In Chapter 1, the committee defines the electronic health record (EHR) as the electronic version of the patient health record, and the EHR system as both the database of that information and the tools used in various workflows to collect the information and support decision making and analysis. The addition of the recommended panel of social and behavioral measures to the EHR has implications for workflow changes to collect and review the new information and for workflow changes needed to address the problems surfaced through these measures. Both types of changes will require modifying how clinical teams operate and how patients report on their own experiences and engage in health-relevant behaviors; however, those details are beyond the scope of this study.
In addition to the challenges that surface when adding any new data to the EHR (particularly the time needed to obtain information), some challenges are specific to the addition of social and behavioral data. Responsibility for addressing social and behavioral determinants of health that surface as problematic in the EHR generally fall outside of the traditional health care system. However, to the degree that addressing these determinants reduces near-term health care utilization, such as hospital readmissions, as well as improves health and reduces future health care service utilization, the investment of time and resources will be well worth it. This chapter describes some of the anticipated challenges, suggests ways to overcome these challenges, and identifies offsetting benefits of implementing the panel of recommended measures. While some of the barriers will be difficult to resolve, case studies have been included to illustrate successful experiences of implementation.
Although EHRs have great potential to improve quality, coordination, safety, health outcomes, and overall efficiency in health care, many obstacles exist in fully realizing their potential (IOM, 2012). A systematic review identified eight categories of barriers to physician adoption of EHRs: financial, technical, time, psychological, social, legal, organization, and change process issues (Boonsta and Broekhhis, 2010). The review suggests that it is valuable for hospital managers, project leaders, and change managers to understand which of these are of greatest concern to the physicians with whom they work in order to find solutions. For example, if physicians report that their time is overloaded with data entry, workflow could potentially be redesigned. Other members of the clinical team (e.g., nurses, pharmacists, physician assistants) as well as patients are drivers in EHR adoption and can identify workflow solutions. However, physicians’ use of and attitude regarding EHRs are the most commonly studied among an overwhelmingly large number of publications (Junhua et al., 2013).
Successful adoption or modification of EHRs involves sociocultural change. Individuals’ roles, workflows, decision making, and communication will change and adapt over time. Careful reconsideration and redesign is needed to align the changes and achieve the full benefit of the technology. Box 6-1 lists the identified principles for design, implementation, and policy for EHRs from Sinsky et al. (2014).
With the rapid adoption of EHRs in response to the Meaningful Use incentives, many health systems and practices have implemented the technology without pausing to work out these alignments. The fatigue of adapting to new systems is acknowledged by the committee. For example, of the 58,000 Medicare-eligible providers who attested to Meaningful Use in 2011, 16 percent did not re-attest in 2012. It is noteworthy, however, that 44 percent of the latter returned and attested in 2013 (HealthData.gov, 2014).
More than four out of five doctors say they prefer to continue working with this evolving technology that holds the promise of enhancing care rather than return to paper records (Friedber et al., 2013). Programs exist, such as the federally funded regional extension center program, to provide clinical teams with assistance in purchasing and implementing EHRs, training staff, and addressing how teams use EHRs in practice (Hsiao and Hing, 2012; Hsiao et al., 2014).
It is beyond this committee’s charge to address the general challenges of EHR use. Other reports, such as the IOM report Health IT and Patient Safety: Building Safer Systems for Better Care, and the NRC report Computational Technology for Effective Health Care: Immediate Steps and Strategic Directions, better address these issues (IOM, 2012; NRC, 2009).
Principles of EHR Design, Implementation, and Policy
1. The use of an EHR should add value for the patient.
2. The primary function of an EHR is clinical care.
Health care professionals
3. The use of an EHR should improve, or at a minimum not reduce, the well-being of health care workers.
4. The use of an EHR should align the work with the training of the worker.
5. The EHR is a shared information platform for individual and population health.
6. The use of an EHR should minimize waste.
7. Electronic workflows should align with clinical work.
8. Various methods of communication, including nonelectronic forms, will be necessary for optimal patient care.
Regulation and payment
9. Sufficient resources should be available for the new work associated with the advanced use of an EHR.
10. Policies around EHR use should reflect the strength of the evidence base supporting them.
11. Regulatory balance between often competing values (e.g., clinical quality versus security or efficiency versus performance measurement) should be sought.
SOURCE: Sinsky et al., 2014.
