In a culture of evidence-based practice and a striving thereby to produce best practice statements (i.e. research-based practice based on hierarchies of evidence: Thompson 2003), this research looked to describe observed nursing behaviour with respect to the first 24 hours of general postoperative care.
Although this may seem a worthy venture, this research, though not without its merits, had limitations. Not that it is necessarily hard to observe practice, and the research design of non-participant observation was well thought through, nor that it would then be hard to produce various descriptions of what nurses actually did. What was missing, of course, was any sense of why they did as they did or any sense of an interpretive analysis of the significance to the nurses or others of their actions. In defence of the research design, this was not the purpose. It was explicitly only to describe, but to what purpose? Descriptive inductive research seeks to articulate empirical data and subsequently provide conceptual explanations of what is occurring, i.e. substantive theory (Glaser & Strauss 1967). This contrasts, of course, with an exposition of what ought to be and the generation of prescriptive theory. The author rightly acknowledges the prevalence of ‘myths, assumptions and differing perceptions’ as to what nurses ought to be, or are, doing. The simple purpose of this research was to describe current practice, no more, no less, however tempting. The imperative of providing solid and transparent rationales for nursing practices as part of clinical governance, is the driver behind this particular phase of a larger study exploring the relationship between routine observations and postoperative complications.
At the outset, the author identifies two different models of postoperative practice: the first being the strict adherence to predetermined protocols in the first 24 hours and the second, a model that allows for the use of clinical judgement. However, although the selected sites were stated to use each of the models, it was never made entirely clear which model the observed practitioner might be following. The type, frequency and duration of observations, is, arguably, only of significance if it can be accounted for against a specific model of practice. By that, it is meant that a good judgement call for any one patient might have underpinned minimal, average or wider ranging and frequent observations. What, therefore, is the reader to gain from the research findings given?
Before appearing too dismissive, there were some interesting observations. Of concern, and rarely commented upon in the literature, was the finding of the significant period of time between leaving the recovery area and the first occasion for discrete observations in the ward setting. If postoperative observations are, as a matter of common sense, if not extensive level 1 evidence, driven by the goal of prevention or early detection of potentially serious complications, this gap in surveillance at a time when the patient, by definition, is still vulnerable, could be seen as a matter for disquiet.
Another finding worthy of note is the apparent increase in surveillance, whether by design or chance, for patients in shared, as opposed to single, accommodation. Should such a finding influence decisions as to ward geography – or vice versa?
The concept of invisible work, arguably more often associated with the emotional labour of nursing (Smith 1992, McQueen 2000), is identified as a limitation of the study as no data obtained reflected this. More correctly, the author is perhaps referring to the minimally visible observations made by the practitioner in terms of the casual glance or quick scan of the patient and their environment. This form of behaviour sits more comfortably with the model based on clinical judgement as the minimally visible observation may presage decisions as to whether to engage in the more measurable postoperative behaviour.
In discussing the findings, the author makes some useful comments as to what drives this postoperative nursing practice. This includes the utility of routines vs. clinical judgement, the notably paucity in the monitoring of how the patient was actually feeling and, mercifully, the high priority put on assessing pain. This research was not addressing intervention as such but it had to be a disappointing finding that, despite the focus on assessing pain, the assessment of nausea (so often part of the pain and pain relief experience) was so neglected (<20%). Although one was not informed whether this failure was equally distributed in both hospital sites, the complication of vomiting was noted in only one hospital site. One is left to ponder as to whether this vomiting might have been due to the employ of effective analgesic therapy without considering their emetic potential.
To return to the issue of how the patient was actually feeling, this is an interesting observation. Perhaps this was part of the minimally observable assessment embedded in the silent glance or passing touch. However fascinating this conjecture might be, this was not the main purpose of the research. Clearly stated was the fact that the data to be reported were about ‘frequency and patterns not individual practice’ and to look at the distribution of time spent with patients in the postoperative period. The question for the researcher becomes is how far these findings will actually move practice towards the optimal.