Examining the evidence of the impact of health information technology in primary care: An argument...

ARTICLE IN PRESSJB-2898; No. of Pages 8

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s x x x ( 2 0 1 2 ) xxx–xxx

j ourna l homepage: www.i jmi journa l .com

xamining the evidence of the impact of health informationechnology in primary care: An argument for participatoryesearch with health professionals and patients

mmanuelle Bélangera, Gillian Bartletta,∗, Martin Dawesb,haro Rodrígueza, Ifat Hasson-Gidoni c

McGill University, CanadaUniversity of British Columbia, CanadaIDF, Israel

r t i c l e i n f o

rticle history:

eceived 22 July 2011

eceived in revised form

3 May 2012

ccepted 18 July 2012

vailable online xxx

eywords:

nformation technology

rimary care

atient health outcomes

ersonal health records

a b s t r a c t

Purpose: Health information technology represents a promising avenue to improve health

care delivery. How can we use lessons learnt from existing health information technologies

in primary care to inform the optimal design of newer developments such as personal health

records?

Methods: The results of systematic literature reviews about the impact of different infor-

mation systems on health outcomes in primary care are critically discussed in a narrative

synthesis, with a focus on their implications for the development of personal health records.

Results: Given the proliferation of systematic reviews and randomized controlled trials, high

quality evidence for health information technology in primary care is accumulating with

mixed results. The heterogeneity of systems being compared and the quality of research

can no longer account for these findings. One potential explanation may be that systems

originally designed for acute care settings are being implemented in primary care. Early

studies evaluating personal health records suggest that targeting patient outcomes directly

and adapting systems to patients’ needs may be part of the solution.

Conclusion: In order to develop personal health records for primary care, studies are needed

that involve the users, namely patients and primary care health professionals, in the design

and evaluation of these systems from their inception. Participatory research is a recom-

mended methodological approach.

ontents

Please cite this article in press as: E. Bélanger, et al., Examiningogy in primary care: An argument for participatory research with

http://dx.doi.org/10.1016/j.ijmedinf.2012.07.008

1. Introduction – health information technology . . . . . . . . . . . . . . . .2. Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

∗ Corresponding author at: Department of Family Medicine, McGill Univel.: +1 514 398 7375x04587; fax: +1 514 398 4202.

the evidence of the impact of health information technol-health professionals and patients, Int. J. Med. Inform. (2012),

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

ersity, 515 Pine-Avenue West, Montreal, H2W 1S4 Canada.

erved.

ARTICLE IN PRESSIJB-2898; No. of Pages 8

2 i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s x x x ( 2 0 1 2 ) xxx–xxx

3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.1. The poor performance of HIT in primary care as compared to acute care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.2. Strength of explanations for poorer performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.3. The early performance of PHRs in primary care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

4. Conclusion – harnessing the full potential of PHRs in primary care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Authors’ contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

. . . . .

1. Introduction – health informationtechnology

There is no dispute that health information technologies (HIT)represent one of the most promising avenues to modernizeand improve health care delivery [1]. The push for comput-erization has been largely motivated by concerns regardingthe diminishing safety and increased variation of patient care[2–4]. Using HIT to have timely access to complete patienthealth information and to provide evidence-based decisionsupport at the point of care is supposed to reduce medi-cal errors and to enhance clinical decision making [1]. Whilethis is expected to yield clinically significant improvementsin patient health outcomes by reducing adverse events andimproving quality of care, the reported results have not beenstraightforward. For instance, order entry systems have beenshown to reduce rates of medical errors in the hospital set-tings [5], while the effect on safety in primary care has yet tobe demonstrated [6].

A variety of systems have been developed and imple-mented in attempts to improve health care delivery. Thesesystems include simple electronic health or medical records(EHR, EMR) that replace paper records for information storageand retrieval [7], and in the recent literature they are coupledwith more sophisticated computerized physician order entry(CPOE) helping with the management of prescriptions or diag-nostic tests [8]. The latter are often used together with clinicaldecision support systems (CDSS), which provide computer-triggered or on-demand advice according to clinical guidelines[9]. Personal health records (PHR) represent the most recentplatform and allow patients to consult and manage their ownhealth information, and sometimes even to communicateelectronically with their health care providers [10]. With sucha diversity of systems and functions, evaluating the impact ofHIT can be challenging.

Certainly in North America, most HIT systems that havebeen successfully introduced with a demonstrated positiveimpact on patient health outcomes, have occurred in acutecare settings. Only a small proportion of HIT systems havebeing successfully introduced into primary care settings inNorth America in contrast to countries such as the UK,the Netherlands and many European Union countries [11].Acute care is concerned with the management of clinicallyserious acute episodes of care, while primary care gener-

ally manages patients with chronic illness and also providespreventive care [12]. Effective primary care requires access,continuity and comprehensiveness of care that is commu-nity focused [13]. Theoretically, the potential of HIT systems

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

should be even greater in primary care, due to the ability toprovide improvements in all these aspects of care. More con-cretely these factors include the large number of patients withchronic, co-morbid conditions that require therapeutic treat-ment, ongoing monitoring and where evidence and clinicalguidelines are regularly being updated added to the fact that alarge proportion of medications are prescribed by primary carephysicians [12]. However, the evidence supporting improve-ments in health outcomes at the primary care level is neitherstraightforward nor readily convincing [6,14].

