Stay informed: Sign up for eNews Subscribe
Read the Appendix

Technical Appendix 1: Interview Methods

The interviews for this project involved 119 persons and were conducted in the following formats:

  • individual interviews with 34 people who have difficulty walking, almost all with progressive chronic conditions; some interviews involved family members and almost half occurred in interviewee's home or workplace
  • focus groups (i.e., group interviews) involving 22 people with difficulty walking because of progressive chronic conditions
  • individual interviews with 22 practicing physicians
  • separate focus groups with physical and occupational therapists (7 and 6 participants, respectively)
  • individual interviews with 28 other persons, including medical directors of large health insurers, other clinicians (e.g., nurses and social workers), clinical educators, persons with a family member (e.g., a parent or sibling) with walking difficulties, assistive device vendors, disability rights advocates, and policy experts
For a brief discussion about paying for mobility aids (When Walking Fails, chapter 14), I used responses from an additional focus group conducted for another study (funded by the Agency for Healthcare Research and Quality, HS10223). This focus group included eight women, all wheelchair users, and took place on 27 September 2000.

For three years I kept a journal recording informal conversations with my "rolling focus group"—taxi drivers, flight attendants, shop clerks, strangers who approached me. I analyzed only the formal interviews for this work but occasionally offer snippets from the rolling focus group to illustrate points.

The interviewers audiotaped all individual interviews, then an administrative assistant transcribed these audiotapes verbatim. Focus groups were both audiotaped and videotaped, with audiotapes transcribed verbatim by an experienced transcription service. I compared typed transcripts of the focus groups to videotapes of those group interviews to identify speakers and to ensure the accuracy of the transcriptions. We conducted all interviews in spoken English.

Throughout this project, I quote people in their own words based on the typed verbatim transcripts. Routine speech, however, contains many sounds, words, and phrases—such as "um," "ah," "like," and "you know"—that are tedious to read. In addition, people often stop and start sentences, repeat phrases, switch verb tenses or pronouns midstream, and speak thoughts that are understandable in person (i.e., because of gestures and facial expressions) but unintelligible afterward when written verbatim. Because I conducted the vast majority of individual interviews and attended all focus groups except one (by design, as described below), I could decipher most of the confusing portions of the transcripts.

To improve clarity and readability, I therefore edited the transcripts. As much as possible, I kept people's own words and grammatical constructions but generally removed sounds and repetitious patterned phrases (e.g., "uh," "like," "you know"), eliminated false starts of sentences, and clarified as necessary. When comments that I quote sequentially actually occurred at different points during the interview, I note the gap with ellipsis points ( . . . ).

The Beth Israel Deaconess Medical Center's Committee on Clinical Investigations approved this study. We obtained written informed consent from all interviewees with mobility difficulties interviewed in person and from the physical and occupational therapists who participated in focus groups. We obtained verbal consent from other interviewees. As part of the consent agreement, I agreed to protect the anonymity of the interviewees. Therefore, with a few exceptions, I use pseudonyms for all interviewees and change small details about their personal histories (such as number of children and precise job). The exceptions involve several people I interviewed as "experts," from whom I obtained explicit written permission to use their names. I offered persons with mobility difficulties $75 for participating, although ten individuals refused payment.

Unlike the NHIS, which queried nationally representative individuals and households, my interviewees are a convenience sample—people identified through various means (described below) who agreed to participate. Focusing on adults, I aimed for diversity in the ages, races, and chronic conditions of persons with impaired mobility, but the sociodemographic and clinical characteristics of my interviewees certainly do not parallel those of the national population. The majority of interviewees with impaired mobility lived in eastern Massachusetts, as did all of the clinicians. Therefore, the interviewees' views may differ in some respects from those of adults nationwide, especially persons from linguistic minorities.


Interviewing People with Mobility Difficulties

In qualitative studies, the characteristics of the interviewer, moderator, and other participants can affect people's willingness to respond openly. My wheelchair use undeniably influenced how interviewees related to me, facilitating certain interactions but at times having the opposite effect. Based on my rolling focus group encounters, I suspected that many respondents would want to question me and perhaps divert the interview toward their interests. On balance, however, I believe that my wheelchair use made interviewees more open, as others have found (C. Miles-Tapping and J. L. MacDonald, "Lifestyle Implications of Power Mobility," Physical and Occupational Therapy in Geriatrics 12, no. 4 [1994]:31-49). I either conducted or was present at all interviews of persons with mobility difficulties with one exception—a focus group involving African American participants, described below. In interpreting my results, I therefore made respondents' perceptions of my wheelchair use an important consideration.


