The Value of Qualitative Description in Health Services and Policy Research.

Health services and policy (HSP) researchers have long used qualitative research methodologies to explore health system issues. However, the appropriateness of one approach, qualitative description, for HSP research is still often overlooked. In this article, I discuss the role that qualitative description can play in HSP research, and argue for its greater acceptance as a valid form of academic scholarship.

Introduction

Qualitative researchers have made significant contributions to health services and policy (HSP) research, providing valuable insights into the ways we conceptualize health, illness, patients' experiences, the dynamics of interprofessional teams and many aspects of care delivery. Dominant qualitative methodologies, such as grounded theory, ethnography, narrative approaches and phenomenology, are now regularly employed to pursue a variety of HSP topics. There is, however, a potentially important qualitative methodology for HSP research that is often not recognized by qualitative researchers or, at the very least, is seen as an inferior use of qualitative data. In 2000, Margarete Sandelowski highlighted the lack of stature that basic qualitative description had within the wider qualitative research community (Sandelowski 2000). While there are HSP researchers who identify using qualitative description (Granger et al. 2009; Gutierrez et al. 2013; Milne and Oberle 2005), this approach is still not widely acknowledged or known within HSP research despite previous calls for its wider adoption (Neergaard et al. 2009). This lack of appropriate recognition of qualitative description risks us missing a significant opportunity to adopt a methodological approach that is quite well suited for addressing many questions that arise for HSP researchers, and for qualitative HSP researchers to make an even greater contribution within clinical, policy and decision-making settings.

A health services research issue

While qualitative description is applicable to a wide range of HSP topics, to help clarify the approach, I will focus on its use within a recent research project. In 2011, we examined the timeliness of pain treatment at one pediatric emergency department (ED), finding that only 15% of patients received an analgesic within the recommended timeline (Porter et al. 2013). In response, the ED instituted a new pain treatment directive, which resulted in some improvements, but still left approximately 50% of patients not having their pain treated within recommended guidelines (Porter et al. 2015). We wanted to explore the barriers to further improvements in the assessment and treatment of pain which still existed within the ED (Chafe et al. 2016). Given the complexity of the care environment, and the level of understanding we hoped to acquired, we decided that some form of qualitative research which engaged nursing and physician staff within the ED was a reasonable approach to take. There were clearly other qualitative research approaches that we could have adopted, but qualitative description was likely the best approach given that our aim in this project was simply to identify possible barriers which people working in the ED felt still existed.

Sandelowski says that researchers conducting qualitative description studies “seek descriptive validity, or an accurate accounting of events that most people (including researchers and participants) observing the same event would agree is accurate, and interpretive validity, or an accurate accounting of the meanings participants attributed to those events that those participants would agree is accurate” (Sandelowski 2000). Maxwell expands on what is meant by interpretative validity, saying that interpretative accounts “are grounded in the language of the people studied and rely as much as possible on their own words and concepts.” Maxwell then contrasts descriptive and interpretative validity with attempts by qualitative researchers to give more theoretical, evaluative or generalizable accounts of a research topic. It must be reiterated that I am not arguing against qualitative projects which explore more theoretical, evaluative and generalizable interpretations, which again have been shown to provide valuable understandings. Yet limiting ourselves to an account that concerns only descriptive and interpretative validity is in keeping with what we hoped to determine in our research project – a description of the issues that people working within the ED felt were barriers to further improvements in pain management.

Qualitative description, again

It has been almost 17 years since Sandelowski first published her article calling for qualitative description to be considered as an equally valid qualitative methodology. Yet, key introductory texts to qualitative research and qualitative health research still often do not even reference qualitative description, let alone present it as an equally valid method (Creswell 2012; Green and Thorogood 2009; Morse 2012; Patton 2015). These are the same texts that many HSP researchers use in their training. Part of the reason for this continued oversight is that for many researchers the power of qualitative research lies in pushing past more simplistic descriptions of situations, and exposing or challenging the underlying conceptions that groups in society and in healthcare hold. While it is difficult to formulate a single definition, which can capture all the various aspects of qualitative research (Creswell 2012; Denzin and Lincoln 2011), it is often equated with the development of more conceptual understandings of social phenomena (Pope and Mays 1995). If a study is simply using qualitative description, accepting and reporting the concepts presented by participants as they are presented, it is not clear that qualitative description reaches this level. For those aiming at increasing our understanding of the social world, qualitative description can be seen as not much more than a journalistic account of what was seen, what people say happened and their reports about what they thought about it. In other words, for many qualitative researchers, qualitative description can seem as either uninspiring or unfinished qualitative work.

Yet, the goal of many HSP research projects is not to increase our conceptual knowledge, but to bring about change and quality improvements. For example, in our qualitative description study, we were able to report to the ED staff and management that current barriers to increasing the timeliness of pain medications identified by providers within the ED related to accurately capturing the level of pain with the current pain assessment tools, issues in treating specific complicated conditions, and inadequacy of the current initiatives to treat patients with severe pain (Chafe et al. 2016). One of the advantages of qualitative description compared to other qualitative methodologies is that there is a lower level of inference so that participants are more readily able to agree on the account being given (Sandelowski 2000). This is not to claim that the researcher is neutral or outside of the research process. Even in writing a basic description, the author selects to include certain details and exclude others (Sandelowski 2010). Yet claiming that a research participant said, for example, that she did not feel comfortable using certain medications for patients with abdominal pain can be easily verified by referring to the interview tape or transcript in a way that more conceptual interpretations cannot. While there may be a rich evaluative discussion whether the provider should feel uncomfortable using certain medications, participants and others should be able to agree that the research participant indeed said it.

