Feasibility Testing of an Application to Improve the Diagnosis and Management of Febrile Illnesses by Frontline Health Workers in Low- and Middle-Income Countries

doi:10.21203/rs.3.rs-4513509/v1

Download PDF

Research Article

Feasibility Testing of an Application to Improve the Diagnosis and Management of Febrile Illnesses by Frontline Health Workers in Low- and Middle-Income Countries

https://doi.org/10.21203/rs.3.rs-4513509/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

The Febra Diagnostica App (Febra App) is an intelligent medical decision support system for diagnosing and managing febrile illnesses among frontline health workers (FHWs). It addresses the global health workforce shortage, especially in low- and middle-income countries, where diagnosing fevers can be challenging due to limited skilled personnel and resources. Task-shifting to frontline health workers is increasingly common, making accurate diagnostic tools crucial. This pilot study assesses implementation outcomes including feasibility, acceptability, and appropriateness of the Febra App among FHWs in Nigeria.

Methods

A 3-month single-arm pilot study was conducted in government-owned primary health centres (PHC) across four Nigerian states (Akwa Ibom, Cross River, Imo, and Rivers states). Implementation strategies included App development, FHW training, provision of tablets and power banks, utilization of the App by FHWs, and continuous technical assistance. Mixed methods, including a quantitative survey and qualitative interviews, were employed to evaluate the intervention. Data analysis utilized descriptive statistics for quantitative data and thematic coding and analysis for qualitative data.

Results

FHWs demonstrated high acceptability to using the Febra App for diagnosis of febrile illnesses, with the majority expressing approval of its ease of use, appropriateness, and relevance to their clinical practice. FHWs also found it to be feasible for routine use in PHCs, however, technical challenges, such as application crashes and poor network connectivity, posed challenges to the routine integration of the App into workflow processes.

Conclusion

The pilot study highlights the potential of the Febra App to enhance diagnostic accuracy and simplify the management of febrile illnesses in resource-limited settings. Recommendations for improvement included streamlining the App's usability and addressing technical issues. Addressing these implementation challenges will be critical for ensuring the sustainable adoption and scalability of the intervention.

Febrile illnesses

Febra Diagnostica

Decision Support Systems

Feasibility

Acceptability

Appropriateness

Implementation Research

Nigeria

1. What uncertainties existed regarding the feasibility of the FebraApp?

This project aimed to assess the acceptability and appropriateness of the Febra App for frontline health workers to diagnose febrile illnesses in limited settings. It also aimed to assess the feasibility of everyday use of the App by frontline health workers.

2.What are the key feasibility findings?

Frontline health workers expressed strong acceptance of the Febra App and found it suitable for diagnosing febrile illnesses in their setting. They reported it to be user-friendly and directly applicable to their clinical practice. They also considered it feasible for routine use in primary healthcare centers. However, technical issues such as application crashes and network connectivity problems hampered its smooth integration into workflow processes. Other non-technical problems included the challenge of staff timing needed to use the Febra App.

3. What are the implications of the feasibility findings for the design of the main study?

The study findings indicate that the Febra App is feasible in low-resource settings, supporting its scale-up. The larger study will explore the effectiveness of the Febra App in assisting frontline health workers (FHWs) in diagnosing and treating febrile illnesses. Implementation strategies from the pilot study were found to be effective and will be retained for the main study. However, significant improvements will be made to the App before the main study, addressing both technical and non-technical issues to enhance its reliability and easy integration into workflow processes.

The global health workforce shortage, estimated at 15 million, threatens equitable health access, particularly in low-middle-income countries (LMICs) plagued by capital flight and inadequate resources in the health sector [1, 2]. This shortage has increased task-shifting of the burden of diagnosis and management of endemic disease conditions to frontline health workers (FHWs). FHWs in this context were middle cadre workers trained to deliver health services formally such as Public Health Nurses, Community Health Officers, Community Health Extension Workers or anyone trained specifically to deliver community health interventions e.g. Volunteer Health Workers, Community Oriented Resource Persons.

Tropical febrile diseases such as malaria and enteric fever are significant public health problems in many parts of the world, particularly in African countries [3, 4]. Accurate diagnosis with traditional diagnostic methods, such as microscopy and serology are often resource-consuming and impractical in resource-limited settings. Hence, the diagnosis and treatment of tropical febrile diseases can be challenging for FHWs [5]. In many cases, diagnosis is based on clinical symptoms and signs, which can be non-specific and overlap with those of other diseases [6]. This leads to the need for more accurate and efficient diagnostic tools for tropical febrile diseases.

There have been calls for context-specific diagnostic tools for febrile syndromes at the point of care, with emphasis on evidence-based clinical algorithms and smart electronic devices [6]. These are based on the recent fruitful experiences in the use of digital health technologies as diagnostic tools [7, 8]. In cancer care, clinical decision support systems have been found to have the potential to improve the diagnosis, classification, and management of brain tumours [9]. Similarly, digital technologies have shown potential in the diagnosis and management of epidemic-prone diseases like Lassa fever and Ebola [10, 11].

