Journal of Basic & Clinical Medical Sciences2959-92022959-92102022-09-1351334310.58398/0002.000035articleMuneeb GhummanAmbreen SalmanNiaz Ali KhanKhadija Sadaf ZehraMuhammad Abdullah AhmadDepartment of Public Health, University of Health Sciences, PakistanSchool of Creative Arts, The University of Lahore, PakistanWorld Health Organization, PakistanSemnan University of Medical Sciences, IranLiaquat National Hospital and Medical College, PakistanUniversity College of Medicine and Dentistry, PakistanCigarette smoking is a leading preventable cause of death in Pakistan, yet adults’ knowledge of its specific harms and attitudes toward pack warning labels are incompletely characterized. We conducted a comparative cross-sectional survey of 406 adults (260 smokers and 146 nonsmokers) in Ichhra, a densely populated locality of Lahore, Punjab. An interviewer-administered questionnaire adapted from a validated instrument captured sociodemographic characteristics, smoking behavior, a 14-item health-risk knowledge battery, and warning-label attitudes. Knowledge scores were classified by the modified Bloom cutoff, and internal consistency was assessed using the Kuder–Richardson 20 (KR-20) and Cronbach’s alpha coefficients. Categorical characteristics were compared with the chi-square test, weight with the t test, and the knowledge score with the Mann–Whitney U test, applying a Bonferroni correction across the 14 item comparisons. Independent predictors were modeled with binary and ordinal logistic regression, and dimensionality (whether each scale measures one or more distinct constructs) was examined by exploratory factor analysis. The sample was predominantly male (97.8%); smokers consumed 7.6 (SD 3.9) cigarettes daily for 6.2 (SD 4.8) years, and 52.7% had attempted to quit. The knowledge and warning-label scales showed acceptable reliability (KR-20 = 0.71 and 0.72), and the quit-barrier scale had good reliability (Cronbach’s alpha = 0.84). Nonsmokers knew more than smokers did; six items remained significant after the Bonferroni correction, all favoring nonsmokers. Good knowledge was more common among nonsmokers (72.6% versus 53.1%; p = 0.001). After adjustment, smoking (adjusted odds ratio [aOR] 0.42, 95% confidence interval 0.26 to 0.68), a rural background (0.45, 0.28 to 0.71) and lower personal income independently predicted poorer knowledge, whereas a graduate education level predicted being a smoker (3.48, 2.13 to 5.68). Among smokers, feeling concerned about smoking predicted quitting attempts (2.13, 1.26 to 3.60). Health messaging should prioritize smokers and people with a rural background and should address sex-specific harms.