Synthesis of informatics literature to support institutional policy statement development.

The Journal of the Medical Library Association (JMLA) case study column provides health sciences librarians with information searching, retrieval, and filtering scenarios in various settings. In previous issues, cases in this series have explored specific patient-care complex questions from clinical teams [1] and best practices for departmental diagnostic decision making [2]. This issue's case explores synthesizing information from the literature to support administrative decision making in health care.


INTRODUCTION
The Journal of the Medical Library Association (JMLA) case study column provides health sciences librarians with information searching, retrieval, and filtering scenarios in various settings. In previous issues, cases in this series have explored specific patient-care complex questions from clinical teams [1] and best practices for departmental diagnostic decision making [2]. This issue's case explores synthesizing information from the literature to support administrative decision making in health care.

THE CASE
In this hypothetical case, you are approached by one of your hospital's administrators. She is the chair of the hospital's policy development committee and asks you to locate information regarding patient perceptions of their access to their own electronic health records (EHRs) to support the development of an administrative policy statement for your institution. Committee members include clinicians and informaticians who have considerable background knowledge of the matter but also value your expertise in information filtering and synthesis. Thus, the committee requests a comprehensive analysis supporting this initiative.
EHRs are gaining increased media and governmental attention, given their potential to increase patient safety and control health care costs [3][4][5][6][7]. Your institution, a recognized leader in designing and implementing EHR systems, is developing white papers and policy statements related to the technology. The goal is to establish the direction the hospital will take as it further develops and refines systems and models for patient access to medical records. You are asked to synthesize the evidence regarding consumers' views of and concerns about the technology, so that these perceptions are adequately addressed in the institution's patient access system, and the hospital's commitment to excellent patient care is proactively communicated to patients.
The administrator requests that you assess the literature regarding multiple facets of EHRs including adoption rates, potential benefits and adverse effects, costs, data accuracy, and patient perceptions. This case focuses on locating and synthesizing information for the aspect of patient perceptions of ''shared,'' patient-accessible EHRs: specifically, what facets do patients see as barriers or facilitators of the EHR, and how do patients perceive the ability to access their own electronic medical records (EMRs)? Figures 1 and 2 include further commentary from an informatician and a physician on the significance of this question.

DEVELOPING YOUR KNOWLEDGEBASE
As noted in previous cases, the first step in the process requires you spend time refreshing or developing your understanding of topics for which you may not have extensive knowledge (Table 1). To learn more about EHRs, you do a quick PubMed search (see ''Constructing a Literature Search'' section below for suggested terms) to scan facets of research in the area and take a closer look at a few relevant articles [10,18]. Because EHRs are often touted as a way to improve health care quality, you also look at the Agency for Healthcare Research and Quality (AHRQ) website, which includes information related to health information technology (IT) and an overview of EHR research [10], and skim the influential Institute of Medicine reports in the area [11,19]. Understanding broader issues surrounding EHR development, implementation, and utilization will help frame your understanding of patients' experiences with this technology.

Electronic health record (EHR) background
Though the visibility of EHRs has been increasing since the 1990s, the EHR concept has been discussed since the late 1950s [20,21], and the seminal Institute of Medicine Computer-based Patient Record report was published in 1991 [19]. Research indicates that EHRs may reduce medication errors, help standardize care across an institution, and lower health care costs (as a result of standardization and increases in service efficiencies) [3,4,11,[22][23][24], but the adoption rate for the systems or components of systems ranges from 11% to nearly 50% in medical practices and hospitals surveyed in the United States [25][26][27][28][29]. Moreover, interoperability, privacy, and return-on-investment issues are significant concerns, especially in view of the numerous examples of significant failures in implementing EHR components and limited data regarding benefits in terms of cost savings and quality improvement [31][32][33][34][35][36][37][38][39].