Against this background, a multidisciplinary team of computational intelligence experts and medical professionals sought to address the gap in access to effective service delivery by developing and deploying an intelligent medical decision support system to help FHWs diagnose febrile illnesses. The purpose was to develop a decision-support system that is accurate and reliable for diagnosing febrile illnesses and is generalizable across different settings. This would result in improved patient outcomes including reduced morbidity and mortality due to timely and accurate febrile illness diagnoses.

The implementation strategy was designed using the RE-AIM framework, which considers reach, effectiveness, adoption, implementation, and maintenance. A logic model was developed to outline the pathway through which the intervention is expected to achieve its effects (see Fig. 1).

A pilot study is defined as a smaller version of the main study used to test whether the components of the main study can all work together [12]. Pilot studies, although requiring additional time and resources, are important for identifying and addressing any issues before progressing to a larger-scale study, ensuring the success of the intervention in the long term. As stated by Bond et al., “it is important to ensure that a good idea is not rejected because of a failure of one part of the system which should have been identified and addressed before progressing to a definitive study” [13]. Hence, this pilot study was done to determine the acceptability, appropriateness, and feasibility of the Febra App from the perspective of the FHWs.

OBJECTIVES

To determine the acceptability of the Febra App among FHWs in Nigeria.

To assess the appropriateness of the Febra App as perceived by FHWs in Nigeria.

To assess the feasibility of using the Febra App by FHWs to diagnose febrile illnesses in Nigeria.

Study design

A 3-month single-arm pilot study was conducted and evaluated using an observational cross-sectional design with mixed methods of data collection. The mixed methods approach used was the convergent parallel design [14]. This design facilitated the simultaneous implementation of both quantitative and qualitative strands during the research process, with equal priority given to both methods. The results from the two strands were integrated during interpretation to allow for triangulation and mutual corroboration.

Study location and context: The study was set in government-owned primary healthcare (PHC) facilities in both rural and urban areas of Akwa Ibom, Cross River, Imo, and Rivers States, situated in the Niger Delta area of Nigeria. Nigeria's healthcare system operates across three tiers: federal, state, and local governments oversee tertiary, secondary, and primary healthcare facilities, respectively. Frontline healthcare workers (FHWs)1 are primarily situated in public sector PHCs, often operating without direct supervision from senior cadre clinicians. In addition to FHWs, there are several semi-formal and non-formal healthcare providers, including patent medicine vendors, village health workers, and traditional birth attendants at the primary care level.

Several factors were considered as potential barriers to the implementation of a digital health intervention at the PHC level in this context, including poor internet connectivity, unreliable power supply, limited technological proficiency among FHWs, and a scarcity of human resources [15]. Moreover, the 2021 Multiple Indicators Cluster Survey reported an average mobile penetration rate of 87.6% in Nigeria and 83.8% in Akwa Ibom State, with an internet penetration rate of 34.6% in Nigeria and 27.7% in Akwa Ibom State [16]. Hence, while participants were expected to be somewhat tech-savvy due to these penetration rates, they still underwent rigorous training to effectively utilize the app.

Study Population

The study population included all healthcare workers involved in patient management in selected primary care facilities.

Sample size determination and sampling

From the list of PHCs obtained from the State Primary Health Care Development Agencies of Akwa Ibom, Cross River, Imo, and Rivers States, a total of forty-five PHCs were purposively selected for the pilot study. Fifteen PHCs were selected from Akwa Ibom, fourteen from Rivers, and eight from Imo and Cross Rivers states respectively (see Table 1). These selections aimed to encompass both urban and rural settings, with a particular focus on facilities with high patient loads (defined as seeing a minimum of fifty patients with febrile illness per week). Additionally, the selection criteria considered the number of health workers present in each facility (with a minimum of two healthcare workers), as well as the availability of sufficient power supply to charge their power banks.

