This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Inviting patients to read their primary care visit notes may improve communication and help them engage more actively in their health care. Little is known about how patients will use the opportunity to share their visit notes with family members or caregivers, or what the benefits might be.
Our goal was to evaluate the characteristics of patients who reported sharing their visit notes during the course of the study, including their views on associated benefits and risks.
The OpenNotes study invited patients to access their primary care providers’ visit notes in Massachusetts, Pennsylvania, and Washington. Pre- and post-intervention surveys assessed patient demographics, standardized measures of patient-doctor communication, sharing of visit notes with others during the study, and specific health behaviors reflecting the potential benefits and risks of offering patients easy access to their visit notes.
More than half (55.43%, 2503/4516) of the participants who reported viewing at least one visit note would like the option of letting family members or friends have their own Web access to their visit notes, and 21.70% (980/4516) reported sharing their visit notes with someone during the study year. Men, and those retired or unable to work, were significantly more likely to share visit notes, and those sharing were neither more nor less concerned about their privacy than were non-sharers. Compared to participants who did not share clinic notes, those who shared were more likely to report taking better care of themselves and taking their medications as prescribed, after adjustment for age, gender, employment status, and study site.
One in five OpenNotes patients shared a visit note with someone, and those sharing Web access to their visit notes reported better adherence to self-care and medications. As health information technology systems increase patients’ ability to access their medical records, facilitating access to caregivers may improve perceived health behaviors and outcomes.
Patients are often cared for in the home by informal caregivers. In the United States, approximately 39% of adults are caregivers for an adult or child with significant health issues [
OpenNotes is an initiative that gives patients access to the visit notes written by their doctors, nurses, or other clinicians. In the OpenNotes study, one third of patients with easy Internet access to their primary care doctors’ notes were concerned about privacy [
Health policy encourages patient access to their electronic health information. The Medicare and Medicaid Electronic Health Care Record (EHR) Incentive Program provides incentive payments to eligible professionals and hospitals as they adopt, implement, or show meaningful use of certified EHR technology. Stage 2 of Meaningful Use advocates electronic engagement of patients and their families [
Large medical systems, including the Veterans Administration (VA), Geisinger Health System, Mayo Clinic, Beth Israel Deaconess network, and MD Anderson Cancer Center are offering open records to increasing numbers of patients, and in some circumstances also to their delegates [
Decreasing barriers to communication of important medical information and recommendations with caregivers has the potential to improve patient outcomes and decrease medical errors. The OpenNotes study invited patients in three US states to access their visit notes for one year. Here we evaluate the characteristics of patients who reported sharing their visit notes during the course of the study, including their views on associated benefits and risks.
The study was a quasi-experimental intervention that invited patients to access their primary providers’ visit notes via Web-based, secure patient portals for a year between the summer of 2010 and fall of 2011. Study populations included urban and suburban primary care practices associated with Beth Israel Deaconess Medical Center (BIDMC), an academic health center with urban and suburban practices, Geisinger Health System (GHS) in central, largely rural Pennsylvania, and Harborview Medical Center (HMC), an urban safety-net teaching medical center affiliated with the University of Washington. At HMC, both the general medical clinic and a primary care clinic for patients with human immunodeficiency virus (HIV) participated in the study. Patients were initially surveyed about their expectations at the time of enrollment, and a year later surveys focused on their experiences with the intervention. The details of the study are previously described [
All study procedures were approved by the institutional review boards of BIDMC, GHS, and the University of Washington.
Patients of volunteering primary care physicians were invited electronically to participate at BIDMC and GHS, where portals already existed, and were approached individually at HMC, where an existing electronic health record was modified and made available to study participants [
Pre-intervention baseline surveys assessed patients’ demographic data, including education, self-reported health, how the patients felt about gaining electronic access to visit notes [
Data from the baseline and post-intervention survey were analyzed for participants who viewed at least one visit note and responded “yes” or “no” to the sharing question on the post-intervention survey. The proportion of participants reporting that they showed or discussed their note with someone else during the study (sharers) was compared to those who did not (non-sharers). Sharers and non-sharers were compared by patient characteristics from the baseline survey, including demographics, self-reported health, patient-doctor communication measures, and from the post-intervention data, including number of notes available during the study, frequency of portal use, and behavioral perceptions (better self-care, better adherence to medications, concern about privacy) using chi-square tests and Mann Whitney tests when appropriate.