However, the committee’s awareness of these issues set the context in which decisions about adding additional data to the EHR were made. The committee was cognizant that its recommendations could increase the burden on health systems, clinicians, patients, and vendors, in addition to implementers of meaningful use regulations. As a result, the committee’s criteria for selecting domains and measures for inclusion strove to provide a systematic approach to weighing the trade-offs (identified in Chapter 2).
Data for EHRs can potentially be collected in many ways. It may be self-reported or reflect the judgment of a member of the clinical team. It
may be imported through extraction from other data sources (e.g., vaccine registries or community datasets) or via a personal device. Even in low- and middle-income countries, data are now routinely collected via cell phones, personal digital assistants, and other modalities, and use in U.S. health care settings should be feasible (Glasgow et al., 2012). Data may be collected directly from the patient on paper or, preferably, via a computer, or through an interview with a member of the clinical team.
The most appropriate approach to collecting data varies among the social and behavioral identified measures. For example, while some EHRs use racial categories assigned by medical care personnel, the committee endorses capturing race by self-report as it is a cultural construct reflecting the individual’s self-perception. A patient’s residential address may be verified by administrative staff. Discussion of interpersonal violence may be most appropriate as part of the clinical interview. Social exposures can be inferred through geocoding of neighborhood indicators.
Many of the measures of social and behavioral determinants of health identified by the committee are best obtained by self-report from patients or their caregivers. Estabrooks et al. (2012) detail strong support for using self-reported data elements on health behaviors and psychosocial factors for the EHR. Self-reported data are most reliable when the item contexts, stems, and response options are clearly written, reliable, valid, and meaningful to the respondent. If data are collected by self-report, the clinical team needs to take specific steps to ensure that the responses are complete and accurate from the patient’s perspective. For example, for patients with low literacy or who are visually impaired, it may be necessary to have a staff member read and record the response to selected items. Language limitations also should be considered as well as use of alternative mechanisms. For example, audio assists using a patient’s preferred language are currently in use in some settings. Box 6-2 provides an example of a clinic’s success in capturing self-reported data to identify and treat at-risk behaviors.
Interviews during clinical encounters afford a measure of professional oversight but also add time and complexity to the encounter. The clinical team needs to decide whether the questions are best asked by the administrative staff, physician, nurse practitioner, physician’s assistant, or another health professional. These individuals need to consider how to ask the question and how to communicate its importance to the patient. Cultural variations in terminology and meaning may reveal or obscure the exact meaning of the concepts and their role in an individual’s life.
New electronic data collection software, Web-based data entry options, and EHR applications have made clinical implementation easier, allow-
At-Risk Behaviors, Identification, and Treatment in Clinics
In response to a concern regarding underdiagnosis of at-risk behaviors and outcomes, the University of Washington Madison HIV Metabolic Clinic developed a Web-based, self-administered patient-reported assessment tool and integrated it into routine primary care for adult HIV-infected patients (Crane et al., 2007; Fredericksen et al., 2012). The patient-reported assessment included brief, validated instruments measuring clinically relevant domains including depression, substance use, medication adherence, and HIV transmission risk behaviors. Patients complete the assessment just prior to seeing a member of the clinical team, and providers receive the results as they begin the patient visit. The assessment was integrated into routine HIV care with the support and coordination of clinic staff.
Workflow, technology, scheduling, and delivery of assessment results were completed using a plan-do-study-act (PDSA) cycle (Tufano et al., 2010). Researchers found the Web-based self-reported assessments to be a feasible tool that can be integrated into a busy multiprovider HIV primary care clinic. They assessed the impact of self-reported outcomes results on provider behavior and found that it led to increased provider awareness and action for at-risk behaviors and diagnoses (Fredericksen et al., 2011). Automated real-time notification of suicidal ideation was found to be particularly valued by providers (Lawrence et al., 2010).
Critical factors for successful integration of such assessments into clinical care include strong top-level support from clinic management, provider understanding of self-reported assessments as a valuable clinical tool, tailoring the assessment to meet patient and provider needs, communication among clinic staff to address flow issues, timeliness of delivery of results to providers, and sound technological resources.
