Search process
The PRISMA flowchart in Figure 1, as guided by [3], represents the search process that was used to obtain the final results. The databases included were PubMed, ScienceDirect, Springer Nature and Google Scholar. A total of 13754 research papers were obtained from Science Direct (6321), PubMed (5623), Springer Nature (1044), and Google Scholar (766). After the application of the eligibility criteria, a total of 83 articles were available for further screening from all the databases. No duplicates were found from the 83 articles. An analysis of the keywords and associated data was performed thoroughly; reviews, articles with no full texts or articles that did not examine barriers in EHRs were excluded, and a total of 20 articles qualified for this study.
Results of the Analysis
The 20 research papers used in this review were analysed for barriers to EHR implementation, and the countries where barriers were encountered were identified. Table 2 shows a summary of these papers, including the journal names and authors.
Table 2: Summary of articles and countries included in the study
Journal name
|
Authors
|
Year of Publication
|
Country
|
Journal of medical internet research
|
Savai et al.[4]
|
2021
|
Canada
|
Acta clinica Croatica
|
Vekic` et al. [5]
|
2022
|
Crotia
|
BMJ open
|
Sharpe et al. [6]
|
2022
|
England
|
Age and ageing
|
Ribbink et al.[7]
|
2023
|
England
|
Sensors (Basel, Switzerland)
|
Pilares et al.[8]
|
2022
|
Switzerland
|
Military medicine
|
Martin et al.[9]
|
2022
|
England
|
Applied clinical informatics
|
Levy et al.[10]
|
2023
|
Germany
|
Healthcare (Amsterdam,Netherlands)
|
Gold et al. [11]
|
2020
|
Netherlands
|
BMJ open quality
|
Gannon et al. [12]
|
2021
|
Zimbabwe
|
Journal of the American Heart Association
|
Funes Hernandez et al.
|
[13] 2024
|
England
|
BMC medical informatics and decision making
|
Bisrat et al. [14]
|
2021
|
England
|
Applied clinical informatics
|
Bersani et al. [15]
|
2020
|
Germany
|
Saudi medical journal
|
AlSadrah [16]
|
2020
|
Saudi Arabia
|
JNMA; journal of the Nepal Medical Association
|
Agrawal et al. [17]
|
2022
|
Nepal
|
Journal of the American Medical Informatics Association
|
Shi et al. [18]
|
2021
|
England
|
Sustainability
|
Lu et al. [19]
|
2021
|
USA
|
BMC Primary Care
|
Kalkhajeh et al. [20]
|
2023
|
USA
|
Software: Practice and Experience
|
Ebad [21]
|
2020
|
USA
|
BMC Medical Informatics and Decision Making
|
Anthony Luberti [22] Sansanee Craig
|
2024
|
USA
|
Saudi Journal of Nursing and Health Care
|
Amponin [23] and Britiller
|
2023
|
Saudi Arabia
|
Barriers
Initial impressions
A study in Serbia [5], Canada [6], the Netherlands [7], and Irania [20] concluded that the first impression of a new EHR system determines the success of the implantation process. The implementation of a new information system always requires proper change management. One thing to understand is that every organisation has an organisational culture in which things are run according to the institution and rules that are set aside by the organisation. Therefore, the implementation of a system requires not only resources and training but also how to handle workers and how to approach them with new ideas for a new EHR system. A proper presentation is required for employees to understand why the system is needed, its advantages and disadvantages and whether it will bring a positive change to the institution.
Resistance to change
Several studies in Saudi Arabia [16,23,24], , the United States [18], and Switzerland [19] revealed that changing the working culture is the most difficult process. Some workers are not ready for a new environment, and most often, they fear disembarking or failing to catch up. Issues associated with resistance to change include reduced working, which leads to less patient care provision and ineffective quality of care [5,6,15,18,19,21,22].
Computer skills
In Kenya, the Rwanda, Uganda, Zimbabwe and Mozambique mUzima systems [4] document two challenges that are common concerning computer skills. The issue of computer literacy is the main issue because after the introduction of a new EHR system, the consideration of whether workers know that not everyone is not computer literate is usually ignored [4,10,11,23]. Another issue with computers is their limited access. The studies further pointed to a lack of resources and sometimes a lack of computer literacy.
Training
EHR implementation can only be successful if there is enough training, and this is an issue because training in a new system takes time and might require repletion for workers to grasp the concept. Training can also be costly due to the arrangements for workshops, trips and expenses to make trainings successful, as documented in the United States Military, Zimbabwe and Ethiopia [9,10,12,14,18,23]. Another study described the training of system installers, programmers, administrators and users. Some systems are complex in that they need continued training and updates to remind users how the system works.
System-dependent factors
Resources
The resources to be used in Electronic Health Records implementation are costly, and these resources include computers, internet connectivity, servers, infrastructure and sometimes tablets [8,16–18,23]. Several studies in Rwanda, Saudi Arabia, Nepal and the United States have proven that most implementation failures are due to these resources because of the cost of providing them and maintenance.
