Variation in GP decisions on antihypertensive treatment in oldest-old and frail individuals across 29 countries

RESEARCH ARTICLE Open Access Va ia ion in GP decisions on an ihype ensi e ea men in oldes -old and ail indi iduals ac oss 29 coun ies S en S ei 1† , Ma jolein Ve schoo 1† , Nicolas Rodondi 1,2 , Daiana Bon im 3 , Robe A. Bu man 4 , Clai e Collins 5 , Ge asimo ska Ki ano ska Biljana 6 , Sand a Gin e e 7 , Raquel Gómez B a o 8 , Ka h yn Ho mann 9 , Claudia I ode 10 , Kaspe L. Johansen 11 , Ngai e Ke se 12 , Tuomas H. Koskela 13 , Sanda K ei maye Peš ić 14 , Dona a Ku pas 15 , Ch is ian D. Mallen 16 , Hube Maisoneu e 17 , Ch is oph Me lo 18 , Yolanda Muelle 19 , Ch is iane Mu h 20 , Ma ija Pe ek Š e 21 , Fe dinando Pe azzuoli 22,23 , Thomas Rosemann 24 , Ma in Sa le 25 , Zuzana Š adlenko á 26 , A hina Ta sioni 27 , Hans Thulesius 28 , Vic o ia Tkachenko 29 , Pe e To zsa 30 , Rosy Tsop a 31,32 , Tuz Canan 33 , Ri a P. A. Viegas 34 , Shlomo Vinke 35 , Ma go W. M. de Waal 37 , And eas Zelle 36 , Jacobijn Gussekloo 37 and Rosalinde K. E. Poo lie 37* Abs ac Backg ound: In oldes -old pa ien s (>80), ew ials showed e icacy o ea ing hype ension and hey included mos ly he heal hies elde ly. The esul ing lack o knowledge has led o inconsis en guidelines, mainly based on sys olic blood p essu e (SBP), ca dio ascula disease (CVD) bu no on ail y despi e he high p e alence in oldes - old. This may lead o a ia ion how Gene al P ac i ione s (GPs) ea hype ension. Ou aim was o in es iga e ea men a ia ion o GPs in oldes -olds ac oss coun ies and o iden i y he ole o ail y in ha decision. Me hods: Using a su ey, we compa ed ea men decisions in cases o oldes -old a ying in SBP, CVD, and ail y. GPs we e asked i hey would s a an ihype ensi e ea men in each case. In 2016, we in i ed GPs in Eu ope, B azil, Is ael, and New Zealand. We compa ed he pe cen age o cases ha would be ea ed pe coun ies. A logis ic mixed-e ec s model was used o de i e odds a io (OR) o ail y wi h 95% con idence in e als (CI), adjus ed o SBP, CVD, and GP cha ac e is ics (sex, loca ion and p e alence o oldes -old pe GP o ice, and yea s o expe ience). The mixed-e ec s model was used o accoun o he mul iple assessmen s pe GP. Resul s: The 29 coun ies yielded 2543 pa icipa ing GPs: 52% we e emale, 51% loca ed in a ci y, 71% epo ed a high p e alence o oldes -old in hei o ices, 38% and had >20 yea s o expe ience. Ac oss coun ies, conside able a ia ion was ound in he decision o s a an ihype ensi e ea men in he oldes -old anging om 34 o 88%. In 24/29 (83%) coun ies, ail y was associa ed wi h GPs’decision no o s a ea men e en a e adjus men o SBP, CVD, and GP cha ac e is ics (OR 0.53, 95%CI 0.48–0.59; ORs pe coun y 0.11–1.78). Conclusions: Ac oss coun ies, we ound conside able a ia ion in s a ing an ihype ensi e medica ion in oldes - old. The ail oldes -old had an odds a io o 0.53 o ecei ing an ihype ensi e ea men . Fu u e hype ension ials should also include ail pa ien s o acqui e e idence on he e icacy o an ihype ensi e ea men in oldes -old pa ien s wi h ail y, wi h he aim o ge e idence-based da a o clinical decision-making. Keywo ds: Hype ension, Oldes -old, Clinical a ia ion, Gene al p ac i ione s, F ail y, Elde ly * Co espondence: [email p o ec ed] † Equal con ibu o s 37 Depa men o Public Heal h and P ima y Ca e, Leiden Uni e si y Medical Cen e , Hippoc a espad 21, 2333 ZD Leiden, The Ne he lands Full lis o au ho in o ma ion is a ailable a he end o he a icle © The Au ho (s). 