Table 1 Summary of Included Articles

1. Greenhalgh et al. 2013 [5]

Retrospective comparative analysis

Four different shared records systems in each of: Scotland (5.1 million people), Northern Ireland (1.8 million people), England (51 million people) and Wales (3 million people).

230,000 monthly accesses of Scottish ECS (Emergency Care Summary). English (SCR) Summary Care Record temporarily halted.

Nil reported

Nil reported

2. Morris L, Brown C, Williamson M and Wyatt J. 2012 [26]

Survey

5.1 million people

Benefits claimed included more efficient assessment; reduced drug interactions; fewer adverse drug reactions; less duplicate prescribing; 34% of clinicians said the ECS had changed a clinical decision.

3. Ayatollahi H, Bath P A and Goodacre S. 2009 [38]

Qualitative study/survey (2 months)

Hospital with 1100 beds and 5500 staff servicing population of 530,000

Mostly using paper records

Nil reported

Occasional clinical benefits from access to patient information in ED noted.

Patient confidence in the confidentiality of information is paramount.

4. Greenhalgh et al. 2008 [6]

Multi-site, mixed-method case study

51 million people (1 year)

System was discontinued

The study found that there was no direct evidence of improved clinical safety apart from a ‘rare but important positive impact on preventing medication errors.’

The study identified that having pre-existing records delivered a rare but important ‘positive impact on preventing medication errors’.

5. Greenhalgh et al. 2008 [9]

Semi-structured interviews and focus groups

51 million people

The project was halted and is still in the process of being restarted.

Nil Reported

Nil Reported. This study did not study actual usage of the system. It focused on the patients’ attitudes toward use of it.

6. Greenhalgh et al. May 2010 [8]

Retrospective observational study, interviews and ethnographic field observation.

16 Primary Care trust regions with a total population of 29.8 million (1 year).

By 2010, 1.5 million such records had been created. In participating primary care out-of-hours and walk-in centers, an SCR was accessed in 4% of all encounters and in 21% of encounters where one was available.

Nil observed, though risks seen from patient records being incomplete or inaccurate.

Rare but important positive impact noted where system helped in preventing medication errors.

7. O’Malley A S, Samuel D, Bond A M, and Carrier E. 2012 [39]

Interviews

16 US States (6 months)

Very limited - only 29% of general practices make after hours care arrangements.

Nil Reported

Most of the benefits determined were based on cost containment and in some cases revenue generation and in-network referral retention.

8. Vest J R, Gamm L D, Ohsfeldt R L, Zhao H and Jasperson J. 2012 [35]

Retrospective observational study

Integrated care collaboration of central Texas. Population not stated (3 years).

HIE was used for up to 21.1% of encounters.

Nil Reported

Nil Reported

9. Frisse M E, Johnson K B, Nian H, Davison C, Gadd C S, Unertl K M, Turri P A and Chen Q. 2011 [33]

Retrospective matched cohort analysis

Memphis Tennessee, 1.2 million people (2 years)

HIE data was accessed in 6.8% of all visits.

Use of HIE reduced hospital admission rates.

Use of HIE improved management of patients with long-term conditions resulting in $1.07 million annual savings.

10. Hripcsak G, Sengupta S, Wilcox A and Green R A. 2006 [40]

Retrospective matched cohort analysis

2.5 million patients (7 months)

Used for 5–30% of patient encounters.

There were information gaps for one third of ED patients. In one third of those cases, getting that missing information was important.

Closing gaps in the information provided to EDs results in significantly increased efficiencies in care; a reduction of both redundant testing and treatment delays, enabling scare resources to be redeployed for care of other patients.

11. Shapiro J S, Kannry J, Kushniruk A W and Kuperman G. 2007 [41]

Survey

Survey of 216 emergency physicians across 12 New York hospitals

The emergency physicians surveyed believe that reliable availability of information will lead to significant usage.

Nil observed; Please note: This was a survey of opinions and attitudes about future HIE usage rather than a survey of actual HIE usage.

Nil observed, Please note: This was a survey of opinions and attitudes about future HIE usage rather than a survey of actual HIE usage.

