FEATURE ARTICLE
The Use of Electronic Medical Records in Jordanian Hospitals A Nationwide Survey Ahmad Tubaishat, PhD, RN, Omar M. AL-Rawajfah, PhD, RN
No previous nationwide study has estimated to what extent electronic health records have been implemented in Jordanian hospitals. The purpose of this descriptive, cross-sectional study was to explore the level of adoption and use of electronic health records in Jordanian hospitals across all major healthcare service providers. The standardized American Hospital Association annual survey was used. The level of use of electronic functionalities of electronic health records was determined. The association of certain hospital characteristics with the adoption of electronic health records was examined. A proportion of 10.3% of the participating hospitals had a comprehensive use of electronic health records in all units, and 15.5% had a basic system in at least one unit. Most (74.2%, n = 72) had not implemented electronic health records. The hospitals with a higher rate of adoption were found to be larger, government, urban, and teaching hospitals that had coronary care units. The level of adoption of electronic health records in Jordan is relatively low. This fact should impel policy makers to resolve the challenges and obstacles for such adoption. National strategic plans are needed to address the goals and implementation processes of electronic health record systems in all Jordanian hospitals. KEY WORDS: Electronic medical records, Health information technology, Hospitals, Informatics, Jordan
SIGNIFICANCE The introduction of health information technology (HIT) into healthcare industries is gaining pace.1 There is much greater investment in HIT in developed countries than ever before. For example, €544 million was spent in Spain in 2009,2 and in the United Kingdom, it was estimated that the expenditure on HIT reached approximately £12.8 billion in 2011.3 In the United States, almost US $20 billion was allocated to Author Affiliations: Faculty of Nursing, Adult Health Nursing Department, Al al-Bayt University, Mafraq, Jordan (Dr Tubaishat); and Faculty of Nursing, Sultan Qaboos University, Muscat, Oman (Dr AL-Rawajfah). The authors have disclosed that they have no significant relationships with, or financial interest in, any commercial companies pertaining to this article. Corresponding author: Ahmad Tubaishat, PhD, RN, Faculty of Nursing, Adult Health Nursing Department, Al al-Bayt University, PO Box 130040, Mafraq 25113, Jordan (
[email protected]). Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved. DOI: 10.1097/CIN.0000000000000343
Volume 00 | Number 0
support the adoption of HIT in US hospitals in 2009.4 By 2011, this figure had risen to $38 billion. The major force driving this generous expenditure is in fact an effort to control healthcare costs. For example, in the United States, it was estimated that the adoption of HIT could produce efficiency and safety savings of $142 billion in US physicians’ offices and $371 billion in US hospitals for 15 years (2004–2018).5 The integration of HIT generally, and electronic medical records (EMRs) specifically, in healthcare can not only reduce costs but also enhance the quality of care.6 Improvement of quality of care can be the results of different benefits of EMRs such as gathering accurate clinical information and coordinating the care process7; enhancing decision-making services8; minimizing communication errors7; generating electronic reports that are necessary for institutional, private, and public requirements9; and improving patient safety.10–12 The combination of EMRs with the computerized provider order entry (CPOE) reduces the costs and reduces the duplication of laboratory and radiology tests and investigations. The documented effect of implementing EMRs on patient outcomes level4 and the national level9 stimulates research around the world to evaluate the extent of EMR implementation in healthcare systems such as that of the United States,13–16 Austria and Germany,17 Greece,1 Korea,18 and Spain.2 The importance of EMRs depends largely on their adoption, application, and use.19 There is wide variation on the reported levels of adoption of EMRs around the globe.20 This variation could be due to lack of agreement on the definitions of EMRs,13 the adoption of EMRs in terms of its capabilities,13,21 methodological differences in estimating the adoption level,20 and convenience sampling techniques with low response rates from the conducted surveys.14 In 2008, the American Hospital Association (AHA) in partnership with the Office of the National Coordinator for HIT of the Department of Health started to conduct an annual survey to assess the level of adoption of EMRs in US hospitals.16 The first survey was conducted to estimate the level of adoption of EMRs in 3049 US hospitals (response rate, 63.1%).14 They surveyed the hospitals about the presence and absence of certain electronic functionalities and whether these functionalities were implemented in all or some hospital units.
