1. Introduction
Wild poliovirus (WPV) infection leads to paralysis (poliomyelitis) in the at-risk population of children under 5 years of age, significantly impacting the lives of the affected children and their families [
1]. Children affected by polio are most often from low- to lower-middle-income families living in areas with poor water and sanitation systems and weak health infrastructure and are therefore often un- or under-immunized with routine immunization antigens including polio. According to the World Health Organization (WHO), polio cases have decreased by over 99% since the launch of the Global Polio Eradication Initiative (GPEI) in 1988, from an estimated 350,000 cases in more than 125 endemic countries to just 12 cases in 2023 in the two remaining endemic countries of Afghanistan and Pakistan. To eliminate polio from its territory, the government of Pakistan declared polio a national emergency in 1994 and currently implements the program managed by Emergency Operation Centers (EOCs) at the national level and in each of its provinces [
2]. The government of Pakistan spearheads the EOC network. It acts as a coordination function for GPEI partners WHO, Rotary International, the US Centers for Disease Control and Prevention (CDC), UNICEF, the Bill and Melinda Gates Foundation, and Gavi, the Vaccine Alliance (GAVI) [
3].
Main polio program elements in Pakistan include routine immunization (RI), nationwide vaccination campaigns with bivalent oral polio vaccine (OPV) and inactivated polio vaccine (IPV), surveillance to detect signs of acute flaccid paralysis (AFP) in children and detect any poliovirus in the environment, and communications and community engagement to ensure a high level of vaccine acceptance. Routine immunization (RI) against six diseases including poliovirus was commenced in 1978 with the introduction of the National Expanded Programme on Immunizations (EPI) in Pakistan [
4]. RI services are given at fixed locations or immunization centers and with minimal outreach by the vaccinators. Besides four doses of the routine OPV within RI services [
5], Supplementary Immunization Activities (SIAs) including national and sub-national immunization activities (NIDs and SNIDs) are the main strategy to enhance OPV coverage. The Pakistan National Polio program organizes these SIAs, which last from 5–7 days and in which all households with children under five years of age are visited in 5 days and households where children were not available or refused during the 5 polio campaign days are revisited in the remaining 2 days [
6]. During NIDs, mass immunization campaigns are conducted across the country whereas, during SNIDs, campaigns are organized in selected high-priority areas or districts [
7]. Also, to ensure vaccination of all travelers against poliovirus, the Pakistan Polio Eradication Programme has set up a Permanent Transit Point for Polio (PTP) at the Pak–Afghan border [
8].
Since only one in 200 infections leads to paralysis, the Pakistan polio program takes samples using various methods described elsewhere from sewage sites to isolate any virus that may be circulating in the population [
9]. This is to give the program time to conduct targeted vaccination campaigns with high enough coverage levels before transmission levels become high enough to result in paralysis.
In September 2023, wild poliovirus type 1 (WPV1) was isolated from four environmental surveillance (ES) samples collected from different sites in Karachi, Pakistan. All four isolates of WPV1 were linked to the same ES sample collected in Jalalabad, Nangarhar, Afghanistan collected in November 2022: ES-AFG-Nangarhar-Jalalabad-Nov2022. Among these four samples, one was collected from the Haji Mureed Goth ES site in Liaqatabad town, district Central Karachi on 7th September 2023 whose WPV1 isolate was a 98.78% match to the Jalalabad sample. The second sample was collected from the Hirjat Colony PIDC Nalla ES site in Saddar town, district South Karachi on 13th September 2023 whose WPV1 isolate was a 98.56% match with the Jalalabad sample. The third sample was collected from the Machhar Colony ES site in Gadap town, district East Karachi on 12th September 2023; whose WPV1 isolate was a 98.78% match with the Jalalabad sample. The fourth sample was collected from the Sohrab Goth ES site in Gadap town, district East Karachi on 12th September 2023 whose WPV1 isolate was a 98.89% match with the Jalalabad sample [
10]. The investigations into the genetic relatedness of poliovirus isolates are conducted in a regional reference laboratory for poliomyelitis at the National Institute for Health (NIH), Islamabad, Pakistan [
11]. The regional reference laboratory for polio follows WHO-recommended procedures for detecting and characterizing polioviruses from the stool samples of AFP cases and sewage water samples collected from the environment [
12]. The reports of WPV1 detection from any sample are shared internally within the polio program network of Pakistan [
10].