Individual Interviews

I conducted all thirty-four individual interviews of persons with impaired mobility. The interviews occurred from 25 April 1997 through 5 August 1998, and each lasted about two hours. I carried out fourteen interviews in people's homes (one in metropolitan Washington, D.C., the others in eastern Massachusetts), two in their workplaces, one at the person's church, one in a hotel I was visiting in the Midwest, three by telephone (two with persons in the South and one in California), and the remaining thirteen in my office or my hospital's outpatient clinical center.[Note 1] In five instances, spouses participated in the interview, and one of these also included an adult son. In four additional situations, I met the spouse briefly, but he or she did not join the interview.

I used a small dictation tape recorder to audiotape the interviews, turning on the recording only after obtaining written informed consent. After each interview, I wrote extensive field notes, describing the personal appearance and demeanor of the interviewee(s), his or her mobility, the accessibility and attributes of the interview location, conversations before and after the audiotaping, and my personal feelings about the interviews.

No official "registry" exists of persons whose mobility is impaired by progressive chronic conditions. Therefore, I recruited interviewees from various sources that include requests to my physician colleagues to recommend appropriate patients; word of mouth; an e-mail broadcast through a housing advocacy group; and solicitations among parishioners at a church outside Boston. My physician colleagues typically approached their patients first to ascertain willingness to speak with me. In only a few instances, when the physician had not made an initial contact, persons refused to participate, generally denying that they had mobility problems. For my other recruitment strategies, people had to take the positive step of approaching me.

Although I interviewed a convenience sample, my thirty-four interviewees nevertheless had diverse sociodemographic characteristics, as follows:

  • 3 were under 40 years of age; 8 were 40 to 49; 6 were 50 to 59; 7 were 60 to 69; 8 were 70 to 79; and 2 were 80 years old or older
  • 21 were women
  • 25 were white, and 9 were black
  • 10 had a high school education or less; 8 had some college or were currently attending school; 8 had completed college degrees; and 8 held graduate degrees
  • 15 had annual incomes less than $20,000; 7 had incomes between $20,000 and $40,000; and 12 had incomes over $40,000
With four exceptions, all persons had progressive chronic conditions acquired in adulthood, including degenerative joint disease or osteoarthritis, back problems, multiple sclerosis, Parkinson's disease, amyotrophic lateral sclerosis, myasthenia gravis, peripheral vascular disease, diabetes mellitus, obesity, asthma, chronic obstructive pulmonary disease, stroke, and heart conditions. Many people had multiple conditions (the most common combination was diabetes, obesity, and arthritis) and could not identify a single cause for their mobility difficulties.

In four instances, people did not have typical, acquired, progressive, chronic conditions, but their debilities were nevertheless chronic. Jody Farr has a rare form of muscular dystrophy, obviously a congenital condition, but it was undiagnosed until she was in college. From that point onward, her impairments progressed. Barbara Forrest's internist, who recommended that I speak to her, still had no complete explanation for what caused her to become acutely unable to walk a few years previously. Ms. Forrest had recovered much of her ability to stand and walk, but she remains slow and falls frequently, risking injury. Natalie Strong had cerebral palsy, a congenital condition, affecting only her legs. However, repeated falls produced progressive impairment, much like an increasingly debilitating condition. Finally, Joe Warren had scoliosis, a progressive curvature of the spine, which required surgery and prevented him from walking for roughly a year as a young adult. Today, however, his major debility relates to a partial spinal cord injury from a car accident.

Only four people did not use any mobility aid, although each of these persons walked slowly and could go only short distances. Most people used one mobility aid, while others employed different mobility aids in different settings. Thirteen used a cane at least some of the time, while six used walkers, two used crutches, seven used manual wheelchairs, three used power wheelchairs, and seven used scooters.

