Iezzoni LI, Davis RB, Soukup J, O'Day B. Quality Dimensions That Most Concern People With Physical and Sensory Disabilities. Arch Intern Med. 2003;163(17):2085-2092. doi:10.1001/archinte.163.17.2085
Copyright 2003 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2003
People with physical and sensory disabilities face important challenges in obtaining high-quality health care. We examined whether persons who are blind or have low vision, who are deaf or hard of hearing, or who have mobility impairments or manual dexterity problems are satisfied with the technical and interpersonal aspects of their care.
By using a 1996 nationally representative survey of 16 403 community-dwelling elderly and disabled Medicare beneficiaries, we identified persons with disabling conditions. Adjusted odds ratios (AORs) and 95% confidence intervals (CIs) controlled for demographic characteristics and managed care membership in predicting dissatisfaction with 12 dimensions of care.
Of an estimated 33.58 million noninstitutionalized Medicare beneficiaries, 64.1% (an estimated 21.51 million) reported at least 1 potential disability and 29.5% (an estimated 9.89 million) reported more than 1 potential disability. Roughly 98% of people with and without disabilities believed their physicians were competent and well trained. But statistically significantly more people with disabilities reported dissatisfaction with care for 10 of the 12 quality dimensions. Persons reporting any major disability were more likely to be dissatisfied with physicians completely understanding their conditions (AOR, 2.4; 95% CI, 1.9-3.1), physicians completely discussing patients' health problems (AOR, 2.4; 95% CI, 1.9-2.9), physicians answering all patients' questions (AOR, 2.3; 95% CI, 1.7-3.1), and physicians often seeming hurried (AOR, 1.6; 95% CI, 1.4-1.9).
Persons with disabilities generally reported positive views of their care, although they were significantly more likely to report poor communication and lack of thorough care. These findings held regardless of the disabling condition. Thoughtful systematic approaches are required to improve communication and to reduce time pressures that might compromise the health care experiences of people with disabilities.
THE QUALITY of health care is multidimensional, encompassing multiple technical and interpersonal attributes. Ensuring confidence and comfort with all dimensions is important for everybody seeking health services. However, strategies for improving specific aspects of care could vary for patients with differing sensory and physical abilities. In particular, persons with potentially disabling conditions—people who are blind or have low vision, who are deaf or hard of hearing, or who have impaired mobility—confront special physical and communication challenges within the typical general medical setting.
Facilitating physical access and communication for persons with special sensory and physical requirements necessitates careful forethought, including user-friendly physical spaces; accessible furnishings and equipment; appropriate communication strategies, such as braille or large-print educational materials, American Sign Language interpreters, and telecommunication devices; and trained physicians, nurses, and office staff. Nevertheless, despite such mandates as the 1990 Americans with Disabilities Act, persons with disabilities still face numerous physical, communication, and attitudinal barriers within health care settings.1- 7 Overall, people with disabilities are less satisfied with their health care than others.8
Administrative, financial, and organizational factors exacerbate negative experiences, as do reduced appointment times and harried physicians. Although people with disabilities do not necessarily have acute or chronic illnesses, many have narrow margins of health and need more time with their physicians to address complex issues.9- 11 Often, health insurance inadequately covers items or services required by people with disabilities, such as long-term physical therapy or assistive technologies.12- 17 In these circumstances, discussing options and developing alternative approaches to meet patients' needs may require additional time with clinicians.
Little systematic information exists about how persons with specific disabilities perceive their care. Herein, we look at patients' views of technical and interpersonal dimensions. For persons with disabling conditions, technical quality is important because of patients' frequently complicated medical conditions. Interpersonal quality is salient given potential communication barriers and continuing societal discomfort with disabilities. We use a nationally representative survey of Medicare beneficiaries to examine our hypothesis that persons with potentially disabling conditions perceive more problems with technical and interpersonal quality than do other people.
