Knowledge From the Afrikan World, For the World

PMC Articles

Facilitators and Barriers in the Implementation of a Digital Surveillance and Outbreak Response System in Ghana Before and During the COVID-19 Pandemic: Qualitative Analysis of Stakeholder Interviews

PMCID: PMC10625076

PMID: 37862105

Abstract

Background In the past 2 decades, many countries have recognized the use of electronic systems for disease surveillance and outbreak response as an important strategy for disease control and prevention. In low- and middle-income countries, the adoption of these electronic systems remains a priority and has attracted the support of global health players. However, the successful implementation and institutionalization of electronic systems in low- and middle-income countries have been challenged by the local capacity to absorb technologies, decisiveness and strength of leadership, implementation costs, workforce attitudes toward innovation, and organizational factors. In November 2019, Ghana piloted the Surveillance Outbreak Response Management and Analysis System (SORMAS) for routine surveillance and subsequently used it for the national COVID-19 response. Objective This study aims to identify the facilitators of and barriers to the sustainable implementation and operation of SORMAS in Ghana. Methods Between November 2021 and March 2022, we conducted a qualitative study among 22 resource persons representing different stakeholders involved in the implementation of SORMAS in Ghana. We interviewed study participants via telephone using in-depth interview guides developed consistent with the model of diffusion of innovations in health service organizations. We transcribed the interviews verbatim and performed independent validation of transcripts and pseudonymization. We performed deductive coding using 7 a priori categories: innovation, adopting health system, adoption and assimilation, diffusion and dissemination, outer context, institutionalization, and linkages among the aspects of implementation. We used MAXQDA Analytics Pro for transcription, coding, and analysis. Results The facilitators of SORMAS implementation included its coherent design consistent with the Integrated Disease Surveillance and Response system, adaptability to evolving local needs, relative advantages for task performance (eg, real-time reporting, generation of case-base data, improved data quality, mobile offline capability, and integration of laboratory procedures), intrinsic motivation of users, and a smartphone-savvy workforce. Other facilitators were its alignment with health system goals, dedicated national leadership, political endorsement, availability of in-country IT capacities, and financial and technical support from inventors and international development partners. The main barriers were unstable technical interoperability between SORMAS and existing health information systems, reliance on a private IT company for data hosting, unreliable internet connectivity, unstable national power supply, inadequate numbers and poor quality of data collection devices, and substantial dependence on external funding. Conclusions The facilitators of and barriers to SORMAS implementation are multiple and interdependent. Important success conditions for implementation include enhanced scope and efficiency of task performance, strong technical and political stewardship, and a self-motivated workforce. Inadequate funding, limited IT infrastructure, and lack of software development expertise are mutually reinforcing barriers to implementation and progress to country ownership. Some barriers are external, relate to the overall national infrastructural development, and are not amenable even to unlimited project funding.

Full Text

In the advent of affordable and robust IT tools in the 1990s, high-income countries began to switch from paper-based to electronic disease surveillance [1]. The pace of digitalization in public health surveillance and outbreak response has since gained momentum and spread to low- and middle-income countries (LMICs) [2]. This transition is, in part, a result of digital penetration, consolidated by the demonstrated utility of digital tools in public health practice [3-5]. Well-established electronic disease surveillance systems have many benefits. Some of these benefits are improved data quality (completeness, detail, and validity), timeliness of reporting, data standardization, automation of data analysis and visualization, prompt identification of public health events, and timely information sharing for public health action [6,7]. Thus, the benefits of digitalization in public health practice are many but so are the challenges. Frequently reported challenges of adopting electronic health systems include the technical challenges of technology, absence of basic necessary infrastructure such as electricity, implementation costs, workforce capacities and attitudes, and organizational factors [8-10]. These challenges are dynamic and vary in scope and complexity across geopolitical, socioeconomic, cultural, and organizational settings [10-12]. Therefore, some electronic systems have not made it beyond pilots to sustained nationwide institutionalization, especially in LMICs [10,13-18].
The cross-border geographical spread of outbreaks and the multidisciplinary response requirements made it more apparent than ever that conventional paper-based systems were simply inadequate to keep pace with the evolution of outbreaks [3,19-21]. Furthermore, most digital systems are designed for 1 or a limited set of response tasks, such as contact tracing, case management, data collection and transmission, data collation and analysis, communication, and coordination [22]. These limitations require public health systems to use multiple systems to accomplish the complete set of public health emergency response tasks—a detraction from the expected workload alleviation and efficiency of task performance. These challenges notwithstanding, the introduction of digital innovations for surveillance in LMICs has typically been welcomed by public health actors to bridge the service delivery gaps for which such innovations hold great promise. In particular, the occurrence of recent major outbreaks, namely, the West African Ebola Virus Disease outbreak and the COVID-19 pandemic, spurred an astounding transformation of electronic public health surveillance and outbreak response tools in Africa—one of which is the Surveillance Outbreak Response Management and Analysis System (SORMAS) [20,23-27].
The 2014 to 2016 West African Ebola Virus Disease outbreak motivated the invention of SORMAS. Beyond coordination, SORMAS integrates the business process management of disease control measures with the surveillance and detection of epidemics within 1 tool. However, compatibility with the existing Integrated Disease Surveillance and Response (IDSR) system of the World Health Organization (WHO)–Africa region was a prerequisite for adoption to maintain established work processes and offer a possibility of regional integration [28]. SORMAS is an open-source, mobile eHealth system. It is accessed via a web version by administrators and supervisors and via a mobile app version on tablets by field users. There exists a bidirectional communication between the web and mobile app versions for data synchronization. Functionally, it is a laboratory-integrated, case-based management system for routine surveillance and overall coordination of public health emergencies. It has disease process modules for notifiable diseases and nonspecific modules that are easily adaptable for emerging infectious diseases. The workflow is organized into interactive directories for the management of tasks, cases, contacts, events, laboratory samples, immunizations, points of entry, campaigns, statistics, reports, and users. SORMAS supports both indicator-based and event-based surveillance, real-time epidemiological analysis and data visualization, field coordination and response process management, bidirectional data transfer, and mobile offline functionality. It attained the global good maturity 4 years after its first deployment in Nigeria [23,26,29], where it was subsequently deployed for mpox (monkey pox) and other concurrent disease outbreaks [30,31].
