School Students’ Perceptions on Snakes, their Uses and Snakebite in Nepal: Implications for Snake Conservation and Snakebite Prevention

Study area

This study was carried out in seven randomly selected public and private secondary schools from four purposively selected cities from the lowlands of Nepal (Figure 1, Table 1), which are mostly occupied by tropical forests and fertile agricultural lands interspersed with forests and Koshi and Gandaki river systems. Here more than 50% of Nepalese inhabit. Because of its tropical climate (with temperature ranging from 10°C to 40°C and average annual rainfall 1500mm), ample habitats for diverse snakes, and greatly dense human population, people inhabiting the southern Nepal is highly prone to snakebites where humans and snakes encounter often resulting in snakebite envenomations and ruthless killing of snakes.

Click to enlarge

Symbols: % percent, ‡ sample size/ sample population x 100. Table 1: Study sites and study samples (sampling units, sampling unit population and sample size).

Data collection

We conducted cross-sectional surveys in those schools during Aug-Sep 2019 and collected information using self-administered structured and semi-structured questionnaires provided to 72 eight- and ninth-grader students selected using systematic random sampling from the daily attendance-registers of the respective schools (Table 1, Figure 1). Research assistants including authors facilitated respondents defining specific terms used in questionnaires, displayed visual stimuli [3, 27] (i.e., A4-sized color photographs of adult snakes including photographs of neonatez and juvenile snakes for species with ontogenetic variation in color patterns and known to be distributed in the lowlands of Nepal [3, 23, 28] to measure the ability of school students for differentiating venomous, mildly venomous or non-venomous snakes (Figure 2, Table 2). To avoid duplicating responses for the same questions by respondents, PI and research assistants heeded to respondents in a class room setting (Figure 2). To avoid any bias due to mention of respondent´s name and address in questionnaires, we assigned alphanumeric code provided for signed informed consents.

After permission from the respective school principals, we clearly explained the objectives of our research and asked those students (respondents) if they would participate in the survey research. Then, we obtained written informed consent from the participants for publication of the provided information and any associated images. We did not obligate any respondents to participate in this study. This study was approved by the Directorate of Research and Extension, Agriculture and Forestry University, Rampur, Chitwan, Nepal.

Click to enlarge

Figure 2: Illustration of typical setting for data collection: showing facilitation of respondents from Small Heaven School, Bharapur-05, Chitwan District for defining specific terms used in the questions by the authors (photo a) and displaying native snake photos (photo stimuli) to the respondents from St. Xevier Secondary School, Birgunj-11, Sripur, Parsa District by research assistant (photo b) while they were filling in the questionnaires.

Abbreviations: PN = Number of photos displayed (Snakes represented by PN 10 and 11 are presumed to be distributed in the lowlands of Nepal. So, we included this species despite it was not reported from Nepal yet; Nv = Non-venomous; Mv = Mildly venomous; Bf = Back-fanged; Ff = Front-fanged; Vs = Venomous secretion in saliva; NA = Absence of fang; V = Venomous; T = Typhlopidae; B = Boidae; C = Colubridae; E = Elapidae; Vip = Viperidae; Fam = Family; Tox = Toxicity. Note: This checklist was adopted from available published sources [4, 22, 23, 28, 29, 30]. Although Coelognathus radiatus possesses postsynaptic neurotoxin in its Duvernoy´s gland [31], Harris, et al. [32] reported four C. radiatus bites on the feet causing pain and bleeding at the bite site. Therefore, we considered both snakes as mildly venomous while analysing knowledge of students on native snakes. Table 2: Checklist of snake photos displayed while questionnaire survey of school students to assess their knowledge on snakes distributed in the lowlands of Nepal.

Use, hunting or killing of snakes: To understand patterns of using snakes in Nepal, we asked respondents whether or not students or their neighbor killed snakes for food, ethno- medicine, etc. during the recent last one year. We asked them whether they had seen human, road, and predator killed snakes while commuting from school to house or visiting their relatives, etc. since last one year.