Potential drawbacks to EHRs
Despite the potential benefits noted in the literature, you also realize from your reading that the effects of EHRs or EHR components such as computerized provider order entry (CPOE) may not be wholly positive. For example, Han et al. observed an increase in mortality coincident with the implementation of a CPOE system [40], though study investigators, as well as other researchers in the field, observed that the mortality increase could have been linked to temporal effects, the application platform (vs. CPOE itself), or other factors [40,41]. Berger and Kichak reviewed two Harvard studies [5,42] frequently cited to demonstrate the potential of CPOEs and noted instead an increase in adverse drug events and The role of the information professional is central to providing a comprehensive framework for the issues at hand as the medical center leadership team develops a policy statement for patients' access to their own electronic medical records (EMRs). Gathering a balanced view of the available evidence is especially important in an initiative such as this, with potential for a high pay-off, in terms of patient satisfaction and improved patient safety, but also potential for high risks, such as alienating patients if privacy and confidentiality concerns are not properly addressed.
Especially of interest, and likely to be a frequently discussed issue for the team, are concerns regarding patients' ability to cope with receiving potentially distressing or difficult-to-interpret results online, without the benefit of direct interaction with a physician. How would the hospital handle queries from potentially distraught patients who have become so after viewing their medical records? Could the misinterpretation of medical information viewed online provoke an already depressed patient to make potentially negative life choices? How liable would the hospital be in these circumstances? These issues of ''information toxicity'' [8] merit significant consideration and are fundamental to the policy development process. A thorough analysis of reported patient barriers and/or facilitators to use of electronic health record (EHR) systems provided by the information professional would uncover closely related issues, such as potentially untoward effects or undiscovered advantages of access to personal health information. Finally, the committee would undoubtedly be interested in comparing policies already in place at other institutions (covering, for instance, such aspects as whether automatically enforced delays are used for the online disclosure of medical data). While such policy details are unlikely to appear in the primary literature, the information professional should strive to compile a reasoned list of institutions that appear to be active in this area and could be contacted for more directed questions. Such a list, as well as the broader analysis mentioned earlier, would certainly enhance the committee's policy development process.

Figure 2 Clinician commentary
The information professional can play a key role in informing the process of policy development. As medical center leadership may often comprise individuals representing a wide spectrum of interests, not all members will have the appropriate background and expertise in the areas that need to be addressed by the policy statement. The inclusion of a librarian as part of the committee will enable the group to leverage the librarian's systematic and comprehensive approach and extensive skill set. From exposure to other successful models and issues reported in the literature, the committee will avoid ''reinventing the wheel,'' and institutional resources, in terms of energy and time costs, can be utilized much more effectively.
The information that the librarian compiles for the committee is furthermore central to the primary goal of the medical center: to improve patient care processes. As patients' access to their electronic medical records is explored, implemented, and evaluated, patient motivations and satisfaction levels must be well understood and adequately addressed to yield a successful endeavor for both the institution and patients. The librarian's comprehensive and timely overview of the state of the literature surrounding patient access to EHRs can have positive implications for the institution's ability to benefit from lessons learned through previous research and ability to develop a policy statement that reinforces to patients the medical center's commitment to excellent and innovative patient care.
health care costs in the initial years of systems implementation [43]. Similarly, Koppel et al. assessed CPOE's role in facilitating medication errors and determined 22 situations in which CPOE could increase probability of prescribing errors [44]. Ash et al. described ''unintended consequences'' of health care IT adoption, including entering care orders for the wrong patient, cognitive overload caused by highly structured data entry interfaces, and reduced communication among clinicians [30,31]. Other potential negatives include total dependence on the electronic system with subsequent inability to provide care during system downtime and workflow and workload issues [31]. In addition, a retrospective analysis of United States National Ambulatory Care Medical Care Survey data showed no association between better quality ambulatory care and providers' use of an EHR as implemented in the various settings included in the data [38]. Patient opinions may also reflect these quality concerns, so you consider this as you prepare to search the literature.

