts the total numS b pe er c iaof list O eligib ffice V le is iliv ts es in each year. Menta 56% l Health Office Visits Note: ADHD = attention deficit hyperactive disorder. m eve olut ntaion o l heaf he lth p aa ltth p ient la sns . Fu . Va rthe lue rm -bor ased e, t he re im prb eur vaseme lencent of ptrhe om se otc eond s the itions deliv in t ery he of pe op vide -ula 7.8nc *t*ion w *e-bait sed h e , m high ploy -q m ue ant lity -b c aa sed re that the H -12 DH %P, the effects on mental health office visits’ spending were mixed: The measure decreased by $104 per mental he $0 alth disorders — a topic that has received little attention. Two recent papers studied the research question in Fi reg spe ure c t15 O ivbs e, ly es Im . s ivp e-C ac om t of pulsH ivD e Di HP sorder on Use of Various Vaccines Among Individuals Ages 18–64 With Depression or Anxiety and Depa 0.2% rtment Office Visits Visits SteIn ppa he tien n So t Hos um pital eraIn i, pa Dtie en nni t Hos s Ross pital -D W e Eg me ith na rge n, a Dedu nca ynd Chr ctib Pis rima tine rle, y Ca Y2002 r. eLu, " S – E pe 2020 ffce iac lit s tof OffiH ceigh Me -D nta ed l He uc altth ible Offi c Ie nsP ur rea scnc ripe tio n Drug on Health Introduction .......................................................................................................................................................... 4 -48.1*** Source: Employee Benefit Research Institute estimates based on administrative -$500 Primary Care Physician Office Visits Specialist Office Visits Mental Health Office Visits Fr e he nc onst aour lth insu in, P ages us a ra ul, a nc ee nd of is — e M. spe ra Cthr che ia is lly rt op t ha high he n cr r rRoe e ela attb ing iv uc e k b t,o a " rot B rie rhe a rnd s t r m - o Ne — ant m hi ael he g a hnd -a va ltG lue h ec ne ond ser ricv it P ic ions re es. sc , ra Iipti nt nd e on ritv e ha D nt rs b ug ions e Use e tn inc ha Atft im re era p A si rov d ng op e. tp ion o Iant ie our nt f a - Full- ***p< A0 dm .01 is , s **ion p< s 0 ( .0 pe 5, r*p<0.1Da 0. ys (per 100 Department Visits Physician Office Visits 48% Visits Fills memb -60 er among those with ADHD and $58 per member among those with MDD. In contrast, the decline in spending on a sample of HDHP enrollees with bipolar disorder (Wharam et al. 2020 and Lu et al. 2021) and found that the higher Note: ADHD = attention deficit hyperactive disorder. We als 50 -9 o% inv Indivi No est teiga : d AD ua HD tels d = a A w ttg ehe e ntis 5 otnhe d– er fi24 ct it he h W ypree it ra h cw tiA ve eD r de iH s od D rdiff e( r.e preernc 10 e0 P s in latn M he eim e nro m p llb m ae c er t n s) tof a n................................ dH clD aH imP s db ay ta .g ender among ................................ people with MDD a ...... nd 15 Care Use in Bipolar Disorder," American Journal of Managed Care 26 (6): 248–255, 2020. Note-: $5 HD 13 HP = ** high-deductible health plan. Employee-Only Coverage Family Coverage P revalence ra 10 te 0 E s for nrol Elati e t es ng he ) D 13 isorde m E re s nrnt olla ee l he s) alth (ped r 1 is0or 0 E dne ro rls le e asr)e show Vn in isits Figure 2. For the remainder of the paper, we focus 0.2% Source: Employee Benefit Research Institute estimates based on administrative enrollment and claims data. Data ................................ Source: Employee Bene................................ fit Research Institute estimates ba................................ sed on administrative enrollment ................................ and claims data. ..................................... 7 Replacement, Consumer-Directed Health Plan With a Health Savings Account." EBRI Notes (Employee Benefit T a ce na he ntly e dsi rees, w cd lin out ee in t c ob om ser he e s w v pe r: ob hil aeb ili mta yint of afil inilin ng g a aff p or re dsc ab rili ipti tyon ma w ya b s esm found all (1 in t pehe rce for nt m in MD of aD c a lin nd ica ally nxi nua ety, nc 4 p ede he rceant lth in A plaD n th HDa )t; Source: Employee Benefit Research Institute estimates based on administrative enrollment and claims data. m ental health office visits among anxiety patients was not statistically significant, despite the decline in use. cost $4 sha ,00r 0ing appears to have caused patients to prioritize some services over others — -66 p .4* sy **chiatrist visits were anxiety. It w Not ould e: ADH not D = b atte e na tio s n ur defip cr it is hye pe if m ractivee dn isoa rdnd er. women reacted differently to the financial inc-e 68 nt .4* iv **es associated with the ***p<0.01, **p< 0.05, *p<0.10. -68.9*** doi:10.37765/ajmc.2020.43487. on individuaNot ls e w : ADH ith M D = D att D e, ntia onnx deie fict it y h, ypa era nd ctivA e D disH ord D e. r. 39% $3,722 Figure 16, Impact of HDHP on Health Care Spending Among Individuals Ages 18–64 With Depression or Anxiety and 40% -$6 -80 00 Source: Employee Benefit Research Institute estimates based on administrative enrollment and claims data. $3,655 better me Re etsea s w Notr eor c : h I HD ke HP = ns rs’t it hcig lin ut h-e dic e, d a u 2014 lc tia bnd le h) e fin a35 ltha p ( nc la3) n.ia . l needs. however, therS ec h w iza op s a reni c 6 Di –1 sorder 0 pe srcent 0.1% decrease in MDD and anxiety-related medications (Figure 11). Source: Employee Benefit Research Institute estimates based on administrative enrollment and claims data. Me una thod ffectse ................................ d, while other mental he ................................ alth care visits fell. ................................ Stated differently, the................................ authors concluded tha ............................... t the HDHP did not 7 move from a PPO to an HDHP. 33% Rese 34% arch suggests that men are less likely than women to use health care, and women I W n a e looked ddition to im at usep a ac nd ting spe the ndi e ng xteon nt spe of he cifi ac lt h ca presc rer ipti use on , the drug HD cH laP ss w es c as om nea m rly only alw uased ys a in t ssoc he ia tte re da w tm ite h r nte of duc the tions me in nta the l Source: Employee Benefit Research Institute estimates based on aF dm ig inu isr tra e ti v5 e enrollment and claims data. • Moving from a PPO to an HDHP reduced the probability of using health care services. Individuals Ages 5–24 With ADHD .......................................................................................................... 16 $3,396 $3,392 ***p<0.01, **p< 0.05, *p<0.10. 10 $3,500 W haha ver a the m, “J. blun Fratnk ins , tFa rum ng eZh nt”a ng, effe c Ja t m de ie te W cta ella d in ce, mCahr ny is of tine th L eu, ge Cne rar ig al st Eaudie rle, St s p ep re he vious n B. lyS d oum iscus ersed ai, . La A ris sisa de N freom khly this udo w v,or aknd v a he olum re a lt m h condit 30 or e % e of lik he eions a lylt Di t h se sha s oc unde n me ir ati vvic er Di e n st s c stud orde o ons e yr,xpe s um incr 0.04% e luding ie dnc . e I n a bsed ar ll t rie ahr triv s t eees a o che ond nd alt it ot h ca ions herr , e bt e he dnzod u e num to iaze b cost ep rs of ine . s, p Our a rim nxi a fin oly ryd tin cic ag s a rs w e offic nd erhy ee m p vnot ixe isits, spe d ic s, whec n it ialis cta v m isi et s, to Ana Prlyt evaica lenc l e Da of ta Ms ee nt ta C l He on Pra se lt ts ruct h eCn ond c io en it o ion 26% f H s e ................................ alth Conditions, b ................................ y Mental Health D................................ isorder ............... 7 Note: HDHP = high-deductible health plan, ADHD = attention deficit hyperactive disorder. Fronstin, Paul, and M. Christopher Roebuck, "Health Care Spending after Adopting a Full-Replacement, High-Deductible -100 Figure 16 Figure 19 -96.7*** Source: Employee Benefit Research Institute estimates based on administrative enrollment and claims data. $3,069 Fi o n b guripo e 17 laD r, ed Inni m isor ps Ross ad ce t rof s, -H tD he D eH g e na Pff o e nn He c, t" of Vua H lne ltD h C H raP b asle r e a A m Sp nd ong ending L etss ho Vu se , blne y Figure 21 w Ty itrh apb m ele e of nt WH a om l he eae ltn I a h C lth d n High are is or Se - drD e ve r ic s h deuc A atm s not ible ong H b e Ie ndivi ae ltn e h P dxla ua ans m lsine E Axpe gde.s 18 rienc –64 ed gender differences. Women decreased visits to primary care providers and specialists more than men, yet men cut a and ntide mp er nt ess al he anta s, a lth vi nd sist ts a imula ll dnt ecs. lineMor d unde eove r rt, he w eH a Dls Ho P inc (Filude gured 12 the ). e The xtent la tro gew st hic im hp (aacnd t on lev offic els e of vus isit es w and as spe fornding those ow n) it h • The amount of health services used was also lower among patients with mental health disorders who moved Impact of HDHP on Health Care Spen Figure 10 ding Among Individuals Ages 18 –64 With Health P Dela lus n Wit ional Di h a sorder A Hn e sa nlt uh Sa al Uv se ings A of Vcari coun outs : T Ay Fi p F F es v igure igure e-o Ye f a 15 Ou 8 r Stt-udy of-N ," etw EBRI or k Iss Hue ealth Brie Cfare , no. 388 (Employee Benefit C To o A enc na xaly m lus tine icaion l D the a t im aset C pactonst of m ruc ovting ion 0.02% ................................ from a PPO to an HDH ................................ P on use of health ca................................ re services and spending ........................... , we constructed 7 $2,871 20% $3,000 Impact of HDHP on Probability of Receiving Various Types of Health Care Depression Anxiety ADHD DW ela ith D yede B prre ea ss stion o Canc r e Arnxi Ca erte y, " and Hea Indivi lth Afdfa ua irls s 38 Ag e (3) s 5 : – 40 24 8 – W 415 ith ADHD , 2019. ........................................................... 17 b m pa re im cnt ka a on rl he y c the aarlt e irh p , v spe is rit ov c s ia ide to lis Imp rm ts, w , e a act n nd thi D a lc o ep m h w he f eH rnt a ession D e lt arH h p l he eP low ro ov a n lt o e ide h p rr Pro A brhy y n s m b xi 68 si ab et c or , ia iy 66 le in/ ty an t, ha o a o ff dnd f n ic IR n e w 34 ece d om viis v p iit e id e v s w n. u r in al 10 g e P s r0 p r V e e A ari sc g rle a es ro cn me ipti e uiv 5 s on e – T d 24 m y d p out bres W e ug r is in of th o fil f ls ne A H t D he a ealth tw H nd or A DD tkot H . CD a are Fi l he , na MD a lly lt D, h ca , parnd erveea nt sp nxi iv ee e nding t y ca re from a PPO to an HDHP. Office visits, prescription drug fills, inpatient days, and emergency department visits Annual Spen Ser d Iv mp in ices gact on A o H mo f ealth Hn D gH IP C n d are oin vi d U Serv u se als o ices f AV gari es Amo o18 us n –g V 64 acc In W diiin th ves id D uA ep als mo ression A n g g es 18 –64 With -120 ***p<0.01, **p< 0.05, *p<0.10. $2,640 Research Institute, 2013). analytical datasetsSer forv e ices ach o Ff th rom e Ou thre t-eof m -N ent etw al he ork a lt Pr h condit oviderions s Amo unde ngr I n inv die vst id iga ual tion: s Ag es anxi 18 et – y64 (am Wong ith those ages 18– I n the present 11% study, use of health care services among patients with MDD, anxiety, and ADHD declined when they We chose to focus on individuals with major depressive disorder (MDD), anxiety, and attention deficit hyperactivity 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% Study Variables ................................ Charlson Comorbidity In................................ dex ................................ 0.39 0.31 ................................ 0.10 ................. 