Orig­i­nal source pub­li­ca­tion: Meira, B. and F. de Sá-Soares (2013). Trust Fac­tors in the Adop­tion of National Elec­tronic Health Records. Pro­ceed­ings of the 15^th IEEE Inter­na­tional Con­fer­ence on e-Health Net­work­ing, Appli­ca­tions and Ser­vices—Health­com 2013. Lis­bon (Por­tu­gal).

Trust Fac­tors in the Adop­tion of National Elec­tronic Health Records

Bruno Meira and Fil­ipe de Sá-Soares

Cen­tro Algo­ritmi, Uni­ver­si­dade do Minho, Guimarães, Por­tu­gal

1. Introduction

The adop­tion of a national Elec­tronic Health Record (EHR) has sig­nif­i­cant impacts on the way cit­i­zens inter­act with the health sys­tem and the health pro­fes­sion­als. To judge the suc­cess of an EHR, the ini­tial degree of accep­tance of the sys­tem by its intended users should be con­sid­ered as well as the sub­se­quent use of the sys­tem. Actu­ally, the for­mal adop­tion of this type of sys­tems is, in most cases, irre­versible. What this means is that if the sys­tem does not ade­quately fit users ’ needs, it is rather unlikely that the health pro­fes­sion­als and patients revert to what­ever sys­tem they had pre­vi­ously. The most likely sce­nario would be that both health pro­fes­sion­als and patients would make min­i­mal use of the sys­tem, caus­ing the sys­tem to fall short in terms of its promised ben­e­fits.

Besides key tech­ni­cal fac­tors, other issues may influ­ence the suc­cess­ful adop­tion of EHR by its stake­hold­ers, par­tic­u­larly health pro­fes­sion­als and patients. Con­sid­er­ing the com­plex­ity and sen­si­tive­ness of the infor­ma­tion manip­u­lated by such a sys­tem, we pos­tu­late that trust fac­tors also play an impor­tant role. Indeed, the lit­er­a­ture show s there is a strong rela­tion­ship between trust in a tech­no­log­i­cal sys­tem and its use [Gefen et al. 2003; Kas­sim et al. 2010].

This study aims to iden­tify and clas­sify the main trust fac­tors that have an influ­ence in the adop­tion of the EHR by health pro­fes­sion­als and patients. By know­ing these fac­tors, design­ers will be bet­ter pre­pared to project EHR sys­tems aligned with users’ require­ments and expec­ta­tions, and imple­menters will increase their chances of cor­rectly man­ag­ing the change process asso­ci­ated with the intro­duc­tion of this kind of sys­tems.

The remain­ing sec­tions of this paper are orga­nized as fol­lows: in Sec­tion 2 (Con­cepts) we review the core con­cepts for this work; in Sec­tion 3 (Method­ol­ogy) the adopted method­ol­ogy for the study is out­lined ; in Sec­tion 4 (Study Descrip­tion) we pro­vide details about the con­ducted field study; in Sec­tion 5 (Results) we present our main find­ings; fol­lowed by their dis­cus­sion in Sec­tion 6 (Dis­cus­sion); and in Sec­tion 7 (Con­clu­sion) we draw con­clu­sions and sug­gest future work.

2. Concepts

To prop­erly ana­lyze the issue of trust in the EHR, it is impor­tant to con­cisely define both con­cepts.

The EHR is a repos­i­tory of infor­ma­tion regard­ing the health of a sub­ject of care, stored in a com­puter read­able for­mat [ISO/TC215 2005]. Besides defin­ing the EHR in terms of con­tent, it is also impor­tant to define it in terms of pur­pose. “The pri­mary pur­pose of the EHR is to pro­vide a doc­u­mented record of care that sup­ports present and future care by the same or other clin­i­cians. This doc­u­men­ta­tion pro­vides a means of com­mu­ni­ca­tion among clin­i­cians con­tribut­ing to the patient s care. The pri­mary ben­e­fi­cia­ries are the patient and the clin­i­cian(s)” [ISO/TC215 2005]. The clear iden­ti­fi­ca­tion of EHR main ben­e­fi­cia­ries is cru­cial, since it helps defin­ing the bound­aries of what an EHR is or should be.