We examined responses from 16 403 community-dwelling (noninstitutionalized) Medicare beneficiaries interviewed in the 1996 Medicare Current Beneficiary Survey (MCBS). Medicare covers almost all elderly persons (those aged ≥65 years) and people younger than 65 years who have met the Social Security Administration's definitions of disability (unable to work for at least 1 year) and have received disability insurance (Social Security disability insurance) cash payments for 2 years.18 As described elsewhere,19,20 the MCBS is an ongoing longitudinal survey of a representative panel of Medicare beneficiaries, with an oversampling of persons younger than 65 years (n = 2378 [14.5%]) and of persons 85 years and older (n = 2072 [12.6%]). We eliminated 106 persons (0.6%) receiving Medicare under the end-stage renal disease entitlement, including the only respondents younger than 18 years (n = 2).
Persons typically remain impaneled in the MCBS for 4 years, with the sample replenished annually (eg, to replace respondents who died). The MCBS interviews panel members or their proxies in person 3 times yearly, tracking participants wherever they reside and using 2 types of surveys: (1) computer-assisted community questionnaires for persons living in the community and (2) facility questionnaires for respondents in long-term care or institutional settings. With the facility questionnaire, interviewers query administrators or designated staff, not the Medicare beneficiary, and, therefore, do not address respondents' perceptions of care. We used results only from the MCBS community survey, which included questions about demographic characteristics, health status and functioning, perceptions of care, and usual source of care. We considered self- and proxy-reported responses. Among persons younger than 65 years, proxies supplied 22.4% of the responses; 9.0% of persons 65 years and older had proxies.
All findings used MCBS sampling weights to produce nationally representative Medicare population estimates. Our analyses used SAS-callable SUDAAN statistical software, version 7.5 (Research Triangle Institute, Research Triangle Park, NC).
The MCBS offers several ways to identify persons with disabilities. The most obvious is age. At some point, Medicare beneficiaries younger than 65 years (except those with end-stage renal disease) have met the Social Security Administration's employment-related definition of disabled and have received Social Security disability insurance.18 This administrative definition, however, may not reflect functional abilities over time (ie, in 1996, when interviewed by the MCBS).21- 23 The MCBS asks whether specific conditions (eg, heart or lung problems, cancer, mental retardation, or psychiatric disorders) originally caused Medicare eligibility.24 Again, the functional implications of many of these conditions in 1996 are unclear. To identify disabilities, other investigators20,24- 26 have used perceived health status (excellent, fair, or poor) or difficulties performing activities of daily living or instrumental activities of daily living. These general measures do not indicate specific sensory or physical limitations.
Annually, the MCBS asks about specific sensory and physical abilities.19 We used these responses to identify 5 categories of potential disabilities pertaining to vision, hearing, walking, reaching overhead, and grasping and writing (algorithm available from the authors). For each category, we created 2 levels based on answers about the extent of difficulties, assigning people to the most severe level for which they qualified.
We also created 2 summary measures, each with 3 groups (no difficulty, any minor or moderate difficulty, and any major difficulty), and 0, 1, or 2 or more disabling conditions. These summary measures produced similar results, and we report findings using the first approach.
The September to December round of the MCBS includes supplemental surveys on experiences with and perceptions of respondents' usual source of care. The questionnaire asks about 6 aspects of physicians' technical quality (carefulness, competence, understanding of medical history, understanding of clinical conditions, completeness, and patients' confidence in their physician) and 6 dimensions of interpersonal quality (physician seems hurried, physician does not explain, physician acts as if he or she is doing the patient a favor, physician informs the patient completely, physician answers all questions, and whether patients depend on physicians to feel better). Interviewees were asked to respond to 12 statements reflecting these dimensions, with the following response options: strongly agree, agree, disagree, and strongly disagree.