8 Note: HDHP = Ser higv hi -d ces eductiA ble mo hean lthg p laIn n , d ADH ivi D d = u aal tten s tioA n g dees ficit h18 ype – ra64 ctiveW disioth rde r. Depression or Anxiety and $2,491 8% Dep In rd ess io vri d iA on u n al xi or s et A A yg nxie an es d 18 t y In – and d 64 W ivi d Ind u ith al iv s Dep iA du gal es ress s 5 A – iges on 24 or W 5 – i A th 24 nxie A W DiH tth yD and ADHD c a w ohor ls ao s a dtls e s, r c o lin inv ee spe d e st m ciga tor ive e t e ly fo d . . r N w W ot om ee in e tha n th cor t p taor he n for a ste ed r m e indic deuc n.t ions ator s for in use the of re ser cev ipt of ices rb erpera est sent , ce a rv ric ea dluc , atn ion o d col f le ore ss c ttaha l cn anc one er offic scree e nin visg its; 10%all declined. Depression Anxiety ADHD Figure 18, Probability of Using DepOut ression -of-N eo tw r A or n kxi H et ea ylt an h C da Irn ed P iv rov idide ualrs s, A Agm es ong 5 – I 24 ndW ividua ith A lsD A H gD es 18–64 With Depression Source: Employee Benefit Research Institute estimates based on administrative enrollment and claims data. 64), MDD (among those ages 18–64), and ADHD (among those ages 5–24). We chose PPO as the baseline plan type $2,500 m dis ov oredde rfr (om AD H a D P)P . O These to an indiv HDHidua P. Ov ls e arcacll sp oune t nding for the on mhe ajor altith ca y of re p aser $2,322 tiev nt ics w es a ith lso mfe ent ll. a l h Ofe caour lth d se, is or our de fin rs,d a ings a nd the re limited 4% Individuals Ages 5 –24 With ADHD Individuals Ages 5 –24 With ADHD (per 100 Plan Members) $2,220 p I pm r eost rp m aa ce t t e m of -b spe e HrD . cH ifi FP c or on a nt indivi Use igen (P d of uaH ls SA e w a ) lt it ah C h M nd a lipid D rD e Se a tnd erst v ic a s; nxi es a nd ................................ etyt, etthe anus num , pb ne erum s of onia inpa ................................ , inf tielue nt nza hospit , and al a hum dmis a ................................ si nons pap a ill nd omdaavyirsus , a s w (HPe V) ..... ll a s 12 Fronstin, Paul, and M. Christopher Roebuck, "Qua De De litpres pres y of s siion on Healt A Anx h C nxiiet eta y yre A A A DHD DHD fter Adopting a Full-Replacement, High-Deductible or So A urc nxi e: Em ety pl oa yend e BeInndivi efit Res d eua arch ls In A stig tue tes 5 esti– m24 ates W basit ed h AD on adm Hin D is tra ................................ tive enrollment and claims data. .................................................... 18 Charlson Comorbidity Index IndDe icapres tors s:ion Anxiety ADHD since it is the most common type of health plan and is less likely to differ from an HDHP in (unobservable) ways not by the fact that outcomes related to changes in health ca Figure 13 re use during the switch from PPO to HDHP are unobserved. p Erndno evalenctee of s these conditions in the population with employment-based health insurance is also especially high relative 0% De Depres press siion on A Anx nxiiet ety y A ADHD DHD • Utilization of preventive services including canc $1,975 er screenings and some vaccinations was also negatively v t he ac cnum ines. b Hee These ra of lth P em la pe rn Wit e rg ve enc nt h a iv y e d H e m p ee a aa rlttsu h Sa mreent s w v ings A vis erites ,c const w coun ere r uc ta : ll ne tA e d Fi g for va et- iv Y indiv e ea lyr idua a Sss tudy oc lsia ,if t " te Ed he B RI w y it m Ih th ss atue ce he B Hd rD ie tH he f,P no. . re Only c40 om4 (E m hospit ende mpa loy dl a ce rde itm eB r is e ia si ne o set ns fit , Chronic obstructive pulmonary disease/asthma 9% 8% 7% 14 Inpatient Hospital Emergency Primary Care Physician Specialist Office Visits Mental Health Office Prescription Drug Fills Impact of HDHP on He Imp alth C acta o re f S Hp D eH nding P on ................................ Use of Various Types ................................ of Prescription Dru ................................ gs Among ..............16 relat$2 ed ,00 to 0 plan design (e.g., HMO and EPO provider networks). We first selected individuals with each condition in t Suc o ot h o heut r m com ent ea s w l he ould alth d ult Tis ota im or l S a d pe te er nd ly si ng ( bFi e gim urp ea 2) cte . d I n 2019 by the , va a Tm lue otaong l E of mp t c lo h a yose r ee r Srpe e unde c nd eiiv nge r da,g for e 65 w hic with h d ea m t Ta p ota loy is l E una m mpe lont v ya e-e S ila bab pe sed le ndi for ng pla tns his , 8.1 $1,761 Figure 19, Annual Use of Various Ty pes of Out-of-Network Health Care Services Among Individuals Ages 18–64 With Discussion affecte Ad dm b isy s ion the s move from Depa a rtmP en PtO to an H OD ffiH ceP V . isits Visits hospit forth b ayl d the ay s, a U.S. nd Preem ve ent rgiv enc e Se y d rv eic pa ers Ta tm Pe ne sk nt umo For vis n $1,658 ia ic t V s for ea.c cin ati indiv on (W id ith uals with ADHD were not significantly affected by the health $14,0Re 00 sea Menta rc l He h Ial ns th tO itffi ut cD ee ia, b P20 ertima e14 s r(y w ). Ca it h ro er P w hy ith so icu ian t complications) Emerge 8% ncy Departmen5% t Inpatient Hos 1% pital $600 Individuals Ages 18 –64 With Depression or Anxiety and Individuals Figure 3 12.6 $1,931 $1,945 Primar$13,299 y Care Physician Office Visits Specialist Office Visits Mental Health Office Visits 2016 if they were continuously enrV olle isitsd in a PPO plan. Then, we distinguished those who switched into an HDHP in 1 a pna ercly esi nt s. w ere classified as having anxi Coe ng ty e, sti5.3 ve He part F ercai elnt ure , Chro had ni MD c D, and 2.8 percent had ADHD. Otherwise, fewer than 1 Sensitivity Analyses Visi ts................................ Office Visits ................................ Specialist Office Visits ................................ Visits Adm ................................ issions P$1,846 rescription Drug .............. Fills 20 The MedD ica ere pr e Mo ssion o derniz r aA tion nxie Ac tyt a ond f 2003 Indiv (MM idua A) ls c A re gaete s 5 d – H24 SA -W eliig th AD ible he HD alth ................................ plans or HDHPs. HDH ................................ Ps are health plans wit ... h 18 $1,808 Overall, our results are consi Mild lst ivee r nt dis e w ait se h and extend prior research tha 3% t finds that 3% use of hea 0lt .3h ca % re services decline 10% plan change. Ages 5 –24 With ADHD (per 100 Plan Members) Note: ADHD = attention deficit hyperactive disorder. Percentage $1,351 of Obs Pop tructivul e Pat ulmo io na n ry Di Un sede ase, r orAge 65 Diagnosed With $1,696 V HP ariou V Vaccis nation (Females, $1,50 0%0 8% 2017 (c 12 ontinuously enrolled) from those who remained in a PPO in 2017. perc • e nt $4 w 00 The ere r d ed iauc gnosed tions in use with aof ny he of atlt he h ca othe re rser mv eic nteas l he prom alth d pteis d or de dc elin rs.e s in overall health ca $326*r *e * spending. Employer Source: Employee Benefi $1,232 t Research Institute estimates based on administrative enrollment and claims data. 11.4 $321*** deductibles of at least $1,400 Cancfo er r (nind on-m ive id taua stal tic co ov re m ra etg ae s ta atnd ic) $2,800 for fami 3%ly coverag2% e in 2022. 0.2% 11.4 Fronstin, Paul, a Teta nd nuM. s Vac Chr cina istitoop n her Roebuck, " Di The abetes Re ) lationship Betw Ine fle uen n He za Va aclt ch P inatil oa nn Ty $1,541pe, Use of A Sp gee <= c26 ia)lty Medications, S Im tat pac istt ica of l H ADHP nalyso is n Out-of-Network Use when individuals with ch $1,143 ronic conditions move from a PPO to an HDHP. However, we found meaningful $2 d 81 iff **e * rences Discussion ............................................................................................................................................................20 Mental Health Disorders, 2013 –2019 Figure 20, Annua $1,079 l Spending on Out-of-Network HDe ea pres lth C sion are A Se nxr iet vic y es A ADHD mong Individuals Ages 18–64 With Depression or Our r $1 ese 2,0a 00 rch can help employers make targeted benefit desig 6% n decisions. They can also inform policymakers as they 1.5 Rheumatoid disease $11,368 2% $1,353 1% 0.1% spending fell by a $11,273 greater amount and percentage than overall spending. Employee spending increased $958 -1%*** -1%*** Similarly, prescription drug fills fell by 96.7 (ADHD), 68.9 (MDD), and 48.1 (anxiety) per 100 plan members. Nearly and Worker Productivity," EBRI Issue Brief, no. 454 (Employee $1,273 Benefit Research Institute, 2018). b 2 etween our prior work on the general population (Fronstin, Sepulveda, and Roebuck 2013) and our current work on $200 See$1 Exhi ,000bit 5.1 in https://me 5%ps.ahrq.gov/data_files/publications/cb25/cb25.pdf. Next, we m Anxi atcehe tyd a nd swit Ic ndivi hingd ua mels m A bg ee rs 1:1 t s 5–24o W non ith -A swit DHc D he ................................ rs using $a 1 ,1 p6 r7 opensity ................................ score model that include ........................ d 2016 19 grapple with a Seda llow tive/B ing en ze odi m az pepi loy ne er Fs t illso provide Hypn e otinha c/Anx nc ioe lyd tic Fi cov ll20 s e 13 rage 20 for 19 he Antia dep lth ca ressan ret F ser illsvices that pArDHD S event ti mu the lan te Fil xa lscerbation of The Concp lu re siv on ale ................................ nce of MDD, anxiety, ................................ and ADHD has also be................................ en inc $1,123 reasing. Betwe................................ en 2013 and 2019, the ........................... percentage of the 21 Cerebrovascular disease 2% 1% 0.2% 5% We fb ound ecaus tha e tthe only mov 5– e8 p from erc t ehe nt P of PO patto ietnt he s w HD ith -H 3%** P the sh * m ifte ent d a som l heea lt ph d re-d iseor duc detrible s in o spe urnding study ont fileo d us a e cr la s of im for hea altn o h ca utr-e. W 9%e estimated multivariate models of all of the health services utilization and cost measures using difference in 10 $1,025 one-half of the decline for MDD patients was for antidepressant medication. About five fewer sedative/benzodiazepine 0 individuals with mental health disorders. First, in our earlier work, we found no impact on inpatient and emergency 1.0 0.8 -4%*** -4%*** (baseline) values of member age, gender, relations $917 hip to policyholder (self, spouse, child/dependent), region Patients visited mental he Pa elt riph p hera rlov vaide sculr as 4 r dis– e6 t aseimes per year on avera 1% ge and fille1% d an averag 0.e 2 % of 16 prescriptions among 3 p chr oponic ula -5% t ion cond wiit tions h MD . D increased from 4.1 perc $869 ent to 5.3 percent, and the share with anxiety increased from 4.8 percent $0 8.1% Fr Sonst ee$1 Buc in, P 0,000 hmu aul, M eller a( r2009) tin J. ;Se Bund pulvoerf da (2012) , and ;M. Buntin Chrise top t ahe l. (r2006) Roeb ;uc $9,553 Fron k, " sC tin onsu andm Reore -D buc irekc t(e 201 d H 3) e;a lt and h P Fron lans sRe tin, duc See p ulv The edL ao , ng and - Figur d of eiff -21 ne er, e tw nc Im or ep k s (DiD a m cte of nt)a H . l he D TH his P a lt o m h p n P etrhod roob v 2% ide a e br m ili , tula a ynd of te e s a Re vec n fe reaiv ndo ing wem rVa ize ha rious d d- a cont n o Tyr ut p olle e -of s of d- ne trH ia tw el d a or lteh C ksi p gra n in im rea S rty e ha rc vta ic r ve e as Fr lue phy s for om sic ia Out a n vi n o -of si ut t- N c (om 2 e– tw 3 eor for k Re ferences ................................ -1.2 ................................ -5%*** ...........................................................................................21 prescriptions per 100 patients were also measured (Figure 13). Among patients with ADHD, the switch to an HDHP -5%*** department visits. In contrast, in t $714 his pap -2.5 er, we found a significa 8nt reduction in inpatient and emergency department -2.7 $500 Renal disease 1% 1% 0.1% 8% $652 (Northeast, Midwest, South, West), Charlson Comorbidity Index, and a partial vector of the -6% health services utilization those with ADHD, 30 prescriptions among those with anxiety, and 35 prescriptions among those with MDD (Figure 7). to 8.1 • peMovi rcent ng (Fi frgom ure a 3) P.P O A D to Ha Dn HD prevH aP le nc hae d inc a m reixe ased d im op nly ac tsl on ight us lye , fr of om out 2.3 p -of-ne ertcw eor nt kt o he 2.8 p alth ca ercre ent se . rv ices. While the -4.9* $573 0% -6%* Roebuck (2013). -5.2 pe 0.r 5cTe ent r) m or Use spe Of O cialis ut t p va istit ie ( nt 5 –P -6 p 6%** hysi e*r cc ia en Vis nt) (Fi its A gurnd e 18 Pr)e . sc Am ripti ong on pD artugs ient,s " w Hho ealtdh Af id ha fav ires o 32 ut -(of 6)-: ne 11 tw 26 or –k11 offic 34, e20 vis 13 it.s, otherw Pris ov e ide com r $518 s A pam raong ble tIrndivi eatm deua ntls a nd Agecs 18 ontr– ol 64 g rW oup ith D sue bp je re ctss s a ion o re erx A am nxi ine etd y a pnd re- Ia ndivi nd p doua st- ls int Ae grevs 5 ent–ion 24 to quantify was ass -$2oc 00iated with 62.3 fewer fills for stimulants or about two-thirds of the decrease in total drug fills for this cohort. Congestive heart failure -7%*** 1% 1% 0.1% visits a -10 m 8ong $446 people with mental health disorders. Second, in our earlier work, we found that office visits fell 15 Endno0% tes .............................................................................................................................................................23 and cost outcomes described below. Subjects were matched using the nearest-neighbor approach to within one-fifth of Some of these prescriptions may be for non-mental health conditions. There were 14 inpatient hospital admissions per probability of using an out-of-network mental health provider fell among those with ADHD and MDD, neither t t7% he he yt ra W evae it tr h ADHD m ae gnt ed e11 ff e ................................ .4 vis ct. Iit n th s to e m preent sent al he sta udy, th ................................ lth prov e ide DiD rs efor stim the at or MD ................................ ca Dp a tur nd es t AD he HD c ha cohor nget ................................ s ( fr aom nd 12 ba.6 vis selineit s to for follow the ........... -up) 19 4 -8%*** These$8 r,00 ed0uctions in use of Pep ptr ice u sc lcr eipti r on drug fills represent less than 1 1% prescripti 0.on 4% drug fil0 l .p 05 e % r person per year. p e Srecee nt Fig $0 a ure nd 1 pr in esc http ripti s:on //wd wrw u.gnimh. fills nih. fell 16 p gov/he era ce lth/ nt. sta Itis n con tics 9 /ts ruic astid , e in t . his paper, we found that office visits fell between 3 R We e fe corr re ob nc orae te sd our findings with other sources. Using data from the National Survey on Drug Use and Health 0.0 a standa -$40r0d deviation of the predicted propensity score. Given the relatively large sizes of the matching pools, very few 10 Fronst 0 pain, P tientasul, M amoang rtin J. those Sew pulv ith M eda D, Da, nd 10 M. ad m Chr issi isons tophe pe r rR 1 oe 00 b uc pakt,ie "nt Me s a dic ma ong tion U those tilizaw tiit on h a anxi nd e Atd yhe , arnd enc four e in a ad m He isasi lto h ns -10%the number of office visits nor overall spending on out-of-network care were affected by the plan change. 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 -9%** 20 * 13 2014 2015 2016 2017 2018 2019 2020 anxiety group (Figure D e19 me) n. tia Total annual spending averaged $2 059 .3% to $418 0for .1% out-of-n0 e.t 1w %ork primary care in the dependent variable for the group of members who switched from a PPO to an HDHP, compared with the Patients with ADHD had an average of 16 prescription drug fills per year, so a reduction of 96.7 drug fills per 100 plan and 7 percent, while prescription drug fills fell between 1 and 4 percent. Our findings may imply that individuals with -20 ( 5NSD 6% UH), we found that self-reporting of MDD episodes increased from 6.5 percent to 8 percent between 2013 and unmatche 6 d switchers (<0.1 percent) were dropped from the ensuing analyses. We assessed covariate balance across p Ae us r t10 in, P 0 p . aC t.ie , nt "As a n Im ntong rod uc thos tioe n to P with ADHD ropensit . y I t Sc isor w eor Me th noti thods for ng tha Re t d som ucing e m the ent E aff l he ectas o lth ca f Confounding re might be in Ob captur ser ed v ain oth tional er See http Sa s:v //ings Ac www.ec bount ri.org -/Edligib ocs/le de P faul lant." -s o Aur mceer/ic wabn J s/w our wsna -202 l o1/ f M 2021 ana-g w eodrk C pa la rc ee 19 -w e(ll12 ne)s: se -s40 ho0 rt –-e re 40 po 7rt. , -4% 20 pd13 f. . Hemiplegia or paraplegia 0.2% 0.2% 0.2% -5% ide phy -$6 nt si 00 ic cia al c n offi hange ce v for isit s; $44 the g 5.3% r3 ou –$5 p of 11non for -spe swit cc ia he lisrts offi (i.ec., e tvhos isits; a e who nd r $2 em ,381 aine –d $3 in ,13 a 6 for PPO) m . e The ntal he diffe ar lt -e 0.3 h o nce ff ic in t e v he isse its two $5,832 Figur -0.5es members reduces average use to about 15 prescription -0.4drug fills per year. mental health d-is 0.4 orders are less sensitive than the general population to financial incentives when it comes to routine $6,000 W 20he 19n e . These mploy e er st s r im aa iste e s a derdeuc sltight ibles, ly t highe hey d ro tha so t n w o b ha ettt e wre m found anag ein c the lairim cs d osta s. ta , The whic y a h is re tw ryha ing t w to e beaxpe lanc ctee d the to ir fin eff dor , ts to t he two groups and found Acv ue ter y m yfe ocw ar dins ial t ina fa nc rcte ios of n statistically signific 0a .5nt % differenc 0.e 5% s or relativ 0% e bias exceeding 10 percent. service cSt ate udie So guor rce s ie : ,M " se. Mult d icFor al Ex iv pe a exa n rd ia itu m tre ep Pa B lee n, eha lp Su rv im rv