Literature Review and Environmental Scan: Evaluation of Personal Health Records Pilots for Fee-for-Service Medicare Enrollees. Impact on Providers


Providers have different attitudes toward and experiences with PHRs. As with all health information technologies, some providers are more receptive than others. Physicians who are early HIT adopters may incorporate the latest technology into their practices, while other segments of the provider community remain resistant to change. Some providers see the potential utility presented by PHRs, particularly in the areas of patient engagement and chronic disease management.[251], [252] Specifically, PHRs create the opportunity for providers to engage patients outside of traditional health care settings and empower them with additional information.[253], [254] Despite the potential benefits, the broader provider community has not played a central role in personal health record adoption efforts.

Some closed systems such as Kaiser Permanente and the Veterans Health Administration have engaged their providers in the implementation and ongoing use of PHR systems.[255], [256] Preliminary studies of these implementations suggest positive results, yet closed networks are atypical as most payers do not directly employ physicians.[257], [258]

Without the type of support and mandate created by closed systems, physicians are left with many barriers that could prevent their active support for the adoption of PHRs. These barriers include the challenge of integrating large volumes of patient data into their clinical workflow and decision making processes, discerning the accuracy of patient-provided data, connecting PHR systems with existing EHR systems, legal liability risks related to incorporating PHRs in the delivery of care and the issues surrounding the creation of new electronic communication channels with patients.

Provider membership associations such as the American Medical Association (AMA) remain skeptical of PHRs. One AMA trustee expressed hope that PHRs could save time for both patients and providers, help prevent medication errors, and prevent duplicate laboratory tests.[259] However, legal concerns and the notion that PHRs would simply amount to ‘garbage in, garbage out’ prevent the AMA from fully embracing PHRs.[260]

The extent to which the concerns expressed by some providers and administrators are typical of the entire provider community is not easily verified in the PHR literature. For example, there are evaluations of PHR use that have shown very high levels of satisfaction with PHRs among providers.[261], [262] One health care network even received more complaints from providers who were not initially given access to a PHR than those who were given access.[263] These early experiences with PHRs suggest that providers attitudes towards PHRs may evolve based on first-hand experience and the ongoing evolution of PHRs, HIT and health care as a whole.

Key Factors that Affect Provider Adoption of PHRs

The successful expansion of PHRs is largely dependent upon their uptake and adoption by providers. Tang et al. (2006) identified a number of potential benefits to providers who utilize PHRs.[264] Namely, PHRs can improve communication between the provider and patient; improve communication between the provider and members of their health care team; and enable patients to input more information about their health directly into a health record that can be connected with the clinician’s EMR. PHRs may also help to foster dialogue between the patient and provider about chronic conditions and medication regimens. Tang et al. (2006) have found, however, that the benefits of the PHR for the provider are contingent upon the PHR being integrated into the provider’s EMR. [265]

Despite these potential benefits, there are a number of key factors that affect provider adoption of PHRs – some are enablers and others are barriers.

Tang (2006) cited a number of barriers to provider participation in PHRs including: concerns about liability risks; reliance on proprietary systems; lack of reimbursement or financial incentives to adopt PHRs; concerns about increased workload and negative impacts on provider workflow; costs associated with staff training and implementation of PHRs; uncertainties about the regulations underlying PHRs; fear of cultural change in the organization; and an absence of information about the sustainability of PHRs.[266]

Impact of PHRs on Patient–Provider Interactions

Currently, most patients play a relatively passive role in their healthcare, relying on providers and payers to track clinical information over time. PHRs, along with other patient-centric technologies, could significantly alter this dynamic. As one commentator put it, ‘Providers and patients will need to develop different mindsets and levels of trust [in working with PHRs].’[267]

In some ways, PHRs represent a departure from traditional patient-provider interactions. For example, providers can use the patient-provided medical data from the PHR to inform their treatment decisions. Patients have the ability to hide certain aspects of their medical history from providers. Some might argue that PHRs provide individuals with more direct control over their healthcare. Further, many PHRs include secure messaging between patients and providers, opening a new mode of communication that may dramatically affect relationships. These potential impacts on the patient-provider relationship are discussed below in further detail.

