Use of tandem observations in ambulatory primary care to evaluate physician—nurse teamwork

Ambulatory primary care is defined by care that is, 1) first contact, 2) longitudinal or a sustained partnership over time, 3) comprehensive, taking care of the majority of a patient&s;s health care needs, and 4) coordinated, integrating the care across healthcare settings and specialties ( Donaldson, Yordy & Lohr, 1996 ; Starfield, 1993 ). With the introduction of the Primary Care Medical Home (PCMH) model, primary care systems are being redesigned to better perform these tasks given the increasing complexity of healthcare using information technologies and teamwork structures. Thus, better methods are needed to accurately model the workflows in primary care clinics, especially related to the interactions and effectiveness of care teams. Direct observation of healthcare workers performing clinical care is one method for understanding the actual work performed and the subsequent outcomes of this work ( Battles & Lilford, 2003 ). Typical observation strategies use the approach of pairing a single observer with a healthcare provider; however, the data has limitations. By only observing one provider at a time, the data only reflects how one piece of the system interacts with the whole, which is more reflective of an individual&s;s effectiveness within the system rather than the overall effectiveness of the healthcare system present. Hence, simultaneously observing more than one provider when teams are working together yields data that is more complete and dynamic in nature by capturing the interaction within the system of two providers, who are trying to complete both collaborative and individual tasks. The richness of the data reflects team members who have different roles (e.g. job descriptions) that interact throughout the day to communicate information or coordinate tasks and procedures relative to patient care. This described observation method uses two observers to evaluate the work of and information flow across a physician or physician assistant—nurse team in a primary care clinic. ‘Tandem observations’, conducted jointly by a human factors engineer and a human-factors trained physician, were performed starting at the beginning of the morning or afternoon session of seeing scheduled patients and ending 3–5 hours later at lunch time or the end of the work day. We observed eighteen provider—nurse pairs, six each at three different primary care clinics in Wisconsin for a total of 130 hours. Observation notes were taken free-hand and transcribed by the researchers for analysis. These observations provided a more complete view of the work being performed by the physician—nurse team before, during and after patient care visits as there is often preparatory work done before the patient visit and follow-up work and documentation performed after the visit. We were also much more able to identify information flow between the team members and with the patient, identify failures in information flow, and how these failures were (or were not) recovered from during the patient care session. Examples are provided. Tandem observations were generally well accepted by both patients and healthcare staff.