2017 Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. S ei e al. BMC Ge ia ics (2017) 17:93 DOI 10.1186/s12877-017-0486-4 Backg ound Hype ension is he mos impo an p e en able cause o poo ca dio ascula ou come and is esponsible o dis- abili y and dea hs om s oke, myoca dial in a c ion and o he diseases [1]. T ea ing hype ension is bene icial and (since he 1990s) i is known ha ea men also e- duces s oke a es and myoca dial in a c ion in pa ien s aged >60 yea s [2–4]. As li e expec ancy has inc eased wo ldwide, a new e m was needed o desc ibe hose in he as es -g owing age g oup expec ed o iple wi hin he nex 35 yea s [5], i.e. he g oup ‘oldes -old’is now de ined as hose aged >80 yea s. The popula ion o he oldes -old is he e ogeneous. Some oldes -old a e e y heal hy whe eas o he s a e mul imo bid wi h complex p oblems. Al hough he g oup o mul imo bid oldes -old is apidly inc easing, mos ials s ill exclude hem. Messe li e al. highligh ed his commonly-applied exclusion by applying exclusion c i e ia aken om 13 hype ension ials wi h oldes -old pa icipan s, o a p ima y ca e coho o hype ensi e pa ien s aged >60 yea s [6]: in his case, ≥70% o he oldes -old would ha e been excluded and hey we e bo h olde and sicke . The exclusion o such a la ge pe cen age o oldes -old has caused a se ious gap in ou knowledge and in guide- lines o ea hype ension in pa ien s wi h mul imo bid- i y. E en mo e sca ce a e ecommenda ions o ail pa ien s: o example, o six cu en hype ension guide- lines, only hose o he Eu opean Socie y o Hype en- sion and o he Eu opean Socie y o Ca diology ha e a speci ic ecommenda ion o lea e decisions on an ihy- pe ensi e he apy in he ail and oldes -old pa ien s o he ea ing physician (class I C ecommenda ion) [7]. Due o he cu en lack o clea e idence, he bes managemen o hype ension in he oldes -old e- mains unknown; his may, in u n, lead o clinical a ia ion. Al hough i is di icul o quan i y, a ia ion exis s in he way ha he bes a ailable e idence is applied in clinical p ac ice [8]. Among he di e se easons o his a ia ion, he app op ia eness o guidelines o physicians in ea ing speci ic g oups o pa ien s is o pa icula impo ance. Howe e , o e- duce clinical a ia ion and imp o e quali y o ca e/pa- ien sa e y, he e is a need o assess clinical a ia ion among he oldes -old pa ien s, who a e consis en ly excluded om ials bu su e om bo h mul imo - bidi y and ail y. The e o e, he p esen s udy in es iga es clinical a i- a ion ac oss coun ies o gene al p ac i ione s’(GPs) de- cisions o s a an ihype ensi e ea men in pa ien s aged >80 yea s. Ou hypo hesis was ha ail y would be an impo an ac o in deciding no o s a an ihype - ensi e ea men in clinical p ac ice, al hough his is no speci ically add essed in mos guidelines. Me hods Design GPs om di e en coun ies we e in i ed o pa icipa e in a su ey based on case igne es. Se ing The aim was o ec ui na ional ep esen a i es (de ined as a GP in con ac wi h a na ional GP ne wo k) o 40 coun ies on he Eu opean con inen , and in B azil and New Zealand. We also e-con ac ed six na ional ep e- sen a i es o GP ne wo ks pa icipa ing in a p e ious su ey [9]. Also in i ed o pa icipa e we e: 1) na ional ep esen a i es o WONCA Eu ope (Eu opean B anch o he Wo ld O ganiza ion o Na ional Colleges, Academies and Academic Associa ions o Gene al P ac i ione s/Family Physicians) [10]; 2) he Eu opean Gene al P ac ice Resea ch Ne wo k (EGPRN) [11]; and 3) he Ne wo k o Junio GPs in Eu ope ( he Vasco da Gama Mo emen , VdGM) [12]. The s udy was conduc ed in acco dance wi h he Dec- la a ion o Helsinki [13]. Because he esponses o GPs we e collec ed anonymously, mos coun ies equi ed no app o al om an e hics commi ee. In coun ies whe e app o al was manda o y (Swi ze land, B azil), a wai e om he e hics commi ee was ob ained. In New Zealand, app o al o he s udy was g an ed by he Uni e si y o Aukland E hics Commi ee. Pa icipan s All na ional ep esen a i es we e asked o include as many GPs as possible om hei GP ne wo k. Because p ima y ca e su eys usually sco e low on esponse a es, we egula ly epo ed he numbe s o pa icipa ing GPs o he na ional coo dina o s, so hey could send e- minde s i needed. The only inclusion c i e ium o he su ey was o be ac i ely wo king as a GP; his was asked a he beginning o he su ey. Pa icipan s who did no mee his c i e ium (e.g. due o e i emen ) we e excluded om comple ing he su ey. P ocedu es Be o ehand, we de eloped/ es ed he su ey o op imal echnicali y be ween Su eyMonkey (www.su eymonkey. com, Palo Al o, CA, USA) and S a a, among i e GPs. Then, o es o cla i y/ easibili y, he su ey was pilo ed among a sample o 16 