ior ea y a r Iny l Re s uc raa nse re ce Com aprhy cp h osi n46 ec nia t (M (ns 3) EPS : oft -39 IC), e9 n p d– ata 4r 24 gov en, eide ra 20 te d11 m at e a hd tt.p ic s:/a /dtaion ma tatools.ahna rq.gg ove /m m ep e -0.5* snt -ic. for ment -a 0.5 l health 4.8% 6 ( 5% Figure 20). differences repre -0.6 sent ** s the estimated impact of moving from a PPO to an HDHP. While this technique controls for See Figure 8 in Copeland (2021). Note: Data not collected in 2007. Fi ser gur vicee s 1, (iA .e v.e, roffic age e A nnu visita s a l Em ndp p loy ree sc er-ipti Only on ad nd rugs) Fam aily nd Dte ha duc t the tible y , arAem m oor ng eW sensit orker iv s in P e to fin riva atnc e-ia Se l inc ctor e nt Est iva eb s w lish he mn ent it s co W m iteh s Fronst-in, P $800 aul, Stu -14a %** rt *Hagan, Olivia Hoppe, and Jake Spiegel, "The Mo $4,603 re Things Change, the More They Stay the Same: r be ed cuc aus e -30 elow not -v e av lue er yhe one alt h ca wit Mh or dMD e e rser atD e/ s vsee eic ve ers w e k s t livehil r r e da e ist einc m ase ee nt nt . iv B izeing caus the e t us hee N of SD 0.high 1UH d % -voe alue s not 0se .1% rcvont ices a.in q Our ue 0% st fin ions dings a on a re nxi lim etit ye , d w b e y the p W ae tie cnt ond s d uc ur te ing d thi rout s pine roc offic ess for e v is ea itc s. h o f the following p3.7 airs of years: 2016/2017; 2017/2018; and 2018/2019. Matched -1.0 -15% 4 4.1% It is not surprising that the drop in medication use specific to the mental health condition was a smaller proportion of -0.9 -29.7*** the covariates included in the propensity score matching 3.3 and multivariate DiD models, the effect estimates can -15%*** to 7 less frequently used ser AID vic Ses (i.e., inpatient services and emerge 0.nc 23.0 % y departm 0.2 e% nt visits). 0% A ustin, P. aC D ., e"d Op ucttim ible al C , 20 alipe 02 3.0 –r20 W20 idth ................................ s for Propensity-Scor................................ e Matching When Estim ................................ ating Differences in Me ......................... ans and 4 - 4% 10%An Analysis of the Generosit -$868 y* *of * Employment-Based Health Insurance, 2013–2019," EBRI Issue Brief, no. 545 r fa eICD c lie t d t-ha 9on -tCM ou seconda = tc o Inte merna rs yr r e tional e la sea ted r C cth to c la o scsha ificonfir nge ation s in he mo our f Dis a fin e lta h ca d se ing s,r e 9th of us inc R ee d r ve ur isaion ing sing , Cl the p inic re swit vaa l le Mo cnc h fr de ific . om a One tion, PP r O e as nd te o a H ICD rc D h st H -10 P udy found a -CM re unob = Inte ser trna ha ve ttional d p.r Suc ior th o control $4,0 su 00 bjects were only permitted entry once. Finally, we aggregated the three sets of matched data to form a Switching from a PPO to an HDHP had a mixed impact on use of out-of-network health care services (Figure 21). -$1,000 2.6 -1.2 the overall decline in the MDD and anxiety groups compared with ADHD. Given that the -16 %** MD *D and anxiety samples Empst loy ille b rs ha e bia vs ee b de d eue n inc to runm easing easu derd euc d c tible hara s b cte er cis aus tice s c itor isr e one late of d w the ith b mot ost h th effeic d ieen ctis p ion to s lan desi wg itn ch ch pala nge ns s t (or o not adop ) atnd to -10% -1.5 -$984*** Overall spe ( DEiff m nding eprloy enc e — e es in P B ae nd ne rtfit o he p Re or ptrsea ions opor rc in Ob th I ion o nsser tf itit uv tp e aa , tiona i20 d b 21 y l Stu ). the d e ie m s,p" loy Pha err m —a w ce aut s ia cls al Sta o calc tis ula tictse d 1 0: 150 and pr–e161 sent , e 20 d. 11 W b.he n enrolled t out Cl he as cs C oific OVI me ation s w D-19 ould o f pDis a ult nde ea im sm eas ic t,e , 10th ly a nxi bee R im teyv p is ra a ion c te te s w , d Cl b inic eyr eta he inc l Mo v re a da lue ific sing, aof tion. c ea spe re c re iaclly eiv in y -e 1.4 d*,* oung *for w a hic dult h d s a(tG aood is un win e avatila ab l. le 20 for 20 t )his . L ong ana- ly te sirsm . 1.9 single- 40 analytical dataset, while retaining indicators for each sample pair. This approach allowed for higher-powered Out-of-network mental health office visits declined 18 percent among patients with ADHD and 12 percent among Figure 2, Percentage of Population Under -1.5 Age 65 Diagnosed With Various Mental Health Disorders, 201 2.8%9 .................... 6 were3% muc 2 h old -18%** er, * these Othe part ie Con ntdi s tw ion es r:e more likely to have other chronic diseases, likely treated with pharmacotherapy. -12%*** m Wa he na tn e he ge m us the pe loy of cost eras r nd of ais spe pe rov d nding e iding duc ton ible hehe as, t lta h b lt he h se eyne drfit o vic so t s. es. W o he ben a tten e r mm ana ploy gee c r ost incsr. e aThey a ses the re he tra ylt ing h ptla o n d bala ed nc uc et ible their , it e ff only orts t mo ust -$1,200 4 -$1,137*** in a PPO, average annual total health care spending was over $13,000 for patients with M $1,931 2.3% DD, over $11,000 among -$1,149*** trends in suicide rates were also increasing, a potential consequence of MDD and anxiety. statistical analyses. Our final sample sizes were 27,730 for individuals with MDD, 41,522 for those $1,72w 0ith anxiety, and 8 those with MDD. Other estimates were not statistically significant. Not shown in the charts are the findings -20 $2% ,000 Anxiety 43% 23% See -Cha 2.0 rlson et al. (1987), Deyo, Cherkin, and Ciol (1992), and Quan et al. (2005). -19%*** For e -xa 15% mple, 16 percent of patients with MDD or anxiety were diagnosed with hypertension, and 14–15 percent also c re ha duc nge e low one- vnum alueb he er. a ltIh ca t is m reuc ser h e vic asi es w er thil o inc e inc ree ant se ivdiz eing ductthe ible us s teha of n to d high- o va thin lueg se s lik rve ic e cs ha . nge It is ins onur e a of ncte he c a re rraie sons rs, a w lte hry Brot-Goldberg, Zarek C., Amitabh Chandra, Benjamin R. Handel, and Jonathan T. Kolstad, "What Does a Deductible Do? G Our oe trze ese l, a Ron Z., rch can he Enid lp Ce hun mpg loy Roe ers m mea r, ke C atllope arget e Hd olingue benefit , M de . si Dg an d niele ec is Fa ions llin, . K They c atherine an a Mc lsC o leinf aror y, m W p illolic iamy m E$1,229 a atk on, ers a Jas t cqhe uey lin e Figure 3, Percentage of Population Under Age 65 Diagnosed With Various Mental Health Disorders, 2013–2019 ........... 