The 2018-19 influenza season in Afghanistan: Epidemiology and Virology

Objective This paper aims to provide information on epidemiology and virology of seasonal influenza for the 2018-19 season, and build upon limited information for developing influenza vaccination policy for risk groups in Afghanistan. This paper is based on a retrospective analysis of Afghanistan influenza surveillance data, both from influenza-like illness (ILI) and severe acute respiratory infection (SARI) cases during the 2018-19 season. The data for 2018-19 has also been compared to the data of the previous two seasons (2016-2017 and 2017-2018).


Introduction
Acute Respiratory Infections (ARIs) are the leading cause of death among under-five children in Afghanistan and accounts for 23% of deaths among this age group [1]. An estimated 750,000 children aged less than five are currently suffering from ARIs annually in Afghanistan [2]. Likewise, national disease surveillance and response (NDSR) data shows that pneumonia had the highest number of deaths (1,924 deaths) compared to other notifiable diseases in 2018 in Afghanistan [3].
Influenza viruses (seasonal and avian) cause acute respiratory infections, ranging from mild to severe cases and even to death, in all parts of the world. The global annual attack rate of influenza is estimated to be 5-10% among adults and 20-30% among children.
Worldwide, these annual epidemics are estimated to result in about 3 to 5 million cases of severe illness [4]. However, the new estimates show that influenza-related respiratory deaths are likely higher and estimated to be 291,243-645,832 (4.0-8.8 per 100,000 individuals) [5].
Epidemics of influenza are causing severe illness and death, especially among children, adults, and people with underlying health conditions. Seasonal influenza is caused by influenza A and influenza B type viruses, which circulate around the world and continues to evolve into new strains [6]. Annual vaccination is recommended at the beginning of each season to groups at risk of severe complications and considered the most effective measure for prevention against seasonal influenza and its complications [7].  [8].
According to the World Organization for Animal Health, Afghanistan is also a high-risk country for avian influenza as it lies along the migratory pathways of wild birds. The widespread practice of raising poultry in the home to supplement income, low community awareness and a health system that is in the early stages of delivering effective health services make Afghanistan prone to an outbreak of spill-over events of avian influenza [9].
From 2007, the virological surveillance was existing with no epidemiological surveillance.
At the beginning of 2014, epidemiological surveillance was added, and lab capacity strengthened through the World Health Organization (WHO) Pandemic Influenza Preparedness (PIP) project. As a result, epidemiological surveillance was instigated in nine hospitals and virological surveillance was revived and integrated. The Central Public Health Laboratory (CPHL) located in the capital city of Kabul, has been recognized as a national influenza center (NIC) since 2009 [10].
Although the epidemiology and impact of influenza are well-defined in developed countries, data, and published papers in developing countries are still limited [11]. Due to the lack of appropriate data on the epidemiology and impact of influenza, influenza control programs, including vaccination, are not a priority in most developing countries [12]. Sufficient influenza epidemiological data is required to guide national vaccine policymakers in developing policy, particularly in low-and middle-income countries with limited resources for healthcare. The published literature about seasonal influenza is missing for the previous years in Afghanistan. Also, there is no influenza vaccination policy in Afghanistan and vaccination is not even provided to high-risk groups including the older population, pregnant women, healthcare providers and people with comorbidities. This paper aims to provide information on epidemiology and virology of influenza for the 2018-19 season and contribute to advocate for developing influenza vaccine policy in Afghanistan.