Following the notification of these four positive samples, a team from the Pakistan Polio National Emergency Operations Center (NEOC) joined the Sindh EOC in Karachi to determine the most likely reason why the September samples in districts Central, South, and East were all linked to the same virus. An initial investigation was conducted in the drainage Union Councils (UCs) and adjoining UCs of ES sites, which were positive for WPV1 isolates. The team investigated the hypothesis that the ES sites share the same drainage UCs (i.e., sewage from the same/overlapping population is draining into different sites). Still, no evidence of overlapping of ES UCs/sewage channels was noted. They also investigated the possibility of composite sampling where either the individuals collecting the samples used the same sample to represent different sites or used the same collection materials to collect various samples. Still, no evidence of composite sampling was found. The final hypothesis investigated by the team was that the population from Nangarhar infected with WPV1 settled in UCs in different districts that drain to different ES sites. This arrived as the most likely scenario based on the continued influx of Afghans to Karachi and the events leading up to the September sample collections due to gatherings within the Afghan community and the disbursement of Afghans from camps that were not previously part of a drainage system as part of a government of Pakistan crackdown order on undocumented migrants from Afghanistan. However, the team could not definitively rule out that the virus could have been circulating among the community within Karachi [
10].
Population movement dynamics are critical to understanding communicable disease transmission patterns and determining where, when, and with whom to deliver appropriate prevention interventions. The polio vaccination status of children under the age of five within the Afghan population residing in Karachi is found to be lower than the population’s average coverage of 95% [
13]. This raises concerns as it positions this population as potential carriers of WPV1 actively shedding the virus into the environment. Information such as the vaccination status of children under five years of age belonging to the Afghan population residing in Karachi and population movement patterns is important to determining whether the source of infection is the population recently arriving in Karachi from Afghanistan or whether there is another population in Karachi with low immunity levels who had recently moved from a different area in the same city. Without this key information, the polio program will be more likely to miss opportunities to vaccinate children who reside in or travel to areas with ongoing transmission and risk further spread of the virus when they travel to other locations. To plan a strategy to effectively reach those most at risk for polio with vaccines, the Sindh EOC determined they needed additional information about the target population and to test the hypothesis that WPV1 arrived in Karachi when a specific population from WPV1-infected districts in Afghanistan arrived around the same time, settling within the established Afghan population within the city all across different districts of Karachi that drain into different ES sites.
There were three study objectives. The first was to identify the origin of the Afghan population and their patterns of movement within Karachi and employ social profiling to determine the social characteristics of the Afghan population in Karachi, including linguistic and tribal affiliations. The second was to assess the polio vaccination status of children under the age of five within the study population through comprehensive epidemiological investigations. Finally, the study aimed to investigate the travel history and guest arrival patterns of individuals from Afghanistan and other regions known to be affected by WPV1 within the past six months to ascertain potential links between travel from WPV1-infected areas and the occurrence of poliovirus within the studied population in Karachi.
3. Results
A total of 409 participants (Afghan household family members) were interviewed. A total of 181 participants (44.3%) were from District West, followed by 147 (35.9%) from District East, 51 (12.5%) from District Central, and 30 (7.3%) from District Keamari. The mean age of the participants was 36.43 with an 11.42 standard deviation (SD), which includes a minimum of 18 years to a maximum 80-year-old participant. The majority of respondents (354 (86.6%)) were females. Languages spoken by participants were 298 (72.9%) Pashto, 84 (20.5%) Dari, 20 (4.9%) Turkman, and 7 (1.7%) Uzbek. A total of 63 participants (15.4%) belonged to the Suleman Khel tribe, followed by 48 (11.7%) Aka Khel, 33 (8.1%) Uzbek, and 33 (8.1%) Tajik. A minimum of zero and a maximum of nine children under five years of age were reported in the households. Respondents reported 395 (96.6%) children received OPV during the last campaign. Respondents reported polio teams comprising of one member visited 145 (35.5%) households, teams with two members visited 215 (52.6%), teams with three members visited 46 (11.2%) households, and a four-member team visited 3 (0.7%) households.
Most Afghan household members (331, 80.9%) had lived for ten or more than ten years in their current location residence, 12 (2.9%) for five to ten years, 20 (4.9%) for three to five years, 33 (8.1%) for one to two years, and 13 (3.1%) for less than one year. Additional demographic characteristics of the study are shown in
Table 1. When asked where in Afghanistan they lived before coming to Pakistan, 55 (13.4%) mentioned only the country of Afghanistan but did not specify the district of origin. The most frequently mentioned districts were Kunduz (39, 9.5%), Kabul (20, 4.9%), Kandahar (18, 4.4%), Mazar Sharif (14, 3.4%), Helmand (12, 2.9), and Takhar (10, 2.4%). The full list of districts is available in
Figure 1a. One participant (0.2%) indicated that they were living in Iran before arriving in Karachi. The most frequently reported origin from other districts within Pakistan included among the majority of 71 (17.4%) were reported from Quetta, followed by 22 (5.4%) from Peshawar, 9 (2.2%) from Zhob, 6 (1.5%) from Pishin, 3 (0.7%) from Waziristan, 2 (0.5%) from Bannu, 2 (0.5%) from Chaman, and 2 (0.5%) from Hyderabad (
Figure 1b). A total of 16 (3.9%) participants did not provide information about the district of origin, whereas 56 (13.6%) reported that they were from Karachi (
Figure S1).