Risks of snake being killed by humans and vehicles: We asked four types of questions to understand attitudes of students towards snakes and snake conservation. To scrutinize and measure attitudes, we asked informants pretested questions [3] related to intention of killing snakes, responses to snakes encountered in defined places (i.e., crop fields, roads, premises of house, and indoor), worship of snakes, realizing the need of snake conservation, and snakes as a farmers’ friend.

We phrased the first type of question as, Do you kill this snake (by displaying 58 photographs of 32 snake species distributed in the lowlands of Nepal)?; we coded responses as “yes” for negative attitude and “no” for positive attitudes. We phrased the second type of question as, What do you do when …?; we coded responses as I ignore it, I kill it, I call others to kill it, I kill it only if I know it is a venomous snake, and I just keep it out using sticks (snake hooks, tong, etc.). We phrased the third type of question as, Which of the following do you consider to be…?; we coded “yes/no” responses for All snakes around us should be killed, Only venomous snakes around us should be killed, and All snakes around us should be conserved. The fourth type of questions included whether respondents worshiped snakes and realized the need of snake conservation, and snake as a friend of farmers.

Awareness: To scrutinize the snake and snakebite awareness level among eight- and ninth-grader school students, we asked 32 pre-tested “yes/no” questions [3], which included useful, useless, deleterious, and fictitious information associated to snakes and snakebite management [23, 30]. Of the 32 questions, 26 were considered to test belief of students in popular and deep-rooted traditional beliefs or misconceptions regarding snakes (n = 13) and pre-hospital care of snakebites (n = 13). Two questions tested their belief in doubtful benefits of pre-hospital care in the context of Nepal [16], and four questions were related to first aid measures (e.g., pressure immobilization bandaging (PIB) and local compression pad immobilization (LCPI)) recommended by the World Health Organization and the Government of Nepal [23, 33, 34, 35, 36].

Knowledge: We tested the students’ knowledge on identifying the 58 displayed snakes that are potential to have been collected or killed by locals in the lowlands of Nepal as venomous, mildly venomous or non-venomous and their understanding of the need for snake conservation. Although Coelognathus radiatus is an archetypal nonvenomous snake, it contains postsynaptic colubritoxin [31]. Considering mild effects of its venom, we have grouped C. radiatus and C. helena into mildly venomous snake.

We presented the first type of question as, which one of the following snakes do you think were venomous, mildly venomous or non-venomous? During data entry, we scored their correct responses for the corresponding snake photos 1 through 58 by crosschecking their replies with the aid of list of snakes displayed (Table 2). To measure knowledge of students on the ecological roles of snakes, we phrased questions such as, Do you know ecological roles of snakes? and If you do think so, why? We asked respondents to give five reasons. To quantify knowledge of the need for snake conservation, we phrased questions such as, Do you think snakes need to be conserved? and If you do/don’t think so, why? We asked informants to give five reasons.

Data analysis

Risk-values of snake species: We developed informant indexing technique [adapted from Phillips, et al. [37]] to obtain the risk-value for snake species (i.e., richness of snake species likely to be killed or collected by humans) distributed in the study area and analyzed risk-values for those species using table, descriptive statistics, and non-parametric test. We entered a value of 1 in a spreadsheet (containing respondents in rows and snake species in columns) for “I kill it” response to each of displayed snake species (Table 2) and 0 for those that were not likely to be killed. Then, we obtained an average score for displayed snakes that were likely to be killed by informants. For each species, we calculated the risk-value involving informant indexing technique which is a quantitative method that demonstrates the relative risk of locally distributed snake species.

