Wearables for Pediatric Rehabilitation: How to Optimally Design and Use Products to Meet the Needs of Users

Abstract

This article will define “wearables” as objects that interface and move with users, spanning clothing through smart devices. A novel design approach merging information from across disciplines and considering users’ broad needs will be presented as the optimal approach for designing wearables that maximize usage. Three categories of wearables applicable to rehabilitation and habilitation will be explored: (1) inclusive clothing (eg, altered fit, fasteners); (2) supportive wearables (eg, orthotics, exoskeletons); and (3) smart wearables (eg, with sensors for tracking activity or controlling external devices). For each category, we will provide examples of existing and emerging wearables and potential applications for assessment and intervention with a focus on pediatric populations. We will discuss how these wearables might change task requirements and assist users for immediate effects and how they might be used with intervention activities to change users’ abilities across time. It is important for rehabilitation clinicians and researchers to be engaged with the design and use of wearables so they can advocate and create better wearables for their clients and determine how to most effectively use wearables to enhance their assessment, intervention, and research practices.

A variety of wearables are available for rehabilitation and habilitation (here referred to collectively as rehabilitation). Some of these, such as clothing and orthotics, have existed in a variety of forms for centuries, whereas others, such as exoskeletons and smart wearables with sensors, have evolved more recently. We define “wearables” as objects that interface and move with users, spanning clothing through smart devices.

This article proposes that wearables, just like treadmills and other equipment, can be effective rehabilitation tools if they are used in ways to impact users’ abilities, task demands, or the environment to promote function and participation. With a recent increase in variety of wearables available, it is critical that rehabilitation professionals: (1) are aware of the existing wearables; (2) understand how they can positively impact the design process for wearables; and (3) critically evaluate how wearables can be used to improve the lives of users. This perspective article aims to provide readers with the information and framework to meet these 3 goals.

Traditionally, rehabilitation tools have been designed with a primary focus on function. Because wearables are donned and transported with users, they must meet a broad range of user needs in addition to function. A novel design approach merging interdisciplinary perspectives with users’ needs is presented here as the optimal approach for designing wearables for rehabilitation.¹ Then, 3 categories of rehabilitation wearables are explored: (1) inclusive clothing (eg, altered fit, fasteners), (2) supportive wearables (eg, orthotics, exoskeletons), and (3) “smart” wearables (eg, with sensors). For each category, we provide examples of existing and emerging wearables and describe potential applications for assessment, intervention, and research. We discuss how these wearables can change function and participation at the level of the individual, task, or environment.

FEA2 Design Model

Primary advantages of wearables for rehabilitation are that they can assist users in a variety of environments during daily activities, and they can offer the high doses of practice often required for meaningful rehabilitation outcomes.^2,3 The traditional approach to the design of products for rehabilitation has focused on addressing impairments or limitations in users’ functional abilities.⁴ However, other key needs must be addressed for a wearable device to be accepted by a user.^2,3 Devices created using the traditional, function-focused model can be uncomfortable to wear physically, psychologically, and socially, because they often focus attention on users’ impairments.⁵ In addition, devices might provide the desired assistance for function but can be cost prohibitive.⁶ It is critical that wearable rehabilitation designs be overtly guided by a model emphasizing the broad needs of users. In the case of pediatric and other patient populations, the role of and reliance on a caregiver should also be considered when assessing user needs. The caregiver can both play the role of informant for the assessment of user needs as well as ultimately mediate device usage.^7–9 Therefore, it is important to address caregivers’ needs and caregivers’ perceptions of user needs throughout the design process. As advocates for clients, and as liaisons between clients and device manufacturers, rehabilitation professionals have significant potential to transform how rehabilitation wearables are designed and distributed.

The FEA2 (Function, Expressiveness, Aesthetics, Accessibility) model was recently developed in Dr Lobo's Super Suits Program at the Move To Learn Innovation Lab, University of Delaware, to fill the need for a model to guide the design of rehabilitation products. The model is an extension of the existing FEA apparel design user needs model integrated with the design approach of engineering and the philosophy of patient-centered health care.^10,11 The FEA2 model incorporates user needs related to function, expressiveness, aesthetics, and accessibility (Fig. 1).^1,10

Figure 1.