Table 1

List of PHCs used for quantitative and qualitative studies.
S/N	STATE	PHC	Used for qualitative data collection?	Number tag for qualitative data
1	Akwa Ibom	PHC base Okopedi	No
2	Akwa Ibom	PHC Mbak Etoi	Yes	1
3	Akwa Ibom	PHC base Abak	No
4	Akwa Ibom	PHC Nung udoe Ibesikpo	Yes	2
5	Akwa Ibom	PHC Edeobom	No
6	Akwa Ibom	PHC base Okopedi	No
7	Akwa Ibom	PHC Afaha Offiong Nsit Ibom	No
8	Akwa Ibom	Primary Health Centre Abat	No
9	Akwa Ibom	Health Centre Ekparakwa	No
10	Akwa Ibom	PHC Base Idu Uruan	Yes	3
11	Akwa Ibom	PHC Eyotai	No
12	Akwa Ibom	PHC Ikot Okubo	Yes	4
13	Akwa Ibom	PHC Etebi Idun assan	No
14	Akwa Ibom	PHC Eyokponung	No
15	Akwa Ibom	PHC Ikot Ada Idem	Yes	5
16	Cross River	PHC Big Qua	Yes	1
17	Cross River	PHC Ikot Omin	Yes	2
18	Cross River	PHC Ikot Effanga Mkpa	Yes	3
19	Cross River	PHC Anantigha	No
20	Cross River	PHC Ekpo Abasi	Yes	4
21	Cross River	Health Center Uwanse	No
22	Cross River	PHC Ikot Offiong Ambai	No
23	Cross River	Ediba PHC, Calabar	No
24	Imo	Egbu Health Centre	No
25	Imo	Dindi PHC	Yes	1
26	Imo	PHC Nwaorieubi	No
27	Imo	PHC Amakohia	No
28	Imo	PHC Orogwe	Yes	2
29	Imo	Area M Comprehensive Health Centre	Yes	3
30	Imo	Irete Health Centre	No
31	Imo	Umuonoha	Yes	4
32	Rivers	PHC Omuoko Aluu	No
33	Rivers	MPHC Marine Base	No
34	Rivers	MPHC Mgbundukkwu	Yes	1
35	Rivers	MPHC Rumuodumaya	No
36	Rivers	MPHC Eliozu	No
37	Rivers	MPHC Aluu	No
38	Rivers	MPHC Rumuepirikon	No
39	Rivers	MPHC Rumueme	Yes	2
40	Rivers	MPHC College of Health	Yes	3
41	Rivers	MPHC Ozuoba	No
42	Rivers	MPHC Every Woman Centre	Yes	4
43	Rivers	MPHC Churchill	No
44	Rivers	MPHC Pott Johnson	Yes	5
45	Rivers	MPHC Mgbuoshimi	No
PHC: Primary Health Centre; MPHC: Model Primary Health Centre

Out of these facilities, eighteen were further purposively selected to participate in the qualitative interviews, ensuring representation from each state.

Intervention description

The Febra App, a medical decision support system (MDSS), was developed to assist frontline healthcare workers (FHWs) in diagnosing and managing febrile illnesses in low-resource settings. Our development process was iterative, focusing on understanding user needs and perspectives to tailor solutions accordingly. This approach allowed for continuous refinement and customization to meet the needs and perspectives of the users, thereby enhancing user engagement.

The development process involved several steps:

Two systematic reviews focusing on i) the use of soft-computing systems for the development of decision support systems for the diagnosis of tropical diseases [17], ii) the analysis of the actual techniques used in the development of these systems and factors influencing their adoption [8].
An adoption study collected data from 1117 respondents across four Nigerian states to determine factors encouraging the adoption of medical decision support systems [18].
Data collection from 62 medical experts and 4729 patients informed model development.
Intelligent decision support models were developed based on collected data: i) multicriteria decision analysis methods, ii) machine learning algorithms, and iii) fuzzy cognitive maps [19].
The Febra Diagnostica app ("Febra App") was built, incorporating insights from the adoption study, and powered by the best performing model, an analytic hierarchy process (AHP) diagnosis engine which is a multi-criteria decision analysis technique.
Continuous stakeholder engagement and ongoing refinement.

The AHP model was seen to outperform other techniques in diagnosis accuracy and had the added advantage of i) utilizing the experiential knowledge of medical experts; and ii) accepting model tuning based on medical and non-medical risk factors reported by the patient. Our study is one of the first to integrate these risk factors in the diagnosis process. This was informed by our previous study which determined that risk factors impact the accuracy of diagnosis [20].

Our development process also employed joint application development methodology, involving 37 individuals from five countries namely: Canada, Nigeria and Uganda, the United Kingdom, and the United States of America and from various disciplines in the fields of medicine, nursing, computer science, social works, and management. This created inherent robustness and internal checks to ensure that we moved away from theoretical research to real-world application.

Implementation strategies

Four key implementation strategies were employed for this pilot study

Recruitment and Training of FHWs: A comprehensive 2-day training session was conducted at central locations in each of the four states. The training module, crafted by the project team, covered the purpose, content, and operational aspects of the Febra App. Participants from all selected facilities, including facility heads, were invited, and duly trained and onboarded. State coordinators of the project and lead research assistants facilitated the training sessions.

Provision of Android Tablets: Android tablets preloaded with the application were distributed to all FHWs in the selected facilities. Additionally, power banks were provided to ensure continuous usage even in areas with poor power supply.

Utilisation of the Febra App by FHWs: FHWs were encouraged to actively incorporate the Febra App into their routine diagnostic and management processes for febrile illnesses.

Provision of Continuous Technical Assistance: Throughout the intervention period, the project team extended continuous technical support to participants. Open lines of communication via mobile phones and instant messaging apps such as WhatsApp were established, enabling FHWs to promptly relay any difficulties encountered to the project team for assistance.

Survey instruments and Data collection: A structured questionnaire was used for the quantitative aspect of this study. The questionnaire, hosted on “Open Data Kit” [21] collected data on the characteristics of respondents, acceptability, appropriateness, and feasibility of the Febra App. Measurement tools for acceptability, appropriateness, and feasibility were adapted from existing implementation research tools: the Acceptability of Intervention Measure (AIM), Intervention Appropriateness Measure (IAM), and Feasibility of Intervention Measure (FIM) [22].

