Jump to ContentJump to Main Navigation
Next in LineLowered Care Expectations in the Age of Retail- and Value-Based Health$

Timothy J. Hoff

Print publication date: 2017

Print ISBN-13: 9780190626341

Published to Oxford Scholarship Online: October 2017

DOI: 10.1093/oso/9780190626341.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 22 January 2019

(p.197) Appendix How the Study Was Conducted

(p.197) Appendix How the Study Was Conducted

Source:
Next in Line
Author(s):

Timothy J. Hoff

Publisher:
Oxford University Press

This is the second full-length scholarly book in which I have used the straightforward methodological approach of talking with people and dovetailing some of what I am hearing with information gleaned from published sources to create an analysis that merits further attention by researchers, managers, policy makers, and those working in and receiving health care. As an academic who works in both a business school and school of public policy, and as someone who tries to take an applied perspective in all my work, in part because of my own experiences working in the health care industry before coming to academia, I feel compelled to examine research questions that have value for both researchers and practitioners. Thus, like my prior book (Hoff 2010), this one is a sort of hybrid treatise on a complex phenomenon, and the methodological approach reflects that hybrid character.

The research was qualitative. It was approved by the Northeastern University Institutional Review Board. I used data in the form of interviews and archival materials consisting of peer-reviewed journal articles; reports and surveys from legitimate sources such as think tanks, foundations, and professional associations; and news articles from an array of media outlets. During the time period November 2014 through February 2017 I interviewed a total of 80 primary care physicians and individuals representing the patient perspective. Forty-four of these interviews were with doctors, and the remainder with patients. My focus was on adult primary care, so my physician sample was limited to family physicians and general internists. I felt that if strong doctor-patient relationships existed, they would most likely be found within this brand of medicine. My patient sample included individuals ranging from their twenties to seventies in age. I conducted ad hoc literature reviews throughout this same time period on a variety of topics addressed in the book such as the doctor-patient relationship; consumer engagement; retail thinking and principles; health (p.198) care quality; payment reform in health care; workforce issues; trends in health insurance and accreditation, performance measurement, system integration, and health care delivery; and health system innovation.

Qualitative research is inductive and idiographic. This means it is designed to capture the particular, grounded in its normal everyday context, in the hopes of illuminating more general statements regarding specific phenomena. Such an approach helps the researcher know what she or he doesn’t know. Qualitative work requires neither large data sets nor adherence to statistical theory, even though some always mistakenly critique on the basis that it should involve both. Instead, it must meet a litmus test of believability in relation to how the data interpretations resonate for the reader (Van Maanen 1979). Using systematic methods of sampling, data collection, and data analysis (see Miles and Huberman 1994; Patton 2002; Strauss 1987) I used interview and archival data to explore the topics of interest, specifically, the doctor-patient relationship; the relational care elements associated with it such as trust, empathy, and listening; and the alignment of various retail and market-based principles with current views on and experiences with the relationship itself.

Because I was interested in something that by its traditional definition is interpersonal and dyadic, I focused my data collection in the interviews on gaining insights from two groups of individuals as parties to the relationship, obtaining descriptions of their experiences interacting with each other over time. With few exceptions the patients and doctors interviewed were not associated with each other, that is to say, the doctors were not the personal physicians of the patients. This allowed me exposure to a large number of different relationship experiences, although it limited my ability to gain perspective on the same exact relationship from two different vantage points.

I used the interviews to identify key elements of the meaning systems each of the two groups possessed regarding the doctor-patient relationship. I was interested in their views, beliefs, and sensemaking, that is, the personal and highly subjective side of their interpersonal experiences and what they learned from or thought about them. I enabled wide latitude during interviews to expound on their thoughts, trying to elicit from them real-world examples to better appreciate why they might feel or believe a certain way. In this way, I used most of each interview to listen rather than ask questions, keeping them focused on the general topics of interest but allowing what they felt was notable to emerge from our real-time conversation around their thoughts.