With over a decade of research behind us about the imple-mentation of HIT in both acute and primary care settings [15],lessons should be drawn from the continued uncertainty oftheir impact, as we turn to the new, promising HIT currentlybeing developed and implemented with PHR [16]. In otherwords, how can we use lessons learnt from primary care HITto inform the optimal design of PHRs? The purpose of this nar-rative synthesis is to review the evidence of the performanceHIT on patient health outcomes in primary care, and providea coherent case to support the involvement of patients in thedesign of PHRs that directly target patient-level outcomes. Wewill develop our arguments of the potential explanations forthe poor performance and our proposed approach for PHRdevelopment and implementation based on the latest pub-lished evidence. This will be done using the most currentsystematic literature reviews and articles published since thereviews.

2. Methods

An assessment of the findings of overlapping systematicreviews was carried out, with an examination of all the pri-mary and secondary studies that focused on adverse drugevents and patient health outcomes in primary care with adultpopulations. An adverse event was defined as “an unintendedinjury or complication caused by delivery of clinical care ratherthan by the patient’s condition” [17] (p. 1555). In light of therapid development of information systems, the papers thatformed the basis for our narrative synthesis were limited toliterature published between 2005 and 2011. Systematic lit-erature reviews were of primary relevance. We then adaptedour search strategies in Medline and EMBASE to find new lit-erature since the publication of the main systematic reviewsthat might have influenced the state of knowledge.

The results consist of a narrative synthesis criticallydiscussing the evidence and the implications for the devel-opment of PHRs. The themes addressed include the mixedresults of diverse information systems on health outcomes

a l i n

iaqwoemmi

Tagsctrsdsrd

emfutwCtraoittpodatpa

Tebhpo

i n t e r n a t i o n a l j o u r n a l o f m e d i c

n primary care, the explanations that could potentiallyccount for these results in light of an accumulation of highuality evidence, and finally the parallels that can be drawnith the early evaluative literature of PHRs. The conclusionffers recommendations to improve the implementation andvaluation of information technology in the context of pri-ary care specifically, and suggests a research approach thatay improve PHR utilization and thereby increase the positive

mpact on patient health outcomes in primary care.

. Results

.1. The poor performance of HIT in primary care asompared to acute care

here are considerably fewer studies about health outcomesnd patient safety than about provider adherence with clinicaluidelines and satisfaction. The results of a thorough review ofystematic reviews by Black et al. highlights the gap betweenommonly expected benefits of HIT and empirical demonstra-ions [18]. While Buntin et al. found predominantly positiveesults when reviewing evaluations of HIT, studies focusingpecifically on patient health outcomes in primary care tell aifferent story [19]. If we look closely at the care setting, thetudies in the acute care inpatient setting more consistentlyeport significant improvements in reducing rates of adverserug events and selected patient outcomes.

Many overlapping systematic reviews in primary carestablished the lack of clear evidence supporting improve-ents in quality of care or health outcomes. Eslami et al. [6]

ocused on CPOE in primary care and found 4 studies eval-ating their impact on medication safety, with only one ofhese studies showing a reduction in ADEs. Conflicting resultsere also obtained in systematic reviews about the impact ofDSS [14,20,21]. Studying CDSS and patient outcomes in inpa-

ient as opposed to outpatient settings, Sintchenko et al. [15]emarked that only 31% of the studies in the latter group had

positive change after implementation, as opposed to 100%f the inpatient studies. Overall, looking at the original stud-

es reported in systematic reviews and studies published sincehen, it appears that between 2005 and 2011, 5 out of 10 studiesesting various systems have failed to demonstrate improvedatient outcomes in outpatient populations [22–31], while 6ut of 7 studies in the inpatient setting report improvements iniverse areas such as thromboembolism, adverse drug events,nd mortality rates [32–38] (Table 1). In conclusion, the posi-ive impact of HIT on patient safety and health outcomes inrimary care has not clearly been demonstrated as it has incute care settings.

.2. Strength of explanations for poorer performance

he poorer performance of HIT in primary care despite theirven greater potential deserves examination. While there may

e several reasons that might account for these differences, weave focused on the main explanations that have been pro-osed and explore the strength of the rationale behind eachf them in this section.

f o r m a t i c s x x x ( 2 0 1 2 ) xxx–xxx 3

There are several problems affecting the evaluation of HITthat are not specific to primary care. Many authors admit thatwe still have a poor understanding of why certain informationsystems are successfully implemented and yield improve-ments in patient-level outcomes while others do not [18].Electronic health records (EHRs) and similar terms are used todescribe heterogeneous systems that often include inconsis-tent or unspecified applications, such as information storageand retrieval, medication and laboratory order entry, emailfunctions, decision support and reminder alerts. Many ofthese systems are country-specific, homegrown or simplyunspecified in the articles, so that comparison of the researchabout their implementation and impact is inherently problem-atic. As Weir et al. point out, [39] a variable such as CPOE is verycomplex and can vary in terms of functions, decision-supportfeatures, implementation strategies, and customization. Inlight of the poor construct validity of CPOE, no two studiesappear to have implemented the same attributes, even whiletesting the same functions [39]. Two surveys in the US furtherrevealed that over 60 different systems are used by generalpractitioners, most of them being reported by three or fewerrespondents [40,41]. Shekelle and Goldzweig [42] point out thatfew studies exist evaluating commercially developed systems.