Advantages of EHRs
Despite the identification of possible untoward effects, interest in the potential of EHRs and health IT remains strong and other studies show beneficial effects of health IT. In a systematic review of the effects of health IT, Chaudhry et al. found benefits related to increased delivery of care in line with guidelines, particularly in the area of preventive health, and efficiency benefits related to decreased resource utilization [23]. Similarly, Welch et al. noted increased adherence to guidelines in a study of 4 community physician practices implementing EHRs [45]. In a systematic review of CPOE evaluations, Eslami et al. found inadequate evidence of direct cost reduction or safety increases but also observed increased adherence to guidelines, which, the authors hypothesized, could lead to cost savings and safety improvements [39]. EHR users also perceive benefits: over 75% of respondents to the 2005 American Academy of Family Physicians Center for Health Information Technology felt that EHRs improved patients' health [29]. Garg et al. noted the potential of clinical decision support systems (CDSS) to improve practitioner performance [46], and Grieger et al. demonstrated positive return on investment after the adoption of an EHR system in ambulatory care practices [24].
EHRs, particularly patient/clinician-shared EHRs (or personal health records [PHRs], see Table 1) can bring together disparate elements from separate episodes of care to facilitate a continuous care environment and help clinicians and patients track medications and test results more effectively. Patient records linked to alerts to notify patients of upcoming appointments, medication schedules, or other monitoring functions can form a robust system for empowering consumers to take a more active role in their own health care [10,11,47,48]. As EHR systems and approaches mature, their adoption, utility, and potential for interconnectivity, should increase as well [12,13], and a 2007 survey of health care opinion leaders found that 66% felt that EHR and health care IT adoption are top priorities for improving efficiency [49]. Though not a panacea, EHRs seem poised to play an influential role in improving health care quality. A better understanding of consumer attitudes toward the technology may help leaders best direct EHR development efforts.

CONSTRUCTING A LITERATURE SEARCH
You decide to start your search with the primary medical literature to understand what previous research was conducted by other groups who have implemented patient access to EMRs. Use of the Medical Subject Headings (MeSH) database in the Entrez search system helps you locate a few MeSH terms for the con-cepts of interest. The notion of ''patient perceptions'' is a fuzzy concept to address directly; however, some MeSH terms stand out as possible candidates to include in the search statement. A term of particular interest is ''Patient Access to Records,'' first introduced to MeSH in 2003. As this term is only four years old, you note that previous indexing included ''Patient Rights'' and ''Patient Advocacy,'' terms which may be useful for locating older literature. Additional terms to represent the patient's perspective in this search include ''Patients'' and ''Patient Participation.'' Because you are interested in focusing specifically on barriers and facilitators to adoption, keyword terms such as barriers, perception, obstacles, and facilitators may prove useful.
Over the past twenty plus years, the MeSH term representing EMRs has consistently been ''Medical Records Systems, Computerized.'' You note this during your exploratory searching as you build your knowledgebase, and, upon examining the MeSH information for this term, you note the extensive entry term list of more than twenty terms mapped to this heading. Given the number of terms mapped to this heading, you recognize that it reflects a wide variety of terminology used in the literature and that additional keywords and phrases such as ''EMR'' and ''electronic health records,'' which do not map to this MeSH term, may prove helpful. This particular MeSH term hierarchically falls under the more general term ''Medical Records.'' Though ''Medical Records'' is broader and includes concepts that may not be focused on your search, it may helpful locate articles that discuss broader issues involving patient experiences with access to medical records in either paper or electronic form. You also find that the term, ''Personal Health Records,'' a related component, maps to ''Medical Records.'' Combining these separate concepts and combining available MeSH terms with your selected keywords, your search strategy may look similar to the following: (