During the interviews, I asked open-ended questions following a semistructured interview guide. Depending on the interviewees' responses, I often changed the order of specific questions but covered all topics before concluding the interviews. The guide encompassed my areas of interest and reflected literature reviews, insights from the rolling focus group, and personal experiences. I treated the first three interviews as a pilot test but did not modify the interview guide substantially after these interviews. The order and general topics of the questions were as follows:

  1. brief biographical history
  2. description of walking difficulty, physical sensations, causes, and course
  3. experiences with medical care system, physicians, and other clinicians
  4. strategies for getting around, including use of mobility aids
  5. health insurance and financial considerations
  6. effect of mobility difficulties on daily life, family, friends, employment (as appropriate), sense of self, and plans for the future
  7. interactions with strangers around mobility difficulties and perceptions of societal attitudes
  8. suggestions for services that would help
  9. whether persons view themselves as disabled and what they know about the ADA
  10. sociodemographic characteristics
Some interviewees wanted to know something about me and my life with my scooter.[Note 2] I asked to defer questions until after completing the interview protocol and, turning off the tape recorder, then spent as long as people wished answering their inquires. Occasionally, people revealed important personal information during these postinterview sessions, and I recreated these conversations by writing field notes soon afterward.

Depending on the referral source, I attempted to determine how my interviewees were doing several years later. In certain instances, I could not easily trace interviewees without breaching their confidentiality; therefore, my follow-up is incomplete. At least eight interviewees have died, some within months of the interview. Throughout the text, I refer to interviewees who have died in the past tense.


Initial Focus Groups

Individual interviews offer rich insight into individuals' experiences. Group interviews, however, provide not only individuals' insights but also an opportunity for stimulating exchanges—as participants share or challenge opinions, make suggestions, agree and disagree. The goal of the focus group is not to reach consensus but to elicit a full range of viewpoints and experiences. Therefore, I organized three focus groups, each drawing participants who have difficulty walking because of progressive chronic conditions.

Moderating a focus group requires special skill in listening, facilitating open expressions of opinions, encouraging contributions from all participants, and moving the discussion forward to cover the designated topics. I therefore worked with Lisa LeRoy, M.B.A., an experienced focus group moderator, to design the moderator's guide and conduct the initial focus groups. Based on my early individual interviews, we created a focus group moderator's guide specifying open-ended questions to address topics parallel to those in the individual interview during a two-hour period.

From early individual interviews and published literature, I believed that belonging to a racial minority might further compound any social stigmatization generated by having a physical functional debility, such as impaired walking. I therefore wanted to conduct a focus group in which participants could freely raise the role of race. And—though focus group experts disagree on whether participants of groups where race might be salient need all be of the same race—I decided that at least one focus group should involve only black individuals (including moderator, videographer, research assistant, and interviewees), and that I would not attend that focus group.[Note 3] Ms. LeRoy and I identified and trained the moderator, Maritza Hall, instructing her to ask probing follow-up questions about the role of race at appropriate points. A nurse-manager at a neighborhood health center recruited interviewees based on her knowledge of patients at the center and our specifications (black persons with mobility impaired by progressive chronic conditions).

I also wanted one focus group to include only wheelchair users, anticipating that they might speak more freely among themselves about the challenges of wheeled mobility. I recruited participants through outpatient social workers and held the meeting at a wheelchair-accessible facility. Nevertheless, transportation to and from the focus group proved to be difficult for some potential participants, and this focus group involved only four interviewees.

The three focus groups were as follows:

  • 16 March 1998, moderated by Ms. LeRoy, at a church in a working- and middle-class town outside Boston. The ages of the six participants were 37, 49, 66, 69, 70, and 93 years. Five were women; all were white. Three had only high school educations, while three had college degrees. Four participants had arthritis, and two had peripheral vascular disease, but four had multiple conditions (e.g., diabetes and heart disease). Three used a cane (although one also had a scooter), and two did not use mobility aids.
  • 10 June 1998, moderated by Ms. LeRoy, at Beth Israel Deaconess Medical Center. The four participants were 47, 56, 57, and 65 years of age. Three were men; three were black. One had only a sixth-grade education; two had finished high school, one with vocational training; and one had a master's degree in special education. The chronic conditions included diabetes (one with an amputation of his lower leg), arthritis, coronary artery disease, postpolio syndrome, and cerebellar ataxia. Two people had multiple conditions. All four used wheelchairs: two used four-wheeled power wheelchairs, one a manual wheelchair, and one a scooter.
  • 23 July 1998, moderated by Ms. Hall, at a neighborhood health center in Boston. The eleven participants were 46, 53, 55, 59, 61, 64, 68, 71, 73, 74, and 77 years old. Ten were women; all were black. Two had a middle school education; five had graduated high school; and four had some college education. Their chronic conditions included arthritis, stroke, asthma, heart disease, multiple sclerosis, and gout. Five persons had multiple conditions.
After each focus group, the moderator, research assistant (who helped with logistics and witnessing the informed consent), and I had a debriefing session and talked systematically about what we had learned from the group.


