Tessa Ireton

Market Structure and Quality of Service: Investigating Oligopolies and the Quality of Nursing Home Care in California During The COVID-19 Pandemic

March 25, 2021   /  

Student Name: Tessa Ireton
Major: Economics
Advisor: Dr. Brooke Krause

Quality-of-service outcomes in nursing homes are of great social and human importance. However, especially during the COVID-19 pandemic, consistently maintaining markets with high quality care has been a pervading issue in the American nursing home industry. Furthermore, the industry is strongly characterized by oligopolies, a market structure that literature indicates may be less compatible with quality service than competitive markets. With this paper, I aim to investigate the possible intersection of oligopolist market structures and the quality of nursing home care during the COVID-19 pandemic. I start by describing quality of care in nursing homes, particularly during the COVID-19 pandemic, contextualized by literature analyzing nursing home care quality and market structure. Then, I develop a model of profit-maximizing nursing home behavior using theories of oligopolist decision-making, rooted in the basis provided by both Cournot and Bertrand, informed by contemporary models describing the nursing home market. This model demonstrates an opposite correlation between the numberof firms in a market and the quality of nursing home care. To test this prediction, I construct a 26-week panel dataset including nursing home attributes and facilities’ experiences with COVID-19, using data from California’s Agency of Health and Human Services, the California Department of Public Health, the Census Bureau, and the Centers for Medicare and Medicaid Services. Random effects and Hausman-Taylor estimations test the relationship between the number of nursing homes in a market and the outcome of COVID-19 outbreaks in nursing homes, a proxy for quality of care during the pandemic. The results indicate that market structure is not explanatory in understanding differences in quality of nursing home care during the COVID-19 pandemic.

Tessa will be online to field comments on April 16:
2-4 pm EDT (PST 11am-1pm, Africa/Europe: evening)

65 thoughts on “Market Structure and Quality of Service: Investigating Oligopolies and the Quality of Nursing Home Care in California During The COVID-19 Pandemic”

  1. Hi Tessa- What an interesting project! How cool that you could apply your economics to this timely issue. Thank you for sharing this project. All the best to you!

    1. Thank you so much Dr. Pierce! The skills I picked up in your Calculus class were absolutely critical to the process, so it means a lot to me that you enjoyed the presentation.

  2. Tessa, this was a very interesting project! Congratulations on a job well done!

  3. Amazing work, Tessa! You are an outstanding researcher and inspiring scholar. It was an honor to work with you this year.

    1. Dr. Krause, thank you so much! I was my honor to learn from your expertise and scholarship throughout the process.

  4. Great job Tessa. Continuing the Ireton tradition of greatness. I even learned a new word “Oligopolies”.

    1. Thank you Geordie! It is such a high compliment that I might live up to Maddy’s Wooster Legacy!

  5. I’ve learned so much from your research, Tessa! Thank you for sharing your findings. Since your IS examines an issue that is still relevant and ongoing, have you learned anything since its completion that has influenced the way you think about the COVID-19 response in nursing homes? Are there new questions you’d like to answer?

    1. Thank you Lynette, and thank you for helping me with my writing in the Writing Center all throughout the process! I learned so much from your expertise. As the story unfolds on the national stage, I find myself thinking more and more what the implications of the COVID-19 disaster in nursing homes means for the long-term care industry as a whole. Obviously, a massive amount of change needs to take place in the industry to protect patients from current endemic disease and the next pandemic. I wonder what will ultimately be the societal and governmental response to the issue, and even if nursing homes face an existential threat in the long-run as long-term care alternatives like community and in-home care gain more favorable attention and subsidy from governments and industry.

  6. Interesting work, Tessa, thanks for both doing and sharing it!

    I’d be curious to know more about the decision to define nursing home care markets at the county level, rather than at the metropolitan statistical area level (and if you did run this at the MSA level what the results were). I can imagine this could have some negative impact on the size of your data set due to the loss of rural counties, but I ask because it seems reasonable that consumers would shop for non-emergency long-term healthcare services across a broader area than a single county…though I admit that’s an assumption I make as someone who has not shopped for nursing home services!

