Determinants of urban mosquito population density and community responses: A cross-sectional study

Introduction

Vectors are defined as organisms that transmit infectious pathogens from humans to humans, or from animals to humans. Some of the most prevalent vector-borne diseases include malaria, filariasis, dengue, chikungunya, yellow fever, chagas disease, bubonic plague, and leishmaniasis are transmitted by arthropod vectors. These diseases contribute 17% of infectious diseases that affect the human population worldwide and cause approximately 7,00,000 deaths per year.¹

Mosquitoes are one of the most prominent arthropod vectors, representing a significant portion of the vector-borne disease burden, with over 80% of the global population are at risk.¹ Mosquitoes are arthropods of medical importance under the class Insecta and are further divided into Anopheline and Culicine mosquitoes. Anopheline mosquitoes, which are responsible for malaria transmission, exhibit nocturnal biting habits (typically bites between 10 PM and 4 AM) and indoor resting behavior and breed in clean, sunlit water sources. India accounts for 79% of the global malaria burden, underscoring its public health significance.^2–4

Culicine mosquitoes, notably Aedes aegypti and Aedes albopictus, are highly domesticated; they breed in water-filled containers in domestic and peridomestic areas and bite primarily during the day and evening. These species transmit dengue and chikungunya, with Asia accounting for 70% of the global dengue risk, and India being a significant contributor.^5–7 The first chikungunya outbreak in India occurred in the 1960s, followed by a period of dormancy until a major resurgence in 2006, which affected 13 states in the country.^8,9

India’s vulnerability to mosquito-borne diseases is exacerbated by its eco-socio-demographic conditions, making it a major public health concern. Karnataka, a southern state in India, faces a significant burden of mosquito-borne diseases including malaria, dengue, lymphatic filariasis, and Japanese encephalitis, with the prevalence varying across districts. In 2010, Karnataka reported 1,09,118 malaria cases, 28,065 of which were attributed to Plasmodium falciparum.¹⁰

The coastal cities of Mangalore and Udupi together account for approximately 72% of the malaria cases reported in Karnataka.¹¹ In particular, Mangalore—a coastal town in the Dakshina Kannada district with frequent heavy rainfall and high humidity—provides an ideal environment for mosquito proliferation. An initial survey conducted in Mangalore taluk identified 26 mosquito species across six genera, with an annual parasite index of 10–12, indicating that the area was malaria endemic.¹² Rapid urbanization in urban Mangalore, including extensive construction, inadequate drainage, and poor road conditions, has further contributed to the persistence of the endemic nature of mosquito-borne diseases in the area.¹³

Numerous studies have identified environmental, socioeconomic, and behavioral determinants of mosquito proliferation and vector-borne disease transmission. Aquatic habitats, such as pools, streams, and water-filled containers, are critical breeding sites for mosquitoes. For instance, a study by Wilke et al. (2020) in Miami-Dade Florida identified a few of the common aquatic habits that are responsible for harboring 80% of all immature Ae. Aegypti and the increase in proliferation in the presence of those aquatic habitats.¹⁴ Another study by Prashanthi et al. (2007) showed that Anopheles breeds in pools and streams, where people living in close proximity are at high risk of malaria and its transmission.¹⁵ These studies have also revealed that socioeconomic factors exacerbate the vulnerability to malaria, as economically marginalized populations often lack access to anti-mosquito measures, such as mosquito nets or repellents, and may follow age-old traditional practices, such as sleeping outdoors at night amid peak mosquito activity.

The use of preventive measures, such as effective lids over water storage containers and frequent emptying of containers, significantly reduces the incidence of arthropod proliferation, especially in Ae. Aegypti.¹⁶ Studies have shown a linear relationship between growing populations, rising socioeconomic status, and increased mosquito proliferation, particularly in economically marginalized densely populated areas vulnerable to dengue.¹⁷ A study by Srividya et al. (2018), through logistic regression analyses indicated that tiled and concrete dwellings increased the likelihood of an area becoming a dengue hotspot by 2.0 and 2.9 times, respectively, due to the conducive breeding environment.¹⁸

