The Housing Enabler Screening Tool

Abstract

The ambition underlying this study is to provide a valid and efficient screening tool targeting accessibility problems in the ordinary housing stock. The study aims to test the feasibility and interrater agreement of a screening-tool version of the Housing Enabler in a real estate company practice context. Two rater pairs administer the screening in 35 apartments in a Swedish municipality. The overall agreement for the rater pairs was 83% and 78%. For 29 of the 61 items of the tool, the agreement was >80% for both rater pairs; for 8 items, it was <80%. The findings show that the tool is feasible for use in a real estate practice context but that there is a need for more rater training and an instrument manual. This study contributes to the development of research-based strategies for the identification of accessibility problems in ordinary housing, with the long-term aim to increase accessibility in the housing stock, ultimately supporting activity and participation among senior citizens.

Keywords

accessibility environmental barriers home assessment housing provision physical environment transdisciplinary research

Introduction

Currently in many countries, housing provision for senior citizens and persons with disabilities is a burning issue (WHO, 2002). It has since long been known that accessibility problems threaten public health (Christiansen & Townsend, 2004) and increase the demands on society to provide home help services and purpose-built housing (Iwarsson, 1997), but such problems prevail. One challenge is the lack of reliable methods for assessing and analyzing accessibility problems (Gitlin, 2003). The measures that are taken are too rarely based on systematic assessment, and there is often no consideration for the different perspectives that should be included in the analysis preceding the measures. With improved methods, the work of creating a society that is accessible and usable for everyone can be made more efficient (Iwarsson & Slaug, 2010a).

The role of the home environment for maintaining daily activity despite functional limitations is now widely recognized in research and practice (Wahl, Oswald, Fänge, Gitlin, & Iwarsson, 2009). For example, there is a long tradition of linking environmental press to loss of competence (Lawton & Nahemow, 1973), resulting in negative consequences. Applied research has also emphasized the home environment as a critical factor supporting or undermining individual functioning (Stark, 2004). Given that the majority of senior citizens live in private households, with those older than 80 years of age spending most of their time in the home or immediate surroundings (Heyl, Wahl, & Mollenkopf, 2005), provision of accessible housing is a burning issue for society. Also, the majority of older people prefer to “age in place” as long as possible (e.g., American Association of Retired Persons [AARP], 2003). Moreover, several population-based studies suggest that gains in functioning in older adults over the past few decades may be attributable in part to the reduction of environmental barriers (Freedman, Martin, & Schoeni, 2002; Spillman, 2004). Consequently, improved accessibility in ordinary housing would promote health and prevent activity limitations and participation restrictions among senior citizens and persons with disabilities (Iwarsson, 2005) so that relocations to assisted-living facilities could be postponed.

One basic prerequisite for the provision of accessible housing in the ordinary housing stock is problem identification on the basis of valid and efficient screening. With reliable survey data on housing accessibility at hand, housing provision could most likely be made more efficient. Currently the awareness of these issues is increasing, among politicians as well as actors in housing provision. For example, in Sweden, attempts to introduce survey strategies are in progress. However, international literature search reveals a lack of evidence-based, valid, and reliable methods for accessibility screening. Overall, research targeting home environments is time consuming and requires a lot of resources (Gitlin, 2003), and as yet few efforts have been made to implement research-based methodology into practice contexts (Fänge, Risser, & Iwarsson, 2007).

Most relevant from a societal planning perspective, accessibility is a question of compliance with official norms and standards for environmental design and represents an objective perspective (Preiser & Ostroff, 2001). According to Iwarsson and Ståhl (2003), accessibility is a relative concept defined as a person–environment (P-E) relationship, that is, the relationship between the profile of functional limitations of the individual and the demands in the environment. In other words, accessibility comprises a personal component and an environmental component; that is, an aspect of person–environment fit as described by Lawton and Nahemow (1973). Adopting this definition for screening purposes is a challenge, since the methodology must represent a valid operationalization of the construct.

The Housing Enabler instrument (Iwarsson & Slaug, 2001) is based on the aforementioned definition of accessibility and is supported by robust research evidence (see, for example, Fänge & Iwarsson, 2003; Iwarsson, Nygren, & Slaug, 2005; Oswald et al., 2007). It can be applied with great flexibility and is, therefore, suitable for assessments from different perspectives. The fundamental principle for the method is that one assesses both the individual’s functional limitations and the demands of the environment, followed by an analysis of the magnitude of accessibility problems. The version of the instrument published 10 years ago comprises 2 dichotomous checklists, for the personal (15 items) and the environmental component (188 items), respectively (Iwarsson & Slaug, 2001). The environmental component assessment is based on standards and guidelines for housing design as expressed in national legislation and public documents. By means of a scoring matrix with predefined severity ratings of the magnitude of accessibility problems generated when juxtaposing the personal component assessment results with those of the environmental component assessment, the instrument generates a total score that quantifies the magnitude of accessibility problems in each case (see Iwarsson, 2005, for detailed information).

