Adapted Feeding Utensils for People With Parkinson’s-Related or Essential Tremor

Impact of Tremor on the Task of Manipulating Utensils for Self-Feeding

Uncontrolled shaking of the arm or hand is most commonly associated with the diagnosis of idiopathic essential tremor (ET), but it is also a problem for a subgroup of people with Parkinson’s disease (PD; Gövert & Deuschl, 2015; Louis & Frucht, 2007). Several differences exist between the motor manifestations of PD and those of ET. PD tremor is typically associated with a lower frequency but higher amplitude of tremor. In early stages of PD, tremor is manifested unilaterally, whereas tremor in ET is bilateral (Fekete & Jankovic, 2011). Moreover, people with PD experience several motor problems in addition to tremor, including kyphotic posture, bradykinesia, rigidity, and difficulty with balance and walking (O’Connor & Kini, 2011).

Nevertheless, the two groups have many similarities with regard to tremor-related quality of life (Louis & Machado, 2015). This motor impairment affects performance of many activities of daily living; in particular, it leads to difficulty using utensils for feeding. Medical, surgical, and rehabilitative interventions are often inadequate in ameliorating the problems related to feeding, and people are left feeling ashamed of how they appear to others or of their dependency on caregivers for the basic self-care task of eating (Louis & Machado, 2015; Traub et al., 2010). Particularly during daily feeding activities, many people with PD might benefit from the same types of devices as people with ET.

Adaptive Utensils to Improve Feeding Performance of People With Tremor

In 1972, the concept of weighted lead cuffs was tested on 50 participants with tremor associated with a variety of diagnoses, including cerebellar ataxia, ET, PD, stroke, and chronic alcohol abuse (Hewer et al., 1972). These researchers found that the weighted cuffs helped 36% (18) of their participants; however, most of these participants had discontinued using the weighted cuffs at home when interviewed at 6-mo follow-up. The researchers did not articulate any physiological rationale for why this intervention might have been helpful. They also did not analyze which diagnoses were most represented among the participants for whom the intervention seemed to be helpful.

In a follow-up study, Morgan et al. (1975) referred to the use of weighted lead cuffs as a “mechanical damping device,” but they also failed to find considerable beneficial impact. Their recommendation was that the optimal weight for the lead cuffs would be between 600 and 840 g. Presumably because of this research, companies marketing adaptive equipment for people with disabilities began selling weighted utensils. These utensils typically weigh between 5 and 8 oz (141–227 g).

A search of the published literature revealed only two retrievable studies that sought to assess the efficacy of weighted cuffs or weighted utensils for people with PD or ET. Meshack and Norman (2002) used a repeated-measures design with 16 participants who had hand tremor and who were diagnosed with PD. In comparing the use of a weighted wrist cuff weighing 470 g with holding a spoon of typical weight with a large handle (built-up spoon), these researchers found no notable differences in measure of tremor amplitude or frequency. On the basis of their findings, they suggested that there is no support for the clinical recommendation of using either weighted utensils or wrist cuffs to reduce postural hand tremor in people with PD.

Ma et al. (2009) conducted a controlled clinical trial to assess the effect of utensil weight on functional arm movement in people with PD. They used videotape analysis of 18 adults with PD as they performed a food transfer task using spoons of three different weights: lightweight (35 g), control (85 g), and weighted (135 g). Their outcome measures were kinematic variables of arm movement: movement time, peak velocity, and number of movement units. They found no benefit to using weighted utensils.

In addition to weighted utensils, the “swivel spoon” is another adaptive device that is recommended to assist people with tremor during self-feeding activities. The swivel spoon, patented in 1979, has a mechanical counterbalance swivel component that keeps the utensil level regardless of the position of the hand and forearm. The patent application stated that this design keeps food on the utensil when turned at any angle and prevents solid or liquid food from spilling from the utensil even in the presence of tremors (Wasson, 1979). Although this device is available from several self-care companies, there is no retrievable, published literature to support its effectiveness.

Utensils with enlarged handles are also marketed as products to assist people with a variety of disabilities, including tremor. Although there is no research evidence to support their effectiveness, weighted utensils and swivel utensils are commercially available with both standard handles and enlarged handles.

Engineers have proposed several technologies to suppress tremor using external orthoses or hand-held devices (Rocon et al., 2007). To date, only one technology has successfully led to a viable product that is accessible to consumers. In Liftware utensils, active cancellation of tremor (ACT) device technology uses sensors that are embedded in the utensil to detect movements as they occur and to produce quick, precise countermovements to “cancel out” the tremor. In a pilot study with 15 participants using the Liftware spoon, Pathak et al. (2014) compared the participants’ performance as they completed tasks with the ACT device turned on and off. The three tasks were holding the utensil, simulating eating (filling the spoon with foam blocks and lifting the load to their lips), and transferring a spoonful of foam blocks into an empty cup 75 cm away. Tremor severity was rated by neurologists using the Fahn–Tolosa–Marin Tremor Rating Scale (Fahn et al., 1993). Participants rated their perceptions of subjective improvement by self-report using the Clinical Global Impression Scale for Improvement (CGI–I; Busner & Targum, 2007). Tremor amplitude was measured with accelerometers embedded in the devices. Statistical analysis showed significant reductions in tremor severity across all three tasks and significant improvements in participants’ perceptions in the eating and transferring tasks. Accelerometer recordings provided general support for the impact of ACT on tremor amplitude.