Follow-up Focus Groups

After completing the individual interviews, initial focus groups, preliminary analysis of the interview texts, and analyses of the NHIS-D (see Iezzoni et al. 2000a, 2000b, 2001), I conducted two additional focus groups. My purpose with these groups was to present preliminary findings to participants and ask their opinions—did the results make sense given their personal experiences, viewpoints, and observations? I moderated these focus groups and designed the moderator's guide around the following six questions:

  1. How common are walking difficulties?
  2. How does walking difficulty make people feel emotionally?
  3. What are people's experiences with mobility aids?
  4. How does walking difficulty affect daily life?
  5. How does walking difficulty affect people's ability to get health-care services?
  6. How do society's attitudes toward walking difficulty affect people?
For each question, I assembled brief illustrative quotations from prior interviews or key findings from the NHIS-D. During the follow-up focus groups, I read these findings to participants and asked for their responses.

For the first follow-up focus group (15 July 1999), I invited five people who had been particularly articulate and insightful when interviewed before: Gerald Bernadine, Lester Goodall, Michael Sexton, Stella Richards, and Louisa Delarte.[Note 4] For the second follow-up focus group (28 July 1999), I reinvited the eleven persons who had participated in the neighborhood health center focus group.[Note 5] One person had died, and she was replaced by another woman who had not been interviewed previously. Therefore, the two follow-up focus groups involved fifteen previous interviewees and one new person.


Interviews with Clinicians and Experts

All interviews with clinicians were conducted in person, but many expert and other interviews occurred over the telephone. Given my experiences in medical school and as a faculty member in primary care medicine, I anticipated that physicians might hesitate to speak openly about mobility difficulties to someone using a wheelchair. Telephone interviewees would not necessarily know that I use a scooter (unless I told them or they knew of me already), but clinicians interviewed in person certainly would. I therefore enlisted the assistance of David Stone, Ph.D., who does not use a mobility aid, to conduct a set of individual physician interviews. Ms. LeRoy moderated the focus groups of physical and occupational therapists, although I was present and identified as the project director.



I conducted twelve interviews with practicing physicians, and Dr. Stone conducted ten. My interviews occurred from 29 July 1996 through 13 May 1999, while Dr. Stone's ranged from 10 February 1998 through 15 July 1998. I chose a purposeful sample of clinicians I knew (six primary care physicians and six specialists), while Dr. Stone had a more "random" sample of primary care physicians. We took the roster of a large integrated health delivery system with physician practices throughout metropolitan Boston and randomly selected names in geographically dispersed areas. However, that strategy had limited success. Most physicians said they were too busy for the interview; those who agreed to participate were often younger doctors who did not yet have full patient panels. We identified six of Dr. Stone's ten physician interviewees using this approach; the other four were practitioners with whom Dr. Stone was already familiar, two of whom also knew me.

The physician interviews took from half an hour to just over an hour and followed a semistructured script of open-ended questions I had prepared after several initial interviews and reviewed with Dr. Stone. The script covered the following topics:

  1. training (in medical school, residency, and afterward) in assessing lower-extremity mobility, including descriptions of specific patients who personified this topic
  2. current practices around evaluating and addressing lower-extremity mobility
  3. referrals to other physicians and physical and occupational therapists
  4. experiences with mobility aids
  5. observations about patients and their attitudes about walking, including questions about specific issues (e.g., "malingering," "use it or lose it")
  6. effects of the current health-care environment, including financial and time pressures
My fears about the physician interviewees' reticence facing a wheelchair user proved groundless. My colleagues spoke remarkably openly, often confessing almost complete ignorance on the topic or virtual indifference.

In addition to the formal, individual physician interviews, I also moderated two physician focus groups:

  • nine fellows training in general internal medicine and three attending primary care physicians (19 February 1997)
  • six fellows training in general internal medicine and two attending primary care physicians (20 January 1999)
Unlike the other focus groups, which were both audiotaped and videotaped, these focus groups were only audiotaped (the second one had very poor quality sound). The groups did yield useful insight into medical school and residency training, which the fellows had only recently completed. However, because of technical problems with the transcripts (they did not always show who was speaking), I did not formally analyze these focus groups and relied on them only for background information.