The Swedish version of the Housing Enabler was issued in English in 2001, in Danish in 2008, and in Finnish in 2009. Besides this, project-specific research versions of parts of the manual and the rating forms are available in German, Hungarian, and Latvian, as well as Icelandic and Portuguese. Aspects of reliability of the very first version of the instrument were tested back in 1994-1995. Seventy-seven Swedish occupational therapists took part, and the results showed that the interrater agreement of the instrument was high or very high. Between 1994 and 2000 the Housing Enabler was chiefly used by occupational therapists in research, tuition, and working practice. In the version subsequently published (Iwarsson & Slaug, 2001), a number of changes were made to further improve the reliability and validity of the instrument. A project-specific version that was used in the European ENABLE-AGE Project was tested (Iwarsson, Nygren, & Slaug, 2005) and applied in empirical research involving 1,918 older adult participants in 5 countries (see, for example, Oswald et al., 2007). Since then a version of the instrument in which the environmental component had been adapted for use in four Nordic countries has been tested in a comparable way (Helle et al., 2010). Moreover, the construct validity was tested in a Swedish study with the focus on housing adaptations (Fänge & Iwarsson, 2003).

The Housing Enabler was originally intended primarily for use in occupational therapy: in clinical practice, research, education, and in-service training. Occupational therapy practices are still focused chiefly on the rehabilitation of patients after injury or disease or with reduced functional capacity on account of aging. However, there is increasing awareness of and interest in developing strategies for health promotion and preventive measures, and therapists also try to influence the building process and public planning as a whole to promote health and prevent restrictions on activity and participation in the population. Having trained and cooperated in recent years with raters from different professional backgrounds, we now recommend that the Housing Enabler should also be used by other professional groups, such as architects, real estate staff, and so on, to create potential for influences on housing provision at large.

Although the tool is considered robust for research and for detailed assessment of accessibility problems in individual cases, the main concerns for its widespread use in practice contexts are the time consumption for on-site assessments, the level of detail of the environmental assessment (Fänge et al., 2007), and needs for rater training. As yet, no studies involving nonacademic staff with a technical background, employed in, for example, real estate companies or equivalent have been carried out.

Since a reduced version of the Housing Enabler with potential for screening purposes would be of great value, by way of two parallel, systematized, and iterative approaches utilizing expertise from a range of different disciplines, in a previous study we identified a set of core environmental barrier items (Carlsson et al., 2009). The reduced set (N = 61) of items demonstrated near-perfect correlation to the 188-item original version. The simulations in large databases with personal as well as environmental component data underlying the results of the previous study produced a list of the environmental barrier items contributing the most to the generation of accessibility problems. Therefore, utilizing this list implies that assessment of individuals by means of the personal component of the original instrument is not necessary; that is, screening of environmental barriers causing accessibility problems by means of a tool based on this list does not require professional health care or social services competence, opening up for a more widespread use of such methodology. Although we assume that such screening could be accomplished by persons employed in the real estate sector, that is, with a technical competence profile, this assumption remains to be tested.

Informed by 20 years of research and practice experiences based on the development and application of the Housing Enabler instrument in various contexts in several countries (Iwarsson & Slaug, 2010a), we aim to provide a valid and efficient screening tool targeting environmental barriers causing accessibility problems in the ordinary housing stock. Such development is a lengthy process and needs to be conducted in a stepwise manner. Starting out with a focus on basic prerequisites of assessment instruments, the present study aimed to test the feasibility and the interrater agreement of a pilot version of a screening tool version of the Housing Enabler in a real estate company practice context.

Method

Study Context

This study was accomplished in collaboration with the public real estate company in a medium-sized municipality in south Sweden. This company was selected on the basis of the criterion that high representatives had shown active interest for participation in the project. The total stock of apartments of the company was approximately 8,700 and represented a wide variety of age, building designs, and so on. Only a few apartments were situated in the city centre, and the rest were situated in the outskirts of the city. Most commonly, apartments owned by the company were situated in clusters of buildings of the same type; that is, there were often several or many identical apartments in one area. As for all of Sweden, the standard of all apartments was high.

