Table 2 Summary of literature included. A summary of the data extracted from the included studies, themed around nature of study and research question answered.

Questionnaires

Giving Doctors’ Daily Progress Notes to Hospitalized Patients and Families to Improve Patient Experience (Weinert, 2017) [17]

University hospital

Questionnaire of patient and healthcare providers’ response to a trial giving patients a daily copy of ‘progress notes’

(Experiential)

Patient, family members and providers (attendings, residents or medical students) involved.

Pilot study: 12 patients, 6 providers, 70 notes

2nd study: 73 patients, 6 providers, 677 notes given of 2011 notes printed, (33.7%)

Quantitative analysis of 2 survey results (one for patients and family members, one for providers)

Most patients (76%) responded favourably to reading notes as it improved their understanding and feeling of control.

Some providers (9–28%) thought it affected their practice: they were more careful about phrasing things and it lengthened the consultation mildly. The majority of providers (up to 72%) disagreed with the idea that it affected practice and 3–16% were neutral.

There were occasions where what was written caused discord between patient and doctor, ultimately leading to better understanding: a patient with severe pain complained she did not have ‘mild pancreatitis’ and the doctor explained it was biochemically mild.

The sample size and duration of this study limited its external validity.

Is Canada ready for patient accessible electronic health records? A national scan. (Urowitz et al., 2008) [18]

Emails to CEOs of general and acute medical hospitals in Canada

Questionnaires given to CEOs to measure national readiness for adoption and implementation of EHRs. In addition, the CEOs were asked to forward the questionnaire onto chiefs of medicine, nursing and informatics staff or other relevant people.

(Experiential)

No patients were involved.

83 CEOs surveyed.

Only 3% of these CEOs had assessed staff perceptions directly.

Statistical analysis and descriptive statistics.

Interview participants included 9.4% CEOs, 3.8% Chiefs of Medicine, 11.3% Chiefs of Nursing, 7.6% Chief Information Officers and 67.9% listed as ‘Other’ (the majority of which had more administrative roles: managers, privacy officers, etc).

54.2% of hospitals had some sort of EHR. Barriers to patient access were identified as hospital finances, patient computer literacy and clinician buy-in.

Staff perceptions (data from 3 hospitals only): Less than 25% thought patients want access to EHR and only 16% thought they’d want lab results, in contrast with other Canadian studies that say patients/public would like access. Only 3.6% said staff would be willing and eager to provide access, 28.6% said staff would be hesitant but willing to provide access, 17.9% thought staff would support partial access.

Medical staff were surveyed minimally in this study, with the emphasis on those in more managerial roles.

The response rate may be inaccurate as the distribution of the questionnaire by each CEO was not tracked.

Interviews & Focus groups

Building and testing a patient-centric electronic bedside communication center (Dykes et al., 2013) [22]

Acute care in 2 academic medical centres, USA

Focus groups to identify improvements for the electronic prototype.

Pilot testing electronic bedside communication prototype including scheduled events, daily routine and space to write notes.

(Experiential, real time access)

No healthcare practitioners were involved.

Focus groups included former patients, family caregivers and hospital volunteers who had recently been inpatients or caregivers

Pilot testing and interviews were done by 11 participants (8 inpatients and 3 family members) .

Mixed methods: Focus groups and bedside interviews

The majority of patients said they would use the device. An 82 year old said she does not want ‘one more thing to worry about’ but her family said they would use it. A 90-year-old patient said he would prefer to speak to humans directly. Accessibility issues were noted: most patients older than 64 had trouble with the touch-screen hardware.

Recommendations included videoconferencing tools and voice recognition.

The method of the analysis of the interview and focus group results is unknown. The pilot testing is also limited by its small sample size.

Designing Patient-Centric Information Displays for Hospitals.” Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (Wilcox et al., 2010) [24]

Emergency Department, USA

Trial of implementing in-room displays based on medical records.

(Experiential, real time access)

The design of these displays involved comments and observations from collaborating physicians.

18 patients, 11 visitors (all female),

16 physicians.

Patients were interviewed after deploying the prototype, then again when all further medical updates had been made to their prototype.

Semi structured interviews with patients, family and healthcare practitioners.

17 patients overwhelmingly positive about the poster display (which included a health profile, vitals, what’s next, medications, care team) because it helped them keep track, especially the ‘What’s Next’ section. One participant said: “yeah, they come in and update me but, I mean, I can’t keep track of it all. That’s why I really like this.”

Physicians were also positive but expressed concerns over display of lab results and vitals.

The study gives limited detail about the structure and method of analysis of each interview, particularly for visitors and physicians.

Patients were recruited for the study by collaborating physicians, which may have introduced selection bias.

Why do people want a paper copy of their electronic patient record (Wibe et al., 2010) [20]

2 hospitals in Norway

Three papers from one study.

Interviews with Norwegian adult patients who have requested access to their notes

(Experiential, retrospective access)

A convenience sample of 17 volunteers following an inpatient stay: 16 female, 1 male.