physicians wo king in Swi ze land. Na ional ep esen a i es ansla ed he su ey om English o hei own language. Finally, he su ey was a ailable in 21 languages. Na ional ep esen a i es o G eece, Is ael and Finland decided o dis ibu e he su - ey in English. The co ec ness o all ansla ions was e alua ed by he eam o collabo a o s. The su ey can be accessed online (see Addi ional ile 1). Fi s , we asked he GP’s gende , o ice loca ion (ci y, subu - ban, u al), and yea s o expe ience wo king as a GP (in 5- S ei e al. BMC Ge ia ics (2017) 17:93 Page 2 o 7 yea s bands). Second, GPs we e asked o es ima e he p opo ions o pa ien s aged >80 yea s a ending hei GP o ice. Thi d, eigh case igne es we e p esen ed o oldes - old pa ien s o bo h gende , p esen ing o a ou ine isi in a GP o ice wi hou blood p essu e- ela ed symp oms and no ecei ing any an ihype ensi e ea men . Fo each case igne e, GPs we e asked o decide i hey would s a an i- hype ensi e ea men . All case igne es di e ed in h ee p ima y cha ac e is ics: sys olic blood p essu e (SBP), ca - dio ascula disease (CVD), and ail y (see Addi ional ile 2). SBP was ei he 140 mmHg o 160 mmHg. CVD was ei- he p esen (e.g. case igne es wi h a his o y o myoca - dial in a c ion o s oke) o absen . Because he condi ion o ail y lacks a common de ini ion [14], we s a ed ha ail y is de ined as pa ien s wi h a leas wo o he ollow- ing c i e ia: unin en ional weigh loss, exhaus ion, low le el o ac i i y, muscle weakness, and slow gai speed. Thus, a pa ien wi h a low le el o ac i i y and unin en ional weigh loss was conside ed o be ail. To acili a e illing in he su ey, o each case igne e we indica ed one o he ol- lowing s a emen s: “You conside his pa ien o be ail”o “You don’ conside his pa ien o be ail”. The su ey was dis ibu ed by email be ween Ma ch 9 and July 31 2016. As he only excep ion, Uk aine dis ib- u ed he su ey on pape du ing a egional GP mee ing because he e is insu icien in e ne access o GPs in Uk aine. S a is ical analysis To desc ibe baseline cha ac e is ics, p opo ions we e cal- cula ed o dicho omized o ca ego ized da a, and means we e calcula ed o con inuous da a. To assess in e na ional a ia ion in decisions o ea - men , pe coun y he c ude p opo ions and con idence in e als (CI) we e calcula ed o GPs who would s a ea men . To assess he ole o ail y in he decision o s a ea - men pe coun y, odds a ios (ORs) and CI we e calcu- la ed pe coun y using a mixed-e ec s model adjus ed o GP’s gende , yea s o expe ience, o ice loca ion, p e a- lence o oldes -old in he GP p ac ice, guideline compli- ance, SBP, and CVD. The mixed-e ec s model was used o accoun o he mul iple assessmen s pe GP. The es i- ma e o each coun y was p esen ed on a o es plo . Fo each case igne e, we calcula ed he c ude p o- po ions o GPs s a ing ea men and also compa ed wo co esponding case igne es (e.g. in Case 1 he pa- ien is no ail, whe eas in Case 2 he pa ien is ail). To assess he o e all in luence o SBP, CVD and ail y, he same mixed-e ec s model was used bu , in addi ion, clus e ing wi hin coun ies was aken in o accoun . A wo-sided p- alue o 0.05 was conside ed s a is ically signi ican . Analyses we e pe o med wi h STATA 14.2 (S a aCo p, College S a ion, TX, USA). Resul s F om Ma ch h ough July 2016, we con ac ed 40 na- ional ep esen a i es om Eu ope, B azil, Is ael, Russia, and New Zealand and ecei ed eplies om 29 coun- ies. O e all, 13,671 GPs we e in i ed, o whom 2585 esponded. Subsequen ly, 42 esponden s we e excluded because hey we e no longe wo king as a GP, esul ing in 2543 pa icipan s. The median esponse a e was 26% (IQR 10–62%) (see Addi ional ile 3). Table 1 p esen s he baseline cha ac e is ics o he pa - icipa ing GPs; 52.3% we e emale, 50.8% li ed in a ci y, and 37.6% had >20 yea s o expe ience. The majo i y o GPs (61.3%) es ima ed he p e alence o he oldes -old pa ien s in hei p ac ice o be >10%. O e all, he c ude p opo ions o ea men a ied conside ably be ween coun ies (Fig. 1). Fo example, he lowes p opo ion o ea men was ound in he Ne he lands (34.2%; 95% CI 32.0–36.5%) whe eas