Instrument

Based on the reduced list of environmental barrier items identified in our previous study (Carlsson et al., 2009), we constructed a study-specific assessment form (i.e., a pilot version of the Housing Enabler Screening Tool) was constructed. To link to the original Housing Enabler instrument (Iwarsson, & Slaug, 2001), we chose to retain the original numbering of the items, but based on former experiences, we adjusted the actual order of the items on the forms to accommodate a screening situation. In the original instrument, the assessment of the environmental barrier items is dichotomous (“present”/”not present”), but in the study-specific form two alternatives were added: “not assessed” and “not relevant.” Since it might not always be possible during a screening situation to gain access to all parts of a building included in the instrument, we anticipated that “not assessed” would be necessary as a response alternative. The rationale for including “not relevant” was that there are apartments that do not have all building features included in the instrument. For example, most apartments in Swedish public real estate companies do not have any basement or upper floor and, consequently, no indoor stairs. The first and last authors (SI & AMF) piloted the study-specific assessment form in a few different apartments, followed by revisions concerning layout, ordering of the items, and accuracy of the item definitions and instructions.

Raters

To test the feasibility of the pilot version of the Housing Enabler Screening Tool for use in a real estate company context, two employees were selected and recommended by the head of the company’s marketing division. Both of them had a technical background from education equivalent to that of a senior high school, complemented with training for housing management and caretaking. Since many years, each of them had the responsibility for several estate districts in the municipality. Their tasks were, for example, to serve as team leaders for service staff and contact persons for persons living in their districts and to make home visits to monitor the status of apartments when tenants were about to move.

To test the interrater agreement, SI and AMF served as two of the raters. They are both senior scientists, university teachers and registered occupational therapists, and very experienced in using the original Housing Enabler instrument. The first author constructed the original instrument and has >10 years experience as the course leader of rater courses and in different countries. These two authors had leading roles in the identification of the environmental barrier items to be included in the screening tool (Carlsson et al., 2009).

Rater Training

According to the longstanding experiences from courses and practice application of the Housing Enabler instrument, for valid and reliable use rater training is imperative (Helle et al., 2010; Iwarsson et al., 2005). Consequently, at the prospect of the data collection planned, SI and AMF accomplished a 6-hr introduction session with the 2 real estate company employees. By way of introduction, a short lecture on basic concepts and assessment principles was given. Since assessment according to the Housing Enabler Screening Tool requires a familiarity with assessing physical barriers in the built environment and up-to-date knowledge of standards and functional demands from the point of view of accessibility according to national guidelines, the training included reading of and discussion on such material. Since at the time for this study no manual for the Housing Enabler Screening Tool had been produced, the raters were instructed to use selected parts of the manual for the original Housing Enabler instrument (Iwarsson & Slaug, 2001). Step by step, they were guided through this manual. The introduction session also included practical training in an authentic rating situation, followed by clarifying discussions based on the raters’ questions related to the assessment task. The two real estate company employees thereafter tested the pilot version of the Housing Enabler Screening Tool on their own in 1 to 2 apartments each, followed by two 2-hour follow-up session led by AMF. Different types of assessment challenges were discussed, clarifications were given, and so on. Upon completion of the training, the rater constellation for the data collection was determined. Each rater pair consisted of one of the scientists and one of the real estate company employees; the constellation was the same during the entire data collection process.

Data Collection

The data collection was accomplished integrated with the routine visits made for inspection of apartments when a tenant had decided to move. During a 3-month period in the fall of 2008, and in the housing areas where the two real estate employees were responsible, the two raters of each pair together visited the apartments consecutively coming up for such routine inspections. In all, 35 apartments of varying size, design, and age were screened. The apartments had been constructed during the period from 1959 to 2007, but most of them were constructed during the 1980s or 1990s. There were apartments situated on ground level, as well as those situated in buildings with several floors, with or without elevators. Thus, the apartments represented a wide range of the Swedish stock of apartments in blocks of flats in publicly owned real estate companies.

During most of the visits, the tenant and sometimes one or several family members were present, in fully furnished apartments. In a few cases, the apartments were empty and nobody except the two raters was present. The main purpose of the visits was the routine inspection induced by relocations, and prior to the visits where a tenant was expected to be present, the real estate employee rater asked for permission to bring along the other rater (scientist). By way of introduction at the actual visit, the scientist presented herself and briefly spoke about the purpose of the study, and informed consent was established. First, the real estate employee administered his routine inspection targeting the general status of the apartment, and the other rater (scientist) started to administer the pilot version of the Housing Enabler Screening Tool. As soon as the real estate employee had completed his routine inspection, he administered the screening as well. Each visit lasted 20 to 45 min. No discussions concerning the rating of individual items took place between the two raters during or after the screening.

During and in between the data collection visits, the scientists documented comments and considerations raised by the real estate employee raters, as well as by themselves. To ensure the quality of the data collection procedures over time, the scientists continuously checked with each other. After completion of the data collection, the four raters met during a concluding session. The focus of this discussion was on aspects of feasibility, but all kinds of considerations of relevance for the study aim were of interest. One of the scientists (AMF) documented the discussions and conclusions.