The initiative was expanded into clinical care into seven HIV clinics as part of the Centers for AIDS Research Network of Integrated Clinical Systems (CNICS) cohort. With the addition of each clinic, a tailored integration was developed to meet that clinic’s needs particularly related to clinic flow and provider feedback and differences in electronic health records. As part of CNICS, HIV-infected patients across the United States have completed the assessment approximately 34,000 times providing a wealth of clinically relevant data to improve clinical care, and population health, and to facilitate clinical research (Crane, 2014).
ing for immediate scoring that can be displayed for review during clinical encounters. Clinicians and patients prefer electronic collection (Valderas et al., 2008), which is associated with lower rates of unanswered questions than with paper forms and higher rates of reporting risks such as violence in the home and substance abuse (Gottlieb et al., in review). Reproducibility of electronic data collection is high, reducing missing data and allow-
Priority Considerations for Using Self-Reported Data
- Specifying the goals for data collection (screening, diagnostic, outcome assessment)
- Selecting the patients, setting, and timing of assessments
- Determining which questions to administer
- Choosing a mode for administering and scoring the questionnaire
- Designing processes for reporting results
- Identifying aids to facilitate score interpretation
- Developing clinical strategies for responding to issues identified by the questionnaires
- Evaluating the impact of the patient-reported outcomes intervention on the practice
SOURCE: Adapted from Snyder et al., 2012.
ing complex skip patterns. Patients interacting with modern systems can experience a consistent look and feel across content and selection methods. Internet connectivity is rising in both private and public clinical locations. Approximately 80 percent of American households indicate regular Internet use (ESA and NTIA, 2011). This rise in connectivity has increased the range of locations where patients can complete questionnaires (e.g., at home, on waiting room kiosks, or on a personal smartphone).
Beyond the time required for data acquisition, it also takes time to interpret and develop appropriate clinical responses to issues identified from the data. The clinical team needs to consider the time during the course of an encounter that is most appropriate to collect or review the information. Some clinics start patient appointments 20 to 30 minutes in advance of the physician encounter to provide enough time for completion of self-reported data before the physician visit begins. If data are collected at home via a patient portal, personal health record, or an email link to a Web-based survey, they may need to be obtained close in time to an encounter so responses are relevant when physicians review the results. Whatever the sequence, time has to be set aside for these steps. It is also important for health care systems to help patients understand the purpose and the value of the information being collected by self-report, when that method is used (Greenhalgh et al., 2005; Lohr and Zebrack, 2009).
Acquiring social and behavioral data at the point of care may generate the expectation that the clinician will, in turn, act on that information.
Indeed, best practices for acquiring information about some social and behavioral data require that an intervention plan be in place (Ockene et al., 2007). This is especially true for problems that fall within the traditional health care system, such as depression. Otherwise, the patient may be left with a positive clinical finding, but not the tools needed to address the health need. Even if it is not possible to address some domains within a primary care setting, efficient and effective intervention resources often exist through referrals. Shared decision-making aids may be indicated that would use data to help patients and their health care teams collaborate to make informed decisions (Glasgow et al., 2012). The International Society for Quality of Life Research’s 2012 guide for implementing a self-reported data collection system in clinical settings identified eight key design considerations for self-reported data collection systems, as seen in Box 6-3.
Using the EHR as a repository for social and behavioral domains is challenging. EHRs originated as the legal record of medical encounters and admissions. Thus, beyond their role in informing diagnosis and treatment, EHRs are legal institutional archives of care events organized at the level of the individual. The data stored in the EHR still largely reflect the care experience and rarely present a complete view of the patient’s health state. Elements of the EHR that document patient history and progress notes may be unstructured narrative or structured as text insertions into structured forms or numeric data. While rich narrative may be the best way to tell parts of the patient story or a clinician’s assessment of medical information and its meaning, structured data and standardized measurement are needed to enhance retrieval, analysis, and interoperability to support clinical care, population management, and clinical research (Fridsma, 2013).
The committee’s criteria for selecting a domain include availability of a standard measure. Several projects are facilitating standard representation of behavioral data, including the Grid-Enabled Measures (GEM) database (Min et al., 2014); consensus measures for Phenotypes and eXposures (PhenX) (NIH, 2012); the Patient Reported Outcomes Measurement Information System (PROMIS) (PROMIS Network, no date); the NIH toolbox (NIH, 2006–2012); and the National Collaborative on Childhood Obesity Research, which have developed standard measures and common definitions (NCCOR, 2014). The committee strove in Chapter 4 to provide as much guidance as possible in order to support consistent acquisition of social and behavioral data according to an interoperable standard. Chapter 4 identifies common metrics, where available, and standard measures for use in EHRs.
Interoperable standards are needed for health information exchanges
(HIEs) to succeed. HIEs commence when one health care professional or health system shares data electronically with another. Reducing the burden of phoning, printing, scanning, and faxing potential sensitive documents will improve the quality, safety, and efficiency of health care delivery (HealthData.gov, 2014). Once information is stored in an EHR it is subject to federal and state laws and regulations and to institutional policies and procedures, which may place significant barriers on the efficient reuse of the data outside the point in which the data are captured. Appendix B provides a more robust description of privacy protection issues.
