Assignment 2

Assignment 2

by Jonathan Lee -
Number of replies: 3
Jonathan Lee Qualitative Methods Assignment #2 Subcritical test result follow-up focus groups Code: Subcritical definition This was the first code I created and I used it for any discussion of the definition of subcritical results. I used this quite often initially. However, after reviewing data from several transcripts I realized that it was too broad and did not capture finer points of the discussion. This led to the creation of multiple codes that were more specific to the data. Code: Differs by department This is used for excerpts discussing how the subcritical definition varies by whether the test department is laboratory or radiology. Code: People performed involved in defining This is used for excerpts discussing how those actually performing the tests (i.e., radiologists and laboratory personnel) should be involved in defining what constitutes subcritical results. Code: Time of follow-up This is used for excerpts discussing how the time frame of follow-up fits into the subcritical definition. Code: Clinical interpretation This is used for excerpts discussing how interpretation of the test result within the clinical scenario is necessary for results to be considered subcritical. Code: No uniform subcritical management system This is used for excerpts discussing either the lack of or variability in the systems currently in place for management of subcritical results. These more specific codes were used multiple times across the 5 focus group transcripts. Memo: There is no consensus subcritical definition or uniform management system and it may be difficult to define because of the need for clinical interpretation What struck me across multiple transcripts was how difficult it was for the participants to give a concise and clear definition of subcritical results. There was also disagreement about whether subcritical could even be defined because of the need for interpretation of the result within a clinical context. Within the larger context, this need for clinical interpretation would make the development of test management protocols for use by non-clinicians more difficult. Code: Dedicated non-MD tracker This is used for excerpts discussing the assignment of a non-physician provider (nurse practitioner or physician assistant) to be the individual primarily responsible for tracking and following up test results. Memo: Following up test results not clinical care? The concept of a non-physician provider being in charge of following-up results was brought up by participants in several focus groups and was in fact already in place in varying form at some sites. The idea of “outsourcing” follow-up to a non-physician running protocols seemed very attractive to the physicians in the focus groups. It raises the more general question of how physicians regard test follow-up in the hierarchy of their clinical duties. Do physicians not consider follow-up as important clinical work? Code: Rotating providers This is used for excerpts discussing the rotating schedule of providers as a barrier to communication of test results to the responsible provider. This was also brought up in several focus groups. Code: Team vs individual follow-up responsibility Code: Easier to contact clinic/team than individual These codes which likely can be combined refer to discussion of responsibility for test results using a team approach rather than individual responsibility. Code: Trusting others to follow-up This is used for excerpts discussing trust when handing off responsibility that the person given the responsibility will actually follow-up on the test results Memo: Clinical culture of responsibility We are all trained to be independent but also to work within systems (hence a core competency in residency being systems-based practice). However, relying solely on individual people is prone to error. Team-based approaches make sense in sharing responsibility. However, providers must be willing to share the responsibility and trust others. I think this is an important piece because in the larger sense this is also where healthcare as a whole is heading. Code: Assigning responsibility for results Code: Handoff responsibility Code: Difficulties identifying responsible person Code: Finding someone to take responsibility Code: Closed loop Code: Discussing follow-up responsibility Code: Responsibility based on time of follow-up Code: Responsibility contract based on test All of these codes refer to responsibility in some way, either assigning, trying to assign, handing off, discussing, or closing the loop of responsibility. I need to combine some of these together or all of them to simplify coding. Memo: The hot potato of responsibility/ownership There’s tension about who owns the results and who should follow-up. The person who initially owns the results is often not the person who eventually follows-up and there are steps in between to handing off this ownership/responsibility from the initial owner to the next person. It is often difficult to identify who this next person is, to actually speak to them or communicate in some way, and then to verify that they have accepted responsibility. Overall I think that I probably created too many codes especially given that there are only 5 focus groups. However, the number of codes is also likely related to the fact that while each focus group is homogeneous within the group (e.g., from the same department), they are heterogeneous compared to the other focus groups. So certain issues that come up in each focus group are specific to that focus group.
In reply to Jonathan Lee

Re: Assignment 2

by Victoria Tang -

Hi Jonathan, I thought that while you had a larger number of codes than you'd like, it was appropriate and never too hard to combine, if fitting. Your thoughts re: why there were so many is likely true and insightful of you to have thought about. I had a similar question about the underlying reason that groups wanted non-physicians following-up labs. Whether they feel this is part of their role in patient care or not is a good question. You memoing is different from others in the class and I'm gathering appreciation for that. I thought your method of coding and memoing is helpful for me to gather a sort of structure to how I could discuss mine in the future. Thanks! 

In reply to Jonathan Lee

Re: Assignment 2

by Laura -

Thanks for sharing your codes-- super interesting to see what you actually used to code your data.  Having analyzed a fraction of your transcripts, I could see some of the themes that your codes allude to, particularly provider responsibility.

I thought it was interesting that you created separate codes for "definition of subcritical results" and "no uniform system (/definition) of subclinical results", since they were so related, but I can see how the additional granularity of the second code would be useful in finding that idea. I hadn't thought of that, creating both sides of the variable, but had come across a similar situation, where someone would be discussing something related to a code, but essentially saying the opposite of the code's idea.  

I also agree that collapsing some of the "responsibility" codes could be useful, and liked the way that last week's presenter did this, using a main idea, colon, sub-idea so that you could group them together (in nvivo). Not sure if the program you are using has similar functionality.  In a way, it seems like all of the codes could represent different tangents that the groups went off on related to responsibility.  If there was repetition across groups, could be useful to identify in this way. But if not, then perhaps trying to find some aspect that was less specific about responsibility (something like responsibility:  relationships) that covers a lot of these ideas might help you see patterns across groups, as opposed to being able to name each idea that each group came up with. 

Great work!

In reply to Jonathan Lee

Re: Assignment 2

by Daniel Dohan -

I like the way you've communicated how the codes have developed (or perhaps proliferated) over time. This is a classic moment in qualitative analysis -- when you realize that the "master" code you've developed needs to be broken down and then you find myself making a gazillion codes and ponder if it is time to roll things back up. In terms of process issues: the definitions you've provided in this short memo are pretty concise and it may be as you push yourself to more thoroughly define codes you'll discover which can be easily combined or which really aren't different after all. And then a substantive thought: you seem to have some codes which are pretty rich and analytic and others that are a bit more concrete and descriptive. It may be worth clustering those groups together (or seeing if other clusterings emerge). These types of codes do different things in a project. Often you need both types but it is also typically helpful to be clear about which is which.