Univariate statistics (percentages, means, or medians in the case of highly skewed distributions) were calculated for the variables, according to type.

Pearson’s correlation coefficient was computed between level of agreement with the statement, “Mosquitoes are a serious threat to my health” and perceived level of concern about potentially contracting a serious disease from mosquito bites in the respondent’s province of residence. The individual risk perception index (RP index) was then created by calculating the mean value of these two variables. The two variables refer to perceived susceptibility, i.e., the perception of being at risk for the condition, and the perception of risk severity, i.e., the seriousness of the condition. Severity and susceptibility dimensions were selected to compose the RP index because some of the main behavior change models, such as the Health Belief Model and Protection Motivation Theory, identify these constructs as the components of the threat appraisal process, which is a key element in influencing behaviors (31).

Variation in the RP index across sociodemographic variables was investigated using the independent samples t-test for two independent groups, and one-way analysis of variance (ANOVA) with Gabriel’s post-hoc correction for comparisons of more than two groups.

The “protective behaviors” variable was built by summing 14 dichotomous variables that captured the adoption of protective behaviors. For the protective behaviors variable, the missing data correspond to individuals who indicated that they do not use any mosquito-protection system. Their answers were therefore coded as individual “I do not use it” responses for each protective behavior.

A knowledge variable was created by summing the correct answers to 6 questions (true/false/I do not know) about mosquito ecology and biology. For knowledge variable “do not know” responses were recoded together with the incorrect ones as “other” responses.

After verifying the linear relationship between variables, in the absence of outliers and given the large sample size. Pearson’s correlation coefficient was calculated between the RP index and continuous variables, such as respondents’ knowledge, and protective behaviors.

Cronbach’s alpha coefficient was calculated to assess the internal consistency of the multi-item variables measured on a quantitative scale.

Data were analyzed using Statistical Package for Social Sciences (SPSS) software (version 25.0) for Windows (SPSS Inc., Chicago, Illinois). The level of statistical significance was set at 5% (α = 0.05).

The majority of the sample, consisting of 1,001 respondents, was female (54.6%), aged between 45 and 64 years (36.6%), had children (60.8%), had a high school diploma (37.9%), and defined themselves as employed (47.7%). Table 2 provides a detailed overview of the respondents’ characteristics. When asked to describe the area and setting in which they lived, most respondents (72.4%) reported residing in a low-lying area, while 13.1% lived in hilly areas, 6.7% in mountainous areas, and 7.8% along the coast or on islands. Furthermore, 63.3% reported living in an urban environment and 36.7% in a rural setting. Approximately 81% of respondents declared to have never traveled to tropical countries.

Most respondents (82.8%) perceived a high presence of mosquitoes in their area. When comparing the current situation to the past 2–3 years, 44.5% perceived an increase in the presence of mosquitoes, 45.6% reported perceiving no change, and 10% perceived a decrease. Those reporting an increase attributed it primarily to environmental factors, such as climate change and the presence of standing water, and secondarily to human behaviors, including the lack of disinfestation efforts or insufficient preventive measures.

In general, respondents expressed a relatively high level of agreement with the statement, “I think mosquitoes are very annoying” (median = 7; IQR = 17), while there was less agreement with the statement, “Mosquitoes are a serious threat to my health” (mean = 3.95; SD = 1.948).

Respondents were less concerned about contracting a serious MBD in their own province compared to contracting one elsewhere in Italy (Table 3).

Cronbach’s alpha coefficient indicated satisfactory internal consistency for the two items presented in Table 3 (α = 0.876).

Findings showed a strong positive correlation between the statement, “Mosquitoes are a serious threat to my health” and perceived level of concern about contracting a serious disease from mosquito bites in own province of residence (Pearson’s correlation coefficient r = 0.528, p-value<0.001).

The most commonly reported PPBs against mosquito bites included the use of skin-applied repellents (62.6%), followed by environmental repellents such as mosquito coils, electric diffusers (60.1%), and mosquito nets (53.5%).