Doctors and patients described the “best” kinds of relational experiences with one another and provided their analysis as to why they felt they were the best. I asked them about whether or not they valued these experiences and if so how and why, trying always not to lead them in too directed a manner by naming things like trust or listening. I inserted some probes about various forms of service delivery innovation, such as the use of electronic medical records and teams to see if and how these things shaped their relational care experiences. I also tried as much as possible to get each group to “put their cards on the table” so to speak, in terms of articulating their (p.199) personal beliefs regarding the worth of and possibilities for doctor-patient care that contained heavy relational elements. When contradictions or paradoxes related to their beliefs or views emerged in the conversations, which they did, I probed individuals to try and better understand what these meant for the relationship as a whole. Similar to most other interviews I have performed in the course of prior qualitative research, for many subjects, doctor and patient alike, it took no more than 10 minutes or so to get them talking candidly about their relationship experiences and views.

That said, asking people to self-reflect on something they may not think a lot about during a normal day or week, and something that may conjure up dissonant feelings of frustration, anger, or other negative emotions when they do think about it requires allowing people to pick a day and time of their choosing to talk. Above all else, I have learned from doing qualitative data collection with physicians to fit myself into their schedule, which I did. Even for patients, I was respectful of when they could be available (e.g., not during the day when working). Thus, I did interviews at all different times of the day, especially during evening hours and early morning. To facilitate flexibility in scheduling, I conducted approximately 75 percent of the interviews by phone, with the rest done in person. In addition, I spread the interviews out over a two-year period. In part, I spread the interviews out over a longer time period as part of the analytic strategy, which is detailed below.

Before beginning the book, I perused the extant literature to see if there was anything similar already available. There may have been, but my search did not yield any books of the type that I was setting out to write. I found a couple of books, for example, on how the use of computers and information technology was changing medicine (cf. Wachter 2015), but nothing specifically focused on the evolving doctor-patient relationship, told from the perspective of the participants, in the midst of an industry making a market-focused turn in delivering health care. Surprisingly, much of what is written on the doctor-patient relationship continues to take the form of opinion-editorial type laments across a range of journals, which, while helpful in identifying key big picture issues shaping the landscape, rests little on extended or systematic analysis of real data.

Information on the Interview Sample

Both the physician and patient sampling strategies were guided by a purposive, snowball sampling approach in which I sought out interviewees across a range of demographic characteristics (see Patton 2002). For physicians, my primary stratification variables of interest were age, gender, career stage (positively correlated with age), and type of employment setting (see Table A.1). The sampling strategy involved using key contacts in the primary care physician community to help introduce me to other physicians whom they knew as colleagues, even if they worked in different medical practices. Thus, there was also a convenience aspect to sample selection. (p.200) (p.201) (p.202) There was one physician practice where I recruited 16 doctors for interviewing, with the remaining 28 interviewees spread out across 18 different medical offices.

