Molecular Detection and Genetic Diversity of Measles and Parainfluenza Viruses Circulating in Humans in Bangladesh: A Cross‐Sectional Study

ABSTRACT

Background and Aims

Among Paramyxoviruses (PMVs), measles virus (MeV) and human parainfluenza virus (HPIV) are major contributors to respiratory tract infections worldwide. Despite recurrent outbreaks of MeV and the significant health impact of HPIV, data on their molecular epidemiology and genetic diversity in Bangladesh remain scarce. This study aimed to detect and genetically characterize circulating MeV and HPIV strains among humans to better understand their distribution and genetic diversity.

Methods

We conducted this study as part of Global One Health Surveillance between 2017 and 2019 across three communities and one hospital in Bangladesh. We enrolled community participants, who had the potential to harbor viruses, and hospital participants with Influenza‐like illness (ILI). Survey data and biological specimens were collected to assess behavioral risk factors and virus spillovers. Samples were tested using consensus PCR, targeting the polymerase (pol) gene to detect PMV. Positive samples were confirmed by sequencing. Firth's logistic regression and phylogenetic analyses were used to evaluate associations and viral diversity.

Results

Overall PMV prevalence was 1.2% (n = 10; 95% CI: 0.5–2.1), while MeV and HPIV were the same: 0.58% (n = 5; 95% CI: 0.2–1.3). MeV was found in participants having symptoms of ILI (5/508), while HPIV was detected both in ILI cases (3/508) and in asymptomatic individuals (2/354), respectively. The multivariable model showed MeV was significantly associated with participants' age, livelihood, and season, while no association was seen for HPIV. Phylogenetic analysis revealed MeV‐D8 and B3, and HPIV‐1, 2, 3 circulating regionally and clustered with Asian strains, suggesting continued viral transmission and evolution.

Conclusion

Our study provides molecular characterization of the circulation of diverse MeV and HPIV strains in humans across Bangladesh. Despite their endemicity, continued surveillance concerning vaccination coverage and genome sequencing is recommended to better understand the spatiotemporal dynamics of these key respiratory viruses in humans.

Summary

• Molecular characterization of diverse MeV and HPIV strains circulating in humans across Bangladesh.

• MeV genotypes B3 and D8, alongside HPIV 1–3, indicate the co‐circulation of diverse Paramyxoviruses in the study area.

• MeV infection is reported among adults, particularly in the dry season, while HPIV is identified in both symptomatic and asymptomatic individuals, suggesting potential silent transmission.

1. Introduction

Paramyxoviruses (PMV), notably measles virus (MeV) and human parainfluenza viruses (HPIV), are among the foremost viral agents responsible for a wide spectrum of respiratory illnesses worldwide [1]. Both viruses represent a substantial burden of disease, causing severe illness, lifelong complications, and death, especially among young children, the elderly, and immunocompromised individuals [2, 3]. MeV, the etiological agent of measles, comprises 8 clades with 24 genotypes, is among the most contagious viruses (R ₀: 12–18), and has a high potential to trigger large epidemics [4]. Clinical manifestations of measles include fever, rash, cough, coryza, and conjunctivitis, which may lead to common complications, like diarrhea, otitis media, as well as severe outcomes, such as blindness, encephalitis, and pneumonia [5]. Despite a remarkable global decline in measles incidence and mortality during the early 21st century, measles resurfaced around the world between 2017 and 2019 [6]. The global resurgence of measles in recent years has been exacerbated by disruptions in immunization during the COVID‐19 pandemic, causing a dramatic upsurge in cases between 2022 and 2023 worldwide [7]. Bangladesh experienced multiple sporadic measles outbreaks between 2000 and 2016, even with substantial progress in national vaccination programs, including the introduction of the second dose of measles vaccine in the Expanded Program of Immunization [8]. Although vaccination coverage reduced overall measles incidence, outbreaks and fatalities continued, driven by immunity gaps, uneven vaccine coverage, suboptimal surveillance, and COVID‐19‐related setbacks [9]. The steadily increasing measles fatality over the period of time implies an imminent threat in every region of the world, including Bangladesh [10].

