Community Health Workers and Disease Surveillance in Tanzania: Promoting the Use of Mobile Technologies in Detecting and Reporting Health Events

Abstract

This cross-sectional study was conducted in the Kilosa, Morogoro Urban, Ngorongoro, and Ulanga districts of Tanzania to investigate the practices of community health workers (CHWs) related to disease surveillance functions and to establish their needs and technology capacities. We also established the strength of mobile phone networks and internet connections in the study areas to inform the feasibility of using mobile-based applications in community-based disease surveillance. A total of 135 CHWs from 85 villages participated in the study. Health events captured at the community level were entirely paper-based. CHWs submitted reports to higher-level health authorities mainly on foot (100%), but they also used public transport (65%) and telephone calls (56%). The median number of days between the onset of a suspected disease outbreak at the community level and reporting to a primary healthcare facility was 10 days (interquartile range [IQR] 2-30). The median number of days between submitting a report and receiving a response was 7 days (IQR 2-30). Of the 53 CHWs who reported the most recent health events to a higher-level health authority, 39 (74%) never received feedback. All 85 villages had a reliable mobile phone network and 74 (87%) had a mobile phone internet connection that was strong enough to support data transmission using digital technology. Almost all (n = 132, 98%) of the CHWs owned mobile phones. The practices related to detection and reporting of health events could be improved to enhance early warning disease surveillance. Reliable mobile networks and internet connections and the ownership of mobile phones among CHWs in the study areas present opportunities to strengthen community event-based surveillance using mobile-based solutions.

Introduction

In recent decades, the world has witnessed an apparent increase in emerging and reemerging infectious diseases, with the largest burden observed in Africa due to weaker health systems.^1,2 Tanzania is among the countries that have experienced a high burden of infectious diseases, characterized by high mortality and morbidity rates. Such diseases include malaria, respiratory tract infections, HIV/AIDS, and diarrheal diseases.^3,4 Diseases that have the potential to become epidemics include dengue, chikungunya, Rift Valley fever, influenza A (H1N1), cholera, and novel coronavirus disease 2019 (COVID-19).^5-13 Six diseases transmissible between animals and humans (zoonoses) that have been characterized as high priority in Tanzania are rabies, anthrax, Rift Valley fever, zoonotic avian influenza, human African trypanosomiasis, and brucellosis.¹⁴

Effective early warning and response systems are instrumental to combat disease outbreaks at the source to minimize associated negative impacts.^15,16,17 Because disease outbreaks typically erupt at the community level, it is plausible to suggest that community members could play a pivotal role to enhance early warning systems. The current human and animal health surveillance systems and strategies in the World Health Organization member states are based on the 2005 International Health Regulations and the World Organisation for Animal Health, which mandate the flow of information from primary sources of events to the global level to ensure national and global health security.^18,19 However, in most low- and middle-income countries, disease surveillance systems are weak.^20-22 Suboptimal performance of surveillance systems in these countries has been associated with poor linkages to the community level.^21,23 Community event-based surveillance has been underscored as an important component of early warning systems for the early containment of disease outbreaks at the source before they spread further into a regional epidemic or become a global pandemic threat.^24-26 To realize this, improvement in timeliness has been highlighted as one of the key disease surveillance metrics.²⁷

As approximately three-quarters of emerging human infectious diseases have animal origins,²⁸ community event-based surveillance grounded in One Health principles (which address public health events at the interface of human, animal, and environmental health) has the potential to enhance early detection of human and animal diseases at their sources. This could be achieved by empowering communities to take ownership and control over local decisions and have a stake in maintaining the surveillance structures and practices, and thus improving community One Health security (ie, the measures that protect humans, animals, and their environment against health threats).^29,30 One of the key elements in the functionality of disease surveillance for early warning systems is the use of community health workers (CHWs).³¹ CHWs are trained to assist in the communication or provision of healthcare³² at the community level, particularly in underserved or remote populations. Most often, CHWs are trained in the context of specific health interventions to carry out defined functions related to healthcare delivery.³³ CHWs are selected by their communities, provided with short training, and held accountable to their communities.³⁴ Because they live within the communities and have gained the trust and good understanding of the communities they serve, they can make important contributions to human health.^31,35,36

Despite the growing realization that the role of CHWs can enhance disease early warning,^30,31,37-39 few have studied their recruitment and practices related to detection, recording, reporting, and response to health events in human and animal populations in Tanzania. Some studies have evaluated the effectiveness of CHWs but mainly for specific programs such as improving the uptake of maternal healthcare services in low- and middle-income countries.^31,40-42 A few studies have reported on how CHWs spend their time on health-related services at the community level.^43,44 One study in West Africa recently explored and documented the roles of CHWs in pandemic preparedness.⁴⁵

Recognizing the importance of CHWs in bridging the community and healthcare system to enhance early detection and containment of diseases outbreaks at the source, we conducted the current study to establish benchmark values of community event-based surveillance performance in Tanzania. Specifically, we aimed to: (1) identify the presence of CHWs at the village level and their recruitment terms of reference; (2) investigate CHW practices related to disease surveillance; (3) establish the needs and technology capacities of CHWs in disease surveillance; and (4) establish the availability and strength of mobile phone networks and internet connections. This data will likely provide guidance in strengthening community-based One Health surveillance systems in Tanzania and beyond. The findings can be used to inform the adoption of technological innovations to enhance the efficiency of CHWs to improve One Health security, ultimately contributing to national, regional, and global health security.

Methodology

This study involved engaging with local authorities to identify CHWs. Data were collected through in-person interviews with CHWs. Interpretation of the data and generation of inferences were made based on descriptive analysis.

Study Area and Design

We conducted a cross-sectional study in the Morogoro Urban, Kilosa, Ulanga, and Ngorongoro districts of Tanzania from January 2016 to December 2017 (Figure 1). The districts were selected based on specific characteristics. Morogoro Urban forms the hub of a network of major roads with settlements by people from different areas of the country. Kilosa and Ulanga are characterized by extensive agriculture and pastoralism, which increases the risk of disease transmission between animals and humans. Ngorongoro is located in the cross-border ecosystem between Tanzania and Kenya and is characterized by pastoralism with frequent interaction between humans, domestic and wild animals, and their cross-border movements, which could influence disease transmission dynamics. According to the Tanzania National Population and Housing Census,⁴⁶ Morogoro Urban has 64 health facilities (48 dispensaries, 13 health centers, and 3 hospitals) that serve 295 streets (equivalent to villages in the rural setting), with a total population of 315,001 people. Kilosa has 65 health facilities (56 dispensaries, 6 health centers, and 3 hospitals) that serve 139 villages with a total population of 500,001 people. Ulanga has 24 health facilities (21 dispensaries, 2 health centers, and 1 hospital) that serve 59 villages with a total population of 151,001 people. Ngorongoro has 33 health facilities (26 dispensaries, 5 health centers, and 2 hospitals) that serve 72 villages with a total population of 174,278 people. The number of CHWs in the study districts is unknown.

Figure 1.

Map of study districts in Tanzania.

Selection of Study Villages and Streets

In each district, introductory meetings were held with the respective district executive directors, district medical officers, district veterinary officers, and district Integrated Disease Surveillance and Response focal persons. Each district nominated a focal person to coordinate the study in the respective district. In each district, discussions were held with officials responsible for human and animal disease surveillance to agree on study implementation, including the number and location of villages and streets to include. The location of villages and streets included in each district was based on their remoteness in terms of distance from district headquarters, poor accessibility, history of disease outbreaks in human and animal populations, and perceived high vulnerability to disease outbreaks. Using these criteria, the final number of villages and streets was influenced by available resources. A total of 85 villages and streets were selected: 49 in Kilosa, 18 in Ulanga, 11 in Ngorongoro, and 7 in Morogoro Urban.

The research team spent 2 days in training on study protocol, data collection tools, research methods, and ethics. This training was followed by 1 day of piloting data collection tools in an area in Morogoro Urban that was not included in the actual study. Data collection tools were revised and updated before data collection began.

Data Collection

All of the CHWs in the study sites were targeted for interviews. Research teams and district focal persons visited the selected villages and streets and introduced the study. In each village, ward and village leaders were consulted to obtain a list of CHWs and their contacts. During these consultative meetings, the research team collected information from the leaders on how the CHWs had been recruited and what roles and responsibilities were included in their operational terms of references. The research team visited CHWs for scheduled in-person interviews. Using a semi-structured questionnaire, the team documented CHW practices in detection, recording, reporting, and response related to health events in human and animal populations. Individual interviews were conducted in Kiswahili for each CHW and took approximately 90 minutes to complete. The researchers made active calls to assess the strength of the mobile phone network in the villages. Internet strength was assessed by attempting to submit dummy data to a server using the AfyaData digital disease surveillance app installed in smartphones. The research coordinator was responsible for data quality assurance in the field by performing a random examination of 20% of data collection forms completed in a day.

Data Analysis

The study team entered data into a Microsoft Excel spreadsheet and verified the data by checking each individual variable for coherence, missing observations, and potential mistakes. The cleaned data were transferred to Stata version 13 (Stata Corporation, College Station, TX) for descriptive analysis. Frequency distribution of variables were produced, and results were summarized in graphs. Continuous variables were summarized into median and interquartile range and categorical variables into proportions. Data analysis output was aggregated by district, as the unit responsible for coordinating human and animal disease surveillance activities.

Ethics Statement

This study was approved by the Tanzania Medical Research Coordinating Committee of the National Institute for Medical Research (NIMR/HQ/R.8a/Vol. IX/2037). Approval for its implementation was sought from the respective regional, district, ward, and village authorities. Verbal informed consent was sought from participants prior to their enrollment in the study. Anonymous data collected were stored in password-protected computers with access limited to only the researchers involved in the study.

Results

Analysis of collected data revealed common approaches to CHW recruitment and assignment of duties in the study areas. CHWs were primarily involved in universal health programs including programs related to HIV/AIDS, tuberculosis, and prenatal, maternal, and postnatal healthcare. Their terms of engagement revealed commonalities across the study areas—none had job descriptions or were aware of their terms of references for the work they were performing. The strength of mobile phone networks and internet connections in the study areas present an opportunity for technology-driven approaches that can strengthen community event-based surveillance.

Recruitment and Terms of Reference

Each of the 85 study villages and streets in the study area had 2 CHWs. They were selected by community members through a voting process. Except for Ulanga, the selection process involved community members nominating potential candidates during general public meetings and then voting on the candidates. Predefined criteria set by the village and street health committee and local leaders helped guide the recruitment process. These criteria were communicated at the general public assembly on the day of recruitment. The minimum selection criteria included residing in the village where one would serve; at least 18 years of age; respected, trusted, and acceptable to the community; having at least primary school education (7-year education); and agreeing to work on voluntary basis. Additional demographic considerations were for each village to recruit 1 male and 1 female CHW. There was no evidence of a plan for preservice training or in-service training for the work CHWs were expected to do except when they had taken part in specific public health programs.

For Ulanga, a similar approach was adopted but with some modifications. CHWs were recruited through a process that involved the local government authority in the screening of applications, with minimum qualifications including the completion of secondary school (11-year education). The names of selected candidates were presented to the general village assembly for voting. Before working in their home villages, the selected individuals underwent a 9-month preservice training on a child survival program, an initiative that was made through the Connect Project.⁴⁷

In all study districts, the names of selected CHWs were submitted to their respective village councils for approval and registration. The CHWs received information regarding the catchment areas they were required to serve, which was defined by specific village and street boundaries. With the exception of those in Ulanga, however, information was missing about their roles and responsibilities and mechanisms for carrying out daily activities. For CHWs not working with a specific health program, it was not clear who was responsible for their supervision and monitoring of their performance. CHWs working on specific programs received supervision from primary healthcare facilities and program coordinators.

Sociodemographic Characteristics

Overall, a total of 170 CHWs in 85 villages and streets were included in this study. Of these, 135 (79%) participated based on their availability during the visit to the village or street and willingness to take part in the study. The majority (56.3%) were male. All of the CHWs were working with human health domain and were from Kilosa (n = 73), Ulanga (n = 33), Ngorongoro (n = 22), Morogoro Urban (n = 7) districts. A total of 97 (72%) CHWs had a primary school level of education and 38 (28%) had completed secondary school. CHWs from Ulanga received a salary, whereas the rest were volunteers. Over half (58%) had 5 to 10 years of experience, 22% had less than 5 years of experience, and 20% had more than 10 years of experience. Those with more than 10 years of experience were from Kilosa and Ngorongoro (Figure 2). The overall median years of experience working as a CHW was 7 (interquartile range [IQR] 2-35). The median years of experience reported by CHWs from each district were 9 (IQR 4-17) in Kilosa, 8 (IQR 3-13) in Ngorongoro, 5 (IQR 2-7) in Morogoro Urban, and 4 (IQR 3-7) in Ulanga.

Figure 2.

Number of years of experience among community health workers by district.

Roles and Responsibilities

The main roles of CHWs consisted of community health promotion and education on various health-related issues including sanitation and hygiene practices, participation in specific disease control programs such as HIV/AIDS and tuberculosis, and prenatal, maternal, and postnatal healthcare. All 135 CHWs had obtained training on specific health programs they had worked with including mother and child health (98%), care for people living with HIV/AIDS (95%), sanitation and hygiene practices (24%), and care for tuberculosis patients (16%). CHWs who had received training on sanitation and hygiene practices were from Ulanga and those who received training on care for people living with tuberculosis were from Ngorongoro (Figure 3).

Figure 3.

Type of training experience among community health workers by district.

Roles and responsibilities related to disease surveillance were infrequently performed by CHWs. Only the CHWs from Ulanga (n = 33) reported having received training on disease surveillance and response in human populations, and the topics covered included detection, recoding, reporting, and responses (actions) on reported public health events. The CHWs from Kilosa (n = 73) and Ulanga (n = 33) were scheduled to work at primary healthcare facilities 1 day per week during antenatal care clinics to assist with maternal and child healthcare services. The activities they performed at health facilities included sensitization on family planning, immunization, HIV testing, prevention of malaria, and nutrition. They also provided support by recording weight and growth scores of babies at health facilities on antenatal care clinic days.

None of the CHWs interviewed in Morogoro Urban, Kilosa, and Ngorongoro had written job descriptions, and were likewise unaware of their terms of references for the work they were performing. In addition, they were not explicitly aware of the duties they were required to perform. A common perceived challenge related to implementing activities reported by all CHWs who were working with specific programs was the lack of involvement during the design stage of the program they were working with.

Experience with Detection of Disease Outbreaks

In-person interviews were used to explore the experience of CHWs with disease outbreaks in the past 10 years. Of the 135 CHWs, 125 (93%) reported having ever experienced a disease outbreak in their community. All CHWs reported clinical manifestation as an indicator of the presence of diseases in humans and animals. An increase in the number of cases (97%) and deaths (72%) beyond what was perceived as normal were reported as the basis for them to suggest that the observed health events indicated a possible disease outbreak. However, the CHWs could not consistently recall the number of cases. Based on clinical manifestations presented, they suspected the outbreaks to be chicken pox and cholera in humans (2015) in Morogoro Urban and Kilosa; peste des petitis ruminants in goats in Ngorongoro (2015) and Kilosa (2016); and contagious bovine pleuropneumonia in cattle in Ulanga (2010), Kilosa (2014), and Ngorongoro (2015). Others included foot-and-mouth disease in cattle and Newcastle disease in chickens in Morogoro Urban (2016). It was not possible, however, for CHWs to share precisely the number of cases for each outbreak.

Interactions with Community Members

CHWs were informed by community members about the occurrence of health events (100%) and through regular visitations to households (58%). Half (51%) of the CHWs who reported visiting households regularly were from Kilosa (Figure 4). Visits to households were made weekly (37%), biweekly (27%), and monthly (36%). The mechanism by which CHWs received information from community members about the occurrence of health events was mainly when a community member visited the CHWs (100%) but also by phone (48%).

Figure 4.

Frequency of community health worker visits to households by district.

Data Capture and Record Keeping

All data regarding health events and outbreaks captured at the community level were captured on paper, a practice reported only for human health events. Almost one-third (31%) of CHWs reported having ever recorded a health event, including recording suspected disease outbreaks in notebooks or loose papers. One CHW reported having recoded events in the phone as a short message text (SMS). The recording of health-related events was reported to happen infrequently mainly due to the absence of structured forms (96%) and inadequate training (72%). For those who reported having ever recorded a health event (31%), frequently recorded information included the date of the event, village name, ward name, clinical manifestations, age, sex, district name, and number of deaths (Figure 5). All 33 CHWs from Ulanga reported using their training skills to report suspected disease outbreak data, whereas only 9 of the 102 (9%) CHWs from Kilosa, Ngorongoro, and Morogoro Urban reported using their previous experience to take part in disease surveillance. None of the CHWs reported having guidelines, job aids, manuals, standard case definitions, or any other instructions for the proper recording of health events. Lack of recording forms was the most frequent challenge encountered by all CHWs while documenting health events.

Figure 5.

Frequency and type of health information recorded by community health workers.

