| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to Professor Susanne Iwarsson, Division of Gerontology and Caring Science, Lund University, Box 157, SE-221 00 Lund, Sweden. E-mail: Susanne.iwarsson{at}med.lu.se
 |
Abstract |
|---|
 TOP Abstract Design and MethodsResultsDiscussionReferences |
|---|
Key Words: Accessibility • Docility hypothesis • Functional limitations • Home assessment • Occupational therapy
Environmental gerontology represents pluralism in regard to theoretical approaches (Wahl & Wiesman, 2003), and placing the study of home environments within a theoretical framework is a major challenge (Gitlin, 2003). However, the frequently cited idea of person–environment (P–E) fit (Lawton, 1980) is applicable to research concerning older adults' housing situation (Steinfeld & Danford, 1999). When it comes to theoretical foundations, Lewin's early definition (1951) of behavior as a function of the person and the environment is often referred to as the basis for P–E fit research. One important aspect of behavior of high relevance for P–E fit research is the performance of activities of daily living (ADLs; Iwarsson, 2004).
The model most often referred to in many different disciplines is Lawton's ecological model (Lawton & Nahemow, 1973; Lawton & Simon, 1968), which represents a landmark in environmental gerontology (Golant, 2003; Scheidt & Norris-Baker, 2003; Wahl & Wiesman, 2003) and which "explores the interplay between individuals and their environments" (Nahemow, 2000, p. 23). The person is defined in terms of a set of competencies, and the environment is defined in terms of demands. P–E fit comprises two interactive components, the personal component and the environmental component. As defined in the original model, the personal component, denoting "competence," includes biological health, sensory and motor skills, and cognitive function. Adopting Scheidt and Norris-Baker's recent interpretation of competence in the context of the ecological model, we can say that competence is equivalent to functional capacity. The environmental component, denoting "press," is described in terms of five broad classes: the physical environment, the personal environment, the small-group environment, the suprapersonal environment, and the social or megasocial environment (Lawton & Nahemow). The surface of the model labeled "adaptation" is intended to reflect the matching between competence (personal component) and press (environmental component). Interpreting adaptation as manifested in real-life situations, we can see that one important aspect of adaptation relevant for gerontological research is ADL performance (Iwarsson, 2004). Further, this kind of interpretation is in line with current terminology for the definition of functioning and disability widely used in health care contexts (World Health Organization [WHO], 2001), supporting the idea that positive outcomes in environmental gerontological research can reasonably be expressed as independence in ADLs (Iwarsson, Isacsson, & Lanke, 1998).
The original ecological model is decidedly interactionist in that the personal and environmental components are assumed to be conceptually and empirically distinguishable, and in this perspective adaptation is viewed as the outcome domain. According to Diehl and Willis (2003), competent behavior resides neither within the person nor within the environment, but in the P–E interaction. That is, one outcome of P–E interaction is ADL performance. The original ecological model was very comprehensive, but over the years Lawton (1989) elaborated it to encompass several additional dimensions (e.g., distinctions between objective and subjective aspects of the components, proactivity). Even though the model and its different versions are widely cited and underlie practical efforts in planning environments that support healthy aging, empirical testing of the model is relatively sparse (for an overview, see Nahemow, 2000). Given the model's complexity, research to test it in its entirety is beyond the reach of current methodology. As advised by Scheidt and Norris-Baker (2003), research on P–E fit issues could well be based on variables representing the original ecological model's major components. The present study is principally focused on P–E fit in terms of functional capacity in the person (personal component), the physical housing environment (environmental component), and independence or dependence in ADLs (adaptation level).
When Lawton's ecological model is applied to the physical dimension of the home environment, physical barriers in the environment are not necessarily problems per se. Instead, they cause different magnitudes of problems for different people, depending on each person's functional capacity (Iwarsson, 1997; Iwarsson & Ståhl, 2003). According to the docility hypothesis (Lawton & Nahemow, 1973; Lawton & Simon, 1968), individuals with lower competence are more sensitive to the demands of the environment than those with higher competence. Individuals with lower competence spend so much of their energy overcoming and adapting to environmental demand that they have a reduced ability to change their situation on their own. Negative environmental changes have a greater impact on them, but also minor improvements result in a more positive outcome in regard to activity than in those with higher functional capacity. When the personal and environmental components are matched, adaptation is achieved; this is manifested in real life, for example, in terms of independent ADL performance (Iwarsson et al., 1998).