In November 2014, Google purchased Lift Labs, the startup initiated by Pathak and colleagues (Depra, 2014). Since then, Liftware utensils have been accessible to consumers; yet, there have been no subsequent publications assessing the efficacy of this device compared with weighted or swivel utensils. In 2017, with the introduction of a second device for a different consumer population, the original Liftware utensil became known as Liftware Steady™.

This study is the first to compare the various devices available to consumers who have difficulty with self-feeding because of tremor associated with ET or PD. The objective was to provide participants with the opportunity to try various devices in a simulated feeding task, compare speed of performance among the devices, and—using the CGI–I (Busner & Targum, 2007)—gather their feedback about the values and limitations of each of the devices.

Participant Recruitment

After securing approval from the institutional review board at the State University of New York (SUNY) Downstate Medical Center, we posted recruitment flyers on Fox Trial Finder (https://foxtrialfinder.michaeljfox.org/), and we received assistance from the International Essential Tremor Foundation (http://www.essentialtremor.org/) and the Brooklyn Parkinson Group (http://www.brooklynparkinsongroup.org) in recruiting potential participants. These efforts yielded 22 volunteers who met the study’s inclusion criteria. Recruitment materials included an information sheet, a flyer, and a screening test used by Liftware and published on the company’s website (http://www.liftware.com/test; Miocinovic et al., 2016). The screening test is a printable page, consisting of three circles of different diameters. Following the protocol for this screening test, potential participants understood that to participate in this study, they had to be able to hover a pen for 5 s within the largest of the three circles (diameter of 1.65 in. [4.2 cm]).

Inclusion criteria were (1) self-reported difficulty managing eating utensils because of tremor; (2) diagnosis of ET, PD, or ET and PD; (3) between ages 18 and 85 yr; (4) ability to converse in English with researchers; (5) ability to maintain an erect seated posture in a standard chair; and (6) ability to meet the physical criteria on the screening assessment, described earlier.

Materials

Materials included a stable, adjustable cardboard platform; two standard bowls; 1 cup (250 ml) of dry Cheerios cereal; and four adapted utensils: a Deluxe KEatlery™ weighted spoon (weight, 8 oz [227 g]), a Good Grips™ weighted spoon with built-up handle (weight, 6.5 oz [184 g]; diameter, 1.5 in. [3.91 cm]), a plastic handle swivel spoon, and a Liftware Steady spoon. The first three utensils were purchased from an online vendor of rehabilitation and self-care products. The Liftware Steady spoon was loaned to us, on our request, by Liftware. Assessment tools included the recruitment screening test and the CGI–I, with questions designed to give participants the opportunity to assess each utensil.

The CGI–I is a 7-point Likert scale, originally developed to enable health professionals to briefly assess the degree of change (improvement or worsening) in overall function before and after implementing an intervention (Busner & Targum, 2007). The CGI–I is commonly used as an outcome measure in PD research (Ahn et al., 2007; Genc et al., 2016). In this study, the CGI–I questions were designed to provide participants with the opportunity to reflect on their performance of the simulated feeding task after using each of the four adapted utensils. Four of the five questions prompted participants to express the extent to which they agreed or disagreed with the following comparisons: (1) “This is easier to use compared with a standard spoon,” (2) “I feel that my performance is faster compared with a standard spoon,” (3) “I feel that my performance is neater compared with a standard spoon,” and (4) “I feel that my performance looks better compared with a standard spoon.” The fifth question prompted participants to express the extent to which they agreed or disagreed with the statement, “I like the appearance of this utensil.”

Procedures

Each participant attended a single 1-hr session in a private room equipped with a standard-height table surface with a minimum of 3 ft × 1.5 ft (0.91 × 0.46 m) of working space and a standard-height stationary chair without arms. Eleven participants were tested in a university conference room; 6 participants were tested at the Brooklyn Parkinson Group site, and 5 participants were tested in their own homes. If the participant became fatigued, rest breaks were provided. Participants were free to terminate the session at any time.

First, we administered the Liftware assessment of tremor amplitude. Then, each participant answered demographic questions and verbally described the problems that he or she had with the physical actions of feeding. In addition, participants reported which medications they were currently taking to control tremor and the approximate time of their last dosage.