As described earlier, select elements for some determinants of health may be found in other sources related to the patient, such as EHRs from other institutions; personal health records, third-party data integrators, such health risk appraisals gathered by insurers or employers or clinical data registries, community agency datasets, national surveys, and datasets from other sectors like retail. Presently there are few straightforward ways to transfer data from external data sources to EHRs or vice versa. Importing data from external sources requires the importing institution to determine the provenance of the data, its accuracy, and its validity. An additional challenge in this area arises from the absence of data standards and terminologies that ensure the meaning and interpretation of the data remain true to their original source.
Open architecture models of health information systems, such as that advocated in the report by JASON/MITRE Corporation A Robust Health Data Infrastructure (AHRQ, 2014), hold the best promise for ensuring the data flows needed to make social and behavioral determinants of health are accessible to the patient, in clinical care encounter, to the health system, and to society. This document lays out a health information infrastructure that represents a significant departure from the one(s) existing today. Today’s health information infrastructure can best be described as a series of hub- and-spoke configurations, where the hub represents an institution’s EHR system and each spoke represents a one-to-one pathway to an authorized business partner where the business partners may be another health care delivery organization, a health information exchange operation, a clinical laboratory, or a physician’s office. Sharing data devolves to a process of opening a trusted channel of information flow and creating a point-to-point connection. Fine-grained access control and data exchange are nearly impossible, as records are exchanged in totality, not as individual data elements.
In the robust data infrastructure envisioned in the JASON report, data are stored at the point of acquisition and integrated at the point of need. Record systems are separated from the tools that operate on them, and information integration is driven by clinical or policy need, not by acquisition strategies. With such an open architecture, the committee’s recom-
mended data elements could be acquired from a wide variety of sources. Integration and updating at the point of care would be feasible but not restricted by the constraint of the clinical information systems.
Several EHR vendors are beginning to collaborate in order to achieve interoperability of records (Bresnick, 2013; Moukheiber, 2014; ONC, 2012). These are important steps and offer the possibility that priority health-relevant social and behavioral domains collected in a clinical encounter at one institution could be available to clinicians at a different institution.
Risks to the patient in some sensitive areas such as substance use or violence represent considerable challenges to collecting data. A recent pilot project examining interstate behavioral health data exchange demonstrated that some privacy concerns need to be addressed to facilitate exchange of sensitive behavioral data nationwide (Parker et al., 2014). However, basic safety measures are widely used. Covered entities (e.g., providers, health systems) and their business associates need to be in compliance with the Health Insurance Portability and Accountability Act of 1996 (HIPAA). HIPAA’s Privacy Rule1 establishes the rules governing the use and disclosure of identifiable health information in either paper or electronic format. HIPAA’s Security Rule2 establishes the security safeguards to be adopted to protect electronic identifiable health information. Other laws govern public health authorities, and state laws are also applicable (see Appendix B for a commissioned paper on privacy concerns). When possible, data can be de-identified to better protect anonymity. For example, in syndromic surveillance, the public health entity only needs to know how many cases there are and, perhaps, associated information such as age, sex, neighborhood, but it need not know the specific names of individuals. In cases where there is a need to individually link EHRs to a public health registry, the data cannot be de-identified, thus raising privacy concerns. However, the data that are transmitted can be encrypted.
Institutions should inform patients about the specifics of data sharing. For example, if data are being shared with public health officials, patients should not only be informed that this is occurring; they should be informed about the rationale and benefits of that information being shared as well. Further protections include asking persons who handle confidential data to sign oaths of confidentiality with clear penalties spelled out for unauthorized release of protected information and making sure that all information
1 45 CFR Part 160 and Subparts A and E of Part 164.
2 45 CFR Part 160 and Subparts A and C of Part 164.
is password protected within the system. Audits of attempts to access the data can be conducted to assure that only those who have a legitimate purpose in looking at the data can do so.
With the above protections in place, routine collection of these types of potentially sensitive data may not only provide important information for diagnoses and treatment but may have the additional benefit of normalizing or destigmatizing discussion of sensitive issues in clinical practice.