EXPLORING THE LITERATURE
This focused search strategy retrieves approximately thirty results. In the search results, you notice a few citations that have been assigned the publication type of ''news'' as well as some that are congress publication types (papers delivered at congresses, symposia, and meetings). These you exclude from initial consideration as a more formal research study is likely to be of greater interest to the policy committee.
Several of the remaining citations are of interest, and you select five research articles to examine further [50][51][52][53][54]. The objective of each of these studies was to eval-uate patient experiences in accessing EMRs, and a quick perusal of the study abstracts reveals data extracted from patient surveys and questionnaires. You also identify a white paper [47] that summarizes discussions from a symposium on PHRs held by the American College of Medical Informatics in February of 2005. Though not primary research, the concepts explored by this internationally recognized informatics group may interest the policy committee.
While selecting terms to use in your initial search strategy, you previously noted that the term, ''Personal Health Records,'' was an entry term for the MeSH term, ''Medical Records.'' Because of this, you decide to try broadening your search strategy to include ''Medical Records'' to account for variations in indexing practices, in the event that an article was assigned this term instead of the more specific ''Medical Record Systems, Computerized.'' Adding the term, ''Medical Records,'' to your search strategy as below broadens your search and retrieves approximately ninety records. Using the ''NOT'' operator and the History feature of PubMed, you are able to remove the duplicate citations between the searches, resulting in a subset of approximately sixty citations. From looking through this set, you quickly note a few studies focused on EHR access via handheld devices [55,56]. As this is not a primary focus of the policy committee, you decide not to examine these further at this time. You also see several articles from the 1990s and late 1980s that discuss patient access to medical records in paper format, before the advent of widespread EMR systems [57][58][59][60]. From the knowledgebase you have developed during your background research, you realize that many of the issues regarding patient access to their health data (e.g., privacy, confidentiality) may be similar, regardless of format. Though you will likely not summarize these items in great detail due to their tangential connection to the information request, you consider this body of literature for mention in your written response to the policy committee so they understand that the literature reflects patient concerns overall in regard to their medical records, regardless of format.
From this broader set, you also identify 2 articles for further consideration. The first is a 2004 Archives of Internal Medicine study in which researchers surveyed 4,500 adults to determine characteristics of those interested in examining their medical records [61]. This may be applicable as the committee seeks to understand facilitators of EMR use by patients. The second, a 2003 review article, collates outcomes from previous studies on patient access to medical records [62]. This may be an effective supplement to the primary research. Table 1 Concepts often discussed in conjunction with electronic health records

Concept Definition
Electronic medical records (EMR) or electronic health records (EHR) The Agency for Healthcare Research and Quality (AHRQ) provides a useful overview of EMRs and EHRs: ''An EMR (electronic medical record) is generally defined as the set of databases (or repositories) that contains the health information for patients within a given institution or organization. Thus, an EMR contains the aggregated datasets gathered from a variety of clinical service delivery processes, including laboratory data, pharmacy data, patient registration data, radiology data, surgical procedures, clinic and inpatient notes, preventive care delivery, emergency department visits, billing information, and so on. Furthermore, an EMR contains clinical applications that can act on the data contained within this repository-for example, a clinical decision support system (CDSS), a computerized provider order entry system (CPOE), a controlled medical vocabulary, or a results-reporting system. In general terms, EMRs are clinician-focused in that they enhance or augment the workflow of clinicians or administrators.
EMRs are said to be interoperable if they are able to exchange (transmit and receive) data using standardized data transmission (coding and messaging) formats (standards). An EHR (electronic health record) extends the notion of an EMR to include the concept of cross-institutional data sharing. Thus, an EHR contains data from a subset of each institution's EMR (that is agreed upon by the institution). An EHR may also reside 'entirely within one institution' and link the various affiliated practice sites together. The EHR is generally patient focused and spans episodes of care rather than a single encounter. An EHR can only be present if the participating sites all have an EMR in place that is interoperable'' [9].
Personal health record (PHR) The term ''personal health record'' describes several approaches to patient access to the medical record, typically including read-only provider-, payer-, employer-, or health system-maintained electronic personal health information that patients can access in whole or in part; a personal file in print or electronic form that is maintained by patients and shared at their discretion with providers and may or may not be created using specialized PHR software; or a ''direct-to-consumer'' marketed system to which patients can subscribe and input their medical information, family history, etc. The variety of systems and the sensitivity of personal health information lead to questions about ownership of information, privacy, and security of data. Ideally, the PHR should be portable, interoperable or sharable across systems, and secure [10,11].
Interoperability standards With the variety of EHR systems that are on the market, a key issue involves interoperability among systems. Health Insurance Portability and Accountability Act of 1996 (HIPAA) regulations regarding patient access to the medical record HIPAA, public law 104-191, was enacted in 1996 to protect the privacy of health care information, increase access to health insurance, and promote standardization and efficiency in health care [14]. ''Covered entities'' or organizations/ individuals who provide care or transmit health information, provide third-party payment for health care (such as insurance companies), or transform health care transactions (such as a billing agency) are subject to HIPAA regulations. Organizations such as software development firms that work with covered entities must take HIPAA rules into account as well. HIPAA includes numerous regulations, but those related to privacy of health information essentially mandate that patients have access to copies of their medical records, can request corrections of errors, and must be notified of how their medical information may be used. Providers must maintain written privacy procedures to share with patients and must train employees in the procedures. The regulations also stipulate limits on the use of personally identifiable health information and require that covered entities take reasonable steps to communicate confidentially with patients at a patient's request [15].