    1. Thank you so much! That’s a really good question, and is definitely something I, and the relevant literature, have grappled with. Much of the literature uses the county-level definition of a NH market due to findings indicating that most nursing home patients reside in facilities that are within their county of origin (the county they lived in before they needed long-term care). However, a couple studies I found used radii of certain miles or kilometers to identify spatial markets for nursing home facilities, but I was unable to replicate this kind of analysis due to data constraints and my own lack of expertise in the tools required to assess and identify those data points. California itself is probably not the best place to settle for county-level markets though, since the number of nursing homes in many California counties is so high, which suggests that there are multiple NH “markets” within the same county. This issue is something I hope to address in the future with data from other states besides California, and possibly by using geospatial analysis.

      1. Thanks for the response Tessa – I’m sure you’ll be plotting in GIS and building spatial weight matrixes to better understand how and why space matters sooner than you think!

        1. Thank you for your question! I think I will! That is definitely one area I hope to grow my knowledge and develop expertise.

  7. Awesome project, Tessa! So cool to hear about what you’ve been working on this year. What are your next steps?

    1. Thank you so much Maya! I hope to investigate the relationship between market structure and quality of care further using new data and further angles of analysis, such as information transparency in the markets. Professionally, in the Fall I will start as a consulting development analyst for Accenture plc. in Columbus, OH. From there, I hope to develop my expertise and skillset in helping clients in the health and public service sectors.

    1. Thanks Andreas! That was probably the hardest part of the project so I’m glad it turned out okay!

  8. Very interesting, and definitely timely. It was fascinating to see you how you constructed the model, and great that you were able to get access to that data. The husband (Jonathan Cohn) of a friend and colleague of mine (Prof. Amy Cohn) wrote the book “Sick” about the American health care system. A very interesting, if sobering, read. Congraulations!

    1. Thank you, Professor Powell! Part of the reason I decided to hone on the topic was the abundance of data, (albeit flawed data,) on long-term care facilities. Thank you as well for mentioning that book, I’ll be sure to check it out as I grow my knowledge on the topic and American health care further.

  9. Congratulations Tessa! I found your topic very interesting, especially given the timing and the fact that most Americans will find themselves making a decision about elder care. Whether it be for ourselves our our loved ones.

    1. Thank you, Kathlee! Yes I totally agree, I think that the topic of the quality of care patients in long-term care facilities receive is surprisingly niche considering the immense impact it has on nearly every American’s life at some point.

  10. Hi Tess! Great job on presenting this information in a way that is easy to follow and understand. This topic is very important to understand and talk about, and I really appreciate the way in which you presented this information. Too often, people report on topics like these in a very detached manner, so I appreciate the fact that you always put your project in the context of people and their lives.

    1. Thank you Julia! I’m so glad that you appreciate the importance of the topic of long-term care quality as more than just a topical news story, but as a human issue. I really tried to center my analysis in this fact and remain connected to the broader importance of the topic, and I’m relieved that you found that coming through in my presentation.

  11. Awesome job Tess! It’s so cool that you were able to pursue such a relevant topic to the current times. If you could continue your research how do you think your findings would change with the increased availability of vaccines?

    1. Thanks Lydia! I think if I continued using COVID-19 cases as a proxy for quality, I would need to control for vaccination rates somehow. I also think that the pandemic has been uniquely detrimental to the information symmetry (or lack thereof) in the nursing home industry, and that possibly my finding that market structure had little impact could be changed under identical analysis in the future time periods where people are permitted to visit nursing homes again at pre-pandemic levels.

  12. Tess, congratulations on a huge achievement! Not only is your work meaningful and timely, but it is especially important as an aging population will need more and higher quality care.

    1. Thank you Shauna! I totally agree, I hope that, at the very least, the disaster COVID-19 has brought to nursing home facilities will draw society to focus more on the kind of treatment our elders deserve.

  13. What an interesting presentation! Why did you decide to have county lines be the boundary? Do you think your findings would be different if you compared between states?

    1. Thanks Maggie! Due to data constraints and my lack of expertise in using tools for geospatial analysis, I opted to use county boundary. I justified this decision with the fact that much of the literature on this topic uses county boundaries, too. In the future I will grow my expertise in that kind of analysis. I totally think that the results of my estimations would be different between states – California’s counties are so populous that there could quite easily be multiple markets for nursing home care within the same county.