To ade­quately con­cep­tu­al­ize trust, it is impor­tant to first rec­og­nize the mul­ti­ple dimen­sion­al­ity of the con­cept. “It has dis­tinct cog­ni­tive, emo­tional and behav­ioral dimen­sions which are merged into a uni­tary social expe­ri­ence” [Lewis and Weigert 1985]. In this paper trust is defined as the will­ing­ness of a party to be vul­ner­a­ble to the actions of another party based on the expec­ta­tion that the other will per­form a par­tic­u­lar action impor­tant to the trustor, irre­spec­tive of the abil­ity to mon­i­tor or con­trol that other party” [Mayer et al. 1995]. The con­cept of risk is essen­tial to this def­i­n­i­tion: by trust­ing some­one, one is essen­tially tak­ing a risk.

Trust in the EHR implies trust at very dif­fer­ent lev­els. The doc­tor patient rela­tion­ship is a del­i­cate one, and the imple­men­ta­tion of an EHR adds new com­plex­ity to that rela­tion­ship. Both clin­i­cian and patient (albeit for dif­fer­ent rea­sons) need to trust in the EHR. If they do, both patient and clin­i­cian need to trust in who designed the EHR they are using, in who is respon­si­ble for main­tain­ing the infor­ma­tion and in the sys­tems man­ag­ing that infor­ma­tion, among oth­ers. To trust in the EHR is to del­e­gate dis­cre­tionary pow­ers (whether con­sciously or not) to addi­tional enti­ties.

Trust fac­tors may be clas­si­fied accord­ing to three dimen­sions: Dis­po­si­tional, Insti­tu­tional and Inter­per­sonal [McK­night 2001].

The dis­po­si­tional dimen­sion relates to an indi­vid­ual’s dis­po­si­tion to trust oth­ers, and intends to cap­ture ones asser­tions about its peers. In this dimen­sion the entity that is the tar­get of the trust is irrel­e­vant.

The insti­tu­tional (or struc­tural) dimen­sion cap­tures the envi­ron­ment in which trust rela­tion­ships occur. Sim­i­larly to the prior dimen­sion, the insti­tu­tional dimen­sion relates to exter­nal fac­tors that may influ­ence, gov­ern or limit any given trust rela­tion­ship. In the par­tic­u­lar case of the doc­tor patient rela­tion­ship, the indi­vid­ual trust placed by a par­tic­u­lar patient in a par­tic­u­lar clin­i­cian is influ­enced by the eth­i­cal code to which the clin­i­cian adheres to.

The inter­per­sonal dimen­sion is the per­sonal side of the trust rela­tion­ship, and there­fore relates to the indi­vid­ual per­cep­tions and inten­tions of the trustor in regard to the trustee. In con­trast to the pre­vi­ous dimen­sions, the inter­per­sonal dimen­sion is based on judg­ments and per­cep­tions about spe­cific enti­ties. In a broad sense, the envi­ron­ment in which trust occurs and one’s per­cep­tion about oth­ers influ­ence one’s per­cep­tions and inten­tions in rela­tion to a par­tic­u­lar sub­ject. This dimen­sion may be sub­di­vided into per­cep­tual, inten­tional and behav­ioral cat­e­gories. Of these, the per­cep­tual cat­e­gory is par­tic­u­larly rel­e­vant for this study since it cap­tures the trust­ing beliefs of trustors regard­ing trustees. At the oper­a­tional level it includes four aspects, namely com­pe­tence (to be able to do for one what one needs done), benev­o­lence (to act in one’s inter­est instead of oppor­tunis­ti­cally), integrity to make good faith agree­ments, tell the truth and ful­fil l promises) and pre­dictabil­ity (to fore­cast trustee’s actions based on the con­sis­tency of pre­vi­ous actions).

