Diabetes is a major health issue for individuals and health services globally. In New Zealand diabetes is one of the most important conditions routinely seen in primary care settings, and disproportionately affects Maori and Pacific peoples, and those from low socioeconomic backgrounds [1–3]. There is a considerable literature internationally on the management of diabetes and also on communication in primary care consultations [4–6]. However, there are few studies which combine these two topics, whether generally or more specifically in relation to nurse-patient communication.
Optimal management of diabetes in primary care in New Zealand involves patients engaging with the combined and coordinated efforts of several health professionals (GP, nurse, podiatrist, dietician and retinal screening services), with referral to secondary services as needed [2, 7]. In New Zealand, diabetes care is largely funded through CarePlus , a chronic care initiative introduced in 2004 as part of the New Zealand Primary Health Care Strategy . This initiative provides additional funding for primary health organisations to give care for people with high needs because of chronic conditions or terminal illness. The specific aims are to improve management of chronic conditions, reduce inequalities, improve teamwork among health professionals and reduce cost of services for high need users .
Practice nurses have a vital role in the initial management of Type 2 diabetes in primary care, which is largely directed towards assisting patients to understand the nature and possible trajectory of the disease, and to become self-managing. The approach taken by practice nurses involves assessments, goal-setting, and information-sharing about self-management in more extended face to face consultations which inevitably turn to conversations concerning lifestyle, behaviour modification and risk reduction .
However, this approach is, as yet, unfamiliar to many patients who are more used to a problem-oriented medical system which deals with immediate rather than long-term concerns, even in the context of long-term conditions. A related problem arises from the complexity of diabetes and diabetes management. Previous studies have reported that misunderstandings and mismatches in expectations between patients and health professionals may arise due to the different priorities of disease management protocols in diabetes care and a patient centred consulting approach [11–14]. This is compounded by the fact that there may be very few symptoms at the time of initial diagnosis, or ones which are of gradual onset.
From a nursing perspective, an important part of diabetes care lies in eliciting and imparting information in a timely way through face-to-face consultation. Nursing consultations are supported by New Zealand guidelines for best practice in diabetes care, including various checklists  which have been widely disseminated and taken up by health practitioners. The guidelines are intended to inform primary care practitioners on targets of treatment and management and therefore go some way to support delivery of a comprehensive and consistent service.
It is easy for practitioners to regard the guidelines with their associated checklists as prescriptive of what must be accomplished in a given consultation or series of consultations, rather than starting from the patient’s existing knowledge and immediate concerns . Many patients have difficulty assimilating the knowledge that diabetes can lead to significant adverse effects on many different parts of the body. Similarly, making changes to lifetime habits of diet and exercise is not easily accomplished and this is especially so when there is no immediate apparent need. As a consequence, close adherence to guidelines by means of a checklist or e-protocol may become unhelpful in individual consultations [15, 16].
Making checklists the focus of a consultation is also at odds with how nurses prefer to relate to patients more generally. Everyday nursing work, including diabetes management, is mediated through talk , and there is increasing recognition in the research literature that nurse-patient encounters have both a content component and a relational component, both of which are important [18–20]. However, in spite of an increased awareness of the need for patient-centred approaches, research suggests that in practice such approaches do not occur as often as might be beneficial [20–24].
It is well-accepted that the quality of communication between health practitioners and patients can make a significant difference to health outcomes [20, 25]. There is also good evidence that practitioners can improve the quality of care they provide by better understanding the consultation process and focussing on effective communication both within consultations and over time [20, 26–29].
In teasing out exactly how communication affects health outcomes, Street et al.  describe what they call the “proximal outcomes of the interaction (e.g., satisfaction with care, motivation to adhere, trust in the clinician and system…) that could then affect health or that could contribute to the intermediate outcomes (e.g., adherence, self-management skills, social support) that lead to better health.”p.297.
The study reported here set out to examine the actual talk and perspectives of nurses and patients who were newly-diagnosed with diabetes in order to describe the features of effective interaction and to identify areas for reflection and possible improvements to practice. Criteria for effective communication included both: (1) video evidence of smooth comfortable interaction at an interpersonal level within the consultation, as demonstrated by verbal and non-verbal cues and (2) recorded evidence over the 6 month period of some retention and appropriate recall of information exchanged.