HPIVs, another respiratory pathogen, are classified as four serotypes (HPIV 1–4), pose an ongoing challenge to global health due to their prevalence, disease severity in vulnerable populations, and their contribution to respiratory outbreaks worldwide [11]. These viruses affect individuals of any age group, are typically self‐limiting, but can cause reinfections throughout life [12]. Results from multicountry pneumonia case‐control studies illustrate that HPIV is one of the leading viral causes of childhood acute lower respiratory illness (ALRI) [13]. HPIV is accountable for around 19 million ALRI cases, 725,000 hospital admissions, and 34,400 pediatric deaths annually [14]. In Asian countries, the prevalence of HPIV infection ranges from 1% to 66% in children [15]. HPIVs serotypes are responsible for asymptomatic or mild upper respiratory infections to life‐threatening lower respiratory tract infections, such as pneumonia and bronchiolitis [16, 17], and contribute to recurrent seasonal outbreaks. Epidemiological studies also highlight seasonal variations and frequent co‐infections with other respiratory viruses [18]. These global resurgences of infections underscore the urgent need for enhanced surveillance and vaccine development.

The epidemiology of these viruses is shifting as a result of changing human behaviors, vaccination impact, and genetic mutation [19]. High population density, high birth rate, and suboptimal immunity further facilitate transmission of these viruses among infants and older individuals [20]. Although the global epidemiology of MeV and HPIV, as well as the global diversity of MeV, has been well documented [21, 22, 23] but data on comprehensive genomic sequencing of HPIV is limited [24]. Despite existing epidemiological research, data on the genetic diversity and molecular research on these viruses in Bangladesh remain scarce. To address this gap, this study aimed to detect and molecularly characterize MeV and HPIV circulating in humans in Bangladesh. To our knowledge, this is the first genetic epidemiological analysis of MeV and HPIV within the country. These findings provide valuable insights into genetic diversity and transmission dynamics, enhancing surveillance and guiding regional efforts toward the effective control and elimination of these pathogens.

2. Materials and Methods

2.1. Study Location and Design

This cross‐sectional study was conducted as part of the Global One Health Surveillance initiative, aimed at strengthening the in‐country capacity to detect and characterize the viruses with known epidemic and unknown pandemic potential with zoonotic spread [25]. We enrolled 862 participants across 3 community study sites, Dhaka, Madaripur, and Dinajpur, and 1 hospital catchment site at Faridpur Medical College Hospital, from 2017 to 2019 (Supporting Information S1: Figure 1). Study sites were selected as emerging infectious disease hotspots, reflecting high population density, rapid environmental changes, and socioecological conditions that may favor pathogen emergence and spread [26, 27].

2.2. Study Population and Sample Collection

In each community, we enrolled study participants purposively, according to the case definition of study protocol, that is community members having frequent animal contact who could be potential carriers of viruses and later categorized them as (i) apparently healthy individuals without symptoms who may carry viruses, (ii) symptomatic participants with self‐reported suspected Influenza‐like Illness (ILI; acute onset of fever ≥ 38°C (100.4°F) within the last 10 days and cough) [28]. In the hospital, we recruited patients from the outpatient department (OPD) of the Medicine Unit diagnosed with ILI by the attending OPD physician. Participants under 2 years of age, prisoners, or unable/unwilling to provide consent or assent, including those with mental impairments, were excluded. We interviewed the enrolled participants through a standardized behavioral questionnaire to identify the behavioral risks and collected biological specimens of blood, oral swab, nasal swab, and rectal swab using standard personal protective measures [29]. Swab samples were collected through sterile larger polyester swabs (Fisher) and stored in a cryotube (Corning) with viral transport medium (VTM). Blood was collected into EDTA tubes, and later serum was separated by centrifugation and stored in cryotubes without media. All tubes were immediately frozen in liquid nitrogen and then transferred to an ultralow freezer (−80°C) for further processing. A unique identification number was allotted to each questionnaire and biological specimen collected from each participant. No personal identifying information was collected. The authorized study personnel who received human subject research training were allowed access to the questionnaire and biological data.

2.3. Ethical Statement

Ethical approval was obtained from the Institutional Review Boards of the Institute of Epidemiology, Disease Control and Research (IEDCR; behavioral study IEDCR/IRB/2017/06), and the University of California, Davis (Protocol Number 804522‐15). Prior to the interview, informed written consent and age‐appropriate assent were taken from all participants after a brief explanation of the study.

2.4. Statistical Methods

All statistical analyses were performed using R version 4.2.0 within RStudio version 2022.02.2. The prevalence and 95% confidence intervals (CIs) were calculated using the exact binomial test. For univariate analysis, we considered “Age,” “Gender,” “Season,” “Education,” “Livelihood,” “No. of people living in the same dwelling,” “Health Status,” as categorical variables, were assessed using the χ ² test to examine associations with MeV, HPIV, and PMV, with a significance threshold set at 0.2 to be selected for multivariable analysis. Multicollinearity among the explanatory variables was assessed using Cramer's V, considering a cutoff value of 0.4 [30]. Subsequently, a generalized linear model (package glm) was applied to evaluate the effects of risk factors. Although logistic regression initially proved challenging due to the small number of positive participants with MeV and PMV, resulting in convergence issues, we addressed this by employing Firth's logistic regression (package logistf). Firth's method is particularly beneficial when dealing with rare events or binary outcomes with small sample sizes, offering a solution to convergence problems and potential biases encountered in traditional logistic regression [31].