Modified Poisson regression with robust error variance was used to determine perceived relative risks of sharing notes for each of the aforementioned patient characteristics in univariate models. Characteristics were found statistically to be significantly associated with sharing visit notes were then included in multivariable models. The resulting characteristics associated significantly with sharing visit notes: age, gender, employment status, and study site were incorporated into relative risk models to assess the association between sharing, frequency of portal access, and behavioral perceptions, respectively. Data analyses were performed using SAS software, version 9.3.
Across the three study sites, 22,703 patients were invited to participate, 19,371 (85.32%) completed the intervention, and 11,155 of those (57.59%) had at least one note available during the study period [
Over half (55.43%, 2503/4516) of post-intervention survey respondents agreed that they would like the option of letting family members or friends have their own access to their visit notes. In fact, 21.70% (980/4516) of participants reported showing or discussing their visit note with someone else. Among those that shared their visit notes, the persons with whom they shared included (the survey allowed reporting of more than one individual) a family member, friend, or relative who takes care of them (349/980, 35.61%), another family member (554/980, 56.53%), another friend (95/980, 9.69%), another doctor (87/980, 8.88%), a nurse or health professional (83/980, 8.47%), or someone else (107/980, 10.92%).
Multiple characteristics were significantly associated with sharing visit notes during the intervention in unadjusted analyses: being 60 years of age and older, male, having less than or equal to a high school education, being retired or unable to work, having poor or fair self-reported health, and participating at a study site other than BIDMC (
Characteristics of patients who reported sharing or did not report sharing their visit notes with someone else during the study.
Patient characteristicsa | Did share visit notes | Did not share visit notes |
|
||||
n | % | n | % | ||||
Total number of participants (N=4516) | 980 |
|
3536 |
|
|
||
|
|||||||
|
18-39 | 97 | 9.9 | 488 | 13.8 | <.001 | |
|
40-49 | 154 | 15.7 | 691 | 19.5 |
|
|
|
50-59 | 284 | 29.0 | 1189 | 33.6 |
|
|
|
60-69 | 292 | 29.8 | 849 | 24.0 |
|
|
|
≥70 | 153 | 15.6 | 319 | 9.0 |
|
|
|
|
|
|
|
|
||
|
Women | 466 | 47.6 | 2256 | 63.8 | <.001 | |
|
Men | 514 | 52.5 | 1280 | 36.2 |
|
|
|
|||||||
|
White | 758 | 77.4 | 2781 | 78.7 | .41 | |
|
Black or African American | 15 | 1.5 | 78 | 2.2 |
|
|
|
Other or multiracial | 50 | 5.1 | 169 | 4.8 |
|
|
|
Unknown | 157 | 16.0 | 508 | 14.4 |
|
|
|
|||||||
|
High school/GED or less | 158 | 16.1 | 429 | 12.1 | <.001 | |
|
Some college | 181 | 18.5 | 616 | 17.4 |
|
|
|
College graduate | 106 | 10.8 | 543 | 15.4 |
|
|
|
Post college | 257 | 26.2 | 1047 | 29.6 |
|
|
|
Unknown | 278 | 28.4 | 901 | 25.5 |
|
|
|
|||||||
|
No (Retired/unable to work) | 311 | 31.7 | 746 | 21.1 | <.001 | |
|
Yes (Employed/self-employed/homemaker) | 374 | 38.2 | 1843 | 52.1 |
|
|
|
Unknown | 295 | 30.1 | 947 | 26.8 |
|
|
|
|||||||
|
Poor/Fair | 118 | 12.0 | 323 | 9.1 | .001 | |
|
Good/Very Good | 523 | 53.4 | 2005 | 56.7 |
|
|
|
Excellent | 61 | 6.2 | 308 | 8.7 |
|
|
|
Unknown | 278 | 28.4 | 900 | 25.4 |
|
|
|
|||||||
|
|
|
|
|
|
|
|
|
|
HIV clinic | 21 | 2.1 | 33 | <1 | <.001 |
|
|
Adult medicine clinic | 10 | 1.0 | 12 | <1 |
|
|
GHS | 460 | 49.9 | 1567 | 54.4 |
|
|
|
BIDMC | 489 | 46.9 | 1924 | 44.3 |
|
aPatient characteristics were obtained from the pre-intervention survey, response rate 51.90% (5789/11,155).
b
Patient-doctor interaction and patient confidence in communication with doctor scores for patients who did share or did not share visit notes with someone else during the study.
Measures of interaction | Did share visit notes | Did not share visit notes |
|
||
Mean (SD) | Median (IQR) | Mean (SD) | Median (IQR) | ||
Ambulatory care experiences survey scorea | 5.2 (0.9) | 5.6 (4.8, 6.0) | 5.1 (0.9) | 5.4 (4.6, 5.8) | .009b |
Perceived efficacy of patient-doctor interactions scorec | 42 (7) | 44 (23, 24) | 42 (7) | 43 (23, 24) | .42b |
aPatient report of patient-doctor interactions; range of 0-6, with a higher score indicating better communication.
b
cPatient level of confidence about communicating with his or her physician; range of 5-50, with a higher score indicating more confidence.
Behavioral perceptions of patients who reported sharing their visit notes with someone else during the study (N=4516).
Behavioral perceptions | Did share visit notes (n=980) | Did not share (n=3536) |
|
||||
n | % | n | % |
|
|||
In thinking about what it was like to read your doctor’s visit notes | |||||||
|
|
||||||
|
|
Agree/somewhat agree | 843 | 86.02 | 2737 | 77.40 | <.001 |
|
|
Disagree/somewhat disagree/don’t know | 137 | 13.98 | 799 | 22.60 |
|
|
|
||||||
|
|
Agree/somewhat agree | 698 | 71.22 | 2103 | 59.47 | <.001 |
|
|
Disagree/somewhat disagree/don’t know | 203 | 20.71 | 1058 | 29.92 |
|
|
|
Do not take medications | 79 | 8.06 | 375 | 10.61 |
|
|
|
||||||
|
|
Agree/somewhat agree | 347 | 35.41 | 1345 | 38.04 | .13 |
|
|
Disagree/somewhat disagree/don’t know | 633 | 64.59 | 2191 | 61.96 |
|
a
When demographic, health, and study site characteristics were placed into a multiple adjusted regression model, the characteristics that remained independently associated with sharing visit notes were being male, being retired or unable to work, and attending the general medicine clinic at Harborview Medical Center, an urban safety-net hospital (
Unadjusted and adjusted association between patients who shared visit notes and demographic characteristics, self-reported health, and study site.
Characteristic | Unadjusted RRa,b | 95% CI | Adjusted RRa,b | 95% CI | ||
|
||||||
|
18-39 | 1 |
|
1 |
|
|
|
40-49 | 1.10 | 0.87-1.38 | 0.96 | 0.72-1.28 | |
|
50-59 | 1.16 | 0.94-1.43 | 0.97 | 0.75-1.26 | |
|
60-69 | 1.54 | 1.25-1.90 | 1.09 | 0.83-1.44 | |
|
≥70 | 1.95 | 1.56-2.44 | 1.22 | 0.89-1.66 | |
|
||||||
|
Men | 1.67 | 1.50-1.87 | 1.61 | 1.41-1.85 | |
|
Women | 1 |
|
1 |
|
|
|
||||||
|
High school/GED or less | 1.37 | 1.15-1.62 | 1.19 | 0.97-1.46 | |
|
Some college | 1.15 | 0.97-1.36 | 1.10 | 0.92-1.32 | |
|
College graduate | 0.83 | 0.67-1.02 | 0.86 | 0.70-1.05 | |
|
Post college | 1 |
|
1 |
|
|
|
||||||
|
No (Retired/unable to work) | 1.74 | 1.53-1.99 | 1.39 | 1.18-1.64 | |
|
Yes (Employed/homemaker) | 1 |
|
1 |
|
|
|
||||||
|
Poor/Fair | 1 |
|
1 |
|
|
|
Good/Very good | 0.77 | 0.65-0.92 | 0.88 | 0.74-1.05 | |
|
Excellent | 0.62 | 0.47-0.81 | 0.78 | 0.59-1.04 | |
|
||||||
|
|
|
|
|
|
|
|
|
HIV clinic | 1.92 | 1.36-2.71 | 1.35 | 0.90-2.01 |
|
|
Adult medicine clinic | 2.24 | 1.41-3.57 | 1.67 | 1.01-2.76 |
|
GHS | 1.12 | 1.00-1.25 | 0.96 | 0.82-1.13 | |
|
BIDMC | 1 | 1 |
aEstimates derived from modified Poisson regression with robust error variance.
bAdjusted for age, gender, education, employment, self-reported health and study site.
