Abstract
Objectives: To investigate correlations between power seat functions (PSFs) usage and wheelchair discomfort.
Design: Quasi-experimental design: Time series design.
Setting: In-home trial in participants’ home/community.
Participants: Thirteen power wheelchair users who independently used power wheelchairs equipped with PSFs as their primary means of mobility.
Main Outcome Measures: PSF usage variables include the frequency of performing repositioning and using PSFs (tilt, recline, legrests and seat elevation), wheelchair occupancy, and driving distance. The Tool for Assessing Wheelchair disComfort (TAWC) were used to evaluate general discomfort and discomfort intensity.
Results: Spearman correlation coefficient showed that the frequency of using tilt, recline, and legrest is significantly correlated with discomfort intensity. Multiple regression analysis with backward stepwise indicated that these functions can explain 43.8% of the variance (R2 = .438, F(3,33) = 8.588, P < 0.01) in the discomfort intensity score. The best-first decision tree shows that the frequency of using the legrest function was the top node, followed by recline and tilt functions. The overall accuracy of prediction with ten-fold cross validation for discomfort intensity was 79.4%.
Conclusions: For people who used power wheelchairs equipped with PSFs, correlation analysis and regression modeling provided evidence from the quantitative data that increasing the frequency of using PSFs may decrease wheelchair discomfort. Future studies should include interventions to encourage people to use their PSFs appropriately.
Keywords: Wheelchairs, Rehabilitation
Introduction
Historically, a dynamic sitting posture with frequent small postural changes has been viewed as a healthy sitting posture.1 Previous studies have shown that there is not an optimal static sitting posture, because sitting postures increase the strain on active (muscle) and passive (tendon or ligament) spinal structures.2,3 People naturally change their postures frequently while sitting, including head, trunk and pelvis angles. Spontaneous postural change can potentially decrease sitting discomfort, since a prolonged stretched condition on contractile and non-contractile tissues and a long duration of low muscular activity might be risk factors for pain or injury.4
Wheelchair seating discomfort is a very common problem for wheelchair users.5–8 This problem could be worse for power wheelchair users who spend most of their time in their wheelchair and may not be able to adjust their sitting posture as frequently as needed independently. In addition, other issues such as prolonged shear force on the back and buttock area or a slumping posture due to weak trunk control ability9 might cause wheelchair-sitting discomfort. For this reason, power seat functions (PSFs) become an important means for power wheelchair users to change their posture dynamically. PSFs include tilt-in-space, backrest recline, and legrests/leg support, all of which allow power wheelchair users to reposition themselves in their wheelchair, to redistribute pressure on areas that have a high probability of skin breakdown, and to stabilize their position while moving.9 Previous studies have mainly focused on the effect of using PSFs on preventing pressure ulcers and improving driving safety,10–14 and a few studies reported that wheelchair users tend to use tilt and recline functions to improve seating discomfort.8,15,16 The results from these studies are, however, based on users’ reports or surveys, and no study has attempted to investigate if changing the frequency of adjusting posture in a wheelchair is able to alleviate wheelchair sitting discomfort. The aim of this study is to investigate correlations between PSF usage and wheelchair discomfort.
Methods
Participants
Thirteen individuals participated in an 8-week experiment in their home/community (Table 1). The inclusion criteria for selection of power wheelchair users was 18 years or older and with the ability to operate a power wheelchair equipped with PSFs as their primary means of mobility. All the participants had been approved for a new power wheelchair with PSFs when we recruited them for this study, and we used the window of the time while they were waiting for their new power wheelchair to conduct this study. In order to accommodate preferences for using different driving wheel configurations, we prepared front- (Permobil C500, seat width: 19″W × 18″D; Permobil AB, Timrå, Sweden), mid- (Pride Quantum 6400Z, seat width: 18″W × 18″D; Pride Mobility Products Corp., Exeter, PA, USA), and rear-wheel (Permobil Street, seat width: 19″W × 18″D) drive wheelchairs for this study. When a person agreed to participate in this study, our power wheelchair was delivered to his/her home environment. A clinician who was also an Assistive Technology Professional (ATP) with more than 10 years of experience providing power wheelchairs with PSFs verified that the power wheelchair would be suitable in his/her environment. This study was approved by the Institutional Review Board at the University of Pittsburgh and the Department of Veterans Affairs Institutional Review Board.
Table 1.