2.5. Laboratory Testing and Analysis

2.5.1. Laboratory Analysis

At the outset, all samples were screened for PMVs using consensus PCR following the protocol described by Tong et al. [32], with detailed information provided in Supporting Information S1: Table 1. RNA pellets were suspended in 60 µL of DNase‐free, RNase‐free water and stored at −80°C until tested (Bio‐Rad C1000). PMV RNA targeting the polymerase (pol) gene was amplified using two rounds of forward and reverse primers. PCR products were subsequently purified using the Omega Gel Extraction kit.

2.5.2. Sequence Analysis

Positive PCR products of the expected size were confirmed and sequenced by traditional Sanger dideoxy sequencing, followed by analysis with appropriate bioinformatics tools. In total, six MeV and five HPIV sequences were obtained and submitted to the GenBank database (available at https://www.ncbi.nlm.nih.gov/genbank/) under the following accession numbers: MeV‐ MT063360, MT063415, MT063647, MT063648, MT063646, MT063649; HPIV‐ MT063414, MT063439, MT063440, MT063650, MT063651. After sequencing the polymerase (pol) genes of MeV and HPIV, highly similar sequences were identified in GenBank using the Basic Local Alignment Search Tool (BLAST). Sequences sharing ≥ 98% identity with previously reported sequences were considered strains of known viruses and allotted the identical name as the corresponding top match sequence [33].

2.5.3. Phylogenetic Analysis

We utilized the bootstrapping method to determine whether the sequences included in this study had undergone recombination, and no recombinant strains were detected (data not shown). We used ClustalW to achieve a multiple sequence alignment of the studied sequences with closely similar sequences retrieved from the National Center for Biotechnology Information (NCBI) via BLAST. A total of 60 MeV and 77 HPIV gene sequences were checked for the best BLAST match based on E‐value, percent identity, query coverage, and aligned them to construct phylogenetic trees incorporating our study sequences. The best‐fitting evolutionary model was determined using Molecular Evolutionary Genetics Analysis (MEGA) version 11, with the Kimura‐Nei model selected according to the Hierarchical Likelihood Ratio Test [34]. Evolutionary relationships were inferred using a maximum likelihood method, and branch support was estimated using 1000 bootstrap replicates [35, 36].

3. Results

3.1. Demographic Characteristics of Study Participants

Between October 2017 and January 2019, a total of 862 participants were enrolled in this study from 4 sites in Dhaka (n = 212), Madaripur (n = 200), Dinajpur (n = 200), and Faridpur (n = 250). The mean age of participants was 34 years (SD ±17.6), ranging from 2.2 to 98 years. Overall, 64.6% of participants were male, and ~40% had attained primary education or less. Primary livelihoods of the participants included animal production, trading and market business, crop production, hunting, and community health work, and other nonanimal businesses such as construction, driving, or daily labor. We conducted interviews biannually during the wet (52%, May through October) and dry (47%, November through April) seasons. Nearly all participants (99%) shared their living space with one another. Participants who reported ILI symptoms were categorized as “symptomatic” (59%), whereas apparently healthy individuals without reported any symptoms were classified as “asymptomatic” (41%) (Table 1).

Table 1.

Demographic characteristics and health status of the study participants.

