Article Text
PDF
Abstract
Objectives A thorough understanding of user needs and behavioural intent-to-use underpins the development of a responsive health information system. This study aimed to examine health workers’ intent-to-use an electronic health record (EHR) system in an urban, rural and remote setting in the Philippines.
Methods Following the Unified Theory of Acceptance and Use of Technology framework, user acceptance and the factors influencing intent-to-use the EHR were examined through a self-administered questionnaire. A total of 128 EHR users, comprising physicians, nurses, midwives, barangay health workers and administrative staff, were surveyed. Median scores for each domain were compared across the sites using the Kruskal-Wallis test. Ridge regression analysis was used to identify factors associated with behavioural intent-to-use.
Results Over 94% of users across all sites reported their intent-to-use the EHR in the near future. Of the seven predictor variables examined, only self-efficacy was found to be significantly associated with behavioural intent-to-use. Intent-to-use the EHR increased by 31% (p=0.007) for each unit increase in self-efficacy score among participants.
Discussion Acceptance was high across the three sites, with self-efficacy being a predictor of intent-to-use the technology. This suggests that users are more likely to adopt an EHR if they believe they have the capacity to successfully navigate the technology and perform their designated tasks with it.
Conclusion Co-producing interventions with primary care providers is crucial in sustaining EHR systems. Rather than developing a technology based on the surveillance needs of policymakers, an EHR developed from the grassroots was shown to be well-received by end-users.
- Electronic Health Records
- Primary care
- Health services research
- Information technology
- Patient-centred care
Data availability statement
Data are available upon reasonable request. Data from this study are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
- Electronic Health Records
- Primary care
- Health services research
- Information technology
- Patient-centred care
WHAT IS ALREADY KNOWN ON THIS TOPIC
Electronic health record (EHR) systems have garnered considerable research attention in high-income settings, revealing their positive impact on health system integration.
However, the feasibility and user acceptance of EHRs in resource-limited settings that have still relied on paper-based systems remains under-represented in existing literature.
WHAT THIS STUDY ADDS
In a cross-sectional study conducted across three diverse Philippine contexts—urban, rural and remote—high user acceptance towards the transition to EHR systems was observed amidst infrastructural barriers experienced.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The findings of this study demonstrate the high acceptability of user-oriented EHR designs in resource-limited settings, which has the potential to augment operational efficiency and service quality given sustained use.
As such, policymakers or implementers targetting the expansion of health information systems in resource-limited settings must consider user acceptance and readiness in intervention design.
Introduction
Fragmentation in service delivery networks (SDNs) has contributed to a decline in the quality of healthcare services in the Philippines.1 According to the Philippine Health System Review, the lack of a facilitated patient referral system has led to disintegrated SDNs and patients bypassing the primary level of care.2 To address these, the Universal Health Care (UHC) Law of 2019 was passed in the Philippines to strengthen primary care with the development of continuing, comprehensive and coordinated health information system (HIS) as one of its key components.3 Its implementation requires multiple electronic health records (EHRs) to be interoperable for consolidating and allowing the exchange of health information across SDNs. An effective EHR system is, therefore, expected to support SDNs by providing centralised patient information available across multiple access points.4
EHR systems can contribute towards process optimisation and improved patient outcomes in primary care settings. Ongkeko et al reported that EHR systems increased work efficiency, improved data quality and enhanced morale among rural health unit workers in the Philippines.5 Despite its benefits, structural challenges have inhibited the transition from paper to digital records. Barriers to adoption often include inadequate capacity building, user resistance and difficulties in fulfilling existing HIS regulations.6 These obstacles are exacerbated by the latent vulnerabilities posed by weak infrastructure in remote or underserved settings.6 7
Understanding user acceptance and the factors influencing EHR adoption is necessary for the full realisation of UHC. User acceptance of information technology is defined as ‘the demonstrable willingness within a user group to employ information technology for the tasks it is designed to support.’8 Since the 1960s, multiple frameworks have been developed to assess user acceptance of technological innovations. In 2003, Keane et al integrated eight models to form the Unified Theory of Acceptance and Use of Technology (UTAUT).9 The UTAUT model has since been applied to various user acceptance studies, including the acceptance and use of EHR platforms in resource-limited settings.10
The Philippine Primary Care Studies (PPCS) programme worked to address system fragmentation and establish a robust primary care system through various interventions. One of its interventions was to pilot an EHR system in three settings in the Philippines: an urban site (University of the Philippines Health Service, Diliman), a rural site (Municipality of Samal, Province of Bataan) and a remote site (Municipality of Bulusan, Province of Sorsogon). Acknowledging the existing barriers to EHR adoption, PPCS provided technical resources and ensured that EHR development focused on healthcare worker (HCW) needs rather than the surveillance needs of policymakers. This led to a user-oriented format that was designed to be compatible with existing workflows and thought processes.