Similarly, measures of the actual usage of the systems dur-ing the intervention is often lacking in evaluations despite thefact that it would be important to know whether the resultsare non-significant simply because providers did not use theCDSS. Sintchenko et al. [15] further highlight the difficulty ofproducing clinically relevant changes with interventions thatare only indirectly targeting patient-level outcomes. Improv-ing provider compliance with clinical guidelines may onlyhave diluted effects on the health of patients. However, fromthe evidence reviewed, we cannot conclude that these limita-tions and the difficulty of evaluating complex interventions inhealth organizations [43] are any less problematic for researchthat occurs in acute care settings. The lack of convincingresults can no longer be attributed to the later introduction ofthese systems in the context of primary care as a considerablebody of high quality evidence is accumulating. Given the pro-liferation of systematic reviews and high level of evidence viarandomized control trials (RCTs), the later introduction of thetechnology and the quality of research are no longer plausibleexplanations of the poorer performance of HIT in primary care.

We suggest that the remaining explanation concerns thedifferent nature of primary care delivery and the misfit thatmay occur with information systems that were originallydesigned for acute care settings. Primary care physicians andother health professionals have been documented as notingthe poor fit between the cognitive processes they go throughin their clinical work and the implemented HIT [44]. Bateset al. [45] concur that primary care providers often expect HITto interfere with their workflow. In other words, the imple-mentation of systems that were designed for acute care maypresent one of the most compelling reasons for their disap-pointing performance, as these systems may simply not betailored appropriately for improving downstream health out-

comes and safety in primary care. For instance, HIT as weknow it may be effective in reducing the risks of medica-tion errors engendered by the cross-coverage of many healthprofessionals in the event of intensive periods of care [1],

4 i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s x x x ( 2 0 1 2 ) xxx–xxx

Table 1 – Studies of HIT measuring patient-level health outcomes 2005–2011.

Outpatient Design Outcomes

CPOE and/or CDSSBassa et al. (2005) Prospective pre-post design Cholesterola

Cleveringa et al. (2008) RCT Glycemia, blood pressurea, cholesterola

Feldstein et al. (2006) RCT Osteoporosisa

Gandhi et al. (2005) Prospective cohort study Adverse drug eventsHicks et al. (2008) RCT Blood pressureHolbrook et al. (2011) RCT Vascular eventsO’Connor et al. (2005) Prospective pre-post design Glycemia, cholesterolO’Connor et al. (2011) RCT Glycemiaa, blood pressurea, cholesterolSamal et al. (2011) Cross-sectional survey Blood pressurea

Steele et al. (2005) Prospective pre-post design Adverse drug-events

PHRFonda et al. (2009) RCT Diabetes distressa

Grant et al. (2008) RCT Glycemia, blood pressure, cholesterolHolbrook et al. (2009) RCT Clinical composite score (including glycemia)a

McMahon et al. (2005) RCT Glycemiaa, blood pressurea, cholesterola

Quinn et al. (2011) RCT Glycemiaa, blood pressure, cholesterol, depressionRalston et al. (2009) RCT Glycemiaa, blood pressure, cholesterol

Inpatient Design Outcomes

CPOE and/or CDSSAustrian et al. (2011) Retrospective cross-sectional study Heparin-induced thrombocytopenia antibody+, mortalityBertsche et al. (2010) Prospective cohort study Adverse drug-eventsa

Colpaert et al. (2006) Controlled cross-sectional trial Adverse drug-eventsa

Guerra et al. (2010) Retrospective cross-sectional study Glycemiaa

Kucher et al. (2005) RCT Venous thromboembolisma

Lester et al. (2006) RCT Cholesterola

Mann et al. (2011) RCT Glycemiaa

a Significant improvement.

which may be less relevant in primary care. In fact, while stud-ies in the inpatient settings are more likely to address medica-tion errors, adverse-drugs events, secondary prevention, andmortality rates, in primary care the health outcomes stud-ied concern chronic diseases, such as diabetes, hypertension,elevated cholesterol, and osteoporosis. As we will explore, tar-geting patient-level outcomes directly in primary care withsystems that are specifically designed to suit the needs ofpatients and primary care providers holds the most promise.

In sum, the challenges of comparing different systems andof coming up with robust research designs are equally impor-tant in the acute as in the primary care settings. What differsis the added difficulty of adequately implementing systemsthat were designed for intense episodes of care. In primarycare, the variety of conditions, the volume of patients, diseasepresentation and the length of follow-up all differ from acutecare [12]. In this context, part of the solution may be to designsystems that target health outcomes and quality of care issuesthat are specific to primary care. As Buntin et al. concluded,the human element and staff “buy in” are crucial to the suc-cess of HIT [19], and offering personalized HIT systems wouldcertainly contribute to securing the interest of stakeholders. Ifwe are to demystify what exactly is failing in the mechanism ofimproving patient safety and patient health outcomes, evalu-ative research has to clarify the different steps that would lead

from the introduction of a given system to tangible improve-ments in a specific patient-level outcome. There is no denyingthat health indicators are influenced by a wide range of factorsbeyond primary care delivery, and this is why incorporating

different sources of information that may come from a newHIT such as personal health records (PHRs) is so promising forprimary care.