6 those 2% with anxiety, and approximately $5,800 for those with ADHD (Figure 8). Employers’ spending was 85 percent of -$1,305*** 11,406 -$1 ,40 for 0 those with ADD H eD pr. es One sion -half of each sample switched from the PPO 29% to the HDHP 13% , while the other half re -5 la 0ted to the number of office visits among those using out-of-network health care providers and the findings Given the non-normal distributions of our dependent variables, we estimated nonlinear DiD models. For had dyslipidemia. 9 e ne mtp wloy ork es, m rThe s ha ov Ivm e e p e ta o m ca tb of rhigh acC eost d - pte he -Sha rfor conc r m ing ing ep on tne of H tw e va or alt lue k h C , -m b aov a re sed e P tr o ic ins e as, Qua ur lim anc itee d nt d ne it eie si tw s, a gn or , nd kw , hic cSp ha he nge nding allow the s D efor y m na p m loy m ula icesrr,y s t " , N offe o Bp ErR Wor rov hide ealt k thhe ing -ris rk P ig aht pe r 0Agnew, et al.,"Mental Health in the Workplace: A Call to Action Proceedings from the Mental Health in the The consiC dOVI er aD llow -19ing pande emp m loy ic e ha rs s to exa prcov eride bat e ed nha me nc nt ea dl he cov ae ltrh iss age ue fors he naatlt iona h ca lly re a se nd rv in t icehe s t ha wor t p krpela vc ee nt . tB he et w exa eec n Au erbagtus ion o t 20 f 20 See Austin (2011a) and Austin (2011b). total spending for MDD and anxiety and 79 percent of spending for patients with ADHD — leaving member cost sharing -2.5 Schizophrenia/bipolar disorder 5% 5% 4% remained in the PPO. -$1,463*** Figure 4, Demographics, by Mental Health Disorder ................................................................................................. 8 inpa relatteie dnt to hospit the im aliz paa cttion of s a mov nd ing day frs a omnd a e Pm PO ertg o ea nc n HD y deH pP a ro tm n o ent ve v ra isll spe its, wnding e use d on tw out o-p -of ar-tne mtod wor els k ( he pa arlt t h ca 1: prre ob . it; Primary Care Physician Office Visits Specialist Office Visits Mental Health Office V -18 is%** its * 1% -$1,60$0 0 a mss ix of essm inc e Nnt e o. ns, offe t21 ive63 s t2 ro fin (b Ne a atnc ttiona eia r e l inc nga l Bur eg nt e ea iv p u e as for of tieE nt c s in t ono biom m he eic tr ir ic Re he sc sea arlt erh ca ecnin h, r20 ges, a 15 (Cudit he ). ht rne ttp he w :// , p Rose w la w n for wn, .nb anon e nd r.or - Fe e gligib ndr /pap ic le e k r d s/ 20 ew p07 e 21 nd ).63 ent 2.p s, e df.t c. c ahr nd onic Feb c rW ua ond or ryikt p ions 20 la21 ce . , : tPhe ublic prop He or atltion h Sum of am dult it."s sh Jour ow na ing l of Occ sympupa tomts of iona l a MD nd D or Env air nxi onm etye nt inc arl M eased edic ine from 60 36 (4) .4 p : 32 erc 2e –nt 330 to , 41.5 Hypertension 16% 16% 1% accounting for 15 percent of total health care spending for patients with M -2.6D **D * and anxiety and 21 percent for those 10 -20% A s we have previously re Tota por l Stpe ed nd w ing ith the general popula Tota tion, l Emp Hlo D yH erP Ss pea nd ls ing o negatively affect T ota usl E e mp of lop ye re e S ve pe nt ndi iv ng e services, SeThese e https r:e //su ww lts w w.ke ffr.e o rg als /w o onot mens st-a hte is atlth ica -lly po lisi cg y/ni fact fica -snt he . et/gender-differences-in-health-care-status-and-use-spotlight-on- pa -60 rt 2: negative binomial regression); and for all other visits and prescription drug fills measures, we employed -25% -3.0 Figure 5, Presence of Hea Dlt ysh lipC idond emiaitions, by Mental Health Disorder15% ................................ 14% ................................ 1% .......... 8 percent (2018 Vahr **. *p< ad tia 0oi:10.1 .0n e 1, **t p < a 009 l. .0520 , 7/ *p< 21 JOM. 0.1)0. . 00 Ac00 cor 00 ding 0000 to 00 the 12 7 20 1.21 EBRI/Greenwald Research Workplace Wellness Survey, one-half As a side note, we also conducted a sensitivity analysis that examined the impact of moving from a PPO to a plan with with ADHD 0% . While few patients used inpatient services, because of their high cost, they accounted for a large Note: ADHD = attention deficit hyperactive disorder. e me vens n w -he he alth/ n th **.*p< ey 0 .0 a 1r , e ** pc < ov 0.0e 5, r*p< ed0 .1 in full 0. pre-deductible (Fronstin, Sepulveda, and Roebuck 2013). This was the case in the negativeNot be inom : HDHP = ial r hig e hg -dree dss uctiion. ble hea A lthll d planic , ADH hot D om = att ou entis a on dny efic ius t hye pe ra or cti vp er de isv oe rdnt er.ive care measures were modeled using probit. W B Muc uc he hm h r n HD ue esea lle HP r rc , s w h ha Thom ere s b a fir s C est en d .int , "on C ro od ens uc to um eunde de in 200 r-Or rst ie a3 a nt nd es p d the Hae rim a t lt of p h C atche ta r t ha e Me Re t d tfor ic he a m rd ee St P druc reasc ttible erg ipti ie in s: A on he D Re a rug ltv h sa ie Iw mv p of ings a rov the em c E e cvnt ount ide , a nc nd (e H - 62 S Mo on A .3* )d Ma - *e *eligib rna nizg a le e tion d Note: ADHD = attention deficit hyperactive disorder. ***p<0.01Ma , **jp o< r 0 Depres .05, *p< s0 iv .1 e D 0. isorder (MDD) Anxiety Disorders 5 Attention Deficit Hyperactivity Disorder (ADHD) This study w Source: Em as c ployond ee Be uc net fie t Res d t eh arc roug h Instih th tute ee sti m EaB teRI