Provider Access to Patient Generated Clinical Information

PHRs allow providers to access patient health data in new ways. For example, PHRs could provide data that would eliminate the onus of filling out and filing “clip board” questionnaires at a provider’s office. A single electronic registration would make it easier for individuals to provide their information and for clinicians to use it. The American Health Information Community’s (AHIC) Consumer Empowerment Work Group (CEWG) is making recommendations that will guide the development of a consumer-directed secure electronic registration summary; AHIC’s CEWG Work Group has also been charged with facilitating the widespread adoption of PHRs. With a registration summary, the consumer will be able to complete his/her medical paperwork one time electronically and share the information with all of his/her health care providers.[7]

In 2006, AHIC CEWG tasked CMS with piloting programs that measure and demonstrate the value of a Registration Summary and Medication History PHR for patients with chronic conditions and their clinicians. To meet the goals of the CEWG, CMS has contracted with AHRQ to evaluate the uptake, use, and utility of the pilot tool for Managed Care beneficiaries.[268]

In addition to a registration summary, there are tracking tools for health indicators such as blood pressure and blood glucose levels that could show the provider a more complete picture of a patient’s health status between visits.

Although the ability for providers to access these new data may be promising, the sheer volume of information included in PHRs could also have negative consequences.[269] Some in the provider community anticipate patients bringing large volumes of self-entered health data to office visits.[270] While often cumbersome and repetitive, clipboard questionnaires allow providers to control the types of data provided by patients from the very first moment of contact. PHRs theoretically reverse that relationship, empowering consumers to bring personal data in various forms and in limitless quantity to their in-person visits.

In addition, some commentators suggest that providers could be held legally responsible for the material in a patient’s PHR.[271], [272] Providers fear being held legally liable for overlooking key information in patients’ PHR medical histories, even if this information may not ultimately prove relevant to their delivery of care.[273] Conversely, some suggest that providers could be held legally negligent if they provide low-quality care, based on inaccurate patient-provided PHR data.[274] In this way, PHRs present providers with a dilemma: they must either ignore what they consider to be questionable medical data and be sued if it turns out to be accurate and significant, or use inaccurate patient data and be held responsible if the data dictate inappropriate or dangerous care. Similarly, providers need to know the origin of information; whether it is from a fellow provider or originally from a provider and since modified by the patient. This issue has not been consistently addressed across all PHRs.[275]

Some findings suggest that the utility and relevance of PHR data to providers varies by situation and the type of care needed. For example, LifeLedger and other PHR products offer a wallet-sized reference card to direct providers to the PHR’s URL.[276] Directing providers to the PHR using this method could be useful in situations where patients are not able to communicate with providers. Some approaches have proven less effective in care settings. In a study of a Minnesota emergency room, patients could not recall their PHR user name or passwords, rendering the PHR system useless in emergency situations.[277] Other PHRs offer specific reports that patients can print out and carry in their purses and wallets.[278] This type of report could include information relevant in an emergency (such as drug allergies, blood type, advanced directives, etc.), but would exclude many classes of data housed in the PHR. One study suggests that this type of report has proven very useful for emergency medical technicians.[279]

There are also examples of patients utilizing PHRs that allow them to add their own annotations of official clinical EHR records. Indivo, an open source personally controlled health record application that was developed at Children’s Hospital in Boston, allows patients and caregivers to annotate the record in a controlled manner. At the time of this review, Indivo is in use at Children’s Hospital Boston and other testbeds. Users are allowed to add to a record, but are not allowed to delete any content. Information added by the user is clearly marked so providers and other caregivers accessing the record have strong assurance regarding the veracity of the information they use to guide treatment decisions. Indivo is being adapted for use by the Dossia Consortium. The source code for Indivo is available for free download and use under the LGPL license.