CIN: Computers, Informatics, Nursing
1
FEATURE ARTICLE The results of this study revealed that only 1.5% of the US hospitals had a comprehensive EMR system and 7.6% of the hospitals had a basic EMR system. In a follow-up study, these numbers were updated, where hospitals with a comprehensive system increased to 2.7%, and those who implement the basic system increased to 9.2%.15 Thereafter, the AHA annual survey reported that the rate increased from 1.5% in 2008 for the comprehensive system to 34.4% in 2014, a more than 22-fold increase. For the basic system with clinical notes, the increase ranged from 7.6% in 2008 to 41.1% in 2014, a more than five-fold increase.16 A similar approach was followed in a study conducted in Spain to measure the use of EMRs in Spanish hospitals.2 In an observational cross-sectional study, an electronic questionnaire was sent via email to 214 hospitals that refer to the National Health Services in Spain (response rate, 30%). The results found that 39.1% of the hospitals had a comprehensive system, and 32.8% used a basic system. In Korea, a survey was conducted using the same methodology used in the United States to measure the prevalence of EMRs in tertiary teaching and general hospitals in Korea (N = 313).18 From the responses of 122 hospitals (response rate, 39%), they found that 5% of the hospitals had a comprehensive system, and 32.2% had a basic system. Another study located in the literature but with a different methodology, which could make the comparison, is more difficult. One of these studies was conducted to measure the level of adoption of clinical information systems in Greek public hospitals.1 A Web-based survey was sent out to 107 hospitals that refer to the Greek National Health System, with a return of 70 questionnaires (response rate, 65.4%). The questionnaire, which was developed for the study based on a literature review, was completed by chief information officers (CIOs) in the hospitals. The findings indicated that the adoption of inpatient EMRs was 22.9%, without specifying the proportions of basic or comprehensive systems. In another different study, Hübner et al17 compared the prevalence of nursing and medical systems in Austrian and German acute care hospitals (N = 130 and 2172, respectively). All hospitals in both countries received an identical questionnaire, with a relatively low response rate, which was 12.4% in Germany and 34.6% in Austria. The 40-item questionnaire covered the hospital characteristics, technological infrastructure, EMRs, and nursing information system. The results showed that Austrian hospitals used more clinical IT systems than their German counterparts, despite that both countries have a similar budget for HIT (2.6% in Austria and 2% in Germany). The authors tie this difference to the presence of technological infrastructure and organizational changes that promote the IT-friendly environment in Austria compared with Germany. In Jordan, healthcare services are primarily delivered though four major health sectors: government, royal medical services
2
CIN: Computers, Informatics, Nursing
(military), private, and university affiliated. In 2014, the Jordanian government allocated 8% of the total national budget to the Ministry of Health.22 The population in 2014 was thought to be approximately 6.5 million with an estimated per capita income of US $5357.22 In 2009, the Jordanian government initiated the first E-health program in Jordan (Hakeem Program). The Hakeem project was derived from an open-source health information system known as the VistA program, which was originally developed by the US Federal government.23 VistA consists of systems, linked databases, and end-user interfaces such as patient administration records, radiology, pharmacy, pathology systems, and nursing documentation systems. The Hakeem Program started in 2009 as a pilot scheme in one public hospital and one health center. After the pilot, the project intended to cover all hospitals operated by the Ministry of Health and the Royal Medical Services. The main goal was to improve the documentation system in these hospitals, as well as to improve the quality and safety of healthcare services.23 Despite these efforts to improve the computerization of healthcare services, to date, there has been no nationwide, scientific study that has estimated the level of implementation of EMRs in Jordanian hospitals. Therefore, the primary aim of this nationwide survey was to explore the level of adoption and use of EMRs in Jordanian hospitals across all major healthcare service providers. Secondary to this was the desire to investigate possible associations between the level of adoption and hospital characteristics.
METHOD Design This survey used a descriptive, cross-sectional design. The cross-sectional design is an appropriate design to report the adoption rate of EMRs in the Jordanian hospitals. Furthermore, the intention of this study was to generate findings that can be generalized to the status of EMRs in Jordan and other countries that have a similar healthcare system, and this design is capable to meet this purpose.24
Setting and Sample The sampling frame of the study was composed of all Jordanian hospitals listed by the most recent Ministry of Health annual statistical report.22 For easy management of data collection, Jordan was divided into three main geographical regions: north, middle, and south. Second, a complete list of all hospitals from each healthcare provision sector (government, military, private, and university affiliated) in each region was created. The final sampling frame consisted of 104 hospitals, of which 31 were government (29.8%), 59 were private (56.7%), 12 were military (11.5%), and two were university-affiliated (1.9%) hospitals. The hospitals serve different populations such as adult, pediatrics, maternity, and psychiatrics. For the purpose of this study, hospitals were categorized into major teaching,
Month 2017
minor teaching, and nonteaching hospitals. A major teaching hospital was defined as a hospital that is recognized by the Ministry of Health as a teaching institution and regularly accepts health-related students for training. Minor teaching hospitals, on the other hand, are those that are not officially recognized by the Ministry of Health as teaching hospitals but accept health-related students on a nonregular basis. Finally, a nonteaching hospital is one that does not accept health-related students for training and is not officially recognized as a teaching institution.
Data Collection Procedure After establishing the sampling frame and gaining ethical approval, contact information including addresses and telephone numbers were recorded for each hospital. All hospitals were invited to participate in the survey by means of a phone call to a manager or nursing director. After the initial agreement to participate, an on-site visit by a research assistant was arranged. During the on-site visit, the CIO was asked to fill out the survey. For hospitals that do not employ a CIO, the nursing directors or head nurses were asked to fill out the survey. No biases in terms of participation or responses were noted from the survey respondents.