In November 2019, the Ghana Health Service (GHS) started the pilot implementation of SORMAS for routine surveillance [31]. Then, 4 months into the pilot implementation phase, the COVID-19 pandemic spread to Ghana. As a consequence, the planned pilot implementation and progress evaluations that were aimed to improve strategies for a stepwise nationwide scale-up were interrupted. Nonetheless, given its demonstrated utility in the early phase of the pilot, Ghana adopted SORMAS for the COVID-19 pandemic response and, further, as the national eSurveillance and outbreak response management system for all its notifiable diseases. Thus, apart from the anticipated challenges of the implementation of digital innovations in normal times, the opportunity for a rapid nationwide scale-up of SORMAS during pandemic came with additional challenges.
We conducted a qualitative study among resource persons involved in the use and management of SORMAS in Ghana. These individuals represented GHS at the district, regional, and national levels; teaching hospitals; research laboratories; public health experts; and software developers for SORMAS (Table 1). We conducted the study in 16 districts across 2 administrative regions between November 2021 and March 2022. We purposively selected the Greater Accra and Ashanti regions for being the 2 most populous regions and COVID-19 epicenters of Ghana, where the use of SORMAS was most intensive. The Greater Accra Region was 1 of 2 pilot regions for SORMAS before the COVID-19 pandemic (2019), whereas the Ashanti Region started using SORMAS when the pandemic spread to Ghana.
We adapted the model of diffusion of innovations in health service organizations by Greenhalgh et al [32] to assess the implementation process. The model describes the dynamic interactions of factors, namely, the nature and value of the innovation, relevant health system antecedents and readiness for the innovation, method of adoption and assimilation, process diffusion and dissemination, implementation and consequences thereof, and relevant external influences on implementation and sustainability. Thus, the model provides a comprehensive and systematic approach to evaluating all aspects of the implementation of SORMAS from the innovation, the health system adopting it, its users, and the influence of local and international partners. We used these constructs to identify stakeholder perspectives about the barriers, facilitators, and sustainability strategies for system integration and institutionalization of SORMAS in Ghana.
The Helmholtz-Zentrum für infektionsforschung (HZI; the Helmholtz Centre for Infection Research) Braunschweig–Germany, in collaboration with the African Field Epidemiology Network and the Nigeria Centre for Disease Control, developed and piloted the first version of SORMAS in 2015. With funding from the Deutsche gesellschaft für internationale Zusammenarbeit (GIZ; German Corporation for International Cooperation), HZI adapted the open-source version of SORMAS in 2016. On the basis of specifications of multinational team of epidemiologists and IDSR experts, the software developers integrated all change requests and lessons learned from the pilot to release an upgraded and first open version [31].
In 2017, a multistakeholder workshop was conducted in Ghana to consider the suitability of SORMAS as a national eSurveillance and outbreak response tool. Following a decision in 2018 to adopt SORMAS, a tripartite public-private partnership business model was adopted. In this model, GHS was the implementing public institution, Ghana Community Network System provided in-country IT service, and HZI was responsible for SORMAS development and maintenance. These 3 partners signed a memorandum of understanding, and the pilot implementation started in November 2019 in the Greater Accra and Upper West regions of Ghana. On January 29, 2020—a day before WHO declared COVID-19 as a public health emergency of international concern, HZI rolled out a COVID-19–specific module [33,34]. Ghana adopted this module and subsequently adapted it based on national response needs.
We recruited 22 study participants, distributed across the abovementioned 7 broad professional categories. These included officers at the national, regional, and district surveillance units and COVID-19–testing laboratories (Figure 1). All the users of SORMAS in the 2 study regions were invited to participate. We randomly selected equal numbers of frontline users based on region and distributed them equally between the urban and rural districts within each region. We purposively selected national officers, surveillance supervisors, and software developers based on their leading roles in implementation.
We designed a reference in-depth interview guide based on the model by Greenhalgh et al [32] such that some questions were the same for all study participants, and others were specific to their professional groups to match their roles and responsibilities in the implementation. We piloted the interview guides with a pretest for surveillance supervisors and frontline users in the Upper West Region, which had a mix of users from the pilot phase and users recruited for the pandemic response. Following this pilot, we revised the interview guides to improve their completeness in scope and the clarity of questions. We did not pilot the interview guides for the remaining 3 professional groups as their availability was limited by their roles and numbers. However, the research team discussed and reached consensus on the usability and soundness of their interview guides.
We coded the data deductively with a priori parent codes and subcodes consistent with the constructs of the model by Greenhalgh et al [32]. Our approach for coding was flexible to include emerging subcodes that related to some parent codes but were not explicitly outlined in the model. We assigned descriptive memos to all subcodes consistent with the constructs of the model, to ensure consistency of coding. BBK and FL performed the coding separately. CJK-T reviewed the coding performed by BBK and FL. BBK, FL, CW, and CJK-T compared and discussed the differences, built consensus where possible, and merged codes where the data did not make a practical difference to keep them separate. For example, we merged adoption and assimilation, diffusion and dissemination of the innovation, and system antecedents and system readiness for adopting the innovation to generate 1 composite parent code each. We also included access to the system, motivation, and trust as subcodes of diffusion and dissemination. We calculated the crude percentage agreements and Cohen κ coefficients at 95% confidence level to estimate the intercoder reliability for 7 final parent codes (Multimedia Appendix 1). We used the guidelines by Landis and Kock [35] to interpret intercoder agreement as follows: poor (κ≤0), slight (κ=0.01-0.20), fair (κ=0.21-0.40), moderate (κ=0.41-0.60), substantial (κ=0.61-0.80), and almost perfect (κ=0.81-1). We exported the coded segments as Excel sheets. We paraphrased the coded segments to obtain descriptive summaries. Next, we performed interpretative analysis on the descriptive summaries to identify key messages related to barriers and facilitators and how they influenced the implementation of SORMAS. We identified user recommendations for sustainability of implementation and categorized them according to their targeted stakeholders. GK reviewed and challenged the exhaustiveness of some of the interpretations we assigned to segments of the transcripts. We discussed further and refined the clarity and scope of some interpretations.
