How adults with developmental disabilities perceive quality of the care and service they receive was investigated and these perceptions compared with those of their parents and primary caregivers. The sample was 34 adults with developmental disabilities, one of their parents, and their primary caregiver, all from a small Swedish municipality. We adapted Quality From the Patient's Perspective by using simple words and fewer response choices. Results showed it was possible to assess how adults with developmental disabilities evaluated actual care and service conditions. Evaluations of the subjective importance ascribed to these conditions proved too difficult. A high degree of similarity was found in ratings of perceived reality of care and service conditions by the adults with developmental disabilities, their parents, and caregivers.
In somatic health care, patients' evaluations have increasingly been used as an indicator of quality (Calnan, 1988). However, for most people with developmental disabilities, physicians and other professionals remain dominant (Sands, Kozleski, & Goodwin, 1991). One reason for this is the cognitive limitations of individuals with developmental disabilities; another is the lack of suitable measurement instruments. In a recent review of the literature, Walsh and Kastner (1999) concluded that “we are aware of no reports of the development of health care satisfaction measures specifically for people with developmental disabilities” (p. 6). Research on quality of life is a related area. In this domain, investigators have frequently included assessment of satisfaction with health care and generic services. However, also within this research tradition, the need for more specific assessment tools for people with developmental disabilities has been pointed out (Goode, 1994; Schalock, 1994).
Excluding people with developmental disabilities from quality of care assessments is ethically problematic. Including them via family members or caregivers may also introduce measurement errors. Thus, there is a need to develop measures of quality of care from a caretaker perspective that can take into account the functional communication levels of people with developmental disabilities (Walsh & Kastner, 1999).
In somatic health care, criticism has been leveled at various scales designed to measure patients' satisfaction with, or perception of, quality of care (Calnan, 1988; Gabbard-Alley, 1995; LaMonica, Oberst, Madea, & Wolf, 1986; van Campen, Sixman, Friele, Kerssens, & Peters, 1995; Sitzia & Wood, 1997). Although several of these scales may be creatively designed, a major criticism leveled at the majority of them is the lack of a theoretical foundation. The selection of indicators has generally not been related to empirically based models of patients' conceptions of the area (Bond & Thomas, 1992; Hall & Dornan, 1988; Meterko, Nelson, & Rubin, 1990). Consequently, one cannot be sure that the attributes chosen in the scales are the most important with regard to quality of care. This validity-related weakness also makes the interpretation of results more difficult. In the present study, we based the assessment of quality of care on a patient-centered model. High-quality caring requires similar perceptions of actual care conditions by the enactor and the recipient (Bergman, 1968). As in all human interaction, there is a potential for perceptual incongruity in caregiver–caretaker relationships. In most studies on this matter, investigators tend to report some inconsistencies (Larson, 1987; Larsson & Wilde Larsson, 1998; McCullough, Willson, Teasdale, Kolpakchi, & Skelly, 1993; von Essen & Sjödén, 1991; Young, Minnick, & Marcantonio, 1996). The correspondence between the perceptions of caretakers and caregivers appears to vary across different care settings (Larson, 1987). The settings focused on in the present study concern situations in which care was provided for people with developmental disabilities. Our aim was to systematically investigate how these individuals perceive the quality of the care and service they receive and to compare these perceptions with those of their parents and primary caregivers.
Model of Quality of Care
First, using a grounded theory approach with a qualitative method of analysis, Wilde, Starrin, Larsson, and Larsson (1993) developed a theoretical model of quality of care from a patient perspective in an attempt to reflect a deeper understanding of the phenomenon. The model was generated from 35 in-depth interviews with voluntarily recruited and verbally communicative patients (mean age 60 years) with infectious diseases. Interviews were interpreted in accordance with the advice given in the literature regarding grounded theory on successively more abstract forms of coding (Glaser, 1978; Starrin, Dahlgren, Larsson, & Styrborn, 1997). As a first step, indicators of quality of care were identified from patients' interview responses (about 900 altogether). As a second step, these indicators were grouped into 27 categories. Third, these categories were coded as comprising four dimensions. As a final step, these four dimensions were classified into a model. All steps were based on the researchers' understanding and interpretation of the data (see Wilde et al., 1993, for further details).
The model stipulates that patients' perceptions of what constitutes quality of care are formed by their encounter with an existing care structure and by their system of norms, expectations, and experiences. Quality of care can be understood in light of two conditions: the resource structure of the care organization and the patients' preferences. The resource structure of the care organization consists of person-related as well as physical and administrative environmental qualities. The patients' preferences are assumed to have a rational and human orientation. Within this framework, patients' perceptions of quality of care may be considered from four dimensions: the medical–technical competence of the caregivers, the physical–technical conditions of the care organization, the degree of identity-orientation in the attitudes and actions of the caregivers, and the sociocultural atmosphere of the care organization. A diagram of the model is given in Figure 1.
Second, Wilde, Larsson, Larsson, and Starrin (1994) developed the questionnaire called Quality From the Patient's Perspective from the theoretical model, using a conventional factor analytical approach. Third, the dimensionality of the Quality From the Patient's Perspective was further developed using structural equation modeling (Larsson, Wilde Larsson, & Munck, 1998). The basic version of this questionnaire, which is intended particularly for patients in ordinary acute care wards, consists of 68 items. Approximately 25 additional versions have been adapted from the basic version to suit a variety of specific contexts, for instance, care and service for older people living in their own home (Wilde Larsson, Larsson, & Starrin, 1998).
The psychometric properties of the Quality From the Patient's Perspective have been extensively studied and found favorable. In the study in which researchers used structural equation modeling (Larsson et al., 1998), reliability was estimated by a method developed by B. Muthén and described by Huang (1997), resulting in a measure called determinacy. This is a measure of the correlations between the true and estimated factor scores. Cronbach alpha coefficients have also been computed several times to assess the internal consistency of the scales (see, e.g., Wilde Larsson & Larsson, 1999).
Three groups of individuals participated: people with developmental disabilities, one of their parents, and their primary caregivers. The operational definition of developmental disability was that the person should have been judged by the social welfare authority of the municipality to fulfill the criteria of the LSS (the Swedish law that regulates the rights of persons with disabilities). This law is applicable to persons
whose development has been retarded . . . or who have other persisting physical or mental functional deficiencies which are obviously not attributable to normal aging, and if they are substantial and cause considerable difficulties in daily life and therefore also create an extensive need for support or service. (our translation)
All people with developmental disabilities in a small-sized Swedish municipality who met the following inclusion criteria were invited to participate: (a) the person should be 19 years or older and receive help from the municipality; (b) the person should be able to communicate in an interview situation, which was judged by a responsible supervisor of the caregivers; and (c) the person was willing to participate. There were 87 persons with developmental disabilities in the municipality who were 19 years or older. Of these, 46 (53%) were judged as able to communicate, and of these 34 (74%) agreed to participate. For each of these persons, one of their parents and the primary caregiver were selected.
The mean age of the adults with developmental disabilities was 43.9 years (standard deviation [SD] = 11.4, range = 22 to 62); 49% were women and 97% lived alone. The parents' mean age was 64.7 years (SD = 12.2, range = 44 to 80), and 86% of the parents were women. The mean age of the primary caregivers was 42.2 years (SD = 11.3, range = 23 to 61), and 90% were women.
Fifteen percent of the individuals with developmental disabilities received help from the municipality less than once a day; 15%, once a day; 50%, two or three times per day; and 20%, four times or more per day. The most common types of help were washing (84%), cleaning (83%), shopping (83%), and cooking (81%). Regarding activities, 22 (73%) attended a day care center, 2 (7%) were at school, 5 (17%) were in special employment, 1 reported “other” activities, and 4 did not respond to this question.
The empirical phase of the study took place in March and April 1999. Data for individuals with developmental disabilities were collected through personal interviews that were structured from the questionnaire Quality From the Patients' Perspective. The questions and response categories were read aloud by a female interviewer. She was a former caregiver of children and adults with developmental disabilities and had 20 years of practical experience. To make it easier for some of the participants to respond, we enlarged the response scales considerably on a separate sheet of paper. The interviewer attempted to strictly follow the questionnaire. However, if the participant did not understand a question, the interviewer tried to illustrate its meaning by briefly elaborating on it and/or by giving an example.
Most interviews lasted for about 60 minutes, including pauses if the respondent felt tired, with a range from 45 to 90 minutes, and took place in the homes of the adults with developmental disabilities. In order not to be identified with the staff, the interviewer wore regular clothes, and no parents or caregivers were present during the interviews. Each participant was given oral information about the study. Informed consent was obtained from participants by the interviewer. The parents and the primary caregivers were given oral and written information about the study.
Data for the parents and the primary caregivers were collected via administration of a questionnaire. The study was conducted according to the ethical guidelines of the Swedish Nurses' Association and the Swedish Psychologists' Association.
Quality of care and service
Data were collected using a modified version of the Quality of Care From the Patient's Perspective described earlier. The modification included a simplification of the wording of most items, a simplification of the wording of the response alternatives, and a reduction of the number of choices. All items designed to measure the quality dimensions physical–technical conditions and sociocultural atmosphere were dropped because they refer to institutional characteristics, and no participant in the present study lived in an institution. The modified questionnaire contained 39 items designed to measure the two remaining quality dimensions: medical–technical competence and identity-oriented approach. Eighteen of the modified items came from the basic version of the Quality From the Patient's Perspective (Larsson et al., 1998) and 16 from the version adapted for care and services for older people living in their own homes (Wilde Larsson et al., 1998). Five new items—dealing with work, free time, and vacations—were constructed.