National Health Information Infrastructure (NHII)
The NHII is national initiative outlined by the US National Committee on Vital and Health Statistics to the Department of Health and Human services in a final report in 2001 [16]. The NHII encompasses not only technology, but also the values, practices and relationships, laws and regulations, standards, privacy safeguards, and systems applications to comprise a ''comprehensive knowledge-based network of interoperable systems of clinical, public health, and personal health information that would improve decision-making by making health information available when and where it is needed'' [17]. The NHII operates in the patient, health care provider, and population health dimensions, interconnecting all three. EMR and EHR systems can play a pivotal role in the development of the NHII.
From your initial searching, you realize that there seem to be few citations within the past couple of years. Between your first two strategies, you find seven articles published in 2006 to 2007. You realize that this is an opportunity to leverage PubMed's Related Articles feature. It may help retrieve relevant articles not picked up by your original search strategies due to nuances in terminology and indexing. You decide to use this feature for each of the articles you have selected at this point for further review and to examine more closely results from the previous few years. From this process, you locate a few additional articles for consideration in your packet [63][64][65].
Given the topicality of EHRs and the technology's broad appeal in the public sector, you realize that there is likely a significant body of gray literature-such as web-published reports, white papers, and existing policy statements-related to EHRs. A quick web search using variations of the keywords EHR, EMR, electronic medical record, electronic health record, and computerized patient record-in combination with search terms like patient access, attitudes, and so on in major search engines like Google and AllTheWeb-returns links to numerous articles in PubMed as well as relevant documents from organizations like AHRQ [9]. Other organizations' papers include documents from the American Health Information Management Association [66] and the Healthcare Information and Management Systems Society [67].
Given your knowledgebase on the issue, you are also already aware of the Commission on Systemic Interoperability's Ending the Document Game report [12] and its extensive discussion of the benefits of the EHR for patients. Because your synthesis for the policy committee will largely focus on patient perceptions of EHRs, you decide to list the systemic interoperability report as a valuable background reference.
To ensure that your search process has been com-