Electric power interruption
This has become one of the major drawbacks of EHR implementation, especially in underdeveloped countries. A study in Zimbabwe, Saudi Arabia [12] [23], concluded that the scarcity of electricity disturbs the proper operation of servers and internet connections, thereby causing poor working conditions with large amounts of data not entering into systems.
Privacy and security concerns
Most of the studies highlighted the issue of privacy and security, although it is debatable. Others have concluded that it is not secure to use EHRs because data can be lost or sent to the wrong systems and people. In the United States and China, studies have shown that there are challenges in the use of health data in artificial intelligence (AI) and machine learning (ML), and in some cases, there are legal battles. Studies in Nepal, the United States, Iran and Saudi Arabia noted that security concerns can be avoided by providing robust security measures, although doing so is very costly, leading to another problem [6,18,20–23].
Interoperability issues
The connection of EHR systems to other health systems and management systems is a major issue because having a system that cannot share information is not satisfactory. The interoperability process requires experts and is also costly, and most of the studies in Rwanda, Kenya, Uganda Mozambique, the Netherlands and the United States show that most implemented EHRs are not interoperable with other systems [4,7,9,13,15,21–23].
Documentation issues
Studies in Switzerland and Saudi Arabia have shown that the implementation of EHR has been widely documented. In some studies, there was parallel implementation and more documentation in two-system comparisons. More information is also collected as a requirement by the system and thus makes workers tired and work more [19,21]. This is a challenge because sometimes they end up using the system less or entering information that is not relevant enough to give proper quality of healthcare and leads to incomplete health records.
Social-related factors
Patient-healthcare provider relationship
In Saudi Arabia and the United States, serious challenges have been documented because healthcare providers have issues with eye-to-eye contact with patients [23]. This has made healthcare workers lose proper interaction with patients because most of the time, the focus is on computers or tablets and understanding the new system [19,21].
Time-related issues
The time that is taken for documentation is a serious problem because patients are waiting in a queue. It was noted in Switzerland, Saudi Arabia, the Netherlands and the United States that users who are new take time to use the system, whereas users who have used EHRs before they find it reliable and useful with less time taken [7,15,19,22,24]. This has allowed institutions using EHRs to be overcrowded, especially if most patients register for the first time.
Frequency of barriers
Table 3 shows a total of 11 barriers identified from the 20 studies selected in this scooping review. The identified barriers are resistance to change, lack of computer literacy, limited computer access, training issues, lack of resources (computers, internet, servers, tablets), electric power interruption, privacy and security concerns, interoperability issues, documentation issues, patient-healthcare provider relationship issues and time-related issues.
Table 3: Barriers to the Implementation of EHR and Frequency of Occurrences in the Selected Articles
Category
|
Barriers
|
Frequency
|
%Frequency
|
Initial impression
|
Resistance to change
|
8
|
40
|
|
Lack of computer literacy
|
2
|
10
|
|
Limited computer access
|
1
|
5
|
|
Training issues
|
6
|
30
|
System-dependant factors
|
Lack of resources (computers, internet, servers, tablets)
|
5
|
25
|
|
Electric power interruption
|
1
|
5
|
|
Privacy and security concerns
|
6
|
30
|
|
Interoperability issues
|
8
|
40
|
|
Documentation issues
|
2
|
10
|
Social-related factors
|
Patient-healthcare provider relationship
|
1
|
5
|
|
Time-related issues
|
1
|
5
|
These barriers are shown using the frequency of occurrence in a bar graph in Figure 2. Resistance to change and interoperability issues are the most common barriers, with a relative frequency of 40% each. Training issues and privacy and security had a relative frequency of 30% in 6 studies each. A lack of resources was also a common problem in 5 studies, with a relative frequency of 25%. Documentation issues and lack of computers had a relative frequency of 10% each in two studies. The fewest barriers that appeared in 1 study each were patient-healthcare provider relationship issues, time-related issues, electricity interruption and limited computer access, with a relative frequency of 5%.
A close analysis of countries where the challenges were documented provided insight. The content and thematic analysis of the barriers from different countries are summarised in Table 4.
Table 4: Barriers to EHR Implementation grouped according to Income level of countries.
Low-income countries
|
lower-middle Countries
|
upper-middle Countries
|
High Income Countries
|
Ethiopia, Uganda Mozambique, Rwanda
|
Ghana, India, Malawi, Zimbabwe, Iran
|
Cape Verde, China Iraq, Russian, Serbia
|
United States, Netherlands, Britain, Canada, Croatia
|
Electricity power disruption
Poor internet connection
Lack of resources
Resistance to change
No enough hardware
|
Documentation issues
time-related issues
Training issues
Lack of skills
|
Interoperability issues, long patients queues, incomplete Health records.
|
Data security,
Privacy from ML and AI use
patient-healthcare provider relationship issues,
|
An analysis of countries where barriers were documented and the stage of implementation shown in Figure 3 shows that the level of income of a country is related to the type of barriers encountered.