For the qualitative component of the study, face-to-face, semi-structured in-depth interviews (IDIs) were used. The interviews focused on gathering feedback from FHWs regarding their experiences with the app, emphasizing ease of use, appropriateness, feasibility, and problems encountered. Trained interviewers conducted interviews, which were audio-recorded. They were then transcribed verbatim by a trained research assistant.

Data analysis

For the quantitative component, data was analyzed using IBM SPSS version 22 (New York). Descriptive statistics were used to summarize the demographic characteristics of the participants, and their responses to the implementation research tools. Results of the acceptability, appropriateness and feasibility measures were presented in bar charts.

For the qualitative analysis, transcribed interviews were manually coded following familiarization with the data. The analyses used thematic (deductive) codes. The coders drew the deductive codes from implementation outcomes (acceptability, appropriateness and feasibility) [23]; The project team then discussed the codes and grouped them into themes as presented in this paper.

Results from the qualitative and quantitative components were then triangulated.

This paper was written using the Standards for Reporting Implementation Studies (StaRI) guidelines [24] and CONSORT guidelines for (extension to randomised pilot and feasibility trials) [12].

A total of 52 respondents participated in the quantitative aspect of this study, while 17 respondents were interviewed for the qualitative aspect. In the quantitative survey, most of the respondents were females (97.6%). A higher proportion hailed from Akwa Ibom state (46.3%), were Community health extension workers (CHEWs) (56.1%) and had ≤ 10 years of experience (51.2%). The mean age of the FHWs was 40.1 years (see Table 2).

Table 2

Characteristics of participants
Characteristics	Number (%) (n = 52)
Age (years) ≤ 40 > 40 Mean + SD	27 (51.9) 25 (48.1) 40.6 + 9.3
Sex Male Female	1 (1.9) 51 (98.1)
State Akwa Ibom Cross River Imo Rivers	19 (36.5) 8 (15.4) 10 (19.2) 15 (28.8)
Job Title Nurse/Midwife CHEW Junior CHEW CHO	9 (17.3) 30 (57.7) 3 (5.8) 10 (19.2)
Years of Experience ≤ 10 > 10 Mean + SD	26 (50.0) 26 (50.0) 12.6 + 8.0
CHEW: Community health extension workers; CHO: Community health officer

Acceptability of the Febra App

The quantitative data showed that all respondents completely agreed or agreed that the Febra App met their approval, and they welcomed the App. The majority completely agreed or agreed that the App was appealing (98.1%), they liked using the App (98.0%) and it was easy to use (92.3%). Similarly, the majority either completely agreed or agreed that it helped them accomplish their tasks quickly (82.8%), although 13.5% of FHW disagreed with this (see Fig. 2).

Similarly, the qualitative study revealed that FHWs were comfortable using the application, had good experiences, and expressed appreciation for the tool. Some of the interviewed FHWs advocated for the integration of the app into the health system, emphasizing its role in helping them explore previously overlooked aspects of patients’ history that play a role in patient diagnosis and management.

“… more questions that were ignored before are being asked and the patient now knows most of the things that they used to overlook when it comes to a normal one…” Cross River 1

“I feel it should be introduced into the health system, into all the health clinics.” Akwa Ibom 3

“If the manufacturer and the producers of this App and those who are managing this program can help us at the facility level to be using the App on a daily basis, it will go a long way in helping us.” Cross River 2

When questioned about their clients' reactions to the Febra App, FHWs provided mixed responses. While some FHWs noted that their clients were impatient with the questions, others reported that clients were pleased with the integration of technology into their management and expressed satisfaction with their treatment.

“…it will make patients complain that it is really delaying their time just like the one I experienced, patients were saying, please leave it just attend to me let me go, it’s delaying my time, I have somewhere to go.” Rivers 2

“They were excited. The joy in them was there. Somebody asked them questions; they want to know how they are faring. If you go more in detail, knowing if they wash their hands, they sleep under (insecticide-treated) net, they like feeling that care more than before.” Rivers 3

Appropriateness of the Febra app

In the quantitative study, most surveyed respondents completely agreed or agreed that the Febra app was relevant (98.0%), and the tool applied to their routine diagnosis and management of febrile illnesses (98.1%). Likewise, all respondents completely agreed or agreed that the Febra App was suitable for managing febrile illnesses and it was a good match (see Fig. 3).

The qualitative study provided additional insights and identified nuanced perspectives. Interviewed FHWs expressed that the application was beneficial and enhanced their management of febrile illnesses. This was attributed to its role in augmenting their knowledge of the different tropical febrile illnesses, providing a structured approach to essential questions, and occasionally identifying comorbidities.