Uk aine had he highes p opo ion (88.3%; 95% CI 85.3–90.9%). Figu e 2 shows he GPs’ ea men p obabili y in ail oldes -old compa ed o non- ail oldes -old o each o he 29 coun ies. O e all, he ea men p obabili y o all coun ies was OR 0.59 (95% CI 0.47–0.75) and he p obabili y pe coun y anged om OR 0.11 in New Zealand o 1.78 in he Czech Republic. In 8/29 (28%) coun ies (i.e. New Zealand, Finland, Denma k, he Ne he lands, I eland, Swi ze land, F ance and Is ael) we a e 95% con iden ha GPs would be less likely o s a an ihype ensi e ea men in he ail oldes -old pa ien s compa ed o he non- ail oldes -old pa ien s. In 16/29 (55%) coun ies, an OR <1 was ound bu a Table 1 Baseline cha ac e is ics o pa icipa ing GPs om 29 coun ies Baseline cha ac e is ics (N= 2543) n(%) Female GP 1341 (52.3) P ac ice loca ion Ci y 1292 (50.8) Subu ban 599 (23.6) Ru al 651 (25.6) Expe ience as GP < 5 yea s 471 (18.5) 5–10 yea s 445 (17.5) 11–15 yea s 341 (13.4) 16–20 yea s 328 (12.9) > 20 yea s 956 (37.6) Sel -es ima ed p e alence o pa ien s >80 yea s a own p ac ice < 10% 851 (38.7) 10–20% 865 (39.4) 21–30% 323 (14.7) > 30% 159 (7.2) S ei e al. BMC Ge ia ics (2017) 17:93 Page 3 o 7 95% CI including 1; his la ge 95% CI was due o he lowe numbe o esponden s pe coun y (<30 pe coun y in 45% o all coun ies). In 5/29 (17%) coun ies, he OR was >1 bu ( o a la ge ex en ) he 95% CI included 1. GPs’decision o ea hype ension in he oldes -old a ied conside ably, anging om 17.3% o 96.8% ac- co ding o he speci ic case igne e (Table 2). The low- es le el o ea men decision was sco ed in hose case igne es ha included no ail y, no CVD, and a SBP Fig. 1 Na ional pe cen ages in which gene al p ac i ione s decide o s a an ihype ensi e ea men in all eigh cases o oldes -old pa ien s (unadjus ed) Fig. 2 In luence o ail y on 2053 gene al p ac i ione s (GPs) when deciding o s a an ihype ensi e ea men pe coun y (adjus ed a ). a Adjus ed o GP cha ac e is ics (gende , expe ience, loca ion, p e alence o oldes -old, guideline compliance) and pa ien cha ac e is ics (ca dio ascula disease, sys olic blood p essu e). A mixed-e ec s model was used o accoun o mul iple assessmen s pe GP. Al hough 2543 GPs pa icipa ed, missing da a on GPs’decisions o ea he eigh cases means ha only 2053 GPs a e included he e S ei e al. BMC Ge ia ics (2017) 17:93 Page 4 o 7 140 mmHg (17.3%; 95% CI 15.7–19.0%). The case i- gne es ha included CVD, SBP 160 mmHg and no ail y sco ed he highes (96.8%; 95% CI 95.9–97.5%). Besides ail y (adjus ed OR 0.53; 95% CI 0.48–0.59), a SBP o 140 mmHg (adjus ed OR 0.01; 95% CI 0.01–0.01) and no CVD (adjus ed OR 0.29; 95% CI 0.26–0.32) we e also independen ac o s ha caused GPs no o s a ea men . Discussion A e sampling >2500 GPs in 29 coun ies, his s udy e- ealed la ge clinical a ia ion in s a ing an ihype ensi e ea men ( anging om 34 o 88%) based on case i- gne es o oldes -old pa ien s. As hypo hesized, ail y p o ed o be an impo an pa ien cha ac e is ic o GPs in deciding whe he o no o s a an ihype ensi e ea men in 24/29 (83%) coun ies. The p obabili y o a GP ea ing a ail pa ien was almos hal ha compa ed wi h a GP managing a non- ail pa ien . Cu en guide- lines a e clea e abou he le el o SBP ela ed o ini ia - ing ea men ; his was con i med in he p esen s udy in which GPs we e less inclined o s a ea men in he case o SBP 140 mmHg compa ed o SBP 160 mmHg. Ne e heless, how o manage ail y will become inc eas- ingly impo an o an inc easingly olde and mul imo - bid popula ion. When speci ic da a om u u e ials ha include ail pa ien s become a ailable, hype ension and o he guidelines can be upda ed acco dingly. Scien i ic and clinical con ex o he esul s T ea men goals o hype ension a e cons an ly chan- ging [15]. Recen ials including oldes -old pa ien s indi- ca e aiming a he lowe le els o SBP [3, 16]. Howe e , hese la e pa ien s may di e om he gene al popula- ion ha GPs a e managing, due o he ex ensi ely ap- plied exclusion c i e ia o he olde and sicke pa ien s [6]. The e o e, i emains unclea whe he lowe ing SBP in mul imo bid and ail pa ien s does in