Demographic data of participants
| Subject | Sex | Age (year) | Height (Inch) | Weight (lb) | Diagnosis | Year of power wheelchair use | New PSFs Users |
|---|---|---|---|---|---|---|---|
| 1 | M | 48 | 72 | 145 | Muscular Dystrophy | 11 | No |
| 2 | M | 59 | 71 | 192 | Multiple Sclerosis | 3 | Yes |
| 3 | M | 44 | 66 | 140 | Cerebral Palsy | 5 | No |
| 4 | F | 62 | 66 | 280 | Poliomyelitis | 0 | Yes |
| 5 | M | 47 | 68 | 153 | Muscular Dystrophy | 0 | Yes |
| 6 | M | 45 | 67 | 244 | Spinal Cord Injury | 10 | Yes |
| 7 | F | 67 | 64 | 174 | Inclusion Body Myositis | 4 | Yes |
| 8 | F | 44 | 65 | 140 | Multiple Sclerosis | 5 | No |
| 9 | M | 39 | 69 | 215 | Spinal Cord Injury | 0 | Yes |
| 10 | M | 57 | 70 | 248 | Spinal Cord Injury | 10 | No |
| 11 | F | 57 | 63 | 200 | Degenerative Joint Disease | 11 | Yes |
| 12 | F | 65 | 65 | 180 | Multiple Sclerosis | 0 | Yes |
| 13 | F | 43 | 66 | 158 | Multiple Sclerosis | 0 | Yes |
| 52.07 ± 9.40 | 67.08 ± 5.51 | 189.92 ± 91.07 | 4.54 ± 9.11 |
Measurement of PSF usage
Each study power wheelchair was equipped with rotary encoders (A2K Absolute Optical Encoder, US Digital, Vancouver, WA, USA) to record angle changes for tilt, recline, legrests, and seat elevation. Two rotary encoders (E2 Optical Kit Encoder, US Digital) mounted on both wheels recorded the wheelchair driving distance. A customized seat switch was placed between the seat cushion and seat pan to detect wheelchair occupancy. All the sensors were mounted on the wheelchair securely with customized cases and were calibrated before each participant received the wheelchair (Fig. 1). A personal computer was attached behind the backrest of the wheelchair and the sensors were all interfaced to the computer.
Figure 1.
The sensors to record PSFs for instruction group in tablet virtual seating coach study (colour online)
Measurement of wheelchair discomfort
This study used the Tool for Assessing Wheelchair disComfort (TAWC) for measuring wheelchair discomfort.17,18 The TAWC consists of three sections. Section I is general information about factors that directly affect wheelchair-related discomfort. Section II contains eight statements related to discomfort and five statements related to comfort. These 13 statements are rated on a seven-point Likert scale where 1 is “strongly disagree” and 7 is “strongly agree” (score range: 13–91). In Section III, seven body areas (back, neck, buttocks, legs, arms, feet, and hands) are rated for a degree of discomfort intensity on a scale of 0 (no discomfort) to 10 (severe discomfort). It also provides spaces which allow users to list any additional body areas, and the general discomfort intensity is also rated on the same 0–10 scale. This study only calculated the seven default body areas (score range: 0–70). Section II and III form two measures of discomfort—the 13 items in Section II are called the General Discomfort Score (GDS) and the seven items in Section III are the Discomfort Intensity Score (DIS). The TAWC has been reported as a reliable and stable tool for quantifying wheelchair seating discomfort.19 For the responsiveness measures, the average change on the GDS and DIS were 8.7 and 7.7.18
Procedure
The total length of the study was 8 weeks. The study period included 5 visits (4 phases: baseline and phases 1, 2 and 3), and the interval between each visit was 2 weeks. During the study, participants were instructed to go about their daily activities as they normally would. The study wheelchair tracked their usage of the wheelchair (e.g. driving distance and wheelchair occupancy) and seating habits (e.g. frequency of using PSFs and frequency of repositioning exercise). In addition, they were asked to fill out the TAWC each day.
Visit 1 was either performed on the same day of the participants’ initial wheelchair assessment or scheduled at a time convenient for the participants within 2 weeks after their initial assessment. Our research wheelchair was brought into the participant's home environment and was fitted and tuned to the subject's needs by the clinician. Once the participant transferred to our research wheelchair, the clinician instructed him/her to drive the wheelchair and access the PSFs to allow the participant to be familiar with our study wheelchairs. If our study wheelchairs were unable to fit participants’ environment, the participants were withdrawn from the study.