3.3. The early performance of PHRs in primary care

The personal health record (PHR) is a relatively new HIT thatis generally defined as some type of electronic platform thatallows patients to manage their health information. This mayinclude access to EHRs and to various other functions, suchas electronic communication with health care providers [46].With the disappointing performance of other HIT in primarycare and growing dissatisfaction with the health care systemon the part of patients, PHRs are now expected to yield sig-nificant outcomes in terms of quality of care, patient healthoutcomes, and most importantly, they are proposed as onemeans of involving patients in the therapeutic relationship[10,47]. Active involvement of patients in the patient–providerrelationship, also called shared decision-making [48], is beingincreasingly advocated in preference-sensitive care [49] and inpatient-centered care [50]. PHRs are unique in that they targetpatient outcomes more directly than previous HIT by involvingpatients in checking the accuracy of their information and inself-management plans. Even before the use of electronic plat-

forms, patient involvement in self-management has yieldedimpressive results in terms of control of chronic illnesses[51,52]. Patients with chronic conditions are likely to benefiteven more from having access to their personal health records,

a l i n

omrc[ybt[tb0lpaam

Arvdbmtdoitah

iumcismutbthb[sppmBod

i n t e r n a t i o n a l j o u r n a l o f m e d i c

aking it easier to track their health, their treatment plansnd appointments.

In terms of fulfilling its promises, early evaluations of PHRsn patient outcomes and quality of care are encouraging albeitodest [52–54]. Looking at original studies, we found 5 RCTs

eporting small but significant improvements in some clini-al outcomes, especially glycemic control in diabetic patients55–59]. The study by Grant et al. is the only one that did notield significant results: for diabetic patients using the PHR,aseline glycemic control did not change significantly in eitherhe intervention or control groups over the duration of the trial60]. Statistical significance is not a panacea, however, andhe changes reported are often small, with the interventiony Holbrook et al. decreasing glycated hemoglobin by a mere.2% [56]. Archer et al. conclude that the results on patient-evel outcomes remain disappointing, and point to a lack ofatient-oriented functionalities as one explanation. System-tic reviews also warn that interventions tend to be complexnd multi-faceted, so that the direct impact of patient involve-ent is not always straightforward [53].

. Conclusion – harnessing the fullotential of PHRs in primary care

s previously outlined, there is a need to develop optimalesearch designs to study health impacts with better internalalidity and better construct validity by providing thoroughescriptions of the systems being studied [39]. While these wille important for the appropriate evaluation of PHRs, the argu-ents we have presented indicate that if PHRs are to improve

he health of patients in general, then systems need to beeveloped for and targeted to patients specifically [52,61]. Inrder to be certain this is achieved, as supported by a grow-

ng body of literature on knowledge translation, we proposehat the involvement of the users of PHR, namely the patients,re essential for the development of platforms that will helparness the full potential of PHRs in primary care.

The current movement toward patient involvementn decision-making and self-management represents annprecedented avenue to target patient health outcomesore directly. It is fair to assume that if patients and primary

are providers had been systematically involved in designingnformation systems tailored to their needs, there would becientific literature documenting this research and develop-ent process. To our knowledge, only one study has reported

sing patients’ feedback in PHR development and found thathey can make significant contributions to their developmenty highlighting different elements of the systems [62]. Givenhe scarcity of evidence, it appears that most HIT systemsave not only been designed for acute care, but they have alsoeen designed for the use of providers rather than patients

52]. It is reasonable to expect that replicating these earlierystems will produce PHRs that are poorly suited to involvingatients in primary care delivery. Patient participation in therovision of accurate and relevant information is one of the

issing links to improve health outcomes in primary care.ut more than being only involved in checking the accuracyf information in their PHRs, patients should participate inesigning systems that they will be willing and able to use.

Early research about poor uptake of PHRs and about securityconcerns demonstrates the need for such involvement. PHRshave the potential to improve patient–provider relations andto promote shared decision-making by taking the patient’sperspective into account [10]. Chronic care would greatly ben-efit from information about the patients’ lifestyle and beliefsabout the condition, as this could contribute to the devel-opment of feasible treatment plans that can be realisticallyadhered to by patients. Involving a diverse group of patientsmay also help offset the divide that has been observed inadoption rates of PHR in ethnic minority groups [63]. Thiswould foster the development of systems that make room fora variety of patients’ perspectives, and thus also contributeto promoting shared decision-making and self-managementof long-term illness.