Timeliness of Reporting and Response

Overall, 53 (39%) CHWs had reported the most recent public health events to higher levels, including reporting suspected disease outbreaks that had occurred in their villages or streets. About three-quarters (77%) reported to primary healthcare facilities and almost one-quarter (23%) reported to the district level (Figure 6). Despite the reported ability of CHWs to detect the presence of diseases in animals, they did not report suspected outbreaks in animals because they were not unaware of where to report them.

Figure 6.

Proportion of community health workers who ever reported health events to higher-level authorities by district.

Due to the inconsistency of awareness among CHWs regarding the onset date of detection of the index case for health events or suspected disease outbreaks, timeliness from detection to reporting was estimated as the number of days between the earliest date of detection of any case and the earliest reporting to health authorities (ie, the date when the CHW became aware of health event and the date the event was reported to health authorities). Overall, the median number of days between the onset of a suspected disease outbreak at the community level and reporting to a health facility or the district level was 10 (IQR 2-30) and 30 (IQR 5-30), respectively. The median number of days between detection and reporting was 6 (IQR 4-7) in Ulanga, 4 (IQR 3-6) in Kilosa, 3 (IRQ 1-5) in Ngorongoro, and 1 (IRQ 1-2) in Morogoro Urban.

Timeliness related to receiving a response from health authorities was estimated as the number of days between the date when event or suspected outbreak was reported by a CHW and the earliest date that actions were initiated to manage or control the event. Of the 53 CHWs who reported the most recent health event or suspected disease outbreak to a higher level, 14 (26%) received a response from a primary health facility and none received a response from the district level. The median number of days between submitting a report and receiving a response was 7 (IQR 2-30). The median number of days between submitting a report and receiving a response was 17.5 (IQR 5-30) in Ngorongoro, 7 (IQR 3-30) in Kilosa, 6.5 (IQR 6-7) in Ulanga, and 4.5 (IQR 2-7) in Morogoro Urban.

Report Submission

The mechanism by which CHWs submitted reports included walking on foot (100%), public transport (65%), telephone calls (56%), and mobile phone SMS (3%). The challenges encountered by CHWs related to report submission included an absence of reliable transport or the cost of transport (96%), lack of feedback or delayed response from higher levels (96%), long distances to case location (95%), inadequate training (93%), and poor cooperation from the community (56%). Other challenges included lack of awareness to whom they should report (37%) and being busy with other activities (31%) (Figure 7). The unspecific nature of clinical manifestations was also highlighted by a large proportion of CHWs (97%) as the main challenge when attempting to report on suspected health events at the community level. Moreover, no specific forms were provided for recording or reporting health events. Most often, reports were submitted from notebooks or loose papers or through verbal communication. All CHWs were unaware of who the responsible person was for data analysis and had never received a data analysis report from higher levels.

Figure 7.

Types of challenges reported by community health workers on recording and reporting of health events by district. Abbreviations: CHWs, community health workers; EBS, event-based surveillance.

Challenges related to the paper-based system reported by CHWs included delayed report submission (97%) and response (85%), system being laborious (73%), and high operational cost (40%). However, the paper-based system was reported to be easy to use (50%) and easy to keep the report for long periods (32%). All CHWs were of the opinion that the paper-based system used to collect and submit health event reports in their areas should be replaced with an electronic system to improve the efficiency of disease surveillance. The reported benefits of an electronic surveillance system included timely submission of reports (100%) and prompt response (82%). The challenges associated with an electronic system included being unsuitable for locations with a poor mobile phone network or internet connection (96%), the need for credit recharge to maintain its functionality (92%), unsuitability for locations without electric power (74%), and vulnerability to viruses (49%) and hackers (21%).

Availability and Strength of Connections

All villages (N = 85) had a reliable mobile phone network as verified by making active calls. An attempt to submit dummy data to a server using the AfyaData digital app was effective in 74 (87%) villages, suggesting that these villages had a mobile phone internet connection that was strong enough to support data transmission using digital technology. The number of villages with strong internet included 43 (88%) out of 49 in Kilosa, 16 (89%) out of 18 in Ulanga, 8 (73%) out of 11 in Ngorongoro, and 7 (100%) out of 7 in Morogoro Urban.

About 98% (n = 132) of the CHWs owned mobile phones, of which 119 owned feature phones and 13 owned smartphones, all with Android operating systems. Low ownership of smartphones among CHWs was observed in the rural districts of Ulanga (7 out of 33), Ngorongoro (2 out of 22), and Kilosa (4 out of 73) (Figure 8).

Figure 8.

Types of mobile phones owned by community health workers by district.

When they were asked which function in the mobile phone they would prefer most if mobile phone technology were introduced for event-based surveillance in their areas, the majority (88%) preferred SMS (text messages). The rest (12%) preferred both voice calls and SMS. SMS was preferred for confidentiality reasons and for the ability to keep a copy of the report, while use of both SMS and voice calls was reported as supporting active interaction between CHWs and the individual receiving the report.

Discussion

To provide guidance on strengthening community engagement in event-based surveillance, the current study provides baseline information related to the presence, recruitment terms of reference, and practices of CHWs in human and animal health surveillance functions Tanzania. We also established the strength of mobile phone networks and internet connections, and the ownership of mobile phones among CHWs to inform future strategies that use mobile phone solutions in community event-based surveillance.