Our estimate of the risk-value of displayed snake species “s” for being likely to be killed by each of informants “i” is defined as RVis = ∑ Ris/nis, where Ris equals the number of intentions of killing of given species “s” by informants “i” and nis equals the number of events for species “s” with informants “i”. Our estimate of the total risk-value for each species “s” is calculated using the formula: RVs = ∑i Ris/ns, where RVs = risk-value of the snake species displayed; Ris = number of “I kill it” responses by each informant “i” per displayed snake species “s”; ns = number of informants interviewed for species “s”. The calculation of the risk-values of any snake species was based objectively on the responses by the informants themselves, and was not dependent on the opinion of the researcher.

Score analysis: We analysed scores for awareness and knowledge of students using the nonparametric Wilcoxon test with median scores as the dependent measure [38]. We used the one-sample Wilcoxon signed rank test to understand median scores for each of demographic groups (i.e., male and female students, public and private schools) for awareness and knowledge of students about snakes and prehospital care of snakebites, the two-tailed unpaired Wilcoxon rank sum tests to compare differences of scores among demographic groups, and the one-tailed unpaired Wilcoxon rank sum test to compare maximum scores among these groups. We did not conduct analyses for sample sizes lower than six to avoid problems associated with measurement error.

We analysed awareness based on the percentage of median scores of respondents after conducting the Wilcoxon test. We classified students as “highly aware” (HA) in snake and snakebite issues if ≥75 % students rejected believing or accepting traditional beliefs of snakes and snakebite care, refusing to seek doubtful medical care for snakebites, and acceptance of WHO recommended measures of pre-hospital care. Similarly, we considered respondents “aware” (A), “mildly aware” (MA), and “unaware” (UA) if 50-74 %, 25- 49 %, and 0-24 % respondents responded to the awareness test questions, respectively. We considered all tests to be significant at α=0.05. We rounded p-values (p) to three digits after decimal (values less than three significant digits were represented as p = < 0.001). We performed all analyses using the R statistical package (R Version 3.5.2, © 2018 The R Foundation for Statistical Computing Platform).

Use or killing of snakes

Only 26% (n = 19) of the respondents reported knowing the use of snakes for eight categories. These included supply of snake venom to modern/ayurvedic medicine producers (14%, n = 10), blood and other body parts of snakes for the traditional medicine (11%, n = 8), getting snake skin for leather items (e.g., belt, wallet, etc.) (11%, n = 8), food (4%, n = 3), display by snake charmers (3%, n = 2), worshipping in religious function such as “Nagpanchami” (i.e., a serpent festival) (1%, n = 1), keeping them in museum (1%, n = 1), study in biology laboratory (1%, n = 1) in study sites and other areas in Nepal since last one year. Snakes used were either live or dead specimens. The rest of the respondents (74%, n = 53) did not know any uses of snakes. Only 1% (n = 1) of the respondents saw selling of snake by snake charmers in their locality and 94% (n = 68) did not see selling any snakes or snake products in their localities in open markets (e.g. Haat Bazaar) or registered market (e.g., meat suppliers) or in religious article stores. Item non responses for whether informants had seen selling snakes or snake products were three.

Click to enlarge

Click to enlarge

Figure 4: Box and whisker plots showing scores of Nepalese students for the intention of killing snakes (displayed during survey, see Table 2) with respect to their gender and type of school where they were studying (overall scores for intended killing of displayed snakes: mean± standard error of mean (SEM) = 18±1.9, range = 0-55, standard deviation (SD) = 16, median = 14, Wilcoxon signed rank test value (W) = 1659.5, p = 0.052, 95% confidence interval (CI) = 13.99-21.50, item non-responses = 3, invalid responses = 0).

Only 6% (n = 4) respondents killed a total of 10 snakes since last one year. But, respondents observed/came to know killing of snakes by locals considering them either as a harmful animal (44%, n = 32) or showing stunt (29%, n = 21), by vehicles at road (42%, n = 30), and by predators (38%, n = 27) while commuting from school to house or visiting their relatives and those shared by their parents since last one year. Predators they observed killing snakes were eagle (n = 11 mentions), mongoose (n = 10), cat (n = 4), peacock (n = 2), dog (n = 2), and hen (n = 1).

