NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Goldzweig CL, Towfigh AA, Paige NM, et al. Systematic Review: Secure Messaging Between Providers and Patients, and Patients’ Access to Their Own Medical Record: Evidence on Health Outcomes, Satisfaction, Efficiency and Attitudes [Internet]. Washington (DC): Department of Veterans Affairs (US); 2012 Jul.

Cover of Systematic Review: Secure Messaging Between Providers and Patients, and Patients’ Access to Their Own Medical Record

Systematic Review: Secure Messaging Between Providers and Patients, and Patients’ Access to Their Own Medical Record: Evidence on Health Outcomes, Satisfaction, Efficiency and Attitudes [Internet].

Show details


Are the objectives, scope, and methods for this review clearly described?The dates of the literature review are stated as “1999 through 12/03/2010”…to improve consistency provide start and end dates in same format (e.g. 01/01/1999 through 12/03/2010).This suggestion has been incorporated.
Are there any published or unpublished studies that we may have overlooked ?The exclusion of descriptive qualitative studies appears to be a shortcoming considering the infancy of this field and the limited availability of data about the use and efficacy of Secure Messaging, Personal Health Records, and Web-based Management Systems. The inclusion of at a minimum a summary of the qualitative findings to date could provide insights to the facilitators and barriers to use and possibly inform the “why” factor to these reported quantitative findings.We evaluated the descriptive qualitative papers and identified those that were potentially studies of patient portals that discussed barriers and facilitators. We identified four such studies for inclusion, two of which were of the same system, that we felt were reasonable to include. A fifth study involved focus groups evaluating a “potential” system and was not included. A sixth study evaluated patient opinions regarding access to records from a kiosk in the practice waiting room in London and was not included.
J Gen Intern Med. 2003 Sep;18(9):736–44.
Effect of a triage-based E-mail system on clinic resource use and patient and physician satisfaction in primary care: a randomized controlled trial.
The email system presented does not match our definition of secure messaging. Insecure email was outside the purview of our current scope.
Randomized Trials:
1. McCarrier KP, Ralston JD, Hirsch IB, et al. Web-based collaborative care for type 1 diabetes: a pilot randomized trial. Diabetes Technol Ther. Apr 2009;11(4):211–217. (U of Washington Study)McCarrier: Our original search only found the companion to this article. We have now included this article as well.
2. Simon GE, Ralston JD, Savarino J, Pabiniak C, Wentzel C, Operskalski BH. Randomized trial of depression follow-up care by online messaging. J Gen Intern Med. 2011. (Group Health Study)Simon and Lyles: These articles were too recent for our original search, however our updated search did include them. They have now been incorporated into the report.
Papers Addressing Potential Disparities in Access to SM and Patient Access to Records:
1. Lyles CR, Harris LT, Jordan L, et al. Patient race/ethnicity and shared medical record use among diabetes patients. Med Care. 2011. (Group Health)
2. Roblin DW, Houston TK, 2nd, Allison JJ, Joski PJ, Becker ER. Disparities in use of a personal health record in a managed care organization. J Am Med Inform Assoc. Sep–Oct 2009;16(5):683–689. (Kaiser)Roblin: This article has now been included in a new section on patient characteristics associated with use of a patient portal/tethered PHR.
3. Sarkar U, Karter AJ, Liu JY, et al. The literacy divide: health literacy and the use of an internet-based patient portal in an integrated health system-results from the diabetes study of northern California (DISTANCE). J Health Commun. 2010;15 Suppl 2:183–196. (Kaiser)Sarkar: We have included this article in a new section on patient characteristics associated with use of a patient portal/tethered PHR.
Wald JS, Grant R, Schnipper J, Gandhi T, Poon E, Businger A, Orav E, Williams D, Volk L, Middleton B. Survey analysis of Patient Experience using a Practice-linked PHR for Type 2 Diabetes Mellitus. AMIA Annu Symp Proc 2009:678–82.Wald 2009: This article reports on a subset of patients that are included in the Grant 2008 article below and the previously included Wald 2010 artucle, and is thus an exclude.
{post 12/3/2010}Wright A, Poon EG, Wald J, Feblowitz JC, Schnipper JL, Grant RW, Gandhi TK, Volk LA, Bloom A, Williams DH, Gardner K, Epstein M, Nelson L, Businger A, Li Q, Bates DW, Middleton B. Randomized controlled trial of health maintenance reminders provided directly to patients through an electronic PHR. J Gen Int Med 2012 Jan; 27(1):85–92. Epub 2011 Sep 9. http://www​.ncbi.nlm.nih​.gov/pubmed/21904945Wright: This article is about reminders, which was not part of the revised focus for this report.
{post 12/3/2010}Yamin CK, Emani S, Williams DH, Lipsitz SR, Karson AS, Wald JS, Bates DW. The digital divide in adoption and use of a personal health record. Arch Int Med 2011; 171(6):568–574. {This paper may be relevant since many studies are limited by selection bias in the study participants.}Yamin: This article was too recent for our original search, but was captured by our search strategy in an update search, and is now included.