Among study participants, 234 (57.2%) reported that their children completed their full vaccination course of routine immunization, 158 (38.6%) reported their children received some but not all vaccines, and 17 (4.2%) were not vaccinated by any routine immunization antigen.
The majority of FLWs 207 (50.6%) were working within the area for polio program for 3 to 5 years, followed by 97 (23.7%) for one to two years, 92 (22.5%) for more than five years, 10 (2.4%) for less than 6 months, and 3 (0.7%) for six months to one year. The behavior of the FLWs/polio team with the family was reported as good by 398 (97.3%) participants, satisfactory by 3 (0.7%) participants, and 8 participants (2%) reported bad behavior of the polio team.
Travel of any household member outside the city within the last six months was reported by 105 (25.7%) participants, 140 (34.2%) participants reported they hosted guests who arrived within the last six months, and 92 (22.5%) participants reported that guest children were vaccinated in their households. For the questions assessing travel of household members, 230 (56.2%) participants observed polio teams at relatives’ households within different districts of Karachi and 127 (31.1%) observed polio teams at relatives’ households outside Karachi in different districts of Pakistan and Afghanistan.
Most reported guests came from the east and west districts of Karachi, constituting 25 individuals (28.1%) and 23 individuals (25.8%), respectively. Guests reported from districts outside Karachi included 11 (12.4%) from Quetta, 2 (2.2%) from DI Khan, 1 (1.1%) each from Badin, Peshawar, Hub, Zhob, Qila Saifullah of Pakistan, 4 (4.5%) from Afghanistan, and 1 (1.1%) each from Baghlan and Kunduz (
Figure 2,
Table 2).
The incidences of travel to Afghanistan without specifying a district name were reported by nine individuals (3.6%). Specifically, six individuals (2.4%) reported travel to Kabul, another six (2.4%) to Kunduz, five (2.0%) to Kandahar, two (0.8%) to Nangarhar, two (0.8%) to Mazar-e-Sharif, and others. Additionally, 28 individuals (11.2%) reported travel to Quetta, 17 (6.8%) to Peshawar, 6 (2.6%) to Zhob, 6 (2.6%) to Hyderabad, 5 (2%) to Pishin, 2 (0.8%) to Waziristan, and 1 (0.4%) each from Badin, Mirpurkhas, Hub, Qila Saifullah, and others (
Figure 2,
Table 2). Furthermore, travel within different districts of Karachi was high as most traveled to east district (72 (28.8%)), and 26 (10.4%) traveled to west district.
Figure 2 provides a visual representation depicting the details of travel to and guest arrivals from various districts within the Karachi division, as well as districts in other regions of Pakistan and Afghanistan.
Fair to moderate agreement was observed between information provided by the household members and FLWs on the variable’s duration of living at current residence (Kappa = 0.370), travel history (Kappa = 0.429), guest arrival (Kappa = 0.395), and household children vaccinated for OPV (Kappa = 0.419), and the agreement was statistically significant for these variables with
p-value = <0.001. However, no agreement was observed for guest children vaccinated with OPV (Kappa = 0.000,
p-value = <0.999), but it was also statistically insignificant (
Table 2). Furthermore, substantial or perfect agreement was not observed for any variable.