Data Analysis

Frequencies for the most prevalent environmental barriers causing accessibility problems were calculated, using the ratings from each pair-wise assessment made by the scientist as the golden standard. The level of agreement was estimated by calculating the percentage agreement between the raters of the 2 rater pairs, for each of the 61 items. To estimate the agreement for the total data set, the mean percentage of the 61 items was calculated for the 2 rater pairs. Furthermore, the level of agreement was analysed by classifying items as follows:

Items where the raters of both rater pairs agreed in at least 80% of the cases, either on the fact that the environmental barrier causing accessibility problems in question was present, or not.

Items where the raters in one of the rater pairs agreed in at least 80% of the cases on whether the environmental barrier causing accessibility problems in question was present or not, whereas the raters of the other pair agreed in less than 80% of the cases.

Items where the raters in both pairs agreed in less than 80% of the cases on whether the environmental barrier causing accessibility problems in question was present, or not.

The content of the documentation of comments and considerations, from the data collection process and the concluding discussion session, was analysed using a simple qualitative categorisation approach.

Results

Some of the environmental barriers causing accessibility problems were present only in a few apartments, and a considerable number of them were very common. Three items concerning controls/operable hardware were present in nearly all apartments screened, as was irregular outdoor walking surfaces. High thresholds and/or steps at entrances and to balconies were present in 28 of the apartments (Table 1).

Table 1.

Prevalence of the 20 Most Common Accessibility Problems in 35 Apartments

Environmental barrier causing accessibility problems	N
Exterior surroundings
Irregular walking surfaces	32
No/too few seating places outdoors	27
Entrances
High thresholds or stairs in the entrance	28
High threshold/level difference/step to sitting-out place/balcony	28
Insufficient manoeuvring space at entrance doors	26
Narrow doors	23
No level area in front of entrance doors	23
Narrow door to sitting-out place/balcony	23
Indoors
Use of controls and hardware in the house requires hands	33
Use of controls and hardware in the kitchen requires hands	33
Use of controls and hardware in hygiene areas requires hands	32
Wash basin placed at a height for standing use (81 cm high or more)	31
Storage areas can only be reached via stairs/threshold	30
Narrow doors	29
No surface at a height suitable for sitting work in the kitchen	28
Insufficient manoeuvring areas where turning is necessary	28
No place to sit in shower/bath	28
Wall-mounted cupboards and shelves placed extremely high	25
No grab bars at shower/bath and/or toilet	25
Toilet with standard height or lower	25
Narrow passages/corridors in relation to fixtures/design	22
Bathtub instead of shower stall	21

The overall agreement for the two rater pairs was 83% and 78%, respectively. For 29 items, the raters of both pairs agreed in 80% of the cases or more, on whether the environmental barrier causing accessibility problems was present, or not (Table 2). Out of these 29 items, 8 items concerned the section exterior surroundings (A); 7, the entrance section (B); and 14, the indoor section (C). For 20 of the items, the raters of one of the pairs agreed in 80% of the cases or more, whereas the raters of the other rater pair agreed in less than 80% of the cases. For 8 items the raters of both pairs agreed in less than 80% of the cases. Two of these items concerned exterior surroundings, and two were in the section entrances, whereas four items concerned environmental barriers causing accessibility problems indoors (Table 3). For another four items, there were only data from one of the rater pairs.

Table 2.

Accessibility Problems (n = 29) Where the Agreement was 80% or More in Both Rater Pairs (n = 35).

Environmental barrier causing accessibility problems	Rater Pair 1 (%)	Rater Pair 2 (%)
Exterior surroundings
Narrow paths	100	100
No resting surfaces or too far between resting surfaces on slopes	83	100
Refuse room/refuse bin can only be reached via steps or other difference in level	85	86
Unstable walking surfaces	92	90
Letterbox can only be reached via steps or other difference in level	90	89
Steep gradients	83	90
Routes with steps	92	93
High kerbs	92	84
Entrances
Complicated/illogical opening procedure	92	95
Stairs the only route	100	100
No handrails	100	89
Handrails too short	83	88
High thresholds and/or steps at the entrance	92	100
Narrow door	92	100
High threshold/level in difference/step	83	84
Indoors
Stairs/threshold/differences in level between rooms/floor spaces	92	86
Insufficient manoeuvring areas where turning is necessary	92	85
Inappropriate design of door to laundry room	86	87
Use of controls/hardware in the kitchen requires hands	92	100
More than half of apparatus/controls in the kitchen in very high/inaccessible position	100	95
No place to sit in shower/bath	82	95
Use of controls/hardware in hygiene areas requires hands	100	95
More than half of apparatus/controls in hygiene areas in very high/inaccessible position	91	90
Wash basin placed at a height for use only when standing	100	100
Shower stall with curb/level difference	80	89
Bathtub instead of shower stall/space	100	90
Use of other controls and operable hardware requires hands	92	100
Storage areas can only be reached via stairs/threshold or other difference in level	91	95
Laundry room can only be reached via stairs/threshold or other difference in level	100	88

Table 3.