ac lead o be e ou comes. Fo example, in he SPRINT ial, ail pa ien s showed smalle in e ea men g oup di e - ences in SBP compa ed o non- ail pa ien s, hus a lowe SBP migh be ha de o achie e in ail pa ien s [16]. On he o he hand, he e is e idence ha ail oldes -old need a highe SBP. In a ecen me a-analysis compa ing p o- and e ospec i e coho s udies, Zhang e al. ound ha a highe SBP in ail oldes -old pa ien s had a p o ec i e e ec in lowe ing he isk o o e all mo ali y [17]. Thus, cu en knowledge seems o be well summa ized by Ma e son e al. who sugges ed o e alu- a e and ea ail oldes -old pa ien s indi idually, while he heal hie oldes -old should be ea ed ega dless o hei ch onological age [18]. In he p esen s udy, his wide spec um o ecommen- da ions and lack o clea e idence may pa ly explain he a ia ion ound be ween he pa icipa ing coun ies. Di - e ences in na ional guidelines/campaigns may ha e also led o di e ences be ween he coun ies. Ne e heless, his s udy con i med ou hypo hesis ha ail y is a ac o ha GPs ake in o conside a ion when s a ing an ihype en- si e ea men ; mo eo e , we ound ha GPs we e less likely o ea ail pa ien s, e en a e adjus ing o SBP and CVD. This is in line wi h indings om a Du ch quali- a i e s udy, whe e ulne abili y was an impo an pa ien - ela ed ba ie o GPs when implemen ing guidelines o seconda y ca dio ascula p e en ion in oldes -old [19]. In e es ingly, ou indings sha e some indings and ye show di e ence wi h he only o he published s udy on his opic. Me mans e al. conduc ed a simila su ey among 305 GPs in Belgium. These au ho s also ound la ge di e ences in ea men in en ions o hype en- sion in he oldes -old pa ien s be ween GPs and showed ha he e was a signi ican di e ence in he ea men in en ion o GPs be ween obus pa ien s and s ongly dependen pa ien s. Howe e , he s a ed ha ‘di e ences in he pa ien s’le el o dependency we e no esponsible o he a ia ion in he o e all ea men in en ion’[20]. Table 2 Pe cen ages o gene al p ac i ione s (GPs) s a ing an ihype ensi e ea men o he eigh indi idual cases (n= 2053 GPs) Cases P opo ion o GPs s a ing ea men Case Cha ac e is ics % (95% CI) F ail y CVD SBP 160 mmHg O e all 54.9 (54.1–55.7) Case 1 17.3 (15.7–19.0) −−− Case 2 18.2 (16.6–20.0) + −− Case 3 85.4 (83.7–86.9) −−+ Case 4 75.6 (73.6–77.5) + −+ Case 5 96.8 (95.9–97.5) −++ Case 6 84.9 (83.2–86.4) + + + Case 7 32.5 (30.4–34.6) −+− Case 8 29.5 (27.5–31.6) + + − CVD ca dio ascula disease, SBP sys olic blood p essu e Al hough 2543 GPs pa icipa ed, missing da a on GPs’decisions o ea he eigh cases means ha only 2053 GPs a e included he e S ei e al. BMC Ge ia ics (2017) 17:93 Page 5 o 7 Howe e , on an in e na ional le el, when including many coun ies, ail y was es ablished as an impo an ac o in luencing GPs’ ea men decisions. S eng hs and limi a ions A s eng h o his s udy is he high numbe o coun ies and ela i ely la ge numbe o esponden s ( hanks o collabo a ion wi h WONCA Eu ope, EGPRN, and VdGM). Fu he , he sampled GPs we e expe ienced wi h ea ing oldes -old pa ien s. The inclusion o many coun- ies enabled o p oduce a de ailed map o ea men decision-making in Eu ope and elsewhe e. In addi ion, we could es ablish ha , in mos coun ies, ail y is associa ed wi h a lowe in en ion o ea , e en when aking SBP and ca dio ascula como bidi y in o accoun . This s udy has se e al limi a ions. Fi s , al hough we e- po wha he GPs s a ed hey would do, his is no neces- sa ily he same as wha hey would ac ually do. Howe e , gi en he ealis ic case desc ip ions and he anonymous na u e o he su ey, we a e ela i ely con iden ha his limi a ion has no in oduced a sys ema ic bias. Second, he esponse a e a ied conside ably be ween coun ies and he median a e was only 26%; his is a commonly oc- cu ing p oblem in p ima y ca e su eys [21]. Howe e , ou esponse a e was well wi hin he ange o o he pub- lished su ey among GPs in majo jou nals [22]. Se e al e iews u he no ed ha a low esponse a es in GP su - ey do no necessa ily in oduce selec ions bias [23, 24]. Thi d, in he case igne es, only h ee pa ien cha ac