Grant RW, Wald JS, Schnipper JL, Gandhi TK, Poon EG, Orav EJ, Williams DH, Volk LA, Middleton B. Practice-linked Online Personal Health Records for Type 2 Diabetes: A Randomized Controlled Trial. Arch Int Med 2008; 168(16):1776–82.Grant 2008: This article is now included.
Grant RW, Wald JS, Poon EG, Schnipper JL, Gandhi TK, Volk LA, Middleton B. Design and implementation of a web-based patient portal linked to an ambulatory care electronic health record: patient gateway for diabetes collaborative care. Diabetes Technol Ther. 2006; 8:576–86.Grant 2006: Although our search did capture this article, it was marked as an exclude because it was descriptive qualitative and did not address barrier or facilitators to use.
Wald JS. Variations in Patient Portal Adoption in Four Primary Care Practices. AMIA Annu Symp Proc 2010:837–41. {Supports the idea that it may be difficult to draw strong conclusions from practices/patients where adoption is weak; the implication for future VA work is to establish “level of adoption” metrics for comparability.}Wald 2010: This paper is a descriptive qualitative paper that was included in our search and that is now included in the new section on barriers and facilitators.
The Value of Personal Health Records. David C. Kaelber, MD, PhD, Sapna Shah, MS, Adam Vincent, MPP, Eric Pan MD, MSc, Julie M. Hook, MA, MPH, Doug Johnston, MTS, David W. Bates, MD, MSc, Blackford Middleton, MD, MPH, MSc. © 2008 by the Center for Information Technology Leadership (CITL). Published and distributed by the Healthcare Information and Management System Society (HIMSS). Requests for permission to reproduce any part of this work should be directed to: Ellen S. Rosenblatt, Manager of Operations Center for Information Technology Leadership Partners HealthCare System, Inc. One Constitution Center Information Systems Department, Second Floor West Charlestown, MA 02129 gro.srentrap@ttalbnesore. ISBN: 978-0-9800697-4-7Kaelber: This was already included, and is described in the “Patient Access and Efficiency/Utilization.”42
Not including search terms such as ‘personal health record’, ‘patient portal’, ‘secure email’ and ‘text messaging’ may have reduced identification of potentially appropriate papers.We will incorporate this suggestion into any future updated search, however for this report we rely on the original search and reference mining of included articles and the review by experts to identify potentially important missing studies. In the peer review process just completed, only three such articles (out of more than 60 already included) were identified, supporting a conclusion that the number of additional relevant studies now already identified is likely to be very small (<5%).
Please write additional suggestions or comments below. If applicable, please indicate the page and line numbers from the draft report.There are some run-on sentences in the introduction, page 5.We have edited the introduction.
Recommend tabling key research question results and “GRADES”.We have included such a table in the Summary and Discussion section.
From the automated email, I think there is a lot more research in the health behavior literature, Wayne Velicer, Vic Strecher, and many others have done this. Some has even extended to text messaging. Thus, I think that this section is not complete.Because this was not the focus of the review, and based on the likelihood that our search did not adequately identify much of the literature in this area, we have removed this section from the report.
Please correct all instances of My HealtheVet to be branded as shown here: “My HealtheVet”This change has been incorporated.
It seems that there is logic to the order in which the GRADE of evidence is presented but it is unclear from the report what that logic is? It is clearly not higher to lower grades, but I wondered why? E.g. p 4The order was based on the order of the outcomes in the key questions, and then within each outcome the evidence is sorted with GRADE going from high to low. A new table has been added for clarification.
I find it very surprising that there is insufficient evidence to reach conclusions about the effect of patient access to their own medical record on their attitudes. Is this because most attitudinal studies were not examined since the scope was on hypothesis testing? Should that be qualified if that’s the case?In order to reach conclusions about satisfaction, we required a study to statistically test satisfaction, either between groups (with and without access to their own medical records) or across time (before and after access to their own medical records). Without these data, we can reach only limited conclusions. We started from the position that in order to reach cause-and-effect conclusions a hypothesis would need to be stated and then tested.
I am finding that the comparison of use of PHR portal alone versus PHR portal WITH Secure Messaging has been useful in my own work. Instead the report compares SM with portal versus SM alone. Given that the model in the industry has evolved from PHR portal to portal WITH SM is there any way to reflect this in your analysis?The original focus of the review was to evaluate secure messaging alone and then to evaluate the area of “patient access to medical records.” In the patient access to medical records section, all of the tethered systems described in the ‘Outcomes, Satisfaction and Adherence” section and all but one in the “Efficiency/Utilization” section include secure messaging as a component of their PHRs. In the one instance where this is not the case, it is specifically noted in the text.30