Accessibility Problems (n = 8) Where the Agreement was Less Than 80% in Both Rater Pairs (n = 35).

Environmental barrier causing accessibility problems	Rater Pair 1 (%)	Rater Pair 2 (%)
Exterior surroundings
Irregular walking surface	77	52
Passenger loading zones far from entrance	54	70
Entrances
Insufficient manoeuvring space at doors	25	70
Heavy doors without automatic opening	62	62
Indoors
Insufficient manoeuvring areas around white goods/storage units	75	71
Shelves too deep	00	71
Complex manoeuvers and good precision required (hygiene area)	55	71
Complex manoeuvers and good precision required (other controls and operable hardware)	50	75

When discussing reliability issues in the team of raters, during as well as after the period of data collection, one threat to agreement came out as prominent, namely, the response alternative “not relevant.” The raters interpreted this option very differently, and their interpretation seemed to change over time, resulting in ambivalence and uncertainty in the ratings. Moreover, the two real estate–employed raters stated that there were items they perceived as more difficult than others to decide on, in particular those where there were no references to concrete measurements. To eliminate uncertainty, they asked for a user manual specific for the screening tool. Preferably, they said, most of the instructions should be printed on the rating forms.

In terms of feasibility, the real estate employees stated that the screening is a complex task, requiring much concentration and accuracy. Still, they did not consider the screening difficult. They emphasized that advance training is imperative, to not only optimize agreement but also to be able to handle the practicalities of the screening situation. In hindsight they concluded that they would have needed more pretraining, with subsequent feedback from the instructors (scientists). In addition, they said that it would be necessary to use the screening tool repeatedly and continuously and they would otherwise not be able to keep up their skills. They were very explicit in their recommendation not to accomplish this kind of screening integrated with a visit for other purposes—that the assessment is too complex and demanding to be made in parallel with other duties.

During the concluding discussion session, three main situations when a real estate company could make use of a screening tool for housing accessibility were identified:

When planning new apartments and housing areas,

When planning extensive renovations of existing apartments and housing areas, and

When directing tenants with functional limitations to apartments with less accessibility problems, in particular those adapted for previous tenants, subsidized by the municipality, that is, to make more efficient long-term use of individually adapted apartments.

Furthermore, the discussions touched upon what responsibilities different categories of staff in a real estate company have to provide for accessibility. According to the two real estate employees, in the planning and drawing phases foremost, the engineers could make use of the screening tool. As to the actual data collection, their opinion was staff with duties similar to their own could very well be responsible for accessibility screenings. Still, even if they themselves were positive toward housing accessibility screening, they expressed serious doubts as to whether the idea of implementing such procedures in the company they represented was realistic.

Discussion

Although our study was limited to basic aspects of feasibility and reliability and was based on a small sample of apartments owned by a Swedish public real estate company, its results demonstrate that after additional refinement of the pilot version and the production of a manual, the Housing Enabler Screening Tool (Iwarsson & Slaug, 2010a, 2010b) could be used by nonacademic staff with a technical background. The descriptive part of the results (Table 1) can of course not be generalized, given the small sample used. The current study is a minor pilot investigation, and the frequency list of environmental barriers causing accessibility problems is only an exemplification of the usefulness of this kind of data collection. Still, the selection of dwellings represented a wide range of different types of housing of the publicly owned real estate stock in Sweden. With limited effort, reliable data on the prevalence of the types of environmental barriers causing the most accessibility problems can be collected and presented, serving as priority lists for measures to improve accessibility in the ordinary housing stock.

Currently in Sweden, the provision of accessible housing for senior citizens and persons with disabilities is part of the public debate. Since many years, there is a possibility for persons with functional limitations to apply for a housing adaptation grant in order for them to adapt a privately or publicly owned dwelling according to individual needs (Iwarsson, 2009). The public costs for such adaptations successively increase (currently at least 128 million US dollar/year), and the national building and planning legislation since many years stipulate that housing should be accessible and usable for all citizens. Therefore, in order to foster accessibility in the housing stock, municipalities and real estate companies are being recommended to perform accessibility audits. Against this background, there is a need to come up with screening tools with potential to stimulate such initiatives. With valid and reliable data on accessibility problems at hand, stakeholders have the possibility to plan for renovation projects with an integration of improvement of housing accessibility. As can be seen from the future applications suggested by the two real estate company raters involved in our study, there is an awareness among practitioners that this kind of screening could be useful. Turning to an international perspective, of course there are marked differences between countries, but there is nevertheless an increasing global awareness of the challenges implied by the aging population. However, it should be kept in mind that since accessibility is linked to standards and current legislation (Preiser & Ostroff, 2001), for valid application of instruments such as the Housing Enabler Screening Tool, national adaptation is always required (Helle et al., 2010).