Table A.1 Physician Interview Sample

Pseudonym

Specialty

Description of Physician Participant

Maggie

Internal Medicine

Rural practice, later career

Martha

Internal Medicine

Rural practice, later career

Betsey

Internal Medicine

Rural practice, later career

Amy

Family Medicine

Urban/suburban practice, resident

Cleo

Family Medicine

Urban/suburban practice, resident

Dana

Family Medicine

Urban/suburban practice, early career

Mary

Family Medicine

Urban/suburban practice, early career

Ola

Family Medicine

Urban/suburban practice, resident

Amy

Family Medicine

Urban/suburban practice, resident

Wilma

Family Medicine

Urban/suburban practice, early career

Billy

Internal Medicine

Rural practice, later career

Bill

Family Medicine

Rural practice, later career

Ron

Family Medicine

Rural practice, later career

Steve

Internal Medicine

Rural practice, later career

Sam

Family Medicine

Rural practice, later career

Karen

Family Medicine

Urban/suburban practice, early career

Paul

Family Medicine

Urban/suburban practice, resident

Barbara

Family Medicine

Urban/suburban practice, early career

Annie

Family Medicine

Urban/suburban practice, early career

Liza

Family Medicine

Urban/suburban practice, later career

Rachel

Family Medicine

Urban/suburban practice, early career

Pete

Family Medicine

Urban/suburban practice, early career

Sid

Family Medicine

Urban/suburban practice, later career

Vince

Family Medicine

Urban/suburban practice, later career

Sonny

Family Medicine

Urban/suburban practice, later career

Hal

Internal Medicine

Urban/suburban practice, later career

Ronny

Family Medicine

Urban/suburban practice, later career

Bradley

Family Medicine

Urban/suburban practice, later career

Mike

Family Medicine

Urban/suburban practice, later career

Keith

Family Medicine

Urban/suburban practice, later career

Bobby

Family Medicine

Rural practice, later career

Bridget

Family Medicine

Rural practice, resident

Karl

Internal Medicine

Urban/suburban practice, later career

Burt

Internal Medicine

Urban/suburban practice, later career

Curt

Internal Medicine

Urban/suburban practice, later career

Owen

Family Medicine

Urban/suburban practice, later career

Pam

Family Medicine

Urban/suburban practice, early career

Scott

Family Medicine

Urban/suburban practice, early career

Rick

Internal Medicine

Urban/suburban practice, later career

Uma

Family Medicine

Rural practice, early career

Harry

Family Medicine

Rural practice, resident

Sarah

Family Medicine

Urban/suburban practice, resident

Olivia

Family Medicine

Urban/suburban practice, early career

Claire

Family Medicine

Urban/suburban practice, resident

I did not wish to talk only with older doctors who, while having more career opportunities than younger doctors to develop deep patient relationships, still had cut their relational teeth in a health care system very different from the one they work in today. I wanted them to talk about those contrasting work contexts, but I also wanted to get the perspectives of younger doctors, even residents, who had less of a past frame of reference to guide how they interpreted their current relationship experiences. Thus, I interviewed nine current primary care physician residents as part of the total group of 44 physicians. Overall, 20 physicians (including the residents) could be classified as being “early career,” that is, less than 10 years of experience practicing medicine.

I wanted to survey a fairly equal number of male and female physicians, which I was able to achieve. Twenty-one of the 44 physician interviewees were female. I was particularly interested in seeing whether there were differences around the doctor-patient relationship that would emerge between male and female physicians, given the different practice styles often associated with each group. No gender-based differences emerged, however, during analysis, which surprised me a bit but also spoke to the universal nature of some of the experiences, sensemaking, and beliefs presented here, at least in terms of how most doctors perceived the best types of relationships, and how they saw various innovations, such as team care or electronic medical records. The 44 physicians worked in 19 distinct primary care practices, some of which were owned by larger corporate entities such as hospitals or health systems, and others which were group practices technically owned by the doctors but reimbursed through a variety of insurances, such as Medicare and private payers. The practices were probably equally split in terms of large-, medium-, and small-size practices (i.e., number of providers on staff). I counted any practice with fewer than five doctors as small, six to ten doctors as medium-sized, and more than ten doctors as large.

Geographically, I interviewed 12 rural and 32 urban or suburban-based doctors; with a majority of the latter group working in and around the metro Boston area. No meaningful differences were noted in the data between rural, suburban, or urban doctors. Almost all worked under the same conditions—packed daily schedules with shorter visits; patients churning in and out of their practices; a heavy emphasis on documenting and reporting quality metrics; workflows that increasingly firewalled them from more “mundane” patient care tasks (and thus more from patients); heavy use of the electronic medical record; and team care that gave responsibilities previously involving physicians over to other office personnel, such as medical assistants and nurses. A few of the practices may have had a higher percentage of more complex patients walking through the door each day, or served a more ethnically diverse population, but that was it in terms of differences.

The physician sample is representative of the larger everyday milieu in which primary care physicians now find themselves, and in that key sense getting doctors from another part of the country to talk with me would not necessarily add anything (p.203) new based on the working conditions, which are similar regardless of locale. There is no empirical proof, for example, that a primary care physician working for Kaiser Permanente or Geisinger (often held up as exemplars of physician workplaces) is any less harangued, relationship driven, or relationship-capable than the next doctor, and as I said, most of my doctors worked for larger organizations with some level of “service integration.” Ten of the 44 physicians were general internists (with two trained in both internal medicine and pediatrics), and the remainder were family physicians. The heavy skewing toward family medicine was not intentional, but I did probably assume that such doctors were more likely, because of their training and socialization, to place heavier emphasis on relational care with their patients. They also would be the doctors that patients would most expect to interact with them in this way.