Page 5: Dr. Nazi’s office is Veterans and Consumers Health Informatics Office/Office of Informatics and AnalyticsThis change has been incorporated.
P17. Missing period in para 2, could RVUs be defined?This change has been incorporated.
Consider further review and discussion of the interdependence of secure messaging and patient medical record access in the studies reviewed. Several of the intervention studies and most of the observational studies were in healthcare systems that intentionally tied together secure messaging and patient access to portions of the electronic medical record. Several of these interventions saw these two functions as interdependent for many patients, particularly for those needing self management support and collaborative care for chronic conditions. Parsing out the individual contribution of one of these activities may be less fruitful than seeing the value of the package.As in the response to the comment above, the synthesis of “patient access to medical records” consisted of interventions that also included secure messaging, so the synthesis of the two interventions is already contained in the report. In addition, in the Summary and Discussion section, we have also discussed this point.
Discuss results in the context of evolving definitions of patient medical electronic medical record access across the studies. For some of these studies, the record is a passive document viewed online by the patients. For others, the record is more interactive for patients and part of the ongoing care and communication tasks many patients face. In the latter case, the record can include secure patient provider messaging, medication refill functionality, and structured health risk assessment and feedback. Although current studies are not sufficient to determine the evidence behind the different approaches to patient access of the medical records, these differences may end up playing a role in outcomes as evidence evolves.In order to inform this issue, we have included more detailed descriptions of the tethered systems described in the studies.
Evaluation and discussion of equity in access to SM and patient access to medical records is missing. Secure messaging and patient access to the medical record should be seen as part of how we deliver care to all patients. Some historically vulnerable and underserved patient populations are less likely to use these services. As the VA and other similar organizations consider implementation of SM and patient access to the record, understanding and addressing these differences is essential for equitable care. I have provided a few references above to consider if the reviewers decided to encompass this domain.We have included a new section entitled “Patient Access and Patient Characteristics” which reviews the evidence relevant to this comment.
Page 15, second paragraph, third to last sentence. The randomized study of patients with diabetes referenced was done at the U of Washington, not Group Health (Ralston et al, Diabetes Care).We thank the reviewer and author of the study for this observation, and have made the correction.
Page 15, 3rd paragraph, in reference to Group Health randomized trial of essential hypertension. The review appears to erroneously imply that the control group did not have access to SM and the electronic medical record. All three study arms, including the usual care arm, had access to SM with PCP and other members of healthcare team and had access to similar portion of the electronic medical record. One intervention arm was given a home blood pressure cuff; the other intervention arm was given the blood pressure cuff and additionally access to pharmacist-based care management. All patients in the study were signed up and had access to the SM and electronic records. The strength of the study is showing how pharmacist care management over SM improves to BP control among patient who have access to SM and the services of the electronic record. May be particularly relevant to the VA’s access efforts for mental health care.We thank the reviewer and author for this clarification and have revised the description of the study and our conclusion.
For Key Question #1, the Grant 2008 paper would justify adding to RESULTS (if authors agree): “There is strength evidence that secure messaging (especially as part of a web-based management system) can improve medication decisions during a subsequent visit, reducing clinical inertia (Grant 2008 Arch Int Med).We have added this conclusion (slightly modified) to the conclusion and given it a GRADE classification of “low” due to sparse data and concern about the generalizability of the intervention and practice settings.
For Key Question #1, the association of secure messaging with many things (pt satisfaction, adherence, outcomes, etc.) is tempered by attitudes, workflow, service orientation, and factors beyond the tool itself. This is touched upon in the conclusion.
Not sure if this “finding” can be considered given the report methodology, but I feel it’s important because these factors are critical for understanding the current evidence and will likely impact future evidence as well. One paper that addresses practice (and other) factors are: Wald JS. Variations in Patient Portal Adoption in Four Primary Care Practices. AMIA Annu Symp Proc 2010:837–41.