	Community sites (n = 612)	Hospital site (n = 250)	Total (n = 862)
Age category
≤ 20	149 (24.35)	79 (31.60)	228 (26.45)
21–30	128 (20.92)	61 (24.40)	189 (21.93)
31–40	127 (20.75)	41 (16.40)	168 (19.49)
≥ 41	208 (33.99)	69 (27.60)	277 (32.13)
Gender
Male	353 (57.68)	204 (81.60)	557 (64.62)
Female	259 (42.32)	46 (18.40)	305 (35.38)
Education
No formal education	214 (34.97)	52 (20.80)	266 (30.86)
Primary school	258 (42.16)	85 (34.00)	343 (39.79)
Secondary school	96 (15.69)	62 (24.80)	158 (18.33)
College/university/professional	44 (7.19)	51 (20.40)	95 (11.02)
Primary livelihood
Child/student/unemployed	98 (16.01)	81 (32.40)	179 (20.77)
Homemaker	184 (30.07)	28 (11.20)	212 (24.59)
Animal production/trade/market business	226 (36.93)	23 (9.20)	249 (28.89)
Crop production	49 (8.01)	32 (12.80)	81 (9.40)
Community health workers	4 (0.65)	13 (5.2)	17 (1.97)
Hunter/trapper/fisher	1 (0.16)	3 (1.20)	4 (0.46)
Nonanimal business	50 (8.20)	70 (28.0)	120 (13.9)
Season
Dry	300 (49.02)	110 (44.00)	410 (47.56)
Wet	312 (50.98)	140 (56.00)	452 (52.44)
No. of people living in the same dwelling
None	3 (0.49)	1 (0.40)	4 (0.46)
1–3	228 (37.25)	93 (37.20)	321 (37.24)
4–5	279 (45.59)	94 (37.60)	373 (43.27)
> 6	102 (16.67)	62 (24.80)	164 (19.03)
Health status
Symptomatic	258 (42.2)	250 (100)	508 (58.9)
Asymptomatic	354 (57.8)	0	354 (41.07)

3.2. Prevalence of MeV and HPIV in Community and Hospital Settings

Overall, the prevalence of PMV was 1.2% (n = 10; 95% CI: 0.5–2.1), with MeV and HPIVs each comprising an equal proportion of 0.58% (n = 5; 95% CI: 0.2–1.3), respectively. From the hospital, MeV was identified in four individuals (1.6%), whereas only one case (0.16%) was detected in the Madaripur community. Genotyping revealed that MeV‐B3 was present in three hospital cases, while genotype D8 was found in individuals from both community and hospital settings. Furthermore, HPIV serotypes 1–3 were identified at the community level, with two strains (HPIV‐1: 0.3%) from Dhaka and three (HPIV 1–3: 0.4%) from Madaripur. Notably, we did not find HPIV‐4 in our study samples, even any single HPIV serotypes identified at the hospital site (Table 2).

Table 2.

Laboratory diagnosis of Paramyxovirus subtype infection.

Virus	Virus strain	Community sites	Hospital site	p
Measles (MeV)	MeV B3	0	3 (1.20)	0.02
	MeV D8	1 (0.16)	1 (0.40)
Human parainfluenza (HPIV)	HPIV 1	3 (0.49)	0	0.56
	HPIV 2	1 (0.16)	0
	HPIV 3	1 (0.16)	0

3.3. Epidemiological Characterization of PMVs

MeV was solely detected among participants aged 21–30 years (2.6%), while HPIV was identified among all age categories. The association between virus detection and age was statistically significant for MeV and PMV (p < 0.001), but not for HPIV. PCR‐confirmed viral infections were more frequently observed in male participants (1.08%). In the case of participants' education and primary livelihood section, we found a significant association with the MeV and PMVs (p < 0.05) rather than HPIVs. Seasonal analysis showed a higher prevalence of both MeV and HPIV during the dry season compared to the wet season; however, these differences were not statistically significant. MeV was identified in symptomatic participants having a history of ILI, whereas HPIV was detected both in ILI cases (3/508) and asymptomatic individuals (2/354), respectively (Table 3).

Table 3.

Characteristics and risk factors of PCR‐confirmed diagnosis.