In Visits 2–5, our research team collected the TAWC forms and downloaded the PSF usage data from the computer. In Visit 2, all participants reviewed a study pamphlet, pocket-size reminder cards, and compact disc containing educational materials on PSF usage. These educational materials included information on using PSFs for body position, posture realignment, transfers, and repositioning exercises. The materials included guidelines on driving in different environments, such as slopes. We expected to observe if the participants changed their behavior of using PSFs after reviewing the materials and if these changes in PSF usage were correlated with changes in wheelchair discomfort.
Data analysis
All the variables including the potential predictors and outcome variables (Table 2) within one phase were first averaged. Because the scores from TAWC are subjective to participants’ personal feeling, the data of the other phases was normalized to the baseline. The reason for preprocessing the data with this method is that each participant had their own pattern of using PSFs and wheelchair discomfort is a subjective measurement, so only the changes of all variables from the baseline were analyzed. The total data points in the analysis were 39 (13 participants multiplied by three phases).
Table 2.
The definition of all variables
| Name | Definition |
|---|---|
| Potential Predictors | |
| Freq_Tilt | The frequency of using tilt function within a day. |
| Freq_Recl | The frequency of using recline function within a day. |
| Freq_Legr | The frequency of using legrests function within a day. |
| Freq_Seat | The frequency of using seat elevation function within a day. |
| WOCO | Wheelchair occupancy (hour/day) |
| Freq_Repo | The frequency of performing repositioning exercise within a day (number/day). The repositioning exercise is defined if the users tilt backward to 30 degree and stay in that position for 30 seconds every hour in this study. |
| Distance | The driving distance of wheelchair (mile/day). |
| Outcome Variables | |
| GDS | General discomfort score in TAWC |
| DIS | Discomfort intensity score in TAWC |
A test for normality was performed before other analyses. Correlation coefficients were calculated to determine predictor(s) for the linear regression model. Linear regression analysis using a backward stepwise method was performed to determine if TAWC wheelchair discomfort scores were predicted by the correlated variables. Correlation and linear regression analyses were performed using SPSS 22.0 (IBM Corp, Armonk, NY, USA).
In order to find the cut-off points that make the meaningful changes for GDS and DIS, the GDS and DIS results were classified as discomfort decreased (decreased), discomfort increased (increased), and discomfort not changed (neutral) based on the responsiveness (8.7 for GDS and 7.7 for DIS).18 Weka20 was used to perform the best-first decision tree learning.21 The variables correlated with the GDS or DIS were used as the features to classify discomfort changes. The performance of predicting the GDS or DIS was evaluated using ten-fold cross validation.
Results
Table 3 shows median and interquartile range of all variables in each phase of the study. Most variables except wheelchair occupancy, general discomfort, and discomfort intensity were not normally distributed. Two outliers were removed, because data points were greater than 3 standard deviations beyond the mean. For this reason, Spearman correlation was performed to calculate the correlation matrix among all variables (Table 4). The results show that the frequency of using the tilt, recline, and legrests functions were significantly correlated with discomfort intensity, so these three variables served as the predictors in the linear regression analysis. The assumptions for linear regression were all met. The assumption of multocollinearity among three predictors was not violated, since there was no significant correlation among the predictors (Table 4). The results of backward stepwise regression indicated that these predictors can explain 43.8% of the variance (R2 = 0.438, F(3,33) = 8.588, P < 0.01) in discomfort intensity (Table 5 and Fig. 2). The value of Durbin-Watson test was 1.332, showing that the errors from each predictor were independent. Figure 3 shows the result of homoscedasticity. As can be seen, the residuals were randomly distributed at each level of the predictors and slope of linear curve fitting is close to 0, meaning that the residuals had the same variance at each level of the predictors. The Shapiro-Wilk (value = 0.983, P = 0.825) also indicated the errors of the linear regression model are normally distributed.
Table 3.
