Who is responsible for action?

I have recently been reading Steve Wing's commentary on research ethics in community-driven studies. (published in Environmental Health Perspectives v110, n5) He describes some of the serious challenges that researchers can face when engaging in research that supports communities over corporations, and the rewards that come from meeting those challenges. The choices he was forced to make were difficult, but it is not terribly difficult to recognize what the right choice was. You make a commitment to community members, you keep it. Even if it means some sacrifice, or a lot of sacrifice.

The environmental researcher's goal is to in Wing's words, "conduct studies that have the sensitivity to detect an effect if one exists." Then they work on dissemination. In academic journals, the media, and the community. This process of dissemination is in a sense action, action that may lead to policy change. This is strongly evidenced by the strong reaction of the pork industry, the pork-industry friendly government, and the university to the dissemination of findings from this research.

However, dissemination will get you only so far. Is it appropriate to leave policy action strategies to the community partner? If so, are they equally responsible to engage the researchers in the planning of action, as the researchers are responsible to engage community members in the research process?

Different partnerships do things differently, but if we were to think about best practices what would they be? So many partnerships get hung up at the action stage, and many do not achieve policy change, some because they never develop clarity about the policy change they would like to see. Many researchers are nervous about being seen as overtly political, because it will call their research into question. This is not without reason.

Much of what happens in partnership is not public. The best strategies leverage the diversity of roles and skills to create change. But if we are not clear, from the outset, who is responsible for analyzing policies, and for planning and initiating action, we may never achieve that clarity.

Including policymakers in program evaluation

Have you ever included interviews with policymakers in your program evaluation? If you are running a pilot program that a) relies on the cooperation of implementers or b) is a model that you hope to scale up, engaging policymakers can be a useful step.

Let's look at the first case. You are running a pilot program that relies upon the good hearts of implementers. In other words, you have cobbled together a series of special arrangements. Your program participants are treated differently than others. If these special arrangements have benefited your participants, it might be good to let policymakers know about that. Maybe they already know. By engaging policymakers as part of the program evaluation process you can find out what they already know about your program and the population you work with. You can also provide information about the program and the population. The research/evaluation paradigm is a useful one for this interaction. It takes the pressure off the policymaker, and puts the onus on you to improve your program. At the same time, it engages the policymaker as an ally.

How does all this impact implementers? People who have agreed to help you do things differently are change agents in their own right. Pursuing policy support for their activities and policymaker recognition of their leadership is a great way to reward their collaboration. The one caveat to this: if the implementer is doing something risky to their career by helping you, obviously you shouldn't call attention to it.

Lets look at the second case. You have a program you think is great and you want to replicate it. In this case, you again are going in to gauge the policymakers knowledge of your program, the problems it attempts to address, and the population you work with. The information you gather about how your population, problem, and program are viewed will inform your strategies for garnering support (monetary and other) for replication. Concerns must be addressed. Confusions clarified. Appeal amplified.

I did a series of interviews with policymakers as part of an evaluation of a jail-based re-entry program some years ago. The information obtained in the interviews helped to frame strategies for an advocacy campaign to support re-entry friendly policies. The creation of new relationships as part of the process (or in some cases the enrichment of ongoing relationships) was also extremely valuable.

Motivational interviewing for policymakers

One of the advantages of doing advocacy work in partnership is that different individuals and groups are perceived differently by policymakers. Researchers can be perceived as less threatening than an organized community group or coalition of service providers, particularly when the interaction is framed as research rather than advocacy. There may well be a host of valuable data that can only be gathered from policymakers. In addition to their views of current and proposed policies, policymakers generally have a sophisticated view of the policy environment. They may or may not be aware of various public health issues or community issues.

It is worthwhile for a partnership to think about interviewing policymakers, to collaborate on questions asked, and to consider the value of the research interaction beyond data gathering. This brings me to the construct of motivational interviewing. Designed as a therapeutic and not an investigative technique, its tenets never-the-less hold some value for action researchers. The motivational interviewing goals of establishing rapport, eliciting change talk, and establishing commitment language mirror that of the CBPR policy cycle. Beginning with creating rapport between the policymaker and the research partnership, the researchers have an opportunity to uncover information critical to policy change.

The policymaker will likely fall into one of three general camps
1) understands your issues, and is diametrically opposed
2) doesn't really understand your issue, and is either willing or unwilling to listen
3) supports your issue, but expresses some sense of powerlessness to move it forward

(There is another category of the policymaker who supports your issue and feels empowered to act upon it, but that is rare and unproblematic.)

So the first goal when approaching policymaker interviews is to understand in which camp the policymaker belongs. Then to delve deeper. Why do they disagree? Is it ideological? Practical? The more you understand opposition, the better you will understand your issue. For those who don't understand your issue the research relationship could also be used to educate. In political circles this is sometimes called push polling, where information that the interviewee does not know is built into the question to try to sway their allegiance. In key informant interviews, the approach is to simply provide what data you have already gathered on the issue as a basis for reactions from the policymaker. In the final camp, the researcher needs to explore the barriers that separate the policymaker from effective action. This information can then form the basis of a collaborative effort between the community and the policymaker to raise the level of political will to the point of action.

Throughout the process of a single or multiple interviews, the researcher can establish the reality as seen by the policymaker, and then use questions to elicit the changes in status quo that the policymaker seeks. This information can then be used to target persuasive arguments to this policymaker and others, when the partnership moves to the advocacy phase and begins to seek commitments from policymakers.

Does this feel ethically murky? Sure it does. But the reality is that qualitative research that engages public officials is an area poorly understood by IRBs and others. As this research becomes more common we will no doubt parse out some reasonable ethical guidelines. But to get to that point we have to first give it a try!