The com­bi­na­tion of the three dimen­sions of trust influ­ence ones behav­ior towards oth­ers. There­fore, the behav­ioral per­spec­tive of trust is the phys­i­cal man­i­fes­ta­tion of trust (or the lack of it). As an ex ample, by shar­ing infor­ma­tion with some­one we are behav­ing as if we trust that some­one.

Most of these con­cepts are usu­ally expressed in terms of peo­ple trust­ing other peo­ple. How­ever, this is not nec­es­sar­ily always so. Trust between peo­ple is often medi­ate d by insti­tu­tions (Insti­tu­tional dimen­sion). Trustor and trustee often trust that a given insti­tu­tion can set­tle even­tual dis­agree­ments. In the case of trust in tech­nol­ogy, the same prin­ci­ples apply, with the excep­tion that, con­trary to humans, tech­nol­ogy lacks free will and moral or eth­i­cal prin­ci­ples [McK­night 2005]. Apart from the lack of free will or moral, which impacts the inter­per­sonal dimen­sion of trust, the remain­ing aspects of trust are equally applic­a­ble to the analy­sis of trust in tech­no­log­i­cal arti­facts.

3. Methodology

The meth­ods used in this study will be out­lined in terms of data col­lec­tion and data analy­sis. Regard­ing the col­lec­tion of data, the cho­sen approach was a field study, in which a series of inter­views were con­ducted. The tech­nique used in these inter­views can be described as Active Inter­views [Hol­stein and Gubrium 1995]. The selected tech­nique allows the sub­ject of the inter­view a cer­tain degree of free­dom, and a higher inter­ac­tion between the inter­vie­wee and the inter­viewer. Given the objec­tive at hand, under­stand­ing what raises or low­ers the user s trust in the EHR, a too strict approach (like a ques­tion­naire) would raise some con­cerns. Namely, it would be very dif­fi­cult to ascer­tain if the inter­vie­wee under­stood what was being asked of him and, also, it would not give the inter­vie­wee a chance of pre­sent­ing prob­lems or con­cerns not pre­vi­ously fore­seen by the authors.

For the process of data analy­sis, the cho­sen approach was Con­tent Analy­sis based on Grounded The­ory, in which the propo­si­tions or the­o­ries are derived from a sys­tem­atic data analy­sis process [Myers 2012]. The generic tech­nique used for the data analy­sis process is known as cod­ing , the process in which the inter­vie­wee s answers are grouped together by asso­ci­at­ing them with com­mon ideas or con­cepts [Rubin and Rubin 1997]. This tech­nique involved the def­i­n­i­tion of a code­book: a doc­u­ment con­tain­ing the codes, their mean­ing and instruc­tions for apply­ing them. The basis for defin­ing the code­book was the trust frame­work pro­posed by [McK­night 2001] and briefly reviewed in the pre­vi­ous sec­tion.

4. Study Description

The field study took place between the months of June and August of 2012 in Por­tu­gal. Briefly, and to put this study into con­text, in 2009 the Por­tuguese Gov­ern­ment launched an ini­tia­tive to imple­ment a national EHR, named RSE (Reg­isto de Saúde Eletrónico, in Por­tuguese). In 2011 the ini­tia­tive was recon­fig­ured to the imple­men­ta­tion of a web based elec­tronic ser­vice to access and manip­u­late EHR.

The num­ber of sub­jects inter­viewed was 63, of which 17 (27%) are health pro­fes­sion­als and 46 (73%) are patients. From the 17 health pro­fes­sion­als, six are nurses, five are gen­eral prac­ti­tion­ers, five are spe­cial­ized doc­tors and one is a psy­chol­o­gist.

The aver­age length of the inter­views was 11 min­utes for the patients (with a min­i­mum, max­i­mum and stan­dard devi­a­tion of 5.3, 24.4 and 4.2, respec­tively) and 27 min­utes for the health pro­fes­sion­als (min­i­mum, max­i­mum and stan­dard devi­a­tion of 14.4, 47.1 and 7.2, respec­tively). All inter­views were con­ducted in places and times of the day cho­sen by the inter­vie­wees. All inter­views were audio recorded.