The business model for capturing social and behavioral domains and measures into the EHR has yet to be fully realized because few examples exist. If care is planned mindful of the patient’s social and behavioral profile, the cost savings from social and behavioral interventions could be substantial as described in earlier chapters. However, those who bear the costs of collecting and acting on social and behavioral determinants of health may not be the ones who benefit from the cost savings. These benefits accrue to society, health care payers, and health systems that are reimbursed for population management. While some of these benefits are near term, many accrue over a period of years. The costs of adding social and behavioral domains to EHRs, such as programming, modifying workflows, and intervening on positive screens, often fall on the clinical practice or hospital. The movement toward population management and accountable care organizations may address this misalignment over time. In the meantime, misaligned costs and benefits remain a barrier.
Costs and benefits are not just financial, and they will be experienced differently depending on the clinical practice. A common question is the time needed to capture the measures or manage the self-reported information. In addition, time and resources will be needed to address the risks identified. Some care settings may be better equipped than others to meet these needs. Large health systems are more likely to have access to specialized programs such as stress management or smoking cessation than small practices. However, addressing these determinants is an important aspect of quality care which is equally relevant for all practices, small and large. Over time, the movement toward patient-centered medical homes, population health management, and health care data exchange will reduce these differential burdens. In the meantime, the committee’s recommendation of a parsimonious set with the fewest measures that would provide a balanced psychosocial vital sign minimizes the burden.
Four key stakeholder groups are likely to be affected by the inclusion of social and behavioral determinants of health in the EHRs: individual patients, clinicians, health care providing institutions, and society in general. Each of these groups stands to benefit in unique but interconnected
TABLE 6-1 Stakeholder Concerns and Examples of Mitigation Strategies for EHRs
Mitigation depends on public education and clinician attitude and explanation, for example:
Mitigation involves adaptation of workflow and clinical strategies, including
|Health care system||
Mitigation involves a population management strategy, including
Mitigation involves community health assessment and improvement at local and national levels, including
ways, and each is also likely to have unique concerns which can be mitigated by careful attention to implementation strategies. Table 6-1 outlines implementation questions from various stakeholder perspectives and examples of mitigation strategies discussed by the committee.
Linking data from EHRs to local public health departments and community agencies provides several advantages to patients, providers, and the broader community. Information can flow in both directions. For example, data in EHRs can enable public health practitioners to identify groups of persons affected by environmental pollutants and identify areas that may need environmental mitigations. Clinicians can use geocoded environmental data to coach individual patients on risk mitigation or to tailor treatment. Reports of symptom constellations can help public health authorities to recognize potential epidemics or toxic exposures much earlier than in the past. Conversely, local immunization registries can be used to feed immunization history records to all local EHRs to know about vaccinations, and other registries can be created to identify medication adherence or interpersonal violence reports.
Public health departments or community agencies are often in the best position to address certain problems, such as food insecurity, lack of housing, and social isolation. The manners in which social and behavioral domains may be addressed fall far outside the typical interventions found in health care. For example, food insecurity may be alleviated by access to government-funded food assistance programs, but patients may need help in navigating the enrollment process. Individuals may benefit from health interventions such as group visits, but some may also need community-level interventions. Box 6-4 describes a promising initiative by an organization to address basic resource needs.
Data in the EHR can also help public health departments to assess the success of community interventions in areas such as increasing physical activity, improving diet, and substance use issues. A better understanding of the smoking prevalence, exercise levels, and dietary habits in a community would enhance development of interventions to decrease community-level cardiovascular disease. This additional data may also help in understanding transmission of communicable diseases. For example, diseases spread by air droplets (e.g., tuberculosis) are more likely to be transmitted in areas where people are living closely together. As mentioned in earlier chapters, public health agencies can use geocodable data to create neighborhood and community health information maps that overlay information on health outcomes (e.g., obesity, diabetes, cardiovascular disease) with neighborhood characteristics (e.g., walkability, food index scores, poverty level).
Health Leads Connections to Community Resources
Innovative groups like Health Leads, headquartered in Boston, Massachusetts, work to enable clinical health teams to “prescribe” basic resources like food and heat just as they do medication. They recruit and train college students to “fill” these prescriptions by working with patients to connect them with the basic resources they lack. Health Leads receive referrals by clinical teams, which are also recorded in the patient’s EHR. By completing a full intake with patients to see what their needs are, they are able to work with resources in the community to address those needs. In the case of food insecurity, Health Leads may direct a patient in need of provisions to a food pantry and will follow up to see if the patient went and received food. If not, they will seek out additional resources until the patient’s needs are met.
SOURCE: Tirozzi, 2014.