No physicians or other healthcare practitioners were interviewed.

Qualitative content thematic analysis

A main priority for patients is the secure transmission of information between healthcare personnel; they want to take it upon themselves to be the ‘messenger’.

The population of this study represents less than 1% of those admitted in the same time frame.

The retrospective nature of this study may have increased the recall bias.

Patients reading their health records - what emotional factors are involved? (Wibe et al., 2009) [37]

Distrust is an important motivator for asking for record access, but some have more practical reasons – e.g. insurance.

Discrepancies (mixing up records of 2 different patients) and lack of openness cause irritation and resentment in individual health care worker and in the system.

Lay people’s experiences with reading their medical record (Wibe et al., 2011) [21]

Very few patients (1%) of those admitted to hospital request their records – and this research interviewed only this selected sample.

Patient reasons for requesting records include desiring a sense of control, taking responsibility and examine inaccuracies.

Problems reading the record included patients feeling like they weren’t being taken seriously and the record stigmatising their lifestyle problems.

Note Analysis

Patient-centric medical notes: Identifying areas for improvement in the age of open medical records (Lee et al., 2017) [27]

Tertiary care centre, USA

Retrospective analysis of patient notes looking at barriers to patient access

(Hypothetical)

337 inpatient admission notes.

No direct information from patients or healthcare practitioners contributed to the results.

Statistical analysis of characteristics of interest (e.g. offensive medical language) identified in notes.

The notes that create confusion, generate offense or impact perceptions/professionalism were those that used medical words which may have judgemental connotations in (e.g. ‘complains’, ‘claims’, ‘denies’), typographical errors and use of jargon.

This study did not look at patients directly accessing notes, but rather what within the notes creates barriers to access. These characteristics were determined by clinicians, rather than involving patients.

This study was conducted at a single site which reduces its eternal validity.

Portal Evaluation

Implementation of acute care patient portals: recommendations on utility and use from six early adopters. (Grossman et al., 2017) [28]

6 hospitals in the USA.

Evaluation of implementing acute care patient portals.

(Experiential, real time access)

6 hospital portals serving 1065 patients overall.

Literature review of 27 studies looking at the characteristics of acute care patient portals.

Analysis of characteristics, usage and tools of each portal.

Literature review of other studies on patient portals.

The purpose of most portals is to engage patients and facilitate transition to outpatient management, clinician relationship, transparency of information and patient safety.

They take a patient centric approach with a care plan and daily schedule, and tools to facilitate messaging and accessing results; half showed the diagnosis.

Problems with portals were also identified, including timing of lab result release and the sharing of differential diagnoses. There was a concern that portals would lead to ‘overwhelming amounts’ of patient contact needs.

The study focuses of large academic medical centres meaning the results may not be valid across smaller hospitals.

Not all the portals focussed on communicating medical advice; some were used for non- medical information.

Clinical Trials

The effect of tablet computers with a mobile patient portal application on hospitalized patients’ knowledge and activation. (O’Leary et al., 2016) [29]

Controlled trail in a large teaching hospital

Patient portal with personal health information (e.g. names and pictures of team members, scheduled tests, and list of active medications) presented on 15 iPads, given to 100 intervention patients.

(Experiential, real time access)

100 intervention patients,

102 control patients.

Each physician in charge of the care of each patient was also interviewed.

Structured and semi-structured interviews with patients and their physicians separately – responses recorded verbatim and compared to each-other and the medical record.

Short form measuring Patient Activation (PAM-SF) qualitatively.

A larger percentage of intervention patients (56%) named 1 or more physicians compared to the control group (29.4%). Similarly, more intervention patients (47%) could name the role of one or more physicians compared to the control group (15.7%).

There was no difference in patient groups knowledge of all their planned tests or procedures. Patient activation (the level of knowledge, skills and confidence a patient has in managing their own healthcare) from the PAM-SF score remained unchanged between the groups.

Of those who had access to the portal, 57% used the portal more than once a day. The majority of patients thought it was useful and easy to use. Reasons for lack of patient activation could include terminology interpretation and lack of time. Further improvements include working to engage patients more and make portals more accessible.

Study did not assess health literacy of their patients and was limited to English-speaking patients. The study was conducted on one site only.

Mixed Methods

A tablet computer application for patients to participate in their hospital care. (Vawdrey et al., 2011) [25]

Cardiac Step-down unit, USA

Mixed methods: semi-structured interviews and questionnaires.

(Experiential, real time access)

5 patients interviewed (all male)

Clinician reception of the pilot study was noted although not assessed formally.

Analysis using subscales for satisfaction and usefulness within the questionnaire, and anecdotal evidence from interviews.

Patients found portal helped them engage more with their care and form a more personal relationship with the MDT. None of the patients raised privacy concerns.

Small population of study on one specialised unit, so results may not be consistent across all acute care settings.