s b aC see dnt one ar d mfo inir s tra Rtie ve sea enro rlc lm h o ent a n He nd claa imlt s h B data. enefits Innovation (EBRI CRHBI), of workers a Nrot ee: a AD t le HDa = st at te m ntod ion e derfia ct it e hy ly pe c rac onc tive e dir sne order. d about their emotional well-being. And workers and employers strongly Source: Employee Benefit Research Institute estimates based on administrative enrollment and claims data. a n HRA on services use and spending. For this work, we simply re-executed the analytical dataset construction process proportion o Note*: ** ADH f tot p<0.D = 01a , l sp a ** ttp< en ti 0. e o05 nding n d , e *p< fici0. t h 10 ( yFi p .eg raur ctive e d9) iso. rd eP r.rescription drug fills accounted for a sizable percentage of total spending as Note: HDHP = high-deductible health plan, ADHD = attention deficit hyperactive disorder. Figure 6, Proba **b *p< ili 0t .0 y1 , of **p< Re 0.0c 5e , *p< iving 0.10.Various Types of Health Care Services Among Individuals Ages 18–64 With Depression present study as well (Figure 14). For example, we found reductions in breast cancer screening for women with MDD, Source: Employe So e u Be rce ne:fi Em t Res ple oa yrc ee h In Be sti ntu efte it R ese tis m eaa te rch s b a In ss etd it u on te a e ds m tiim nisa tra teti sv b e a es ne rod ll m on en a t d am ndi n cils aitm rast id ve ata e. nrollment and claims Finally, for inpatient hospitalization and emergency department spending, we used two-part models (part 1: he Acta, ltth p he -70ir la C a ns om b ili — pte EBRI Is y ta itt nd o ion a p high r sov uend ide Bri -d C ee fonsu c d iov s re uce g tr m ible ist ae g erre e -he d D for iir na e t lt hc ch p ee t U e rtd .S la a .in H ns Pe ap t a e(r n lt H e th I v D an eH d nt ns P Tr iv s) ur ae d a m e ser nc mor ae rv k e, ic " O ge fMilba e fs p in cee . r rIior a S nk lly SN t Qua :o — 08sa 8hr 7t a t –is e s 1r fa 3ha ly 7c X t87 d /io 9 0o n o 0 ( n use 4) 88f th 7 : –82 1 3 e of 7 0 X he –he /841 9a 0a lt $lt h p , .h ca 52 000 + la .5n d r9 0e . se edruc vic tible es a nd Sourc Not e: e Em : HD plHP = oyee Be hign he -d fit eRes ducti eb alrc e h h e In as lth titu pte la n e, sADH timate D s= ba att se en dti o on n a dd em ficin it is htra ype tiv ra ec e tin vro e d llm iso erd nt e ar. nd 6 claims data. Source: Employee Benefit Research Institute estimates based on administrative enrollment and claims data. with the fun Notd e: ing HDHP = sup hip ghor -detd of uctibtlhe e he afollow lth plan, ing ADHD or = g atta eniz ntiona dte ions ficit h: ypA era on ctiv, eB dilue sord eC r.ross Blue Shield Association, ICUBA, JP a gree that emp Soloy urcee : r Em s ha ployv eee Be an erfie t sponsibilit Research Institu yte for esti m w aor tes k be asr es d’ om n ae dm nt ina isl he trativea elt nh. rollm e E nt m anp d loy claime sr ds a ata. re looking for ways to address the described above, as well as the statistical analyses presented below. data. w ell. Source: Employee Benefit Research Institute estimates based on administrative enrollment and claims data. 3 or Anxiety and Individuals Ages 5–24 With ADHD ...................................................................................... 9 cervical cancer So sc urcree : e Em nin plog ye efor Ben w efit om Rese ean w rch Init sh M titute e D sD tim aa tend s b aa senxi d one atdy m, inP isSA trati vt ee est nroing llmenfor t and m clae im n w s data it. h anxiety, and lipid testing for both probit; part 2: generalized linear model with gamma family and log link); and for all other visits and prescription w spe as lim nding ite .d . E a Und rly e wro trhe k foc IRS usg ed uida onnc re ese , unt arcil t h q he ue d steions duc tp ible ert a isini m ng et, to cov pla er n ch ageoic doee s not (e.g., inc enr lude olle e " acny ha rser actveic ris e tor ics, r beis ne k fit doi:10.1111/j.1468-0009.2009.00580.x. Morgan Chase, Pfizer, and PhRMA. mental health needs of workers given the current economic climate. And they are especially interested in addressing © 2022, Employee Benefit Research Institute –Education and Research Fund. All rights reserved e e e e e e e e e e e e e e e e e e e e e eb b b b b b b b b b b b b b b b b b b b b br r r r r r r r r r r r r r r r r r r r r ri. i. i. i. i. i. i. i. i. i. i. i. i. i. i. i. i. i. i. i. i. i.o o o o o o o o o o o o o o o o o o o o o or r r r r r r r r r r r r r r r r r r r r rg g g g g g g g g g g g g g g g g g g g g g IIIIIIIIIIIIIIIIIIIIIIs s s s s s s s s s s s s s s s s s s s s ss s s s s s s s s s s s s s s s s s s s s su u u u u u u u u u u u u u u u u u u u u ue e e e e e e e e e e e e e e e e e e e e e B B B B B B B B B B B B B B B B B B B B B Br r r r r r r r r r r r r r r r r r r r r rief ief ief ief ief ief ief ief ief ief ief ief ief ief ief ief ief ief ief ief ief ief A re • • • • • • • • • • • • • • • • • • • • • • s e Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar Mar arcc c c c c c c c c c c c c c c c c c c c c c hh h h h h h h h h h h h h h h h h h h h h h r e 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1p 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0,,,,,,,,,,,,,,,,,,,,,,o r 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2t0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 f2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 ro 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2m • • • • • • • • • • • • • • • • • • • • • • tN N N N N N N N N N N N N N N N N N N N N N he o o o o o o o o o o o o o o o o o o o o o o ......................E 555 555 555 555 555 555 555 555 555 555 555 555 555 555 555 555 555 555 555 555 555 555 BRI Education and R esearch Fund © 2022 Employee Benefit Research Institute 11 22 13 14 15 16 17 18 10 19 20 12 21 23 8 2 3 5 4 6 9 7