In addition to PPBs, respondents reported taking actions to reduce the presence of mosquitoes, as removing standing water from outdoor containers (49.5%), mowing the lawn (35.3%), and applying larvicides to drains or containers (18.7%) (Table 5).

The local entities most frequently identified as key participants in the fight against mosquitoes were municipal authorities (63%), followed by land reclamation authorities (58.8%), and local health authorities (52.6%). Additionally, 40.8% of respondents identified citizens as important stakeholders in mosquito control efforts.

Using the independent t-test, the RP index proved higher in females compared to males (p = 0.002) and among respondents who reported having recently read or heard news about mosquitoes in the newspaper or on television compared to those who had not (p < 0.001). ANOVA analysis revealed that the RP index varied by educational level (p = 0.001). According to Gabriel’s post-hoc correction, respondents with a primary/middle school (mean RP index = 3.753) or high school diploma (mean RP index = 3.630) showed higher risk perception than those with a university degree (mean RP index = 3.244, p = 0.001 and p = 0.007, respectively). Regarding employment status, students reported lower risk perception (mean RP index = 2.859) compared to the employed (mean RP index = 3.529, p = 0.004), unemployed (mean RP index = 3.795, p = 0.009), homemakers (average RP index 3.905, p = 0.001), and retirees (mean RP index = 3.633, p = 0.002).

No significant relationships were observed between the RP index and respondents’ age (p = 0.333), the presence of children in their household (p = 0.066), the setting (urban, suburban, rural) in which they lived (p = 0.989), and whether they had traveled to tropical countries (p = 0.940).

Respondents’ knowledge of mosquito ecology and biology was inversely correlated with the RP index, indicating that those with a higher level of knowledge expressed a lower degree of risk perception (p = 0.001).

The RP index was positively correlated with: (1) the “protective behaviors” variable, indicating that higher risk perception levels were associated with a higher tendency to adopt protective behaviors (p = 0.001), and (2) descriptive social norms, i.e., the perception that other people living in the area considered mosquitoes to be very annoying (p < 0.001) and to be a serious threat to the community (p < 0.001).

Respondents perceived a high presence of mosquitoes in their areas of residence, with nearly half observing an increase over the past 2–3 years. However, these perceptions are not supported by entomological surveillance data for WNV vectors, which show that mosquito populations have remained relatively stable over the past 10 years (32). In the study, perceived abundance did not automatically translate into heightened perceptions of health risk. While mosquitoes were widely considered a nuisance, the perceived threat to personal health remained moderate. This discrepancy between nuisance perception and risk perception has also been observed in other European settings and may reflect low levels of perceived vulnerability (29, 33, 34). Furthermore, respondents expressed significantly lower concern when assessing risk at the local compared to the national level. This suggests that respondents underestimated the risks to themselves and their local environment, which may be linked to psychological factors such as perceived invulnerability and perceived control over own surroundings (35, 36).

Knowledge levels varied across domains. Respondents demonstrated good knowledge about environmental factors driving mosquito proliferation, including climate change. They also showed strong awareness of mosquito breeding sites—particularly water-filled containers and natural stagnant water, consistent with findings from other studies where standing water was widely acknowledged to be a potential breeding site (34, 37). However, some misconceptions did emerge about mosquito biology, since a large proportion of respondents incorrectly believed that mosquitoes feed exclusively on blood or that both males and females bite.

The majority of respondents correctly identified the presence of both native and invasive mosquitoes, although nearly one in three expressed uncertainty. Good knowledge about the presence of invasive species is due to the establishment of Aedes albopictus (the Asian tiger mosquito) in Italy for over 30 years (38). However, the more recent introduction of other non-native species is still ongoing and the introduction of additional invasive species (Aedes koreicus and Ae. japonicus) presents new challenges for mosquito management and control.

