The purpose of this study was to identify the diagnostic reasoning strategies deployed by beginning and experienced nurses when reasoning about a clinical problem. Sixty student and registered nurses took part in a simulated nursing diagnostic task. “Think-aloud” protocols enabled the identification of clinical information, clinical concepts and diagnostic hypotheses used by nurses along with underlying reasoning strategies. Analyses revealed four diagnostic reasoning groups discriminated by differences in the timing and breadth of diagnostic activity and outcomes, and in the use of conceptual and clinical information. Registered nurses were more successful in this process than student nurses. Implications for nursing education and professional development are discussed.

Aims

The purpose of this study was to identify the diagnostic reasoning strategies deployed by beginning and experienced nurses when reasoning about a clinical problem. The processes underlying proficient diagnostic reasoning are both complex and highly metacognitive, involving not only the screening and co-ordination of potential diagnostic information, but also integration of this information into a coherent and meaningful form if appropriate nursing diagnoses are to be generated. Previous research into diagnostic reasoning has demonstrated that experts and novices differ profoundly in the way they approach and process diagnostic information, and thus, unsurprisingly, differ significantly in the accuracy and appropriateness of their diagnostic outcomes. These differences suggest that there may also be fundamental differences in the kinds of cognitive schemes used by more expert-like nurses when confronted with a diagnostic reasoning task compared to those used by more novice-like nurses when confronted with the same task. We suggest that these schematic differences are likely to be located in the depth and generality of the underlying clinical knowledge available to the nurse, and in the patterns of strategy use underlying the accessibility and deployment of that knowledge. Using a think-aloud procedure in a simulated clinical environment, we aimed in this study to determine the extent to which novice and experienced nurses could be discriminated on the basis of their strategic approaches to a clinical reasoning task.

Methodology

Sixty student and registered nurses took part in the study. All testing was conducted individually. Nurses were given a written case history of a “patient” who was to be the subject of a clinical interview and were told they would be viewing a short video-tape of the (simulated) clinical interview. They were instructed to think about a primary nursing diagnosis for this patient. They were to then verbalise their thoughts whilst reading the notes and watching the interview. These thoughts were audio-taped. Transcribed audio-tapes were then analysed for the presence of relevant clinical information and associated clinical concepts as well as for evidence of diagnostic hypotheses. In addition, verbal protocols were analysed for evidence of strategic behaviours underlying the associations drawn between clinical information, concepts and diagnoses. Finally, the analysed protocols were mapped on to schematic diagrams in order to provide a visual representation of the diagnostic process.

Findings

The 60 schematic diagrams were analysed for patterns in the way clinical information, concepts were used and for how diagnostic outcomes were generated. Analyses indicated that four “diagnostic groups” could be discriminated on the basis of the clinical information and concepts used, the types of cognitive strategies used, and the number and appropriateness of diagnostic hypotheses generated. These four groups were labelled, in descending order of sophistication, as expert, competent, intermediate and novice diagnostic reasoning groups. The expert diagnostic reasoning group (n=14) and the competent diagnostic reasoning group (n= 12) consisted entirely of registered nurses. The intermediate diagnostic reasoning group (n=16) consisted of 12 student nurses and 4 registered nurses. The novice diagnostic reasoning group (n=18) consisted entirely of student nurses.

In general terms, the data revealed a greater use of conceptual knowledge as organising and interpretative schemes for clinical information amongst the more sophisticated expert and competent groups, and a greater focus on clinical information by the less sophisticated intermediate and novice groups. Groups also differed in both the timing of diagnostic hypothesising and in the accuracy and appropriateness of those diagnoses: the greater the clinical sophistication of the nurse, the greater the likelihood of earlier diagnosis and the greater the use of confirmatory strategies. For the novice group in particular, reasoning strategies were significantly constrained by the ongoing need to clarify the meaning of clinical information as it emerged.

Theoretical and educational significance of the research

We proposed that the processes underlying effective diagnostic reasoning in nursing were both complex and highly metacognitive. The data revealed a pattern of responses amongst participants that support the view that expertise may be associated with the use of more sophisticated diagnostic schemes than is the case with the more novice-like reasoning. One critical aspect of the expert group is the reliance upon conceptual rather than purely clinical knowledge as the basis for diagnostic decisions. This undoubtedly emerges from the automaticity with which incoming clinical information is able to be contextualised. A strategy of pattern recognition is only effective where the underlying knowledge is at such a point of automaticity that memory space is available for associations (reasoning strategies) to be generated. The driving schemes of the more expert nursing diagnostician, then, allows for the use of efficient top-down processes. Amongst the expert group, this was evident in the accuracy and embraciveness of their diagnostic outcomes. For the competent group, this allowed for the generation of accurate, but less comprehensive diagnoses. By contrast, the attentional focus of the more novice-like nursing diagnosticians appeared constrained to “knowledge acquisition”, suggesting that the task was completed on the basis of incomplete diagnostic schemes. For these nurses, the lack of conceptual understanding of the clinical information encountered limited diagnostic activity to either symptom repetition (novice group) or the generation of overly narrow nursing diagnoses Intermediate group).

There are questions emerging from these data to do with understandings of what are reasonable expectations of student and perhaps less experienced registered nurses. The fusion of theoretical and clinical information, a necessary prerequisite to effective diagnosis, may require ongoing professional development activity – it may simply not be adequate to assume that all nurses will with time reflect the attributes of the expert diagnostician. The metacognitive attributes of expert diagnosis are complex and do imply a different way of thinking about clinical tasks. In a controlled task where all participants were exposed to common information, differences in outcome become best explained by reference to the metacognitive basis of effective clinical decision-making. It would seem a reasonable outcome from the data to suggest the inclusion of domain-related higher-level metacognitive instruction in both pre-registration programmes and post-registration professional development. In a profession where lives may be at stake, the randomness of experience may not be enough.