The FEA2 model was created to guide the design of rehabilitation products so they address the broad needs of users in relation to function, expressiveness, aesthetics, and accessibility.

Functional needs for a product pertain to its fundamental use. Does the device facilitate and/or improve movement, function, or participation? Does it fit the user appropriately? Is it safe to wear or use? When designing a device, one must consider how the device will be used, how that use determines device demands and constraints, and ultimately the metrics the device must meet in order to be successful.^11–13

Expressive needs relate to the socially communicative or symbolic aspects of wearables.¹⁰ Wearables communicate varied social messages about the wearer, such as social group membership, personal identity, and sense of self.¹⁴ The expressive quality of a device can vary across users yet is important for conveying personal expression and self-concept.¹⁵

Aesthetic needs for products relate to the users’ interest in the product as an object of beauty.¹⁰ For example, does the object have a pleasing shape, color, or pattern? Different users will have different aesthetic criteria and these can change over time or among cultures.¹⁴ Aesthetics are a vehicle by which a wearer can mediate the social environment.^14,16 For example, the appearance of a wearable can be aesthetically engaging to others, creating opportunities for social interaction and acceptance.

Accessibility needs for products deal with availability, affordability, and inclusivity.¹ Relatively limited choices of rehabilitation devices are available in the traditional retail environment¹⁷ or through medical manufacturing companies.^1,18 Moreover, these products are often cost prohibitive in a manner that limits users to those with sufficient financial means.^19,20 A recent shift toward more inclusive design practices that focus on serving the maximum number of users, regardless of age or ability level,²¹ has the potential to allow users with disabilities to access wearables via the general retail market rather than through niche medical venues. Retail marketing of inclusive designs has the potential to significantly impact availability and affordability of wearables for people with disabilities. The do-it-yourself and maker movements are also impacting the way devices are designed and obtained as well as empowering end users and members of the broader community.²²

The FEA2 multidisciplinary framework presents a model for developing innovative and impactful rehabilitation devices for use in the real world. For example, the FEA2 model has been used to design the Playskin Lift, the first exoskeletal garment for pediatric users with upper extremity impairment.²³ Throughout the design process Dr Lobo's team involved children who had discontinued use of a hard exoskeleton.²³ The final solution improved functional ability for children while also being aesthetically pleasing, personally expressive, easy to use, and accessible via a do-it-yourself manual.^1,23

Rehabilitation professionals should be aware of the FEA2 model for use in their own designs and so they can guide manufacturers to create products that address clients’ broad needs. Improved design can increase adherence with prescribed device use²⁴ leading to potentially better client outcomes and overall satisfaction.^25,26

Inclusive Clothing

Clothing itself is a wearable rehabilitation tool. In the early 20th century, the US medical community began to identify clothing as a rehabilitation tool, using tests of donning and doffing of clothing to assess physical and cognitive improvement of clients.²⁷ At present, it can be challenging to find clothing that meets desired criteria for style, fit, price point, and self-expression, especially for users with disabilities.²⁷ The mainstream apparel industry historically neglects this market, due to misconceptions that the market is small (and therefore not profitable), is not interested in style or aesthetics,^21,28 and requires a drastically different production process.²⁷ Clothing designed for people with physical disabilities has traditionally fallen into the apparel industry category of “functional clothing.”¹⁰ User satisfaction with function-focused designs has been mixed.^29,30

The newly emerging inclusive clothing design perspective can serve the needs of people with disabilities better than the traditional functional or adaptive clothing perspectives. Inclusive clothing is designed with a broad range of users in mind, with no population excluded, making it more universally appealing and marketable.²¹ Inclusively designed clothing has the potential not only to allow for assessment of clients’ changing abilities, but also to impact rehabilitation outcomes at the level of the individual, task, or environment (Fig. 2).

Figure 2.

Examples of ways in which wearables can be used to improve the lives of users by impacting the individual, the task, or the environment.