“I want to commend the coordinators…. The app has been a wow experience. It has expanded our skills.” Imo 2

“…there is a patient that was brought here with a complaint of a very high fever... Come and do tests, she said no. So, this Febra Diagnostica made us know what the cause of fever is. With it, we were able to discover that the patient is an HIV patient but hiding it from us. So, when we finally tested, we discovered that the patient was (sero-) reactive. We enrolled the patient, and everything was okay.” Cross River 4

“…getting to ask them those questions, it helps us to quickly suspect, this is the cause of this...” Rivers 5

Additionally, the App was noted to be particularly suitable for health centres that lacked laboratory facilities.

“….so that even if there is nothing like lab technician around the area, the device will always give you after you’ve seen the patient the likely and not likely will come up and you’ll know where to start up treatment.” Cross River 1

The App was also noted to be cost-saving in that it rightly diagnosed the patients’ illnesses, and suggested appropriate medications when the patients had no funds to carry out investigations.

“Some persons tell us they don’t have money for tests... By the time you give them the drugs that have been suggested by the App, you see that you rightly treat these persons without most times doing the tests… for them to spend further money.” Rivers 5

However, some FHWs found the app to be unsuitable for their settings, particularly due to the high volume of patients they dealt with daily. They expressed that going through all the questions in the application was impractical given their workload.

“…it is a bit time-wasting… some (patients) are uneducated and if you ask questions, they don’t give you the right word, …by the time you come down to their level, if you were using like 10 mins to see a patient before you go through all these, it will get to 20–30 mins...” Rivers 1

“When using this tool, we need more hands in order for us to succeed in what we’re doing because one person cannot attend to a crowd with Febra Diagnostica.” Cross River 3

Feasibility of using the Febra App routinely by FHWs

Out of 52 respondents, the majority either completely agreed or agreed that the Febra App was suitable for everyday use (94.2%). A similar percentage also completely agreed or agreed that the app seemed possible to use (96.2%) and could be used at a large scale (92.3). All respondents agreed that the App seemed easy to use (see Fig. 4).

The qualitative study also revealed that overall, the App was simple and easy to use, particularly after the training that FHWs received before the study commenced.

“I feel so comfortable using it because all the information is already stored up there and data also goes directly to the appropriate place. It is very easy. It’s not difficult to use because all the buttons and everything about the device is clear, so I find it very easy to use.” Cross River 1

However, several feasibility issues with the current design of the App were unveiled. These were split into technical and non-technical issues. The major non-technical issue raised by several FHWs was the time required to navigate through all the questions in the App. They noted that with numerous patients, it was impractical to consistently go through every question.

“Since the App involves so much in terms of documentation, it took so much time comparatively because we had to take a comprehensive history of each client we needed to see and that usually takes us more than 30 minutes compared to the conventional method.” Akwa Ibom 5

Technical issues were highlighted that could impact the routine use of the App. They noted instances where the tool crashed while attempting to submit completed forms, and consultations were occasionally interrupted by poor network connections. Additionally, several FHWs pointed out the inability to minimize one client’s form and open another while awaiting laboratory results as another significant issue that could hinder routine use.

“… at a point, it was giving some issues like…sometimes it seems like it’s out of data and then I have to use my phone to hotspot the App. Sometimes, I also notice that my work is not going to the cloud.” Rivers 4

“…this one you have to wait until the lab result is out before you attend to another patient… that’s why we could not meet up with 100 patients because if you have many patients in a day, you won’t be able to meet up with the App...” Akwa Ibom 3

Recommendations for Improvement from FHWs

The FHWs offered several recommendations for improving the application. Common suggestions included shortening the questions, enabling the minimization of patient tabs, and allowing the addition of new medications to those suggested by the App.

“…the questions should be straight forward and simple and short so that we’ll be able to accommodate, treat more patients…” Cross River 3

“They should make provisions so that when we stop before the lab, when the lab result is out, we come back to where we stop and key in the result, so that you will be able to meet up with enough patients….” Imo 3

“...there should be provisions to add more medications… Provisions should be made so that you can add as many as possible medications based on the patient’s symptoms and signs.” Imo 4

The findings of this pilot study shed light on the acceptability, appropriateness and feasibility of the Febra App, an intelligent medical decision support system designed to aid FHWs in diagnosing and managing febrile illnesses in resource-limited settings. The study found that overall, the Febra App was acceptable, appropriate, and feasible for use by frontline health workers to diagnose febrile illnesses. However, non-technical issues like time consumption, and technical issues like poor internet connectivity and occasional crashes were raised.

The high level of acceptability and appropriateness reported by FHWs in both the quantitative and qualitative aspects of the study highlight the potential value of the Febra App in the context of PHC facilities in Nigeria. The positive feedback regarding the ease of use, relevance, and suitability of the app for routine diagnosis and management of febrile illnesses is a strong foundation for its integration into routine healthcare delivery. Moreover, the qualitative insights threw light on the usefulness of the Febra App in augmenting FHW’s knowledge and streamlining diagnostic processes. These findings mirror other studies that have highlighted the potential of digital health interventions to enhance healthcare delivery in settings where access to traditional diagnostic methods is limited [7, 10]. Studies examining similar interventions have reported positive outcomes in terms of healthcare worker satisfaction [25].

