The paper by Gardner et al. (2010) in this issue of JAN has added to the relatively sparse literature addressing the actual activities performed by nurse practitioners and other registered nurses. A related study of neuro-rehabilitation nurses’ work activities was published in an earlier issue of JAN (Williams et al. 2009). The importance of these two studies must be underscored because the findings help nurses, other healthcare team members and the general public better understand what nurses do as they provide services to human beings with diverse health conditions across the life span. I have for many years been interested in the work activities of nurse practitioners, with special concern for how nurse practitioners’ work activities differ from those of physicians and other healthcare team members. The findings of studies such as Gardner et al.’s help us to unpack the ‘black box’ of nursing practice, a metaphorical box into which nurses’ work activities are placed only in the broad nursing process phases of assessment, planning, intervention and evaluation. Gardner et al. expanded the categories to direct care activities, which encompasses all four phases of the nursing process; indirect care activities; service-related activities; and personal activities, and identified specific activities within each category. The study finding that nurse practitioners spent as much time on service-related activities as direct care activities is troubling. As Gardner et al. noted, nurse practitioners in some counties are not always able to perform more direct care. This situation may be due to restrictions on the scope of nursing practice imposed by physicians and politicians who do not fully understand what nursing really is, what nursing practice should be and the positive impact that the full scope of nurse practitioner practice can have on the health of the public. However, it is, I submit, unfortunate that Gardner et al., like many other clinicians and researchers, singled out nurse practitioners’ inability to prescribe drugs as a ‘significant barrier to [their] practicing according to the dimensions and scope of their role’ (p. 2167). Nurse practitioner prescribing of drugs has long been contested by physicians, at least in the USA. One solution to this ongoing issue is to limit prescribing to pharmacists and pharamacologists, who have in-depth knowledge of drug actions and interactions. If they were to collaborate with nurse practitioners, physicians and patients, it is likely that drug prescriptions would be clearly consistent with each patient’s particular health-related beliefs, values and lifestyle. The rigorous work sampling methodology used by Gardner et al. is a relatively new approach to identifying the work activities of nurse practitioners. Indeed, my search of the Cumulative Index to Nursing and Allied Heath Literature (CINAHL) conducted in July 2010 yielded just four citations for the search terms, work sampling and nurse practitioners. A special feature of their methodology is that the work activities sampled were derived from established nurse practitioner competencies. I encourage researchers who are interested in refining the work sampling methodology to focus on measurement of the health-related and cost effectiveness of each activity, as well as measurement of time spent performing each activity and the frequency of performance of each activity across all encounters with a particular patient. As research about nurses’ work activities advances, researchers may want to begin to uncover the reasons why a nurse performs certain activities in certain situations. I believe that that the why of practice comes from the implicit or explicit conceptual model of nursing that guides each nurse’s practice, as well as from more specific middle range theories. When those theories are empirically and pragmatically adequate, they serve as the scientific evidence for evidence-based practice.