There is a significant relationship between rapid, unplanned urbanization and the proliferation of mosquitoes. One study revealed that inadequate urban infrastructure and sanitation play important roles in the transmission and reproduction of vectors, especially Aedes aegypti. These unplanned disorganized cities aggravate mosquito proliferation by creating artificial breeding grounds, such as stagnant water pools, and increasing disease transmission.¹⁹ Climate change also has a considerable impact on vector proliferation. It reduces larval development time and rapidly increases mosquito populations. This also leads to a reduction in the extrinsic incubation period of pathogens in mosquitoes, thereby increasing their infectiousness.¹⁹

Community knowledge and behavior are critical for effective vector control. A study by Garbin et al. (2021) revealed that while 76% of respondents believed that their neighborhood was likely to be infected by a disease spread by mosquitoes, but no action was taken by them, highlighting a gap between awareness and actions.²⁰ Another study by Madeira et al. (2002) demonstrated that didactic interventions among schoolchildren increased knowledge about mosquito breeding sites and vector proliferation, leading to heightened awareness.²¹

Various determinants of mosquito proliferation have been identified across different studies, and the present study builds on this evidence by exploring the specific environmental, socioeconomic, and behavioral factors driving mosquito proliferation in Mangalore and assessing community measures to mitigate vector-borne disease risk. By examining vector nuisance, disease prevalence, and community engagement, this study aligns with Sustainable Development Goal 3.3—ending epidemics of malaria and other communicable diseases. The results provide valuable evidence to support urban vector control programs and the development of tailored community-based interventions.

Objectives

To identify the key determinants of mosquito presence in urban settings and assess community-measures taken to prevent an increase in mosquito density and control mosquito-borne diseases. Also to evaluate the community perceptions towards mosquito-borne diseases and quantify the disease burden from self-reported cases.

Methodology

This community-based cross-sectional study was conducted in Mangalore, a coastal city in the South Indian state of Karnataka, between September and October 2024, among Community households present in the urban field practice area of the Department of Community Medicine. The sample size was calculated based on a previous study conducted in Mangalore, Karnataka,²² it was reported that 83% of the people used preventive measures such as mosquito nets to prevent mosquito bite, using this as our anticipated proportion and 10% relative precision, 95% confidence interval, and adding 20% non-response rate as 95 sample size was calculated as follows:

N = Z α 2pq/d2: Z = 1.96, 95% confidence interval; p = 0.83; q = 1-p = 0.17, and d is 10% relative precision and 20% of non-response rate.

The study protocol was approved by the Institutional Ethics Committee, Kasturba Medical College Mangalore with No: IEC KMC MLR 09/2024/587, followed by permission from the Head of the Institute. The written consent was taken on an ‘Informed consent form’ which was provided to the participants ≥18 years of age for signature, and the participants <18 years of age were excluded from the study. All participants had the right to withdraw at any stage of the study, and all incomplete responses were considered withdrawal and excluded from the analysis. The study participants were approached from a household that was selected from five wards out of the total 60 wards present under the Mangalore City Corporation based on convenience; one reliable informant residing in the household for more than at least 1 year, who was aware of the household conditions and consented were included in the study, whereas temporary residents of less than a year, apartment complexes, and households without adult personnel were excluded.

The number of households to be taken for our study was divided equally among each ward; that is, a total of 19 households from each of the selected wards were considered for the study. A random street was selected from each ward and then, standing at the end of each randomly selected street, every 3^rd house present on the left side of the lane was considered for the study, which was continued in a clockwise order; in case of failure to meet the inclusion criteria, absence, or denial to take part in the study, the next house was considered. When the head of the household was absent during the time of data collection, information was collected from the oldest adult residing in the household. A pretested, semi-structured, internally validated physical questionnaire designed after extensive literature review was used for data collection, which comprised details such as sociodemographic data, mosquito proliferation and nuisance, determinants of mosquito breeding, behavioral and preventive measures, perception related to mosquito control, and self-reported cases of mosquito-borne diseases. Data were recorded after obtaining informed consent from the heads of the households.

The data collected were entered into MS Excel and analyzed using Jamovi version 2.6.26. Descriptive statistics are represented using frequencies and proportions. The association between two categorical variables was assessed using the chi-square test. The strength of association between the independent and dependent variables were measured using odds ratios (OR) with corresponding 95% confidence intervals (CI) and significant p value of <0.05. Multivariate logistic regression was performed to assess the independent effect of overlapping risk factors.

