Early warning and response systems for respiratory disease outbreaks: lessons learnt from cluster-associated cases of acute respiratory illnesses in Gilgil subcounty, Nakuru County, Kenya, 2021

Abstract

Investigating acute respiratory illnesses (ARIs) is difficult due to non-specific symptoms, varied health-seeking behaviors, and resource limitations; yet early detection is critical to global health security. Kenya's Ministry of Health (MOH) uses the Integrated Disease Surveillance strategy for public health surveillance, incorporating event-based surveillance (EBS) and indicator-based surveillance (IBS) for early warning system. MOH, supported by the US-CDC, established Influenza Sentinel Surveillance (ISS) in 2006 and later launched community EBS (CEBS) and health facility EBS (HEBS) pilots to enhance surveillance for COVID-19. On March 2, 2021, the CEBS system detected a signal of “Two or more people presenting with similar signs and symptoms in a community within a week” in a county. Investigations launched on March 4, 2021, investigations revealed unreported ARI cases which had been missed by both the ISS and IBS. A total of 176 ARI cases were line-listed with 91/176 (51.7%) aged <5-years and 46/176 (26.1%) hospitalized. RT-PCR tests confirmed 34/79 (43.0%) SARS-CoV-2 and 1/7 (14.3%) A/H3N2 cases. Of the CEBS, HEBS, IBS, and ISS systems deployed by the county to strengthen the early warning for respiratory diseases, CEBS detected a signal of unreported ARIs that facilitated further investigations and response.

SUMMARY BOX.

• Strengthening early warning and response systems to promptly detect and rapidly respond to respiratory events is a key part of global health security strategies.

• In March 2021, community event-based surveillance (CEBS) detected a cluster of acute respiratory illnesses (ARIs) in Nakuru County, Kenya.

• Detection of ARI through CEBS facilitated rapid investigation and control by public health authorities.

• We highlight the contributions of CEBS in ARIs cluster detection, as well as challenges to public health investigation and response during the COVID-19 pandemic.

Introduction

Clusters of acute respiratory illnesses (ARIs) can be challenging to investigate due to non-specific clinical presentations, variability in health-seeking behaviours and resource limitations.^{1 2} As highlighted by the COVID-19 pandemic, effective strategies for early detection and control of respiratory disease outbreaks are critical for global health security.³ Event-based surveillance (EBS), defined as the ‘organised collection, assessment and interpretation of mainly unstructured ad hoc information regarding health events’, can enable early detection and real-time reporting of public health risks.^{4 5} EBS complements indicator (disease)-based (IBS) and sentinel surveillance systems by strengthening early detection of outbreaks and emerging threats from multiple sources including community, healthcare facilities and media platforms.

Respiratory disease surveillance in Kenya

The Ministry of Health, Kenya (MOH) implements public health surveillance using the integrated disease surveillance and response (IDSR) strategy. IDSR detects priority diseases and conditions using standard case definitions in IBS, and public health events using signals in EBS.⁶ A signal is a set of information used in early warning and response (EWAR) systems to indicate a potential public health risk.⁵ EWAR uses both IBS and EBS data for signal detection, verification and risk analysis to facilitate rapid response. Signals, reported by community health promoter (CHPs) and healthcare workers (HCWs), are considered events when verified true. Verification considers the relevance (accuracy, active and new) and authenticity of the signal information. Events are then assessed for risks including investigation and laboratory confirmation (linkage to IBS) to guide public health response. Reports of signals, obtained from multiple sources, are not dependent on laboratory/technical evaluation hence often precede confirmation of diseases/outbreaks enabling EBS to improve the sensitivity and timeliness of the IBS system.

