The Role of Architecture and Design in the Management of Parkinson’s Disease: A Systematic Review

Literature search

We searched MEDLINE and Embase from inception to May 2020, using the electronic search strategy presented in the Supplementary Table 1.

Study selection

We included observational studies (case-control, cross-sectional, and prospective cohort) and experimental studies that evaluated architectural or design features of the built environment and its impact on PD outcomes across any setting (community, rehabilitation, acute care, and long-term care).

Studies were excluded when no specific details of architectural and design features of the environment were provided and when the study population was other than people diagnosed with idiopathic PD. Additionally, narrative reviews, study protocols, abstracts, and conference proceedings were excluded. Only English-language studies were included.

Two reviewers (JBR and GD) independently screened the titles and abstracts identified from searches. Any paper identified as potentially relevant by at least one review author was retrieved. Two reviewers (JBR and GD) independently screened full-text articles, with discrepancies resolved by discussion. The references of relevant studies were cross-checked for additional studies not identified by the electronic search.

Data extraction

We extracted information including the study characteristics (publication year, study period, study design), participant characteristics (number, age, sex, disease duration, PD-related impairment, and relevant treatment information), in addition to other disease specific-characteristics of interest.

Fall-related outcomes

OPEN IN VIEWER

Fig.2

Architectural plan with a summary of the most significant architectural/design features of the built environment. Homestyler online software was used. PD, Parkinson’s disease.

Pavement characteristics were the most commonly assessed architectural features of the built environment (n = 11), followed by confined spaces (n = 3), and housing type (n = 2).

Seven articles reported falling [30–36], four studies reported fear of falling [37–40], and one study reported postural instability [13] as outcomes. Among these, nine studies were observational, namely cross-sectional (n = 6) [30, 37–40] and prospective cohort (n = 3) [31, 36] designs, and two studies used an experimental design (n = 2) [13, 33].

To survey the circumstances surrounding falls, Stack et al. [30, 34] developed two cross-sectional design studies where questionnaires were used, and found trip hazards to be the main driver of sudden falls or near falling, namely uneven pavements, curbs, steps, skirting boards, doorways, and carpets. In addition, steps and doorways were reported to be the primary locations of such events within the home. Similarly, in a prospective-cohort study where fall diaries were used to assess fall related characteristics, Gray et al. [31] found carpets to be the most common location of falls, referring further confined spaces to be also important fall locations. However, neither the use of stairs nor the housing type was found to increase the risk of falling.

Additionally, research has been conducted on where falls occur (outdoors versus indoors). In a cross-sectional study where interviews were conducted, Gazibara et al. [35] found a small though statistically significantly higher frequency of outdoor falls compared with indoor falls among persons with PD, although a prospective cohort study conducted by the same authors found indoor falls to be more common than outdoor falls [36]. In both studies, outdoor falls were mostly preceded by extrinsic factors, notably slipping and tripping while walking over a curb or an object on the ground, such as a carpet. Moreover, in a prospective cohort study developed by Ashburn et al. [32] where fall diaries were recorded, 80% of falls were found to happen at home, mostly in bedrooms, living rooms, and kitchens, with tripping during walking being the biggest single cause of falls. Additionally, more falls happened outdoors in gardens than in the bathroom, hallway, or on the stairs, a finding that was attributed to challenges posed by uneven surfaces, unanticipated trip hazards, and difficult maneuvers. Furthermore, Cole et al. [33], in an experimental study where gait was assessed while walking on two different surfaces (firm and foam surfaces), also suggested an increased risk of tripping and falling on less stable surfaces, particularly on compliant or uneven surfaces, where the height of the walking surface is not uniform.

Four studies reported findings regarding fear of falling, all using a cross-sectional design. For instance, through semi-structured interviews Jones et al. [37] found that turning in confined spaces, steps, and stairs universally increased the fear of falling, whether outside or around the home. In addition, by collecting data by a postal survey, Nilsson et al. [39] found walking difficulties to be the strongest factor contributing to fear of falling, among which climbing stairs seemed to be of particular importance. However, one study by Jonasson et al. [40] suggested that housing type and residential area have no significant impact on people’s fear of falling, despite suggesting that these factors remain a concern. Finally, Rahman et al. [38] recognized, through the use of several scales, that some activities are actively avoided by PD patients due to fear of falling, such as going outside in slippery conditions, going to crowded places, going up and down stairs, taking a shower, and either reaching for something above head level or bending down.