While analyzing risks of human-caused mortality of snakes, we found a high degree of snake killing attitudes of school students while they encountered snakes indoors or in areas with high human activity, such as yard or backyard, roads, etc. in the snakebite prone lowlands of Nepal (Figure 3). These attitudes varied by localities (Figure 3), school types, and sex (Figure 4). Only 3% (n = 2) informants considered that all snakes around us should be killed (item non-response = 1), 50% (n = 36) considered only venomous snakes around us should be killed, and 82% (n = 59) all snakes around us should be conserved (item non-response = 1). Overall, the median number of snakes likely to be killed by each of informants was 14 (mean±standard error of mean (SEM) = 18±1.9, p = 0.052, 95% confidence interval (CI) = 13.99-21.50). Overall risk-value of displayed snakes was 0.151 (median = 0.0045, Wilcoxon signed rank test value (W) = 401, p = 0.004, 95% CI = 0.0045-Inf, range = 0.003-0.011, standard deviation (SD) = 0.001, SEM = 0.0002). Among all species, Lycodon aulicus was found to be at the highest risks and Amphiesma stolatum and Boiga siamensis were at the least risks of being killed by humans in the study sites (Table 2).

Awareness of students about snakes and snakebites

Altogether 1-67% respondents believed traditional beliefs or misconceptions on snakes, which were potential to cause snakebites (Table 3a). In snakebite event, 50-93% respondents would believe to follow up non-recommended interventions for pre-hospital care of snakebite (i.e., sucking wound, squeezing the wound, ingesting herbal medicine, applying herbal medicine topically, applying (tight) tourniquet) (Table 3b). 15-92% students were aware of recommended pre-hospital care of snakebite (Table 3c,d).

c. Seeking medical help of doubtful use

1 Visiting medical person 69 96 3 4 0

2 Visiting any hospital or healthcare center 70 97 2 3 0

d. Recommended measures of pre-hospital care I know (N) I know (%) I don´t know (N) I don´t know (%) Item non- responses (N)

1 Visiting healthcare facilities supplied with antivenom 66 92 5 7 1 HA

3 Pressure immobilization bandaging (PIB) 52 72 20 28 0 A

4 Local compression pad immobilization (LCPI) 35 49 36 50 1 MA

2 Availability of nearby snakebite treatment center 11 15 60 83 1 UA

General responses Median, range W p- value Median* Median* (%) LA†

Awareness (rejecting traditional beliefs on snakes and prehospital care of snakebite and medical help of doubtful use and accepting modern measures of pre-hospital care); H0(A): M = M0(33), Ha (A): M>M0(33) a. Respondents (n = 72, Table 3) Awareness (rejecting traditional beliefs on snakes and prehospital care of snakebite and medical help of doubtful use and accepting modern measures of pre-hospital care); H0: M = M0(17), Ha: M > M0(17) b. Scores (responses to 32 test questions, Table 3)

41,2-71 365 0.03 33 46 MA

Unawareness (accepting traditional beliefs and medical help of doubtful use and rejecting modern measures of pre- hospital care); H0(UA): M = M0(24), Ha (UA): M>M0(24) 31,1-70 358 0.04 24 33 Item nonresponses (Not answered to questions), H0(IN): M = M0(0), Ha (IN): M>M0(0) 0,0-2 NA NA NA NA