e is- ics we e aken in o conside a ion. Howe e , because we ocused on a ia ion in ea men decision and he ole o ail y in ha decision, i was beyond he scope o his s udy o add ess all possible easons ela ed o GPs’ ea - men decision-making. Fou h, we mainly ec ui ed one GP ne wo k pe coun y, which is a selec ion o GPs dependen on hei egion o o igin o a ea o in e es ; howe e , by adjus ing ou analysis o GP cha ac e is ics we aimed o ake his possible con ounde in o accoun . Implica ions This s udy has se e al implica ions o esea ch and clin- ical p ac ice. Fi s , he la ge a ia ion in s a ing ea - men in hype ensi e oldes -old calls o high-quali y coho s udies o (ideally) new hype ension ials speci - ically including ail pa ien s o acqui e e idence as o whe he ail y is indeed an impo an ac o when ea ing hype ension in oldes -old pa ien s. Second, u u e s udies should in es iga e whe he ea men a ia ion migh be ex- plained by e.g. he ecommenda ions in guidelines ha indi- idual GPs ollow. Thi d, quali a i e s udies could help us o unde s and mo e o he a ia ion we ha e ound. I ea- sons o he in e na ional a ia ion in ea men a e es ab- lished, educa ional campaigns can be launched o uni y he quali y o ca e in Eu ope (and elsewhe e) based on he cu en body o e idence. Finally, u u e hype ension guidelines should s a i y hei ecommenda ions no only o age, blood p essu e le el and ca dio ascula como bid- i y, bu also o ail y. Conclusions In Eu ope, B azil, Is ael and New Zealand, GPs’deci- sions conce ning s a ing an ihype ensi e ea men in he oldes -old a ied conside ably. Independen ly, he ail oldes -old pa ien s had an almos 50% lowe p ob- abili y o hei GP o conside hem eligible o ecei e an ihype ensi e ea men . Fu u e hype ension ials should also include ail pa ien s o acqui e e idence on he e icacy o an ihype ensi e ea men in oldes -old pa ien s wi h ail y, wi h he aim o suppo and uni y clinical decision-making. Addi ional iles Addi ional ile 1: Su ey. (DOCX 49 kb) Addi ional ile 2: Cha ac e is ics o he eigh case igne es used in his su ey. (DOCX 18 kb) Addi ional ile 3: Pa icipa ing coun ies: numbe o in i ed GPs and esponse a es pe coun y. (DOCX 21 kb) Abb e ia ions CI: Con idence in e al; CVD: Ca dio ascula disease; EGPRN: Eu opean Gene al P ac ice Resea ch Ne wo k; GP: Gene al p ac i ione ; IQR: In e qua ile ange; OR: Odds a io; SBP: Sys olic blood p essu e; VdGM: The Vasco da Gama Mo emen ; WONCA Eu ope: Eu opean B anch o he Wo ld O ganiza ion o Na ional Colleges, Academies and Academic Associa ions o Gene al P ac i ione s/Family Physicians Acknowledgemen s The au ho s hank Gun a Ticmane o helping o ec ui he gene al p ac i ione s in La ia and hank all he pa icipa ing gene al p ac i ione s om he pa icipa ing coun ies. Funding D . S ei ’s esea ch is suppo ed by g an s (P2BEP3_165353) om he Swiss Na ional Science Founda ion (SNF) and he Go ied and Julia Bange e -Rhyne Founda ion, Swi ze land. This s udy was suppo ed by he Swiss Uni e si y Con e ence and he S a e Sec e a ia o Educa ion, Resea ch and Inno a ion (SUC p ojec P-10). A ailabili y o da a and ma e ials The da ase used and analysed du ing he cu en s udy is a ailable om he co esponding au ho on easonable eques . Au ho s’con ibu ions D . SS and Ms. MV con ibu ed equally. D . SS, D . JG, D . RKEP had ull access o all da a in he s udy and ake esponsibili y o he in eg i y o da a and he accu acy o he da a analysis. S udy concep and design: SS, MV, JG, RKEP. Acquisi ion, analysis, o in e p e a ion o da a: all au ho s. D a ing o he manusc ip : SS, MV, JG, RKEP. C i ical e ision o he manusc ip o impo an in ellec ual con en : all au ho s. S a is ical analysis: SS, JG, RKEP. Ob ained unding: SS, JG, NR. Adminis a i e, echnical, o ma e ial suppo : MV, SS, JG, RKEP. S udy supe ision: JG, RKEP. All au ho s ead and app o ed he inal manusc ip . Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. Consen o publica ion No applicable. S ei e al. BMC Ge ia ics (2017) 17:93 Page 6 o 7 E hics app o al and consen o pa icipa e The s udy was conduc ed in acco dance wi h he Decla a ion o Helsinki. [13] Because he esponses o GPs we e collec ed anonymously, mos coun ies equi ed no app o al om an e hics commi ee. GPs exp essed hei consen o pa icipa e by esponding o ou su ey. In coun ies whe e app o al was manda o y (Swi ze land, B azil), a wai e om he e hics commi ee was ob ained. In New Zealand, app o al o he s udy was g an ed by he Uni e si y o Aukland E hics Commi ee. Publishe ’sNo e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. Au ho de ails 1 Ins i u e o P ima y Heal h Ca e (BIHAM), Uni e si y o Be n, Be n, Swi ze land. 2 Depa men o Gene al In e nal Medicine, Inselspi al, Be n Uni e si y Hospi al, Uni e si y o Be n, Be n, Swi ze land. 3 Hospi al Is aeli a Albe Eins ein, São Paulo, B azil. 4 Vennesla P ima y Heal h Ca e Cen e, Be gen, No way. 5 I ish College o Gene al P ac i ione s, Dublin, I eland. 6 Depa men o Neph ology and Depa men o Family Medicine, Uni e si y Clinical Cen e, Uni e si y S . Cy il and Me odius, Skopje, Macedonia. 7 Facul y o Medicine, Depa men o Family Medicine, Riga S adiņs Uni e si y, Riga, La ia. 8 Ins i u e o Heal h and Beha iou , Resea ch Uni INSIDE, Uni e si y o Luxembou g, Luxembou g, Luxembou g. 9 Depa men o Gene al P ac ice and Family Medicine, Cen e o Public Heal h, Medical Uni e si y o Vienna, Vienna, Aus ia. 10 Timis Socie y o Family Medicine, Sano Med Wes P i a e Clinic, Timisoa a, Romania. 11 Danish College o Gene al P ac i ione s, Copenhagen, Denma k. 12 School o Popula ion Heal h, Uni e si y o Auckland, Auckland, New Zealand. 13 Depa men o Gene al P ac ice, Uni e si y o Tampe e, Tampe e, Finland. 14 Family Medicine Depa men , Heal h Cen e Tuzla, Medical School, Uni e si y o Tuzla, Tuzla, Bosnia and He zego ina. 15 Family Medicine Depa men , W oclaw Medical Uni e si y, W ocław, Poland. 16 P ima y Ca e and Heal h Sciences, Keele Uni e si y, Keele, S a o dshi e ST5 5BG, UK. 17 P ima y Ca e Uni , Facul y o Medicine, Uni e si y o Gene a, Gene a, Swi ze land. 18 Ins i u e o P ima y and Communi y Ca e Luce ne (IHAM), Luce ne, Swi ze land. 19 Ins i u e o Family Medicine Lausanne (IUMF), Lausanne, Swi ze land. 20 Ins i u e o Gene al P ac ice, Goe he-Uni e si y, F ank u / Main, Ge many. 21 Depa men o Family Medicine, Medical acul y, Uni e si y o Ljubljana, Ljubljana, Slo enia. 22 SNAMID (Na ional Socie y o Medical Educa ion in Gene al P ac ice), P a a Sanni a, I aly. 23 Depa men o Clinical Sciences in Malmö, Cen e o P ima y Heal h Ca e Resea ch, Lund Uni e si y, Malmö, Sweden. 24 Ins i u e o P ima y Ca e, Uni e si y Hospi al Zu ich, Uni e si y o Zu ich, Zu ich, Swi ze land. 25 SSLMG, Socie é Scien i ique Luxembou gois en Medicine gene ale, Luxembou g, Luxembou g. 26 O dinace Řepy, s. .o., P ague, P ague, Czech Republic. 27 Resea ch Uni o Gene al Medicine and P ima y Heal h Ca e, Facul y o Medicine, School o Heal h Sciences, Uni e si y o Ioannina, Ioannina, G eece. 28 Family Medicine, Depa men o Clinical Sciences, Lund Uni e si y, Malmö and senio esea che Region K onobe g, Växjö, Sweden. 29 Depa men o Family Medicine, Ins i u e o Family Medicine a Shupyk Na ional Medical Academy o Pos g adua e Educa ion, Kie , Uk aine. 30 Depa men o Family Medicine, Semmelweis Uni e si y, Budapes , Hunga y. 31 LIMICS, INSERM, U1142, F-75006 Pa is, Uni e si é Pa is 13, So bonne Pa is Ci é, UMR_S 1142, F93000 Bobigny, So bonne Uni e si és, UPMC Uni e si é Pa is 06, UMR_S 1142, F75006 Pa is, Pa is, F ance. 32 Leeds Cen e o Respi a o y Medicine, S James’s Uni e si y Hospi al, Becke S ee , Leeds LS9 7TF, UK. 33 Family Medicine Specialis , Kemaliye Town Hospi al, E zincan Uni e si y, E zincan, Tu key. 34 Family Doc o , In i ed Assis an o he Depa men o Family Medicine, NOVA Medical School, Lisbon, Po ugal. 35 Sackle Facul y o Medicine, Tel A i Uni e si y, Tel A i , Is ael. 36 Cen e o P ima y Heal h Ca e (uniham-bb), Basel, Swi ze land. 37 Depa men o Public Heal h and P ima y Ca e, Leiden Uni e si y Medical Cen e , Hippoc a espad 21, 2333 ZD Leiden, The Ne he lands. Recei ed: 8 Feb ua y 2017 Accep ed: 11 Ap il 2017 Re e ences 1. Wilson PW. Es ablished isk ac o s and co ona y a e y disease: he F amingham s udy. Am J Hype ens. 1994;7(7 P 2):7S–12S. 2. S aessen JA, Faga d R, Thijs L, Celis H, A abidze GG, Bi kenhage WH, Bulpi CJ, de Leeuw PW, Dolle y CT, Fle che AE, e al. Randomised double-blind compa ison o placebo and ac i e ea men o olde pa ien s wi h isola ed sys olic hype ension. The sys olic hype ension in Eu ope (Sys -Eu ) ial in es iga o s. Lance . 