Inclusive clothing can be designed such that the garment facilitates donning and doffing. Individuals with disabilities, medical conditions, or age-related impairments comprise a heterogeneous group, yet have common dressing challenges that can be addressed through inclusive design.^17,31,32 Most research in this area is either dated^33–35 or strictly theoretical,^17,36 but proposed solutions include altering the fit to accommodate comfortable positioning in wheelchairs,¹⁷ altering the location and type of fasteners,^13,14 and constructing clothing using comfortable, nonrestrictive material.¹⁷

Inclusive clothing is becoming more readily available in the retail marketplace. Two examples are the Adaptive Clothing Collection by Tommy Hilfiger in collaboration with the Runway of Dreams organization,³⁷ and the Cat & Jack line available at Target.³⁸ The Tommy Hilfiger collection features adjustable sleeves and pant lengths for improved fit, and magnets that replace challenging buttons and zippers.³⁷ The Target collection is specifically designed for children with sensory sensitivity.³⁸ The line features tagless garments, soft fabric, and side openings for donning and doffing.³⁸ These collections represent 2 industry leaders bringing positive change to the historically overlooked special needs market, by making available inclusively designed clothing that is functionally appropriate for the user, while also being aesthetic, expressive, and accessible.

Finally, inclusive clothing can be designed to facilitate social participation, thereby changing the social environment of the user. Like any wearable, clothing has a psychosocial component.^14,39 Choice of clothing presents the wearer's sense of self in the social environment, thus impacting interaction with others. Clothing can communicate one's culture, socioeconomic status, group identification, gender identity, and other key perceiver variables.¹⁴ Individuals with disabilities can feel socially stigmatized due to distorted body image and/or negative self-image.¹⁴ For this reason, clothing can be used to mediate potential social stigmatization by concealing physical differences, reducing perceived impairments, or enhancing self-expression¹⁵ thereby increasing self-confidence and security.¹⁴ Therefore, rehabilitation professionals can consider clothing as a tool to assess change across time, to increase independence with activities of daily living, or to positively impact user confidence, self-expression, and social interaction.

Supportive Wearables

Rehabilitation professionals often use supportive wearables that aim to provide movement assistance to users. Examples include passive and active orthoses or exoskeletons. Passive orthoses or exoskeletons provide a fixed amount of movement support at any given moment. Active orthoses or exoskeletons can be controlled to provide variable levels of movement support.⁴⁰ Although the distinction between orthoses and exoskeletons is blurred, we use the term “orthosis” here to refer to wearable devices with support as their main function, and the term “exoskeleton” for devices that primarily aim to augment movement.

Passive orthoses generally provide support for a specific region of the body. One common orthosis for the lower extremities is the ankle-foot orthosis. The goal of an ankle-foot orthosis is to stabilize the ankle in dorsiflexion. This can allow users to achieve improved gait velocity or prevent falls, facilitating greater participation and safety in daily activity. Although ankle-foot orthoses usually improve function in clients with poor volitional dorsiflexion,⁴¹ they are made of rigid thermoplastic and might not address user needs related to comfort, expressiveness, and aesthetics. A systematic review showed that ankle-foot orthosis adherence is often low, in some cases just 20%, due to complaints related to pain, discomfort, and appearance of the wearable.^42,43 To improve patient adherence, common orthoses should be redesigned using the FEA2 model that accounts for users’ broad needs. Interestingly, some companies have already begun to alter their designs to incorporate softer, comfortable, more aesthetically pleasing materials. For example, knee braces have traditionally used metal components on the medial and lateral aspects for stability. In recent years, companies like DonJoy, known for high-stability knee braces, have been offering softer fabric- and silicone-based brace options (https://www.donjoyperformance.com/knee).

In contrast to traditional plastic orthoses, dynamic elastomeric fabric orthoses, such as Theratogs (Theratogs Inc, Telluride, CO, USA) and Dynamic Movement Orthoses (Boston Orthotics and Prosthetics, Avon, MA, USA),⁴⁴ create compression forces on a specific part of the body for added stability. These orthoses can be fitted to most body segments, but are primarily used for postural support and stability around the torso and hip. Because these garments are fabric based, they can be more comfortable than traditional rigid orthoses, and can result in higher usage. Preliminary evidence demonstrates some potential of these wearables to improve comfort, posture, or gait.^44–47