Each dimension, exemplified with one item, is as follows:
Medical-technical competence (physical care) (Cronbach alpha = .72) was assessed by 5 items. Example: “I get the help I need/require when I visit the bathroom.”
Identity-oriented approach (Cronbach alpha = .88) was assessed by 34 items. Example: “The staff respect me.”
Each item was evaluated in two ways by the respondent. One was designed to measure perceived reality of quality of care and service. The wordings of all items in the questionnaire for the individuals with developmental disabilities are shown in Tables 1 and 2. The parents and the primary caregivers related each item to the sentence” ——— (name of individual) gets/can etc…” The following response scale was used: yes (3), sometimes (2), no (1), don't know, and not applicable. The second kind of evaluation concerned the subjective importance the person ascribed to the various aspects of care and service. Here, each item was related to the sentence “This is how important it is for me that. . . ” (e.g., I can decide myself when I am going to go to bed). The parents and the primary caregivers related each item to the sentence “This is how important it is for ——— to. . . ” A 3-point response scale was used: very important (3), important (2), and not important (1).
We used Kruskal-Wallis one-way analysis-of-variance by ranks to test differences in means between the persons with developmental disabilities, their parents, and their primary caregiver. All tests were two-tailed and statistical significance was assumed at the .05 level of significance.
Responses of Individuals With Developmental Disabilities: Summary of Don't Know Answers, Not Applicable Answers, and Missing Values
Perceived reality questions
On the 5 questions designed to measure physical caring, one don't know response was obtained. On 2 questions, 50% or more checked the not applicable alternative. The number of missing values was 2 on all these questions. On the 34 questions designed to measure identity-oriented approach, 4 had 50% or more don't know answers, no questions had more than 5 not applicable answers, and no questions had more than 6 missing values.
Subjective importance questions
All 5 questions designed to measure physical caring, and 31 of 34 questions designed to measure identity-oriented approach, had at least 50% missing values. This indicates that most participants were unable to respond to these questions, and they were not analyzed any further.
Comparison of Individuals With Developmental Disabilities, Their Parents, and Primary Caregivers
Comparisons of the three groups of participants on the quality of care scales are summarized in Tables 1 and 2. Statistically significant overall mean differences were found on 5 out of 39 possible items. All 5 were designed to measure the identity-oriented approach of the caregivers. On the following 4 questions, the adults with developmental disabilities gave the most favorable ratings, and the caregivers the least favorable: I can decide myself when I am going to get up, I can decide myself when I am going to go to work, I can decide myself what I am going to do in my free time, and I can decide myself what I am going to do during my vacation. On the following question the parents gave the highest rating, followed by the caregivers, and, finally, the individuals with developmental disabilities: I get good information on which members of staff are coming to work and when they are going to come.
Table 1, which is focused on the scores of the individuals with developmental disabilities, also shows that they perceive that they do get the help they need regarding all mapped aspects of physical care. Table 2 further shows that these participants feel they can decide themselves when to get up, go to bed, or take a shower. The vast majority also feel that the staff care about them, knock or ring on their door before entering, give sincere answers to questions, and act towards them in a good manner. The adults with developmental disabilities experience their decision latitude as restricted on some aspects of work and meals as well as in deciding what staff members they prefer, and in getting help from the same staff members.
The participants with developmental disabilities had a favorable perception of all mapped aspects of help with physical caring. Most aspects of the identity-oriented approach in the attitudes and actions of the caregivers were also favorably evaluated. The lowest ratings were found on the questions dealing with the possibility to decide which staff members they wanted to have.
Most of the mapped aspects of care and service were regarded as applicable. Evaluations of the subjective importance ascribed to each care aspect were, however, more problematic. The majority of the participants were not able to respond to these questions in a meaningful manner.
The following conditions may have contributed to what, in our opinion, was the successful attempt to systematically investigate the opinions of adults with developmental disabilities. First, a selection factor is applicable: Nearly half of all potential study participants in the chosen municipality were excluded from the study. They would not have been able to respond in a meaningful way according to judgments of experienced caregivers.
Second, an interviewer factor may have contributed. We believe that the experience of the interviewer, who had worked with individuals who had developmental disabilities for many years, made it easier for the remaining participants to respond.
Third, an instrument factor may be relevant. The wording of the Quality From the Patient's Perspective items was simplified and the number of response choices reduced. The response scale was also enlarged on a separate response sheet. The model underpinning this questionnaire is based on a patient perspective of quality of care. Although this questionnaire could be improved in terms of making the questions more contextually relevant, the care and service aspects and wordings chosen appear to have been of great relevance for the participants.