We grouped agree and strongly agree responses (and likewise disagree and strongly disagree responses). The directionality of the 12 statements varied. Agreement sometimes suggested poor experiences and sometimes good care. We modeled responses with negative connotations. One statement ("depends on doctor in order to feel better both physically and emotionally") offers mixed messages: neither agreement nor disagreement necessarily indicates poor experiences, simply different views of physicians' roles. For this statement, we model disagreement.
People enrolled in managed care might have different experiences and satisfaction levels than beneficiaries enrolled in traditional Medicare.27,28 In 1996, the MCBS added a special supplement of persons enrolled in "risk" health maintenance organizations (HMOs), 22.9% of our sample. In addition, Medicare administrative records indicated that 22.7% of MCBS respondents had been enrolled in an HMO for at least part of the previous calendar year. About 2.6% of persons had discordant information about HMO participation. We designated persons as having managed care membership if they were in the HMO supplement or if Medicare administrative records indicated HMO membership during the prior year.
We used the direct standardization method of SUDAAN statistical software to adjust for age, using 7 groups (18-44, 45-64, 65-69, 70-74, 75-79, 80-84, and ≥85 years). For each disability category and each quality dimension, we produced 2 multivariable logistic regression models predicting negative perceptions of care based on the following: (1) age group, sex, and presence and extent of the disabling condition; and (2) age group, sex, disabling condition, race, ethnicity, residence location (urban vs rural), educational level, annual household income (<$25 000 and ≥$25 000), and managed care participation. The second model, thus, adjusts for other patient and organizational attributes that could affect perceptions of care to isolate the contributions of disability. We report adjusted odds ratios (AORs) with 95% confidence intervals.
We also produced separate multivariable models for persons 65 years and older and those younger than 65 years (ie, persons receiving Social Security disability insurance). We failed to find consistent important numerical differences by these broad age groups, although in some instances a nonsignificant trend suggested that younger persons were less satisfied than older persons. Herein, we report the results from the models combining elderly and younger respondents. Only 117 MCBS respondents were blind; results from this group are unstable, rarely reaching statistical significance.
We used the Cronbach α coefficient to assess whether the directionality of responses agreed across sets of questions.
Of an estimated 33.58 million noninstitutionalized Medicare beneficiaries, 64.1% (an estimated 21.51 million) reported at least 1 of 5 potentially disabling conditions (Table 1) and 29.5% (an estimated 9.89 million) reported more than 1 potentially disabling condition. Among people younger than 65 years (10.9% of noninstitutionalized Medicare beneficiaries, or 3.67 million persons), 73.1% (an estimated 2.68 million) noted at least 1 of the 5 disabling conditions and 45.4% had more than 1 condition.
Across the 5 disability categories, from 93.3% to 96.7% of persons 65 years and older reported having a usual source of care. At that usual care source, 88.8% to 94.0% saw a specific physician (Table 1). Among persons younger than 65 years, 4.5% to 8.8% did not have a usual source of care, with roughly 34.6% of blind persons lacking a usual care source. Younger Medicare beneficiaries were less likely to have specific physicians at their usual care site (Table 1).
Among persons 65 years and older, those with more severe impairments were older, on average, than those with less severe limitations. After adjusting for age, higher percentages of women than men reported all disabling conditions, except hearing difficulties. After adjusting for age and sex, higher percentages of black than white persons reported all disabling conditions, except hearing difficulties. Adjusted percentages for Hispanic persons and people of other races varied by disabling condition. Persons with disabling conditions were more likely than others to have only high school educations and annual incomes of less than $25 000.
Table 2 and Table 3 show percentages of persons (adjusted for age and sex) reporting negative views of their care for those without the specific disability or with only minor difficulties, those with moderate difficulties, and those with major limitations. Among the technical quality dimensions across all disabilities (Table 2), respondents most commonly reported that they "often have health problems that should be discussed but are not," with 12.1% to 15.0% of persons with major disabilities expressing this view, significantly higher than for persons without disabilities (P<.001), except for the vision category. The next most common concerns were generally that physicians are not "very careful to check everything when examining [them]" and do not have "a complete understanding of the things that are wrong with [them]," cited by 7.0% to 11.0% of persons with major limitations. Few people reported problems with other technical dimensions of care, with more than 90% saying they have "great confidence in [their] doctor."