We found that each study village had 2 CHWs, regardless of the size of the area and population they were serving. The number of CHWs that is proportional to the population they serve, however, has been associated with a higher rate of timely reporting of health events.⁴⁸ The large proportion of CHWs with a primary level of education in this study is not exceptional, as other programs elsewhere have relied on those with a minimal level of education to serve at the community level.^49,50 While attainment of secondary-level education has been positively associated with the performance of CHWs,⁵¹ this qualification remains debatable, especially in rural settings like those in Tanzania, where primary school is the highest level of education attained by the majority of the country's population.⁴⁶ With the exception of CHWs from Ulanga, CHWs from other study districts had not obtained training on the disease surveillance they were performing, which could negatively influence the quality of services they provide. Although the current study did not assess the performance of CHWs compared with their training levels, evidence shows that education and training (including in-service training) of CHWs improves their performance.^31,52,53

The CHWs from Ulanga received a salary, whereas those from Morogoro Urban, Kilosa, and Ngorongoro worked on a voluntary basis without pay. Whether CHWs should be engaged as volunteers or remunerated remains contentious and there is no evidence on the efficiency of volunteerism and whether it can be sustained for long periods.³² A recent study in Morogoro, Tanzania, reported that providing a form of financial incentive is critical for strengthening CHW motivation and sustaining health services programs.⁵⁴ Some form of remuneration would not only motivate CHWs but would also make them accountable and thereby enhance their contribution to community event-based surveillance and other community healthcare services.³¹

The CHWs in our study were selected by community members and were part of the community they were serving, suggesting that they had strong community ties, experience with the communities they served, and were therefore in a good position to directly observe and communicate community events. Their selection approach fits largely with World Health Organization guidelines that define CHWs as individuals selected by members of the community to work in the communities where they live.³² Similar consideration in the selection process of CHWs is reported in a systematic review by Scott et al.⁵⁵ In some countries, however, the selection criteria for CHWs have been established and all CHWs are recruited according to the same list of requirements, regardless of where they live and where they will serve. For example, in Mozambique, CHWs are a nationally recognized cadre of the health workforce, for which training, qualifications, and selection criteria have been standardized.⁵⁶ An attempt to standardize their selection criteria should be made in conjunction with the use of common training curricula.⁵⁷

In our study, most of the CHWs were not knowledgeable about their scope of work, roles, and responsibilities, which is likely to contribute to the inefficiency of community-based surveillance strategies.⁵⁸ Defining the CHW job description would enable program designers to specify the selection criteria, qualifications, and requirements that correspond with the work that CHWs undertake and would enhance monitoring of their performance.⁵⁹ Developing specific guidelines that define optimal training, roles and responsibilities, working terms and conditions, remuneration, supervision, monitoring, and evaluation has been reported to improve CHW performance.^60,61 The activities that CHWs were performing were largely not directly related to disease surveillance. By working on specific health programs, CHWs in the study areas were less likely to be in a position to detect and report all potential signals, events, and diseases at the community level. This observation is similar to those reported in other studies highlighting the limitations of community-based surveillance programs that focused on reporting single diseases.^26,62-65 The International Health Regulations framework for disease reporting broadened the scope of CHWs to focus on identifying any public health emergency rather than a few selected diseases.¹⁸ A review of CHW programs in 46 countries⁴³ has indicated a wide range of functions, including home visits, environmental sanitation, provision of water supply, first aid and treatment of simple and common ailments, health education, nutrition, surveillance, maternal and child health and family planning activities, communicable disease control, community development activities, referrals, record keeping, and collection of data on vital events.

Within the Global Health Security Agenda,⁶⁷ community event-based surveillance and the One Health approach are priority areas that together have been suggested to provide an effective mechanism for early detection and rapid response to health events including disease outbreaks. Community event-based surveillance using the One Health approach has been reported to enhance early detection of signals associated with zoonoses.^48,67,68 Because about three-quarters of emerging infectious diseases of humans have animal origins, we recommend that Tanzania should develop a One Health training curriculum for CHWs to equip them with the necessary skill sets and clearly defined roles to enhance their efficiency as the primary sensors for early detection of health events in animals, humans, and their environment.

Timeliness of reporting and response and interactions observed in this study suggest that current CHW practices and mechanisms of operations cannot support early warning of signals and events to promptly reach responsible authorities. Gaps identified including the absence of event and outbreak recording forms, use of paper-based reporting as the notification system, hand delivery of reports to higher levels, and lack of training and guidelines. These gaps should be addressed to improve the efficiency of community event-based surveillance.^20,24,48,69 The current study suggests that the majority of CHWs were likely to report to a primary health facility, a practice that offers opportunities for prompt response to health events. However, the observed overall median number of days between detection and reporting, and reporting and response from primary healthcare facilities, were within 2 and 30 days in all study districts. Although the events reported by CHWs were not characterized by their severity, the lack of timeliness observed does not provide an enabling environment for identifying and controlling outbreaks quickly to ensure community, regional, national and global health security.¹⁸

In light of the observed practices of CHWs related to detection, recording, and reporting of health events, the adoption of electronic reporting systems that support seamless data collection, repository, simple automated analysis, and real-time data access is likely to improve the timelines of reporting and response. The current study demonstrates a high level of ownership of mobile phones among CHWs and the presence of reliable phone networks and internet connections, both of which offer an opportunity for mobile-based solutions to improve efficiency of community-based disease surveillance systems. Previous studies have demonstrated the impact of electronic systems in improving the timeliness of outbreak detection, reporting, and response.^23,70,71 With the rise of information and communication technologies,^72,73 introduction of early warning applications and tools—powered with simple automated analysis outputs, visualization, and transmission of signals—is therefore recommended.^74,75

While interpreting the results of this study, it is worth noting important limitations. The study was not designed to be statistically representative of the population in the study areas as a whole, and therefore generalizing from the limited sample size used should be made with caution. The study site selection process using the criteria presented in this study resulted in unequal sampling of villages, streets, and participants. The study focused more on remote locations characterized with poor infrastructure and access to healthcare services. In this setting, the CHWs assume the frontline position to enhance detection and reporting of public health events and provision of basic health services and education. In this situation, an overall descriptive analysis was largely adopted and where attempt was made on comparison analysis based on distribution of parameters of interest across the study districts should be interpreted cautiously. The quality of interview-based data generated might have been affected by recall bias among the study participants. Triangulation of results, especially related to the timeliness of response by the level of authorities receiving data, would provide more informative data. A nationally representative CHW study would complement the findings of this study and increase transferability of the results.

