The complexity conundrum: Nursing interventions for complex health problems

David R Thompson
Faculty of Health Sciences, Australian Catholic University, Melbourne VIC

Alexander M Clark
Professor, Faculty of Nursing, University of Alberta, Edmonton, Alberta, Canada

PP: 277 - 280

Article Text

What is nursing if not complex? Nurses and nursing exist and operate not in isolation but in complex contexts and as complex parts of complex health care systems. Both are focused on caring for patients or clients with complex health problems, which can be addressed through complex, multi-component interventions. With these multiple facets of complexity, nursing is not then only complicated, but is the quintessential complex intervention (Richards & Borglin, 2011).

Complex interventions

Yet, it is easy to fall into vague, circular and even nihilistic positions when describing and researching nursing's many layers of complexity. Though arguments for complexity are in vogue and are often supported by general appeals to the various complexities of the world, conceptualization of what this complexity actually consists of and evidence to support it can appear scant (Paley & Eva, 2011). Proponents of qualitative research frequently purport to explore this complexity, but too often generate findings that are excessively formulaic (Hammersley, 2008; Thorne, 1997).  While the complexity movement challenges the overly neat and de-contextualized prescriptions for nursing care evident in clinical guidelines, meta-analyses and randomized trials (Pawson & Tilley, 2007; Pawson, 2006), might it leave the discipline and its professionals impotent to change a world so complex?

Indeed, many of the concerns about the complexity-turn in nursing and health care are well placed. However, rather than reflecting inherent problems with the complexity-turn, such shortcomings more likely reflect the relative youthfulness of the application of complexity to health care. Where have we come in this movement towards complexity thus far?

Health care interventions or programs, including those enacted by nurses, can be described as treatments, procedures or actions (Chien & Norman, 2009; Sidani & Braden, 1998) but are increasingly considered as 'complex interventions' since they usually comprise several components, which may act independently and inter-dependently (Medical Research Council, 2008; Campbell et al., 2007), and rely on active participation of both patients and health professionals (Campbell et al., 2000).  These components include not only the behaviors and their parameters and methods of organising and delivering them of the care providers, but also variations in the client group or population that receive the interventions and the nature of the intervention itself from site-to-site and/or recipient-to-recipient, as well as the number and variability of potential health outcomes (Craig et al., 2008).

Yet, in the midst of such complexity, though nursing interventions vary tremendously, they are often poorly conceptualized and described with little rationale or theory to underpin them.  For example, patient teaching, information-giving, education, and counseling are terms often used synonymously used to describe a single intervention, even though these are fundamentally different components with varied complexity (Abraham & Michie, 2008). Factors such as the strength (dosage) of an intervention and the resources required for delivering it (e.g., personnel, setting and equipment) are rarely described, nor are extraneous variables such as contextual or environmental features and patient characteristics, which can significantly influence the intervention process and consequently their health outcomes (Sidani & Braden, 1998).

Further, many of the interventions that nurses develop and implement are not only complex in themselves but also are provided by organizations, which themselves reside in wider place-related and socio-cultural contexts that constitute discrete systems. These systems are characterized by change, adaptation and uncertainty (Mohammadi, 2010; Pawson & Tilley, 1997).  As such, patients (being the complex bio-psychosocial systems themselves) are part of a series of nested systems that each can exercise influence in and across system boundaries. Single factors, such as a patient's age or sex, do not gain influence over health outcomes inherently or via simplistic direct effects, but due to the cumulative effects of interpersonal, social, cultural, and organizational factors that combine to more likely render discrimination. In this ontology, the degree of complexity of the nursing intervention and, crucially, its outcomes are influenced by the interplay of the components of the intervention (and their interactions); the values and behaviors of its providers and recipients; its organizational elements, and elements of the context in which it is provided (Medical Research Council, 2008). As such, people (patients and professionals), places and packages of care co-create effects of the intervention provided.            

The implications of this ontology for research methods are wide-ranging. Methodological issues in evaluating complex healthcare interventions in nursing have been increasingly emphasized (Forbes, 2009; Blackwood, 2006). Although the Medical Research Council (2008) framework is useful and applicable, a key challenge remains the use of an appropriate research design.  For example, 'Can complex nursing interventions be evaluated using randomized controlled trials?' (Richards & Hamers, 2009a, 2009b; Rolfe, 2009; Thompson, 2004; McCormack, 2004). A checklist of study protocol development deemed to have potential for reducing bias has been proposed, covering six areas: study concept, definition of intervention, selection of comparisons, randomization or allocation, integrity of intervention and ascertainment of outcomes (Campbell-Yeo et al., 2009). More pointedly, should such trials designs be complemented with qualitative methods (Lewin et al., 2009), or even replaced altogether with interrupted time series and action research or case studies (Seers, 2007)?

View references

Navigate: Archives | Contents | Top of page

Search

Translate

Related Publications from eContent Management