Passive exoskeletons augment how a user is able to move. The Wilmington Robotic Exoskeleton from Nemours AI DuPont Hospital (Wilmington, DE, USA) is an example that uses 3D-printed components, metal components, and elastic bands to assist with elbow and shoulder flexion (Fig. 3A).⁴⁸ The device can improve arm function for children with muscle weakness due to muscular dystrophy, arthrogryposis multiplex congenita, or spinal muscular atrophy.⁴⁸ However, it has limitations associated with bulkiness, hardness, aesthetics, and affordability that can constrain families’ access to the device and limit use once obtained. The Playskin Lift developed by Dr Lobo's Super Suits Program is a passive upper extremity exoskeletal garment that incorporates supportive bundles of flexible metal wires running through vinyl tunnels underneath each arm (Fig. 3B).²³ It can improve arm function in children with arm weakness within a session, and pilot data suggest it can also improve unassisted, independent function across time.²³ Because the garment looks and feels like a typical shirt and is easy to don, doff, and clean, users report high rates of adherence. Low-profile soft exoskeletons like this have a variety of benefits including improved comfort, aesthetics, and affordability. However, current designs have key limitations. For instance, the Playskin Lift cannot support the arms of children over 3 years of age. Passive exoskeletons are also limited in not incorporating mechanisms allowing users to regulate the amount of support they receive.

Figure 3.

Examples of 2 passive exoskeletons currently available to assist upper extremity function for children with arm movement impairments. A, The Wilmington Robotic Exoskeleton (WREX) (Wilmington, DE, USA) uses plastic or metal components for support and elastic bands to lift the arms. It is available through Nemours A.I. DuPont Hospital for Children and JAECO Orthopedic Specialties (Hot Springs, AR, USA). B, The Playskin Lift is a garment with vinyl tunnels under each arm to house bundles of elastic wires to lift the arms. It is available via an open-access DIY manual (http://sites.udel.edu/move2learn/how-todiy/).

Active/powered orthoses and exoskeletons incorporate mechanisms to allow users to customize and regulate the level of support. Robotic exoskeletons have been created to enhance human performance, as assistive devices for individuals with disabilities, and as therapeutic devices for rehabilitation.⁴⁹ The Defense Advanced Research Projects Agency has funded a variety of exoskeleton projects over the past 2 decades⁵⁰ to enhance human performance. Many of the exoskeletons produced aim to allow users to carry a heavier load with greater ease. Unfortunately, problems such as bulkiness, power demands, and human energy exertion from altered kinematics have prevented these exoskeletons from becoming mainstream.⁴⁹ The military is now shifting from hard to soft exoskeletons, like the exosuit from Harvard University, which consists of 2 motors that pull cables to assist in hip extension and ankle plantar flexion.⁵¹ Using this device, participants’ net metabolic use decreased by 7.3%. Reductions in metabolic cost of this magnitude could have a significant impact on endurance, fatigue, and agility when walking over large distances.

Powered exoskeletons, like the Ekso (Ekso Bionics Holdings, Inc, Richmond, CA, USA) and ReWalk (ReWalk Robotics, Marlborough, MA, USA), are also currently being tested for individuals with disabilities. In 2014 ReWalk was the first exoskeleton to receive US Food and Drug Administration approval for people with spinal cord injury (http://rewalk.com/rewalk-robotics-announces-expansion-to-india-with-saimed-innovations-2-2-2-2-3-2-3/).⁵² This exoskeleton can allow people with paraplegia to walk while standing upright because it supports their body weight and augments their leg motion through motors at the hip and knee. Although they are becoming increasingly lower in profile, powered exoskeletons remain expensive and as a result are typically clinic-based. LiteRun Inc. (St Paul, MN, USA) has begun to shift the design of rehabilitation exoskeletons from hard to soft, mirroring the shift observed in military exoskeleton design, creating a lower extremity pneumatic exoskeleton that uses soft air bladders. Soft exoskeletons like this have the potential to more broadly address needs related to aesthetics, comfort, and ease of use. They also can be more affordable, increasing the likelihood for use in natural environments.