4. Discussion
Simultaneous isolation of WPV1 from four different sites of Karachi with genetic linkages related to an ES isolate from Jalalabad, Nangarhar, Afghanistan (AFG) in November 2022, was an alarming situation for the polio program of Sindh, Pakistan [
10,
18]. All four VP1 genomes differed by <1.5% from the November 2022 isolate. They were isolated within a very short time, and their genomes differed by <0.5%. This is within the margin of error of 1% divergence per year of circulation. The small difference in genomes suggests short-term local circulation; however, there is a possibility that this circulation might still have occurred elsewhere and then been introduced in Karachi along with the extensive population movement to the different catchment areas. Karachi has remained one of the country’s greatest historical poliovirus exporters, with evidence of viruses linked to Karachi isolates appearing in many districts across Pakistan and Afghanistan once Karachi becomes infected [
19,
20,
21]. In the past, genetic sequence data analysis indicated that most WPV1 lineages were transmitted between Pakistan and Afghanistan [
22,
23,
24]. Due to the uncontrolled population movement across the borders between the two countries, the Pakistan–Afghanistan block is considered a single poliovirus reservoir that shares multiple poliovirus lineages as determined from the genetic data analysis of the past few decades [
22,
24]. From this study, it is evident there is heavy population movement among Afghan household members, with 26% of household members reporting recent travel within Karachi and from WPV1-infected areas of Pakistan and Afghanistan. Also, 56% of participants observed polio teams at relatives’ households within different districts of Karachi and 31% at relatives’ households outside Karachi in different districts of Pakistan and Afghanistan. In addition, 34% of guests arrive from different districts within Karachi, other districts of Pakistan, and Afghanistan. This suggests extensive population movement from Afghanistan to Pakistan. The virus is traveling into Pakistan from Afghanistan, as a majority of the WPV1 found in ES of Punjab, KP, and Karachi originated from Afghanistan [
25]. In the current scenario, it is an import in Karachi, due to the high-risk population movement from Afghanistan to Pakistan [
25]. In a previous study, a high number of high-risk and mobile populations (HRMP) with substantial links with Afghanistan and throughout Pakistan were also reported overall; 84% of children originated outside of their current district, including 29% from Afghanistan [
26].
Extensive population movement within different districts of the Karachi division and across Pakistan in the six months preceding the study was also observed. This includes movement to districts that have already reported polioviruses such as Peshawar and Waziristan. We have also noted reported population movement to Hyderabad, Mirpurkhas, Pishin, Hub, and Quetta districts that should be considered the next destinations of poliovirus spread within the country. Karachi does not share borders with Afghanistan and the distance from Karachi to different travel sites is as far as Peshawar (1554.5 km), Waziristan (1181.1 km), Pishin (735.8 km), Quetta (685.4 km) Mirpurkhas (232.8 km), and Hyderabad (163.7 km) and as close as Hub (22.9 km). Epidemiological analysis of WPV1 shows that in the past, most of the poliovirus burden was shared by three major reservoirs including Karachi, Peshawar, and Quetta block (64.2% in 2015, 75.4% in 2016, and 76.7% in 2017) in Pakistan [
21,
27]. The past genetic data reflect sustained transmission within reservoir areas, further expanded by periodic importations, which is also evident from population movement patterns within core reservoir areas as identified in this study. Before the isolation of these four WPV1 from Karachi, only two ES samples came back positive for WPV1 from Karachi, but after this event, several ES sites became positive in Karachi and all were linked to each other. A total of 32 ES isolates were reported from Karachi in the year 2023. Moreover, two human-confirmed polio cases were also reported from Karachi in October 2023.
The United Nations estimates that between 3.7 and 4.4 million Afghans reside in Pakistan [
28]. Sindh province has 73,789 registered Afghan citizens with most living in Karachi [
28]. In October 2023, the interior minister of Pakistan ordered all undocumented Afghans to voluntarily leave the country by 1st November 2023, which impacted the approach to conducting this survey. Due to these repatriation activities initiated by the government of Pakistan, the investigators anticipated that the Afghan population could be reluctant to give information about their origin, travel, and guest arrival from Afghanistan. Therefore, the questionnaire included questions like whether participants observed polio team visits at relatives’ houses in other districts of Pakistan and Afghanistan to give an indirect indication of the travel of the household members to cross-check with directly reported travel information. We also selected the survey timeline carefully to avoid conflict between Afghan household members and polio workers in the same area due to multiple knocks on their doors within a short timeframe. The survey was conducted in the second week of November after the completion of the outbreak response campaign and after the deadline for repatriation so that Afghan families would not speculate the polio worker was connected to the repatriation activities to avoid negatively impacting the polio campaign. More than 80% of families lived in a place for more than 5 years, but they have also maintained their connections with their hometowns in Afghanistan. They are traveling to Afghanistan, and guests from Afghanistan are also coming to visit them. This frequent travel in and out of the country is associated with the introduction of WPV1 in Karachi.
It is encouraging to know from the household members that the behavior of the polio team with the families was good as stated by 97%, and only 2% were not happy with the behavior of FLWs. The satisfactory behavior of the polio team was also reported by the majority of participants in a previous study conducted in high-risk unions in Karachi [
14]. Fair to moderate agreement was observed between information provided by the household members and FLWs on the duration of living at the current residence, travel history, guest arrival, and household children vaccinated for OPV. However, no agreement was observed for guest children vaccinated with OPV. Also, substantial or perfect agreement was not observed for any information the household members and FLWs provided. It indicates that Afghan household members are less likely to give complete or accurate information to FLWs when asked about travel or guests.