Although we do not claim (nei­ther intended) to have formed a sta­tis­ti­cally rel­e­vant sam­ple (the selected meth­ods alone would make this task sig­nif­i­cantly dif­fi­cult), a n effort was made to assure there was a min­i­mal cor­re­spon­dence between the patients’ pro­files and national indi­ca­tors regard­ing age and edu­ca­tional level.

Each of the two groups inter­viewed (patients and health pro­fes­sion­als) had dif­fer­ent inter­view scripts. The main dif­fer­ence between the two scripts was the focus of the top­ics. The patient script was more focused in deter­min­ing the level of com­fort with infor­ma­tion tech­nol­ogy, the level of aware­ness to patient rights regard­ing health infor­ma­tion, the inter­ac­tion between the patient and the health sys­tem and the patients’ opin­ions and thoughts on what they val­ued as impor­tant in the EHR. The health pro­fes­sional script was more focused in deter­min­ing the level of aware­ness to the national e-Health ini­tia­tives, the nor­mal work­flow used by the pro­fes­sion­als, the dif­fer­ent kinds of records they worked with (elec­tronic, paper or both), the accept­able uses given to the infor­ma­tion and, as in the case of the patient script, their thoughts and opin­ions about the EHR. Both scripts approached some com­mon top­ics such as with whom should this infor­ma­tion be shared and what level of con­trol should patients have. All inter­vie­wee s were wel­comed to dis­cuss top­ics not cov­ered by the script (but con­nected to the inter­view theme) and the inter­viewer guar­an­teed anonymity for all par­tic­i­pants.

The cod­ing process was sup­ported by appro­pri­ate soft­ware, and all the inter­views were directly coded in audio for­mat. On this sub­ject, it is impor­tant to make a note regard­ing the dimen­sions used. Despite the warn­ing made by [McK­night 2005], the integrity aspect of the inter­per­sonal dimen­sion was con­sid­ered when the sub­ject was refer­ring to a com­puter sys­tem. That aspect was used as a mea­sure of the user s expec­ta­tions. This may be eas­ily illus­trated with regards to infor­ma­tion access. There is a sig­nif­i­cant dif­fer­ence between the sys­tem s capac­ity (or com­pe­tence in the ter­mi­nol­ogy used by [McK­night 2001]) to han­dle mul­ti­ple accesses to patient data and the right­eous access to that same data. In the case of this study, sys­tem integrity serves as a medi­a­tor between the expected behav­ior of the sys­tem, and the even­tual real­ity of the same sys­tem. The deci­sion to include the integrity aspect was made dur­ing the analy­sis of the inter­views, when it was noted that the focal point of the sub­jects was not the sys­tem s capac­ity, but their own per­cep­tions of what the sys­tem should or should not per­mit. In line with [McK­night 2005], we did not con­sider that the benev­o­lence aspect would make sense when ana­lyz­ing trust in tech­nol­ogy.

5. Results

A. Demographic Data

1) Patients

Regard­ing access to health ser­vices, the major­ity of inter­viewed patients (78%) use the national health ser­vice. Some patients resort both to pub­lic and pri­vate health ser­vices.

In terms of fre­quency of access to health ser­vices, the aver­age is 3.6 times per year. The lower reg­is­tered fre­quency is once every three years and the max­i­mum reg­is­tered fre­quency is between 24 to 36 times per year (cases of hemodial­y­sis or sim­i­lar sit­u­a­tions).