Supportive wearables can potentially serve as assistive devices, improving function by providing support and/or movement assistance for performance of various tasks. If used in conjunction with intervention activities, supportive wearables can also be used as rehabilitative devices. By grading the level of assistance provided during intervention across time, exoskeletons can potentially facilitate permanent improvements in independent performance and function beyond the wear time.⁵³

Smart Wearables

Smart wearables, or wearables that incorporate sensors, afford the opportunity to collect continuous data across time, tasks, and environments.⁵⁴ This contrasts with standard clinical tools that provide a snapshot of behavior at a specific time in a particular environment, often failing to capture true performance or variability in performance. Smart wearables are becoming more common, and a variety are available both for the research market and general population. These include wearables that track body temperature, heart rate, step number, foot pressure, position, and acceleration.⁵⁵ The data measured depend on the sensor type(s) integrated in the wearable. Accelerometers have been commonly used in smart wearables in recent years. They provide information about the acceleration of the person or body part to help us understand physical activity and can be especially useful for rehabilitation professionals.

Both commercial and research wearables incorporating accelerometers exist. Commercial devices include the Fitbit (Fitbit Inc., San Francisco, CA, USA) and other fitness trackers sold for personal use. Research devices include accelerometer-only devices, such as Actical (Philips Respironics, Inc, Murrysville, PA, USA) and Actigraph (ActiGraph, LLC, Pensacola, FL, USA), designed to measure the intensity of physical activity.⁵⁶ Other research devices, for example the Stepwatch (Modus Health, LLC, Washington, DC, USA), have embedded software to calculate the number of steps taken.⁵⁷ Finally, some research wearables measure acceleration as well as other signals, such as angular velocity, electromyography, cardiac activity, respiration, and/or galvanic skin response (eg, MC10, MC10 Inc, Lexington, MA, USA; and Shimmer, Shimmer Sensing, Dublin, Ireland).

A key consideration for rehabilitation professionals using smart garments/sensors is the validity of those wearables for measuring the variable of interest. One challenge relates to how the products process the sensor data. Activity monitors like Actigraph and Actical use proprietary software to determine activity counts and classify physical activity intensity. In effect, the type of movement required to produce an activity count is unknown to users. Different types of data output can have varying levels of validity. For instance, Actigraph data in 31 ambulatory children aged 12 to 30 months showed a positive correlation between number of Actigraph activity counts and observed activity level measured by the Observational System for Recording Physical Activity in Children—Preschool, but poor to fair relations between classification of activity levels based on activity counts.⁵⁸ It is important to understand the process and accuracy of software classification systems associated with wearables, especially if they are going to be used to prescribe or evaluate the effectiveness of interventions.

Another challenge to the validity of wearables for measuring variables of interest is that a device designed and calibrated to measure physical activity in a healthy adult might not be accurate for an adult with a disability or for pediatric populations. Accelerometer thresholds (cutpoints) to classify intensity of physical activity validated by visual observation have been established for 5- to 6-year-olds but do not yet exist for other ages to the best of our knowledge.⁵⁹ Actigraph activity counts have been measured for nonambulatory infants, but no validation of activity counts to a gold standard measure has been reported.^60–62 Step-counting devices such as FitBit and Stepwatch have been designed and calibrated to measure steps in ambulatory persons. Whether they are able to capture kicking or stepping movements in nonambulatory infants is unknown. Stepwatch has been validated for use in ambulatory youth with cerebral palsy and Gross Motor Function Classification System levels I to III between 10 and 13 years of age.⁶³ Regarding nonambulatory populations, a recent study showed that 3 commercially available wrist-worn activity trackers generally performed poorly for accuracy/precision in counting strokes per minute when users were propelling a manual wheelchair.⁶⁴ The Apple Watch (Apple Inc, Cupertino, CA, USA) is the only mainstream step-counting product that has a mode to measure activity while rolling.⁶⁵ Other studies have also shown that participants did not find current activity tracker data relevant to their specific activities.^66,67

Although using data from accelerometers to classify stationary versus movement periods or to measure number of steps can be fairly straightforward, identifying or classifying other types of movements is not as clear-cut.⁶⁸ For many sensor-derived metrics, variability is high, both within and across participants (eg, arm movement characteristics in infants⁶⁹ and adults⁷⁰). Alternatively, sensor data can look similar across a variety of activities, making it difficult to classify activity performance. Therefore, interpretation and classification of data can be complicated (Fig. 4).