Conclusion

This study offers an important baseline description of the presence of CHWs in the study districts and their practices related to detection, recording and reporting of health events, responses, and associated timelines. Timeliness of reporting and response and interactions observed in this study should be improved to enhance timely communication of early warning signals and events to responsible authorities. The presence of reliable mobile phone networks and internet connections in the study areas offers an opportunity for deployment of mobile phone-based solutions to improve community-based disease surveillance.

Acknowledgments

Very sincere gratitude and appreciation are extended to district executive directors, district medical officers, and community leaders in the respective study districts for their invaluable support during the whole period of this study. We wish to express our deepest gratitude to study participants whose understanding and cooperation made this study a success. This work was supported by Skoll Global Threats Fund (grant numbers 14-02688 and 17-41812).

References

World Health Organization (WHO). Disease burden and mortality estimates: cause-specific mortality, 2000-2016. Accessed October 19, 2020. https://www.who.int/healthinfo/global_burden_disease/estimates/en/

Braack

, Gouveia de Almeida

, Cornel

, Swanepoel

, de Jager

. Mosquito-borne arboviruses of African origin: review of key viruses and vectors. Parasit Vectors. 2018; 11(1):29.

Mboera

LEG

, Rumisha

, Mlacha

, Mayala

, Bwana

, Shayo

. Malaria surveillance and use of evidence in planning and decision making in Kilosa

District

, Tanzania. Tanzania J Health Res. 2017;19(3).

Mboera

LEG

, Rumisha

, Chiduo

, et al. Cause-specific mortality patterns among hospital deaths in Tanzania. PLoS One. 2018; 13(10):e0205833.

Karimuribo

, Mboera

LEG

, Mbugi

, et al. Are we prepared for emerging and re-emerging diseases? Experience and lessons from epidemics that occurred in Tanzania during the last five decades. Tanzania J Health Res. 2011; 13:387-398.

Mmbaga

, Mwasekaga

, Mmbuji

, et al. Results from the first 30 months of national sentinel surveillance for influenza in Tanzania, 2008–2010. J Infect Dis. 2012; 206(suppl 1):S80-S86.

Chipwaza

, Mugasa

, Selemani

, et al. Dengue and Chikungunya fever among viral diseases in outpatient febrile children in Kilosa District Hospital, Tanzania. PLoS Negl Trop Dis. 2014; 8(11):e3335.

Sindato

, Karimuribo

, Pfeiffer

, et al. Spatial and temporal pattern of Rift Valley fever outbreaks in Tanzania; 1930 to 2007. PLoS One. 2014; 9(2):e88897.

Mboera

LEG

, Mweya

, Rumisha

, et al. The risk of Dengue virus transmission in Dar es Salaam, Tanzania during an epidemic period of 2014. PLoS Negl Tro Dis. 2016; 10(1):e004313.

10.

Kajeguka

, Kaaya

, Mwakalinga

, et al. Prevalence of dengue and chikungunya virus infections in north-eastern Tanzania: a cross sectional study among participants presenting with malaria-like symptoms. BMC Infect Dis. 2016; 16:183.

11.

Kinimi

, Shayo

, Bisimwa

, et al. Evidence of chikungunya virus infection among febrile patients seeking healthcare in selected districts of Tanzania. Infect Ecol Epidemiol. 2018; 8(1):1553460.

12.

World Health Organization Regional Office for Africa (WHO AFRO). WHO AFRO outbreaks and other emergencies, week 22: 26 May - 1 June 2018. ReliefWeb website. Published June 1, 2018. Accessed October 24, 2020. https://reliefweb.int/report/world/who-afro-outbreaks-and-other-emergencies-week-22-26-may-1-june-2018-data-reported-1700

13.

World Health Organization. Coronavirus disease (COVID-19) Weekly Epidemiological Update and Weekly Operational Update. Accessed May 13, 2020. https://www.who.int/emergencies/diseases/novel-coronavirus-2019/situation-reports/

14.

Government of the United Republic of Tanzania. Guidelines for Surveillance of Prioritized Zoonotic Diseases for Human and Animal Health in the United Republic of Tanzania. Internal document. Dodoma, Tanzania: Government of the United Republic of Tanzania; 2018.

15.

World Health Organization (WHO). Protocol for Assessing National Surveillance and Response Capacities for the International Health Regulations (2005) in Accordance with Annex 1 of the IHR: A Guide for Assessment Teams. Geneva: WHO; 2010. Accessed March 26, 2018. http://www.who.int/ihr/publications/who_hse_ihr_201007_en.pdf?ua=1

16.

World Health Organization (WHO). Using Climate to Predict Infectious Disease Outbreaks: A Review. Geneva: WHO; 2004. Accessed March 26, 2018. http://www.who.int/globalchange/publications/en/oeh0401.pdf

17.

Thomson

, Connor

. The development of malaria early warning systems for Africa. Trends Parasitol. 2001; 17(9):438-445.

18.

World Health Organization (WHO). International Health Regulations (2005). 2nd ed. Geneva: WHO; 2008. Accessed May 22, 2016. http://www.who.int/ihr/publications/9789241596664/en/index.html

19.

World Organisation for Animal Health. One Health. Accessed April 18, 2019. http://www.oie.int/en/for-the-media/onehealth/

20.

Mtema

. Integrated Disease Surveillance and Response Systems in Resource-limited Settings. Glasgow: University of Glasgow; 2013. Accessed October 19, 2020. http://theses.gla.ac.uk/5224/

21.

Phalkey

, Yamamoto

, Awate

, Marx

. Challenges with the implementation of an Integrated Disease Surveillance and Response (IDSR) system: A systematic review of the lessons learned. Health Policy Plan. 2013; 30(1):131-143.

22.

Mandyata

, Olowski

, Mutale

. Challenges of implementing the integrated disease surveillance and response strategy in Zambia: a health worker perspective. BMC Public Health. 2017; 17(1):746.