The present study aimed to examine the behavioural intent-to-use an EHR among healthcare providers in an urban, rural and remote setting in the Philippines. Additionally, we identified the factors associated with behavioural intent-to-use an EHR system.
Methodology
Study context, design and population
This cross-sectional study was undertaken across three distinct geographical locations in the Philippines. Each setting featured a comprehensive healthcare provider network encompassing outpatient facilities, satellite centres, laboratories and pharmacies. Prior to the implementation of the PPCS programme in 2017 for the urban site and 2019 for the rural-remote cohort, these networks were reliant on self-initiated paper-based records. This method of operations lacked a unifying digital backbone and operated primarily through manual or person-to-person interactions. While a paper-based approach can enable facilities to customise workflows according to their unique contexts, relegating the responsibility of coordinating care to patients and healthcare providers can compromise quality of care. As reported in prior studies, fragmentation in operations has been observed to complicate patient journeys and ultimately disrupt care plans.2
The present study used a self-administered questionnaire to survey all 128 HCWs who used the PPCS EHR in the three primary care sites (100% response rate). Of these HCWs, 37 were from the urban site, 48 were from the rural site and 43 were from the remote site. These included physicians, nurses, midwives, barangay health workers (ie, lay community volunteers) and administrative staff. In the urban site, the survey was conducted in March 2019, 3 years after the deployment of the EHR system. In the rural and remote sites, the survey was conducted in February 2020, 10 and 9 months after the implementation of their EHR systems, respectively.
Implementation of the EHR system
PPCS developed an EHR system for its research sites to accomplish the following: (1) to streamline patient record-keeping, (2) to automate repetitive tasks like prescribing drugs and diagnosing diseases based on the International Classification of Diseases, (3) to manage public health surveillance, (4) to monitor the quality of care and (5) to track and validate expenses from a pilot run of the Philippine Health Insurance Corporation Primary Care Benefit package—a financing model that supported universal health coverage. The EHR system was deployed at the urban site in October 2016, then at the rural and remote sites in April 2019 and May 2019, respectively. Throughout the operationalisation of the EHR system, feedback from end-users at the facilities was regularly elicited and integrated into its development.
The EHR system used the Open Medical Records System, a collaborative open-source platform. Hands-on EHR training workshops were conducted to support HCWs during the transition from paper to digital records. During these workshops, case studies were presented to simulate situations for participants to navigate various health service pathways. These pathways included the patient queue, the Subjective Objective Assessment Plan chart, and pharmacy, laboratory and administrative data entry modules. Technical interventions to support implementation of the EHR system included the installation of local database servers in the urban and rural sites, and a cloud server in the remote site. The university health service (for the urban site), rural health units, barangay health stations and SDN partners such as pharmacies and diagnostic centres (for the rural and remote sites) were connected to their corresponding servers. Routers, laptops and printers were likewise provided for all facilities.
Research framework and hypotheses
Venkatesh et al proposed the UTAUT framework to measure the factors influencing user acceptance of new technologies. The UTAUT model initially included seven domains influencing Behavioural Intention, a key criterion in user acceptance and a significant predictor of usage behaviour.11 The seven domains were Performance Expectancy, Effort Expectancy, Social Influence, Facilitating Conditions, Attitude Towards Technology, Self-Efficacy and Anxiety.
In Venkatesh et al’s study, only our domains of the UTAUT model were shown to have a significant influence on Behavioural Intention, namely, Performance Expectancy, Effort Expectancy, Social Influence and Facilitating Conditions. On the other hand, the influence of Attitude Towards Technology, Self-Efficacy and Anxiety on Behavioural Intention was found to be statistically insignificant. Venkatesh et al also hypothesised that age, sex and length of EHR experience of end-users were significant moderators on the influence of each domain.11 Another study by Conrad and Munro suggested that the inter-relationship in three domains, namely, Attitude Towards Technology, Self-Efficacy and Anxiety, played an important role in technology acceptance.12 Specifically, the Conrad and Munro study showed that improved self-efficacy contributed to more positive attitudes towards technology and reduced anxiety.