18,9-26 1387 0.035 17 53 A

Unawareness (accepting traditional beliefs and medical help of doubtful use and rejecting modern measures of pre- hospital care); H0: M = M0(13), Ha: M > M0(13) 14,6-23 1472 0.002 13 41 Item nonresponses (Not answered to questions), H0(IN): M = M0(0), Ha (IN): M>M0(0) 0,0-2 NA NA NA NA

a. Scores for awareness (rejecting traditional beliefs and medical help of doubtful use and accepting modern measures of pre-hospital care, n = 32) Demographics All respondents 18,9-26 1387 0.035 17 17-Inf 14,6-23 1471.5 0.002 13 13.5-Inf Male (one- tailed) 18,14-19 460.5 0.008 16.5 17-Inf 14,6-18 350 0.008 13 13-Inf Gender Female (one- tailed) 18,9-24 469.5 0.038 16.5 16.5-Inf 14,8-23 405 0.033 13 13-Inf Male and female (two- tailed) - 654.5 0.941 - minus 2-2 - 636 0.901 - minus 2-2 Government run (Public) (one-tailed) 18,14-26 377 0.042 17 17-Inf 14,6-18 386 0.030 12.5 12.5-Inf School type Private (Institutional) (one-tailed) 17,9-23 493 0.006 16 16.5-Inf 14,9-23 490.5 0.002 13 14-Inf Government and private schools (two- tailed) b. Scores for unawareness (accepting traditional beliefs and medical help of doubtful use and rejecting modern measures of pre-hospital care, n = 32) Median, range W (Aware) p- value Median* 95% CI Median, range W (Unaware) p- value Median* 95% CI - 717.5 0.403 - minus2-2 - 579.5 0.471 - minus2-1

In general, respondents demonstrated adequate awareness about procedures in case of snakebite. 92 % (n = 66) reported seeking medical attention visiting a healthcare facility supplied with antivenom in case of snakebite (Table 3d). However, only 15% (n = 11) knew where the nearest snakebite treatment centre to their residential areas. Similarly, 99 % of respondents rejected the belief that all snakes surrounding us are venomous and 92 % refused to seek treatment from traditional healers (Table 3 a,b). 49- 72% of respondents accepted widely recommended first aid measures for venomous snakebites (PIB and LCPI) (Table 3d). We found that more than 46% of the respondents (median = 33, p = 0.030, Table 4.a) were aware of snake and snakebite care based on rejecting traditional beliefs on snakes and snakebite care and medical care of doubtful use, and accepting modern measures of pre-hospital care of snakebite victim. These respondents scored more than 53% (median = 17, p = 0.035, Table 4b). In contrast, more than 33% of respondents (median = 24, p = 0.040, Table 4.a) were unaware of recommended pre-hospital care of snakebite as indicated by accepting traditional beliefs and medical help of doubtful use, and rejecting modern measures of pre- hospital care who scored more than 41% (median = 13, p = 0.002, Table 4b). The level of awareness was not significantly different among male and female students and respondents between public and private schools (Table 5).

Knowledge on identifying snakes and their ecological roles

Students received a median score of >12 (60%), >5 (30%), and >3 (13%) for identifying 58 displayed snake photos (Table 2) correctly as venomous, mildly venomous, and nonvenomous snakes, respectively. Similarly, they received a median score of >3.5 (10%), >18 (78%), and >11 (31%) for their claim of venomous or mildly venomous snakes to be non-venomous, non-venomous snakes to be venomous or mildly venomous, and venomous snakes to be mildly venomous and vice versa, respectively (Table 6).

Symbols and Abbreviation: * = Median Significantly Greater than (After Hypothesis Test); % = Percent; N = Number of Snake Photos Displayed; W = Value of One-Tailed One-Sample Wilcoxon Signed Rank Test. Table 6: Familiarity of students with native venomous, mildly venomous, and non-venomous snakes.

Male and institutional (private) school students had greater score than that of female and Nepal Government run (public) school students for identifying venomous and mildly venomous snakes, respectively. The median scores for identifying non-venomous snakes were similar among males, females, and private and public (institutional) school students (Figure 5). But, median scores for the incorrect responses to displayed snakes were greater among female and institutional school students (Figure 6).