Using of Geocoding for Supporting Public Health Surveillance
of Social and Behavioral Determinants of Health
The Denver Public Health Department is working on a project called the Colorado Health Observation Regional Data Service (CHORDS), with a goal to support public health surveillance and engage with communities. Using CHORDS, they are able to extract body mass index (BMI) data from electronic health records (EHRs) from various partners, using minimal data to protect patient privacy. This information is then geocoded—including the demographic data with the exact location, which can then be linked to census tract such as income and other social and environmental data. The data allow the Denver Public Health Department to superimpose factors such as walkability, availability of food, restaurants, green space, and poverty on top of the BMI information from EHRs. This information can be used to create specific maps, such as percentage of child obesity. Creating this registry allows communities to examine the health issues in their own neighborhoods and gives the public health community insight into population health.
The Denver Public Health Department also hopes to implement personal prescriptions using community resources. An example of this would be to create a walking map for an individual in their own neighborhood, highlighting the route as well as the health effects such as calories burned. This resource can also use facilities within the community, such as alerting community members to exercise classes near their home.
SOURCE: Davidson, 2014.
This information can be linked back to the health care systems and clinical teams informing them of how well or how poorly the populations they are serving are doing. Box 6-5 details the Denver Public Health Department’s use of geocoding from the EHRs to engage communities.
Patients need to understand the role of public health agencies and the links that the agencies have to one’s clinical team. Patients are not fully aware of the responsibilities of local health departments and may be surprised to discover that their clinical information (e.g., notification of a communicable disease) has been shared with the health department. They might feel that their doctor has compromised the confidentiality of their health record if they receive a call from the health department asking them about their history of a contagious disease, such as a specific food-borne illness. These risks can be mitigated by ensuring that patients are notified about shared data and the roles that health departments play in safeguarding community health. While data regarding domains in this report are more likely to be used in the aggregate, there may be concerns about the sharing of and use of this information.
In some instances, the introduction of the EHR has led to unintended consequences (Ash et al., 2004), including increases in medication errors and data entry failures. Adding social and behavioral domains to the EHR may aggravate existing unintended consequences as well as create new ones. As described earlier, most data (but not all) will be provided by patient self-report, but there will be a need to provide assistance or accommodate patient preferences in doing such. There will also be a change in the clinical workflow that requires the clinical team to verify and interpret rather than simply acquire information. Additionally, the inclusion of new screening tools may inadvertently lead clinicians to minimize or skip previously well-established parts of the clinical evaluation. For example, the use of a three-question screen of alcohol abuse might deter a clinician from undertaking a complete alcohol history. As with any change in clinical information flow, careful planning can mitigate some unintended consequences, and constant surveillance and evaluation are needed to detect those that were not anticipated.
The ultimate value of incorporating the social and behavioral domains of health in the EHR lies in engaging the patient and aligning health service and care. Such redesign is a long-term answer to facing and addressing the implementation challenges summarized in this chapter. The barriers and suggested interventions highlighted are intended to act as a reference to guide stakeholders along this journey.
AHRQ (Agency for Healthcare Research and Quality). 2014. A robust health data infrastructure. AHRQ publication no. 14-0041-EF. Washington, DC: AHRQ.
Ash, J. S., M. Berg, and E. Coiera. 2004. Some unintended consequences of information technology in health care: The nature of patient care information system-related errors. Journal of the American Medical Informatics Association 11(2):104–112.
Boonsta, A., and M. Broekhhis. 2010. Barriers to the acceptance of electronic medical records by physicians from systematic review to taxonomy and interventions. BMC Health Services Research 10(231):10.1186/1472-6963-1110-1231.
Bresnick, J. 2013. Top EHR vendors join Commonwell Alliance to boost interoperability. EHR Intelligence. http://ehrintelligence.com/2013/03/04/top-ehr-vendors-join-commonwell-alliance-to-boost-interoperability (accessed August 5, 2014).
Crane, H. 2014. Comments from Heidi Crane from the Madison HIV Metabolic Clinic on HIV clinical care, assessment, and electronic health records to the IOM Committee on Recommended Social and Behavioral Domains and Measures for Electronic Health Records. Washington, DC: Institute of Medicine.
Crane, H. M., W. Lober, E. Webster, R. D. Harrington, P. K. Crane, T. E. Davis, and M. M. Kitahata. 2007. Routine collection of patient-reported outcomes in an HIV clinic setting: The first 100 patients. Current HIV Research 5(1):109–118.
Davidson, A. 2014. Linking EHRs between public health departments, social service agencies, and other relevant organizations: How to create information systems with data that flow both ways. Presentation on Denver Public Health, Denver Health to the Committee on Recommending Social and Behavioral Domains and Measures for Electronic Health Records. Washington, DC: Institute of Medicine.