In many studies the physical home environment has mainly been targeted in terms of risks of falling, focusing on fall prevention in different patient groups (e.g., El-Faizy & Reinsch, 1994; Gill, Robison, Williams, & Tinetti, 1999; Steinfeld & Shea, 1993). Moreover, P–E interactions have been studied in the context of care units and sheltered housing (e.g., Wahl, 2001), whereas far less research has targeted ordinary housing in the community (Diehl & Willis, 2003). Among the few studies focusing on P–E fit in community housing, some evidence from investigations in Germany that analyzed relationships between objective housing conditions and patterns of capacity confirmed that reduced ADL ability was due to unfavorable conditions in the physical home environment (Schmitt, Kruse, & Olbrich, 1994). Wahl, Oswald, and Zimprich (1999) demonstrated that, in contrast to that for older adults without visual impairments, unfavorable P–E fit negatively affected individuals with impairments. Applying novel methodology capturing P–E fit in great detail, in prior studies, Iwarsson and colleagues explored and identified relationships between functional capacity, environmental factors, and ADL dependence among older adults (Iwarsson & Isacsson, 1997b; Iwarsson et al., 1998). These studies also showed that disability rates are affected by environmental factors (Iwarsson & Isacsson, 1997a). That is, prior results demonstrated that dependence on assistance from others in more complex activities (e.g., shopping and transportation) is affected by whether environmental factors such as the design of the home entrance and the availability of public transportation meet the needs of individuals with functional limitations. The results also demonstrated that the relationship between P–E fit problems and ADL dependence was significant only among the frailest half of the sample investigated, lending some empirical support for the docility hypothesis (Iwarsson et al.).
One of the greatest challenges for use of Lawton's general ecological model is incorporating temporal dimensions. Hence, in the current study I incorporate time and focus on activity as recently suggested by Golant (2003), investigating how the home environment affects healthy aging over time.
A New Approach to Conceptualizing and Measuring P–E Fit
In general, research targeting home environments lacks psychometrically sound measures (Gitlin, 2003). Methods for assessing P–E fit problems should provide evidence of the degree to which a particular physical environment prevents or supports daily activity (Steinfeld & Danford, 1999). In most prior studies, however, measures used are study specific, with unknown reliability and validity.
The Enabler Concept is an American idea, originally published to serve as a basis for the design of accessible housing and public environments (Steinfeld et al., 1979). It was developed in connection with research into issues of accessibility that gradually led to the housing standards of the American National Standards Institute (ANSI, 1980). Because the purpose of the Enabler Concept is to predict the extent of P–E fit problems arising as a consequence of the combination of a person's functional limitations and barriers in the physical environment, it agrees well with Lawton's ecological model (Iwarsson & Slaug, 2001). More specifically, in the Enabler Concept (Steinfeld et al.), P–E fit is operationalized in terms of functional capacity (personal component) and environmental demands (environmental component; see Iwarsson & Ståhl, 2003). Iwarsson and colleagues have developed a methodology for the assessment and analysis of P–E fit in housing that is based on the Enabler Concept, rendering predictive, objective, and norm-based assessments and analyses of P–E fit problems in the physical home environment possible (Iwarsson & Slaug).
When it comes to the assessment of ADLs, substantial bodies of assessment instruments as well as ADL research literature are available, representing different instrument administration procedures, measurement levels, and scaling approaches (Wade, 1994). For example, note that there is a distinction between performance potential and actual performance (Diehl & Willis, 2003). Further, self-reports and assessments based on observations yield different results in the same study sample (Jette, 1994). Finally, far from all ADL instruments are conceptually pure; in several instruments, for example, basic functional capacities such as body functions are included together with activities.
For a reliable and valid assessment of P–E fit in housing and of ADL dependence to be made possible, methods meeting basic methodological requirements are needed. Further, the assessment tools must reflect the targeted theoretical constructs. In order to differentiate between the personal component of P–E fit and the adaptation level of the general ecological model, physical capacity as well as ADL dependence must be validly operationalized. When it comes to testing relationships between different variables, all variables involved should reflect the same perspective; that is, if P–E fit is based on objective observations of actual circumstances, ADL dependence should be assessed likewise.
Applying the new approach to conceptualizing and measuring P–E fit as described, based on Lawton's ecological model and the Enabler Concept, my objective in this study was twofold. My first aim was to give a long-term description of P–E fit problems in housing and ADL dependence among older adults living in the community. My second aim was to study the relationship between P–E fit problems in housing and ADL dependence and to test the docility hypothesis at two points in time.
|
Design and Methods |
|---|
|
TOPAbstractDesign and MethodsResultsDiscussionReferences |
|---|
Study Sample
The study district was a rural municipality situated in the south of Sweden. At baseline, the municipality had 49,458 inhabitants. Those aged 75–84 years numbered 3,504 (i.e., 7% of the population). I performed a random sampling procedure, using the Swedish national population register. The baseline sample comprised all individuals born on Days 15 or 16 of any month of the years between 1910 and 1919, resulting in 222 individuals. Of these, I excluded 12 elders who were living in sheltered housing facilities or staying at an address outside the district, leaving 210 possible participants. Four of those died before the study began, and another 8 people moved to sheltered housing or out of the district, resulting in 198 individuals. Among these, 133 (68%) agreed to participate (Iwarsson, 1997; Iwarsson et al., 1998). For the follow-up 6 years later, I identified the survivors by means of the Swedish national population register. Since baseline, 42 individuals (32%) were deceased and 2 had moved out of the study district. Out of the 89 participants possible to reach, 72 (81%) agreed to participate. That is, at baseline as well as at follow-up, the current study was based on the 72 participants possible to follow at two points in time (Figure 1). The study sample is further described in Table 1.
|
|
Procedures
At baseline as well as follow-up, we sent information letters to intended participants, followed by telephone calls in order to ask for informed consent and confirm an appointment. Data collection was performed during home visits by one of two experienced occupational therapists, trained to administer a study-specific interview questionnaire as well as two previously established instruments (subsequently described) for interview and observation. Besides descriptive data, the data collection comprised information on functional limitations in the person, use of assistive devices for mobility, eventual home modifications accomplished or planned, environmental barriers in the home and its close surroundings, and dependence in personal (ADL) and instrumental (IADL) activities of daily living. Therapists recorded the actual situation at each data-collection occasion.
Measures
To assess P–E fit, I had therapists administer the Housing Enabler instrument (Iwarsson & Slaug, 2001). Reliable administration requires the use of trained interviewers or observers who are able to collect valid information on the personal component of P–E fit during interviews, as well as to validate interview responses by means of observations as respondents move around and act in different parts of their home environments. In addition, interviewers or observers have to be trained for reliable assessment of the presence of environmental barriers, on the basis of national norms and regulations.
The Housing Enabler is administered in three steps.
The first step is the assessment of functional limitations and dependence on mobility devices (personal component of P–E fit). This is a combination of interview and observation, used to assess the presence or absence of functional limitations (13 items) and dependence on mobility devices (2 items). The majority of the functional limitation items concern physical capacity, whereas 1 item touches on cognitive capacity and 3 concern perception (vision and hearing; see Table 2).
|
The third step is the calculation of the P–E fit score. For each environmental barrier item, the instrument comprises predefined severity ratings (Steinfeld et al., 1979), operationalized as points quantifying the severity of the problems predicted to arise in each case. The accessibility severity scale is scored from 1 (potential accessibility problem) to 4 (very severe accessibility problem or impossibility). On the basis of the assessments accomplished in Steps 1 and 2 and with the use of a complex matrix comprising the predefined severity ratings from 1 to 4, the profile of functional limitations identified in each person is juxtaposed with the environmental barriers found present in the home environment. This analysis is run item by item, and each P–E fit incongruency is quantified by means of the scale. The sum of all the predefined points yields a score outlining the problems anticipated. Thus, the magnitude of problems caused by a particular combination of functional limitations and environmental barriers (i.e., the degree of objective, norm-based P–E fit problems in the home) can be calculated. In cases in which no functional limitations or dependence on mobility devices are present, the score is always zero. In cases in which the person has functional limitations or is dependent on mobility devices, higher scores mean more P–E fit problems. For reliable and valid computations, the Housing Enabler software is recommended (Slaug & Iwarsson, 2001; demo version available at http//:www.enabler.nu).
Content validity and interrater reliability have been established for functional limitations and mobility devices, 
= 0.87, as well as for environmental barriers, = 0.68 (Iwarsson & Isacsson, 1996a). The instrument has been used in different studies, and based on accumulated empirical experiences, it subsequently has been revised for improved reliability and validity (Iwarsson & Slaug, 2001).
For assessment of dependence in ADL and IADL performance, I used the ADL Staircase (Sonn & Hulter-Åsberg, 1991), revised version (Iwarsson, 1998; Iwarsson & Isacsson, 1997a). The nine-item ADL Staircase was based on Katz's ADL index (Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963) and Swedish studies adding IADL items to the original Katz instrument (Sonn & Hulter-Åsberg). The ADL Staircase comprises the following activities: feeding, transfer, going to the toilet, dressing, bathing, cooking, shopping, cleaning, and using transportation. In the original version the transportation item was defined for use in urban areas, whereas in the revised version there was an alternative item intended for use in rural areas. The measurement level of the instrument is the level of dependence a person has on other people. I ensured that the assessment was administered by means of a combination of interview and observation (Sonn & Hulter-Åsberg) as originally recommended by Katz and associates and that the results were scored on a three-graded scale: dependent, partly dependent, or independent.
Reliable administration of the ADL Staircase requires trained interviewers or observers, who are able to collect valid information on actual ADL dependence during interviews as well as to validate interview responses by means of observations of basic mobility as respondents move through different parts of their home environments. In line with earlier studies (Iwarsson, 1998; Iwarsson & Isacsson, 1997a), the ADL Staircase data collected for the current study fulfilled internal validity requirements (coefficient of scalability, baseline = 0.68, follow-up = 0.76; coefficient of reproducibility, baseline = 0.95, follow-up = 0.96).
Data Analyses
For functional limitations and dependence on mobility devices (personal component of The P–E fit) as well as environmental barriers in the home (environmental component of the P–E fit), I analyzed changes by means of McNemar's test. I analyzed magnitudes of P–E fit problems by means of the Housing Enabler software (Slaug & Iwarsson, 2001). Based on the raw data collected and the predefined severity ratings in the complex instrument, the software generates total P–E fit problem scores. Thereafter, I analyzed changes in magnitudes of P–E fit problems from baseline to follow-up by means of the sign test.
For ADL dependence, I investigated changes from baseline to follow-up on the basis of the three-graded scale of the ADL Staircase (Fänge, Lanke & Iwarsson, 2004), computing ADL ranks (Iwarsson et al., 1998, Iwarsson & Lanke, 2004). The ranking approach represents an alternative data-treatment principle respecting ordinal scale properties. The raw data outcome of an ADL Staircase assessment is a nine-item variable describing response patterns resulting from the three-graded scale. For analysis of changes between two assessment occasions using ADL ranks, all data from both assessments should be considered simultaneously. The rank of all nine items is computed by comparing one assessment with all other assessments in the 2n sample, one by one, and then counting the number of times that the first assessment is higher than the second, and the number of times the two assessments are equal. That is, the assignment of ranks is based on an itemwise comparison of response patterns. If one response pattern has every item coded at least as high as that of the response pattern with which it is compared, with strict inequality for at least one item, then the former should be considered to be more dependent than the latter. In all other cases, the two response patterns are considered to be noncomparable and treated as tied. After computation of ADL ranks, I used the sign test for the investigation of changes between baseline and follow-up.
I computed the correlation between P–E fit problems in housing and ADL dependence by use of Spearman's rank correlation. Thereafter, in order to test the docility hypothesis, I performed separate correlation analyses of participants with higher or lower P–E fit problem scores. My rationale for this approach to analysis was that the total Housing Enabler score is an estimation of the P–E fit, in congruence with how the relation between competence and environmental press is expressed in Lawton's general ecological model (Lawton, 1980; Lawton & Nahemow, 1973). As outlined in the introduction, in this study ADL dependence represents the adaptive level in the Lawton model (Iwarsson et al., 1998). Thus, I used the median P–E fit problem scores at baseline and follow-up as cutoffs for dichotomization, and thereafter I calculated the correlation between ADL dependence and P–E fit problem scores in each of the subgroups of participants at baseline and at follow-up. I considered results to be significant at p <.05.
|
Results |
|---|
|
TOPAbstractDesign and MethodsResultsDiscussionReferences |
|---|
Personal Component of P–E Fit
At baseline, 24% of the individuals in the sample had none of the functional limitations assessed, and they were not dependent on mobility devices; at follow-up, this percentage had decreased to 12% (p <.0001). On both occasions, the most prevalent functional limitations were difficulty in bending and kneeling (baseline = 43%, follow-up = 46%) and limitations of stamina (baseline = 31%, follow-up = 33%). From baseline to follow-up, the prevalence of severe loss of hearing (p <.0001) and loss of upper extremity skills (p =.025) increased. At baseline, none of the participants used a wheelchair, whereas 3 participants did at follow-up. Reliance on walking aids increased between the two data collection waves, from 8% at baseline to 47% at follow-up (p <.0001; see Table 2).
Environmental Component of P–E Fit
Most participants lived in homes with high housing standards, but I identified environmental barriers in every home assessed. To exemplify, in at least 90% of the homes' kitchens, no work surface at a height suitable for sitting was available, the bathroom sinks were placed at a height for use only when standing, with pipes causing obstruction under them, and telephones with amplified sound were lacking. At baseline as well as follow-up, in virtually all bath and laundry rooms as well as in the dwelling in general, control and hardware use required hand function.
At follow-up, 13 participants (18%) had taken concrete measures for making it possible to continue living in their present home, and 11 of those had applied for and been granted economic support for their home modifications. The most common measures concerned bathrooms (corner shower instead of a bathtub), the removal of thresholds, and changes at entrances. Eleven percent of the participants reported that they had taken some security measures, for example, removing loose carpets, and 7% had rearranged furniture so that they now slept on the ground floor instead of upstairs. Despite the fact that some participants had undertaken concrete measures, the prevalence of single environmental barriers was stable between baseline and follow-up; that is, in 158 of the 188 barriers assessed (84%), I identified no significant changes. Among the 30 barriers changing significantly in prevalence between baseline and follow-up, 15 were significantly more prevalent at follow-up, whereas for the remaining 15 barriers the prevalence decreased. For example, in 82% of the cases at baseline but in 71% at follow-up, there were stairs at the entrance, without an accessible alternative way (p =.001). Further details are presented in Table 3.
|
Relationships Between P–E Fit and Functional Dependence
At baseline, the correlation between ADL–IADL dependence and P–E fit in housing was nonsignificant, whereas at follow-up it reached significance at rs = 0.52 (p <.0001). At baseline, the analyses on the subsamples (based on dichotomization of the total sample) resulted in very low, nonsignificant correlation coefficients. At follow-up, rs = 0.45 (p =.0059) in the subsample with more P–E fit problems in housing, whereas among those individuals with fewer P–E fit problems, the correlation was nonsignificant (rs = 0.30).
|
Discussion |
|---|
|
TOPAbstractDesign and MethodsResultsDiscussionReferences |
|---|
Nevertheless, the changes I found in the environmental component give detailed information on the types of environmental barriers in housing contributing to P–E fit problems. The results demonstrate highly significant relationships between P–E fit problems and functional dependence at follow-up, and the results of the subsample analyses lend support to the docility hypothesis (Lawton & Nahemow, 1973; Lawton & Simon, 1968). That is, changes in one of the components of the ecological model (e.g., personal competence) affect the adaptive level, operationalized here as ADL–IADL dependence. Further, individuals with more P–E fit problems are more vulnerable, demonstrated by the fact that the relationship between P–E fit problems in housing and ADL–IADL dependence at follow-up was significant only in the subsample of individuals with more P–E fit problems. Because the operationalization of the docility hypothesis was limited in terms of the variables covered by the instruments used, these results are at best suggestive. Further, as originally expressed by Lawton and Nahemow, docility means that individuals with low competence are much more vulnerable to environmental demands than those with high competence, and details are crucial as to what they can manage in everyday life. As demonstrated by the current study, the major reason for increasing P–E fit problems in housing was not changes in the environmental component but in the personal component, to some extent contradicting the statement by Lawton and Nahemow.
In a previous article based only on baseline data from the current study (N = 133; Iwarsson et al., 1998), subsample analyses revealed a significant relationship between P–E fit problems and ADL–IADL dependence among participants with more P–E fit problems in housing, whereas among participants with fewer problems no such relationship was found. Presumably because of the larger sample size and more variation in terms of functional limitations and ADL dependence earlier on in the aging process, those results already supported the docility hypothesis. Current results demonstrate that, to determine the kinds of changes in P–E fit that cause docility, a follow-up design is required. However, the high mortality rates in old age reduced the possible follow-up sample to approximately one third, and some of the dropouts declined follow-up because of ill health. The results of any longitudinal gerontological study have to be interpreted bearing such well-known facts in mind, and the participants at follow-up of course represented the proportion of the aging population with overall higher health status and less rapid functional decline. Our understanding of P-E fit relationships and how they are related to ADL/IADL dependence throughout the aging process is still insufficient, in particular when it comes to such relationships in sub-samples displaying different levels of health. Data collection for a 10-year follow-up is in progress, and forthcoming results will shed additional light on the complex relationships targeted by the current study. The data also provide a basis for additional analysis, such as survival analysis, which could generate important knowledge regarding P–E fit and mortality.
Given the impact Lawton's general ecological model has had over the years (Scheidt & Norris-Baker, 2003; Wahl & Wiesman, 2003), empirical results elucidating its mechanisms contribute to current knowledge. Even though the current study demonstrated a significant relationship between P–E fit problems in housing and ADL–IADL dependence at follow-up, the correlation coefficient was only moderate. Thus, this relationship only explains a minor facet of the complex problem of functional dependence. In developing an integrated perspective on how environmental variables affect human functioning, it is important to consider the connection between physical environments and their social and cultural implications (Iwarsson et al., 1998), as well as many other dimensions.
Lawton's general ecological model has been criticized and further developed over the years, because it deals neither with people's needs, attitudes, knowledge, preferences, and perceptions, nor with the fact that the environment is characterized by resources as well as demands (Scheidt & Norris-Baker, 2003; Wister, 1989). For example, instrumental disabilities do put pressure on people to change their living environments, but features of living arrangements, such as with whom one lives, are a vital part of the system of environmental support, transcending the physical housing environment (Jackson, Longino, Zimmerman, & Bradsher, 1991). It should be kept in mind that the assessment of the environmental component used in this study was limited to one of the five environmental classes as defined by Lawton and Nahemow (1973), that is, the physical environment. Thus, it was not possible to distinguish between older adults who functioned at their highest possible level because of a supportive social environment and those whose functions could have been increased if the physical environment were modified. Even if there were differences in rates of living alone versus living with others between baseline and follow-up, given the limited sample size I did not considered it valid to run subgroup analyses. Further, it should be kept in mind that the current study was based on the general ecological model, with no ambitions to embrace the much more transactional and dynamic perspectives of the complex, largely taxonomic articulation that emerged from its revisions. The results, however, serve as a guide for future research involving a wider range of variables.
The study of home life is necessarily time intensive, requiring face-to-face contact with single individuals, and the research efforts involved in a study like this are substantial. Even if the sample used for the current study was limited, the response rate at follow-up was notably high, and the study exemplifies Gitlin's statement (2003) that the home environment is a potential laboratory for developing and testing constructs that may extend to environmental gerontology as a whole.
When it comes to the methodology used in the current study, according to Gitlin (2003), the Housing Enabler (Iwarsson & Slaug, 2001) is the only promising research tool available in this field. Further, I thoroughly considered the choice of an appropriate ADL–IADL instrument. Among patients with considerable reduction in functional capacity, instruments measuring ADL ability are sufficient. For discriminative power among those with fewer functional deficits, as for example in the current general population study of older adults living in the community, assessments comprising IADL are necessary (Iwarsson, 1998; Iwarsson & Isacsson, 1997a; Sonn & Hulter-Åsberg, 1991). Further, in order to match the objective assessment of P–E fit problems, I found it necessary to use an ADL instrument based on interview and observation instead of self-reports. Such an assessment tool must be reliable and valid, and in longitudinal studies it is imperative that the scale is sensitive enough to detect changes over time. Even if a multitude of ADL and IADL assessments exist (see, e.g., Wade, 1994), by applying these criteria for choosing an instrument, one restricts the arsenal. The ADL Staircase is not widespread outside of Scandinavia, but as it stands the instrument underwent successive application and testing of appropriate analysis methods (Iwarsson, 1998; Iwarsson & Isacsson, 1997a; Iwarsson et al., 1998). The rank data treatment principle applied in this study is novel and therefore scarcely known, but it optimally utilizes the discriminative power of the three-graded scale of the ADL Staircase. In addition, the ordinal properties as well as the hierarchical order between items in each sample are fully utilized, thus respecting the ordinal character of the data (Iwarsson & Lanke, 2004).
Based on the notion that P–E fit is an interactive construct (Iwarsson, 2004), the results of this study demonstrate that P–E fit problems in housing are related to older adults' ADL–IADL dependence, and the relationship grows stronger with advancing age. The results indicate that the increase of P–E fit problems in housing among older adults over time mostly results from functional decline, although the increase also to some extent results from changes in the environmental component. The results are based on detailed, reliable, objectively collected information on the personal and environmental components of P–E fit as well as on functional dependence, and they lend support to the docility hypothesis. That is, the study implies that P–E fit problems in housing have a more negative impact on daily activity among older adults with more functional limitations and dependence on mobility devices. An important contribution of this study is that it demonstrates that research utilizing a methodology that allows for operationalization of aspects of Lawton's ecological model, such as the Housing Enabler, has the potential to increase our understanding of how the ecological model's main components change and interact over time.
When it comes to practical implications, knowledge on how housing accessibility problems in the aging population increase with advancing age can be used to reduce the need for relocation and sheltered housing. Because the notion of P–E fit implies that the relation of the personal and environmental components can be influenced by interventions, the fact that the increase of accessibility problems mainly is caused by functional decline is a strong argument for preventive measures targeting the environmental component instead. A major advantage of the analysis of accessibility problems based on the Housing Enabler is the possibility to identify which environmental barriers cause the most accessibility problems (Iwarsson & Isacsson, 1996b; Iwarsson & Slaug, 2001). Such analyses identify the environmental barriers that, in relation to the prevalence of functional limitations and dependence in an individual or a sample or population, generate the highest accessibility scores. For example, based on baseline data from this study (N = 133; Iwarsson & Isacsson, 1996b), an analysis of weighted environmental barriers demonstrated that among the single items of the environmental component of the Housing Enabler, the environmental barrier "no grab bars at shower/bath and/or toilet" generated the most accessibility problems. Such detailed information about the nature of the environmental barriers causing accessibility problems is directly applicable in the planning of housing for senior citizens as well as in individual home-modification cases, representing a concrete measure to reduce poor P–E fit among older people living in the community.
|
Footnotes |
|---|
1 Division of Gerontology and Caring Sciences, Faculty of Medicine, Lund University, Sweden.
Decision Editor: Linda S. Noelker, PhD
Received for publication March 5, 2004. Accepted for publication October 27, 2004.
|
References |
|---|
|
TOPAbstractDesign and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. X. Marquez, E. E. Bustamante, B. J. Blissmer, and T. R. Prohaska Health Promotion for Successful Aging American Journal of Lifestyle Medicine, January 1, 2009; 3(1): 12 - 19. [Abstract] [PDF]
|
|
|
|
 |
|
 S. Iwarsson, H.-W. Wahl, C. Nygren, F. Oswald, A. Sixsmith, J. Sixsmith, Z. Szeman, and S. Tomsone Importance of the Home Environment for Healthy Aging: Conceptual and Methodological Background of the European ENABLE-AGE Project Gerontologist, February 1, 2007; 47(1): 78 - 84. [Full Text] [PDF]
|
|
|
|
|
|
C. Nygren, F. Oswald, S. Iwarsson, A. Fange, J. Sixsmith, O. Schilling, A. Sixsmith, Z. Szeman, S. Tomsone, and H.-W. Wahl Relationships Between Objective and Perceived Housing in Very Old Age Gerontologist, February 1, 2007; 47(1): 85 - 95. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Oswald, H.-W. Wahl, O. Schilling, C. Nygren, A. Fange, A. Sixsmith, J. Sixsmith, Z. Szeman, S. Tomsone, and S. Iwarsson Relationships Between Housing and Healthy Aging in Very Old Age Gerontologist, February 1, 2007; 47(1): 96 - 107. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