In the original Quality From the Patient's Perspective, ratings of perceived reality can be related to evaluations of subjective importance (Wilde et al., 1994). The idea is that this opportunity gives an improved assessment of quality of care. A low evaluation of a given aspect of care, for instance, is assumed to have a different meaning if it is accompanied by a high evaluation of subjective importance than if the latter rating is low. A similar line of reasoning is found in quality of life research (Cummings & McCabe, 1994; Ferrans & Powers, 1985; Schalock, 1994). Results of the present study suggest that this potential advantage of the Quality From the Patient's Perspective would have to be dropped, however, because most participants with developmental disabilities were unable to complete the subjective importance ratings. On the other hand, successful attempts to employ subjective importance ratings, using Cummins Comprehensive Quality of Life Scale—Intellectual Disability, have been reported (Cummins, 1992). Intuitively, we feel that the cognitive demands posed by this scale are comparable to the ones posed by the Quality From the Patient's Perspective. Thus, the only explanation we can think of is that the participants in the present study were lower functioning than those in the Cummins study. Within our study group, at least, it was obvious that perceived reality ratings were easier to obtain than were ratings of subjective importance. Further studies are needed to determine how different levels of cognitive functioning are related to different kinds of questions on quality of care and service.
The reliability (internal consistency) of the two mapped quality of care dimensions was acceptable. However, we note that the relationship between the items designed to measure the identity-oriented approach dimension in the original Quality From the Patient's Perspective and the corresponding items in the present modified version is not known. Due to the exclusion of some of the original hospital-specific items, and the inclusion of some new ones with work-related con-tent, it seems likely that a different factor structure is to be expected. The low number of participants prevented us from analyzing the dimensionality any further. Regarding reliability, we also note that there is a risk of preference for agreement in interview situations, such as the ones in the present study. In this case, although all participants were informed that only the independent researchers would have access to their answers, an acquiescence scale would have been desirable. The Lifestyle Satisfactory Scale (Heal & Chadsey-Rusch, 1985), for instance, contains such a measure.
Summing up the measurement discussion, we feel that the present study could be regarded as a step in the development of tools applicable to adults with developmental disabilities. The lack of such tools has repeatedly been pointed out (Walsh & Kastner, 1999). Items could be adapted better to the situation of individuals with developmental disabilities than was the case in this study. We suggest that this could be done if researchers first conducted qualitative studies and then operationalized the resulting qualities into standardized questions worded simply.
The comparison of the ratings of the adults with developmental disabilities, their parents, and their caregivers indicated a high degree of perceptual consensus. No statistically significant differences were found (perceived reality ratings) between the three groups on any of the items designed to measure physical caring. Within the large item pool designed to assess the quality dimension identity-oriented approach, 5 significant differences out of a possible 34 were noted. On 4 of these items, the individuals with developmental disabilities gave the most favorable ratings of the quality of care; and their caregivers, the least favorable. Due to the high internal drop-out rate on the subjective importance questions, we do not find it meaningful to comment on them any further.
The similarity in the ratings of perceived reality differs from what has frequently been found in studies of care of older individuals (Larsson & Wilde Larsson, 1998) and in studies in acute somatic care (von Essen & Sjödén, 1991). We offer two possible suggestions that could account for this difference. The first is psychometric: When the number of response choices is reduced, the probability of increased similarity between different groups of raters is increased. The second suggestion, and the one we attribute most importance to, is related to radically different care situations. Even as adults, individuals with developmental disabilities spend much time with their parents and their caregivers. The caregivers have typically known and worked with the individuals for several years. Thus, compared to most other kinds of care, they objectively share more experiences. This could be reflected in the similarity of their subjective reports. Another consequence of the high degree of similarity between rater groups is that among adults with developmental disabilities, ratings from parents and caregivers may be more acceptable from a validity point of view. However, replication studies are needed because the present results may apply only to a Scandinavian situation. In a previous quality of life study in the United States, low agreement was reported between individuals with developmental disabilities and their parents or attendants (Sigelman et al., 1981).
Finally, we have not commented on the practical implications of these results because this was not our aim in this report. In the municipality concerned, the results were presented to the caregivers. They, in turn, presented them to the individuals with developmental disabilities and their parents. Based on the findings, a priority of the quality improvement changes was established and implemented.
Authors:Gerry Larsson, PhD, Professor, Department of Leadership, National Defence College, Järnvägsgatan 6, SE-652 25 Karlstad, Sweden ( email@example.com). Bodil Wilde Larsson, PhD, Associate Professor, Division for Health and Care, Karlstad University, SE-651 88 Karlstad, Sweden.