Variables		Measles virus			Parainfluenza virus			Paramyxovirus
		n (%)	95% CI of prevalence	p	n (%)	95% CI of prevalence	p	n (%)	95% CI of prevalence	p
Age	≤ 20	0	0	< 0.01	2 (0.88)	0.10–3.13	0.89	2 (0.88)	0.10–3.13	0.03
	21–30	5 (2.65)	0.86–6.06		1 (0.53)	0.01–2.91		6 (3.17)	1.17–6.78
	31–40	0	0		1 (0.60)	0.01–3.27		1 (0.60)	0.01–3.27
	> 40	0	0		1 (0.36)	0.009–1.99		1 (0.36)	0.009–1.99
Gender	Male	3 (0.54)	0.11–1.56	0.83	3 (0.54)	0.11–1.56	0.83	6 (1.08)	0.35–3.32	0.76
	Female	2 (0.66)	0.07–2.35		2 (0.66)	0.079–2.34		4 (1.31)	0.39–2.32
Education	No formal education	—	—	< 0.01	1 (0.38)	0.009–2.07	0.74	1 (0.38)	0.009–2.07	0.024
	Primary school	—	—		3 (0.87)	0.18–2.53		3 (0.87)	0.18–2.53
	Secondary school	1 (0.63)	0.01–3.47		1 (0.63)	0.016–3.47		2 (1.27)	0.15–4.49
	College/university/professional	4 (4.2)	1.15–10.4		—	—		4 (4.21)	1.15–10.43
Primary livelihood	Child/student/unemployed	1 (0.56)	0.01–3.1	< 0.01	2 (1.12)	0.13–3.97	0.49	3 (1.68)	0.34–4.81	< 0.01
	Homemaker	0	0		1 (0.47)	0.01–2.59		1 (0.47)	0.01–2.59
	Animal production/trade/market business	0	0		0	0		0	0
	Crop production	0	0		0	0		0	0
	Community health workers	2 (11.76)	1.45–36.4		0	0		2 (11.76)	1.45–36.4
	Hunter/trapper/fisher	0	0		0	0		0	0
	Nonanimal business	2 (1.67)	0.20–5.89		2 (1.67)	0.20–5.89		4 (3.33)	0.91–8.31
Season	Dry season	4 (0.98)	0.26–2.47	0.14	3 (0.73)	0.15–2.12	0.57	7 (1.71)	0.68–3.48	0.15
	Wet season	1 (0.22)	0.005–1.22		2 (0.44)	0.053–1.58		3 (0.66)	0.13–1.92
No. of people living in the same dwelling	≤ 4	3 (0.56)	0.12–1.62	0.91	3 (0.56)	0.12–1.62	0.91	6 (1.12)	0.41–2.41	0.84
	> 4	2 (0.62)	0.07–2.21		2 (0.62)	0.07–2.21		4 (1.23)	0.34–3.13
Health status	Symptomatic	5 (0.98)	0.32–2.28	0.06	3 (0.59)	0.12–1.71	0.96	8 (1.57)	0.68–3.08	0.17
	Asymptomatic	0 (0)	—		2 (0.56)	0.06–2.03		2 (0.56)	0.06–2.03

Note: Bold values indicate statistical significance of p < 0.05.

The univariable analysis identified five factors were significant (p < 0.2) for PMV and MeV (Table 3). As Education and Age had Cramer's V > 0.40, we excluded Education from the multivariable model (Supporting Information S1: Figure 2). Additionally, “Health Status” was excluded from the model as it represents a host characteristic rather than a predictive factor. We used the remaining three variables as explanatory variables in generalized linear models to find out the factors that significantly affect the presence or absence of PMV and MeV in the participants (Table 4).

Table 4.

Estimates with standard error and p value of Firth's logistic regression model of presence/absence of PMV and MeV.

Variable	Category	Paramyxovirus				Measles virus
		Estimate	Std. error	OR (95% CI)	p	Estimate	Std. error	OR (95% CI)	p
Age [ref: ≤ 20]	21–30	1.79	0.85	6.01 (1.13–31.69)	0.053	3.47	1.43	32.17 (1.95–529.96)	0.009
	31–40	0.62	1.13	1.85 (0.20–17.47)	0.631	1.15	1.89	3.17 (0.08–127.81)	0.610
	> 40	0.22	1.04	1.25 (0.16–9.65)	0.852	0.96	1.75	2.62 (0.09–78.60)	0.650
Season [ref: Wet]	Dry season	1.47	0.60	4.34 (1.34–14.25)	0.024	2.15	0.86	8.59 (1.57–47.19)	0.022
Livelihood [ref: Homemaker]	Child/student/unemployed	1.98	1.04	7.24 (0.95–55.11)	0.080	2.90	1.52	18.26 (0.90–364.28)	0.068
	Crop production	0.13	1.55	1.14 (0.05–25.14)	0.940	1.38	1.83	3.96 (0.11–147.73)	0.517
	Animal production/ trade/market business	−1.25	1.52	0.29 (0.01–5.64)	0.411	−0.17	1.82	0.85 (0.02–37.20)	0.933
	Community health worker	3.28	1.05	26.58 (3.39–208.10)	0.004	4.67	1.53	106.99 (5.32–2140.51)	0.001
	Nonanimal business	1.75	0.90	5.73 (1.00–33.02)	0.048	2.26	1.42	9.54 (0.62–149.87)	0.089

Note: Bold values indicate statistical significance of p < 0.05.

Firth's logistic regression analysis indicated that the risk of PMV infection was approximately four times higher during the dry season compared to the wet season. Additionally, the odds of PMV were 26 times higher among community health workers and 5 times higher in participants involved in other nonanimal businesses compared to homemakers (Figure 2). In contrast, for MeV, participants aged 21–30 years had a 32 times higher risk compared to those under 20 years. Furthermore, the odds of MeV were nearly 9 times higher during the dry season compared to the wet season. Remarkably, community health workers showed a 106.99‐fold increased risk of MeV infection (Figure 1). We depicted the predictive risk of PMV and MeV, along with 95% CI, illustrated in Figures 2 and 3 for clearer insight.