Our study participants represented 7 categories of resource persons involved in the implementation of SORMAS in Ghana—73% (16/22) of whom were men. The median duration of the interviews was 59 (range 37-175) minutes. The intercoder reliability was almost perfect for adoption and assimilation (crude agreement=92.9%; κ=0.86, 95% CI 0.76-0.96), and substantial for the other 6 codes (crude agreement range 80%-87.5%; κ range 0.62-0.75; Multimedia Appendix 1).
We considered the barriers to and facilitators of implementation under 7 broad themes, each organized into subthemes consistent with the adapted model of diffusion of innovations in health service organizations (Multimedia Appendix 2). Overall, the identified barriers to and facilitators of implementation were similar across the 2 regions. The main facilitators resulting from the COVID-19 pandemic included the political support at the level of the presidency for the national scale-up, creation of Ghana-specific SORMAS development branch for national pandemic response needs, introduction of sample barcodes and scanners to speed up the workflow, and adoption of SORMAS by non-GHS facilities including private and state-owned research laboratories. In contrast, the main barriers to implementation occasioned by the pandemic included reduced quality of user trainings, shortage of tablets and other logistics arising from increased demand, and truncation of planned evaluation of the pilot implementation (Multimedia Appendix 2).
Perhaps, the most important facilitator of the diffusion and dissemination of the system was the frontline users’ intrinsic motivation to use it. Consistent with the concept of balance of the “pains” and “gains” of adopting digital health innovations as espoused by Shaw et al [36], although some users were dissatisfied with delays in receiving internet data bundles, they took up the cost themselves—convinced that the benefits (gains) of using the system are worth their sacrifices (pains). Such levels of user altruism informed by their passion to use digital health innovations have also been reported from experiences of implementing mobile health tools in Ghana; Malawi [37]; Kenya [38]; Myanmar [39]; Brazil [40]; Los Angeles County, California [41]; Texas [42]; and the Netherlands [43]. However, this form of altruism is unsustainable, as aptly expressed by one of the frontline health workers:
The committed leadership of the health service managers and political endorsement at the level of the presidency translated to a boost in the commitment and confidence of the workforce. They had the assurance that their use of the system was appreciated by the highest office of the country. Users also considered themselves as being on a national assignment in the transition from paper-based surveillance and outbreak response to an electronic system. The positive user attitude derived from the high level of stewardship is consistent with user behaviors in the implementation of innovations in public health systems, as reported by Ginsburg et al [44] in their feasibility study of a digital innovation (mPneumonia) for pneumonia treatment in Ghana and an acceptability and usability study of a mobile health intervention for eye care in Kenya [45]. In addition, the all-inclusive management style of the GHS leadership on the implementation process, effective internal communication among supervisors and frontline users, trust of users in their leadership and the system, and field support by enthusiastic peer champions of SORMAS all contributed to sustaining the optimism for success in the face of operational challenges. Rightly so, this style of leadership and the support of champions have been recognized by digital health system implementers as important success conditions in Swaziland [46], Kenya [38], and Norway [47].
One of the key benefits of adopting SORMAS is to improve the overall efficiency of task performance and reduce the workload. DHIS2 serves as the repository for all health data in Ghana, into which surveillance and outbreak data collected with SORMAS should be fed. However, the delays in achieving stable interoperability between these 2 systems undermine the expected efficiency of data transfer, as users are required to manually enter summary data already captured in SORMAS onto DHIS2. This challenge of system interoperability in the adoption of digital innovations in health care settings is rather common in both high-income countries and LMICs [9,42,43,48]. In the current case of Ghana, it poses a threat to a speedy completion of health system integration and institutionalization of SORMAS. A closely related barrier to long-term sustainability of the system is the yet inadequate technical expertise and infrastructural capacity of GHS to host the data collected with SORMAS. The availability of a private local partner to fill this gap is helpful, but its downside is that it deprives GHS of full control of its data. The external service also comes at an extra recurrent cost and competes for already stretched resources.
The challenge of synchronization undermines the real-time reporting and communication—the very properties of the system that make it a good management tool. Delayed or failed synchronizations were mostly a result of poor internet connectivity. Although the challenge of internet connectivity was more common in the rural districts, some of the study participants from urban districts experienced frequent episodes of poor connection. The challenge of reliable internet has improved over the years in LMICs but still poses a threat to the implementation of digital health tools in many settings, as corroborated by health worker experiences in Ghana [37,49], Kenya [38,45], Tanzania [50], Bangladesh [51], Myanmar [39], Brazil [40], India [52,53], and Ireland [54]. We recognize that the challenge of internet connectivity rests largely with the telecommunication industry in most countries and is not within the control of the health systems, even if they have unlimited funding to procure the best service subscriptions. Furthermore, the investments in national communication infrastructure are capital intensive, are consumer driven, and require good collaborations between the private and public sectors. In the past 2 decades, the leading internet providers in Ghana (Vodafone, Mobile Telecommunication Network, Expresso, Globacom, Airtel, and Tigo) have been largely privately owned and hence profit driven. As part of strategies to improve the penetration and quality of information and communications technology services, there have been calls for infrastructure sharing to contain the cost of investments and harness synergies among these service providers [55]. However, there have been challenges with mergers epitomized by the postmerger tensions between Airtel and Tigo that forced the government of Ghana to intervene and take over this merger [56]. This government takeover does not provide guarantee for better services because there is mixed evidence regarding the question of who provides more effective telecommunication services—the state or the private sector [57-59].
Cognizant of this multilayer of prerequisites for attaining the ideal quality of internet service vis-à-vis the ever-increasing need for real-time reporting in surveillance and outbreak response, the invention of Low-Bandwidth Database Synchronization by HZI offers a pragmatic solution, especially for LMICs [60]. Its implementation would overcome this challenge because it provides 3 alternate low-bandwidth options in 1 tool for data transmission and synchronization between mobile devices and central databases.