1997;350(9080):757–64. 3. Becke NS, Pe e s R, Fle che AE, S aessen JA, Liu L, Dumi ascu D, S oyano sky V, An ikainen RL, Niki in Y, Ande son C, e al. T ea men o hype ension in pa ien s 80 yea s o age o olde . N Engl J Med. 2008;358(18):1887–98. 4. P e en ion o s oke by an ihype ensi e d ug ea men in olde pe sons wi h isola ed sys olic hype ension. Final esul s o he Sys olic Hype ension in he Elde ly P og am (SHEP). SHEP Coope a i e Resea ch G oup. JAMA. 1991, 265(24):3255–3264. 5. The Uni ed Na ions: Wo ld Popula ion P ospec s: The 2012 Re ision. A ailable a h p://esa.un.o g/unpd/wpp. Accessed 31 Ma 2017. 6. Messe li FH, Sulicka J, G yglewska B. T ea men o hype ension in he elde ly. N Engl J Med. 2008;359(9):972–3. au ho eply 973-974 7. Hype ension EETF MoA: 2013 P ac ice guidelines o he managemen o a e ial hype ension o he Eu opean Socie y o Hype ension (ESH) and he Eu opean Socie y o Ca diology (ESC): ESH/ESC Task Fo ce o he Managemen o A e ial Hype ension. Jou nal o hype ension 2013, 31(10):1925–1938. 8. Kennedy PJ, Lea hley CM, Hughes CF. Clinical p ac ice a ia ion. Med J Aus . 2010;193(8 Suppl):S97–9. 9. den Elzen WP, Le eb e- an de Flie AA, Vi gini V, Mooijaa SP, F ey P, Kea ney PM, Ke se N, Mallen CD, McCa hy VJ, Mu h C, e al. In e na ional a ia ion in GP ea men s a egies o subclinical hypo hy oidism in olde adul s: a case-based su ey. B J Gen P ac . 2015;65(631):e121–32. 10. h p://www.woncaeu ope.o g/, Accessed 31 Ma 2017. 11. h p://www.egp n.o g/, Accessed 31 Ma 2017. 12. h p:// dgm.woncaeu ope.o g/, Accessed 31 Ma 2017. 13. Wo ld Medical A. Wo ld medical associa ion Decla a ion o Helsinki: e hical p inciples o medical esea ch in ol ing human subjec s. JAMA. 2013; 310(20):2191–4. 14. S e nbe g SA, We sho Schwa z A, Ka unanan han S, Be gman H, Ma k Cla ield A. The iden i ica ion o ail y: a sys ema ic li e a u e e iew. J Am Ge ia Soc. 2011;59(11):2129–38. 15. P e e MA, McMu ay JJ. Lessons in unce ain y and humili y - clinical ials in ol ing hype ension. N Engl J Med. 2016;375(18):1756–66. 16. Williamson JD, Supiano MA, Applega e WB, Be lowi z DR, Campbell RC, Che ow GM, Fine LJ, Haley WE, Haw ield AT, Ix JH, e al. In ensi e s s anda d blood p essu e con ol and ca dio ascula disease ou comes in adul s aged >/=75 yea s: a andomized clinical ial. JAMA. 2016;315(24): 2673–82. 17. Zhang XE, Cheng B, Wang Q. Rela ionship be ween high blood p essu e and ca dio ascula ou comes in elde ly ail pa ien s: a sys ema ic e iew and me a-analysis. Ge ia Nu s. 2016; 18. Ma e son BJ, Ga cia-Es ada M, P es on RA. Hype ension in he ail elde ly. J Am Soc Hype ens. 2016;10(6):536–41. 19. an Pee PG, D ewes YM, Gussekloo J, de Ruij e W. GPs' pe spec i es on seconda y ca dio ascula p e en ion in olde age: a ocus g oup s udy in he Ne he lands. B J Gen P ac . 2015;65(640):e739–47. 20. Me mans E, Deg yse J, Vaes B. T ea men in en ions o gene al p ac i ione s ega ding hype ension in he oldes old: a igne e s udy. BMC Fam P ac . 2016;17(1):122. 21. McA oy BR, Kane EF. Gene al p ac ice pos al su eys: a ques ionnai e oo a ? BMJ. 1996;313(7059):732–3. discussion 733-734 22. Hyman DJ, Pa lik VN. Sel - epo ed hype ension ea men p ac ices among p ima y ca e physicians: blood p essu e h esholds, d ug choices, and he ole o guidelines and e idence-based medicine. A ch In e n Med. 2000; 160(15):2281–6. 23. Kelle man SE, He old J. Physician esponse o su eys. A e iew o he li e a u e. Am J P e Med. 2001;20(1):61–7. 24. Asch DA, Jed ziewski MK, Ch is akis NA. Response a es o mail su eys published in medical jou nals. J Clin Epidemiol. 1997;50(10):1129–36. S ei e al. BMC Ge ia ics (2017) 17:93 Page 7 o 7