Figure 2.

Estimated marginal means (predicted values adjusted for all other effects) and their 95% confidence intervals for significant explanatory variables that have effects on Paramyxovirus. Panel A shows the effect of season (wet vs. dry), and Panel B shows the effect of livelihood categories on PMV prevalence.

Figure 1.

Odds ratio with 95% confidence interval of the variables of the Firth's logistic regression model relating demographic characteristics with presence/absence of Paramyxovirus and measles (*p < 0.05, **p < 0.01 ).

Figure 3.

Estimated marginal means (predicted values adjusted for all other effects) and their 95% confidence intervals for significant explanatory variables that have effects on Measles, and Panel A presents livelihood effects, Panel B presents age, and Panel C presents seasonal effects on MeV prevalence.

3.4. Molecular Characterization of MeV and HPIV

Phylogenetic analysis based on the nucleoprotein (N) and hemagglutinin (H) genes revealed that the Bangladeshi MeV sequences clustered within two distinct genotypes, B3 and D8 (Figure 4). Genotype D8 sequences from both Faridpur and Madaripur were closely related to strains previously detected in China, India, South Korea, Thailand, Indonesia, and Vietnam, as well as in the United States, United Kingdom, France, and Australia between 2015 and 2019, suggesting widespread global circulation of this lineage. Conversely, B3 genotype sequences from Faridpur clustered with contemporary strains from India, the United States, Italy, and South Korea, reflecting regional and intercontinental transmission links. The co‐circulation of B3 and D8 genotypes underscores persistent endemic MeV transmission in Bangladesh, despite ongoing vaccination and elimination efforts.

Figure 4.

Phylogenetic trees based on partial nucleotide sequences of the Polymerase (pol) gene from MeV, family Paramyxoviridae, using the maximum likelihood method and 1000 bootstrap replications with the use of MEGA 10.0. MeV sequences obtained from humans in this study are marked with blue diamond‐shaped.

Phylogenetic analysis of the polymerase (pol) gene showed that Bangladeshi HPIV strains clustered within three distinct serotypes: HPIV‐1, HPIV‐2, and HPIV‐3 (Figure 5). The HPIV‐1 sequences from both Dhaka and Madaripur grouped with reference strains from the United States, India, China, Thailand, Argentina, Cambodia, and Vietnam, exhibiting complete sequence similarity with isolates from China and the United States. Two additional sequences from Madaripur represented HPIV‐2 and HPIV‐3, which clustered closely with epidemic strains from India, the United States, and other Asian and American regions (2010–2018), indicating possible cross‐regional introductions and adaptation of Asian lineages to South Asian environments. Notably, all three HPIV serotypes were detected in Madaripur, while only HPIV‐1 was found in Dhaka, suggesting greater viral diversity and local circulation in rural settings.

Figure 5.

Phylogenetic analysis indicates partial nucleotide sequences of the Polymerase (pol) gene from HPIV, family Paramyxoviridae, using the maximum likelihood method and 1000 bootstrap replications with the use of MEGA 10.0. MeV sequences obtained from humans in this study are marked with red, blue, and green diamond‐shaped.

4. Discussion

Between 2017 and 2019, we conducted a Global One‐Health Surveillance among the three district communities and one hospital setting, aiming to detect the known and novel zoonotic viruses in Bangladesh. Our findings revealed the presence of PMV in both community and hospital settings, from which isolates of MeV and HPIV were detected. PMVs showed a significant effect on season and livelihood of those involved in nonanimal business and community health workers, while we found a positive association of MeV with the participants' age, season, and livelihood. Our study reported HPIV infection among the asymptomatic cases, though no significant association was found with HPIV, which might be the reason for low sero‐positivity rates. Phylogenetic analysis revealed, Measles genotype D8 and B3, and HPIV genotypes 1, 2, 3 were circulating in the study area (Figures 4 and 5).