For the 36 patients, I sought to stratify the sample based primarily on age and gender. I thought these variables would be partial proxies for other important stratifying criteria such as health status and types of care needed, which would be more difficult to obtain in advance from potential interviewees (Table A.2). The sample as a whole reflected a variety of ages and was almost equally split by gender. I did ask each patient interviewee, to the extent comfortable, to provide a snapshot of their health status at the outset of the interview, that is, whether they were in generally (p.204) (p.205) good health, suffered from one or more chronic diseases, had a more serious disease like cancer, or accessed the primary care system on a frequent basis. From this knowledge, I categorized eight of the interviewees as being in poorer health compared with others. These patients generally discussed accessing the health care system more than others in the sample.

Table A.2 Patient Interview Sample

Pseudonym

Description of Patient Participant (senior is someone over 65 years of age)

Hallie

Multiple conditions/diseases, fifties

Hadley

Multiple conditions/diseases, fifties

Krystal

Healthy, fifties

Winona

Healthy, fifties

Scarlett

Healthy, twenties

Raymond

Healthy, senior male

Will

Single chronic disease, senior male

Edward

Healthy, thirties

Wanda

Healthy (past history of disease), senior female

Teddy

Healthy, thirties

Susan

Healthy, twenties

Lucy

Healthy, twenties

Bette

Healthy, thirties

Tonya

Healthy, twenties

Wayne

Multiple conditions/diseases, senior male

Fran

Healthy (past history of disease), thirties

Darlene

Healthy, thirties

Raul

Healthy, fifties

Cliff

Healthy, fifties

Barry

Healthy (past history of disease) sixties

Bart

Multiple conditions/diseases, sixties

Janell

Healthy, forties

Jerry

Healthy, forties

Matt

Healthy, senior male

Renee

Healthy, thirties

Pete

Healthy, twenties

Rebecca

Healthy, forties

Siobhan

Healthy, fifties

Rich

Healthy, senior male

Gabrielle

Healthy, forties

Carly

Multiple conditions/diseases, thirties

Tommy

Multiple conditions/diseases, sixties

Anthony

Multiple conditions/diseases, fifties

Mia

Healthy, fifties

Grady

Multiple conditions/diseases, fifties

Donna

Healthy, forties

Patients were recruited through personal contacts I had with several individuals, whom I interviewed for the study. These individuals introduced me to friends or family members for interviewing, although in no case did I recruit more than a couple of individuals from the same person. While pursuing the snowball sampling approach, I also took care to try and gain enough numbers of patients possessing the types of stratification features (e.g., age, gender) that interested me. There were some limitations to the patient sample. For example, almost all of the patients were white and could be considered “middle-class” in terms of how they presented such things as their employment status, insurance status, and where they lived. Only a handful of patients came from lower socioeconomic backgrounds. In theory, overall care experiences for patients from lower socioeconomic backgrounds might be different compared with white, middle-class patients, and this could affect the findings. For example, one might expect stronger relational features and deeper, longer-term doctor-patient relationships in the higher socioeconomic status (SES) group given “better insurance” and better access to doctors in their geographic locales. Or, perhaps the opposite, that is, one might think that lower SES patients, who might have more health issues, could have better ongoing relationships with a single doctor. In comparing the interview data across the two groups, however, no differences of this type emerged.

As mentioned earlier, statistical generalizability was not required for a qualitative study like this, and so a non-probability sampling scheme like that outlined here was appropriate. That said, could leaving out of the sample patients or doctors with other types of demographic characteristics bias the findings in some way? I conducted a series of comparisons in the data, for instance, between patients who were healthier versus sicker; younger versus older; male versus female; and living in urban versus suburban areas. No meaningful differences in any of the major themes or descriptions presented in the book were observed across these patient groups. They all reported similar experiences, expectations, sensemaking, and beliefs about their relationships with doctors; and what they ultimately took away from those relationships. In fact, it was striking to me how similar these types of findings were across the data set as a whole, with the physician sample also uniform across different subgroups as to the themes identified and experiences and views described.