The median and interquartile range (IQR) of all variables in each phase of the study
| Baseline | Phase 1 | Phase 2 | Phase 3 | ||
|---|---|---|---|---|---|
| Freq_Tilt (numbers/day) | Median | 3.94 | 4.33 | 4.00 | 4.78 |
| IQR | 2.02 | 3.45 | 3.13 | 2.88 | |
| Freq_Recl (numbers/day) | Median | 3.30 | 2.58 | 3.38 | 3.16 |
| IQR | 1.67 | 4.25 | 2.02 | 3.03 | |
| Freq_Legr (numbers/day) | Median | 4.76 | 3.31 | 3.38 | 3.31 |
| IQR | 4.97 | 8.99 | 7.19 | 6.71 | |
| Freq_Seat (numbers/day) | Median | 0.31 | 2.46 | 0.42 | 0.73 |
| IQR | 1.32 | 1.33 | 2.15 | 2.26 | |
| WOCO (hours/day) | Median | 5.30 | 6.08 | 6.32 | 5.31 |
| IQR | 5.13 | 3.76 | 3.74 | 5.01 | |
| Freq_Repo (numbers/day) | Median | 0.19 | 0.78 | 0.50 | 0.71 |
| IQR | 0.54 | 1.58 | 0.46 | 1.22 | |
| Distance (miles/day) | Median | 0.87 | 1.20 | 1.26 | 1.15 |
| IQR | 1.18 | 1.39 | 1.13 | 1.70 | |
| GDS (score/day) | Median | 40.11 | 39.21 | 40.95 | 42.15 |
| IQR | 16.54 | 11.40 | 14.36 | 13.02 | |
| DIS (score/day) | Median | 20 | 18.13 | 13.86 | 14.97 |
| IQR | 12.33 | 14.69 | 16.57 | 15.28 |
Table 4.
The correlation matrix of all variables
| Freq_Recl | Freq_Legr | Freq_Seat | Freq_Repo | WOCO | Distance | GDS | DIS | |
|---|---|---|---|---|---|---|---|---|
| Freq_Tilt | 0.281 | 0.280 | 0.104 | 0.661** | 0.140 | 0.075 | 0.030 | –0.353* |
| Freq_Recl | –0.277 | 0.095 | 0.0.94 | –0.242 | –0.093 | –0.002 | –0.543** | |
| Freq_Legr | 0.065 | 0.450** | 0.203 | –0.025 | –0.186 | 0.353* | ||
| Freq_Seat | 0.390* | 0.104 | 0.272 | –0.003 | –0.081 | |||
| Freq_Repo | –0.017 | 0.121 | 0.0027 | –0.066 | ||||
| WOCO | –0.139 | –0.287 | 0.114 | |||||
| Distance | –0.011 | –0.149 | ||||||
| GDS | 0.139 |
*: P < 0.05
**: P < 0.01
Table 5.
The result of multiple regression using backward stepwise method to predict DIS
| Unstandardized Coefficients | Standardized Coefficients |
||||
|---|---|---|---|---|---|
| Model | B | Std. Error | Beta | t | Sig |
| Constant | –1.524 | 0.880 | –1.732 | 0.093 | |
| Freq_Tilt | –0.252 | 0.105 | –0.356 | –2.414 | 0.022 |
| Freq_Recl | –0.897 | 0.384 | –0.344 | –2.3363.360 | 0.026 |
| Freq_Legr | 0.734 | 0.303 | –0.357 | 2.425 | 0.021 |
Figure 2.
The result of multiple linear regression model shows that the score change of discomfort intensity can be 43.8% predicted by the frequency change of using tilt, recline and legrests functions
Figure 3.
The assumption of homoscedasticity shows that the residuals had the same variance at each level of the predictors
Best-first decision tree learning was performed to classify discomfort intensity. The frequency of using the tilt, recline, and legrests functions was the feature in the learning algorithm. The results of the decision tree are shown in Figure 4. The frequency of using the legrests was the top node of the decision tree, followed by recline and tilt. The overall accuracy of prediction of discomfort intensity with ten-fold cross validation was 79.4%.
Figure 4.
The result of Best first decision tree
Discussion
This study investigated the correlation between PSF usage and wheelchair discomfort. The results from correlation analyses showed that wheelchair discomfort intensity is correlated with the frequency of using tilt, recline, and legrests functions. Correlation analysis also showed that the frequency of using the tilt, recline, and legrests functions were not correlated with each other. This may indicate that all three PSFs contribute to changing wheelchair discomfort intensity. Multiple linear regression with backward stepwise modeling supported this result, since none of the functions were eliminated from the model. Also, the assumptions for the regression model were all met, so the model was able to show an unbiased result. The negative correlation indicates that increasing the frequency of using tilt and recline decreases wheelchair discomfort intensity. This result confirms previous studies showing that users tend to use the tilt and recline functions to improve seating discomfort.8,22,23 The tilt function is a gravity-assisted way to realign users in their wheelchairs.9,24 It is also an important method for relieving seating pressure and restoring skin perfusion.16 In addition, the posture achieved using the recline function is believed to relax the lower back muscles.25 Using the recline function can change the length and tension of contractile and non-contractile tissue in the lumber and pelvic area, which is able to prevent prolonged low-level muscle activity which might impair muscle oxygenation and cause pain,26 thus resulting in sitting discomfort.