Figure 4.

One full day (∼11 hours) of acceleration data from a 7-month-old with wearable sensors on both ankles. Each sensor contains triaxial accelerometers collecting actively synchronized data at 20 samples per second. Left and right leg data are overlaid. The sensors were put on around the 1.5-hour mark and stayed on until around the 12.5-hour mark. The infant made a variety of movements across different activities during the data capture window, but this is not readily apparent by viewing the acceleration time series alone; advanced analyses are needed.

As these examples demonstrate, research to determine the validity of the different types of data output by wearables is critical. Although algorithms for processing data from wearables might need to be created and validated for users with specific disabilities, it might not be possible to validate all types of movements produced by users with each unique type or category of disability. As smart wearables become more integrated within practice, rehabilitation professionals should drive the design and testing not only of the wearables but also of the associated data processing approach and software-user interface to ensure that products meet the key needs of clients and rehabilitation professionals as end users.

Once the validity of smart wearables has been established for the intended use, these wearables can provide useful information about activity in the natural environment. For instance, Dr Smith's Infant Neuromotor Control Laboratory at the University of Southern California has used wearable sensors to capture the variability and repertoire of limb movements that infants produce across full days. Wearable sensors containing triaxial accelerometers and gyroscopes (Opal; APDM Inc, Portland, OR, USA) were used to create and validate algorithms to count the thousands of leg movements infants make in a day,⁷¹ as well as the duration, peak acceleration, average acceleration, and type (unilateral or bilateral) of each movement produced.⁷² Dr Smith's group has taken a similar approach to create and validate an algorithm to identify bouts of arm movements.⁶⁹ They have measured infants with typical development (1–12 months of age) and infants/children at risk for developmental delay (1–15 months postterm age). Although these are modest samples (12–20 per group) they suggest the ability to use data from wearables to discriminate between typical and at-risk groups as well as to predict neurodevelopmental outcomes. Preliminary outcomes demonstrate that leg movements in infants at risk show smaller accelerations,⁷³ more repeatability (less variability), and computationally less information.^74,75 Further, infants at risk for delays with poor neurodevelopmental outcomes at 24 months of age demonstrated smaller acceleration values in their leg movements as infants than did infants at risk for delays with good neurodevelopmental outcomes at 24 months of age.⁷³ These types of data from smart wearables could facilitate early identification of atypical neuromotor development and assessment of change across time in response to intervention.

Collecting repeated measures of activity from large, culturally diverse samples of infants and young children over extended periods of daily life could improve our understanding of typical and atypical health and development and how to optimize the delivery of early intervention services. Collecting full-day and full-week datasets from wearable sensors will also increase our understanding of the influence of environmental and cultural contexts on infant development in support of global health and development efforts. In low- and middle-income countries, 43% of all young children are at risk of not achieving their developmental potential due to malnutrition, poverty, and lack of early stimulation.⁷⁶ Improving the ability of health workers in low-resource settings to efficiently identify which children are most at risk and could benefit from intervention is an important global health priority. However, existing psychometric tools are time-consuming, require significant training and support for testers, and might not always measure culturally appropriate developmental end points.⁷⁷ Smart wearables could help to address these practical needs.

To this end, the Infant Neuromotor Control Laboratory and Maya Health Alliance collaborated to collect wearable sensor data from infants in rural Guatemala at risk for stunting of growth and developmental delay. For example, pilot data on a full day of leg movement from a 1-month-old indigenous infant in a remote part of Guatemala were compared with a 1-month-old infant in the United States with typical development. The infant in Guatemala had a movement rate of 400 movements per hour awake, compared with 800 movements per hour awake for the US infant. This preliminary observation suggests that infants experiencing stunting of growth due to undernutrition can exhibit a greater proportion of sleep time and lower quantity of movement when awake. However, infants in rural Guatemala are traditionally worn swaddled against the caregiver's body and this would be expected to impact physical activity of the limbs. Smart wearables can serve as useful tools to gather data in a variety of populations like that in Guatemala to better understand how movement experience might impact future development and to identify movement characteristics that are early indicators of stunted development.