The MOH with the support of US Centers for Disease Control and Prevention (CDC) established a national Influenza Sentinel Surveillance (ISS) system in 2006.⁷ The ISS involved eight hospital-based sites including Nakuru County Referral Hospital (NCRH) and focused on surveillance for severe acute respiratory infection (SARI) defined as a history of fever or measured fever ≥38°C and cough, with onset within the last 10 days, and requires hospitalisation. Later, MOH supported by CDC launched a community EBS (CEBS) pilot in two counties including Nakuru in September 2019, and health facility EBS (HEBS) pilot in three counties including Nakuru in August 2020.^{8 9} CEBS is detection and reporting of events within a community-by-community members, for example, ‘unexpected large numbers of animal deaths’ while HEBS tracks events at health facilities by the HCWs, ‘severe illness requiring hospital admission in healthcare workers after caring for patients with similar symptoms’.⁵ With the onset of the COVID-19 pandemic, the MOH developed COVID-19 working case definitions and signals to enhance surveillance. Respiratory samples for SARS-CoV-2 were collected and tested by real-time reverse transcriptase PCR (RT-PCR) in identified laboratories nationwide.

A cluster of respiratory illnesses in Nakuru County: February–March 2021

On 2 March 2021, a CHP from Gilgil subcounty, Nakuru County, received a report from a community member whose two children had reported fever, cough and difficulty breathing for 2 days. The member confided in the CHP that her companion tested positive for SARS-CoV-2 approximately 3 weeks earlier but did not list her as a contact with public health authorities. The CHP identified this incident as CEBS signal ‘two or more people presenting with similar signs and symptoms in a community within a week’ and reported to the assigned community health assistant (CHA) who triaged and verified the signal as an event and reported to the subcounty team.

Field investigation methods

The subcounty team initiated preliminary investigations on 4 March 2021, which established that many ARI cases had been seen at the local health facility from mid-February 2021 but were not reported. ARIs are notifiable within 24 hours under IBS. The Nakuru County health team was notified who then collected nasopharyngeal/oropharyngeal swabs from active ARI cases and submitted them to CDC-supported MOH laboratory for SARS-CoV-2 testing by RT-PCR. The county health team alerted the national surveillance office which prompted a field investigation. A team from Kenya Field Epidemiology and Laboratory Training Program, MOH, Washington State University and CDC investigated the reported ARI cases by reviewing Gilgil health facilities records on 13 April 2021–19 April 2021. A case of ARI was defined as acute onset of respiratory symptoms and fever in any person in Gilgil from 1 February 2021 to 31 March 2021. Qualitative interviews and reviews of EBS and ISS data were also done to assess SARI trends.

Clinical, laboratory and surveillance findings

The investigation identified 176 ARI cases in Gilgil with index case on 14 February 2021 and a peak on 1 March 2021 (figure 1). Up to 91/176 (51.7%) cases were <5 years old; 46/176 (26.1%) were hospitalised due to clinical pneumonia and no deaths. Cases were spread over 24 villages within two administrative wards of Gilgil. SARS-CoV-2 was detected in 34/79 (43.0%) patients and A/H3N2 influenza in 1/7 (14.3%) patients through RT-PCR. Contacts (n=188) were traced and placed on home-based isolation and care. ISS data from the NCRH showed a peak in SARI cases during late March 2021 (figure 2). Of 132 SARI cases reported from 14 February–31 March, 7 (5.3%) were among residents of Gilgil; the remainder (94.7%) were from other subcounties of Nakuru County.

Figure 1. Epidemic curve of cluster-associated cases+, by SARS-CoV-2 test result and hospitalisation status and timeline of community event-based surveillance (CEBS) reporting—Nakuru County, Kenya, 2021.

Figure 2. Severe acute respiratory infection (SARI) cases at Nakuru County Referral Hospital* and cluster associated ARI cases in Gilgil subcounty٭٭—Nakuru County, Kenya, 2021. *SARI data were routine reports to the national Influenza Sentinel Surveillance system. Nakuru County SARI cases include those from Gilgil Sub-county (location of cluster) which are also visualized separately to depict increase in SARI cases around the outbreak period. ٭٭Cluster associated ARI cases data were data collected through health facilities records review during the outbreak investigation.