Respondents were aware that mosquitoes can transmit pathogens causing disease and that different species can transmit different pathogens. This becomes even more significant in view of their limited knowledge of the MBDs present in their own area of residence. The best-known disease was WND, which has become endemic in the two regions considered (9, 39). Conversely, public perception of malaria did not correspond to scientific evidence: one in five respondents believed it was still present in their area in recent years and the same percentage of respondents did not know whether or not it was present. Although malaria historically affected some areas of the Po River Valley, Italy has officially been declared malaria free by the World Health Organization (WHO) since 1970. Nonetheless, mosquitoes of the Anopheles maculipennis complex—vectors of the Plasmodium parasite—are still present in the area (7). Several European studies have shown similar gaps in MDB knowledge, likely reflecting the sporadic nature of local transmission and the predominance of outbreaks linked to imported cases (40, 41). A study conducted in central Italy comparing the KAP of an Italian sample with two Indian communities showed that while people from endemic areas were more concerned and more knowledgeable about MBDs, they perceived mosquitoes less as a nuisance (42).

Encouragingly, respondents reported engaging in a broad set of PPBs, particularly the use of repellents and physical barriers, such as mosquito nets. Respondents also practiced environmental prevention measures, including the removal of standing water. One in five respondents avoided going outdoors in the afternoon or evening to reduce mosquito bites. This is in line with other studies where people reported changing their daily activities to avoid bites (19).

A growing body of evidence has demonstrated that the effectiveness of mosquito control interventions and the prevention of MBDs is strongly influenced by the adoption of appropriate PPBs among the population. Several studies (34, 43) have identified individual awareness of health risks as a key determinant in promoting mosquito control practices and encouraging the adoption of PPBs. Therefore, understanding population attitudes toward mosquitoes and MBD becomes key to the design of effective, targeted health communication strategies.

Given the absence of vaccines for most MBD, primary prevention relies on vector control strategies and the adoption of PPBs (44). Notably, empirical studies have estimated that regular application of PPBs can reduce the risk of WNV infection by approximately 50%, making them the most effective form of personal prophylaxis (45).

Municipal authorities and health services were widely seen as responsible for mosquito control, while only 40.8% identified citizens themselves as key stakeholders. This reveals a gap in public understanding of integrated vector management, where community engagement is considered a cornerstone of effective, sustainable mosquito control. On the other hand, the way citizens are involved and the effectiveness of interventions is a pivotal issue (33, 46). Hence, intervention planning requires not only institutional leadership but also systematic evaluation of citizen involvement in terms of the cost-effectiveness of control strategies.

The study found that individuals with higher risk perception scores (RP index) were more likely to engage in PPBs, confirming the central role of perceived threat in shaping health-related actions. This result is consistent with psychological frameworks that posit perceived susceptibility and severity as key predictors of protective behavior (47, 48). Importantly, PPBs are not determined solely by risk perception; they are also associated with disease-related knowledge and sociodemographic variables (19, 20). A systematic review on KAP and their determinants found that concern about WNV circulation appeared to be a predictor of all examined preventive behaviors (49). The correlation between risk perception and behavioral practices is consolidated by other studies on different MBDs (34, 43).

With regard to sociodemographic variables, the survey revealed that gender, educational attainment, and occupational status significantly influenced perceived risk. Women reported higher risk perception than men, a finding consistent with existing literature, suggesting that women tend to express greater concern toward health threats (27, 29, 50). Respondents with lower educational levels expressed higher perceived risk compared to those with tertiary education. While this may reflect unequal access to scientific information, it could also suggest a heightened sense of vulnerability among individuals with less formal education (51, 52). Furthermore, students reported significantly lower risk perceptions compared to employed or retired respondents. This could reflect both lower feelings of vulnerability or less attention to vector-borne disease threats in public communication interventions targeting young people. A study conducted in Australia found that younger people exhibited low risk perception and limited knowledge of MBDs, identifying them as an important target for tailored risk communication (29).