Regard­ing appoint­ment sched­ul­ing, the pre­ferred method of sched­ul­ing is per­sonal atten­dance (50%), fol­lowed closely by sched­ul­ing via phone call (44%). The less used chan­nels of appoint­ment sched­ul­ing are in loco (sched­ul­ing from one appoint­ment to another 24%) and Inter­net sched­ul­ing (11%). Curi­ously, reg­u­lar access to a com­puter, one of the indi­ca­tors authors expected to be low, is actu­ally quite high (93%), espe­cially con­sid­er­ing the age aver­age of the inter­vie­wees (47 years). Even more curi­ous is the fact that 63% of the patients claim they fell rel­a­tively con­fi­dent in using a com­puter. The low use of the Inter­net for appoint­ment sched­ul­ing may be explained by the gen­eral lack of knowl­edge of that pos­si­bil­ity (only 24% of the sub­jects knew of that pos­si­bil­ity) and for other fac­tors like bad expe­ri­ences in using the e-health plat­form for doing so, or by the fact that the patient does not have a State appointed gen­eral prac­ti­tioner.

Regard­ing the knowl­edge of the leg­is­la­tion applic­a­ble to per­sonal health data, only 7% of the inter­viewed patients are aware of it.

From all the 46 patients, none of them con­sid­ered the RSE as a non-impor­tant e-Health ini­ti­ate and all of them would, if they were given that choice, allow that their health infor­ma­tion to be part of a national EHR.

2) Health Professionals

From the 17 inter­viewed pro­fes­sion­als, 11 work in the pub­lic sec­tor, one in the pri­vate sec­tor and five in both sec­tors. From the 17 pro­fes­sion­als only one (the psy­chol­o­gist) had not worked with elec­tronic health records and did not have a daily access to his patients’ health records. From the remain­ing 16, all of them have a daily access to elec­tronic health records, and 13 of those also access paper based health records. The aver­age years of expe­ri­ence of the health pro­fes­sion­als inter­viewed was 12 years, and the age aver­age was 37 years.

The less-known e-Health ini­tia­tive is the RSE, fol­lowed by the eAgenda ini­tia­tive (the pos­si­bil­ity of patients sched­ul­ing appoint­ments with the State appointed gen­eral prac­ti­tioner).

The applic­a­ble leg­is­la­tion to health data is also con­sid­er­ably unknown, 15 of the 17 pro­fes­sion­als did not know the actual leg­is­la­tion.

Like the patients, all health pro­fes­sion­als con­sid­ered e-Health ini­tia­tives very impor­tant and all of them would make use of the RSE if it was imple­mented, some of them not­ing how­ever that the def­i­n­i­tion and select ion of the appro­pri­ate tech­nol­ogy and tools is not their con­cern, only the cor­rect usage of those tools.

B. Trust Factors

1) Patients

Regard­ing patients, a total of 152 trust fac­tors and 25 dis­trust fac­tors were found. Dis­trust fac­tors are con­cep­tu­ally equal to trust fac­tors (the only dif­fer­ence is the man­ner in which they are pre­sented), and there­fore were ana­lyzed in the same way trust fac­tors were. From these fac­tors, 142 and 7 fac­tors of trust and dis­trust respec­tively, are related with the use of infor­ma­tion sys­tems. Addi­tion­ally, a total of 138 opin­ions about infor­ma­tion shar­ing were reg­is­tered and 56 opin­ions about patient access and con­trol of their own health records. The rel­a­tive fre­quency of the iden­ti­fied trust fac­tors can be seen in Table 1 in the fol­low­ing tables all fig­ures were rounded to the units).

Table 1. Rel­a­tive Fre­quency of Trust Fac­tors (Patients)

Based on that Table, it is pos­si­ble to observe that the most val­ued fac­tors for the patients are insti­tu­tional fac­tors, fol­lowed by the inter­per­sonal fac­tors if taken as an aggre­gate, rep­re­sent­ing 35%. The dis­po­si­tional fac­tors also form a sig­nif­i­cant share of all the fac­tors iden­ti­fied. These, how­ever, are very dif­fi­cult to fur­ther sub­di­vide and cat­e­go­rize. They are a sort of “default” stance, where there is no clear rea­son or jus­ti­fi­ca­tion for trust, apart from the indi­vid­ual posi­tion assumed by the trustor in any given rela­tion­ship.

The insti­tu­tional fac­tors were sub­di­vided in the cat­e­gories included in Table 2.