So far, given the level of technical and political stakeholder support, ongoing consolidation of system integration, and enthusiasm of frontline users, what remains as a serious threat to institutionalization and sustainability is a reliable national funding. There is good political support and appreciation of the utility of SORMAS. However, these have not translated to commensurate funding. The immediate reason is that the political support for the rapid nationwide scale-up of the implementation came owing to the need to respond to the COVID-19 pandemic. Thus, resources are unplanned and fragmented. A more systemic reason may simply be that health systems in LMICs, similar to most other sectors, are substantially dependent on external support, for which, new initiatives such as SORMAS do not become exceptions [61,62]. Although the rapid nationwide scale-up caused an increased demand on logistics, resulting in shortages, the continued support from the external partners, private sector, and government abated the logistics crisis over the course of the pandemic. This experience constitutes a reminder for governments and their national health institutes to maintain emergency funds and stockpiles as part of their emergency preparedness strategies [63-66]. Thus, to achieve this, our findings suggest 3 complimentary and mutually reinforcing approaches, namely, the commitment of central and local governments, broad national participation involving active financial and technical support from the private sector, and resource pooling from programs within GHS and among all health agencies in the country. Regarding the approaches to national ownership of health care financing in low-income countries, Kiendrébéogo and Meessen [67] also proposed a similar model, which they describe as “a journey with more than one pathway”—requiring broad stakeholder support including central governments, parliaments, health institutions, and the private sector. The private sector has also been identified as a key potential player in complementing the efforts of governments in health care financing in Nigeria [68,69]; Zimbabwe [70]; South Africa [71,72]; and more generally, for the African setting [73,74].
However, the timing of the introduction of SORMAS in Ghana could benefit from the national digitalization agenda that seeks to digitalize the operations of all public sectors, critical among which is health. This digitalization agenda is already receiving technical and funding support from in-country United Nations agencies and the World Bank for upgrading the overall national digital infrastructure [75-77]. We anticipate that the private telecommunication sector will remain a key player in realizing this goal, while also benefiting from a close public-private partnership. Furthermore, we anticipate that a joint stewardship of the United Nations, World Bank, private telecommunication sector, and government of Ghana could increase the chances of successful execution of the digitalization agenda. Ultimately, the envisaged penetration and quality of service of this agenda should minimize the challenges of internet connection in the use of SORMAS. A further boost to the sustainability outlook of SORMAS is Ghana’s formal adoption of eSurveillance consistent with WHO recommendations is expected to attract dedicated organizational budgetary support for its operationalization. Given that SORMAS is now the national electronic tool for surveillance, there is reason to be confident that its implementation will be sustained in the medium to long term. The ultimate institutionalization would benefit from the suggestions of users about galvanizing support from local government agencies and private business establishments at the lowest reporting levels.
We note that these stakeholder identified barriers and facilitators cut across all constructs of the model by Greenhalgh et al [32]. Thus, the outcome of the implementation is contingent on a dynamic nonlinear interaction of interdependent factors. This phenomenon has been widely recognized by other models and frameworks of implementation science [78], exemplified by the number and variety of barriers encountered in the implementation of a tuberculosis contact investigating system in Uganda [17] and the substantial gains of web-based disease monitoring and management system in the Netherlands [43]. The facilitators and barriers also cut nearly uniformly between urban and rural districts, except that internet connectivity was generally better in urban districts as would be expected. The uniformity of facilitators and barriers regarding implementation between the 2 study regions could be explained by 2 factors. First, the truncated piloting of the implementation in the Greater Accra Region did not allow for systematic evaluation and addressing of early bottlenecks. Second, the need for mass recruitment and equipping of new users across the country for the pandemic response posed similar challenges to the performance of SORMAS, workforce training, and related demands, all of which depend on a common resource pool.
From the foregoing, we infer that the facilitators and barriers identified by the study participants regarding the implementation of SORMAS in Ghana are not entirely unique to this tool and country. For example, just as its compatibility with IDSR would make it readily acceptable to countries of the WHO–Africa Region, so will the challenge of interoperability with the DHIS2 pose barriers to its integration with public health systems of these countries. The challenge of poor internet connectivity will confront the implementation of any digital tool in any country to the extent of the insufficiencies of the national telecommunication services—be it in the quality of bandwidth, geographical penetration, or other context-specific factors. Moreover, as in the case of Ghana, many LMICs receive financial support from international development partners, some of which are either short term or inconsistent [79]. Hence, the threat of implementation failure for lack of reliable and sustainable local funding should concern any such country that undertakes the institutionalization of digital systems for surveillance and outbreak response. The timing of adoption, organizational capacities and work culture, workforce skill sets, public-private partnerships, and overall political and business climate have relevance for the implementation of SORMAS in Ghana, as would be expected for the implementation of similar digital tools at scale in comparable settings. The interactions among these factors and how they influence the implementation are certain to vary to various extents depending on the specific contexts of adopting countries in normal times or during public health crises.
These limitations notwithstanding, our study had many strengths. First, the timing of the study allowed us to evaluate the implementation process in both normal and pandemic times. Thus, the richness of the findings has relevance for implementing similar tools in both scenarios. Second, the model by Greenhalgh et al [32] that we adapted as part of for our study design is a comprehensive model built from evidence obtained from an extensive systematic review of implementing innovations in health care organizations such as GHS. Thus, this model enabled us to identify a broad range of factors across all relevant aspects of the implementation, namely, technology, interactions of systems (internal, national, and international), and workforce behaviors. Third, our findings are also relevant in their timing as they could feed into evidence for funding prioritization for the adoption of eSurveillance as part of a national digitalization agenda. Fourth, our findings raise some important questions for further studies. An in-depth investigation of the business and political complexities of the telecommunication industry in Ghana could provide more insights about and possible solutions to the problems of poor information and communications technology services and how the private telecommunication industry could support public institutions in tackling the challenges of system interoperability that hamper the implementation and integration of digital systems. A review of funding models for the digitalization of surveillance in LMICs would also provide insights for adopting and adapting proven funding models to promote country ownership. As our findings reveal, the implementation of SORMAS in Ghana has so far benefited from a wide range of financial and technical contributions from the state, health workers, private sector, and international partners. Hence, a cost-benefit analysis would be a useful follow-up study to examine the direct, indirect, intangible, and opportunity costs of the implementation so far and could provide further insights for planning sustainable strategies.
We provide specific and targeted recommendations for the sustainable institutionalization of SORMAS in Ghana, which could also be useful for comparable LMICs (Table 2).