Interpersonal quality seemed more problematic than technical quality (Table 3). Respondents most commonly reported that the "doctor often seems to be in a hurry," with 15.8% to 24.1% of persons with major disabilities noting this (vs <11.6% of those without disabilities). The next most frequent concern was that the "doctor often does not explain medical problems," expressed by 12.9% to 16.7% of those with major limitations. Regardless of disability, more than 85% of persons do not rely on their physician "in order to feel better both physically and emotionally."
We tested whether responses to the 6 interpersonal quality questions pointed in the same direction (negative vs positive views). The Cronbach α for these questions exceeded .85 for all disability categories; persons dissatisfied with one interpersonal dimension tended strongly to be dissatisfied with other dimensions.
Table 4 and Table 5 show the AORs of reporting concerns about technical and interpersonal dimensions of care, after controlling for demographic characteristics and HMO enrollment. We did not find statistically significant and consistent associations between satisfaction and basic demographic characteristics (age <65 or ≥65 years, sex, race, ethnicity, educational level, income, and urban vs rural residence) across the dimensions of care and disability groups.
Adjusted odds ratios were generally statistically significant for all disability categories for all 12 dimensions of care with 2 exceptions: perceptions of competence and whether people rely on their physicians to feel better. Although the AORs tended to be somewhat higher for interpersonal than technical dimensions, persons with disabilities seemed much more concerned than others about both broad aspects care. Adjusted odds ratios were generally more significant for the 3 disabilities relating to impaired mobility compared with the vision and hearing categories, but all 5 disability groups reported problems.
We considered whether proxy respondent status affected perceptions of care by adding proxy status to our multivariable regression models. Proxy respondent status rarely reached statistical significance, although we found a trend toward proxies reporting more negative views of care. Even these trends, however, varied across dimensions of care and disability groups.
Persons enrolled in HMOs reported significantly more problems for all dimensions of care except for physician competence and whether physicians seem hurried. For all 12 quality dimensions, the AORs associated with HMO membership were similar across the 5 disability categories.
Most persons with disabilities were satisfied with their care and equally likely as those without disabilities to perceive their physicians as competent and well trained. The most common concern was that physicians seem hurried, articulated by roughly one fifth of respondents, regardless of disability and HMO membership. Nevertheless, even after accounting for various factors, people with disabilities were significantly more likely than those without disabilities to report other problems, including incomplete understanding of clinical histories and conditions, lack of thoroughness, and inadequate communication. These findings held across disabling conditions. People with disabilities were much less likely than others to have confidence in their physicians.
Given the breadth of quality concerns, devising strategies to improve the experiences of patients with disabilities requires careful thought. Two threads possibly linking the various concerns are communication and time. Although people with disabilities believe their physicians are competent, they are simultaneously less satisfied with their physicians' thoroughness and communications. These latter issues may not reflect reservations about physicians' technical competence, but instead worries that physicians do not take the time required to understand fully patients' clinical concerns and communicate effectively.