For the present study, all seven predictor domains were included to re-examine their influence on Behavioural Intention as anticipated by the UTAUT model. A description of the domains used in the study is summarised in table 1. Age, sex and length of EHR experience of end-users were included in the analysis as demographic covariates.
Description of UTAUT domains
Questionnaire development
A previously validated UTAUT questionnaire was adapted for this study (see online supplemental appendix A).11 The first section included sociodemographic data such as age, sex, role in the health system, years in service and length of EHR use. The second section contained items from the eight UTAUT domains with a total of 31 statements. These items were measured using a 4-point Likert scale (1=strongly agree, 2=agree, 3=disagree, 4=strongly disagree).
Supplemental material
To evaluate the internal consistency of the questionnaire, a Cronbach’s alpha reliability test was conducted (table 2). A Cronbach’s α value exceeding 0.70 is generally understood to suggest satisfactory reliability. However, it is worth noting that other studies have also contended that α thresholds between 0.60 and 0.70 are an indication of acceptable internal reliability.12 13 In this study, results from the reliability test showed that seven of the eight domains had acceptable reliability (see table 2). The Facilitating Condition domain revealed poor reliability, with a Cronbach’s α of approximately 0.545. On inspection, removing item FC3 improved internal reliability for this domain to an α of 0.764. Responses for the FC3 item were, therefore, excluded in the final regression model based on a priori intent.
Results of Cronbach’s alpha reliability test
Statistical analyses
Responses to items were dichotomised on analysis by combining ‘strongly agree’ and ‘agree’ responses into ‘generally agree’, and ‘strongly disagree’ and ‘disagree’ into ‘generally disagree.’ Negatively worded statements in the questionnaire, specifically items in the Anxiety domain, were reverse-coded. Demographic data were presented in frequencies and percentages for categorical variables. Agreement to UTAUT domain statements was summarised in frequencies and percentages of ‘generally agree’ responses over the total number of responses. Median scores for each domain across the three locations were also compared using the Kruskal-Wallis test. Ridge regression was used to identify factors that are significantly associated with intent-to-use. The tuning parameter for the model was set to 0.024 to address multicollinearity issues. Data were analysed using Stata V.12.0 and SAS V.9.4.
Ethical considerations
Participants did not receive any form of financial incentive to answer the questionnaire. Data were collected and analysed on a confidential basis. Written informed consent was obtained from all respondents.
Results
Demographics
Table 3 shows the demographic characteristics of respondents across all three research sites. The majority of respondents were female (85%) and were below the age of 40 years (55%). While the urban site had the greatest number of physician respondents at 35%, midwives were most represented in the rural (42%) and remote sites (28%). Length of tenure was shortest in the remote site, with 77% of remote staff being in service for less than 5 years. Urban respondents reported the most extensive EHR usage, with 87% of HCWs having used the EHR system for over 12 months by the time the survey had been undertaken.
Demographics of respondents
Overall UTAUT domain percentage agreement of EHR across primary care settings
Table 4 shows the percentage of general agreement of respondents per site for each item under the eight UTAUT domains. Respondents from the urban site had the lowest percentage of general agreement in four domains, namely, usefulness at work (PE1-4) ≤76%, positive attitude towards the EHR (ATT2-4) ≤76%, availability of technical support (FC1-2,4) ≤92% and personal competence (SE1-3) ≤95%. Respondents from the rural and remote sites showed higher levels of percentage agreement in these four domains.
Percentage of ‘generally agree’ responses across the three sites
All sites showed high levels of agreement for two domains—ease of use and social desirability. The percentage agreement for ease of use (EE1-4) ranged from 94% to 100%, while the percentage agreement for social desirability (SI1-4) ranged from 91% to 100%. Apprehension was generally low across all sites. Only 14% of urban, 29% of rural and 46% of remote site respondents regarded the EHR as intimidating (AN4). Overall intent-to-use the EHR was high at 96–98% across all sites.
Comparison of domain scores across primary care settings
The median scores for all of the UTAUT domains significantly differed across the three sites (table 5). Rural site respondents had the highest median agreement scores (1–1.5) in all domains except for the Anxiety domain (2.25). Respondents from the remote site were the most anxious about EHR use, with a median score of 2.5 (range 1–4), followed by the rural site at a median score of 2.25 (range 1–4). The urban site respondents had the least anxiety towards EHR use at a median score of 2 (range 1–3).
Comparison of median domain scores among sites
Factors influencing EHR acceptance and use
Only self-efficacy (personal competence) was found to be significantly associated with intent-to-use the EHR after adjusting for confounding (table 6). Intent-to-use increases by 0.310 arbitrary units for every 1 arbitrary unit increase in self-efficacy while controlling for other variables (β=0.310, adj. p=0.007).
Regression model of UTAUT domains on Behavioural Intention
Discussion
This study demonstrated that intent-to-use and user acceptance of the EHR system was high for respondents in urban, rural and remote sites. We used the UTAUT model to assess the acceptance of healthcare workers of the EHR system, with intent-to-use as the primary outcome variable.11 Of the seven domains evaluated, only self-efficacy was found to be significantly associated with intent-to-use. Intent-to-use increased by 31% for every unit increase in self-efficacy among health workers.
The findings of this study differed from the findings of prior studies that also used the UTAUT model, where other factors such as usefulness at work and social desirability had a more prominent effect. In the study by Shiferaw and Mehari in 2019 that evaluated EHR acceptance and use in Ethiopia, usefulness at work, social desirability and ease of use were significantly associated with intent-to-use.10 Another study by Ayaz and Yanartaş in 2020 reported that 61% of the intent-to-use of EHR was explained by usefulness at work and social desirability.13 This difference may be due to several factors, including differences in EHR system used, implementation of the EHR (eg, provision of training and assistance) and sociocultural differences. In addition, a variable that has uniformly high results is unlikely to be predictive of an outcome. In this study, ease of use and social desirability had the highest percentage of agreement across all sites. This may explain why these two domains were not found to be significant predictors of intent-to-use. This may have resulted in a loss of power to detect determinants of this outcome.
The EHR system in this study was developed particularly with the HCWs in mind, with the primary intention to reduce their workload and improve work efficiency. The study results demonstrate that the EHR system was able to achieve this goal since the study respondents had the highest percentage of agreement for the items related to ease of use. Among the respondents in this study, 96% in the urban site and 98% in the rural and remote sites agreed that they will use the EHR if support and assistance were available (SE2 item). This suggests that the confidence of HCWs in their ability to fulfil their tasks through the EHR system is bolstered by the availability of assistance. Several studies report that the provision of technical support for early end-users is vital in the implementation of new technologies. This enables end-users to gain computer experience and the ability to use the technology in diverse ways, thus boosting their perceived personal competence.14–16
Despite varied responses on the seven UTAUT predictor domains, user acceptance of the EHR system remained high across all three sites. Over 97% of HCWs in urban, rural and remote sites intended to use the EHR in the next 12 months (BI4). The intention of HCWs to use the EHR system and their expected actual use of the platform could potentially facilitate the integration of service delivery networks and ultimately improve patient experience.17 As reported in a literature review by Uslu and Strausburg, 78% of studies probing into the influence of EHRs on patient care revealed its positive impact on augmenting quality care. Moreover, 56% of studies included in the review reported a notable reduction in system costs after EHR implementation.18
EHR acceptance bolsters utilisation, thereby maximising the potential impact of the technology on service delivery. McAlearney et al showed that EHRs bring operational efficiencies in clinical processes in urban areas—settings that encounter increasingly complex and dense patient populations.19 In rural and remote settings in the Philippines, research has shown that the introduction of an EHR system still improved information management and data protection. These results were mirrored in the present study which likewise revealed that EHRs can be accepted by end-users provided adequate infrastructural and operational support.5 Therefore, despite the heterogeneity of environments among these areas, the use of EHRs has been demonstrated to be beneficial in integrating SDNs and fostering continuity of care.
Limitations
Several limitations were encountered in this study. First, the adapted questionnaire was not translated into Filipino or the lingua franca spoken in the regions of the study settings. The translation of questionnaires is recommended since this minimises selection bias favouring those who are literate in English. In this study, we opted to retain the questionnaire in its English form since the HCWs in the study sites were bilingual. Second, Chyung et al suggested not to use a mix of positively and negatively worded items ‘as doing so can create threats to validity and reliability of the survey instrument.’20
Self-reported scores were used for measuring computer experience and proficiency. The respondents of the survey were HCWs who remained employed in their respective facilities from the introduction of the EHR to the first point of data collection. These HCWs may be more likely to give a more positive perception of the EHR system. As such, survivor bias may have had an impact on the results given the high staff turnover in the remote setting. It is crucial to recognise that the conclusions drawn from this cross-sectional study are pertinent solely to the contexts under investigation. While the results reported bear resonance to existing literature on EHR acceptance and can potentially inform HIS policy in low-middle-income settings, the extent to which these findings can be generalised to other settings remains uncertain. Furthermore, as no baseline assessment of health workers’ experience with paper-based records was conducted, this study could not quantify the impact of the EHR implementation on service efficiency in relation to the users’ previous experience.