An efficient way of developing user-oriented PHRs wouldbe to carry out a study that involves patients in the designand use of these systems from their inception, in order tomake sure that they meet their expectations in terms of func-tions and security. Participatory research is the process ofproducing new knowledge by “systematic inquiry, with thecollaboration of those affected by the issue being studied, forthe purposes of education and taking action or effecting socialchange” [64]. Such a study could involve patients in everystep of the research process, to develop systems in consulta-tions with patients, and to come to an agreement about whatpatients want to achieve with an access to their health infor-mation. When stakeholders collaborate in research as equals,the approach has been found to maximize community and layinvolvement [65]. Asking users what they want may representan excellent strategy to improve the uptake, enrich innova-tions and increase the impact of diverse HIT in primary care. Inlight of the evidence reviewed, unless both patients and healthcare providers are involved in the design and use of new gen-erations of PHRs specifically for primary care, the problems ofpersistent uncertainty over their impact on health outcomeswill recur, and the tremendous potential of patient participa-tion in chronic disease management at the primary care levelmay never materialize.

Authors’ contributions

EB conducted the literature search and drafted the articleunder the supervision of GB. All authors reviewed the papermultiple times during the writing process and contributed sig-nificant ideas to the overall argument.

Conflict of interest

The authors declare that they do not have any conflict of inter-est.

Acknowledgments

This paper was presented as a poster for the student session of

research-in-progress at the 2010 North American Primary CareResearch Group Annual Meeting, November 13, 2010, Seattle,Washington (Ambiguous Impact of Health Information Tech-nology on Patient Outcomes in Primary Care: Learning Lessons

6 i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i

Summary points

What is already known about the topic?

• Health information technology has great potential toimprove patient health outcomes in primary care.

• High quality evidence is accumulating that healthinformation technology is not having a clear impact onpatient health outcomes in primary care in contrast tothe consistent benefits seen in acute care.

What the study adds?

• The lack of fit between health information technol-ogy and the specific needs of patients and health careproviders in primary care represent the most plausibleexplanation for the lack of impact on health outcomes.

• Both patients and providers should be involved in thedesign and use of new generations of PHRs if we are toharness their potential in primary care.

in the Age of Personal Health Record. Emmanuelle Belanger,MSc, McGill University; Gillian Bartlett, PhD; Charo Rodriguez,MD, PhD; Martin Dawes, MD).

e f e r e n c e s

[1] D.W. Bates, A.A. Gawande, Improving safety withinformation technology, N. Engl. J. Med. 348 (June (25)) (2003)2526–2534.

[2] R. Romanow, Building on Values: The Future of Health Carein Canada – Final Report, Commission on the Future ofHealth Care in Canada, Saskatoon, 2002.http://dsp-psd.pwgsc.gc.ca/Collection/CP32-85-2002E.pdf(accessed October 2010).

[3] Institute of Medicine, Crossing the Quality Chasm: A NewHealth System for the 21st Century, Committee on QualityHealth Care in America, 2001.

[4] J. Car, A. Black, C. Anandan, K. Cresswell, The Impact ofeHealth on the Quality & Safety of Healthcare. A SystematicOverview & Synthesis of the Literature. Report for the NHSConnecting for Health [Internet]. 2009. Available from:http://www.haps.bham.ac.uk/publichealth/cfhep/documents/NHS CFHEP 001 Final Report.pdf.

[5] S. Eslami, N.F. Keizer de, A. Abu-Hanna, The impact ofcomputerized physician medication order entry inhospitalized patients—a systematic review, Int. J. Med.Inform. 77 (June (6)) (2008) 365–376.

[6] S. Eslami, A. Abu-Hanna, N.F. de Keizer, Evaluation ofoutpatient computerized physician medication order entrysystems: a systematic review, J. Am. Med. Inform. Assoc. 14(April (4)) (2007) 400–406.

[7] K. Häyrinen, K. Saranto, P. Nykänen, Definition, structure,content, use and impacts of electronic health records: areview of the research literature, Int. J. Med. Inform. 77 (May(5)) (2008) 291–304.

[8] E. Ammenwerth, P. Schnell-Inderst, C. Machan, U. Siebert,

The effect of electronic prescribing on medication errorsand adverse drug events: a systematic review, J. Am. Med.Inform. Assoc. 15 (June (5)) (2008) 585–600.

n f o r m a t i c s x x x ( 2 0 1 2 ) xxx–xxx

[9] R. Kaushal, K. Shojania, D. Bates, Effects of computerizedphysician order entry and clinical decision support systemson medication safety—a systematic review, Arch. Intern.Med. 163 (12) (2003) 1409–1416.

[10] D.C. Kaelber, A.K. Jha, D. Johnston, B. Middleton, D.W. Bates,A research agenda for personal health records (PHRs), J. Am.Med. Inform. Assoc. 15 (August (6)) (2008) 729–736.

[11] Harris Interactive [European Union EuroBarometer June, July2001 and Harris Interactive Surveys for USA in June 2001 andJanuary/February 2001]. http://www.harrisinteractive.com/(accessed March 2010).

[12] Institute of Medicine, in: M.S. Donaldson, K.D. Yordy, K.N.Lohr, N.A. Vanselow (Eds.), Primary Care: America’s Health ina New Era, Committee on the Future of Primary Care, 1996.

[13] B. Starfield, L. Shi, J. Macinko, Contribution of primary care tohealth systems and health, Milbank Q. 83 (3) (2005) 457–502.