Galna et al. [13] was the only study to report findings related to postural instability. This experimental study performed a center of mass analysis during gait, suggesting that environmental hazards, such as ground-based obstacles, may accentuate postural instability and the subjective impression of near falls in people with PD.

Gait impairment

Twenty-one articles reported gait as an outcome [16, 41–47], mostly concerning FOG. Among these, five studies were observational, namely case-control (n = 1) [41], cross-sectional (n = 3) [16, 42], and prospective cohort (n = 1) [55] designs, and sixteen studies used an experimental design (n = 16) [33, 44–51].

Lee et al. [42] conducted a cross-sectional study using questionnaires, and found that 70% of participants had problems moving through confined or narrow spaces (e.g., doorways) within the home, predominantly due to bumping and FOG (78% overall). Rahman et al. [16] also conducted a cross-sectional study and, through the use of questionnaires, recognized other factors that potentially induce FOG, namely bright lights and crossing roads. On the other hand, climbing stairs was found to improve walking and overcome FOG. In addition, Jones et al. [37] used semi-structured interviews to examine perceived walking challenges. Busy outdoor environments were reported to pose unpredictable challenges, while in indoor environments doorways were specifically a major trigger for FOG. Furthermore, furniture was found to challenge gait, especially for people with higher Unified Parkinson’s Disease Rating Scale and FOG Questionnaire scores. Both studies identified attentional and external cueing strategies that helped improve gait, such as following lines on the floor, walking over tile edges or paving stones, looking at pavement cracks, and using a strip across a door threshold [16, 37].

Additional research was undertaken regarding confined or narrow spaces. For instance, Giladi et al. [41] assessed 990 PD patients in a case-control study and found narrow spaces (e.g., doorways) to be the cause of motor blocks in 25% of cases. Later, Nieuwboer et al. published two articles [43, 45] using an experimental design where gait was assessed under three conditions (normal, stop, or block/freezing conditions), and proposed that in people with PD the confrontation with visual stimuli suggesting a limited space (e.g., narrow walkway) or a change of direction (e.g., obstacle course) can potentially induce FOG. Additionally, Schaafsma et al. [44] also conducted an experimental study in a gait laboratory, finding FOG to be elicited by walking in narrow spaces, particularly during “off” states.

Later, Almeida et al. [46] and Lebold et al. [48] developed two experimental studies aiming to assess gait under three doorway conditions, namely narrow, normal, and wide. They suggested that narrow doorways may result in gait differences between PD individuals and healthy controls, finding that the narrow doorway condition was significantly associated with gait changes and FOG experienced among the participants. In addition, Cowie et al. [47, 49] developed two experimental studies where gait kinematics were measured under four doorway conditions: no door, narrow door, medium door, and wide door widths. They identified abnormal walking responses to doorways in PD patients who regularly experience FOG, with disturbances becoming more pronounced as door width decreases (p < 0.001). Furthermore, Kataoka et al. [50, 54] assessed gait patterns in a suddenly narrowed path and confined spaces, and showed that the moments preceding these were characterized by gait variability in patients with Hoehn-Yahr stage III PD. More recently, another study developed by Kataoka et al. [55] prospectively studied 26 patients with Hoehn-Yahr stage III PD for six years, aiming to evaluate the disease progression and to analyse gait parameters under two circumstances: a suddenly narrowed path and a straightly narrowed path. They found an increase in the number of steps associated with FOG when walking on narrowed paths, particularly on a path that narrows suddenly.

Additionally, Ehgoetz et al. [56] developed an experimental study aiming to evaluate gait in the dark under five doorway conditions, namely walking through a doorway to a confined space in complete darkness, with door frame illuminated, or with both the door and limbs illuminated, and walking away from the doorway into an open space in complete darkness or with limbs illuminated. FOG occurrences were found to be four times more often when walking toward a standard doorway to a confined space, than walking away from the doorway into a large open space. In addition, illuminating the door frame was found to reduce FOG by 38% compared to walking through the door in complete darkness. Additional experimental research was developed by Almeida et al. [57], who assessed gait in complete darkness in comparison to a regularly lighted room, suggesting that individuals with PD adapt walking in the dark presumably to optimize safety.