Data Collection Tool The data collection instrument used in this study was based on the one used by the AHA annual HIT survey,16 which is deemed to be of high quality.14 This tool was chosen for the current study because it underwent through rigorous developmental techniques. The tool developers appraised the existing tools in the past 5 years. Then, it was reviewed by field experts. After that, it was pilot tested by CIOs and hospital managers, and comments were solicited from health informatics experts. In 2008, the AHA, in partnership with the Office of the National Coordinator for HIT in the Department of Health, began to conduct an annual survey to assess the level of adoption of EMRs in US hospitals.16 Permission to use the tool was granted before this study commenced. This data collection tool covers 24 electronic functionalities related to the use of EMRs. The electronic functionalities are organized under four major areas: clinical documentation (seven functionalities), for example, patient demographic, physician and nursing assessments, medication lists, and so forth; results viewing (six functionalities), for example, viewing laboratory, radiology and diagnostic test results, and consultant reports; CPOE (five functionalities), for example, nursing orders and medical orders including laboratory, radiology, and medication orders; and decision support (six functionalities), for example, drug allergic or interaction alerts, clinical guidelines and reminders, and drug dosing support. The list of electronic functionalities in the original tool was established based on the
Volume 00 | Number 0
consensus of a panel of experts in the fields of HIT, health policy, health services, and survey research.14 In the current study, the original version of the survey was used, without any modification or translation, since the English language is widely used and understood by medical and technical staff in Jordanian hospitals. Besides using the original survey, information about the hospitals’ characteristics was also collected, namely, bed capacities, location and region, type of hospital, teaching status, and the availability of coronary care units, which is considered a marker of a high technology setting.14 Respondents were asked to indicate the presence or absence of 24 electronic functionalities of the records system in their hospitals and the extent of their implementation or whether there were any future plans to implement them. Furthermore, respondents were asked to indicate whether their hospital had fully implemented the functionalities in all major hospital units, had implemented them in one or more (but not all) major hospital units, or had not yet implemented them in any unit. In the current survey, we followed Jha et al’s14 criteria for classifying hospitals according to the extent of their use of EMRs. Hospitals were classified as having a comprehensive EMR system if they demonstrated the use of all 24 functionalities in all units. Conversely, hospitals were classified as having a basic EMR system if they demonstrated the use of eight specific functionalities in at least one major unit. Jha et al14 further categorized the basic EMR system into basic with or without clinical notes (physician and nursing assessment notes). The current survey also followed the same categorization of the basic EMR system (Table 1).
Ethical Considerations The study was approved by the research and ethics committee of the School of Nursing and the Deanship of Academic Research at Al al-Bayt University, as well as by the participating hospitals. The act of completing the survey was taken as a proof of consent. Participation was completely voluntary, and anonymity was ensured; no personal identification data were required from either the hospital or the personnel who filled in the survey.
Statistical Analysis The prevalence of the adoption of EMRs was calculated using three definitions of EMRs: comprehensive, basic with clinical notes, and basic without clinical notes. Bivariate analyses were used to examine the relationship between hospital characteristics (size, region, location, type of hospital, teaching status, and presence of coronary care units) and adoption of a basic or comprehensive EMR systems. The P value was set at the level of .05 to designate the statistical significance.
RESULTS According to Jha et al’s14 definitions, the vast majority of participating hospitals (72, 74.2%) were found to not implement
CIN: Computers, Informatics, Nursing
3
FEATURE ARTICLE
Table 1. Electronic Functionalities Required for Classifying Hospitals as Having Basic or Comprehensive EMR Systems (N = 97) EMR Functions Required Comprehensive EMR Basic EMR With Clinician Notes Electronic clinical information ( ( Patient demographics Physician notes ( ( Nursing assessments ( ( Problem lists ( ( Medication lists ( ( Discharge summaries ( ( Advance directives ( Computerized provider order entry Laboratory tests ( Radiology tests ( Medications ( ( Consultation requests ( Nursing orders ( Results management View laboratory reports ( ( View radiology reports ( ( View radiology images ( View diagnostic test results ( ( View diagnostic test images ( View consultant reports ( Decision support Clinical guidelines ( Clinical reminders ( Drug allergy alerts ( Drug-drug interaction alerts ( Drug-lab interaction alerts ( Drug dosing support ( No. (%) hospitals 10 (10.3) 9 (9.3)
any EMRs and relied completely on paper records. Ten of the participating hospitals (10.3%) were classified as having comprehensive EMRs and implemented the system in all major units. The remaining 15 hospitals (15.5%) were classified as having a basic EMR system (Table 1). Of a total of 104 hospitals in Jordan, 97 (93.2%) agreed to participate in the study. Of the total participating hospitals, 56.7% were private hospitals. Only one university hospital participated in the study. Most participating hospitals (56, 57.7%) were small hospitals with a capacity of less than 100 beds. Most of the hospitals (62.9%) were located in the middle region, where the capital is located. Private hospitals accounted for most of the participating hospitals (56.7%) in this survey. Most of the participating hospitals (64.9%) were nonteaching hospitals (Table 2). Bivariate analysis revealed that the adoption rate of EMRs was significantly associated with the hospital size (χ2 = 14.7, P = .005). Small hospitals (