Figure 3
Overall summary of the literature The literature examining how patients perceive access to their EMRs mostly comprises studies that use a mix of quantitative and qualitative research methods. This is to be expected, as qualitative research methods are valuable for eliciting subjective experiences and attitudes. While the body of literature that focuses specifically on this aspect is not large, several research articles provide insight into possible patient motivations and perceived benefits of EMRs and into areas where patients express concern and may be barriers to access. This packet comprises three sections: 1. two review articles [47,62] that provide an overall framework for patients' access to EMRs 2. a selection of research articles that have larger numbers of participants (ranging from approximately 100-1,400 patients) with specific emphasis on understanding how patients interact with and respond to having access to their EMR 3. additional references to supplement the review and primary literature Overall, the research reflects that patients recognize numerous benefits to having access to their EMR and value the ability to communicate directly with health care providers, access laboratory or test results, and look up information to further understand the information provided in their record.
Areas of patient concern tend to be largely focused on issues of privacy, confidentiality, security, and accuracy-issues also reflected in older literature regarding patient access to medical records in paper format. The studies indicate that among other issues, patients are concerned about the ability of unauthorized personnel, including employers and other third parties, to view their medical records and the possibility that records in an electronic format may result in increased data errors.
The geographic locations for these studies have been explicitly indicated in the first column of the summary table. Further contact information for these research groups, as well as other groups that may be involved in EMR implementations with combined patient access, can be prepared for the committee on request.
Limitations of this body of research include the frequent use of surveys, interviews, and focus groups, which present inherent study design limitations such as recall and self-selection bias. Additionally, each study describes challenges in generalizing study results given nonrepresentative study population types when compared to the overall population.
Additional references are included that provide further background information for the committee. Those references have smaller study populations but are often cited and describe patient concerns for EMRs [8,52]. An Archives of Internal Medicine paper [61] provides extensive analysis of the socio-demographic characteristics of patients who express interest in reading their medical records. The report from the Commission for Systemic Operability provides a series of stories of experiences directly from users of EMRs that personalizes the role of the consumer in the EMR process. The report outlines the advantages and disadvantages of patient EMR access.
prehensive, you would also likely elect to search additional resources and databases that may contain relevant information, such as Web of Science, EMBASE, and ScienceDirect.

SUMMARIZING THE INDIVIDUAL REFERENCES
You decide on a final selection of articles to summarize for the committee. To set the overall framework for the research you have located, you decide the first two articles will be the white paper by Tang et al. [47] and the review paper by Ross et al. [62]. As a symposium synopsis, the Tang paper can provide the committee with an overview of the issues involved in patient access and PHRs. You decide to include the Ross review as it provides a tabular synopsis of outcomes of studies with patients who received access to their medical records. As this review also covers the body of literature you noted regarding access to medical records in paper format, it will serve as a useful representation of this literature, while enabling you to focus the primary literature that you provide on research conducted specifically in EMR systems.
In selecting the primary literature to include, you consider the location of the study, the types of systems used, and the study sample size. Given the interest in understanding patient responses to medical record access via the EHR, study design types that involve questionnaires and/or surveys are likely most appropriate as qualitative research can yield rich data sets that can provide insights into a group's thinking. With these considerations in mind, you select five research articles [50,53,54,64,65] (Table 2).
To supplement this core selection of articles, you also decide to include three additional articles and the previously mentioned Commission on Systemic Interoperability report [12] for presentation to the committee as additional references. Your searching reflects that research from Masys et al. [68] and Cimino et al. [52] makes important contributions to the body of published work in this area. A third article from Fowles et al. reports an in-depth analysis of clinical and socio-demographic characteristics of 1,293 patients who reported interest in reading their medical record [61].

PRESENTING THE RESULTS
To effectively organize these selected articles given the nature of the question, you synthesize your findings in a tabular format to list the various outcomes in a quickly ''scannable'' format for the committee. You determine that your table will include the following column headers: Ⅲ Author: providing enough detail for the reader to quickly glean the citation information; in particular, you will include the institution reporting the research Ⅲ Study design and patient population: including details of study methodologies, types of instruments used, and outlining the purpose for the research; providing specifics on the patient demographics will also be useful for the committee to understand what type of patients have been included in the studies and help them assess applicability and generalizability to their own patient population Ⅲ Results/outcomes: outlining the specific results from the work that are most relevant for the question asked; to indicate facilitators and/or barriers, you will include subheadings in this section to indicate under which category the results fall; in this column, you will also note study limitations For the overview articles and articles selected to be included as additional references, you will not provide tabular summaries of their content, but instead write brief descriptive statements detailing their contribution to thinking in this area. Table 2 provides an example of how you might format your summaries for the five research articles you have chosen.