Methods
This paper is based on a retrospective analysis of Afghanistan influenza surveillance data, both from influenza-like illness (ILI) and severe acute respiratory infection (SARI) cases during the 2018-19 season, compared to the data of the previous two seasons as both epidemiological and laboratory quality data are available since 2016.
A total of 10,604 ILI and 5,261 SARI cases were reported to the influenza sentinel There are nine surveillance sentinel sites (all are both ILI and SARI ), operating in nine provinces of Afghanistan. Among the sentinel sites, five are regional hospitals, three provincial hospitals and one national hospital (Fig. 1). The criteria for selecting these sentinel sites was representativeness by different geographic region and high population density, which could represent the whole population. The nine SARI and ILI sentinel sites submit specimens and epidemiological data weekly. Data is being entered in EMFLU (regional platform for sharing of epidemiological and virologic data on influenza). Since 2014, all sentinel sites collect throat and nasal swabs from two ILI cases and three SARI cases per week per site selected systematically and send them under the cold chain condition to NIC. The sample collections are increasing by twofold during the influenza season (October to March) (four ILI samples and six SARI cases are being collected per site and per week). Besides, during any suspected outbreak or cluster of ILI or SARI, the surveillance staff collect epidemiologic data and specimens from all cases immediately and share them with surveillance department and the NIC for timely confirmation.
a. Case definition Influenza surveillance in Afghanistan uses the WHO case definition of influenza-like illness (ILI), which is "an acute respiratory infection with a measured fever of ≥ 38 °C, and cough, with onset within the past ten days". Similarly, severe acute respiratory infection (SARI) is defined as: "an acute respiratory infection with a history of fever or measured fever of ≥ 38 °C, and cough, with onset within the past ten days, and requiring hospitalization" [13].

b. Laboratory diagnosis
All of the laboratory tests for influenza viruses were conducted by CPHL, which is a WHOdesignated NIC. The type/subtype of the influenza virus was identified by real-time c. Statistical data analysis Data were summarized in the form of proportions and frequency tables for all categorical variables. Proportions of ILI and SARI were calculated as the number of ILI cases per 1,000 outpatient consultations and SARI cases per total inpatient admissions. The variables included in the study were ILI and SARI cases, positive influenza and sub-types, demographic characteristics (e.g., sex, age), hospitalization and outcome. Data from EMFLU was exported to SPSS software version 20 for statistical analysis. The epidemic threshold for the proportion of ILI and SARI cases during recent years was defined using the WHO Average curve Shiny app with the two seasons of historical data [14]. As the current WHO Average curve Shiny app was not able to demonstrate the best estimation of the beginning and end week of the season, the Moving Epidemic Method (MEM) 2.15 was used for determining the beginning and end week of the season.
Almost half (44%) of SARI reported cases were among under-two children. A large proportion of SARI associated positive influenza cases aged less than five years old (46%) with 32% aged less than two years old and 14% aged 2-5 years. It was followed by 11% among 5-15 years old, 13% among 15-50 years old, 12% among 50-65 years old and 18% among 65 + years old patients ( Table 2). The influenza A (H1N1) pdm09 virus was more common among under-two children with 23 positive viruses, and 65 + years old groups with 12 positive viruses (Fig. 8).
The SARI-associated influenza cases were slightly higher among males (53%) than females (47%) ( Table 2). For both SARI and ILI, the influenza-positive cases higher for males (59% for ILI and 53% for SARI) than females (41% for ILI and 47% for SARI) which can be due to males either being exposed to the outside environment or seeking health care services more than females in Afghanistan.
Among the total ILI positive cases, more than half of influenza cases were caused by A(H1N1)pdm09 virus (57%). This was followed by influenza A(H3N2) (22%) and influenza B   Figure 1 Influenza surveillance sentinel sites in Afghanistan. Figure 1 shows the nine SARI/ILI sentinel sites geographical distribution in Afghanistan.  Temporal distribution of the number and proportion of influenza viruses from ILI specimens, by type and week. Figure 3 shows the number of influenza viruses (vertical axis) type from ILI specimens during seasonal weeks (horizontal axis) in which the A(H1N1)pdm09 is the predominant virus during the seasonal weeks.

Figures
The red line chart shows proportion of influenza positive on the right vertical axis.

Figure 4
Distribution of Influenza virus subtype by age group from ILI positive cases, week 48 2018-week 14 2019. Figure 4 shows the distribution of influenza virus by age group. The vertical axis shows six categories of age groups and the vertical axis shows the number of influenza viruses by type.     Figure 8 shows the number of viruses from SARI specimen by age group.
The horizontal axis is the six categories of age group and the vertical axis the number of viruses.