  14. Tess! Two years ago I was so sad when you left the computer science department to be an economics major, but now I’m so glad you did, because it’s clearly where you belong.
    You’re doing great things!

    1. Thank you Nicole! That is so nice of you to say, I was sad to stop taking CS classes, but I’m glad that I was able to continue coding in Econ. I’m so happy to see what you have accomplished since then – you are such a brilliant expert in what you study, that it is plainly obvious that you belong where you are too!

  15. Tessa:
    Fascinating project for these times. Thank you for sharing your research. As someone who works in healthcare, and more directly in the monoclonal antibody space for COVID disease, I am curious if you investigated what effect the lack of clear guidelines or treatment options had as an intervening variable on quality of nursing homes care. I know you are looking at all care, not just COVID-19 care in the nursing homes channel, but thought I would ask anyway.
    Great stuff, thank you!

    1. Thank you Arvind! In the beginning of my research, I spent a good deal of time familiarizing myself with the story behind COVID-19’s rise in long-term care facilities. While I did not empirically test for relationships in this area specifically, the narrative seems to reflect that absent, unclear, or inconsistent guidelines (especially early on in America’s experience with the pandemic) played a role in COVID-19 spread in facilities and in mortality. One of the areas I wish I could have controlled for in my research was the impact of administrative decision making on the facility level on the outcomes of outbreaks in facilities. My robustness tests, not included in my video presentation, indicated that there could be a good deal of factors influencing outbreak outcomes outside of my model, including the uniqueness of each facilities’ reaction to the pandemic, especially in how they address infection control.

  16. Tessa!

    This is an interesting project that intersects patient care and business. Did you have any way to “normalize” the data if one NH had say 70 beds compared to another with 30 beds? If you had to do this again, would you pick a different geographic location to account for the county variation in terms of number of nursing homes?

    I am so happy for you! You rock and like your family will do great things in life!

    1. Thank you so much Sam! And thank you for your supportive words. To answer your questions, (1) I controlled for patient census days and the number of occupied beds in my analysis, since the literature indicates that the scale of the facility’s operations and the number of patients do impact overall care quality. (2) I am looking at conducting a similar analysis on the national level, not just using one state’s data!

  17. Very interesting! Thank you for teaching us something new and meaningful Tessa. Very Proud of all you have accomplished at Wooster!
    Mom & Dad

    1. Mom and Dad, thank you so much for all of your love and support throughout my education. You two are my heroes.

  18. Great presentation Tessa! Also really impressed that you were able to present your work at the Eastern Economics Association Conference earlier.

    I was wondering if you could shed more light on why you think for-profit nursing homes had worse outcomes? Is there a theoretical reason? Did you find literature that corroborated your empirical findings?

    1. Thank you so much Professor Moledina! I am glad I had the chance to learn from the Conference and that I had the support of Wooster’s Economics department to do so. The literature corroborates this finding, and the theories behind why center on the incentive structure for-profits face. The general idea is that there is a low level of information symmetry in the nursing home market, (due to the infirmity of patients and the opaqueness of care options) and that for-profits might be incentivized to opportunistically extract more profit by keeping quality-associated costs low. Because patients might not be able to detect lacking care quality (or because they might not have the income level necessary to respond in the market to lacking care quality), for-profits may lack incentive to maintain a quality level beyond what is necessary to compete with outside industries.

  19. Congratulations Tessa! I am impressed with your methodology and approach to your research question. Wonderful job!

  20. Hi Tess! Awesome presentation! You did a great job explaining a complicated concept in an easy to understand way. If you had the chance, would you continue this research. If so, how?

    1. Thank you Erica! I’m very lucky that I do have the opportunity to continue to investigate this research question – I hope to do so by adjusting my empirical specification and using more comprehensive data from prior to the pandemic. I also hope to use different metrics for quality of care besides proxying for quality with COVID-19 outbreak outcomes.

  21. Congrats Tess, this is really important and valuable work! My question is what are the benefits and pitfalls of analyzing quality care as a quantifiable?