ESA (Economics and Statistics Administration) and NTIA (National Telecommunications and Information Administration). 2011. Exploring the digital nation. Computer and Internet use at home. Washington, DC: U.S. Department of Commerce.
Estabrooks, P. A., M. Boyle, K. M. Emmons, R. E. Glasgow, B. W. Hesse, R. M. Kaplan, A. H. Krist, R. P. Moser, and M. V. Taylor. 2012. Harmonized patient-reported data elements in the electronic health record: Supporting meaningful use by primary care action on health behaviors and key psychosocial factors. Journal of the American Medical Informatics Association. doi:10.1136/amiajnl-2011-000576. http://jamia.bmj.com/content/early/2012/04/16/amiajnl-2011-000576.abstract (accessed April 3, 2014).
Fredericksen, R., B. J. Feldman, P. K. Crane, J. Tufano, R. D. Harrington, S. Dhanireddy, T. Davise, T. D. Brown, and M. M. Kitahata. 2011. The impact of same-day pre-visit patient-reported outcome (PRO) collection on provider assessment of sexual risk and other behaviors of HIV-infected patients in routine clinical care. Paper presented at 6th International Conference on HIV Treatment and Prevention Adherence, May 22–24, 2011, Miami, FL.
Fredericksen, R., P. K. Cran, J. Tufano, J. Ralston, S. Schmidt, T. Brown, D. Layman, R. D. Harrington, S. Dhanireddy, T. Stone, W. Lober, M. M. Kitahata, and H. M. Crane. 2012. Integrating a web-based, patient-administered assessment into primary care for HIV-infected adults. Journal of AIDS and HIV Research 4(2):47–55.
Fridsma, D. 2013. EHR interoperability: The structured data capture initiative. HealthIT Buzz. http://www.healthit.gov/buzz-blog/electronic-health-and-medical-records/ehr-interoperability-structured-data-capture-initiative (accessed August 5, 2014).
Friedber, M. W., P. G. Chen, K. R. Van Busum, F. M. Aunon, C. Pham, J. P. Caloyeras, S. Mattke, E. Pitchforth, D. D. Quigley, R. H. Brook, F. J. Crosson, M. Tutty, F. J. Crosson, and M. Tutty. 2013. Factors affecting physician professional satisfaction and their implications for patient care, health systems, and health policy. Santa Monica, CA: RAND Corporation.
Glasgow, R. E., R. M. Kaplan, J. K. Ockene, E. B. Fisher, and K. M. Emmons. 2012. Patient-reported measures of psychosocial issues and health behavior should be added to electronic health records. Health Affairs 31(3):497–504.
Gottlieb, L., D. Hessler, D. Long, A. Amaya, and N. Adler. In review. Maximizing caregiver social needs disclosure in an urban pediatric emergency department: The iScreen Study. Paper presented at Innovations in Family Medicine. 10th Annual UCSF Department of Family Medicine Rodnick Colloquium, May 22, 2014, San Francisco, CA.
Greenhalgh, J., A. F. Long, and R. Flynn. 2005. The use of patient reported outcome measures in routine clinical practice: Lack of impact or lack of theory? Social Science & Medicine 60(4):833–843.
HealthData.gov. 2014. Data: Medicare. http://www.healthdata.gov/data/dataset/cms-medicare-and-medicaid-ehr-incentive-program-electronic-health-record-products-used (accessed July 8, 2014).
Hsiao, C.-J., and E. Hing. 2012. Use and characteristics of electronic health record systems among office-based physician practices: United States, 2001–2012, NCHS data brief. Atlanta, GA: Centers for Disease Control and Prevention.
Hsiao, C.-J., E. Hing, and J. Ashman. 2014. Trends in electronic health record system use among office-based physicians: United States, 2007–2012. Atlanta, GA: National Center for Health Statistics.
IOM (Institute of Medicine). 2012. Health IT and patient safety: Building safer systems for better care. Washington, DC: The National Academies Press.
Junhua, L., A. Talaei-Khoei, H. Seale, P. Ray, and C. R. MacIntyre. 2013. Health care provider adoption of eHealth: Systematic literature review. Interactive Journal of Medical Research 2(1):e7.
Lawrence, S. T., J. H. Willig, H. M. Crane, J. Ye, I. Aban, W. Lober, C. R. Nevin, D. S. Batey, M. J. Mugavero, C. McCullumsmith, C. Wright, M. Kitahata, J. L. Raper, M. S. Saag, and J. E. Schumacher. 2010. Routine, self-administered, touch-screen, computer-based suicidal ideation assessment linked to automated response team notification in an HIV primary care setting. Clinical Infectious Diseases 50(8):1165–1173.
