Tuesday 7 October 2014

Context, complexity, mixed methods and multidisciplinarity

Posted by Emily Henderson

As the new Research Associate lead on Fuse’s Complex Systems Research Programme, I was funded by Fuse to attend a training course on Evaluating Complex Public Health Interventions in Cardiff, 25-27 June. It was delivered by our partners in DECIPHer (Development and Evaluation of Complex Interventions in Public Health), one of five UKCRC Public Health Centres of Excellence (PHCoE). As we know from Prof Simon Murphy’s lecture, presented at the Complex Systems Research Programme’s inaugural event last May, DECIPHer have a successful record of doing localised, translational research. So it struck me and Prof David Hunter (Programme Leader, Complex Systems Research Programme and Fuse Deputy Director) as a great opportunity to exchange experiences with other PHCoE.

As a mixed methods medical anthropologist, I was hoping on this course I would 1) make new and interesting acquaintances, which I did, and 2) learn about how the diversity of methods can be best used in addressing complex public health issues. The three-day course, which followed its two-day counterpart on Developing Complex Public Health Interventions, covered: process evaluations, feasibility studies, large-scale effectiveness studies and alternatives to randomisation and data linkage. The DECIPHer approach was clearly outlined as adopting 1) the MRC Framework on developing and evaluating complex interventions; 2) Ecological Thinking, 3) Co-Production Models and 4) Complex Systems Thinking. And echoing the MRC Framework on process evaluation - coming to a theatre near you - that, despite the range in methods that can be adopted, there is “no hierarchy of evidence”.

This was welcome news - given the diversity of methods adopted within Fuse, e.g. embedded ethnography, realist evaluation, economic evaluation of trials - excepting that the focus of the training was exclusively on trials. In certainly what was not an unusual experience for me, I was the odd-(wo)man-out at this event (emphasis on odd). But this lonely position has its rewards, as participant-observer:

“The magic of anthropology is precisely its ability to self-critically live with the almost schizophrenic contradiction of adhering to two worlds simultaneously”. Messac et al. Soc Sci Med 2013. p185
 
It seemed my chummy classmates were by-and-large RCT researchers, and many told me they left the course wanting to re-read the notebooks to apply what they learned to their research. Pfft, over-achievers. The task I left feeling that I needed to address, was to consider in greater depth this question:

‘In adopting socio-ecological models, how can we as public health researchers, if at all, adequately consider context?’

Just last week, I was asked by a medical anthropology PhD student from the University of California Los Angeles to use my PowerPoint slides as a complement to a podcast of a talk I gave at the University of Oxford, which, dear reader, I wouldn’t shamelessly draw your attention to here. This bright and motivated student wanted to illustrate to her undergraduates how anthropologists can address “real world” problems using mixed methods approaches.

 Induced to dig out my own mixed methods lecture (delivered for Durham’s Methods and Analysis anthropology module), I was reminded of the premise I taught: *only* use mixed methods when the sum of methods reveals more than the methods as parts, or some such or other. I used Durham City’s biennial Light Festival as an analogy of how mixing unrelated media (a cathedral, a slide projector and spooky music) can enhance one’s experience of learning about Durham’s history. But why the proviso “only”? Why not have a light festival every night? Indeed Durham City, why not? Because mixed methods approaches also have many drawbacks. They require: added time, added expertise and multidisciplinary working, to name but a few, not to mention the added funding all this entails. 
 
'Lumiere' - Durham Light Festival
 
No, it’s not possible to create time - we as academics know this all too well - but somehow the latter point - multidisciplinary working - seems more challenging. The debate on ‘Realist RCTs’ for example raises the important issue of the researcher’s epistemological lens. The what? That is, the researcher’s understanding of knowledge and how knowledge is constructed. Do you believe reality can be objectively measured, or rather that reality exists only as a subjective experience, or something in between? Disagreements on this fundamental of doing research can make combining methods - for example embedding ‘qualitative’ methods within trials to provide context - a challenge and at times implausible. However, as with epistemological views, there is a continuum of mixed methods approaches, including triangulation - no, not strangulation - where methods complement each other but retain their disciplinary roots.

Given the range in my own multidisciplinary background, from neuroscience to biocultural anthropology to health policy, my tendency is to at least consider, albeit critically, the range of approaches to researching humans and human health. As social animals we are, after all, complex beings (or at least we’re quite happy to get ourselves into a muddle over what it all means), thus one perspective surely cannot paint the entire picture. Enter multidisciplinarity. However, as with all disciplines, multidisciplinary working itself requires skills, training and experience, especially interpersonal skills to facilitate collaboration, or to recognise when this is impracticable. 

All views expressed are exclusively those of the author.


No comments:

Post a Comment