OBJECTIVES: Food safety is regarded as an integral public health concern. Socio-economic status of people is being strongly influenced by food poisoning in developing countries. A survey was carried out to evaluate the knowledge, attitude and practices of people living in District Swat, Khyber Pakhtunkhwa, Pakistan about food poisoning. METHODS: Paper based questionnaire were used as a tool to collect data from general population in seven tehsils of Swat district, Khyber Pakhtunkhwa, Pakistan. One thousand respondents were recruited by using convenience sampling method from bus stops, terminals, colleges and universities. RESULTS: Majority of the participants (38.2%) were in between 18 to 27 years old. Out of 1000 participants, 73.4% were male, 26.6% were female, 84.2% were rural and 15.8% were urban. More than half (60.6% and 96.5%) of the respondents knew that microbial toxins are heat resistant and unwashed vegetables and fruits are extremely risky for health respectively. Out of total, 36.4% respondents were agreed that raw milk is healthier and nutritious than pasteurized milk. 14% of the respondents thought that raw food can be placed next to the cooked food in refrigerator. 98.8% participants believed that food hygiene training can be very beneficial in reducing food poisoning. More than half (55.6%) often wash their hands with soap before preparing foods. 34% participants often look for expiry date before buying food items. Statistically significant association was observed between occurrence of food poisoning with gender (0.017), education status (0.001) and geographical location (0.032). CONCLUSIONS: The enrolled participants were having necessary knowledge about food poisoning but their practices need to be improved. Certain behaviors such as not washing hands with detergent before preparing and eating food are persistent. Customs, cultural and habitual attitudes are the actual dynamics affecting these behaviors. Adequate training may improve their knowledge, attitude and practices towards food poisoning.