Sections

"[{\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref1\", \"ref2\", \"ref3\", \"ref5\", \"ref6\", \"ref7\", \"ref8\", \"ref10\", \"ref10\", \"ref12\", \"ref10\", \"ref13\", \"ref18\"], \"section\": \"Background\", \"text\": \"In the advent of affordable and robust IT tools in the 1990s, high-income countries began to switch from paper-based to electronic disease surveillance [1]. The pace of digitalization in public health surveillance and outbreak response has since gained momentum and spread to low- and middle-income countries (LMICs) [2]. This transition is, in part, a result of digital penetration, consolidated by the demonstrated utility of digital tools in public health practice [3-5]. Well-established electronic disease surveillance systems have many benefits. Some of these benefits are improved data quality (completeness, detail, and validity), timeliness of reporting, data standardization, automation of data analysis and visualization, prompt identification of public health events, and timely information sharing for public health action [6,7]. Thus, the benefits of digitalization in public health practice are many but so are the challenges. Frequently reported challenges of adopting electronic health systems include the technical challenges of technology, absence of basic necessary infrastructure such as electricity, implementation costs, workforce capacities and attitudes, and organizational factors [8-10]. These challenges are dynamic and vary in scope and complexity across geopolitical, socioeconomic, cultural, and organizational settings [10-12]. Therefore, some electronic systems have not made it beyond pilots to sustained nationwide institutionalization, especially in LMICs [10,13-18].\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref3\", \"ref19\", \"ref21\", \"ref22\", \"ref20\", \"ref23\", \"ref27\"], \"section\": \"Background\", \"text\": \"The cross-border geographical spread of outbreaks and the multidisciplinary response requirements made it more apparent than ever that conventional paper-based systems were simply inadequate to keep pace with the evolution of outbreaks [3,19-21]. Furthermore, most digital systems are designed for 1 or a limited set of response tasks, such as contact tracing, case management, data collection and transmission, data collation and analysis, communication, and coordination [22]. These limitations require public health systems to use multiple systems to accomplish the complete set of public health emergency response tasks\\u2014a detraction from the expected workload alleviation and efficiency of task performance. These challenges notwithstanding, the introduction of digital innovations for surveillance in LMICs has typically been welcomed by public health actors to bridge the service delivery gaps for which such innovations hold great promise. In particular, the occurrence of recent major outbreaks, namely, the West African Ebola Virus Disease outbreak and the COVID-19 pandemic, spurred an astounding transformation of electronic public health surveillance and outbreak response tools in Africa\\u2014one of which is the Surveillance Outbreak Response Management and Analysis System (SORMAS) [20,23-27].\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref28\", \"ref23\", \"ref26\", \"ref29\", \"ref30\", \"ref31\"], \"section\": \"Background\", \"text\": \"The 2014 to 2016 West African Ebola Virus Disease outbreak motivated the invention of SORMAS. Beyond coordination, SORMAS integrates the business process management of disease control measures with the surveillance and detection of epidemics within 1 tool. However, compatibility with the existing Integrated Disease Surveillance and Response (IDSR) system of the World Health Organization (WHO)\\u2013Africa region was a prerequisite for adoption to maintain established work processes and offer a possibility of regional integration [28]. SORMAS is an open-source, mobile eHealth system. It is accessed via a web version by administrators and supervisors and via a mobile app version on tablets by field users. There exists a bidirectional communication between the web and mobile app versions for data synchronization. Functionally, it is a laboratory-integrated, case-based management system for routine surveillance and overall coordination of public health emergencies. It has disease process modules for notifiable diseases and nonspecific modules that are easily adaptable for emerging infectious diseases. The workflow is organized into interactive directories for the management of tasks, cases, contacts, events, laboratory samples, immunizations, points of entry, campaigns, statistics, reports, and users. SORMAS supports both indicator-based and event-based surveillance, real-time epidemiological analysis and data visualization, field coordination and response process management, bidirectional data transfer, and mobile offline functionality. It attained the global good maturity 4 years after its first deployment in Nigeria [23,26,29], where it was subsequently deployed for mpox (monkey pox) and other concurrent disease outbreaks [30,31].\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref31\"], \"section\": \"Background\", \"text\": \"In November 2019, the Ghana Health Service (GHS) started the pilot implementation of SORMAS for routine surveillance [31]. Then, 4 months into the pilot implementation phase, the COVID-19 pandemic spread to Ghana. As a consequence, the planned pilot implementation and progress evaluations that were aimed to improve strategies for a stepwise nationwide scale-up were interrupted. Nonetheless, given its demonstrated utility in the early phase of the pilot, Ghana adopted SORMAS for the COVID-19 pandemic response and, further, as the national eSurveillance and outbreak response management system for all its notifiable diseases. Thus, apart from the anticipated challenges of the implementation of digital innovations in normal times, the opportunity for a rapid nationwide scale-up of SORMAS during pandemic came with additional challenges.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"table1\"], \"section\": \"Study Design\", \"text\": \"We conducted a qualitative study among resource persons involved in the use and management of SORMAS in Ghana. These individuals represented GHS at the district, regional, and national levels; teaching hospitals; research laboratories; public health experts; and software developers for SORMAS (Table 1). We conducted the study in 16 districts across 2 administrative regions between November 2021 and March 2022. We purposively selected the Greater Accra and Ashanti regions for being the 2 most populous regions and COVID-19 epicenters of Ghana, where the use of SORMAS was most intensive. The Greater Accra Region was 1 of 2 pilot regions for SORMAS before the COVID-19 pandemic (2019), whereas the Ashanti Region started using SORMAS when the pandemic spread to Ghana.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref32\"], \"section\": \"Study Design\", \"text\": \"We adapted the model of diffusion of innovations in health service organizations by Greenhalgh et al [32] to assess the implementation process. The model describes the dynamic interactions of factors, namely, the nature and value of the innovation, relevant health system antecedents and readiness for the innovation, method of adoption and assimilation, process diffusion and dissemination, implementation and consequences thereof, and relevant external influences on implementation and sustainability. Thus, the model provides a comprehensive and systematic approach to evaluating all aspects of the implementation of SORMAS from the innovation, the health system adopting it, its users, and the influence of local and international partners. We used these constructs to identify stakeholder perspectives about the barriers, facilitators, and sustainability strategies for system integration and institutionalization of SORMAS in Ghana.