Conversely, risk perception was negatively associated with level of scientific knowledge about mosquito biology and ecology. More knowledgeable respondents tended to report lower levels of perceived risk. This pattern may reflect informed individuals’ greater ability to contextualize the actual health threat posed by mosquitoes considering that the probability of infection is still relatively rare in Italy. Another possible explanation is that knowledge can foster protective behaviors, which increase feelings of control and safety, thereby reducing risk perception (53). This finding is consistent with other studies that showed that higher levels of knowledge can sometimes lead to more rational, and thus lower, risk appraisal (54, 55).

The study also revealed that descriptive social norms were positively associated with perceived risk. This aligns with Ojala and Lidskog (50), who found that individuals who perceived their community to be highly disturbed by mosquitoes were more likely to develop heightened risk perceptions, corroborating broader evidence on the impact of descriptive norms on risk appraisal (56, 57). Moreover, results showed that exposure to media reports influenced risk perception. Annual outbreaks of diseases such as WNV and dengue in certain areas, often amplified by national media coverage, can increase perceived risk even among individuals with no direct personal experience. According to the theory of social amplification of risk, exposure to mosquito-related news and shared information in informal networks was associated with increased concern, suggesting that media coverage and social relationships remain powerful tools for amplifying risk perception and fostering public engagement (58). This effect can become even stronger when the issue has a local dimension, as in the case of mosquito-related risks, in which case the community shares information, management initiatives, and interventions.

The results of this study reinforce the idea that risk perception is shaped by social and emotional cues, including media salience and perceived social attitudes, which may serve as important triggers for action (21, 26, 29).

When interpreting the results, it is important to contextualize the period in which the survey took place and recognize the possible impact of the COVID-19 period on the public perception of risks and especially of MBDs. In the spring of 2020, the rapid spread of COVID-19 in Italy led to profound changes in individual behaviors and required the adoption of extensive personal and collective protective measures. How this global event—with its substantial health, economic, and social consequences—affected vector-borne diseases has been the focus of several studies. These analyses highlight a contraction in resources dedicated to prevention and control activities, potentially resulting in underdiagnosis and underestimation of cases on the one hand, and a possible increase in vector populations on the other (59). With regard to public awareness, the interruption of information campaigns and community-engagement activities may have reduced public attention to MBDs and a downgrading of their perceived risk relative to the pandemic emergency (60). Moreover, media coverage of mosquito-borne diseases may also have been influenced by the predominance of COVID-19 as the major public health concern during that period. It would be interesting to further investigate how COVID 19 may have affected the relationship between risk perception, protective behaviors and other factors such as social norms in the MBDs domain.

The results of this study showed a perceived increase in mosquitoes presence that may be linked to an actual rise in their numbers due to the reduction of control interventions in urban areas during the pandemic, but it could also be the result of heightened threat perception caused by the pandemic itself. However, these hypothesis cannot be confirmed by entomological data in the considered territory, at least for Aedes albopictus (the main urban species in Italy), as it was not possible to carry out monitoring in 2020 due to restrictions imposed by COVID-19. On the other hand, entomological surveillance for WNV in 2020 (conducted regularly) did not report a significant difference in the number of Cx. pipiens captured.

Understanding people’s perceptions and opinions about mosquitoes, as well as their attitudes toward these insects is an essential part in the design of risk communication strategies and in planning citizen engagement in preventive activities. These efforts should build awareness that everyone can contribute to protecting public health. By adopting correct PPBs, citizens play an important role in mosquito density control and, consequently, in the incidence of MBDs. To be effective, strategies to mitigate health risks associated with mosquitoes must therefore adopt an interdisciplinary, multi-stakeholder approach that combines epidemiological surveillance, education, communication, and community mobilization. Based on the study findings, several recommendations can be proposed to improve mosquito control efforts and public health communication strategies.

Enhanced public awareness campaigns:

Community engagement in vector control:

Leverage media and social norms:

Support local authorities in the promotion of sustainable control measures: