Structure and process of the patient records computerization project in hospitals of Burundi

INTRODUCTION

The health information system (HIS) is one of the pillars of the health system, providing reliable and timely data to enable appropriate decision-making in the day-to-day management of a health facility. The computerization of the patient record is one of the means of making the information system within a hospital efficient.^1,2

With the emergence of new information and communication technologies, most Sub-Saharan African countries have implemented computerized patient record projects. Nevertheless, it has been observed that those projects remain limited by multiple factors, the main ones being the cost of equipment, human resource skills, the unilateral adoption of software without consulting professionals, which leads to the social unacceptability of the tool deployed, the absence of the prerequisite’s evaluation for the implementation of the project, the absence of strategic planning at the beginning of the project, etc.^3-5

In Burundi, the government's objective was to strengthen the HIS by using new information and communication technologies. Many actions have been taken to achieve this goal, including the implementation of DHIS 2 (District health information software version 2), which is a routine health data management software. This software, which is only limited to the management of aggregated data, should be completed by other software that allows the management of individual patient information and networking within a hospital. It is in this context that a project to computerize patient records using the OpenClinic GA software was initiated in Burundi hospitals. OpenClinic GA is an OpenSource software for hospital information management. It was developed by a team of doctors and computer scientists from the company MXS SA of Belgium since 2006. It can be redistributed or modified, and marketable versions are also available.⁶

Since the implementation of the computerization project, no evaluation has been made. In order to identify the challenges to be met to ensure the success and the sustainability of the patient record computerization project, it was important to evaluate the six components (resources, indicators, sources, management, information products, and data dissemination and use) of the HIS according to the method of the health metrics network, the results of which are already available.^7,8 On the other hand, it was necessary to evaluate the implementation, hence the interest and objective of the present study. The objective of this study is to evaluate the implementation (structure) and the implementation (process) of the patient record the computerization project in the Burundi hospitals.

METHODS

Study design

This is a descriptive cross-sectional study in 11 hospitals whose patient records were computerized between 2015 and 2017 with the OpenClinic GA software. Beneficiaries of the ENABEL-funded project and belonging to the 3 levels of the health pyramid in Burundi, were included in the study. The data collection was done in the February 2020.

Study instruments

An evaluation grid composed of 64 items was used to evaluate the two components, namely the "structure" and the "process" of the computerization project. The grouping of elements and items in the two components was based on the evaluation model proposed by Donabedian.⁹ The "structure" component included 4 sub-components and 21 elements, while "process" component included 4 sub-components and 43 elements that were evaluated (Table 1).

Data collection and analysis

In each hospital, 3 key informants were identified to answer the questions of this evaluation grid: the director of care, the head of the nursing department and the person in charge of the health information system. For each item, the evaluation was based on scores ranging from zero to three (0=not at all satisfactory; 1=moderately satisfactory; 2=satisfactory and 3=very satisfactory). The selected answer was the result of a consensus of the 3 selected key informants (director of care, head of nursing and health information system manager). If there was a discrepancy between the 3 key informants, all responses were retained to calculate an average score. The average score for each component was calculated, reported to the maximum score of three, and converted into quartiles: 0-25%=not at all satisfactory; 26-50%=moderately satisfactory; 51-75%=satisfactory; and 76-100%=very satisfactory. The judgment criteria were based on the HIS evaluation tool of the health metric network.⁷ The first comparison variable was the year in which the computerized patient record began to be used, the second comparison variable was the referral level of the study hospitals. Data entry, processing and analysis were performed with the excel tool. Mean scores were calculated for each component, before calculating them by hospital group.

The assessment of the completeness of the data filled on the patient record was done with the PRISM (Performance of routine information system management) tools, whose section on data completeness was adapted for the present evaluation.¹⁰ It focused on a single item "principal diagnosis at discharge". It was carried out in four hospital services chosen in a reasoned way from the services already computerized in all 11 study hospitals, namely the gynecological-obstetrical consultation service, the pediatric service, the adult consultation service, and the emergency service. One fifth (1/5) of the medical records were randomly selected in each hospital service, for the four months (March, June, November, December) also randomly selected from the 12 months of the year 2019. Data entry, processing, and analysis were performed with the excel tool. The percentage of records with the item "main diagnosis at discharge" was calculated by service using EPI INFO 7.2 software, the level of completeness was compared between hospitals according to the reference level and according to the year in which the computerized patient record was introduced, using the Pearson Chi-2 statistical test with a p value of 0.05. Odds ratios were also calculated to determine the probability of finding the item "principal diagnosis at discharge" for a group of hospitals.

Ethical approval

This article is part of a research project entitled Evaluation de l'informatisation du système d'information hospitalier: structure, processus, effets, enjeux et perspectives au Burundi (Evaluation of the computerization of the hospital information system: structure, process, effects, issues and perspectives in Burundi which has received the approval of the national ethics committee for the protection of human subjects of biomedical and behavioral research.

RESULTS

Structure assessment

Overall, the structure component was moderately satisfactory with a mean score of 53%. Governance (coordination, availability of standards and planning documents) was the lowest scoring component with a score of 21% while availability of materials and equipment was the highest scoring component with a score of 72% (Figure 1).

Regarding the governance sub-component, the results showed that none of the hospitals had the standards and procedures manuals (guidelines) for managing the electronic patient record (0%). For the reorganization of services before the start of computerization, the score obtained was 55%. With regard to the availability of human resources, the score obtained for the availability of qualified human resources with skills in health information science and computer science was 51%, while the capacity building of personnel at the beginning of the project obtained a score of 58%. For financial resources, the contribution of the project's beneficiary hospitals in terms of staff recruitment scored 91%, while the existence of strategies for mobilizing funds scored 27%. Regarding material resources and equipment, the availability of the server and data security scored 74% and the availability of computers (desktop/laptop/tablet) scored 65% (Table 2).

Process evaluation

The process component was satisfactory with a score of 51%. Maintenance had the lowest average score of 44%, while the management of the computerized patient record obtained the highest average score of 66% (Figure 2).

For the coordination, support, monitoring and evaluation sub-component, supervision of hospital offices by the health districts or provinces and/or the central level scored 18%, while support from partners scored 100%. With regard to the management of the electronic patient record (EPR), the use of documents on the standards and procedures for managing the EPR scored 37%. In terms of human resource capacity building, capacity building in network and computer maintenance scored 12% and capacity building in software use scored 100%.

Arrangements and activities to maintain servers, local area networks, OpenClinic GA software, and computer equipment scored 37%, 49%, 40%, and 51%, respectively (Table 3).

Evaluation of the level of completion of the computerized patient record

Of the 11 hospitals, the gynaecology-obstetrics service had 68% of the records bearing the diagnosis at discharge, whereas the emergency service had only 28% (Table 4).

The results of the comparison of the level of patient record completion by year of the project’s start showed that hospitals that were computerized in 2015 were more likely to have the patient records with the diagnosis at discharge than hospitals computerized in 2017 for all four services. The large difference was observed for the emergency service (OR: 12.5 [10.9 -14.4]; p<10-4). Regarding the referral level of the study hospitals, the 1st referral level hospitals (district) were more likely to have the patient records with the diagnosis at discharge than the 2^nd and 3^rd level hospitals (regional and national level). The large difference was observed for the emergency services (22.1 [19.3 -25.2]; p<10-4) (Table 5).

*Significant at p<0.05.

DISCUSSION

The average score recorded for the "structure" component was 53% and was at the lower end of the third quartile (Q3). The element of this component that recorded a low average score was governance. During the assessment, it was noted that none of the hospitals had the standards and procedures manuals (guidelines). The literature has shown that managing the diversity of stakeholders in the provision of care and services within a hospital, the completeness and consistency of patient information, the balance of decision-making and good coordination within the institution requires standards and procedures documents. Thus, the absence of such a document creates a significant gap in the process of implementing the project of computerization of the patient record.^11,12 In addition, two of the eleven hospitals had an action plan containing the implementation, monitoring and evaluation activities of the computerization project. However, this plan was not regularly used for monitoring and evaluation of implementation activities. The absence of a coordination framework in the majority of hospitals could also explain the low score recorded for this component (Table 2).

The "process" component was satisfactory with a score of 51% and was also at the lower end of the third quartile (Q3). Human resource capacity building (46%) and maintenance (44%) were the sub-components that had the lowest average score. Indeed, capacity building in maintenance had the lowest average score of 12%. The explanations given by the providers showed that the hospitals use the software designers for maintenance. As for development and update of patient record, the average score obtained was 25%. The hospitals' IT staff did not have the capacity to add the necessary information to the software, and any changes to the software required external assistance. Although there have been trainings aimed at empowering the IT staff, there is still a need for capacity building. Concerning the maintenance, the strategies adopted by the hospitals to ensure the maintenance, especially of servers, local networks and software, were deficient. The skills of the hospitals' IT staff were limited, and they had to seek external support, which either required a significant budget or did not respond in time (Table 3).

