Let’s talk real talk: An argument to include conversation in a D-COS for aphasia research with an acknowledgment of the challenges ahead

Dietz and Boyle (this issue) present a compelling argument for the development of a core outcome set for discourse (D-COS) in aphasia research. As has been demonstrated repeatedly in other medically-oriented research fields, and recently proposed by Wallace and colleagues for aphasia research (Wallace et al., 2014), the development and use of a COS is likely to improve our ability to compare and synthesize experimental treatment outcomes and to increase transparency in the reporting of those outcomes. Dietz and Boyle argue forcefully for the principle of a D-COS, or subsets of D-COS geared toward the various levels of discourse analysis. As they reason, an abundance of tools for measuring discourse in aphasia has led to a tipping point, wherein the reliability and validity of some discourse measures are questionable, and the comparison of outcomes between treatment studies, and systematic reviews of aphasia treatment research, are difficult, if not impossible to conduct.

We agree in principle with this view and, like the other participants at a recent Clinical Aphasiology Conference (CAC) Round Table (Boyle & Dietz, 2015), who unanimously supported the establishment of a D-COS, we also recognize that it might nonetheless be a lofty goal. Dietz and Boyle acknowledge the potential difficulties, but may underestimate the challenges in this initial unveiling of the concept. Much like the impediments that need to be addressed to successfully implement the development of a COS, as Hula, Fergadiotis, and Doyle (2014) point out, establishment of a fruitful D-COS would have to overcome several obstacles, notably achievement of consensus on constructs to be measured both within the community of aphasia researchers and between all stakeholders. In fact, Hula and colleagues warn against a consensus-based process that risks codifying an unfocused, stale, or imperfect COS and that may fail to advance scientific progress or be adaptable to the many stakeholders. Moreover, MacWhinney (2014) also raises important considerations for overcoming the vast challenges of establishing and implementing a COS, among them, misjudging the complexity of the problem, and neglecting to integrate state-of-the-art methods for data-sharing and data archiving.

All of the challenges facing development of an aphasia COS are also inherent in establishment of a D-COS. At the same time, developing a D-COS may be a first step in reducing the complexity of an aphasia COS, by offering a methodology for analyzing outcome subcomponents, i.e., those specifically related to discourse. Regardless of the obstacles, we believe that inclusion of conversational discourse must be an essential ingredient if the core outcomes are to reflect meaningful outcomes in aphasia treatment, even though we acknowledge that its inclusion is likely to exacerbate the challenges of developing a reliable D-COS. Clinicians and aphasiologists can learn a great deal about an individual’s fluency, word retrieval, propensity to produce paraphasias, grammatical form, discourse structure, etc., the basic characteristics and severity of a person’s expressive aphasia, just by listening to their Cookie Theft description (Goodglass et al., 2001) or other similar picture description. Nonetheless, to capture information about a person’s functional discourse, we need more than just “tell me about this or that” monologic speech elicitation tasks. We need to understand how persons with aphasia (PWA) negotiate the most basic currency of everyday communication, what Clarke and Wilkes-Gibbs (1986) called “the fundamental site of language use”, i.e., conversation.

By observing how PWA manage the co-construction of shared meaning in conversation (Sacks, Schegloff, & Jefferson, 1974), with its dyadic, real-time, interactive nature (Carragher et al., 2012), we can learn how PWA manage to overcome their linguistic deficits in the arena deemed to be their most frequent communicative activity of daily life (Davidson, Worrall, & Hickson, 2003). Importantly, we are not proposing that every researcher and clinician become an expert in conversation analysis (CA; Sacks, Schegloff, & Jefferson, 1974). This qualitative methodology enables the examination and description of the ways in which orderliness, e.g., in turn-taking, sequencing, and repair organization, usually permeates conversation. Although CA is a potent tool for investigating individual compensatory strategies for negotiating the pervasive ‘trouble in talk’ in aphasia (Klippi, 1996), the labor intensive nature of discourse analysis in general, and CA in particular, cannot be ignored. Thus, when Dietz and Boyle make a statement beginning with, “Transcription and coding issues aside,…”, it seems difficult to ignore their competing desires to put ‘Ivory Tower’ analogies to bed with the stark reality that there are currently no clinically feasible applications of discourse analysis in aphasia. Even the creative solution of measuring transactional success in conversation partners (Ramsberger & Rende, 2002) may not be easily implemented by clinicians in the real world, given the need for naïve conversation partners.

To reiterate, it may seem overly optimistic for Dietz and Boyle to suggest that the community of aphasia researchers will agree on a D-COS, let alone one that is inclusive of the needs of other stakeholders such as clinicians and persons with aphasia and their significant others, one that includes measures of conversation, and one that resists growing stagnant over time. Certainly, new collaborative approaches to aphasia research will be required if we are to realize a “step change” that could result in increased reliability, validity, and transparency of aphasia research (Brady et al., 2014). In their article of support for Wallace and colleagues’ (2014) call for an aphasia COS, Brady and colleagues urged that aphasia researchers and clinicians have a moral and ethical responsibility to PWA to develop a common database for aphasia research and high quality transparent reporting in parallel with the establishment of a COS. This work has already been started on more than one continent, in the form of the AphasiaBank (www.aphasia.talkbank.org) in the U.S. and the Collaboration of Aphasia Trialists (CATS; www.aphasiatrials.org) in the European union. We suggest that these collaborative efforts may hold the key to some of the stubborn challenges of developing and implementing COS and D-COS in aphasia. The archiving and sharing of aphasic discourse samples – including conversational samples – across labs, languages, and cultures, may one day lead to Big Data analysis solutions that we can’t even imagine today, just as we couldn’t have imagined we would all be carrying computers in our pockets 20 years ago.

Finally, we wish to acknowledge that incorporating conversation and other functional communication abilities reflective of language used in real-life settings, is not a new idea, but rather a timely revival of some very good old ideas, not yet realized (e.g., Holland, 1991; Sarno, 1993). Promoting a D-COS that includes one or more measures of conversational success will expand upon the sparse but promising recent findings suggesting that impairment-focused aphasia therapy can positively impact conversation in aphasia (Best et al., 2011; Carragher et al., 2012). Importantly, as MacWhinney (2014) notes, there is no question that efforts at international standardization, such as development of COS, will ultimately be adapted by those who perform clinical research in aphasia. Thus the question is not whether to get on board with development of a D-COS, but how to guarantee maximum benefit. As we argue over the methods, let us remember the hopeful advice of Simone de Beauvoir, that “…each idea not yet realized curiously resembles a utopia; one would never do anything if one thought that nothing was possible except that which exists already” (de Beauvoir, 1954, p.193).