Results

Discussion

In the present study, a majority (95%) had some level of education, and 61% reported nearby healthcare facilities within 2 km, indicating relatively good access to healthcare in the study area. These findings match those of prior studies highlighting that educational level and proximity to health services may influence awareness and preventive behaviors related to vector-borne diseases.^20,21 Approximately 42% of participants reported an increase in mosquito breeding sites in the past year, and 60% of participants reported experiencing moderate to severe mosquito nuisance in their locality over the past year. This contrasts with the findings of Davidson et al., who noted that 80% of respondents in St. Johns County, Florida, felt bothered by mosquitoes daily or several times a week.²³ Notably, 71.5% of the participants indicated that evenings were the peak time for mosquito activity, and 30.5% reported that mosquito bites disrupted their sleep, which is consistent with Aedes aegypti’s day-biting habits, whereas Anopheline mosquitoes are known for nocturnal activity.¹⁵

Additionally, 67.4% of the participants described a moderate to severe presence of mosquitoes indoors, while 87.3% reported similar conditions outdoors. This finding is comparable to data from Kampango et al., who reported an average of 85.93% An. gambiae s.l. bites per night, with 66% occurring indoors and 34% outdoors, peaking between 22:00 and 03:00 in a rural community in Chókwè District, southern Mozambique.²⁴

In the present study, 50% of the mosquitoes were often found inside the house, whereas 48.3% of the population rarely found mosquitoes inside the house after the use of mosquito repellents or coils. A total of 64.5% of the population rarely found mosquitoes inside houses, whereas 32.8% often found mosquitoes inside houses after regular repair or checking holes in window screens or mosquito nets. A total of 61.2% of the population rarely found mosquitoes inside houses without usage of insecticide sprays at home. A total of 84.3% often found mosquitoes inside the house when they stored water in closed containers. Those who often repaired or checked holes in window screens and mosquito nets were more likely to report a lower presence of mosquitoes than those who rarely performed the repair.

Considering the environmental factors, 69.4% of the participants identified the rainy season as the period when mosquitoes were most prevalent. This aligns with a cross-sectional study by Mahgoub et al. in Barakat and El-Kareiba, Sudan, which reported a high number of positive habitats during the rainy season, whereas the lowest numbers were reported during the hot season followed by the dry season, corroborating findings that climate and seasonal variation significantly influence mosquito density and disease transmission.^19,25

The primary breeding sites identified in our study included water stagnation (74.7%), dense vegetation (54.7%), and nearby water bodies (20%), which is consistent with previous studies demonstrating the importance of stagnant water and vegetation in vector proliferation.^14–16 In contrast, Mahgoub et al. reported that the main breeding sites for various mosquitoes in Sudan were leaking water pipes (51.5%), followed by irrigation channels (34.2%), hoof prints (6.4%), tire tracks (5.5%), and water tanks (2.4%).²⁵ Logistic regression analysis in this study revealed that water stagnation (OR 2.5, 95% CI: 1.1, 5.7 p value = 0.033) was significantly associated with a greater presence of mosquitoes, suggesting that stagnant water plays a crucial role in mosquito proliferation, and the absence of construction (OR 3.0, 95% CI: 1.0, 8.8 p value = 0.014) activity was significantly associated with a lower presence of mosquitoes. Multivariate logistic regression analysis, adjusting for water stagnation, showed that only the presence of construction sites remained significantly associated with increased mosquito density (OR 3.0, 95% CI: 1.0–8.8 p value = 0.042). These findings suggests that construction sites may serve as important breeding grounds for mosquitoes and that areas without construction activity experience considerably lower mosquito density. These findings highlight the need for targeted vector control measures at construction sites to reduce mosquito proliferation.^18,19

Preventive measures among participants were notable, as 91.5% used chemical repellents (insect sprays, vaporizers, and smoke), 81% kept doors and windows closed, 55.7% utilized mosquito nets or screens, and 15.7% employed other personal measures. This contrasts with a study in urban northern Gujarat by Mahalakshmi et al., in which 67.3% used chemical measures such as repellent creams or liquids, 22% used bed nets, and only 2% reported taking no precautionary measures.²⁶