Bangladesh has a high population density with 1333 people per square kilometer [37] and is a hotspot for the emergence and transmission of infectious diseases. Despite achieving relatively high measles vaccination coverage compared with many other countries, the disease still represents a considerable public health burden for the country [38]. In this study, MeV was detected among young adults aged 21–30 years. While measles traditionally affects children under 5 years, recent evidence from China and Sri Lanka revealed an increasing number of adult measles cases among people more than 18 years of age [39, 40]. Researchers found that the level of antibodies in the human body tends to decrease over time, with the lowest titers often observed among individuals aged 16 and above [41]. This suggests that waning immunity, incomplete vaccination coverage, or suboptimal vaccine response may contribute to shifting susceptibility to adolescents and young adults. Our study samples were collected throughout the year, covering both dry and wet seasons. Bangladesh naturally experiences a tropical monsoon climate, with distinct seasonal variation. PMV, including measles, mumps, parainfluenza, and other respiratory viruses, can survive and transmit in various climates and commonly show seasonal patterns depending on virus type and geographical context. Though our study detected PMVs in all months of the year, an increasing trend of PMVs, including measles, was observed in the dry season. The periodic occurrence of these viruses is strongly correlated with seasonal factors. Similar seasonal trends have also been documented in some countries; reported measles cases coincide with the end of the wet season and peak during the dry season [42, 43, 44, 45]. Few studies also reported the seasonal distribution for PMVs [46, 47]. Several factors, like low humidity, crowded living conditions, and contact patterns, can contribute to enhancing viral stability and transmission. Comparatively, seasonal peaks in wet months have also been reported in other tropical regions [13, 47, 48], suggesting that behavioral and environmental factors play a substantial role in sustaining virus circulation.

Our study detected PMV among community health workers and individuals engaged in other job categories involved in nonanimal business, such as construction, driving, and day laborers, while for MeV, a positive association was observed among community health workers, indicating potential occupational exposure and community‐level risks transmission. These viruses are a concern among the people due to their frequent interpersonal contact and mobility, which may increase exposure risk, contributing to sustained virus transmission in the community through respiratory droplets from infected individuals, particularly in densely populated areas. Some similar studies also reported measles infections among the healthcare workers [45, 49, 50], while PMV association was found in individuals who were involved in nonanimal business for their livelihoods [51]. Community health workers can be susceptible to measles if they're unvaccinated or have not acquired immunity through previous infection. Healthcare workers need to be vaccinated against measles and use infection control measures like proper hand sanitizing, wearing personal protective equipment, and implementing isolation protocol measures to protect themselves and prevent the viral spillover to the vulnerable population. Additionally, maintaining high vaccination coverage among healthcare workers will help safeguard both the individuals providing care and the communities they serve.

Phylogenetic analysis revealed the co‐circulation of MeV genotype B3 and D8 in Bangladesh, reflecting ongoing transmission and multiple lineage introductions. The isolates clustered with globally circulating D8 and B3 lineages, consistent with multiple international importations and reflecting global interrelatedness of circulating viruses and the dynamic nature of regional transmission patterns. Similar genotypic patterns have been reported from neighboring countries, including India, Myanmar, and other regional countries such as China, Nepal, Sri Lanka, and Thailand [52, 53, 54], indicating the cross‐border spread in South and East Asia. The detection of genotype B3, traditionally predominant in Africa and increasingly reported in Asia, while genotype D8, a lineage frequently observed in South and Southeast Asia [55]. MeV genotyping commenced in Bangladesh in 2014, with genotype B3 detected during 2014–2015 and genotype D8 in 2016 [8]. Since then, only these two genotypes have been reported in the country [56], although their precise distribution remains unclear, and no other strain has been detected yet.

Globally, MeV genotypes B3 and D8 have driven major outbreaks. Mirroring these trends, Bangladesh has witnessed recurrent outbreaks over the years. MeV genotypes B3 and D8 were identified in China, imported from Egypt and the Philippines [57]. Across Europe, particularly in Romania, France, Italy, Greece, and the United Kingdom, outbreaks from 2016 to 2019 were mainly caused by genotypes D8 and B3 [58]. In the United States, during 2001 and 2019, 29% of measles cases were imported, resulting in 158 outbreaks, whereas all 2019 cases were caused by wild‐type D8 and B3 genotypes of measles [59, 60].

We identified five HPIV isolates, demonstrating the co‐circulation and genetic diversity of three HPIV genotypes in Bangladesh, with greater heterogeneity observed in Madaripur than in Dhaka. HPIV‐1 was closely related to strains from Australia, the United States, Mexico, and Vietnam, while HPIV‐3 isolates clustered with strains from the United States and the United Kingdom. The HPIV‐2 exhibited higher similarity to strains from the United States, Vietnam, and Greece than those isolated from Bangladesh (Figure 5). These clustering patterns suggest intercontinental lineage connections, indicating, possibly facilitated by international travel or migration. Our results are aligned with studies from other countries that have reported the global spread and genetic diversity of HPIVs. For instance, a study from Germany found that all the strains of HPIV were distributed across multiple phylogenetic clades, indicating widespread circulation and genetic diversity [24]. Similarly, research from China identified serotype HPIV‐3, highlighting the occurrence of genetic recombination and the emergence of novel variants [61].