A total of 29% (n = 21) informants replied knowing the need for snake conservation, but only 10% (n = 7) knew the ecological roles of snakes in some extent. A total of 79% (n = 57) informants replied worshipping of snakes during Naagpanchami (a snake worship day in Hindu culture worldwide). But, only 11% (n = 8) considered snakes as a friend of farmers.

Click to enlarge

Click to enlarge

Demographics of the Respondents

We surveyed 72 randomly selected students attaining eight (58 %, n = 42 students) and nine (42 %, n = 30) grade in public (n = 33 students, 46%) and private schools (n = 39 students, 54%) with a mean age of 14 years (range = 12-18, median = 14, SD = 1.1, SEM = 0.13). Among all informants, 49% were males (n = 35, males and females ratio = 0.95) and 88% (n = 63) were Hindus, 10% (n = 7) were Muslims, and 3% (n = 2) Buddhists (Figure 7, Table 1).

Click to enlarge

Use or killing of snakes in Nepal

Categories of using snakes in the lowlands of Nepal known during this study were similar to usage mentioned in a similar study carried in Mexico [39].

Snake as medicine: In Nepal, snakes are used for ethnomedicine only to a minor extent. We found the use of a cobra’s fat to supposedly cure a wound by a single ‘Yadav’ family in Rajdevi Municipality in Rautahat District, southcentral lowlands of Nepal. Pandey, et al. [3] reported use of fats and gall bladder of pythons and fats and intestine of cobras for ethnomedine in the vicinity of Chitwan National Park, southcentral Nepal. Similar ethnomedicinal use of snake product is practiced widely. In India, fat from Pythons is used to cure ailments in joints and muscular pain [40, 41] and flesh from cobras is used in various folk medicine [42]. In Latin America, fat from boa (Boa constrictor) is used in traditional folk medicine [43]. In China, snake gallbladder is used to treat damp and fire, relieve a cough, reduce phlegm, and improve eyesight. It is essential to document any snake products being used for the modern or ethnomedicinal use in Nepal more comprehensively including higher lands.

Snake as food: Snake is not favored bushmeat in Nepal although snakes have served as an important source of protein for human populations in certain parts of the world [5]. Only three percent of the informants had been eyewitness of snake meat being consumed in the lowlands of Nepal. In support of our findings, Pandey, et al. [3] reported fractional use of snakes as bushmeat particularly by ethnic people such as ‘Tharu’, ‘Mushahar’, ‘Kusunda’, and ‘Newar’ in the vicinity of Chitwan National Park, southcentral Nepal. Shah [13] also reported rat snakes being used for food by some ethnic people of Nepal. In Camboida, large number of Enhydris enhydris are exploitated for bushmeat [44]. It is likely that youths who experienced eating snake meat while they worked abroad can relish the bushmeat. This depicted potential use of snakes for high protein rich food as well as medicine in Nepal. However, the magnitude of the exploitation and consumption of snakes as bushmeat varies from country to country depending on cultural ban, influence by governmental policies on wildlife hunting, and socio-economic status of community people. In Nepal, mostly ethnic people and those from lower socio- economic background depend on snake-meat for the supply of animal protein infrequently.

Commercial use of snakes: Despite widespread commercialization of snakes [10, 12, 45], there is a general lack of information about the snake use in Nepal, which makes it difficult to evaluate magnitude and impact of use of snakes on their populations. The Nepalese Government has allowed commercial use of snakes recently (https:// thehimalayantimes.com/kathmandu/new-rules-allow- commercial-farming-reproduction-of-wild-animals/; accessed in 17 Feb 2020). The impact of snake farming policy should be evaluated periodically.

Use of snakes by charmers: Display of snakes by charmers in Nepal is occasional. Snake charming activities are also decreasing in India due to political ecology of snakes [46]. In Nepal, Natuwa tribals mainly adhere to snake charming (cited in: https://www.youtube.com/watch?v=5rwr7y3CvkY, “Sajha Sawal”, BBC Nepali Service, accessed in 13 Oct 2019). Nepal Government authorities are planning to shift their occupation without proper studies. It is evident that some of the snake charmers illegally involved in supplying live venomous snakes and/or their venom to neighboring countries of Nepal.