Next, the cardboard platform was set up at midline at a height approximating mouth level. One empty bowl was placed on the platform, and a second bowl—filled with 1 cup of dry Cheerios cereal—was placed on the table surface, 5.9 in. (15 cm) from midline (toward the arm that the person would use for the simulated feeding task). Each participant completed five trials of transporting 10 spoonfuls of Cheerios from the bowl on the table to the bowl positioned at mouth level. The first trial was with a standard soup spoon. The second to fifth trials were with the four adapted utensils (weighted spoon, weighted spoon with built-up handle, swivel spoon, and Liftware Steady spoon). The sequence was randomized according to a card that the participant selected. Twenty-four cards with all possible permutations of the four-item sequence were placed in an envelope (e.g., A, B, C, D . . . D, C, B, A), and each participant selected one card from the envelope. After a sequence was used, that card was removed as an option. All but 2 participants completed the tasks within the 1-hr session.

For each utensil, using an iPhone™ stopwatch, we recorded speed of performance for completion of 10 repetitions of transferring a spoonful of Cheerios from the bowl on the table to the bowl at mouth level. After using each adapted utensil, the participant compared the device with the standard spoon, using the CGI–I, for ease of use, speed, neatness, performance appearance, and aesthetics of the utensil. Finally, after trying all four devices, each participant shared narrative comments with the data collector.

Data Analysis

We used Friedman’s test to compare participants’ performance speed among the four adapted utensils and to compare their ratings of these utensils for the outcomes assessed on the CGI–I. Friedman’s test is the nonparametric equivalent of repeated-measures analysis of variance (ANOVA). It was used for the self-reported responses to items on the CGI–I because these were interval data, and it was used to analyze data for performance speed because the outcomes did not have normal distributions. The non-normality of the distributions was determined by visual inspection of the corresponding histograms and the Kolmogorov–Smirnov test. To conduct pairwise comparisons among the utensils on each of the variables we assessed, we used two-way ANOVA on the basis of ranks, an established approach with Friedman’s test (Conover, 2010). We performed pairwise comparisons using Tukey’s adjustment for multiple comparisons, separately for each outcome. We considered p < .05 to be statistically significant. We performed the data analysis using SAS (Version 9.4; SAS Institute, Cary, NC).

Research Participants

A total of 22 participants completed the study: 6 with ET and 16 with tremor related to PD. These participants were active, community-dwelling adults who presented, primarily, with mild tremor. Demographic information and scores on the Liftware screen are summarized in Table 1.

Fourteen participants reported taking medication for their tremor. For participants with ET, these medications included anticonvulsants (topiramate, primidone, and gabapentin), benzodiazepam (diazepam), and the β-blocker propranolol. For participants with PD, medications included dopamine replacement therapy (carbidopa–levodopa), pramipexole (a dopamine agonist), selegiline and rasagiline (monoamine oxidase inhibitors), and amantadine. All participants complied with our request that they refrain from taking their medication dose within several hours before their study session.

Participants’ Reports of Compensatory Strategies That They Currently Used

None of the participants had any experience with Liftware or swivel utensils. Three participants reported using either the weighted spoon (KEatlery) or the weighted spoon with built-up handle. Two participants stated that the device did not help much, but they still used it because there was nothing better available. Another person complained of feeling arm fatigue shortly after eating using a weighted utensil with built-up handle. Several participants reported that when they are in stressful situations or when they are fatigued, their tremor increases and affects their feeding performance.

Speed of Performance on the Simulated Feeding Task

In comparing actual speed of performance between the standard spoon and all four adapted spoons for the simulated feeding task, Friedman’s test revealed a statistically significant difference. Pairwise comparisons, with Tukey adjustment for multiple comparisons, revealed that this difference was attributable to performance speed with the swivel spoon, which was significantly slower than with the other utensils (Table 2).

Participants’ Perceptions of the Four Adapted Utensils

On responses to the CGI–I, Friedman’s test revealed statistically significant differences in rankings among the four adapted utensils for ease of use, performance speed, neatness, performance appearance, and aesthetics of the device. For ease of use, pairwise comparisons, with Tukey adjustments for multiple comparisons, indicated that the Liftware spoon, the weighted spoon, and the weighted spoon with built-up handle did not differ significantly from one another; however, the swivel spoon was ranked significantly lower than both the Liftware spoon and the weighted spoon.

For performance speed and performance appearance (“performance looks better”), post hoc pairwise comparisons revealed that the Liftware spoon and weighted spoon were rated significantly higher than the other utensils; however, there was no statistically significant difference in ratings between the two. For aesthetics, the weighted spoon was rated significantly higher than the other three utensils (Table 3).

Participants’ Verbal Feedback About the Devices

After performing the simulated feeding task with a standard spoon and then with all four adapted utensils, participants shared their thoughts about the positive and negative attributes of each utensil.