In terms of community cleaning practices, 51.5% of individuals reported cleaning their surroundings daily; additionally, 38.9% checked for water stagnation weekly; in contrast, 84.6% never reported water stagnation to authorities. Unfortunately, 58.9% indicated that municipal cleaning was infrequent in the locality. Notably, 25.2% of the participants regularly checked window screens or mosquito nets for damage, and 50.5% stated that there had been no anti-fogging efforts by local authorities. This is compared with Mahalakshmi et al.’s findings in northern Gujarat, where 88% cleaned their homes daily, 57.3% cleaned their surroundings weekly, and 82% actively avoided water stagnation.²⁶

Despite these measures, approximately 30% of the participants reported suffering from a mosquito-borne disease in the past year, with dengue accounting for 71.4% of the cases and malaria accounting for (28.6%). Common symptoms included fever (96.4%), headache (78.5%), joint pain (60.7%), and muscle pain (46.4%), and 82% of the patients sought treatment at private clinics. This aligns with findings from a study conducted by Kumar et al. in a tertiary hospital in Udupi district, which indicated that 83.9% of cases were due to dengue fever, presenting symptoms such as fever (99.1%), myalgia (64.6%), vomiting (47.6%), headache (47.6%), abdominal pain (37.5%), and breathlessness (17.8%).²⁷

Finally, 92.3% of the participants in our study reported experiencing complications postrecovery, primarily weakness, whereas 7.6% experienced cold, leg pain, headache, or eye pain. This contrasts with a study in Vietnam by Tam et al., who reported that 12.5% of participants experienced alopecia, 11.1% had blurred vision, 9.5% faced concentration difficulties, and 8.5% suffered from fatigue following dengue infection.²⁸

The present study found that 82.1% of the population perceives water stagnation as a significant factor in mosquito proliferation, indicating that it is a major factor in the prevalence of mosquito-borne diseases. Construction sites, water bodies, dense vegetation, and water storage in open containers were identified as significant determinants. Additionally, 17.8% of individuals perceive water leakage from tanks and taps (p = 0.166) as a key determinant, indicating that water leakage from household stores is a significant factor in the prevalence of mosquito-borne diseases. Overall, these factors contribute to the prevalence of mosquito-borne diseases.

Interestingly, the use of mosquito repellents or coils was significantly associated with a higher incidence of mosquito-borne diseases (OR 3.7, 95% CI: 1.4, 9.6, p = 0.006). This odd finding may indicate that individuals resort to chemical measures instinctively in areas with high mosquito density rather than those measures that effectively reduce disease risk. This finding also suggests possible improper usage or overreliance on repellents without addressing environmental breeding sites. Similar gaps between knowledge and action have been reported in prior studies.²⁰

Conclusion

In conclusion, the study adopts a novel, integrated approach by identifying key determinants of mosquito presence while simultaneously evaluating community measures, perception and self-reported disease burden, provide context specific insights from Mangalore that extend beyond previous studies in other settings. The study showed the knowledge and attitudes regarding mosquito borne diseases in the community are apt and good. It correlates with good practice but only to a certain extent. Practicing the basic requirements, for example, using mosquito sprays and closing windows, is to par. However, proper intervention by local authority is necessary to combat the main factors responsible for mosquito breeding, such as water stagnation and the presence of dense vegetation. This proves the presence of gaps identified by our study, where despite most people being aware of mosquito-borne diseases, respondents recognized the lack of awareness and good preventive practices. Hence, apart from awareness, there is a dire need to provide the right personnel and services to combat mosquito-borne diseases.

Recommendation

Ethical approval

The study protocol was approved by the Institutional Ethics Committee, Kasturba Medical College Mangalore with No: IEC KMC MLR 09/2024/587, followed by permission from the Head of the Institute. The written consent was taken on an ‘Informed consent form’ which was provided to the participants ≥18 years of age for signature, and the participants <18 years of age were excluded from the study. All participants had the right to withdraw at any stage of the study, and all incomplete responses were considered withdrawal and excluded from the analysis.