Contrary to our hypothesis that the polio vaccination status of children under five years of age belonging to the Afghan population in Karachi is sub-optimal, 97% of children received OPV in the last SIA. High vaccination coverage of 98% in HRMP including Afghan refugees was also reported in a previous study [
26]. Vaccination of children of HRMP is a priority of the polio program in Pakistan. Household members also indicated that 23% of guest children were also vaccinated by the polio teams at their homes. However, we cannot rule out recall bias or deliberate overreporting of polio vaccination status by the household members. It is also a limitation of our study that we have not asked about the age of the household member who traveled to Afghanistan as a person of more than 5 years old who was vaccinated or naturally exposed to poliovirus might still be an asymptomatic carrier. Reported routine immunization status showed a suboptimal level of vaccination. A total of 57% of respondents reported children in their households had received all the vaccines on the routine immunization schedule, 39% reported partial vaccination, and 4% reported their children had not received any routine vaccines. Earlier studies conducted in Karachi also report a low proportion of fully immunized children [
29]. Another limitation is that we have not collected information separately on each vaccine of the RI in the questionnaire, therefore the proportion of fully immunized children may be overreported. The phenomenon of over-reporting vaccination rates has been documented across various contexts, revealing discrepancies between reported and actual coverage levels. For instance, research in India has highlighted discrepancies between reported and verified immunization coverage rates, emphasizing the necessity for robust verification mechanisms [
30]. In Pakistan, evaluations of polio vaccination campaigns have also encountered challenges in accurately assessing coverage due to factors such as incomplete reporting and logistical constraints during door-to-door campaigns [
31]. These examples underscore the critical importance of implementing independent verification mechanisms to validate reported data, thereby ensuring accurate assessment and facilitating effective public health responses. One of the limitations is the timing of data collection immediately after a vaccination campaign and during a politically sensitive period, such as the repatriation of Afghans, might have influenced participant responses. Despite the reported high coverage from the polio program’s door-to-door OPV vaccination campaigns, the presence of WPV detections in ES sites across Karachi confirms that unvaccinated or under-vaccinated carriers of WPV1 exist in Karachi that are shedding poliovirus in the environment. Several factors may contribute to this discrepancy. Migrants or visitors who are not immunized might introduce or reintroduce the virus to the local population. Furthermore, inaccurate reporting and record-keeping can lead to over-reporting of vaccination coverage due to systemic inaccuracies, social desirability bias, or record-keeping errors, giving a false sense of security regarding the actual immunization status of the population. Also, certain fully vaccinated asymptomatic children can be carriers of poliovirus. Nevertheless, there is a possibility of vaccination failures as reported in previous studies conducted in Uttar Pradesh and Behar, India [
32,
33]. Even some children with 30 doses still were paralyzed as children were vaccinated with OPV but the vaccination was not as effective as expected or OPV did not reach optimal titers due to malnutrition status, overcrowding, high amounts of poliovirus in the environment, and competition by other enteric viruses [
32,
33].
To determine the root causes of continued poliovirus transmission despite reported high vaccination coverage, several approaches can be employed. First, seroprevalence studies measure the actual immunity levels in the population by collecting and analyzing blood samples from a representative sample of both children and adults to assess the presence of poliovirus antibodies. Second, molecular epidemiology traces the source and transmission pathways of the virus through genetic sequencing of poliovirus isolates from infected individuals or environments. Identifying genetic similarities can help pinpoint whether it is due to local transmission or importation from other regions. Third, evaluating vaccine efficacy and whether vaccine failure contributes to transmission by comparing immune responses and protection in vaccinated individuals, considering factors such as the number of doses received and the time since vaccination. Identifying potential issues with vaccine efficacy, such as cold chain failures or vaccine handling issues, can help implement corrective measures.
The following policy steps are recommended to enhance poliovirus eradication efforts in Karachi: firstly, targeted vaccination campaigns should prioritize intensive efforts in districts with high Afghan populations and mobility, ensuring comprehensive coverage among children under five years old. Secondly, improving community engagement is essential. This can be achieved through culturally sensitive approaches to communication and trust-building with Afghan communities, addressing vaccine hesitancy, and promoting immunization. Thirdly, integrating vaccination efforts with routine immunization services is imperative. Special attention should be given to improving coverage and equity in immunization services among Afghan children. Lastly, fostering cross-border collaboration with Afghan authorities is essential. This collaboration should aim to synchronize vaccination efforts and enhance monitoring of population movements across the Pakistan–Afghanistan border, thereby boosting overall poliovirus eradication efforts.