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref31\"], \"section\": \"Context of SORMAS Implementation in Ghana\", \"text\": \"The Helmholtz-Zentrum f\\u00fcr infektionsforschung (HZI; the Helmholtz Centre for Infection Research) Braunschweig\\u2013Germany, in collaboration with the African Field Epidemiology Network and the Nigeria Centre for Disease Control, developed and piloted the first version of SORMAS in 2015. With funding from the Deutsche gesellschaft f\\u00fcr internationale Zusammenarbeit (GIZ; German Corporation for International Cooperation), HZI adapted the open-source version of SORMAS in 2016. On the basis of specifications of multinational team of epidemiologists and IDSR experts, the software developers integrated all change requests and lessons learned from the pilot to release an upgraded and first open version [31].\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref33\", \"ref34\"], \"section\": \"Context of SORMAS Implementation in Ghana\", \"text\": \"In 2017, a multistakeholder workshop was conducted in Ghana to consider the suitability of SORMAS as a national eSurveillance and outbreak response tool. Following a decision in 2018 to adopt SORMAS, a tripartite public-private partnership business model was adopted. In this model, GHS was the implementing public institution, Ghana Community Network System provided in-country IT service, and HZI was responsible for SORMAS development and maintenance. These 3 partners signed a memorandum of understanding, and the pilot implementation started in November 2019 in the Greater Accra and Upper West regions of Ghana. On January 29, 2020\\u2014a day before WHO declared COVID-19 as a public health emergency of international concern, HZI rolled out a COVID-19\\u2013specific module [33,34]. Ghana adopted this module and subsequently adapted it based on national response needs.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"figure1\"], \"section\": \"Study Participants\", \"text\": \"We recruited 22 study participants, distributed across the abovementioned 7 broad professional categories. These included officers at the national, regional, and district surveillance units and COVID-19\\u2013testing laboratories (Figure 1). All the users of SORMAS in the 2 study regions were invited to participate. We randomly selected equal numbers of frontline users based on region and distributed them equally between the urban and rural districts within each region. We purposively selected national officers, surveillance supervisors, and software developers based on their leading roles in implementation.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref32\"], \"section\": \"Data Collection and Management\", \"text\": \"We designed a reference in-depth interview guide based on the model by Greenhalgh et al [32] such that some questions were the same for all study participants, and others were specific to their professional groups to match their roles and responsibilities in the implementation. We piloted the interview guides with a pretest for surveillance supervisors and frontline users in the Upper West Region, which had a mix of users from the pilot phase and users recruited for the pandemic response. Following this pilot, we revised the interview guides to improve their completeness in scope and the clarity of questions. We did not pilot the interview guides for the remaining 3 professional groups as their availability was limited by their roles and numbers. However, the research team discussed and reached consensus on the usability and soundness of their interview guides.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref32\", \"app1\", \"ref35\"], \"section\": \"Data Analysis\", \"text\": \"We coded the data deductively with a priori parent codes and subcodes consistent with the constructs of the model by Greenhalgh et al [32]. Our approach for coding was flexible to include emerging subcodes that related to some parent codes but were not explicitly outlined in the model. We assigned descriptive memos to all subcodes consistent with the constructs of the model, to ensure consistency of coding. BBK and FL performed the coding separately. CJK-T reviewed the coding performed by BBK and FL. BBK, FL, CW, and CJK-T compared and discussed the differences, built consensus where possible, and merged codes where the data did not make a practical difference to keep them separate. For example, we merged adoption and assimilation, diffusion and dissemination of the innovation, and system antecedents and system readiness for adopting the innovation to generate 1 composite parent code each. We also included access to the system, motivation, and trust as subcodes of diffusion and dissemination. We calculated the crude percentage agreements and Cohen \\u03ba coefficients at 95% confidence level to estimate the intercoder reliability for 7 final parent codes (Multimedia Appendix 1). We used the guidelines by Landis and Kock [35] to interpret intercoder agreement as follows: poor (\\u03ba\\u22640), slight (\\u03ba=0.01-0.20), fair (\\u03ba=0.21-0.40), moderate (\\u03ba=0.41-0.60), substantial (\\u03ba=0.61-0.80), and almost perfect (\\u03ba=0.81-1). We exported the coded segments as Excel sheets. We paraphrased the coded segments to obtain descriptive summaries. Next, we performed interpretative analysis on the descriptive summaries to identify key messages related to barriers and facilitators and how they influenced the implementation of SORMAS. We identified user recommendations for sustainability of implementation and categorized them according to their targeted stakeholders. GK reviewed and challenged the exhaustiveness of some of the interpretations we assigned to segments of the transcripts. We discussed further and refined the clarity and scope of some interpretations.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"app1\"], \"section\": \"General Results\", \"text\": \"Our study participants represented 7 categories of resource persons involved in the implementation of SORMAS in Ghana\\u201473% (16/22) of whom were men. The median duration of the interviews was 59 (range 37-175) minutes. The intercoder reliability was almost perfect for adoption and assimilation (crude agreement=92.9%; \\u03ba=0.86, 95% CI 0.76-0.96), and substantial for the other 6 codes (crude agreement range 80%-87.5%; \\u03ba range 0.62-0.75; Multimedia Appendix 1).\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"app2\", \"app2\"], \"section\": \"Overview\", \"text\": \"We considered the barriers to and facilitators of implementation under 7 broad themes, each organized into subthemes consistent with the adapted model of diffusion of innovations in health service organizations (Multimedia Appendix 2). Overall, the identified barriers to and facilitators of implementation were similar across the 2 regions. The main facilitators resulting from the COVID-19 pandemic included the political support at the level of the presidency for the national scale-up, creation of Ghana-specific SORMAS development branch for national pandemic response needs, introduction of sample barcodes and scanners to speed up the workflow, and adoption of SORMAS by non-GHS facilities including private and state-owned research laboratories. In contrast, the main barriers to implementation occasioned by the pandemic included reduced quality of user trainings, shortage of tablets and other logistics arising from increased demand, and truncation of planned evaluation of the pilot implementation (Multimedia Appendix 2).