Based on the small sample size, only an unsophisticated approach to data analysis could be validly adopted. In terms of interrater reliability, agreement is a basic and simple, but yet crucial aspect and methodological challenges are inherent in such analyses (Helle et al., 2010). Results reported in terms of percentage agreement alone are only based on the frequency of observation and have been criticized for not taking chance agreement into account. There are other options, for example, kappa statistics (Fayers & Machin, 2001; Shoukri, 2004). However, since one of the well-known limits of kappa is its dependence on prevalence (Byrt, Bishop, & Carlin, 1993), applying the kappa statistic based on the frequencies displayed in Table 1 would not be a valid approach in the present study. Similar problems have been identified in previous studies on Housing Enabler interrater agreement (Helle et al., 2010; Iwarsson et al., 2005).

As to the interpretation of threats to reliability, one strength of the current study is that we also targeted aspects of feasibility. Application of a new screening tool in a practice field where research-based methodology seldom is being used imply a diversity of challenges (Helle et al., 2010), not necessarily caught by analyses of interrater agreement. Studies in progress indicate that the complex contexts where this kind of assessments is administered need further consideration. For example, disturbance from the tenant and other persons present in the screening situation should be documented and reflected upon. The real estate company–employed raters underscored that they did not recommend accomplishment of the accessibility screening integrated with a routine inspection with another primary purpose. This is an important facet of the results, deserving attention when recommending the Housing Enabler Screening Tool for widespread use.

Even if the mean level of percent agreement presented is quite high, the results on interrater agreement are somewhat mixed. Looking at the item-wise level of agreement, there is a large variation with close to half of the items indicating high interrater agreement, whereas for eight items the agreement is low. Keeping in mind the short rater training provided and the lack of a handy manual for the pilot version of the Housing Enabler Screening Tool used, such results are not surprising. However, it deserves attention that among the items with the highest level of agreement, 27 out of 29 items are assessed by measurement, or are obvious by observation, and only 2 of these items call for some evaluative judgement by the rater. Out of the eight items with the lowest level of agreement (Table 3), for only half of them the rating is based on a concrete measurement. Scrutinizing the item definitions in detail, and considering the comments and questions of the raters during the data collection process, there is a need for further specification. For example, for the item “Insufficient manoeuvring space at doors” it might not have been sufficiently stressed during the training sessions that it concerned the area outside as well as inside the entrance door. Another type of challenge in the rating situation is how to rate environmental barriers that occur repeatedly, for example, “Irregular walking surface.” As to the four items that call for evaluative judgement by the rater, the results accord with our earlier experiences when instructing raters with different types of professional backgrounds (Iwarsson & Slaug, 2010a); that is, persons with no experience of how people with functional limitations interact with features of the physical environment need much training and feedback. For example, assessment and reliable rating of aspects such as “Complex manoeuvers and good precision required” for the use of controls and hardware is not an easy task if you do not have any experience of how persons with upper-extremity limitations function in daily life. To understand and act upon threats to reliability, not only the quantitative results should be focused on. That is, the qualitative information collected along and after the data collection period yielded additional information. For example, for one of the measurable items with a low level of agreement, one rater pair disagreed in all cases. Thanks to the discussion afterwards, we got the explanation for this; there had been an obvious misunderstanding of the instructions. These examples highlight the need for more rater training as well as a manual for the screening tool, as put forward also by the raters involved in the study.

The present study is only one among many of those we have pursued during a period of 20 years, all based on a process of projects targeting the development of methodology for housing accessibility assessments. This research is based on the notions of transdisciplinarity (Kötter & Balsiger, 1999); that is, it targets problems relevant to society by means of a research process of participatory nature that involves researchers from different disciplines as well as practitioners in an integrated manner. The aim of transdisciplinary research is to produce solutions perceived to be the common good, and such a goal requires development of reflexivity through recursiveness (Pohl & Hirsch Hadorn, 2007). This implies introducing recursiveness not just at the end but all through a stepwise research process. Every effort must be observed and the experiences and results used to inform the next step. In fact, several small-scale studies with similar aims and designs might be necessary; thus, preliminary results become a means of targeted learning. Reflexivity means to adjust assumptions on which the production of knowledge is based, in case “the real-world experiments” reveal that our assumptions need to be corrected. On the basis of the study results and experiences from previous studies on transfer of research methodology and results to practice contexts (see, for example, Fänge et al., 2007), we were well aware of some of the challenges, and even the potential resistance among practice representatives, which might be present. However, the representatives and staff from the real estate company involved in the present study were very open and positive and were willing to invest the time and engagement required. Still, it should be noted that implementation of research into practice is a lengthy and demanding process. As stated by the two real estate company–employed raters involved in the current study, they were far from convinced that housing accessibility screening could be easily accepted and implemented in the company they represented.

