Making evidence from scientific studies available to clinical practice has been expected to directly improve quality of care, but this expectation has not been realised. The notion of quality of care is complex, and quality improvement needs medical, contextual, and policy evidence. In primary care, research is needed that takes into account the specific characteristics of its population and the presentation and prevalence of illness and disease. The context of the doctor-patient encounter plays a major part, and needs better understanding. At the policy level, issues of equity must be addressed. The knowledge base for family practice must be expanded by integration of multiple methods of comprehension, so we can bridge the gap between evidence and practice.