Table 2. Rel­a­tive Fre­quency of Insti­tu­tional Fac­tors (Patients)

These val­ues indi­cate that the most import fac­tor for patients is the com­pli­ance with leg­is­la­tion, with their com­mon val­ues (as a soci­ety) and with the national health ser­vice. It is pos­si­ble to observe that these indi­ca­tors have some­thing in com­mon with one another. The exis­tence of a national health ser­vice is some­thing that most Por­tuguese view as an essen­tial com­mon value, and the exis­tence of a national health ser­vice is a right advo­cated in the Por­tuguese Con­sti­tu­tion. Besides these three fac­tors, the remain­ing ones are divided between trust in the med­ical pro­fes­sion, and the accep­tance of the cur­rent trend of mod­ern­iz­ing exist­ing struc­tures through infor­ma­tion tech­nol­ogy (sit­u­a­tion nor­mal­ity).

For the patients, the most sig­nif­i­cant inter per­sonal fac­tors are those related to integrity and to pre­dictabil­ity. In the first group, sys­tem integrity rep­re­sents 43%, infor­ma­tion integrity 30% and sys­tem access integrity 26%. Regard­ing pre­dictabil­ity, the most rel­e­vant fac­tor is access to infor­ma­tion (59%).

Regard­ing infor­ma­tion shar­ing, 75% of patients’ opin­ions are towards full shar­ing of health infor­ma­tion between pro­fes­sion­als. How­ever, this does n o t mean that the patient agrees to give uncon­di­tional access to the pro­fes­sion­als. It means that if the pro­fes­sional requires that access, the access should be to the total­ity of the record. The remain­ing 25% main­tain that access to some of the infor­ma­tion should be restricted. Addi­tion­ally, 76% of opin­ions indi­cate that infor­ma­tion shar­ing with the pri­vate sec­tor should be restricted or non-exis­tent (52% inci­dence on restricted).

On the topic of access and con­trol over one’s EHR, 70% of opin­ions indi­cate that patients would like some form of con­trol over their own EHR. The remain­ing 30% do not wish to have any kind of con­trol over their own EHR.

2) Health Professionals

From the health pro­fes­sion­als’ inter­views, a total of 65 trust and 11 dis­trust fac­tors were gath­ered. From these fac­tors, only two dis­trust fac­tors are not related to the use of EHR. A total of 41 and 27 opin­ions about infor­ma­tion shar­ing and access or con­trol by the patient of his own EHR were iso­lated, respec­tively. The rel­a­tive fre­quency of trust fac­tors is dis­played in Table 3.

Table 3. Rel­a­tive Fre­quency of Turst Fac­tors (Pro­fes­sion­als)

In the case of the health pro­fes­sion­als, there is a sig­nif­i­cant shift towards the char­ac­ter­is­tics of the infor­ma­tion sys­tem, with the inter­per­sonal fac­tors reach­ing 69%. Among these, integrity related fac­tors are the most rel­e­vant, dis­trib­uted into the fol­low­ing three issues: infor­ma­tion integrity (71%), sys­tem integrity (17%), and sys­tem access integrity (12%).

Still, the insti­tu­tional fac­tors are very sig­nif­i­cant. This group of fac­tors can be fur­ther sub­di­vided in the cat­e­gories depicted in Table 4.

Table 4. Rel­a­tive Fre­quency of Insti­tu­tional Fac­tors (Pro­fes­sion­als)

Com­pli­ance with leg­is­la­tion is a sig­nif­i­cantly rel­e­vant fac­tor. In the case of the health pro­fes­sion­als, the National Health Ser­vice fac­tor also relates with the com­pli­ance of the RSE with the prin­ci­ples of pub­lic health. It is impor­tant to note that, in both cases (patients and health pro­fes­sion­als), a rel­e­vant issue included in this indi­ca­tor is the fact that the RSE is a pub­lic ini­tia­tive.