The Febra App demonstrated promising feasibility for routine use by FHWs, however, several technical and non-technical challenges were identified. Technical issues such as application crashes, poor network connectivity, and limitations in multitasking capabilities hindered the seamless integration of the app into daily workflows. These challenges not only disrupted consultations but also contributed to time inefficiencies, particularly in settings with high patient volume. Similar challenges have been found to hinder sustainability [26] and must therefore be addressed going forward. Non-technical challenges, such as the time-consuming nature of navigating through all the questions in the app, further compounded the feasibility issues. Addressing these challenges will be crucial for ensuring the sustainable adoption and scalability of the Febra App in real-world healthcare settings.

Healthcare workers' perceptions about the acceptability, appropriateness, and feasibility of an intervention are good predictors of adoption [8, 27], hence this pilot shows that adoption of the Febra App in PHC facilities is possible. The rigorous pre-intervention training and availability of a continuous technical assistant were key factors that played a role in the success of this pilot. Evidence from previous studies demonstrates that FHW programs are likely to succeed if they are given appropriate training and skills, clear tasks and supportive supervision [28, 29]. However, concerns arise regarding the scalability and sustainability of the intervention, particularly in terms of maintaining continuous technical assistance. While periodic refresher training is feasible within this setting, sustaining ongoing technical support will necessitate continuous funding. Therefore, for long-term sustainability, it is imperative to scale up the intervention and garner support from stakeholders with sufficient resources to sustain the human resources for continuous technical assistance.

One of the implementation strategies in this study was the provision of tablets, which was chosen over using staff’s mobile devices. While being more expensive, this approach ensured that the staff had appropriate devices to work with and delineated the staff’s personal life from work. They were also provided with power banks to ensure the availability of power. The cost of smartphones, tablets and accessories has often been cited as a concern regarding the scalability of similar digital health interventions [30]. However, using the tablets for multiple applications, such as case management, stock control, mobile learning and assessment, and supervision will make them more cost-effective [31].

Strengths and limitations

This pilot study benefits from a robust mixed-methods approach, combining a quantitative survey with qualitative interviews to provide comprehensive insights into the feasibility, acceptability, and appropriateness of the Febra App. Moreover, the multidisciplinary collaboration between computational intelligence experts and medical professionals ensured the development and refinement of the Febra App based on evidence-based clinical algorithms and user needs. Continuous stakeholder engagement and iterative refinement processes contributed to the user-centred design and customization of the App to suit the context and preferences of FHWs.

Despite the strengths of this study, several limitations should be acknowledged. Firstly, the relatively small sample size and convenience sampling approach adopted in this study may introduce selection bias and limit the representativeness of the study population. However, this was a pilot study, and the larger-scale implementation will provide a more robust assessment of the intervention's impact and scalability. Secondly, the fidelity of the implementation was not assessed due to limited resources. Hence, it is not possible to have complete certainty about how clinical consultations went using the App, and any adaptations that were employed during implementation.

Thirdly, this study was done on the assumption that test kits for routine investigations such as malaria and Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome rapid diagnostic tests were available at the PHCs. However, test kits for rarer febrile illnesses yellow fever, dengue fever and Lassa fever are not available routinely at PHCs. Additionally, it was realized after the pilot that the signs and symptoms used by the model to generate diagnoses were not stored on the data cloud, limiting our ability to assess the performance of the model. This has however been rectified.

Despite these limitations, the findings of this pilot study provide a valuable foundation for future research and implementation efforts aimed at leveraging digital health technologies to improve healthcare access and quality in resource-limited settings.

Recommendations

Shortening the questionnaire, enabling the minimization of patient tabs, and allowing flexibility in adding new medications based on patient symptoms were among the common suggestions by the FHWs. These recommendations align with the need for user-centred design principles and iterative refinement to enhance the usability and efficiency of the App.

Addressing technical issues such as application stability, network connectivity, and multitasking capabilities will be paramount for optimizing the feasibility and scalability of the Febra App.

Continuous stakeholder engagement, user feedback mechanisms, and agile development methodologies can facilitate ongoing improvement and adaptation of digital health interventions to evolving healthcare needs and contexts.

This pilot study provides insights into the feasibility, acceptability, and appropriateness of the Febra App as an intelligent medical decision support system for diagnosing and managing febrile illnesses in resource-limited settings. The FHWs were empowered to test the app in their facilities on patients seen in their routine clinics. The study found high levels of acceptability and appropriateness, but feasibility was moderate. This was likely modified by the challenges to using the app, including poor internet connectivity, occasional crashes and prolonged patient contact time.

While the findings highlight the potential benefits of digital health interventions in enhancing healthcare delivery, addressing technical and non-technical challenges will be essential for realizing the full potential of the Febra App in improving patient outcomes and strengthening healthcare systems. Future research and implementation efforts should focus on iterative refinement, user-centred design, and collaborative partnerships to ensure the successful integration and sustainability of the Febra App or similar interventions in diverse healthcare settings.