One unexpected finding is the positive correlation between the frequency of using legrests and wheelchair discomfort intensity. Elevating legrests are a common way for power wheelchair users to deal with edema27,28 and contractures of knee joints.24 When elevating the legrests, it is recommended that the backrest be reclined to prevent tight hamstring muscles from rotating the pelvis posteriorly leading to a slouching posture. This result indicates that caution is needed when using elevating legrests and emphasizes the importance of education.
Best-first decision tree analysis was performed to find cut-off points to understand levels of change in tilt, recline and legrests angles that results in significant differences in discomfort intensity. An advantage of decision tree algorithms is that they can describe a pathway and define thresholds for decision nodes to provide quantitative information on the degree of adjustment of PSF usage that can change wheelchair discomfort. The algorithm selected the frequency of legrests usage as the first decision node. Decreased discomfort intensity occurred when the frequency of using legrests was below 2.41 times per week. Increasing frequency of tilt and recline usage was observed to mitigate discomfort due to increasing frequency of legrests elevation usage.
Study limitations
One limitation to this study is that even though this is a longitudinal observation of PSFs and wheelchair discomfort, the inherent disadvantage of correlation tests is that they do not prove causation. However, anecdotally, it is more likely that participants experienced less discomfort with increased use of PSFs. Another limitation is that there were only thirteen participants. The reasons that some participants dropped out of the study are (1) they received their new power wheelchair before the end of the study, and their living environment did not allow them to have two wheelchairs; (2) the participants could not fit into our research wheelchairs, since we only prepared the research wheelchairs with the size that most people can fit in according to the statistics from the clinics; (3) our study wheelchairs did not fit into users’ environments and the users could not accommodate them; and (4) the participants could not answer the TAWC questionnaire each day. In addition, we provided the educational materials to the participants in visit 2 and expected that they would lead to increased PSF usage, so we would be able to observe the decrease of wheelchair discomfort. However, some of our participants did not increase or decrease PSF usage. Nonetheless, the multiple linear regression model can reflect the increasing and decreasing change of discomfort intensity explained by the changes of tilt, recline and legrests functions usage. The limitation of the decision tree model is that most of the changes were below the responsiveness score (7.7 for discomfort intensity score) and we only observed a few data points with meaningful discomfort intensity decrease (only 7 discomfort intensity decrease from 37 data points). The imbalance of the data points could weaken the generalization of the decision tree model. Future studies should utilize an intervention which could significantly and continuously facilitate users in utilizing their PSFs, so it can further consolidate the relationship between using PSFs and wheelchair discomfort and verify the effect of any intervention. Furthermore, this study reports that power wheelchair users should utilize a dynamic sitting posture (changing their posture periodically by PSFs) in order to have better wheelchair comfort, rather than staying in a posture for a relatively long time. However, the wheelchair discomfort might result from other medical or physiological factors and the maximal improvement of wheelchair comfort that the PSFs can achieve should be further investigated.
Conclusion
This study investigated the relationship between PSF usage and wheelchair discomfort for power wheelchair users. The evidence from the quantitative data indicates that increasing the frequency of using PSFs - especially the tilt and recline functions - can potentially decrease wheelchair discomfort. Future studies should develop an intervention which can significantly change the PSF usage and observe the improvement of the wheelchair discomfort.
Disclaimer statements
Contributors None.
Funding This research was completed with the support of the National Science Foundation (Grant #EEC0540865 - Quality of Life Technology Engineering Research Center), VA RR&D (Merit Review Grant # B6591R - Powered Seating Function Usage among Veterans-Compliance and Coaching), and the University of Pittsburgh.
Conflict of Interest Statement The authors (Y Wu, H Lui, J Pearlman, RA Cooper) wish to disclose that they are the inventors of the Seating Function Monitoring and Coaching System (US Patent Application 14/609,747), which has been licensed by the University of Pittsburgh to Permobil, USA as the Virtual Seating Coach by Permobil, USA. The authors do not have any personal financial interest in Permobil, USA. Federal Disclaimer: The contents of this paper do not represent the views of the Department of Veterans Affairs or the United States Government.