Regard­ing infor­ma­tion shar­ing, 62% of the health pro­fes­sion­als’ opin­ions indi­cate that infor­ma­tion shar­ing should have some restric­tions, namely that a pro­fes­sional should not have access to the total­ity of the record. The topic of pub­lic pri­vate shar­ing between health pro­fes­sion­als is a sen­si­ble one, with 42 % indi­cat­ing that there should n o t be any sig­nif­i­cant dif­fer­ence between both sec­tors, and 42 % indi­cat­ing that there should be restric­tions to pri­vate pub­lic shar­ing. 17% of opin­ions also indi­cate that there should n o t be any shar­ing of any kind with the pri­vate sec­tor (i.e. direct shar­ing of the EHR).

Finally, on the topic of patient access to and con­trol of their own health record, 93% of col­lected opin­ions show that pro­fes­sion­als con­sider patients should have access and/or con­trol over their own EHR. Most of them note how­ever, that cau­tion must be taken to pre­vent sit­u­a­tions where the patient, while exer­cis­ing this right, may inad­ver­tently con­trib­ute neg­a­tively to his own health sta­tus.

6. Discussion

The results of the study sug­gest that the two groups inter­viewed dif­fer in terms of the trust fac­tors they most value. For patients, the most rel­e­vant fac­tors are those of insti­tu­tional nature. For the health pro­fes­sion­als, the inter­per­sonal dimen­sion col­lects the issues most fre­quently reported. These dif­fer­ences may be explained by con­sid­er­ing the dis­tinct con­cerns of the two groups of EHR stake­hold­ers. Health pro­fes­sion­als put greater empha­sis on the EHR sys­tems’ fea­tures, high­light­ing the EHR as a piv­otal instru­ment for their job. With respect to patients, they focus the pro­vi­sion of health care, valu­ing the align­ment of the EHR with the national health ser­vice by show­ing com­pli­ance with leg­is­la­tion and con­gru­ence with com­mon val­ues.

The integrity related fac­tors also hold a strong posi­tion, with a big­ger empha­sis in the health pro­fes­sion­als group. How­ever, these fac­tors are also tightly related to what both groups con­sider as being just, fair and ade­quate. Actu­ally, it may be impos­si­ble to sep­a­rate these fac­tors from the Por­tuguese iden­tity and com­mon val­ues. In any case, that same rela­tion­ship should be valid across dif­fer­ent health sys­tems and dif­fer­ent national iden­ti­ties, only the core val­ues in ques­tion could be dif­fer­ent. This study shows that, in the par­tic­u­lar case of Por­tu­gal, the RSE should not be con­trary to what the sys­tem is try­ing to enhance and improve. In short, the RSE should not con­demn the required con­di­tions for its own thriv­ing.

The method­ol­ogy used in this work also helps on the effec­tive pri­or­i­ti­za­tion of require­ments for the EHR, enabling an effi­cient match between users’ expec­ta­tion s and imple­men­ta­tion pri­or­i­ties. In order to pri­or­i­tize require­ments for the EHR, the find­ings can be matched with stan­dards such as [ISO/TC215 2004].

7. Conclusion

Assum­ing trust is a rel­e­vant fac­tor for tech­no­log­i­cal adop­tion and sys­tem use, this paper pre­sented a study that aimed to find out what the users value most con­cern­ing trust in EHR. The imple­men­ta­tion of a national EHR is a com­plex and del­i­cate process, with a low mar­gin for error. For that rea­son alone, under­stand­ing what makes patients and health pro­fes­sion­als trust an EHR is impor­tant. The value of an EHR is directly linked with its use. If the EHR does not com­ply with its main users expec­ta­tions, it will most likely be mis­used or under used.

The trust fac­tors pre­sented in this work do not rep­re­sent the total­ity of con­cerns about e Health ini­tia­tives. Nev­er­the­less, given the rela­tion­ship between use of a sys­tem and trust in that sys­tem, they form a sig­nif­i­cant group of aspects to con­sider that may con­trib­ute strongly to the adop­tion and use of e-Health sys­tems.