Comparing the two hospital groups by year of EPR introduction, the results showed that those computerized in 2015 scored higher than those computerized in 2017 for the structure component, 54% versus 52%. The deployment of resources would be greater in the pilot phase than in the expanded implementation of the software. In contrast, for the "process" component, hospitals that were computerized in 2017 had a higher score than those computerized in 2015, 52% versus 50%. The experience of the hospitals in the pilot phase undoubtedly contributed to making some improvements during the project extension process.

Assessment of the level of completion of the electronic patient record

Regarding the assessment of the level of completion of the patient records, the low proportion recorded for emergency services (28%) could be explained by the increased frequentation by patients requiring emergency care, so the providers' concern would be focused on the provision of care rather than on the correct filling in of patient information.^13,14 Normally, computerization of the patient record should improve the completion of the record. However, some studies have identified problems with the completion and completeness of computerized patient records. Factors identified that could explain incomplete completion of the EPR include work overload, lack of technical knowledge, lack of sufficient computer equipment, the impossibility of real-time data entry at the patient's bedside, and certain computer system malfunctions.^12,13,15

CONCLUSION

The evaluation of the structure and process of the computerization project in Burundi revealed that the availability of standard documents and procedures, coordination, planning and maintenance of equipment and materials are aspects that require the combined efforts of the stakeholders. In addition, the use of the electronic patient record has not made its completion effective. Improving planning, the level of collaboration between actors at different levels of the health system and capacity building strategies for users will help consolidate the achievements and guarantee the success and sustainability of the patient record computerization project in Burundi hospitals.

ACKNOWLEDGEMENTS

Author would like to thanks coordinators of national institute of public health (NIPH) support project funded by European Union for technical and financial support, and to the data managers and computer specialists of the ministry of public health and the fight against AIDS for data collection.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. Chougrani S, Ouhadj S, Agag F. Evaluation of the Oran university hospital information system. Santé Publique (Paris). 2013;25(5):617-26.
2. 2. AbouZahr C, Boerma T. Health information systems: the foundations of public health. Bull World Health Organ. 2005;83(8):578-3.
3. 3. Sory ST, Baguiya A, Coulibaly A, Kouanda S, Sondo B. Evaluation de la performance du système d’information hospitalier du Centre Hospitalier Universitaire Yalgado Ouédraogo de Ouagadougou, Burkina Faso. Sci Tech Sci la santé. 2017;40(1):135-50.
4. 4. Senafekesh B, Tesfahun Y, Mulusew A, Binyam T. Health Professionals readiness to implement electronic medical record system at three hospitals in Ethiopia: a cross sectional study. BMC Med Inform Decis Mak. 2014;14(115):1-8.
5. 5. Kruse CS, Smith B, Vanderlinden H, and Nealand A. Security Techniques for the Electronic Health Records. J Med Syst. 2017;41(127):1-9.
6. 6. République du Burundi. Ministère de la Santé Publique et de la lutte contre le Sida. Plan National de Développement de l’Informatique de la Santé (PNDIS). 2015;26-9.
7. 7. Donabedian A, Wheeler JR, Wyszewianski L. Quality, cost, and health: an integrative model. Med Care. 1982;20(10):975-92.
8. 8. Munezero F, Ahanhanzo GY, Charles SJ, Leodegal B. Hospital Information System in the Context of the Use of Electronic Medical Record in Burundi. Universal J Public Health. 2021;9(2):94-101.
9. 9. World Health Organization-WHO. Framework and standards for country health information systems. Second edition. Health Metrics Network. 2008;15-45.
10. 10. Anwer Aqil BA, Lippeveld T, Moussa T. Outils PRISM: Guide de l’Utilisateur. 2012 ;41-7.
11. 11. Cucciniello M, Lapsley I, Nasi G, Pagliari C. Understanding key factors affecting electronic medical record implementation: A sociotechnical approach. BMC Health Serv Res. 2015;15(268):1-19.
12. 12. Msiska KEM, Kumitawa A, and Kumwenda B. Factors affecting the utilisation of electronic medical records system in Malawian central hospitals. Malawi Med J. 2017;29(3):247-53.
13. 13. Ologeanu-Taddei R, Paré G. Technologies de l’information en santé : un regard innovant et pragmatique. Systèmes d’information Manag. 2017;22(1):3-8.
14. 14. Kavuma M. The usability of electronic medical record systems implemented in sub-Saharan Africa: A literature review of the evidence. J Med Internet Res. 2019;6(1):1-27.
15. 15. Jawhari B, Ludwick D, Keenan L, Zakus D, Hayward R. Benefits and challenges of EMR implementations in low resource settings: A state-of-the-art review. BMC Med Informatics Decision Making. 2016;16(116):1-12.