Ethics approval None.
Financial Disclosure We certify that no party having a direct interest in the results of the research supporting this article has or will confer a benefit on us or on any organization with which we are associated AND, if applicable, we certify that all financial and material support for this research (eg, NSF or VA grants) and work are clearly identified in the title page of the manuscript.
ORCiD
Yu-Kuang Wu http://orcid.org/0000-0002-6145-5696
Jonathan Pearlman http://orcid.org/0000-0003-0830-9136
Acknowledgement
The contents of this paper do not represent the views of the Department of Veterans Affairs or the United States Government.
References
- 1.Cook C, Burgess-Limerick R. Guidelines for computer workstations. Ergonomics Australia 2003;17(1):19–37. [Google Scholar]
- 2.Andersson E, Oddsson L, Grundström H, Nilsson J, Thorstensson A. EMG activities of the quadratus lumborum and erector spinae muscles during flexion-relaxation and other motor tasks. Clin Biomech (Bristol, Avon) 1996;11(7):392–400. doi: 10.1016/0268-0033(96)00033-2 [DOI] [PubMed] [Google Scholar]
- 3.O'Sullivan PB, Dankaerts W, Burnett AF, Farrell GT, Jefford E, Naylor CS, et al. Effect of different upright sitting postures on spinal-pelvic curvature and trunk muscle activation in a pain-free population. Spine 2006;31(19):E707–12. doi: 10.1097/01.brs.0000234735.98075.50 [DOI] [PubMed] [Google Scholar]
- 4.Shin G, Mirka GA. An in vivo assessment of the low back response to prolonged flexion: interplay between active and passive tissues. Clin Biomech (Bristol, Avon) 2007;22(9):965–71. doi: 10.1016/j.clinbiomech.2007.06.003 [DOI] [PubMed] [Google Scholar]
- 5.Chesney D, Hsu L, Wright W, Axelson P, editors. Immediate improvements in wheelchair mobility and comfort with use of the adjustable back support shaping system. Proceedings of the RESNA ’95 Annual Conference; 1995 Jun 9–14; Vancouver Canada Arlington, VA: RESNA Press; 1995. [Google Scholar]
- 6.Harms M. Effect of wheelchair design on posture and comfort of users. Physiotherapy 1990;76(5):266–71. doi: 10.1016/S0031-9406(10)62219-5 [DOI] [Google Scholar]
- 7.Shaw G. Wheelchair seat comfort for the institutionalized elderly. Assist Technol 1991;3(1):11–23. doi: 10.1080/10400435.1991.10132176 [DOI] [PubMed] [Google Scholar]
- 8.Weiss-Lambrou R, Tremblay C, LeBlanc R, Lacoste M, Dansereau J. Wheelchair seating aids: How satisfied are consumers? Assist Technol 1999;11(1):43–53. doi: 10.1080/10400435.1999.10131984 [DOI] [Google Scholar]
- 9.Dicianno BE, Arva J, Lieberman JM, Schmeler MR, Souza A, Phillips K, et al. RESNA position on the application of tilt, recline, and elevating legrests for wheelchairs. Assist Technol 2009;21(1):13–22. doi: 10.1080/10400430902945769 [DOI] [PubMed] [Google Scholar]
- 10.Corfman TA, Cooper RA, Fitzgerald SG, Cooper R. Tips and falls during electric-powered wheelchair driving: effects of seatbelt use, legrests, and driving speed. Arch Phys Med Rehabil 2003;84(12):1797–802. doi: 10.1016/S0003-9993(03)00467-2 [DOI] [PubMed] [Google Scholar]
- 11.Riordan J, Voegeli D. Prevention and treatment of pressure ulcers. Br J Nurs 2009;18(20):S20, S2, S4–7. [DOI] [PubMed] [Google Scholar]
- 12.Soban LM, Hempel S, Munjas BA, Miles J, Rubenstein LV. Preventing pressure ulcers in hospitals: A systematic review of nurse-focused quality improvement interventions. Jt Comm J Qual Patient Saf 2011;37(6):245–52. doi: 10.1016/S1553-7250(11)37032-8 [DOI] [PubMed] [Google Scholar]