Analyzing the Data and Arriving at Interpretations

Having conducted a lot of interview-based qualitative research over the years, I pursued a similar analytic strategy as in my prior work. Each interviewee received a pseudonym so there was no way of identifying them. All interviews, with the exception (p.206) of five were digitally recorded and transcribed onto computer. I listened to each of the 75 interviews in their entirety at least once, as well as read the transcripts of all 75 interviews. For the five not recorded, I reviewed my notes taken during the interviews. Some interviews were extremely informative and so helped me to articulate preliminary findings that I sought to validate using other interviews. I also was keen on looking for specific storytelling within the interviews, and identifying participant experiences that might help to illuminate the wider themes and patterns that were emerging as I reviewed and coded across the wider data set. For example, the “low expectations” patients held of the doctor-patient relationship in primary care, described and analyzed in Chapter 5 and Chapter 6, were first identified through descriptive coding of multiple interviews, then further validated and made more specific by finding stories in select interviews that captured the essence of the descriptive finding. In this way, I went back and forth in the analytic phase between creating larger, more systematic interpretations and trying to understand anecdotes that were unique to one person’s experience, yet at times held the key to better understanding what the interpretation meant at a more abstract level of analysis.

For a theme, pattern, or interpretation to emerge from the data as meaningful, it required support from a majority or more of the interviews, whether that interview sample was the entire physician and patient data set, only the physicians, or only the patients. Each interview generated approximately 20–25 double-spaced pages of text, so there was a significant amount of qualitative information to review and analyze. For a portion of the interview sample, particularly the first 20 doctor and 20 patient interviews, I used Atlas.ti software to help perform descriptive coding that could begin to isolate views, opinions, beliefs, and experiences that were similar across a portion of the given data set. I also used Atlas.ti tools like the “word find” function to isolate instances in the data where respondents were discussing things like “trust,” “communication,” and “respect.” I did not only look for those dynamics where the actual words were explicit in the data, but instead started with textual chunks that openly talked about these words, and then examined other places in the same interview for additional discussion.

Most of the findings presented in the book have deep and widespread support within the interview sample. There are a few findings of interest, however, that may not have achieved robust support, yet were rich enough in their descriptive power to reveal some phenomenon or insight, which, although remaining somewhat exploratory or nascent, nonetheless informed questions that were of interest in the book. Thus, some findings presented in the book are there not only because of their widespread or descriptive support in the data, but also because they speak to the larger topics that piqued my interest in writing the book in the first place. They also raise questions or suggest implications that are important for thinking about the present and future viability of the doctor-patient relationship within a market-based health system.

The book is organized in the following manner: Several introductory chapters (a) describe and analyze health care delivery system forces to illuminate a contextual (p.207) backdrop in which doctor-patient relationships, relational care, and our overall expectations of these phenomena are embedded; (b) highlight the importance and value of doctor-patient relationships in health care (using systematic research findings); and (c) describe and critique a core foundational philosophy of market-based, consumer-oriented health care in the modern age, namely, that of retail thinking. These first three chapters provide the grist for asking questions regarding how doctors and patients experience their interpersonal relationships, and what these experiences mean for how each group thinks about the other, in terms of expectations and preferences.

Chapters 4, 5, and 6 form the ethnographic core of the book, and the specific findings presented in each of them are analyzed in-depth at the end of the respective chapters for their implications related to management, policy, patient care, the doctor-patient relationship, and further evolution of a consumer-oriented health care system. The final chapter circles back to discuss and analyze the prospects for strong doctor-patient relationships given the experiences and beliefs of doctors and patients presented. It also offers up modest suggestions for saving relational care within a health care delivery system that is moving forward in adopting corporate care, retail thinking, service standardization, and transactional business models.

As I also noted at the end of my 2010 book, the present book is intended for a variety of audiences, and the findings and analyses offered can be construed on several levels. For instance, is the book a sustained critique on the movement of American health care delivery into a more “Walmart-like” phase of existence where patients become “consumers” and retail thinking grows ascendant? Is it about how recent external forces associated with the quality and “value” movements have undermined the more humane, relational elements of our system permanently? One could read it and go in either or both of these directions. At its core, however, the book is about the threats under which the doctor-patient relationship now exists, and how that relationship is eroding in ways that have implications for doctors, patients, and the system as a whole. If the book sounds ideological at times, that tone is justified by the substance of the interview data and analysis presented here, which suggests good reasons to be both concerned and critical. (p.208)