Next, to keeping the analysis at a lower level of inference such that the results given can be more readily agreed to, qualitative description keeps the analysis at a level at which those in the situation being studied should be able to readily understand. The findings of our project give an account of the barriers in the ED that were not known before, because it included perspectives from a range of people involved, but is hopefully one that nursing and physician staff working in the ED can relate to, discuss and act upon. In HSP research, rather than seeming uninspiring, qualitative description can have a powerful role in engaging a range of stakeholders at a level they relate to in order for them to better understand a situation and encourage change.

Qualitative researchers have long had concerns with the type of research project that I am describing, which is largely motivated by its practical or applied, rather than conceptual, goals. Over 20 years ago, Ritchie and Spencer (1994) proposed a framework method to address applied research questions. But they dismissively placed these types of questions into the domain of “commissioned research” to distinguish them from more proper uses of qualitative research (Bryman and Burgess 1994). Green and Thorogood make a similar distinction between “pure” and “applied” research, with applied research again concerned with the aims set by external organizations that want to use qualitative methods to solve their specific, practical problems (Green and Thorogood 2009). This distinction misses the fact that as HSP research has developed into its own domain of study, more independent researchers are formulating and pursuing these types of applied questions themselves, some even doing so while working within healthcare organizations (Chafe and Dobrow 2008). Other qualitative health researchers are more open to the applied implications of their work, but they do so still by working at the conceptual level (Morse 2012; Thorne et al. 1997). One advantage of qualitative description for certain projects is that it is able to motivate action by keeping the description closer to the everyday terms of the people involved.

Rigor

Given that it is not uncommon for HSP researchers to face questions that are mostly concerned with determining what is happening and what are people's reactions to it, qualitative description would likely be appropriate for a range of HSP studies. Better appreciating the distinctness of qualitative description as a methodology is the first step in a longer discussion around how this type of HSP research should be conducted. Being explicit that this is the approach being taken and making methodological choices in line with this direction are key starting points. Milne and Oberle emphasize appropriate interviewing skills, ensuring that participants are free to speak about a topic, and the need to probe for clarification and depth (Milne and Oberle 2005). Our project used a protocol that had included a fairly standard consent process; given the small number of potential subjects (~30 people), we invited all physicians and nurses working in the ED to participate in the project; we developed and revised the interview and focus group guides; interviews and the focus group discussions were recorded and professionally transcribed; the analysis used both deductive and inductive coding; and institution ethics approval was granted for the entire project before it started. In other words, many of the methodological choices we made are fairly common within qualitative research projects.

Yet, this approach also possibly raises some unique methodological questions. One of the reasons we were able to usefully conduct a qualitative descriptive study is that most people in the situation we were concerned with, i.e., the activities of this specific pediatric ED, share basic beliefs about the situation: (1) beliefs about the ED and its function to treat urgent medical needs; (2) that people have certain roles within an ED, e.g., patients, parents, triage nurse, other nurses, physicians; (3) that pain is real, is often associated with an underlying medical condition and is something that is preferably avoided depending on situations; (4) that drugs which reduce pain can be administered, and that these drugs can have other impacts that may be negative and need to be considered; (5) that because of the possible potentially positive and negative impacts the drug might have in specific circumstances, there may be disagreements on whether a drug should be administered to a specific patient at a specific time, but it is usually better to reduce the pain associated with a condition earlier; and (6) that the role of developing departmental policy is that it gives direction for a consistent approach across similar situations. In our interviews with staff and patients, no indication was given that anyone challenged any of these underlying conceptions of the situation; or did we, as researchers, see the need to explore these issues in order to achieve our study aims. It is also likely that readers interested in the barriers to quicker pain management within a pediatric ED share similar views. In other words, it could be argued that there is an “agreement within the community of inquirers about the descriptive or interpretative terms used,” recognizing that if these assumptions are not supported during the study another level of analysis may be needed (Maxwell 1992). It is important for the researcher in this type of qualitative research project to be reflexive and consider the impact that their background and social position may have on the findings they arrive at. Yet, if we are concerned with only providing a description of events that fits with a community's shared understanding of a situation, like trained medical staff working within the same ED, and we do not attempt to move beyond descriptive and interpretative validity in our analysis, it is not clear, in this context, whether there is an added value for the inclusion of ontological and epistemological considerations within the qualitative research process, as suggested in a number of qualitative research texts (Denzin and Lincoln 2011; Leavy 2014). The methodological implications of this possibility need to be further explored.

Conclusion

In this article, I describe qualitative description as a research methodology that is well suited for many HSP research projects. Although distinct from more conceptually focused qualitative methodologies, it is not opposed to them, and clearly topics can benefit from being studied from various qualitative perspectives. The approach I am outlining is one likely familiar to HSP researchers using qualitative methods, even if it is not always recognized as such. It is also quite often used within healthcare organizations to engage with patients and staff around various issues. Better recognizing qualitative description will hopefully encourage researchers to explicitly adopt this methodology when it is appropriate, and to foster greater discussion of what are the most rigorous ways that it should be used within HSP.