Physical and Occupational Therapists

Ms. LeRoy moderated two separate focus groups involving seven physical therapists (3 June 1999) and six occupational therapists (19 May 1999). We recruited therapists from the outpatient and home-care practices affiliated with Beth Israel Deaconess Medical Center, the Boston University's Sargent College of Health and Rehabilitation Sciences , and the local visiting nurses association. All therapists had experience working with people with impaired lower-extremity mobility caused by progressive chronic conditions.


Experts and Others

I conducted telephone interviews with six health insurance representatives, identified largely through contacts at the American Association of Health Plans, Health Insurance Association of America, and Massachusetts Blue Cross. I had hoped to interview more insurance representatives, especially medical directors at health plans. However, health insurance representatives were unenthusiastic about being interviewed, perhaps because of the climate of "managed-care bashing" in 1998 and 1999. In addition, I conducted twenty-two interviews with other persons with an interest in or experiences with walking difficulties, including nine disability- or disease-specific patient rights advocates (e.g., representatives from the Arthritis Foundation, National Multiple Sclerosis Society, and Parkinson's Disease Foundation), persons representing the assistive technology industry persons representing the American Medical Association, health policy experts, clinical educators, nurses, social workers, and wheelchair vendors. Before beginning each telephone interview, I secured permission to audiotape the conversation.



For each category of interviewee, I learned little new from the later interviews. Although the details of lives or experiences obviously differed, people repeated the same broad themes or concepts already raised by others. Qualitative researchers view this absence of new information among late interviewees as a useful sign, suggesting that although the absolute number of interviewees is relatively small, the interviews captured the critical, shared experiences.

Once transcribed, the interviews produced hundreds of pages of texts. I conducted all analyses myself, although I discussed ideas and findings with Ms. LeRoy and Dr. Stone. I explored but decided against using computerized software to code the texts.[Note 6] I read through each text several times, highlighting key concepts in the margins. I then printed out segments of texts (sometimes a paragraph, sometimes several pages) and placed these in a file with other texts relating to the same topic. Through this process, I produced about one hundred subject-specific files for the interviews of persons with mobility problems and thirty for the physician interviews. In preparing to write this book, I reviewed folders applicable to the subject at hand, writing a summary of key points raised by each interviewee's comments and highlighting particularly cogent quotations. I therefore reviewed these texts multiple times over more than two years.

The two follow-up focus groups helped validate my analyses. The major goal of these follow-up interviews was to test my preliminary interpretations and hypotheses. These focus groups generally validated my conjectures but often provided more colorful or pointed ways of articulating the ideas. My analyses, however, were certainly limited by the absence of extensive external review and by being inextricably tied to my own experiences as someone with mobility limitations.


Notes for Technical Appendix 1: Interview Methods

1 In general, people who were poor or African American preferred to meet in my office. For all persons residing in eastern Massachusetts, I offered to come to their homes or the most convenient place for them. To get there, I rode my scooter, took a wheelchair-accessible taxi, or was driven by my administrative assistant Ron. Ron generally dropped me off and made sure I was settled before leaving, then returned about two hours later. He did not participate in the interviews but often generated lively conversation as we left.back to text

2 People asked about the nature of my condition, how long I had had MS, what it was like getting around with my scooter, and other primarily logistical questions. With a few exceptions, they did not ask personal questions like those I had asked them. While securing the informed consent, I had told people that they could refuse to answer any question for any reason, but few people declined to answer my inquires.back to text

3 Although I attended the focus groups moderated by Ms. LeRoy, I did not speak until the very end, when Ms. LeRoy turned to me and asked if I had any questions. My questions generally addressed comments raised earlier by interviewees. I had been introduced as the director of the project, and everybody saw that I use a wheelchair.back to text

4 Other people had been equally articulate and insightful but could not participate (e.g., Sally Ann Jones, who lives in the Midwest; Walter Masterson, who had become ventilator-dependent; Esther Halpern and Tina DiNatale, who have transportation difficulties; and several, like Lonnie Carter and Mike Campbell, who had died).back to text

5 This was the first time that the neighborhood health center interviewees had met me, although I felt I knew them from having watched the videotape of the initial focus group. I introduced myself, updated them on the activities of Ms. Hall and the research assistant they had met the year before, and then explained the purpose of the follow-up focus group.back to text

6 At the time, the software available for conducting text analyses relied heavily on manipulation of computerized text files using mouse-driven actions. Intensive use of a mouse is difficult for someone with limited manual dexterity. My time was better spent reading through texts, rather than wrestling with the mouse.back to text