[14] C. Bryan, S.A. Boren, The use and effectiveness of electronicclinical decision support tools in the ambulatory/primarycare setting: a systematic review of the literature, Inform.Prim. Care 16 (2) (2008) 79–91.

[15] V. Sintchenko, F. Magrabi, S. Tipper, Are we measuring theright end-points? Variables that affect the impact ofcomputerised decision support on patient outcomes: asystematic review, Med. Inform. Internet 32 (3) (2007)225–240.

[16] T. Greenhalgh, H.W.W. Potts, G. Wong, P. Bark, D.Swinglehurst, Tensions and paradoxes in electronic patientrecord research: a systematic literature review using themeta-narrative method, Milbank Q. 87 (4) (2009) 729–788.

[17] G. Bartlett, R. Blais, R. Tamblyn, R.J. Clermont, B. MacGibbon,Impact of patient communication problems on the risk ofpreventable adverse events in acute care settings, CMAJ 178(June (12)) (2008) 1555–1562.

[18] A.D. Black, J. Car, C. Pagliari, C. Anandan, K. Cresswell, T.Bokun, B. McKinstry, R. Procter, A. Majeed, A. Sheikh, Theimpact of ehealth on the quality and safety of health care: asystematic overview, PLoS Med. 8 (January (1)) (2011)e1000387.

[19] M.B. Buntin, M.F. Burke, M.C. Hoaglin, D. Blumenthal, Thebenefits of health information technology: a review of therecent literature shows predominantly positive results,Health Aff. 30 (March (3)) (2011) 464–471.

[20] J.I. Wolfstadt, J.H. Gurwitz, T.S. Field, M. Lee, S. Kalkar, W.Wu, P.A. Rochon, The effect of computerized physician orderentry with clinical decision support on the rates of adversedrug events: a systematic review, J. Gen. Intern. Med. 23(March (4)) (2008) 451–458.

[21] B. Mollon, J.J. Chong, A.M. Holbrook, M. Sung, L. Thabane, G.Foster, Features predicting the success of computerizeddecision support for prescribing: a systematic review ofrandomized controlled trials, BMC Med. Inform. Decis. Mak.9 (1) (2009) 11.

[22] A. Bassa, M. del Val, A. Cobos, E. Torremade, S. Bergonon, C.Crespo, M. Brosa, S. Munio, C. Espinosa, Impact of a clinicaldecision support system on the management of patientswith hypercholesterolemia in the primary healthcaresetting, Dis. Manag. Health Out. 13 (1) (2005) 65–72.

[23] F.G.W. Cleveringa, K.J. Gorter, M. van den Donk, G.E.H.M.Rutten, Combined task delegation, computerized decisionsupport, and feedback improve cardiovascular risk for type 2diabetic patients: a cluster randomized trial in primary care,Diabetes Care 31 (December (12)) (2008)2273–2275.

[24] A. Feldstein, P.J. Elmer, D.H. Smith, M. Herson, E. Orwoll, C.Chen, M. Aickin, M.C. Swain, Electronic medical record

reminder improves osteoporosis management after afracture: a randomized, controlled trial, J. Am. Geriatr. Soc.54 (March (3)) (2006) 450–457.

a l i n

[25] T.K. Gandhi, S.N. Weingart, A.C. Seger, J. Borus, E. Burdick,E.G. Poon, L.L. Leape, D.W. Bates, Outpatient prescribingerrors and the impact of computerized prescribing, J. Gen.Intern. Med. 20 (September (9)) (2005) 837–841.

[26] L.S. Hicks, T.D. Sequist, J.Z. Ayanian, S. Shaykevich, D.G.Fairchild, E.J. Orav, D.W. Bates, Impact of computerizeddecision support on blood pressure management andcontrol: a randomized controlled trial, J. Gen. Intern. Med. 23(March (4)) (2008) 429–441.

[27] A. Holbrook, E. Pullenayegum, L. Thabane, S. Troyan, G.Foster, K. Keshavjee, D. Chan, L. Dolovich, H. Gerstein, C.Demers, G. Curnew, Shared electronic vascular risk decisionsupport in primary care: computerization of medicalpractices for the enhancement of therapeutic effectiveness(COMPETE III) randomized trial, Arch. Intern. Med. 171(October (19)) (2011) 1736–1744.

[28] P.J. O’Connor, Impact of an electronic medical record ondiabetes quality of care, Ann. Fam. Med. 3 (July (4)) (2005)300–306.

[29] P.J. O’Connor, J.M. Sperl-Hillen, W.A. Rush, P.E. Johnson, G.H.Amundson, S.E. Asche, H.L. Ekstrom, T.P. Gilmer, Impact ofelectronic health record clinical decision support on diabetescare: a randomized trial, Ann. Fam. Med. 9 (1) (2011) 12–21.

[30] L. Samal, J.A. Linder, S.R. Lipsitz, L.S. Hicks, Electronic healthrecords, clinical decision support, and blood pressurecontrol, Am. J. Manag. Care 17 (9) (2011) 626–632.