Use of live snakes for worshipping: Like worshipping live snakes as a symbol of god in “Nagpanchami” (the serpent festival) in Nepal, they are worshipped in India as well [10].

Use of snakes for public display in museums: Although dead snakes are deposited in different parts of Nepal, only 1% (n = 1) of the respondents reported using snakes to display in museum. This reflected no exposures of school students to museums in Nepal where snake collections are maintained.

Use of snakes in biological study in laboratory: From this study, it is evident that snakes are rarely used in biological laboratories in Nepal.

Use of snake as pet: Although snakes are kept as pets (e.g., Asian Ratsnake Coelognathus radiatus) and traded widely [31, 47, 48], we found only an instance of keeping a python as a pet in the midwestern hills of Nepal during our study.

The tendency of killing snakes in Nepal and other regions worldwide is markedly similar although we found a smaller number of direct killings of snakes within the recent last year. We found 44% students who observed/came to know killing of snakes by locals considering them as a harmful animal. More than 50% of the respondents from Mexico mentioned killing snakes because they considered them to be venomous and aggressive animals [39]. 49% students from northeastern Brazil showed negativity towards snakes due to potential risks and myths associated with snakes [49]. 43% of locals from Araponga region of southeastern Brazil [50] responded to kill snakes. 49% respondents in Azerbaijan killed a snake or had seen someone killing it [51].

Although our students showed sympathy towards snake conservation (29%), we found witness of killing snakes by locals showing stunt and responses of large number of informants for killing displayed snakes and those encountered at different defined localities (i.e., crop fields, roads, premises of house, and indoor) (Figure 3,4). Such dual public attitudes towards snakes are common in southcentral Nepal [3] and elsewhere [52]. The snake killing attitude was also common among medical students from the middle hills of Nepal [53]. Overall, we suggest that snakes are at the risks of human-caused mortality in Nepal. This risk might have been intensified by vehicular accidents (42%, n = 30).

Greater knowledge and awareness of people about snakes and their ecological and utilitarian roles decreases the fear of and negative attitudes to snakes. Frequent exposure of people with scientific and environmental educational activities seems to have been influential for the higher tolerance to snakes [27, 50]. Therefore, increasing public knowledge on native venomous snakes, the ecological roles of snakes in general, and primary and secondary prevention of snakebite [26] reduces fear of and negativity towards snakes. This diminishes their attitudes of killing snakes, which in turn, contributes to snake conservation.

Risk-values of native snake species

The risk-values of native snake species (Table 2) can be used to assign priority for the conservation efforts and educational interventions targeted to species at the potential risks of human-caused mortality. The number of interviews per informants per snake species and the number of informants giving information on each species are denominators in the calculations. For the more reliable/ improved estimation of risk-value for each species, one can further adopt re-sampling methods such as bootstrapping. Commercial and subsistence values of each of the displayed snake species (which we could not document due to time and financial constraints) will increase the risk-values for these snake species (Table 2).

Awareness of students about snakes and snakebites

Lower level of awareness of school students on snakes and snakebites (Table 3-5) in this study was comparable to medical students from a medical college in the hills of Nepal [53]. Unlike the satisfactory awareness about seeking medical attention in case of snakebite by 78% respondents and use of traditional treatment of snakebite by approximately 21% of respondents in southeastern Brazil [50], our respondents reported to seek both recommended and non-recommended pre-hospital care practices (Table 3). Such confusion was common among military personnel in southeastern China, too [54].