Since the completion of the current study, based on the results presented in this article the Housing Enabler Screening Tool has been revised and optimized. A new book, including the complete Housing Enabler instrument and the Screening Tool (Iwarsson & Slaug, 2010a), and a separate short screening tool manual have been published (Iwarsson & Slaug, 2010b). The Housing Enabler thus exists now in two different versions. The complete instrument is intended for detailed assessments, chiefly at individual level. The tool is intended for screening environmental barriers causing accessibility problems at group level and comprises 60 items concerning the home and its immediate surroundings. The items are divided into three sections: (A) exterior surroundings (20 items), (B) entrance (13 items), and (C) indoor environment (27 items). The results of the current study and several previous studies on the complete Housing Enabler instrument were used for optimization of the list of items, item definitions, and the layout of the screening forms, and to develop a manual intended to strengthen rater agreement (Figures 1 and 2).

Figure 1.

Exemplification of rater instructions, from the Housing Enabler Screening Tool

Figure 2.

Excerpt from the Housing Enabler Screening Tool

In conclusion, given the findings of the current study and the authors’ longstanding experiences in the use of the complete Housing Enabler instrument in research and practice, it can be concluded that the Housing Enabler Screening Tool has the potential to contribute to the development of research-based and efficient strategies for the identification of accessibility problems in ordinary housing. Applying the principles of transdisciplinary research, our long-term aim is to contribute to the solution of a complex problem of high societal relevance. The ultimate goal in this regard is to contribute to the reduction of accessibility problems in the housing stock, in order to promote health and prevent activity limitations and participation restrictions among senior citizens and persons with disabilities.

Footnotes

Acknowledgements

We thank G. Svärd, A. Svensson, and G. Flygare, Lunds kommuns fastighets AB, for fruitful collaboration making the data collection possible.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research and/or authorship of this article:

We are grateful to the Swedish Institute for Assistive Technology for funding the project underlying this study. This article was prepared within the context of the Centre for Ageing and Supportive Environments (CASE) at Lund University, financed by the Swedish Research Council on Social Science and Working Life.

Bios

Susanne Iwarsson, PhD, is registered occupational therapist, professor in gerontology at the Department of Health Sciences, Faculty of Medicine, and director of the Center of Ageing and Supportive Environments, Lund University, Sweden. Her research focuses on older persons and housing and neighbourhood issues, geriatric rehabilitation, and assistive technology, with a particular emphasis on accessibility and usability to the physical environment.

Björn Slaug, BSc, is a research engineer and PhD student in public health science at the Department of Health Sciences, Faculty of Medicine, and the Center of Ageing and Supportive Environments, Lund University, Sweden. His main research interest is development of methods for assessing accessibility problems in the physical environment.

Agneta Malmgren Fänge, PhD, is registered occupational therapist, associate professor in occupational therapy at the Department of Health Sciences, Faculty of Medicine, and the Center of Ageing and Supportive Environments, Lund University, Sweden. She is the study director of the Master of Medical Science program. Her research focuses on home and health in old age, housing adaptations, person–environment–activity transactions, and methodological issues.

References

American Association of Retired Persons. (2003). These four walls. Americans 45+ talk about home and community. Washington, DC: Author.

Byrt

Bishop

Carlin

J. B.

(1993). Bias, prevalence and kappa. Journal of Clinical Epidemiology, 46(5), 423-429.

Carlsson

Schilling

Slaug

Fänge

Ståhl

Nygren

Iwarsson

(2009). Towards a screening tool for housing accessibility problems: A reduced version of the Housing Enabler. Journal of Applied Gerontology, 28(1), 59-80.

Christiansen

C. H.

Townsend

(2004). Introduction to occupation, the art and science of living. Upper Saddle River, NJ: Pearson.

Fayers

P. M.

Machin

(2001). Quality of life, assessment analysis and interpretation. New York: John Wiley.

Fänge

Iwarsson

(2003). Accessibility and usability in housing: Construct validity and implications for research and practice. Disability and Rehabilitation, 25, 315-326.

Fänge

Risser

Iwarsson

(2007). Challenges in implementation of research methodology into community-based occupational therapy—The Housing Enabler example. Scandinavian Journal of Occupational Therapy, 14, 54-62.

Freedman

V. A.

Martin

L. G.

Schoeni

R. F.

(2002). Recent trends in disability and functioning among older adults in the United States. Journal of the American Medical Association, 288, 3137-3146.

Gitlin

L. N.

(2003). Conducting research on home environments: Lessons learned and new directions. Gerontologist, 43, 628-637.

10.

Helle

Nygren

Slaug

Brandt

Å.

Pikkarainen

Hansen

A-G.

Pétersdórttir

Iwarsson

(2010). The Nordic Housing Enabler: Inter-rater reliability in cross-Nordic occupational therapy practice. Scandinavian Journal of Occupational Therapy, 17(4), 258-266.

11.

Heyl

Wahl

H.-W.

Mollenkopf

(2005). Visions, out of home activities and emotional well-being in old age. Do macro-context make a difference. International Congress Series, 1282, 147-151.

12.

Iwarsson

(1997). Functional capacity and physical environmental demand. Exploration of factors influencing everyday activity and health in the elderly population. Doctoral dissertation. Lund, Sweden: Department of Community Health Sciences, Lund University.

13.

Iwarsson

(2005). A long-term perspective on person-environment fit and ADL Dependence among older Swedish adults. Gerontologist, 45, 327-336.

14.

Iwarsson

(2009). Housing adaptations—Current practices and challenges for the future. In Söderback

(Ed.), International handbook of occupational therapy interventions (pp. 63-69). New York: Springer.

15.

Iwarsson

Nygren

Slaug

(2005). Cross-national and multi-professional interrater reliability of the Housing Enabler. Scandinavian Journal Occupational Therapy, 12, 29-39.

16.

Iwarsson

Slaug

(2001). Housing Enabler: An instrument for assessing and analysing accessibility problems in housing. Nävlinge and Staffanstorp, Sweden: Veten & Skapen HB; Slaug Data Management.

17.

Iwarsson

Slaug

(2010a). Housing Enabler—A method for rating/screening and analysing accessibility problems in housing. Manual for the complete instrument and screening tool (2nd rev. ed.). Lund and Staffanstorp, Sweden: Veten & Skapen HB; Slaug Enabling Development.

18.

Iwarsson

Slaug

(2010b). The Housing Enabler Screening Tool—Short manual. Lund and Staffanstorp, Sweden: Veten & Skapen HB; Slaug Enabling Development.

19.

Iwarsson

Ståhl

(2003). Accessibility, usability, and universal design—Positioning and definition of concepts describing person-environment relationships. Disability and Rehabilitation, 25, 57-66.

20.

Kötter

Balsiger

P. W.

(1999). Interdisciplinarity and transdisciplinarity. A constant challenge to the sciences. Issues in Integrative Studies, 17, 87-120.

21.

Lawton

M. P.

Nahemow

(1973). Ecology and the aging process. In Eisdorfer

Lawton

M. P.

(Eds.), The psychology of adult development and aging (pp. 619-674). Washington, DC: American Psychological Association.

22.

Oswald

Wahl

H.-W.

Schilling

Nygren

Fänge

Sixsmith

Iwarsson

(2007). Relationships between housing and healthy ageing aspects in very old age: Results from the European ENABLE-AGE Project. Gerontologist, 47, 96-107.

23.

Pohl

Hirsch Hadorn

(2007). Principles for designing transdisciplinary research. Munich, Germany: Oekom Verlag.

24.

Preiser

W. F. E.

Ostroff

(2001). Universal design handbook. New York: McGraw-Hill.

25.

Scheidt

R. J.

Norris-Baker

(2004). The general ecological model revisited: Evolution, current status, and continuing challenges. In Wahl

H.-W.

Scheidt

R. J.

Windley

P. G.

(Eds.), Annual review of gerontology and geriatrics, 2003. Vol. 23: Aging in context: Socio-physical environments (pp. 34-58). New York: Springer.

26.

Shoukri

M. M.

(2004). Measures of inter-observer agreement. Washington, DC: Chapman & Hall/CRC.

27.

Spillman

B. C.

(2004). Changes in elderly disability rates and the implications for health care utilization and cost. Milbank Quarterly, 82, 157-194.

28.

Stark

(2004). Removing environmental barriers in the homes of older adults with disabilities improves occupational performance. Occupational Therapy Journal of Research: Occupation, Participation and Health, 24, 32-39.

29.

Wahl

H.-W.

Oswald

Fänge

Gitlin

Iwarsson

(2009). The home environment and disability-related outcomes in aging individuals: What is the empirical evidence? Gerontologist, 49, 355-367.

30.

World Health Organization. (2002). Active ageing: a policy framework. Retrieved from http://www.who.int/ageing/publications/en/st common accessibility problems in the housing