Acute care patient portals: a qualitative study of stakeholder perspectives on current practices. (Collins et al., 2017) [26]

Acute care in 5 academic medical centres, USA

Mixed methods: semi-structured interviews, focus groups, site visits and questionnaires of expert leaders at each site to evaluate perceptions of patient portals and identify requirements of patient portals in the acute care setting.

(Hypothetical)

84 participants in total including on average:

3.3 PFAC (patient family advisory council) members

4.5 researchers

3.8 Information system leaders

3.6 clinical leaders

1.3 policy makers and administrators

Thematic analysis of 12 interviews and 18 focus groups to form development of an explanatory model.

Main themes identified from stakeholders include: access and security (with concern over BYOD use outside of hospital); content and functionality encouraging simplistic intuitive displays; the need to minimise the exclusion of those with less health literacy and engagement; using both patients and doctors in the design and training to use the portal.

Patients believed the portal would facilitate face-to-face communication rather than replace it. The portal must be easy to use and ‘familiar’; this was considered particularly important for the acutely-unwell patient.

The study was limited to academic centres and as such, results may not be valid across all acute care settings.

The challenges in making electronic health records accessible to patients. (Beard et al., 2012) [31]

–

Examining concerns arising in relation to patient access to health records.

(Hypothetical)

–

Analysis

1.Cost and security concerns: limited financial resources need to be shared across healthcare organisations and clear regulations regarding access must be communicated.

2.Assignment of responsibilities and rights: there are conflicts regarding the timing of information (release of lab results); use of medical terminology and control of the record.

3.Liability issues:the use of messaging portals may present a liability risk.

4. Tensions between patients and doctors: is messaging appropriate; are physicians proficient with electronic communication.

Ethical Considerations about EHR-Mediated Results Disclosure and Pathology Information Presented via Patient Portals. (Davis and Smith, 2016) [32]

–

Examining ethical issues regarding patient access to pathology and other results

(Hypothetical)

–

Analysis

The main focus was on time delays for different pathology reports to be accessible to patients.

Some results (like HbA1C) are useful for patients to monitor the progress and for peace of mind, improving self-reliance but patients may feel to blame for poor outcomes due to lack of vigilance.

Having access to other results could have negative effects:

Access to abnormal, ‘surprising’ results could lead to patients having to come to terms with a diagnosis without emotional support or the immediate opportunity to ask questions; the (lack of) importance of out-of-range results could be misunderstood; genetic testing results and diagnoses might be seen without sufficient counselling.

Legal, Practical, and Ethical Considerations for Making Online Patient Portals Accessible for All. (Lyles et al., 2017) [33]

America

Examination of the legal, practical and ethical issues regarding patient portals in America.

(Hypothetical)

–

Analysis

Interest in portals is constant throughout populations but portals themselves are mostly small-font, English only, text-based content and therefore disadvantage disabled users and non-English speakers.

There are no specific regulations regarding the accessibility standards of EHRs, but portal designs must take the Civil Rights act and the Digital Accessibility Guidance into account.

Why a shared care record is an official medical record. (Gu et al., 2013) [34]

New Zealand

An argument supporting the validity of a shared care record as an official record and the consequences of this.

(Hypothetical)

–

Analysis

A shared record needs to meet ethical and medico-legal criteria, including regulations for interoperability, clinical responsibility and restrictions on patient and professional access. Issues include balancing empowerment with legislative requirements and how we can foster confidence in healthcare professionals to co-partner the record.

Ethical questions must be considered for electronic health records. (Spriggs et al., 2012) [35]

Australia

Identifying concerns over electronic health records

(Hypothetical)

–

Analysis

This study identified specific questions arising from the move towards personally controlled electronic health records in Australia, including who benefits and who should pay for the system, what uses of the system are legitimate, how we should govern the management and use of the system and how we should implement privacy.

Three main questions were identified for our analysis:

1.Can records be considered trustworthy if patients can lock/change record?

2.Will patients who do not (or can not, for health literacy reasons) use the system be disadvantaged?

3. Who else can access the information? Consumers generally more willing to share information for research and public health uses but not for pharmaceutical use.

Medical records: practicalities and principles of patient possession. (Gilhooly and McGhee, 1991) [36]

UK

Examining the practical and ethical advantages and disadvantages of patient possession of medical records.

(Hypothetical)

–

Analysis

Practical benefits include storage and transfer of notes – this would be less problematic if the patient possessed them and the patient can check/screen notes regularly.

Practical problems include the risk of patients losing records (although research shows otherwise), the extra time doctors may take to explain the record’s contents (although the benefit of the patient understanding may outweigh the cost of time lost).

Ethical benefits include reducing the power imbalance between the doctor and patient, leading to more communication and trust as patients can control confidentiality of records.

Ethical problems include that the doctors may feel they need to censor record so as to not offend patients. There is an argument that the notes are doctors’ property and that patients might feel anxious due to the medical uncertainty portrayed through the notes.