Although our respondents were more knowledgeable on recommended first aid measures (Table 3) compared with high school students in West Bengal of India [48%, [55]] and snakebite patients referred to Ahvaz Razi Hospital in Iran [56], 93% believed to use a tourniquet in case of snakebite. This confusion with regard to our students could be due to the influence of improper texts about snake and snakebite mentioned in their books [16]. This is supported by the report of similar unawareness among medical students in Nepal [53]. This demonstrates an educational opportunity in Nepal. Since improper practices of pre-hospital care for snakebites increase duration of hospitalization, costs for the treatment, risks of infections, etc. [57], there is a need to improve conditions for people inhabiting regions of Nepal where they are at risk of snakebite.

Knowledge on identifying snakes and their ecological roles

As in southeastern Brazil [57% [50]] and Karnataka, the southern India [59-87%, [58]], students in the lowlands of Nepal were unfamiliar to native snakes. More than 78% (median 18, p = 0.001, Table 6) of displayed non-venomous snakes were claimed to be venomous or mildly venomous by these students, more than 10% (median 3.5, p = 0.005, Table 6) venomous or mildly venomous snakes were claimed to be non-venomous, and more than 31% (median 11, p = 0.31, Table 6) venomous snakes were claimed to be mildly venomous and mildly venomous as venomous snakes by these students. Correct identification of more than 12 (60%, p = 0.021, Table 6) venomous snakes by our students might have been influenced by periodic displays of venomous snakes in national and international television channels and social media, limited information on venomous snakes in school textbooks [16], and transfer of parental knowledge on venomous snakes to their children. The cognizance of venomous snakes by these students was similar to recognizing displayed images by locals in Mexico [39].

Greater scores of males and institutional school students for identifying venomous and mildly venomous snakes (Figure 5), respectively, correspond to the more frequent exposure of students to the extra activities in those schools and exposure of male students to media and extra activities in communities. Like the male students recognizing more snake species than female students did in northeastern Brazil [49], male students scored higher for knowing nonvenomous snakes in this study (p = <0.001). In contrast, we found that female students scored higher for identifying venomous snakes (p = <0.001).

Illogical claim of knowing snakes is prevalent in the southcentral Nepal [3, 59]. Wrong identifications by our students of venomous or mildly venomous snakes to be non- venomous, non-venomous snakes to be venomous or mildly venomous, and venomous snakes to be mildly venomous and vice versa (Table 6, Figure 6) may keep them at the risk of envenomations. They demonstrated a greater intention to kill snakes (median = 14, p = 0.052, 95% CI = 13.99-21.50). A total of 29% (n = 21) informants replied knowing the need for snake conservation, but only 10% (n = 7) knew the ecological roles of snakes to some extent. This gives an educational opportunity in the lowlands of Nepal.

Because of the elusive nature of snakes and their random encounters with humans at night, evening, morning, and day hours, direct observation of snakes involved in bite or snakes killed for different purposes are difficult. According to our informants, it is hard to differentiate kraits from similar looking wolf snakes and banded kukri snakes, cobras from Dhamin, and so on for several locals in other parts of Nepal [3, 23]. Correct identification will prevent needless killing of snakes and this in turn will lessen incidence of snakebites and will contribute to biodiversity conservation.

Our findings of L. aulicus at the greater risks of being killed corresponds to the report of this species being mistaken as krait species at snakebite treatment centers in the lowlands of Nepal by Pandey, et al. [60] and extreme fear of snakes and noticeable ignorance of identifying venomous snakes among students, teachers, and locals in southcentral Nepal by Pandey, et al. [3]. Since there is evidence of treating two proven L. aulicus bite cases in Nepal [60], both medical and non-medical students, teachers, and locals should be made well aware in identifying native medically important snakes in snakebite prone zones elsewhere. A total of 38% (n = 27) informants noticed predators (e.g., eagle, mongoose, cat, peacock, dog, and hen) feeding on a snake while commuting from school to house. Therefore, educating people on the ecological roles of snakes and their natural control by predators through educational interventions can help preventing snakebites.