\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref36\", \"ref37\", \"ref38\", \"ref39\", \"ref40\", \"ref41\", \"ref42\", \"ref43\"], \"section\": \"Principal Findings\", \"text\": \"Perhaps, the most important facilitator of the diffusion and dissemination of the system was the frontline users\\u2019 intrinsic motivation to use it. Consistent with the concept of balance of the \\u201cpains\\u201d and \\u201cgains\\u201d of adopting digital health innovations as espoused by Shaw et al [36], although some users were dissatisfied with delays in receiving internet data bundles, they took up the cost themselves\\u2014convinced that the benefits (gains) of using the system are worth their sacrifices (pains). Such levels of user altruism informed by their passion to use digital health innovations have also been reported from experiences of implementing mobile health tools in Ghana; Malawi [37]; Kenya [38]; Myanmar [39]; Brazil [40]; Los Angeles County, California [41]; Texas [42]; and the Netherlands [43]. However, this form of altruism is unsustainable, as aptly expressed by one of the frontline health workers:\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref44\", \"ref45\", \"ref46\", \"ref38\", \"ref47\"], \"section\": \"Principal Findings\", \"text\": \"The committed leadership of the health service managers and political endorsement at the level of the presidency translated to a boost in the commitment and confidence of the workforce. They had the assurance that their use of the system was appreciated by the highest office of the country. Users also considered themselves as being on a national assignment in the transition from paper-based surveillance and outbreak response to an electronic system. The positive user attitude derived from the high level of stewardship is consistent with user behaviors in the implementation of innovations in public health systems, as reported by Ginsburg et al [44] in their feasibility study of a digital innovation (mPneumonia) for pneumonia treatment in Ghana and an acceptability and usability study of a mobile health intervention for eye care in Kenya [45]. In addition, the all-inclusive management style of the GHS leadership on the implementation process, effective internal communication among supervisors and frontline users, trust of users in their leadership and the system, and field support by enthusiastic peer champions of SORMAS all contributed to sustaining the optimism for success in the face of operational challenges. Rightly so, this style of leadership and the support of champions have been recognized by digital health system implementers as important success conditions in Swaziland [46], Kenya [38], and Norway [47].\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref9\", \"ref42\", \"ref43\", \"ref48\"], \"section\": \"Principal Findings\", \"text\": \"One of the key benefits of adopting SORMAS is to improve the overall efficiency of task performance and reduce the workload. DHIS2 serves as the repository for all health data in Ghana, into which surveillance and outbreak data collected with SORMAS should be fed. However, the delays in achieving stable interoperability between these 2 systems undermine the expected efficiency of data transfer, as users are required to manually enter summary data already captured in SORMAS onto DHIS2. This challenge of system interoperability in the adoption of digital innovations in health care settings is rather common in both high-income countries and LMICs [9,42,43,48]. In the current case of Ghana, it poses a threat to a speedy completion of health system integration and institutionalization of SORMAS. A closely related barrier to long-term sustainability of the system is the yet inadequate technical expertise and infrastructural capacity of GHS to host the data collected with SORMAS. The availability of a private local partner to fill this gap is helpful, but its downside is that it deprives GHS of full control of its data. The external service also comes at an extra recurrent cost and competes for already stretched resources.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref37\", \"ref49\", \"ref38\", \"ref45\", \"ref50\", \"ref51\", \"ref39\", \"ref40\", \"ref52\", \"ref53\", \"ref54\", \"ref55\", \"ref56\", \"ref57\", \"ref59\"], \"section\": \"Principal Findings\", \"text\": \"The challenge of synchronization undermines the real-time reporting and communication\\u2014the very properties of the system that make it a good management tool. Delayed or failed synchronizations were mostly a result of poor internet connectivity. Although the challenge of internet connectivity was more common in the rural districts, some of the study participants from urban districts experienced frequent episodes of poor connection. The challenge of reliable internet has improved over the years in LMICs but still poses a threat to the implementation of digital health tools in many settings, as corroborated by health worker experiences in Ghana [37,49], Kenya [38,45], Tanzania [50], Bangladesh [51], Myanmar [39], Brazil [40], India [52,53], and Ireland [54]. We recognize that the challenge of internet connectivity rests largely with the telecommunication industry in most countries and is not within the control of the health systems, even if they have unlimited funding to procure the best service subscriptions. Furthermore, the investments in national communication infrastructure are capital intensive, are consumer driven, and require good collaborations between the private and public sectors. In the past 2 decades, the leading internet providers in Ghana (Vodafone, Mobile Telecommunication Network, Expresso, Globacom, Airtel, and Tigo) have been largely privately owned and hence profit driven. As part of strategies to improve the penetration and quality of information and communications technology services, there have been calls for infrastructure sharing to contain the cost of investments and harness synergies among these service providers [55]. However, there have been challenges with mergers epitomized by the postmerger tensions between Airtel and Tigo that forced the government of Ghana to intervene and take over this merger [56]. This government takeover does not provide guarantee for better services because there is mixed evidence regarding the question of who provides more effective telecommunication services\\u2014the state or the private sector [57-59].\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref60\"], \"section\": \"Principal Findings\", \"text\": \"Cognizant of this multilayer of prerequisites for attaining the ideal quality of internet service vis-\\u00e0-vis the ever-increasing need for real-time reporting in surveillance and outbreak response, the invention of Low-Bandwidth Database Synchronization by HZI offers a pragmatic solution, especially for LMICs [60]. Its implementation would overcome this challenge because it provides 3 alternate low-bandwidth options in 1 tool for data transmission and synchronization between mobile devices and central databases.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref61\", \"ref62\", \"ref63\", \"ref66\", \"ref67\", \"ref68\", \"ref69\", \"ref70\", \"ref71\", \"ref72\", \"ref73\", \"ref74\"], \"section\": \"Principal Findings\", \"text\": \"So far, given the level of technical and political stakeholder support, ongoing consolidation of system integration, and enthusiasm of frontline users, what remains as a serious threat to institutionalization and sustainability is a reliable national funding. There is good political support and appreciation of the utility of SORMAS. However, these have not translated to commensurate funding. The immediate reason is that the political support for the rapid nationwide scale-up of the implementation came owing to the need to respond to the COVID-19 pandemic. Thus, resources are unplanned and fragmented. A more systemic reason may simply be that health systems in LMICs, similar to most other sectors, are substantially dependent on external support, for which, new initiatives such as SORMAS do not become exceptions [61,62]. Although the rapid nationwide scale-up caused an increased demand on logistics, resulting in shortages, the continued support from the external partners, private sector, and government abated the logistics crisis over the course of the pandemic. This experience constitutes a reminder for governments and their national health institutes to maintain emergency funds and stockpiles as part of their emergency preparedness strategies [63-66]. Thus, to achieve this, our findings suggest 3 complimentary and mutually reinforcing approaches, namely, the commitment of central and local governments, broad national participation involving active financial and technical support from the private sector, and resource pooling from programs within GHS and among all health agencies in the country. Regarding the approaches to national ownership of health care financing in low-income countries, Kiendr\\u00e9b\\u00e9ogo and Meessen [67] also proposed a similar model, which they describe as \\u201ca journey with more than one pathway\\u201d\\u2014requiring broad stakeholder support including central governments, parliaments, health institutions, and the private sector. The private sector has also been identified as a key potential player in complementing the efforts of governments in health care financing in Nigeria [68,69]; Zimbabwe [70]; South Africa [71,72]; and more generally, for the African setting [73,74].