Behavioral perceptions in patients who shared visit notes.
Behavioral perceptions | Unadjusted RRa | 95% CI | Adjusted RRa,b | 96% CI | |
|
|||||
|
I take better care of myself | 1.61 | 1.36-1.90 | 1.45 | 1.20-1.76 |
|
I do better with taking my medications as prescribed | 1.55 | 1.34-1.78 | 1.49 | 1.25-1.76 |
|
I am concerned about my privacy | 0.91 | 0.81-1.03 | 0.94 | 0.82-1.08 |
aEstimates derived from modified Poisson regression with robust error variance.
bAdjusted for age, gender, employment, and study site.
Health care systems, prodded by policy drivers and consumer demand, are increasingly moving forward with opening records to patients [
A large proportion of persons in the United States are informal caregivers [
Integrating caregivers of frail patients into each step of patient care will likely become increasingly important for providing high-quality and cost-effective care for these medically complex patients [
Accessing medical information via the Internet has been postulated to increase the “digital divide” between those who are facile with technology and those who are not [
Clinicians often record intimate details of patients’ lives. Electronic access to such information for persons outside of the patient-doctor relationship raises concerns about privacy. Substance use, mental health, and sexual history, for example, are areas that many patients could be reticent to share with caregivers. However, we found that those who shared their visit notes were neither more nor less concerned about privacy than were non-sharers. While this is reassuring for those who advocate open access to patient notes, perhaps the proportion of those who shared would have been greater if elements of the “social history” were restricted. Health information technology vendors need to consider how to protect patients’ privacy while facilitating access to pertinent medical information and recommendations. Currently there are no clear standards for caregiver access to patient portals, and authentication procedures vary widely [
Caregiver stress is well documented [
As patients share notes with caregivers, they may also consider sharing their notes more broadly, such as posting them on social media platforms. Divulging sensitive information, wittingly or unwittingly, could affect personal relationships, job opportunities, or litigation. A doctor’s note freely accessible on the Internet could generate positive or negative comment from a wide variety of viewers. As social media evolve hand in hand with health care transparency, the consequences for the doctor-patient relationship are largely unknown, and adding caregivers to the mix may introduce even more complexity.
This study’s strengths include the large number of patients granted access to their visit notes in geographically and socioeconomically diverse settings. These participants may represent early adopters of technology that may quickly become standard of practice. But limitations derive from a sample that nevertheless represents a small subset of Americans, so these findings cannot be considered widely generalizable. The data are self-reported survey data, without input from caregivers.
We have undergone a revolution in the way personal electronic data are accessed and shared. Future patient portals will need to integrate the preferences of patients, caregivers, and health care providers. Developing separate secure portals for caregivers may help limit access to components that the patient prefers to keep private. Vendors will need to add features allowing patients to share specific information with caregivers based on patient preferences. Policies for organizations seeking to enable “delegation” for patients are needed to address aspects such as authentication of patient delegates and how control of specific access to patient information is supported. Protections against inadvertent over-sharing must also be considered.
Coordinating care for patients is both a tremendous challenge and a core competency for effective care organizations [
Beth Israel Deaconess Medical Center
Electronic Health Records
General Educational Development
Geisinger Health System
human immunodeficiency virus
Harborview Medical Center
Veterans Administration
All investigators were supported by The Robert Wood Johnson Foundation’s Pioneer Portfolio grant No. 65921. Dr Delbanco and Ms Walker were also supported by the Drane Family Fund, the Keane Family Foundation, and the Richard and Florence Koplow Charitable Foundation. Dr Elmore was supported by the National Cancer Institute (K05 CA 104699).
None declared.