OVERALL STATE OF THE LITERATURE
An overall summary of the literature will assist the policy development team in understanding the breadth of the research conducted in this arena. With MyChart patient users who registered for the site but did not log into their account Demographics of the 1,421 survey respondents: Ⅲ 54% were ages 46-54, 24% were ages 31-45, 16% were older than 65, and 5% were ages 18-30 Ⅲ 60% female, 40% male Ⅲ 40% educated beyond high school and less than 4 years of college, 33% high school or less, and 27% 4-year college graduate or beyond Ⅲ 98% white, 2% other Positive experiences Ⅲ On a scale of 1 (hard)-100 (easy), ease of use scored a mean rating of 83 for ease of registration and logging in, 85 for email, 80 for prescription renewals and 78 for first time email use Ⅲ All respondents indicated they could understand their medical information and test results; mean scores for these categories averaged more than 80. Ⅲ Investigators noted that patients were not concerned about becoming worried by the possibility of seeing abnormal test results or getting results prior to having discussions with health care providers. Ⅲ Majority of patients preferred online communication for renewing prescriptions (more than 60%) and getting answers to their general medical questions (more than 50%).
Patient concerns Ⅲ Approximately 50% indicated their PHRs were not complete. Ⅲ Approximately 25% of respondents indicated they felt their medical history was not accurate. Ⅲ Approximately 30% of respondents indicated some concern about the confidentiality and security of their online records.
Limitations Ⅲ Survey response rate of 33%, thus authors raise concerns for the possibility that the sample may not be representative of the population as a whole. Ⅲ Survey instrument was only compatible with Internet Explorer 5.x or higher. Ⅲ All patients were early adopters of the technology.

Ralston JD et al. (2007) [64]
Group Health Cooperative, Seattle, Washington Examined features patients used most frequently, as well as overall patient satisfaction with the health care system's patient website, MyGroupHealth; researchers conducted a retrospective, serial, cross-sectional study from September 1, 2002-December 31,2005 Study design Ⅲ all patients who used the site during this time period included in the study for website usage patterns (nϭ527,603) Ⅲ patient satisfaction survey given to a random sample of users of the portal in August 2004 (nϭ896); demographic characteristics of this group not reported Satisfaction results Ⅲ 81% of survey respondents reported being very satisfied with the portal. Ⅲ 55% reported satisfaction with test results. Ⅲ 96% were satisfied with medication refills. Ⅲ 93% were satisfied with secure messaging.
Limitations Ⅲ Focus of study was limited; future work should more closely examine demographic, health status, and other characteristics of users. Ⅲ Measures of adoption could in the future focus on more clearly delineating the intersection of adoption to patient value and health outcomes.
Authors note their results should inform developers and policy makers; initial development should focus on providing interactive patient-provider communication and engage patients with the information in their record that is highly relevant to ongoing care at the institution. Patient responses Ⅲ 83% think they should have access to their health records. Ⅲ 66% think the advantages of electronic records outweigh the disadvantages. Ⅲ 57% thought it would be useful to have their records on a SMART card (credit card-sized cards that store electronic data).
Areas of patient concern Ⅲ 67% would like to determine who can view their electronic health record. Ⅲ 50.4% were worried about the security of their records as well as possible places their records could be linked to outside of their general practitioners office. Ⅲ 29.3% were worried there may be errors in their records.
Limitations Ⅲ Study was biased toward women and older patients. Ⅲ Researchers indicate their local population includes 9%-10% who speak languages other than English and 98% of their respondents spoke English as a first language; questionnaire was not available in other languages.