[31] A.W. Steele, S. Eisert, J. Witter, P. Lyons, M.A. Jones, P. Gabow,E. Ortiz, The effect of automated alerts on provider orderingbehavior in an outpatient setting, PLoS Med. 2 (9) (2005)e255.

[32] J.S. Austrian, J.S. Adelman, S.H. Reissman, H.W. Cohen, H.H.Billett, The impact of the heparin-inducedthrombocytopenia (HIT) computerized alert on providerbehaviors and patient outcomes, J. Am. Med. Inform. Assoc.18 (October (6)) (2011) 783–788.

[33] T. Bertsche, J. Pfaff, P. Schiller, J. Kaltschmidt, M.G. Pruszydlo,W. Stremmel, I. Walter-Sack, W.E. Haefeli, J. Encke,Prevention of adverse drug reactions in intensive carepatients by personal intervention based on an electronicclinical decision support system, Intensive Care Med. 36(February (4)) (2010) 665–672.

[34] K. Colpaert, B. Claus, A. Somers, K. Vandewoude, H. Robays,J. Decruyenaere, Impact of computerized physician orderentry on medication prescription errors in the intensive careunit: a controlled cross-sectional trial, Crit. Care 10(February (1)) (2006) R21.

[35] Y.S. Guerra, K. Das, P. Antonopoulos, S. Borkowsky, L.Fogelfeld, M.J. Gordon, B.M. Palal, J.C. Witsil, E.A. Lacuesta,Computerized physician order entry-based hyperglycemiainpatient protocol and glycemic outcomes: the CPOE-HIPstudy, Endocr. Pract. 16 (April (3)) (2010) 389–397.

[36] N. Kucher, S. Koo, R. Quiroz, J.M. Cooper, M.D. Paterno, B.Soukonnikov, S.Z. Goldhaber, Electronic alerts to preventvenous thromboembolism among hospitalized patients, N.Engl. J. Med. 352 (March (10)) (2005) 969–977.

[37] W.T. Lester, R.W. Grant, G. Octo Barnett, H.C. Chueh,Randomized controlled trial of an informatics-basedintervention to increase statin prescription for secondaryprevention of coronary disease, J. Gen. Intern. Med. 21(January (1)) (2006) 22–29.

[38] E.A. Mann, J.A. Jones, S.E. Wolf, C.E. Wade, Computerdecision support software safely improves glycemic controlin the burn intensive care unit: a randomized controlledclinical study, J. Burn Care Res. 32 (February (2)) (2011)246–255.

[39] C.R. Weir, N. Staggers, S. Phansalkar, The state of theevidence for computerized provider order entry: asystematic review and analysis of the quality of the

literature, Int. J. Med. Inform. 78 (June (6)) (2009)365–374.

[40] R.L. Edsall, K.G. Adler, An EHR user-satisfaction survey:advice from 408 family physicians, Fam. Pract. Manag. 12(October (9)) (2005) 29–35.

[41] R.L. Edsall, K.G. Adler, User satisfaction with EHRs: report ofa survey of 422 family physicians, Fam. Pract. Manag. 15(February (2)) (2008) 25–32.

[42] P. Shekelle, C. Goldzweig, Costs and Benefits of HealthInformation Technology: An Updated Systematic Review,The Health Foundation, 2009.

[43] E. Murray, S. Treweek, C. Pope, A. MacFarlane, L. Ballini, C.Dowrick, T. Finch, A. Kennedy, F. Mair, C. O’Donnell, B.N. Ong,T. Rapley, A. Rogers, C. May, Normalisation process theory: aframework for developing, evaluating and implementingcomplex interventions, BMC Med. 8 (October (1)) (2010) 63.

[44] L. Poissant, J. Pereira, R. Tamblyn, Y. Kawasumi, The impactof electronic health records on time efficiency of physiciansand nurses: a systematic review, J. Am. Med. Inform. Assoc.12 (5) (2005) 505–516.

[45] D. Bates, M. Ebell, E. Gotlieb, J. Zapp, H. Mullins, A proposalfor electronic medical records in US primary care, J. Am.Med. Inform. Assoc. 10 (1) (2003) 1–10.

[46] P.C. Tang, J.S. Ash, D.W. Bates, J.M. Overhage, D.Z. Sands,Personal health records: definitions, benefits, and strategiesfor overcoming barriers to adoption, J. Am. Med. Inform.Assoc. 13 (March (2)) (2006) 121–126.

[47] C. Pagliari, D. Detmer, P. Singleton, Potential of electronicpersonal health records, Brit. Med. J. 335 (August (7615))(2007) 330–333.

[48] C. Charles, A. Gafni, T. Whelan, Shared decision-making inthe medical encounter: what does it mean? (or it takes atleast two to tango), Soc. Sci. Med. 44 (March (5)) (1997)681–692.

[49] A. Edwards, G. Elwyn, Shared Decision-making in HealthCare, Oxford University Press, USA, 2009.

[50] S.R. Reti, H.J. Feldman, S.E. Ross, C. Safran, Improvingpersonal health records for patient-centered care, J. Am.Med. Inform. Assoc. 17 (2) (2010) 192–195.

[51] J.E. Scherger, Future Vision:, Is family medicine ready forpatient-directed care? Fam. Med. 41 (4) (2009) 285–288.