    1. Thanks so much Stephanie! Great question, a good bit of research was conducted on my part in how quality can be quantified. The benefit of quantifying abstract or subjective characteristics such as quality is that we can develop empirical models to objectively test for correlations, magnitudes of impacts, and in some cases causality. In this way, we can isolate the relationships with variables and develop statistical tests to specifically identify whether we can reject hypotheses with certain levels of confidence in a scientific way, rather than subjective or rhetorical analysis. This approach lends more confidence to policy-making. Of course, there are flaws in this approach – studies indicate that the idea of “quality care” varies widely among nursing home residents which means that quality is not necessarily something that exists on a number line or can be identified as an objective metric without making some sacrifices and removing nuance from the picture. In my analysis, I started with literature that defines quality as “conformity” to a consumer’s expectations for what a good/service should yield them in terms of utility. From here, I simply assume that an “average” or “typical” nursing home patient would expect that the care they would receive would keep them safe from deadly infection. There is a good deal of of literature that investigates the connection between infection control, other output-based metrics of care quality, and COVID-19 outcomes in nursing homes.

  22. What a wonderful presentation, Tessa! This research is really impressive, and so well timed. It was really interesting to learn about elder care and covid through an economics lens that I hadn’t previously considered. And it was really nice to hear at the end of your presentation that things are looking up!

    1. Thanks Georgia! Yes, the situation is very dire, but I think it is important to remember that solutions to these issues *do* exist, we just have to think creatively and collaboratively to find them.

  23. Well done, Tessa.
    Nice presentation and it is obvious that you are passionate about this topic.
    Congratulations on your completion of the IS and your Wooster education.

    1. Thank you Sanne! It’s been a wild ride, but I am thankful for the College and all the amazing people I’ve had the opportunity to learn from here.

  24. Tessa,
    I am so proud of you! What an impressive presentation. As I tell everyone, you are a such great representative for the college and our department. I know that that you are very invested in your research so I will look forward to seeing the direction you will take your work as you move out into the world.

    1. Thank you Denise! And thank you for your enthusiastic support of my professional and academic endeavors here at Woo. So honored I got to both learn from you and from Special Collections these last three years.

  25. Thanks so much Stephanie! Great question, a good bit of research was conducted on my part in how quality can be quantified. The benefit of quantifying abstract or subjective characteristics such as quality is that we can develop empirical models to objectively test for correlations, magnitudes of impacts, and in some cases causality. In this way, we can isolate the relationships with variables and develop statistical tests to specifically identify whether we can reject hypotheses with certain levels of confidence in a scientific way, rather than subjective or rhetorical analysis. This approach lends more confidence to policy-making. Of course, there are flaws in this approach – studies indicate that the idea of “quality care” varies widely among nursing home residents which means that quality is not necessarily something that exists on a number line or can be identified as an objective metric without making some sacrifices and removing nuance from the picture. In my analysis, I started with literature that defines quality as “conformity” to a consumer’s expectations for what a good/service should yield them in terms of utility. From here, I simply assume that an “average” or “typical” nursing home patient would expect that the care they would receive would keep them safe from deadly infection. There is a good deal of of literature that investigates the connection between infection control, other output-based metrics of care quality, and COVID-19 outcomes in nursing homes.

  26. Amazing job, Tess! I was wondering, I know you had to narrow your topic- what was that like? How did you go about it?

    Congrats on such a great project, I know it’s leading you to great places.

    1. Thanks Kath! I narrowed my topic based on the context of the pandemic and the data available. In the spring 2020, I knew I wanted my IS to be a response to COVID-19. Also, I knew I wanted to take advantage of the data on healthcare providers already available, and not collect my own primary data. The is an abundance of data (albeit flawed data,) on nursing homes available, and the long term care industry has obviously been uniquely and catastrophically impacted by the virus.

  27. Tessa,
    Amazing project! You sound as if you have years of experience. A very timely subject and of particular interest to me as I live near a nursing home which had to be taken over by the National Guard at the height of the pandemic. This seems to have a lot of interesting implications which could take you in a variety of directions. Congratulations on all of your hard work.

  28. TESS TESS TESSTESSTESS TESS!! What a cool project, your research is so relevant and you sound like a pro!

Comments are closed.