Regard­ing the lim­i­ta­tions of this study, it is impor­tant to note that its find­ings are par­tially lim­ited to the Por­tuguese cul­ture. The Por­tuguese peo­ple are, in gen­eral, strong advo­cates of a national health sys­tem, which is not nec­es­sar­ily true in other coun­tries. Nev­er­the­less, only the core social val­ues should dif­fer, the method­ol­ogy should remain valid. Addi­tion ally, as it was noted before (cf. sec­tion 4 Study Descrip­tion), we do not claim to have used prob­a­bil­ity based sam­pling meth­ods to form our pool of par­tic­i­pants.

Three avenues for fur­ther research are sug­gested. The first involves extend­ing the study to col­lect the views from health sys­tem man­agers. This could prompt a dis­cus­sion between an even­tual util­i­tar­ian view of the EHR and the ben­e­fits and risks of using the EHR as per­ceived by health pro­fes­sional s and patients. The sec­ond sug­ges­tion for future work is related to the repli­ca­tion of this study in a post imple­men­ta­tion sce­nario of the RSE. This would allow a com­par­i­son between ex ante trust fac­tors and ex post trust fac­tors, enabling to con­trast the intended use of the sys­tem with the actual use of the sys­tem. The third sug­ges­tion regards the shar­ing of health infor­ma­tion between the pub­lic and pri­vate sec­tor. Besides iden­ti­fy­ing the infor­ma­tion that should or could be trans­fer red between the two sec­tors, it would be inter­est­ing to reflect on the tech­ni­cal and orga­ni­za­tional secu­rity chal­lenges and solu­tions involved in that shar­ing process.

References

• Gefen, D., E. Kara­hanna and D. W. Straub (2003). Trust and TAM in Online Shop­ping: An Inte­grated Model. MIS Quar­terly 27(1), 51–90.

  Hol­stein, J. A. and J. F. Gubrium (1995). The Active Inter­view. Thou­sand Oaks: Sage.

  ISO/TC215 (2004). ISO/TS 18308 Health infor­mat­ics—Require­ments
  for an elec­tronic health record archi­tec­ture. Inter­na­tional
  Orga­ni­za­tion for Stan­dard­iza­tion.

  ISO/TC215 (2005). ISO/TR 20514 Health infor­mat­ics—Elec­tronic Health Record Def­i­n­i­tion, Scope and Con­text. Inter­na­tional Orga­ni­za­tion for Stan­dard­iza­tion.

  Kas­sim, E., N. Zamzuri, S. F. A. Kader and H. Han­i­ta­haiza (2010). Inves­ti­gat­ing infor­ma­tion tech­nol­ogy adop­tion suc­cess: the influ­ence of fit and trust. IEEE Inter­na­tional Con­fer­ence on Man­age­ment of Inno­va­tion and Tech­nol­ogy (ICMIT) 2010, 35–39.

  Lewis, J. D. and A. Weigert (1985). Trust as a Social Real­ity. Social Forces 63(4), 967–985.

  Mayer, R. C., J. H. Davis and F. D. Schoor­man (1995). An Inte­gra­tive Model of Orga­ni­za­tional Trust. The Acad­emy of Man­age­ment Review 20(3), 709–734.

  McK­night, D. H. (2005). Trust in infor­ma­tion tech­nol­ogy. Black­well Ency­clo­pe­dia of Man­age­ment 7, 329–331.

  McK­night, D. H. and N. L. Cher­vany (2001). Trust and dis­trust def­i­n­i­tions: One bite at a time. Trust in Cyber-soci­eties, 27–54.

  Myers, M. D. (2012). Qual­i­ta­tive Research in Infor­ma­tion Sys­tems.
  http://www.qual.auck­land.ac.nz/#Qual­i­ta­tive%20Research%20Meth­ods

  Rubin, H. J. and I. Rubin (1997). Qual­i­ta­tive Inter­view­ing: The Art of Hear­ing Data. Thou­sand Oaks: Sage.