Week 3 – Methodology

Mar 20, 2019

My methodology investigates the multi-faceted relationships in the diagnostic healthcare industry, specifically analyzing the three major players of the insurance providers, doctors, and patients. Although there are many dimensions to these three players, I will be employing the following methods to explore the role of insurance companies in a doctor’s decision to administer diagnostic tests to patients and then examine its effect on patients.

First, the interviews.
Data would be collected through in-depth interviews (average length – thirty minutes) that utilized Dr. Allegriti’s approach of open-ended questions with investigator developed-prompts. (See Appendix 1) Although I expect to find overwhelming evidence that insurance companies limit doctor’s ability to provide diagnostic tools, I will script my research questions to avoid a bias towards either side and generally seek a doctor’s experiences with insurance companies, regardless of its positive or negative characteristics. In addition, I will also ask repetitive questions that will cause the interviewee to recount and confirm their previous reports throughout the interview. I will take audio recording of each interview and record any observation notes as the interview occurs. After completing the interviews, I will transcribe the data and begin to carry out the analysis. I will sort the anecdotal evidence into two themes: one where doctors were clearly limited by insurance companies in their abilities to provide diagnostic tests and one where doctors were not limited by insurance companies. Once I analyze the evidence and gain a sense of other motifs in the evidence, I will advance the categorization of the anecdotal evidence into other sub-themes that address what factors limited or expanded the doctor’s ability to provide diagnostic tools, like costs or vague insurance guidelines. Then, data analysis will be assisted by NVivo 1.2 software (Ljungberg et al. 2017). The constant comparative method will be used, where the software compares current data with data that was collected earlier to form sub-themes and explore variations in the data. The results from this stage of research will ultimately provide a narrative to the quantitative results from the next stage – the standardized provider model.

Second, the standardized provider model.
In order to understand how doctors respond to the policies of insurance providers in relation to diagnostic tests, I will be using a standardized provider model. Using the same testing group from the previous interviews, I plan to split the doctors into two groups. The first group will respond to a recording by an insurance provider that employs Aetna’s vague “medically necessary” policy to determine whether or not a diagnostic test should be administered for a specific patient profile. The second group of doctors will respond to a recording by another pre-recorded insurance provider that employs specific guidelines that are modelled from Kaiser Permanente. Both groups will also be asked to provide a rationale for their decision. This model should produce binary results on whether or not the doctor pursued diagnostic testing at the time of the patient’s visit based on the doctor’s interaction with an insurance company’s guidelines for testing. The model should also provide additional qualitative data on the actual thought process and calculus of the doctor. I expect a percentage that reveals a doctor’s decision to pursue diagnostic testing for a specific patient profile. Combined with the anecdotal evidence from the interviews, the standardized provider model will produce a compelling narrative about the different factors that might affect a doctor’s decision to administer diagnostic tests to patients.

Third, the statistical analysis.
Through the three previous methods of data collection, I am confident that there will be sufficient data to extensively analyze the role of insurance companies in a doctor’s ability to administer diagnostic tests. In order to advance this project further, I also plan to investigate the effects of the relationship between insurance providers and doctors on a patient’s debt to a healthcare center. This will be done through a simple correlative analysis of the number of errors by a center and the incurred debt by patients.

2 Replies to “Week 3 – Methodology”

  1. Krystal Y. says:

    Hi Akash! I like how you used a mixed-methods approach to answer your research question. I’m excited to see what results (both quantitative and qualitative) you got from your interviews!

  2. Krip R. says:

    Your methodology is incredibly thorough from interviews to analysis! I look forward to reading your results.

Leave a Reply

Your email address will not be published.