Our study findings provide key insights into the detection and genetic diversity of MeV and HPIVs in Bangladesh, providing crucial baseline information for strengthening surveillance systems and guiding targeted screening strategies. Furthermore, understanding the genetic relationships of local strains with global references can help anticipate the risk of imported lineages and guide vaccination and containment strategies. It is recommended to establish virological molecular surveillance, not only to understand the spillover events but also to ensure accurate diagnosis and differentiation from other respiratory viruses circulating in Bangladesh. The detection of multiple HPIV genotypes suggests long‐standing community circulation, with many cases likely going undiagnosed due to their clinical similarity to the common cold. Despite having robust surveillance for Influenza, the detection of HPIVs, including their seasonal patterns, remains understudied and underreported in Bangladesh, contributing to the unrecognized burden of respiratory illness. Sequence data obtained from molecular monitoring is beneficial for tracking the real‐time pathways of virus distribution in countries over time. Future surveillance, including complete genome sequences, is required to better understand the viral evolution and support the development of effective vaccines and antiviral therapies.

Several limitations should be noted. Since the study was conducted in selected locations, the findings may not be generalizable to the entire population. Due to the limited sample size and study duration, detection of viral strains may be lower. Our study did not collect detailed clinical information related to the viruses, vaccination history, or reinfection history. Therefore, the exact cause of participants' exposure to MeV or HPIV could not be determined accurately. Additionally, sequenced MeV and HPIV strains were limited, which may not fully represent the comprehensive viral diversity across the country. It is recommended to validate these findings in future studies with a prospective approach.

5. Conclusions

This study reports the detection and molecular characterization of human PMV strains, revealing diverse strains of MeV and HPIVs circulating in study areas. Association of MeV was found among age, season, and livelihood, while PMVs were linked to seasons and livelihood factors. Presence of MeV genotypes reflects sustained endemic transmission despite progress toward elimination goals, while HPIV diversity demonstrates continued viral circulation and evolution, underscoring their pivotal role in respiratory disease burden. Strengthening multisectoral surveillance and molecular epidemiological investigation of MeV and HPIVs is crucial to understanding the spatial‐temporal dynamics and guiding effective public health strategies in Bangladesh.

Author Contributions

Conceptualization and supervision: Ariful Islam and Jonathan H. Epstein. Writing – original draft preparation: Shusmita Dutta Choudhury, Md. Arif Khan, and Emama Amin. Formal analysis: Emama Amin, Md. Arif Khan, and Ariful Islam. Writing – review and editing: Sarah Munro, Maria Kaczmarek, Jade K. Forwood, Tahmina Shirin, Meerjady Sabrina Flora, Ariful Islam, and Jonathan H. Epstein. Resources: Jade K. Forwood, Tahmina Shirin, and Meerjady Sabrina Flora. Investigation: Shusmita Dutta Choudhury, Josefina Abedin, and Ariful Islam. Methodology: Shusmita Dutta Choudhury, Josefina Abedin, Tahmina Shirin, Meerjady Sabrina Flora, and Ariful Islam. Project administration: Meerjady Sabrina Flora, Jonathan H. Epstein. Funding acquisition: Ariful Islam and Jonathan H. Epstein. All authors have read and approved the updated version of the manuscript. Dr. Ariful Islam (corresponding author) had full access to all the data in this study and takes complete responsibility for the integrity of the data and the accuracy of the data analysis.

Disclosure

Dr. Shusmita Dutta Choudhury affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Ethics Statement

Ethical approval was obtained from the Institutional Review Boards of the Institute of Epidemiology, Disease Control and Research (IEDCR; behavioral study IEDCR/IRB/2017/06), and the University of California, Davis (Protocol Number 804522‐15).

Consent

Prior to the interview, informed written consent and age‐appropriate assent were taken from all participants after a brief explanation of the study.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Sampling locations pointed in the map of Bangladesh. Figure S2: Correlation heatmap (Cramér's V) among categorical covariates. Each cell shows the strength of association (0–1) between the row and column variables based on Cramér's V; higher values indicate stronger association.

Supporting Figure S2: Matrix plot with Cramer's V values amongst all combinations of explanatory categorical variables. Table S1: Primers used in Paramyxoviruses detection.

Data Availability Statement

All data generated and analyzed during this study are included in this article. Additional raw data sets are available from the corresponding author upon reasonable request. Genome sequence data have been submitted to GenBank and are freely accessible; the corresponding accession IDs are provided in the manuscript.