23.

Jajosky

, Groseclose

. Evaluation of reporting timeliness of public health surveillance systems for infectious diseases. BMC Public Health. 2004; 4:29.

24.

World Health Organization (WHO). Early Detection, Assessment and Response to Acute Public Health Events: Implementation of Early Warning and Response with a Focus on Event-Based Surveillance. Interim Version. Geneva: WHO; 2014. Accessed October 24, 2020. https://apps.who.int/iris/bitstream/handle/10665/112667/WHO_HSE_GCR_LYO_2014.4_eng.pdf

25.

Toyama

, Ota

, Beyene

. Event-based surveillance in north-western Ethiopia: experience and lessons learnt in the field. Western Pacific Surv Resp J. 2015; 6(3):22-27.

26.

Curry

, Bisrat

, Coates

, Altman

. Reaching beyond the health post: community-based surveillance for polio eradication. Dev Pract. 2013; 23:69-78.

27.

Smolinski

, Crawley

, Olsen

. Finding outbreaks faster. Health Sec. 2017; 15(2):215-220.

28.

Jones

, Patel

, Levy

, et al. Global trends in emerging infectious diseases. Nature. 2008; 451(7181):990-993.

29.

Thacker

, Berkelman

. Public health surveillance in the United States. Epidemiol Rev. 1988; 10:164-190.

30.

Rafiq

, Wheatle

, Mushi

, Baynes

. Who are CHWs? An ethnographic study of the multiple identities of community health workers in three rural Districts in Tanzania. BMC Health Serv Res. 2019; 19:712.

31.

Bhutta

, Lassi

, Pariyo

, Huicho

. Global Experience of Community Health Workers for Delivery of Health Related Millennium Development Goals: A Systematic Review, Country Case Studies, and Recommendations for Integration into National Health Systems. Geneva: World Health Organization; 2010. Accessed October 19, 2020. https://www.who.int/workforcealliance/knowledge/resources/chwreport/en/

32.

Lehmann

, Sanders

. Community Health Workers: What Do We Know About Them? The State of the Evidence on Programmes, Activities, Costs an Impact on Health Outcomes of Using Community Health Workers. Geneva: World Health Organization; 2007. Accessed May 24, 2019. https://www.who.int/hrh/documents/community_health_workers.pdf

33.

Lewin

, Dick

, Pond

, et al. Lay health workers in primary and community healthcare. Cochrane Database Syst Rev. 2005;(1):CD004015.

34.

World Health Organization (WHO), WHO Study Group on Community Health Workers. Strengthening the Performance of Community Health Workers in Primary Health Care: Report of a WHO Study Group. Geneva: WHO; 1989. Accessed October 22, 2020. https://apps.who.int/iris/handle/10665/39568

35.

Christopher

, Le May

, Lewin

, Ross

. Thirty years after Alma-Ata: a systematic review of the impact of community health workers delivering curative interventions against malaria, pneumonia and diarrhoea on child mortality and morbidity in sub-Saharan Africa. Hum Res Health. 2011; 9:27.

36.

Johnson

, Noyes

, Haines

, et al. Learning from the Brazilian community health worker model in North Wales. Glob Health. 2013; 9:25.

37.

Jaskiewicz

, Tulenko

. Increasing community health worker productivity and effectiveness: a review of the influence of the work environment. Hum Resour Health. 2012; 10(1):38

38.

Shoo

, Mzige

. A Review of Community Health Worker Cadres' Training Programs in Tanzania. Dar es Salaam, Tanzania: Ministry of Health and Social Welfare and UNICEF; 2011.

39.

Allport

, Mosha

, Bahari

, Swai

, Catley

. The use of community-based animal health workers to strengthen disease surveillance systems in Tanzania. Rev Sci Tech. 2005; 24:(3):921-932.

40.

Gopinathan

, Lewin

, Glenton

. Implementing large-scale programmes to optimize the health workforce in low- and middle-income settings: a multicounty case study synthesis. Trop Med Int Health. 2014; 19(12):1437-1456.

41.

Mangham-Jefferies

, Pitt

, Cousens

, Mills

, Schellenberg

. Cost-effectiveness of strategies to improve the utilization and provision of maternal and newborn health care in low-income and lower-middle-income countries: a systematic review. BMC Pregnancy Childbirth. 2014; 14:243.

42.

Geldsetzer

, Mboggo

, Larson

, et al. Community health workers to improve uptake of maternal healthcare services: A cluster-randomized pragmatic trial in Dar es Salaam, Tanzania. PLoS Med. 2019; 16(3):e1002768.

43.

Tani

, Stone

, Exavery

, et al. A time-use study of community health worker service activities in three rural districts of Tanzania (Rufiji, Ulanga and Kilombero). BMC Health Serv Res. 2016; 16(1):461.

44.

Manzi

, Schellenberg

, Hutton

, et al. Human resources for health care delivery in Tanzania: a multifaceted problem. Hum Resour Health. 2012; 10:3.

45.

Boyce

, Katz

. Community health workers and pandemic preparedness: current and prospective roles. Front Public Health. 2019; 7:62.

46.

National Bureau of Statistics (Tanzania), Office of Chief Government Statistician (Zanzibar). 2012 Population and Housing Census: Population Distribution by Administrative Areas. Dar es Salaam: National Bureau of Statistics; 2013. Accessed October 19, 2020. https://www.nbs.go.tz/index.php/en/census-surveys/population-and-housing-census/162-2012-phc-population-distribution-by-administrative-areas

47.

Ramsey

, Hingora

, Kante

, et al. The Tanzania Connect Project: a cluster randomized trial of the child survival impact of adding paid community health workers to an existing facility-focused health system. BMC Health Serv Res. 2013,13(suppl 2):S6.

48.