Finally, while PHRs allow providers to access patient health data in new ways, the same risks of inaccurate and incomplete information apply. Users do have control over the content of their PHRs, and in some cases, can add or delete information from their medical history. Thus, PHRs are not entirely dissimilar from a patient’s verbally reported medical history. Archelle Georgiou, M.D., an independent consultant, noted that, ‘Just as the doctor uses a verbal medical history as a component of the care decision-making process, the doctor should use the PHR as a valuable data point in care planning. However, PHRs should never be used to make health care decisions. The risks of a poor patient history are the same risks of inaccurate or incomplete data in a PHR.’ Thus, PHRs should be used to inform – rather than to determine – the provider’s treatment decisions.

Secure Communication with Patients

Many PHRs offer users the capability to directly communicate with their primary care providers. In the words of Paul Tang, this function ‘provide[s] an ongoing connection between patient and physician…which changes encounters from episodic to continuous,’ something Tang cited as a ‘critical benefit.’[280] To date, PHRs have largely adopted a secure message functionality that provides greater security assurances and controls than standard email. Research comparing secure messaging with standard email in health care settings shows that messaging provides a number of significant advantages over email in areas such as: encryption, access controls, message templates, prescription routing, and reimbursement.[281]

Many argue that secure messaging capabilities could be abused by patients and that it could pose a liability if users misunderstand the purpose of provider messaging.[282] In many cases, providers express concern about the sheer number of messages and the workflow changes that would be required to answer all messages in a responsible way. These potential negatives represent significant barriers to adoption for providers.

However, some empirical studies show that these concerns are not borne out by the evidence. One study of a secure message system using a patient portal showed that message volume peaked at 8.5 weekly messages per 100 scheduled visits.[283] Studies of secure-messaging done at Beth Israel Deaconess Medical Center (BIDMC) further demonstrate that physicians are not inundated with patient messages, with an average of 20 messages each month per 100 patients 57 These results suggests that access to secure messaging does not automatically lead to abuse or unmanageable message volumes, a potential barrier to adoption. Further, messaging may decrease over time as patients become better acquainted with PHRs and their functionality.[284] In this way, patients may learn more about the appropriate topics for messages and learn to use the message function more efficiently over time. Another study found that providers who receive more messages found integrating message-related work easier than providers who received fewer messages.[285] This finding suggests that message volume may not be a primary workflow obstacle in PHR implementation. Alternatively, provider comfort with PHRs or commitment to using them could cause greater problems for providers.

Anecdotal evidence further suggests that patients respect their providers’ time, and generally do not burden them with unnecessary messages.[286] An independent study of Kaiser’s KP HealthConnect shows that 70% of user sessions result in a message between the patient and provider.[287] This demonstrates that KP HealthConnect and its users have fully embraced secure messaging. At the same time, positive feedback from Kaiser Permanente providers suggests that this embrace has not resulted in a flood of clinically irrelevant messages.[288] Kaiser has also found that the secure message function is more efficient than telephone communication.[289] Additional PHR studies support this idea, although there is not yet a clear consensus in the literature.[290] [291] [292] [293]

Recent implementations of PHRs have created approaches to implementing secure messaging in a manner that could be more palatable to providers. Kaiser Permanente empowers providers to choose the patients they think will benefit most from this form of communication.[294] In this system, the PHR amounts to a tool for patients over which providers retain considerable control. This type of solution may convince skeptical providers who may be resistant to new information systems, especially where they can identity no direct benefit.[295]

Research has shown that successful electronic communication does not have to directly connect patients with their physicians. Kaiser Permanente has given each of its geographic regions the authority to decide which types of providers are authorized to read and respond to patient messages sent through HealthConnect.[296] This means that registered nurses or other types of providers can handle messages, in addition to physicians. This sort of compromise could allay physician fears of being overwhelmed by patient messages.

Other examples in the literature show that sites that utilize secure messaging often have established processes for triaging messages so that office administrative staff deal with scheduling of appointments, nurses and physician assistant staff respond, to some patient email and providers are only expected to respond to email that requires their specific involvement. In addition, most practices establish turn around times for responding to messages, ensuring that patient expectations are appropriately managed. Patients are also informed about the circumstances in which email communication should be used.

View full report


"litreview.pdf" (pdf, 998.47Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®