We have now incorporated this reference in an expanded discussion of this issue.
Not sure if this synthesis should include any high level comments about why the findings are largely indeterminate. Overall penetration of secure messaging and patient access to health records is still quite low, and given this, a paucity of rigorous data is not unexpected.
Some of the findings may suffer from a ceiling effect…. Meaning that patients practicing greater health engagement with providers and in terms of self-management may be more likely to adopter new technology sooner, making it harder to demonstrate strong improvement.
We have incorporated these into the limitations.
Comments on language: specific language and definitions could be helpful early on, with consistency throughout. Secure messaging, per VA-developed language, refers to secure email defined by online communication between patients and providers or healthcare team members. ‘Messaging system’ is less clear (e.g. page 14), and could include automated email, text or other technology such as interactive voice response (IVR). This review is focused on specific functionality delivered largely through patient portals and/or personal health record systems. Consider briefly defining functions early on; using either secure email or secure messaging, and avoid ‘messaging’ in other instances.We have included definitions and have eliminated the section on automated email to avoid confusion.
Comment on category of Efficiency/Utilization (page 16+): This grouping includes study findings across 3 types of measures: patient-level utilization of care or services delivered, provider-delivered care or workload measures, and patient-level characteristics or factors. The 3rd type of measure, characteristics of users and non-users, is important yet not a component of the category title. Impact on care utilization and provider workload should be distinct from one another.We have now incorporated a new section on patient characteristics.
Comment on Automatic Email Systems (page 19): there are several types of automatic messaging systems, including email notifications, text notification and others, such as IVR. Based on the search terms used, it appears the scoping was for the 1st type of notification only. If so, it would be valuable to describe the types of papers that were excluded. If not, there is some concern that the search terms used may have limited the studies identified.This section has been deleted in this version.
Comment on ‘SM users’: it would be helpful to clarify if a study examined patients who were enrolled or authenticated (identity-proofed) to use the PHR or secure email portal, or, whether actual use of SM was employed to identify the individuals. This is nuanced, but creates differences in study denominators.This is an important distinction, however, most of the primary studies are not clear on this point.
Comments on ‘access to their own record’: there are significant differences in the record content available for patients to view through a PHR. The VA offered access to medications and wellness reminders (at end of 2010); Kaiser and Group Health provided lab results and problem lists; few systems offered access to clinical notes. These distinctions should be described, even if there is insufficient evidence to discern the impact related to specific content or increasing level of health record access.We have included more detailed descriptions of the tethered systems described in the studies.
Page 14: ‘web-based pharmacy group’ – suggest modify to web-based system plus pharmacistWe have modified this description.
Various pages: On-line can be one word: online; Diabetic patients should be patients with diabetes, patients with CHF, etc.These changes have been incorporated throughout the text.
Summary and Discussion (page 27): there appears to be a summary but no discussion.We have now put the conclusions and GRADE information in a table and added text to the discussion.
Limitation: all of these studies have relevance for the VA. Comments about fee-for-service or academic centers raise issues of external validity.We have modified this text.
Conclusion: studies showed patient satisfaction and reportedly improved communication with secure messaging alone. ‘Web-based management program’ could gain specificity with personal health record systems having access to online information and services… allowing patients to participate in their health and healthcare.We have revised this section to include more information about this conclusion.
Please provide any recommendations on how this report can be revised to more directly address or assist implementation needs.Include relevant qualitative/descriptive research findings section/summary.We have incorporated relevant qualitative descriptive studies about barriers and facilitators within the patient access section.
Consider adding to the review a discussion of the evidence for coupling secure messaging and patient access to medical records, particularly for chronic conditions.We have further emphasized these points.


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (998K)

Other titles in this collection

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...