12. Vest J R and Jasperson S 2011 [42]

Retrospective observational study

The medically indigent population of central Texas (3.5 years)

105,705 unique users’ sessions is a significant number of observations

Nil Reported

Nil Reported

13. Vest J R, Kern L M, Silver M D and Kaushal R. 2014 [43]

Retrospective observational study

800,000 patients in Rochester, New York (2 years)

6800 records were analyzed.

A significant decline in readmission indicates that the system has value in reducing adverse outcomes.

A significant decline in readmission (57%) indicated that the system has value in improving patient care.

14. Finnell J T, Overhage J M. 2010 [19]

Track log file analysis

1.6 million people in Indiana (6 months)

HIE information sought in 16% of ED admissions.

Nil reported

Nil reported, however the majority of clinicians viewed the availability of information as beneficial.

15. Johnson K B, Unertl K M, Chen Q, Lorenzi N M, Nian H, Bailey J, Frisse M. 2011 [28]

Mixed-method analysis, observation and interviews

1.7 million patients in Memphis (6 months)

Used for 7% of all patients and in 16% of repeat visits.

Detected public health risks on a small number of patient visits (0.8%).

Reduced the time taken to see patients, reduced the need for repeat testing and improved clinicians’ understanding of patients’ overall conditions.

16. Yaraghi N. 2015 [27]

Retrospective Observational study

Unstated (6 months)

737 ED visits

Nil reported

Significant reductions in laboratory tests (52%) and use of radiology services (36%) with the resultant ability to redeploy resources to other aspects of care.

17. Dumay A C M and Haaker T I. 2010 [21]

Interviews

The Twente Region (population 620,000)

The electronic locum report (ELR) system is well used in Twente but efforts to scale it up and expand it across the Netherlands have failed.

Nil Reported

Nil reported. This study did not assess the value of the ELR system as an aid to improving clinical care.

18. Woudstra D P J. 2013 [20]

Interviews

Holland (population 16.8 million).

Only 23.5% of Dutch population have given permission for their electronic health records to be shared.

Nil reported. This study did not look at improvements to patient safety, nor did it measure any impact on the quality of patient care that could be achieved via use of the ELR.

Nil reported. This study did not look at improvements to patient safety, nor did it measure any impact on the quality of patient care that could be achieved via use of the ELR.

19. Ben-Assuli O, Shabtai I and Leshno M. 2013 [24]

Track log file analysis

Seven main Israeli hospitals; 3.8 million patients (HMO) (3 years)

Medical history viewed in 16.2% of referrals.

Access to records improves admission decisions. Good admission decisions have a significant impact on patient safety.

Although many ED physicians believe that the majority of their patients would benefit from longitudinal health information, they attempt to obtain such data less than 10% of the time.

An improved admission decision positively impacts a patient’s clinical care and improves the medical facility’s ability to manage its resources in a manner that enables optimal care delivery. Access to internal records resulted in a 22.9% reduction in single day admissions.

The availability of a patient’s long term health records including information about medications, diagnoses, recent procedures, and recent laboratory tests is critical to forming an appropriate plan of care.

20. Ben-Assuli O, Leshno M and Shabtai I. 2012 [22]

Track log-file analysis

Seven main Israeli hospitals 3.8 million patients (HMO) (3 years)

Medical history viewed in 16.2% of referrals.

Physicians used medical records more when under pressure. Emergency physicians may admit more patients unnecessarily when under pressure, if they do not have time to get access to the information they need.

Records found to be more useful in complex cases.

21. Ben-Assuli O, Shabtai I and Leshno M. April 2013 [25]

Track log file analysis

Seven main Israeli hospitals; 3.8 million patients (HMO) (3 years)

External medical history viewed in 4.3% of cases. Internal medical history was viewed in 26.9% of cases.

Reduced the number of emergency readmissions within 7 days: confirms a clear improvement in patient safety.

Improving admission decisions positively impacts care, freeing up resources to better focus care on where it is needed.

22. Ben-Assuli O, Shabtai I, Leshno M and Hill S. 2014 [23]

Track log file analysis.

Seven main Israeli hospitals 3.8 million patients. (3 years)

Medical history viewed in 24% of all referrals.

Better admission decisions improve patient safety. Availability of blood pressure results increased the likelihood of admitting a patient by 70.6%. Availability of community records increased the likelihood of admitting CP patients by 29.2%.

Better admission decisions improve quality of care. Better decisions are enabled by a more comprehensive patient view.