Lohr, K., and B. Zebrack. 2009. Using patient-reported outcomes in clinical practice: Challenges and opportunities. Quality of Life Research 18(1):99–107.
Min, H., R. Ohira, M. A. Collins, J. Bondy, N. E. Avis, O. T chuvatkina, P. K. Courtney, R. P. Moser, A. R. Shaikh, B. W. Hesse, M. Cooper, D. Reeves, B. Lanese, C. Helba, S. M. Miller, and E. A. Ross. 2014. Sharing behavioral data through a grid infrastructure using data standards. Journal of the American Medical Informatics Association 21(4):642–649.
Moukheiber, F. 2014. Group of electronic health record vendors to become officially interoperable. Forbes. http://www.forbes.com/sites/zinamoukheiber/2014/06/10/group-of-electronic-health-record-vendors-to-become-officially-interoperable (accessed August 5, 2014).
NCCOR (National Collaboration on Childhood Obesity Research). 2014. About. http://nccor.org/about/index.php (accessed September 8, 2014).
NIH (National Institutes of Health). 2006–2012. NIH Toolbox. What and Why. http://www.nihtoolbox.org/Pages/default.aspx (accessed September 8, 2014).
NIH. 2012. PhenX Toolkit. Washington, DC: National Human Genome Research Institute. http://www.genome.gov/27541903 (accessed September 8, 2014).
NRC (National Research Council). 2009. Computational technology for effective health care: Immediate steps and strategic directions. Washington, DC: The National Academies Press.
Ockene, J. K., E. A. Edgerton, S. M. Teutsch, L. N. Marion, T. M. Miller, J. L. Genevro, C. J. Loveland-Cherry, J. E. Fielding, and P. A. Briss. 2007. Integrating evidence-based clincal and community strategies to improve health. American Journal of Preventive Medicine 32(3):244–252.
ONC (The Office of the National Coordinator for Health Information Technology). 2012. Beacon community and EHR vendor collaboration: A catalyst for interoperability and exchange. Washington, DC: ONC and HHS. http://www.healthit.gov/sites/default/files/pdf/ehr-vendor-beacon-topic.pdf (accessed August 5, 2014).
Parker, G., C. Turner, K. Koch, J. Costich, K. Coyle, W. Baker, A. Petak, S. Carter, R. McDonald, A. Greenberg, C. Throop, V. Prescott, K. O’Neill, N. Bashyam, and S. Reynolds. 2014. Behavioral Health Data Exchange Consortium. ONC State Health Policy Consortium Project. Final report. Washington, DC: ONC and HHS.
PROMIS (Patient Reported Outcomes Measurement Information System) Network. no date. PROMIS overview. Rockville, MD: PROMIS Network. http://www.nihpromis.org/about/overview (accessed August 5, 2014).
Sinsky, C. A., J. W. Beasley, G. E. Simmons, and R. J. Baron. 2014. Electronic health records: Design, implementation, and policy for higher-value primary careers for higher-value primary care. Annals of Internal Medicine 160(10):727–728.
Snyder, C., N. Aaronson, A. Choucair, T. Elliott, J. Greenhalgh, M. Halyard, R. Hess, D. Miller, B. Reeve, and M. Santana. 2012. Implementing patient-reported outcomes assessment in clinical practice: A review of the options and considerations. Quality of Life Research 21(8):1305–1314.
Tirozzi, K. 2014. Obstacles in adding measures to EHRs and ways to overcome these for the patient, provider, system, and society. Presentation by Health Leads to the Committee on the Recommended Social and Behavioral Domains and Measures for Electronic Health Records. Washington, DC: Institute of Medicine.
Tufano, J., R. Fredericksen, S. Schmidt, R. Harrington, W. Lober, M. Kitahata, J. Ralston, and H. M. Crane. 2010. Evaluating integration of HIV medication adherence computer-assisted self-administered interview (CASI) with routine patient care. Paper presented at 5th International Conference on HIV Treatment Adherence, May 23–25, 2010, Miami, FL.
Valderas, J. M., A. Kotzeva, M. Espallargues, G. Guyatt, C. E. Ferrans, M. Y. Halyard, D. A. Revicki, T. Symonds, A. Parada, and J. Alonso. 2008. The impact of measuring patient-reported outcomes in clinical practice: A systematic review of the literature. Quality of Life Research 17(2):179–193.