Chhanabhai P et al. (2007) [65] University of Otago, New Zealand
Conducted as part of a nationwide survey of patient perceptions on the security of electronic health records in New Zealand Researchers Ⅲ conveniently identified and sampled physician practices to participate in the study Ⅲ provided practices with 30 surveys to have completed by patients A total of 300 patients representing 4 cities returned the study surveys; demographic characteristics of patients are not reported Patient concerns Ⅲ 42% believed that an EHR would facilitate sharing of their data without their permission. Ⅲ 40.3% perceived that the availability of an EHR would result in sensitive medical information being leaked out. Ⅲ While 41% reported that they agreed with the statement that ''electronic health records increase medical errors,'' 38% indicated they did not agree with this statement. Ⅲ 53% responded that an EHR would contain weaknesses in the security system. Ⅲ 79.4% indicated concerns with possible hackers into the EHR. Ⅲ 54.9% indicated a belief that paper records were more secure than electronic records.
Limitations Ⅲ Authors indicate potential limitations with regard to not having specifically set out to determine variations in perceptions among ethnic groups. Ⅲ Research design may have prompted patient response bias toward agreeing with the list of concerns. Objective Ⅲ investigate how patients and physicians utilized the EHR system, System Providing Patients Access to Records Online (SPPARO), specifically exploring functions used and attitudes towards use Study design Ⅲ randomized controlled trial conducted from January 1-December 31, 2002, with a medical subspecialty clinic; control group patients (nϭ53) received standard of care and were provided access to SPPARO after the study period; intervention group patients (nϭ54) were provided access to SPPARO during the study period Ⅲ participants completed written questionnaires before and after the trial period Ⅲ 16 control group patients participated in semi-structured telephone interviews Ⅲ 13 patients participated in focus group sessions Patients were 77% male, mean age of 56 years, 90% white Quantitative results Ⅲ 90% (CI 82%-95%) of the patients were in favor of a patient-accessible record; 85% (CI 77%-91%) were interested in reading their records in the future. Ⅲ 31% (CI 22%-40%) of the patients had concerns about privacy and confidentiality.
Qualitative results Ⅲ Interviews with patients found that attitudes toward SPPARO were ''overwhelmingly positive'' and the following benefits of having access were identified: learning more about conditions, coordinating care, learning about medical decision making, reinforcing their memory regarding patient education instructions, confirming normal results, confirming the accuracy of their records. Ⅲ Patients expressed few concerns regarding access to their EHR, with some specific concern that their records might be released to employers or government agencies without their permission. Ⅲ Suggestions for the system included aids to help understanding lab values, links to explain technical terms, and ability for the patient to annotate their record when errors were found.
Limitations Ⅲ Trial was conducted in subspecialty clinic and may not be applicable to most practice settings. Ⅲ Patient population reflected a higher level of education and computer literacy than the general public. Ⅲ Much of data was qualitative and statistically significant differences could not be shown in this data.
the knowledge gained from your investigation, you develop the following key points to address: Ⅲ describe the overall quality and quantity of the published work on this topic Ⅲ summarize the key issues reported in the literature regarding patients' access to their EMRs Ⅲ outline the studies selected for inclusion in the packet and the rationale for selection Ⅲ detail the contributions that the articles selected as additional references bring to the overall packet Ⅲ characterize related literature that may interest the committee but is not directly relevant to the question at hand (for example, the older literature you noted about patient access to paper records) Figure 3 provides a representative overall summary of the literature for this packet.

CONCLUDING REMARKS
Your investigation of the requests of the policy development committee has provided you with a broad picture of the ongoing initiatives related to EMR systems and patient interaction processes. The compilation of research that you provide to the policy development committee provides them with an overall understanding of previous work that has been conducted in this area. Furthermore, the list of institutions represented by the summary packet may prove useful to the committee for identifying possible sources of further information and data sharing.
The impact this will have on the health care experience of your hospital's patients is tremendous, for as the committee members consider the experiences of other institutions who sought to understand patient perceptions of access to EMRs, it will further define the role the institution will have in providing patients access to their clinical care documentation. Patients will likely appreciate this commitment from the institution to continually provide an environment in which their voices matter and the knowledge that their opinions and concerns contribute to better services for patient care. As a member of this committee, you are respected for your contribution to this process!