\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref75\", \"ref77\"], \"section\": \"Principal Findings\", \"text\": \"However, the timing of the introduction of SORMAS in Ghana could benefit from the national digitalization agenda that seeks to digitalize the operations of all public sectors, critical among which is health. This digitalization agenda is already receiving technical and funding support from in-country United Nations agencies and the World Bank for upgrading the overall national digital infrastructure [75-77]. We anticipate that the private telecommunication sector will remain a key player in realizing this goal, while also benefiting from a close public-private partnership. Furthermore, we anticipate that a joint stewardship of the United Nations, World Bank, private telecommunication sector, and government of Ghana could increase the chances of successful execution of the digitalization agenda. Ultimately, the envisaged penetration and quality of service of this agenda should minimize the challenges of internet connection in the use of SORMAS. A further boost to the sustainability outlook of SORMAS is Ghana\\u2019s formal adoption of eSurveillance consistent with WHO recommendations is expected to attract dedicated organizational budgetary support for its operationalization. Given that SORMAS is now the national electronic tool for surveillance, there is reason to be confident that its implementation will be sustained in the medium to long term. The ultimate institutionalization would benefit from the suggestions of users about galvanizing support from local government agencies and private business establishments at the lowest reporting levels.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref32\", \"ref78\", \"ref17\", \"ref43\"], \"section\": \"Principal Findings\", \"text\": \"We note that these stakeholder identified barriers and facilitators cut across all constructs of the model by Greenhalgh et al [32]. Thus, the outcome of the implementation is contingent on a dynamic nonlinear interaction of interdependent factors. This phenomenon has been widely recognized by other models and frameworks of implementation science [78], exemplified by the number and variety of barriers encountered in the implementation of a tuberculosis contact investigating system in Uganda [17] and the substantial gains of web-based disease monitoring and management system in the Netherlands [43]. The facilitators and barriers also cut nearly uniformly between urban and rural districts, except that internet connectivity was generally better in urban districts as would be expected. The uniformity of facilitators and barriers regarding implementation between the 2 study regions could be explained by 2 factors. First, the truncated piloting of the implementation in the Greater Accra Region did not allow for systematic evaluation and addressing of early bottlenecks. Second, the need for mass recruitment and equipping of new users across the country for the pandemic response posed similar challenges to the performance of SORMAS, workforce training, and related demands, all of which depend on a common resource pool.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref79\"], \"section\": \"Principal Findings\", \"text\": \"From the foregoing, we infer that the facilitators and barriers identified by the study participants regarding the implementation of SORMAS in Ghana are not entirely unique to this tool and country. For example, just as its compatibility with IDSR would make it readily acceptable to countries of the WHO\\u2013Africa Region, so will the challenge of interoperability with the DHIS2 pose barriers to its integration with public health systems of these countries. The challenge of poor internet connectivity will confront the implementation of any digital tool in any country to the extent of the insufficiencies of the national telecommunication services\\u2014be it in the quality of bandwidth, geographical penetration, or other context-specific factors. Moreover, as in the case of Ghana, many LMICs receive financial support from international development partners, some of which are either short term or inconsistent [79]. Hence, the threat of implementation failure for lack of reliable and sustainable local funding should concern any such country that undertakes the institutionalization of digital systems for surveillance and outbreak response. The timing of adoption, organizational capacities and work culture, workforce skill sets, public-private partnerships, and overall political and business climate have relevance for the implementation of SORMAS in Ghana, as would be expected for the implementation of similar digital tools at scale in comparable settings. The interactions among these factors and how they influence the implementation are certain to vary to various extents depending on the specific contexts of adopting countries in normal times or during public health crises.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"ref32\"], \"section\": \"Limitations and Strengths\", \"text\": \"These limitations notwithstanding, our study had many strengths. First, the timing of the study allowed us to evaluate the implementation process in both normal and pandemic times. Thus, the richness of the findings has relevance for implementing similar tools in both scenarios. Second, the model by Greenhalgh et al [32] that we adapted as part of for our study design is a comprehensive model built from evidence obtained from an extensive systematic review of implementing innovations in health care organizations such as GHS. Thus, this model enabled us to identify a broad range of factors across all relevant aspects of the implementation, namely, technology, interactions of systems (internal, national, and international), and workforce behaviors. Third, our findings are also relevant in their timing as they could feed into evidence for funding prioritization for the adoption of eSurveillance as part of a national digitalization agenda. Fourth, our findings raise some important questions for further studies. An in-depth investigation of the business and political complexities of the telecommunication industry in Ghana could provide more insights about and possible solutions to the problems of poor information and communications technology services and how the private telecommunication industry could support public institutions in tackling the challenges of system interoperability that hamper the implementation and integration of digital systems. A review of funding models for the digitalization of surveillance in LMICs would also provide insights for adopting and adapting proven funding models to promote country ownership. As our findings reveal, the implementation of SORMAS in Ghana has so far benefited from a wide range of financial and technical contributions from the state, health workers, private sector, and international partners. Hence, a cost-benefit analysis would be a useful follow-up study to examine the direct, indirect, intangible, and opportunity costs of the implementation so far and could provide further insights for planning sustainable strategies.\"}, {\"pmc\": \"PMC10625076\", \"pmid\": \"37862105\", \"reference_ids\": [\"table2\"], \"section\": \"Limitations and Strengths\", \"text\": \"We provide specific and targeted recommendations for the sustainable institutionalization of SORMAS in Ghana, which could also be useful for comparable LMICs (Table 2).\"}]"

Metadata

"{}"