[52] N. Archer, U. Fevrier-Thomas, C. Lokker, K.A. McKibbon, S.E.Straus, Personal health records: a scoping review, J. Am.Med. Inform. Assoc. 18 (June (4)) (2011) 515–522.

[53] M. Tenforde, A. Jain, J. Hickner, The value of personal healthrecords for chronic disease management: what do we know?Fam. Med. 43 (May (5)) (2011) 351–354.

[54] C.Y. Osborn, L.S. Mayberry, S.A. Mulvaney, R. Hess, Patientweb portals to improve diabetes outcomes: a systematicreview, Curr. Diab. Rep. 10 (October (6)) (2010) 422–435.

[55] S.J. Fonda, G.T. McMahon, H.E. Gomes, S. Hickson, P.R.Conlin, Changes in diabetes distress related to participationin an internet-based diabetes care management programand glycemic control, J. Diabetes Sci. Technol. 3 (January (1))(2009) 117–124.

[56] A. Holbrook, L. Thabane, K. Keshavjee, L. Dolovich, B.Bernstein, D. Chan, S. Troyan, G. Foster, H. Gerstein, For theCOMPETE II Investigators, Individualized electronic decisionsupport and reminders to improve diabetes care in thecommunity: COMPETE II randomized trial, CMAJ 181 (July(1–2)) (2009) 37–44.

[57] G.T. McMahon, H.E. Gomes, S. Hickson Hohne, T.M.-J. Hu,B.A. Levine, P.R. Conlin, Web-based care management inpatients with poorly controlled diabetes, Diabetes Care 28(July (7)) (2005) 1624–1629.

[58] C.C. Quinn, M.D. Shardell, M.L. Terrin, E.A. Barr, S.H. Ballew,A.L. Gruber-Baldini, Cluster-randomized trial of a mobilephone personalized behavioral intervention for blood

c a l i

M.L. McCabe, C.M. Robbins, P.L. Twohig, Participatory

8 i n t e r n a t i o n a l j o u r n a l o f m e d i

glucose control, Diabetes Care 34 (September (9)) (2011)1934–1942.

[59] J.D. Ralston, I.B. Hirsch, J. Hoath, M. Mullen, A. Cheadle, H.I.Goldberg, Web-based collaborative care for type 2 diabetes: apilot randomized trial, Diabetes Care 32 (February (2)) (2009)234–239.

[60] R.W. Grant, J.S. Wald, J.L. Schnipper, T.K. Gandhi, E.G. Poon,E.J. Orav, D.H. Williams, L.A. Volk, B. Middleton,Practice-linked online personal health records for type 2diabetes mellitus: a randomized controlled trial, Arch.Intern. Med. 168 (September (16)) (2008) 1776–1782.

[61] S.C. Bergeson, M.J.D. Dean, A systems approach topatient-centered care, JAMA 296 (23) (2006) 2848–2851.

[62] P.J. Wagner, S.M. Howard, D.R. Bentley, Y.-H. Seol, P.Sodomka, Incorporating patient perspectives into the

n f o r m a t i c s x x x ( 2 0 1 2 ) xxx–xxx

personal health record: implications for care and caring,Perspect. Health Inf. Manag. 7 (2010) 1e.

[63] C.K. Yamin, S. Emani, D.H. Williams, S.R. Lipsitz, A.S. Karson,J.S. Wald, D.W. Bates, The digital divide in adoption and useof a personal health record, Arch. Intern. Med. 171 (6) (2011)568–574.

[64] L. Green, M. George, M. Daniel, C. Frankish, C.W.R.B. Herbert,Study of Participatory Research in Health Promotion, RoyalSociety of Canada, Ottawa, 1995.

[65] A.C. Macaulay, L.E. Commanda, W.L. Freeman, N. Gibson,

research maximises community and lay involvement. NorthAmerican Primary Care Research Group, BMJ 319 (September(7212)) (1999) 774–778.

Examining the evidence of the impact of health information technology in primary care: An argument...

Documents

Transcript of Examining the evidence of the impact of health information technology in primary care: An argument...

Proposition logic and argument

The Conclusive Argument - Al Islam

How health professionals conceive and construct interprofessional practice in rural settings: a qualitative study

Australian Health Professionals' Attitudes toward Voluntary ...

For professionals

Forest Industry Professionals - NJ.gov

On an Argument for Humility

Directory of One Health Professionals in Chitwan, Rupandehi ...

PRESSURE CLEANERS “for professionals”

KONSISTENSI STRUKTUR ARGUMENT BY COMMITMENT

Recommendations for and warnings against physical activity given to older people by health care professionals

Medical Professionals

The logic behind negotiation: from pre-argument reasoning to argument-based negotiation

Argument Structure Constructions - UC Berkeley

The argument of experience: Experience predicates and argument structure in Catalan and Mayangna

Developing High-Fidelity Health Care Simulation Scenarios: A Guide for Educators and Professionals

code of ethics on interactions with health care professionals in ...

Health professionals’ responses to women’s disclosure of domestic violence

Digital Storytelling: Positive Uses in the Training of Health and Social Care Professionals

Can Spirituality be Taught to Health Care Professionals