CHEW

Community health extension worker

FHW

Frontline Health Worker

LMIC

Low-and Middle-Income Country

PHC

Primary Health Center

Ethics approval and consent to participate: Procedures of our methods were reviewed and approved by the University of Uyo Teaching Hospital Institutional Health Review Ethics Committee, Cross River State Ministry of Health Ethics Review Board, Imo State Ethical Review Board, and Rivers State Ethical Review Board. Permission was obtained from all facility heads, and consent was obtained from all participating FHWs. Verbal consent was also obtained from all patients attended to using the App.

Consent for publication: Not applicable.

Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests: The authors declare that they have no competing interests.

Funding: This project was funded by the New Frontiers in Research Fund (NFRF) grant.

Authors' contributions: conception OR design of the work; OR the acquisition, analysis, OR interpretation of data; OR the creation of new software used in the work; OR have drafted the work or substantively revised it CA conceptualized the study, interpreted the data, and assisted in software creation. VE conducted the data analysis, interpreted the data, and drafted the initial manuscript. NE, OM, OM, UE, and CMI interpreted the data, contributed to software creation, and revised the work. EU, OO, KA, and OU assisted in study design, data acquisition, and software development. KA, BB, and CA contributed to the study design and the creation of the user interface for the new software. F-MU conceptualized the study, provided oversight during data acquisition and software creation, interpreted the data, and revised the manuscript.