Significance of the study and recommendations
Prior research, both within and beyond the studied geographical regions, has highlighted the importance of further integrating healthcare provider networks to enhance service quality by improving system responsiveness and care coordination.2 The implementation of an EHR can stand as a key intervention for facilitating operational integration across healthcare facilities. Yet despite its potential in augmenting service efficiency19 and reducing system costs,18 the operational success of these systems is still informed by user experience and acceptance of the technology. This suggests that, to an extent, the effectiveness of an EHR is inextricably linked with the ways users engage with the system.
This paper investigates the factors influencing user motivations to use an EHR system. While baseline satisfaction towards paper-based systems was not within the purview of this research, the present study outlines the need to focus on designing EHRs that are intuitive and self-efficacious. In practice, this can entail a targeted approach towards accommodating different user types and ensuring each user is thoroughly equipped with the necessary resources for the transition. Designing user-specific portals during the development phase and providing hands-on training before and after deployment could enhance user acceptance. Future research could extend these findings by investigating the impact of EHR systems on key quality of care indicators and performance metrics. These include, but are not limited to, reduced wait times, fewer data errors and increased patient satisfaction.
Conclusion
This study demonstrated that while responses towards UTAUT predictor domains considerably varied, there was high end-user acceptance of the EHR system across settings. This translates to a higher potential for actual usage of the system and consequently allows for the realisation of integrating health systems and strengthening primary care through EHR systems. Of the seven UTAUT domains, only self-efficacy (perceived personal competence) was shown to have a direct and significant association with intent-to-use the EHR.
The implementation of an EHR system in primary care facilities in the Philippines serves as a critical mechanism for integrating the currently fragmented SDNs. Close collaboration with primary care providers is integral in operationalising EHR use across diverse contexts. The co-development of tools to improve workflows and joint decision-making in introducing new technologies must be considered given the specifications of each locality. Ascertaining infrastructural compatibility and engaging end-user needs are ultimately vital in improving EHR interventions in underserved settings.
Data availability statement
Data are available upon reasonable request. Data from this study are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by the University of the Philippines Manila Research Ethics Board (UPMREB) under the code 2015-489-012 and The Philippines’ Department of Health Single Joint Research Ethics Board (SJREB-2029-55). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
We would like to extend our sincerest gratitude to the University of the Philippines Health Service and the Municipalities of Samal and Bulusan for their collaborative efforts in co-developing the on-site EHR system and for their continued partnership with PPCS. Additionally, we would like to express our appreciation to Maria Rhodora N. Aquino for her exceptional management of the study's financial matters, which significantly contributed to the successful implementation of the PPCS programme. Lastly, we wish to acknowledge the unwavering support provided by our dedicated data collection team, namely, Herbert Zabala, Johanna Faye E. Lopez, Zharie P. Benzon, Karl Engelene E. Poblete, Ryan M. Esconde, Miguel Francisco F. Callo, Criselle D. Malibiran and Chad Lester H. Lastrilla.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors RYHDM led manuscript development, developed the manuscript, wrote sections of the first draft and oversaw editorial revisions. CLG wrote sections of the first draft and oversaw substantial editorial revisions. CSCT-L wrote sections of subsequent drafts, appraised the manuscript and oversaw editorial revisions. MAUJ conducted the statistical analysis for the study and provided feedback on the methodology. JMSP and NMF provided substantial revisions on the subsequent drafts of the manuscript. YC participated in data collection and suggested revisions for the manuscript. MPR, NB-S, JTS, LD and RC provided technical oversight on the study design and suggested revisions for the manuscript. ALD oversaw study design, objective setting, provided substantial edits to the manuscript and led the intervention introduced in the study. ALD is the guarantor for the overall content and conduct of the study.
Funding The study was supported by The Philippine Department of Health, the Philippine Health Insurance Corporation, the Emerging Interdisciplinary Research Program, the University of the Philippines Center for Integrative and Development Studies (UP-CIDS, Grant No. C07-001), and the Philippine Council on Health Research and Development. The authors of this study have no conflicts of interest to declare. Moreover, patients were not involved in the design, conduct, reporting or dissemination plans of this research.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.