- 13.Moore Z, Cowman S, Conroy RM. A randomised controlled clinical trial of repositioning, using the 30 degrees tilt, for the prevention of pressure ulcers. J Clin Nurs 2011;20(17–18):2633–44. doi: 10.1111/j.1365-2702.2011.03736.x [DOI] [PubMed] [Google Scholar]
- 14.Makhsous M, Lin F, Knaus E, Zeigler M, Rowles DM, Gittler M, et al. Promote pressure ulcer healing in individuals with spinal cord injury using an individualized cyclic pressure-relief protocol. Adv Skin Wound Care 2009;22(11):514–21. doi: 10.1097/01.ASW.0000305495.77649.ee [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Henderson JL, Price SH, Brandstater ME, Mandac BR. Efficacy of three measures to relieve pressure in seated persons with spinal cord injury. Arch Phys Med Rehabil 1994;75(5):535–9. [PubMed] [Google Scholar]
- 16.Jan YK, Jones MA, Rabadi MH, Foreman RD, Thiessen A. Effect of wheelchair tilt-in-space and recline angles on skin perfusion over the ischial tuberosity in people with spinal cord injury. Arch Phys Med Rehabil 2010;91(11):1758–64. doi: 10.1016/j.apmr.2010.07.227 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Crane BA, Holm MB, Hobson D, Cooper RA, Reed MP, Stadelmeier S. Development of a consumer-driven wheelchair seating discomfort assessment tool (WcS-DAT). Int J Rehabil Res 2004;27(1):85–90. doi: 10.1097/00004356-200403000-00014 [DOI] [PubMed] [Google Scholar]
- 18.Crane BA, Holm MB, Hobson D, Cooper RA, Reed MP. Responsiveness of the TAWC tool for assessing wheelchair discomfort. Disabil Rehabil Assist Technol 2007;2(2):97–103. doi: 10.1080/17483100601178484 [DOI] [PubMed] [Google Scholar]
- 19.Crane BA, Holm MB, Hobson D, Cooper RA, Reed MP, Stadelmeier S. Test-retest reliability, internal item consistency, and concurrent validity of the wheelchair seating discomfort assessment tool. Assist Technol 2005;17(2):98–107. doi: 10.1080/10400435.2005.10132100 [DOI] [PubMed] [Google Scholar]
- 20.Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH. The WEKA data mining software: an update. SIGKDD Explorations Newsletter 2009;11(1):10–8. doi: 10.1145/1656274.1656278 [DOI] [Google Scholar]
- 21.Shi H. Best-first decision tree learning [thesis]. Waikato, New Zealand: University of Waikato; 2007. [Google Scholar]
- 22.Lacoste M, Weiss-Lambrou R, Allard M, Dansereau J. Powered tilt/recline systems: why and how are they used? Assist Technol 2003;15(1):58–68. doi: 10.1080/10400435.2003.10131890 [DOI] [PubMed] [Google Scholar]
- 23.Ding D, Leister E, Cooper RA, Cooper R, Kelleher A, Fitzgerald SG, et al. Usage of tilt-in-space, recline, and elevation seating functions in natural environment of wheelchair users. J Rehabil Res Dev 2008;45(7):973–83. doi: 10.1682/JRRD.2007.11.0178 [DOI] [PubMed] [Google Scholar]
- 24.Lange M. Positioning: it's all in the angles. Adv Occup Ther Pract 2006;22(5):42. [Google Scholar]
- 25.Van Dieen J, De Looze M, Hermans V. Effects of dynamic office chairs on trunk kinematics, trunk extensor EMG and spinal shrinkage. Ergonomics 2001;44(7):739–50. doi: 10.1080/00140130120297 [DOI] [PubMed] [Google Scholar]
- 26.McGill SM, Hughson RL, Parks K. Lumbar erector spinae oxygenation during prolonged contractions: implications for prolonged work. Ergonomics 2000;43(4):486–93. doi: 10.1080/001401300184369 [DOI] [PubMed] [Google Scholar]
- 27.Kreutz D. Power tilt, recline or both. Team Rehab Report 1997;8(3):29–32. [Google Scholar]
- 28.Levy C, Berner TF, Sandhu PS, McCarty B, Denniston NL. Mobility challenges and solutions for fibrodysplasia ossificans progressiva. Arch Phys Med Rehabil 1999;80(10):1349–53. doi: 10.1016/S0003-9993(99)90043-6 [DOI] [PubMed] [Google Scholar]