Lessons learnt

Nakuru County uniquely implements CEBS, HEBS, IBS and ISS complementary surveillance systems providing early warning for respiratory disease outbreaks. Through CEBS, CHPs detected an event that was not picked up by the other systems. Subsequent response by the subcounty COVID-19 response included active surveillance, HBIC of ARI cases/contacts and risk communication regarding COVID-19 prevention. Although limited laboratory testing isolated both SARS-COV-2 and A/H3N2 and had no definitive aetiology, mitigation measures effective across respiratory pathogens, such as social/physical distancing, hand hygiene, contact tracing, isolation and symptom monitoring, facilitated respiratory disease control. Further, the detection of the ARI cluster in Gilgil subcounty, Nakuru County coincided with the third wave of COVID-19 transmission in Kenya which was mainly concentrated within Nairobi, Kajiado, Machakos, Kiambu and Nakuru counties.^{9 10} Of the five counties, Nakuru that had been onboarded onto the EBS (CEBS and HEBS) system, reported an ARI cluster signal. This further underscores the utility of EBS in strengthening the sensitivity of the public health surveillance system.

NCRH ISS site is approximately 40 km from Gilgil. As a result, very few patients from Gilgil were captured in ISS, thus, the trend of SARI cases from ISS was not commensurate with that from the cluster. The purpose of the ISS is not to identify clusters for public health response; rather, its important objectives are to identify circulating influenza including those with pandemic potential and characterise their epidemiology. EBS can complement IBS and ISS by directly engaging communities and primary health facilities. Although a HEBS pilot was ongoing in Nakuru County, it had not been deployed in Gilgil at the time of this outbreak. Notably, CEBS detected cluster-associated ARI cases depicted through records review to have started much earlier, depicting a gap in the use of IBS data in detecting acute public threats, thus the need for HEBS (figure 1). HEBS expansion and personnel training in primary health facilities have been planned to fill this gap. Routine and systematic communication between the CEBS, HEBS, ISS and IBS systems should be initiated to allow for comparison of facility, regional/national trends in respiratory disease incidence, including the detection of both localised and widespread events.

CEBS system challenges highlighted by this outbreak included insufficient workforce capacity to investigate signals. Qualitative interviews with subcounty public health personnel showed gaps in leveraging CEBS to enhance surveillance during the COVID-19 pandemic. HCWs had also not identified any clustering of ARI cases. Although a line list of cluster-associated ARI cases was compiled and updated daily by the Gilgil subcounty Disease Surveillance Coordinator, the data were not analysed before deployment of the field investigation team in April, over a month after event detection. Additionally, although the MOH was notified of the occurrence of this cluster, the event was not shared with neighbouring regions to facilitate active surveillance and implementation of prevention measures across a larger geographical area. Strategies to make EBS data available in real-time through an interactive dashboard and to make data readily available to emergency operations centres are in progress.

Challenges with specimen collection kits and shipment were identified, with less than half of cases (45%) having specimens submitted for testing. Further, multipathogen testing was not performed, limiting the ability to detect other respiratory pathogens; it is possible that multiple concurrent aetiologies could have contributed to the occurrence of the cluster, given the large proportion of young children.^{11 12} Moreover, specimens from at least 20 additional patients submitted for testing had no results due to failure to link patient identifiers with laboratory results. Strengthening laboratory systems is critical for public health EWAR systems.^{13 14} Recommendations to improve Kenya’s EBS system include integration into the routine health systems for sustainability; strengthening laboratory systems including multipathogen testing and links with surveillance systems and data utilisation.

Conclusions

This case study highlights the ability of CEBS to leverage community networks to facilitate EWAR to potential public health threats. CEBS, HEBS and ISS can complement IBS by strengthening EWAR for unusual and emerging disease events given adequate laboratory support. Detection of signals and events is just the first step in the cascade of public health response and must be followed with appropriate epidemiological/laboratory investigation, and mitigation measures which require systematic investment in resources and training.

The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the US Centers for Disease Control and Prevention/the Agency for Toxic Substances and Disease Registry.

Data availability statement

Data are available on reasonable request.