Acknowledgements: Not Applicable

World Health Organization. Global strategy on human resources for health: Workforce 2030 [Internet]. World Health Organization; 2020 [cited 2024 Mar 1]. https://www.who.int/publications-detail-redirect/9789241511131.
Boniol M, Kunjumen T, Nair TS, Siyam A, Campbell J, Diallo K. The global health workforce stock and distribution in 2020 and 2030: a threat to equity and ‘universal’ health coverage? BMJ Global Health. 2022;7(6):e009316.
Adesegun OA, Adeyemi OO, Ehioghae O, Rabor DF, Binuyo TO, Alafin BA, et al. Current trends in the epidemiology and management of enteric fever in Africa: A literature review. Asian Pac J Trop Med. 2020;13(5):204.
World Health Organization. Fact sheet about malaria [Internet]. 2023 [cited 2024 Mar 1]. https://www.who.int/news-room/fact-sheets/detail/malaria.
Mahero MW, Pelican KM, Waila JM, Namusisi S, Rwego IB, Kajura C, et al. There are many fevers: Communities’ perception and management of Febrile illness and its relationship with human animal interactions in South-Western Uganda. PLoS Negl Trop Dis. 2022;16(2):e0010125.
Bottieau E, Yansouni CP. Fever in the tropics: the ultimate clinical challenge? Clin Microbiol Infect. 2018;24(8):806–7.
Keitel K, Kagoro F, Samaka J, Masimba J, Said Z, Temba H, et al. A novel electronic algorithm using host biomarker point-of-care tests for the management of febrile illnesses in Tanzanian children (e-POCT): A randomized, controlled non-inferiority trial. PLoS Med. 2017;14(10):e1002411.
Khan N, Nwafor Okoli C, Ekpin V, Attai K, Chukwudi N, Sabi H, et al. Adoption and utilization of medical decision support systems in the diagnosis of febrile Diseases: A systematic literature review. Expert Syst Appl. 2023;220:119638.
Mukherjee T, Pournik O, Lim Choi Keung SN, Arvanitis TN. Clinical Decision Support Systems for Brain Tumour Diagnosis and Prognosis: A Systematic Review. Cancers. 2023;15(13):3523.
Emperador DM, Yimer SA, Mazzola LT, Norheim G, Kelly-Cirino C. Diagnostic applications for Lassa fever in limited-resource settings. BMJ Glob Health. 2019;4(Suppl 2):e001119.
Oluwagbemi O, Oluwagbemi F, Abimbola O. Ebinformatics: Ebola fuzzy informatics systems on the diagnosis, prediction and recommendation of appropriate treatments for Ebola virus disease (EVD). Informatics in Medicine Unlocked. 2016;2:12–37.
Eldridge SM, Lancaster GA, Campbell MJ, Thabane L, Hopewell S, Coleman CL, et al. Defining Feasibility and Pilot Studies in Preparation for Randomised Controlled Trials: Development of a Conceptual Framework. PLoS ONE. 2016;11(3):e0150205.
Bond C, Lancaster GA, Campbell M, Chan C, Eddy S, Hopewell S, et al. Pilot and feasibility studies: extending the conceptual framework. Pilot Feasibility Stud. 2023;9(1):24.
Creswell JW. Mixed Methods Procedures. In: Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. 4th Edition. SAGE; 2014. p. 215.
Akwaowo CD, Motilewa OO, Ekanem AM. Assessment of Resources for Primary Health Care: Implications for the Revitalization of Primary Health Care in Akwa Ibom, Nigeria. Niger Med J. 2020;61(2):90–5.
National Bureau of Statistics. 2021 Nigeria Multiple Indicator Cluster Survey [Internet]. 2021 [cited 2024 Mar 11]. https://nigerianstat.gov.ng/elibrary/read/1241209.
Attai K, Amannejad Y, Vahdat Pour M, Obot O, Uzoka FM. A Systematic Review of Applications of Machine Learning and Other Soft Computing Techniques for the Diagnosis of Tropical Diseases. Trop Med Infect Dis. 2022;7(12):398.
Sabi H, Akwaowo CD, Umoh E, Ekpenyong N, Isiguzo CM, Maduka O et al. Adoption of medical decision support systems in Nigeria – Evidence from a survey of four states in Southern Nigeria. 2023.
Asuquo D, Attai K, Okure O, Ekpenyong M, Akwaowo C, Uzoka FM. Febrile Disease Modeling and Diagnosis for Improving Medical Decisions in Low- and Middle-Income Countries. In Kumasi Ghana.
Uzoka FME, Akwaowo C, Nwafor-Okoli C, Ekpin V, Nwokoro C, El Hussein M, et al. Risk factors for some tropical diseases in an African country. BMC Public Health. 2021;21(1):2261.
ODK Collect —. Open Data Kit Docs [Internet]. [cited 2020 Mar 12]. https://docs.opendatakit.org/collect-intro/.
Weiner BJ, Lewis CC, Stanick C, Powell BJ, Dorsey CN, Clary AS, et al. Psychometric assessment of three newly developed implementation outcome measures. Implement Sci. 2017;12(1):108.
Proctor E, Silmere H, Raghavan R, Hovmand P, Aarons G, Bunger A, et al. Outcomes for implementation research: conceptual distinctions, measurement challenges, and research agenda. Adm Policy Ment Health. 2011;38(2):65–76.
Pinnock H, Barwick M, Carpenter CR, Eldridge S, Grandes G, Griffiths CJ, et al. Standards for Reporting Implementation Studies (StaRI) Statement. BMJ. 2017;356:i6795.
Burke HM, Wamburi A, Rademacher KH, Wong CM, Were E, Hoppes E et al. A Field Test of the NORMAL Job Aid With Community Health Workers in Kenya to Address Contraceptive-Induced Menstrual Changes. Glob health sci pract. 2023;11(1).
Adepoju IOO, Albersen BJA, Brouwere VD, van Roosmalen J, Zweekhorst M. mHealth for Clinical Decision-Making in Sub-Saharan Africa: A Scoping Review. JMIR mHealth uHealth. 2017;5(3):e7185.
Proctor EK, Bunger AC, Lengnick-Hall R, Gerke DR, Martin JK, Phillips RJ, et al. Ten years of implementation outcomes research: a scoping review. Implement Sci. 2023;18(1):31.
Scott K, Beckham SW, Gross M, Pariyo G, Rao KD, Cometto G, et al. What do we know about community-based health worker programs? A systematic review of existing reviews on community health workers. Hum Resour Health. 2018;16(1):39.
Bakkabulindi P, Ampeire I, Ayebale L, Mubiri P, Feletto M, Muhumuza S. Engagement of community health workers to improve immunization coverage through addressing inequities and enhancing data quality and use is a feasible and effective approach: An implementation study in Uganda. PLoS ONE. 2023;18(10):e0292053.
Free C, Phillips G, Galli L, Watson L, Felix L, Edwards P, et al. The Effectiveness of Mobile-Health Technology-Based Health Behaviour Change or Disease Management Interventions for Health Care Consumers: A Systematic Review. PLoS Med. 2013;10(1):e1001362.
Tomlinson M, Rotheram-Borus MJ, Swartz L, Tsai AC. Scaling Up mHealth: Where Is the Evidence? PLoS Med. 2013;10(2):e1001382.

Globally, Frontline health workers comprise all types of health workers—including nurses, midwives, community health workers, doctors, pharmacists, and more—who provide care directly to their communities. However, in this study, we use the term for the middle cadre workers trained to deliver health services formally; Public Health Nurses, Community Health Officers (CHOs), Community Health Extension Workers (CHEWs), or anyone trained specifically to deliver community health interventions e.g. Volunteer Health Workers, Community Oriented Resource Persons (CORPs).

STROBEchecklistFeasibilitypaper.docx

Download PDF

Reviewers invited by journal
04 Aug, 2024
First submitted to journal
13 Jun, 2024

You are reading this latest preprint version

Feasibility Testing of an Application to Improve the Diagnosis and Management of Febrile Illnesses by Frontline Health Workers in Low- and Middle-Income Countries

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

KEY MESSAGES

BACKGROUND

METHODS

RESULTS

Acceptability of the Febra App

Appropriateness of the Febra app

Feasibility of using the Febra App routinely by FHWs

Recommendations for Improvement from FHWs

DISCUSSION

Strengths and limitations

CONCLUSION

Abbreviations

Declarations

References

Footnotes

Supplementary Files

Status:

Version 1