Although many patients, regardless of disability, viewed physicians as hurried, persons with disabilities may face special risks from time constraints for 3 reasons: complex underlying medical conditions; extra knowledge, skill, sensitivity, or time required because of the disabling condition itself; and discordant perceptions and expectations between physicians and patients, especially about the experience of disability. Examples include informing blind persons about actions during the physical examination or discussing treatment plans with a deaf person through an American Sign Language interpreter. Positioning people with extensive mobility limitations on examination tables generally takes more time than required for other patients. Physicians may focus on the disabling conditions when patients prefer to address wellness care or other clinical concerns.29,30
The MCBS has important limitations for our purpose. We cannot assess whether specific accommodations may resolve dissatisfaction articulated by respondents. All functional deficits are self-reported, raising questions about their validity.31,32 The use of proxy respondents further complicates the interpretation of findings, although research33- 38 offers contradictory evidence about the direction of potential bias. The MCBS does not indicate the extent of respondents' acute and chronic underlying disease nor how long they spent with physicians during office visits. Therefore, we cannot directly test the hypothesis that disabled patients are especially susceptible to time constraints. Finally, the MCBS does not identify instances during which effective communication is impeded by stigmatizing attitudes about disability among physicians.39- 44
Other studies45- 47 suggest that satisfaction with care relates strongly to time spent: patients with longer visits report greater satisfaction. Persons with substantial health problems generally are less satisfied with medical care than healthier individuals, possibly because they believe their needs are not fully met.48,49 Whether visit lengths have diminished in recent years and the relationship between time spent and managed care insurance remain controversial. Nevertheless, most agree that the average office visit lasts less than 20 minutes.50- 52
Anecdotal reports53 raise questions about how much more time might be needed to accommodate persons with sensory or physical disabilities. An internist, for example, described a routine office visit with a middle-aged woman with multiple sclerosis.4 The internist and her assistants tried unsuccessfully to lift the woman onto the high unadjustable examining table; the patient's daughter, familiar with transferring her mother, performed the maneuver. Valuable time, thus, was expended simply positioning the patient before more meaningful interactions could occur. The various interchanges that build patient-physician relationships, confidence, and trust—and may be valued more highly than technical competence54—must await these most basic accommodations. Inaccessible facilities also implicitly suggest that the needs of persons with disabling conditions have not been adequately addressed within a practice.
Our results support the recommendations of Wagner and colleagues,55,56 who examined the research evidence from the United States and Europe about improving health care outcomes for persons with chronic conditions. Providing complete information was 1 of 4 essential elements when designing systems of care, as were practice redesign, patient education, and expert systems (eg, physician education and consultations). Other investigators57- 59 have found that communication, respect, and involvement in decision making are especially valued by patients. Patients who report that their physicians "do not always take enough time to answer questions" or "do not provide sufficient information" are likely to consider changing physicians.60(p37) Perceptions of stigmatizing attitudes about disabling conditions39- 44 certainly could compound these problems.
Thoughtful systematic approaches are required to improve accommodations for communicating with blind and deaf persons and to reduce the time demands and physical efforts of people with impaired mobility (eg, moving around offices, undressing and dressing, and positioning on examination tables). These concerns largely reflect the structure of the health care system, the often-neglected branch of the structure-process-outcome quality triad articulated by the late Avedis Donabedian.61,62 Although analysts frequently reduce structure to "bricks and mortar," Donabedian's original formulation reached more broadly, encompassing physician characteristics, their "tools and resources," and "the physical and organizational settings in which they work. The concept of structure includes the human, physical, and financial resources that are needed to provide medical care."61(p81)
For people with disabilities, many and varied structural accommodations are important to ensure delivery of the highest-quality care. Some do involve bricks and mortar (eg, ramps, widened doorways, and automatically adjustable examination tables), while others represent essential tools (eg, large-print and braille written materials and readily available sign language interpreters), and yet others reflect policies (eg, scheduling longer appointments). Strategies may reach beyond individual practices to the broader health care system, such as paying more for routine visits of persons needing interpreters or special physical accommodations.
Additional investigations must explore why people with disabilities are generally less satisfied with their care than others. Solutions that improve care for persons with disabilities are likely to improve experiences for everybody.63
Corresponding author: Lisa I. Iezzoni, MD, MSc, Division of General Medicine and Primary Care, Department of Medicine, Beth Israel Deaconess Medical Center, East Campus RO-137, 330 Brookline Ave, Boston, MA 02215.
Accepted for publication November 21, 2002.
This study was supported by grant RO1 HS10223-03 from the Agency for Healthcare Research and Quality, Rockville, Md.