Stegemöller et al. [51] evaluated gait kinematics in an experimental study and under two circumstances (normal walking and obstacle crossing), and found that decreased walking ability was present when an obstacle needed to be overcome. In addition, Cole et al. [33] and Oates et al. [52] conducted experimental studies to assess gait while walking on different surfaces, such as firm, soft, and slippery surfaces. They found that people who fell adapt differently to walking on unstable surfaces than age-matched controls, and that PD patients may alter their gait kinetic when walking on a slippery surface. Recently, an experimental study developed by G a ´ l et al. [58] assessed gait under six floor patterns, with the reference pattern being a virtual large transverse chessboard, and the other patterns differing either in size (small floor stones), orientation (diagonal), nature (real paving), regularity (irregular), or no pattern. They observed no direct effect on FOG in overall and no differences in gait parameters between large and small floor stones. Additionally, they found improvement in gait parameters when walking on large, virtual, regular, transverse chessboard floor stones, compared with walking on no pattern or on real, irregular or diagonal floor stones.

Finally, Mak et al. [53] reported gait performance parameters under three different conditions (walking at a natural pace, and either walking while doing a cognitive task with or without the addition of traffic lights). In this study, PD patients when exposed to traffic lights experienced benefits in terms of stride length, cadence, and gait velocity (p < 0.001).

Functional mobility

Six studies reported functional mobility outcomes [42, 59–63], mostly regarding housing accessibility and housing adaptations. All studies were cross-sectional.

Pretzer-Aboff et al. [59] explored the facilitators and barriers encountered by PD patients and caregivers to optimize participation in functional activities and exercise. The home environment, particularly small places, clutter, and stairs, was reported to challenge the functional mobility of people with PD. Additionally, environmental interventions, such as the use of safety bars, shower benches, lift chairs, and raised beds and seats, were described as fostering independence, sense of security and safety, and improving mobility among the participants.

Further, a study by Lamont et al. [60] sought to understand the facilitators and barriers to walking in the community perceived by people with PD and their partners through the use of focus groups. Environmental factors were identified as the main barriers to community walking, including crowded environments, pavement characteristics, bad weather, and reduced or fluctuating lighting. Only one aspect of the physical environment, i.e., signaled pedestrian crossings, was described as a facilitator to community walking.

Nilsson et al. [61] and Slaug et al. [62], using structured interviews, housing observation and self-ratings, and a version of the Housing Enabler Instrument, found that elderly people with self-reported PD have more accessibility problems with their housing and experience less usability of their home than the elderly in general. In addition, elderly people with self-reported PD were also found to be less independent in activities of daily living, to have more functional limitations and to be more dependent on walking aids [61]. Although the number of environmental barriers did not differ between the studied populations, the top 10 environmental barriers that generated the most accessibility problems in PD individuals were identified, and include, among others: wall-mounted cupboards and shelves placed extremely high in the kitchen; few or no seating places in the outside spaces; lack of grab bars in the shower, bath, or toilet; high curbs outdoors; uneven surfaces outdoors; and bathtubs [62]. Moreover, misjudging the space between objects and problems reaching for objects were also problems reported by Lee et al. [42] Finally, Haak et al. [63] used structured interviews to identify the hygiene area as the most common location where PD individuals used housing adaptations.

Limitations

Our study has several limitations. For instance, by restricting our inclusion criteria to English-language studies, we may have missed relevant data from other sources. Additionally, data accuracy may have been limited because of recall bias, as well as omissions of outcomes and sample characteristics. For example, many results were based on PD patients’ subjective perceptions of events and risk situations, thus being more prone to recall bias. Wearable technologies that are able to continuously monitor PD patients may be used in future studies to reduce such subjective elements.

Another potential limitation is that some patient characteristics, such as age, PD-specific disability, and outcome definitions across studies, were variable, meaning that caution must be exercised when pooling the findings. We suggest the creation of a standardized questionnaire, ideally elaborated by a multidisciplinary team, including physicians, nurses, psychologists, architects, and designers, to assess the impact of architecture and design on patients mobility and quality of life.

In conclusion, we found evidence that some elements of the built environment may have a role in fall-related outcomes, gait disturbances, and functional mobility in people with PD. However, the available evidence regarding the impact of person- and context-specific interventions remains sparse. A summary of potential architectural/design adaptations in the management of PD is presented in Table 3.