OBJECTIVES: To examine inequalities in health service utilisation in urban Pakistan. This analysis investigates how household economic status, duration of illness and distance to a provider influence health service utilisation in Pakistan. METHODS: The study uses data from the Pakistan Socioeconomic Survey (PSES) and analysis is based on 1,407 individuals who belong to 855 urban households. Health care providers are classified into public hospitals, other public providers, private doctors/clinics and other private providers. Household economic status is measured by a wealth index constructed using data from the survey on ownership of durable assets and housing conditions. Principal components analysis (PCA) is used to construct the index. Multinomial logistic regression is used to investigate the effects of various characteristics of individuals/households on health service utilisation in Pakistan. RESULTS: Overall, 79.6% of those reporting any health complaint sought health care. A large gap in health service utilisation exists between poorest patients (60.5%) and richest patients (84.9%). Almost three-fourth patients visited private providers; 57.3% visited private doctors/clinics and 15.7% visited other private providers. The remaining one-fourth patients visited public providers (19.7% visited public hospitals and 7.2% other public providers). Multinomial logistic regression reveals that poorest patients are significantly more likely (p<0.10) to visit public hospitals whereas patients of poorest, poor, middle and rich households are significantly less likely to visit private doctors/clinics compared to members of richest households controlling for other factors such as education, occupation, duration of illness,  distance to a provider and residence. An additional day of illness significantly (p<0.01) increases the likelihood of visiting public hospitals and private doctors/clinics. The distance travelled to visit a provider shows a significant positive (p<0.01) and negative association with visiting public hospitals and other private providers respectively. CONCLUSIONS: Large gaps exist in health service utilization in urban Pakistan.