Novel wearable sensors have the potential to make smart wearables more comfortable and common in the commercial market. Stretch sensors made from conductive threads using a coverstitch,⁷⁸ or using a zig-zag stitch,⁷⁹ are being developed and tested to measure movement as alternatives to hard, bulky sensors. These sensors can be stitched directly into clothing across joints to measure movement. Novel silicone capacitive sensors that can be integrated with fabric are also being developed and tested.⁸⁰ Soft sensors like these can address user needs related to comfort, aesthetics, and expressiveness to facilitate motion data capture in natural settings over longer periods of time.

In addition to measuring activity and identifying differences or delays, smart garments have the potential to assist in changing an individual's function. For example, sensors can provide feedback to users, informing them how to change aspects related to their quality of movement, such as pattern, speed, sway, or pressure distribution.^81,82 Smart garment feedback can potentially shape the quality of users’ movement practice so that users learn to perform behaviors in more functional, energy efficient, and safe ways. Smart garments can also provide feedback about amounts of activity performance that can be used to motivate users to improve their endurance, and physiological and mental health.^83,84 Finally, when used in conjunction with other devices, including active exoskeletons, mobility devices, and communication devices, sensors integrated in smart garments can empower users to more independently and fully navigate and participate in their physical and social environment (Fig. 5).^52,85,86

Figure 5.

An example of how smart wearables can allow children to control their environment in meaningful ways. This prototype allows a child to use gross arm movement rather than more challenging fine manual movements to engage in video game play.

Conclusions

Various novel wearables have become available to rehabilitation professionals and clients in recent years. Wearables spanning clothing through smart garments can be used in a variety of ways for assessment and intervention to impact users’ abilities, task demands, or the environment to promote functional performance and participation (Fig. 6). They have the potential to be used as assistive devices to aid functional performance for users when they are worn, and as rehabilitation devices incorporated in activity programs aimed at improving independent function without the devices.⁵³ There is a significant need for additional high-quality research on all wearables to determine their effectiveness in relation to these potential uses.

Figure 6.

Potential uses of wearables for rehabilitation assessment and intervention.

Smart wearables provide the additional ability to continuously monitor a range of variables and so can help in tracking performance across time in natural settings and identifying potential delays or impairments in performance and participation. Smart wearables will likely be most effective when clinicians and clients use them to measure or provide feedback on targeted outcomes based on scientifically supported models for change. In other words, although smart garments can allow us to monitor and provide feedback on any number of variables, we should focus on those variables that the literature suggests are likely to allow us to impact our clients’ goals.^87,88 More research is needed not only to determine the validity of the available smart garments but also to determine how to optimally set goals and provide feedback to users based on data from these devices.

To maximize usage, wearables and other rehabilitation tools should not be designed with function as the sole focus but should consider the broad needs of users related to function, expressiveness, aesthetics, and accessibility, noting the role caregivers play in mediating the assessment of user needs and device usage. Physical therapists are trained to focus on the needs of users for intervention design and they liaise between users and device manufacturers. Thus, they are uniquely positioned to advocate for and to create better wearables for those they serve.¹

Author Contributions and Acknowledgments

Concept/idea/research design: B. Greenspan, M.L. Hall, L.A. Prosser, B.A. Smith, M.A. Lobo

Writing: B. Greenspan, M.L. Hall, L.A. Prosser, P. Rohloff, B.A. Smith, M.A. Lobo

Project management: B.A. Smith, M.A. Lobo

Fund procurement: M.A. Lobo

Providing facilities/equipment: M.A. Lobo

Providing institutional liaisons: P. Rohloff

Funding

The research discussed in this article was supported by the National Institutes of Health, Eunice Kennedy Shriver National Institute of Child Health & Human Development (1R21HD076092-01A1, Lobo PI), the Delaware Economic Development Office Matching Grant Program (109, Lobo PI), and the University of Delaware Research Foundation, Inc. (Lobo PI).

Role of the Funding Source

The funding sources provided funds to support personnel required for this manuscript. They played no part in the scientific processes involved in the development of this article.

Disclosures

The authors completed the ICJME Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.