Clara

, Dao

ATP

, Do

, et al. Factors influencing community event-based surveillance: lessons learned from pilot implementation in Vietnam health security. Health Secur. 2018; 16(suppl 1):S66-S75.

49.

Chopra

, Munro

, Lavis

, Vist

, Bennett

. Effects of policy options for human resources for health: an analysis of systematic reviews. Lancet. 2008; 371(9613):668-674.

50.

Lewin

, Munabi-Babigumira

, Glenton

, et al. Lay health workers in primary and community health care for maternal and child health and the management of infectious diseases. Cochrane Database Syst Rev. 2010; 2010(3):CD004015.

51.

Wanduru

, Tetui

, Tuhebwe

, et al. The performance of community health workers in the management of multiple childhood infectious diseases in Lira, northern Uganda – a mixed methods cross-sectional study. Glob Health Action. 2016; 9:33194.

52.

Vouking

, Takougang

, Mbam

, Mbuagbaw

, Tadenfok

, Tamo

. The contribution of community health workers to the control of Buruli ulcer in the Ngoantet area, Cameroon. Pan Afr Med J. 2013; 16:63.

53.

Raphael

, Rueda

, Lion

, Giordano

. The role of lay health workers in pediatric chronic disease: a systematic review. Acad Pediatr. 2013; 13(5):408-420.

54.

Ormel

, Kok

, Kane

, et al. Salaried and voluntary community health workers: exploring how incentives and expectation gaps influence motivation. Hum Resour Health. 2019; 17:59.

55.

Scott

, Beckham

, Gross

, Pariyo

, Rao

, Perry

. 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.

56.

The United States Global Health Initiative (GHI). Mozambique Strategy 2011–2015. Washington, DC: GHI; 2011. Accessed December 16, 2018. https://photos.state.gov/libraries/mozambique/328671/pepfar/mozambique-ghi-strategy.pdf

57.

George

, Young

, Nefdt

, et al. Community health workers providing government community case management for child survival in sub-Saharan Africa: who are they and what are they expected to do?. Am J Trop Med Hyg. 2012; 87(suppl 5):85-91.

58.

Alam

, Tasneem

, Oliveras

. Performance of female volunteer community health workers in Dhaka urban slums. Soc Sci Med. 2012; 75(3):511-515.

59.

Tulenko

, Møgedal

, Afzal

, et al. Community health workers for universal health-care coverage: from fragmentation to synergy. Bull World Health Organ. 2013; 91(11):847-852.

60.

Campbell

, Buchan

, Cometto

, et al. Human resources for health and universal health coverage: fostering equity and effective coverage. Bull World Health Organ. 2013; 91(11):853-863.

61.

Dahn

, Woldemariam

, Perry

, et al. Strengthening Primary Health Care Through Community Health Workers: Investment Case and Financing Recommendation. Baltimore, MD: Initiatives, Inc.; 2015. Accessed May 30, 2019. https://chwcentral.org/resources/investment-case/

62.

Crowe

, Hertz

, Maenner

, et al. A plan for community event-based surveillance to reduce Ebola transmission—Sierra Leone, 2014-2015. MMRW Morb Mortal Wkly Rep. 2015; 64(3):70-73.

63.

Morse

. Public health surveillance and infectious disease detection. Biosecur Bioterror. 2012; 10(1):6-16.

64.

Cairncross

, Braide

, Bugri

. Community participation in the eradication of guinea worm disease. Acta Trop. 1996; 61(2):121-136.

65.

Ofosu-Amaah

, World Health Organization (WHO). National Experience in the Use of Community Health Workers. A Review of Current Issues and Problems. Geneva: WHO; 1983. Accessed October 24, 2020. https://apps.who.int/iris/handle/10665/39257

66.

Global Health Security Agenda website. Accessed May 29, 2019. https://www.ghsagenda.org

67.

Trang do

, Siembieda

, Huong

, et al. Prioritization of zoonotic diseases of public health significance in Vietnam. J Infect Dev Ctries. 2015; 9(12):1315-1322.

68.

Prasad

, Muraleedharan

. Community Health Workers: A Review of Concepts, Practice and Policy Concerns. London: International Consortium for Research on Equitable Health Systems, London School of Hygiene & Tropical Medicine; 2007. Accessed October 24, 2020. https://www.hrhresourcecenter.org/hosted_docs/CHW_Prasad_Muraleedharan.pdf

69.

Ward

, Brandsema

, van Straten

, Bosman

. Electronic reporting improves timeliness and completeness of infectious disease notification, The Netherlands, 2003. Euro Surveill. 2005; 10(1):27-30.

70.

Mondor

, Brownstein

, Chan

, et al. Timeliness of nongovernmental versus governmental global outbreak communications. Emerg Infect Dis. 2012; 18(7):1184-1187.

71.

Kostkova

A roadmap to integrated digital public health surveillance: the vision and the challenges. In: Schwabe D, Almeida V, Glaser, eds. WWW’13 Companion: Proceedings of the 22nd International Conference on World Wide Web. New York: Association for Computing Machinery; 2013:687-694. Accessed October 24, 2020. https://doi.org/10.1145/2487788.2488024

72.

Kostkova

, Fowler

, Wiseman

, Weinberg

. Major infection events over 5 years: how is media coverage influencing online information needs of health care professionals and the public?. J Med Internet Res. 2013; 15(7):e107.

73.

Widdowson

, Bosman

, van Straten

, et al. Automated, laboratory-based system using the internet for disease outbreak detection, the Netherlands. Emerg Infect Dis. 2003; 9(9):1046-1052.

74.

Chan

, Brewer

, Madoff

, et al. Global capacity for emerging infectious disease detection. Proc Natl Acad Sci USA. 2010; 107(50):21701-21706.

75.

Freifeld

, Chunara

, Mekaru

, et al. Participatory epidemiology: use of mobile phones for community-based health reporting. PLoS Med. 2010; 7(12):e1000376.