Abstract
Background Burns occur frequently in young children. To date, insufficient data are available to fully describe the functional consequences of burns. In different patient populations and countries, the WeeFIM instrument (“WeeFIM”) often is used to measure functional independence in children.
Objective The purpose of this study was to examine the psychometric properties of the WeeFIM instrument for use in Dutch burn centers.
Design This was an observational study.
Methods The WeeFIM instrument was translated into Dutch. All clinicians who rated the children with the instrument passed the WeeFIM credentialing examination. They scored consecutive children (n=134) aged 6 months to 16 years admitted to Dutch burn centers with acute burns during a 1-year period at 2 to 3 weeks, 3 months, and 6 months postburn. To examine reliability, 2 raters scored a child at the same time (n=52, 9 raters) or the same rater scored a child twice within 1 week (n=7, 3 raters).
Results After a few weeks, the WeeFIM assessment could be administered in less than 15 minutes. Clinicians thought it was difficult to rate a child aged between 2 and 4 years as well as the cognitive items. Nevertheless, reliability was good (all intraclass correlation coefficients [1,1] were above .80). The standard error of measurement was 3.7.
Limitations Intrarater reliability was based on only 7 test-retest measurements. Within our clinical setting, it turned out to be difficult to schedule the same rater and patient twice in one week for repeated assessments. Assessments for interrater reliability, on the other hand, worked out well.
Conclusions The WeeFIM instrument is a feasible and reliable instrument for use in children with burns. For evaluation of a child's individual progress, at least 11 points' improvement should be observed to state that a child has significantly improved.
To date, insufficient data are available to fully describe the disabling consequences of burns.1 Burns might have a major impact on population health, because they frequently occur in children aged 0 to 4 years,2,3 and many of these children experience permanent consequences of their injuries from childhood to adolescence. It is important to measure the health outcomes of children and youth with disabilities because public health services and rehabilitation interventions are driven by analysis of appropriate outcomes.4 Therefore, a standard core set for the measurement and reporting of functional outcome in children after burns needs to be developed.1
The functional consequences of burns can be described with the International Classification of Functioning, Disability and Health (ICF) and the International Classification of Functioning, Disability and Health–Children and Youth Version (ICF-CY) as a framework.5 These classifications encompass functioning as a universal human experience that can be conceptualized and classified at 3 different dimensions: body function and structure, the performance of personal activities, and participation in communal life. Burns directly affect body function and structures. In addition, consequences of burns may become apparent at the activity or participation domain. An activity is defined as the execution of a task or action by an individual, and participation is defined as involvement in a life situation. Activity limitations, therefore, are difficulties an individual may have in task execution, and participation restrictions are problems with involvement in a particular life situation. Problems or deviations at one level do not automatically imply problems at other levels, although associations are likely to exist.
Regarding the activity and participation domains, in pediatric burns progress has been made with the development of burn-specific self-report instruments.6–9 To assess the participation/quality of life domain, the EQ-5D instrument is used in patients with burns10 and is undergoing further testing in the Dutch pediatric population. Self-report instruments, however, can be biased by memory or social desirability, which results in other outcomes than if items would be rated by a professional. The availability of a generic instrument concerning the activity domain rated by the professional is needed for children with burns, as it enables a quantitative description of the disabling consequences of burns and comparison of functional outcomes after burns with outcomes in other groups.
In other specific groups of children, several instruments rated by the professional are used, such as the Pediatric Evaluation of Disability Inventory (PEDI),11 the Vineland Adaptive Behavior Scales (VABS II),12 the Gross Motor Function Measure (GMFM),13 and the WeeFIM instrument (“WeeFIM”).14 Of the approximately 15 to 20 instruments regularly used to measure the activity domain, only the WeeFIM instrument and the PEDI are reported by the professional and can be used in young children (0–4 years of age). In children with developmental disabilities and acquired brain injury, high correlations are found between the WeeFIM instrument and the PEDI, indicating that the 2 tests measure similar constructs.15 The advantage the WeeFIM instrument has over the PEDI is that it takes far less time to complete.
Whether the WeeFIM instrument can be added to our standard core set for the measurement and reporting of functional outcome of children with burns depends first of all on the feasibility, reliability, and agreement of the instrument in this population. It is important that measurement instruments be valid and reliable, as they are essential tools for clinicians, therapists, and researchers, not only in identifying children with impairments or disabilities, but also in evaluating development and assessing the efficacy of interventions. In several studies, the WeeFIM instrument has been found to be valid and reliable.16–18 Concepts such as validity, reliability, agreement, and responsiveness, however, are relative, as they are situation-specific and highly dependent on the study population and measurement circumstances.19 The psychometric properties of the WeeFIM instrument have not been studied in the Netherlands, nor in children with burns. Serghiou et al20 used the WeeFIM instrument to track functional outcomes of patients with burns, but they did not study the psychometric properties of the instrument in these patients. Moreover, they misinterpreted the levels of independence among the young age group (6 months–6 years) in their study.20 Therefore, the purpose of this study was to examine the feasibility, reliability, and agreement when the WeeFIM instrument is used in children with burns who have been admitted to Dutch burn centers.
Method
Selection of Participants
In each of the 3 dedicated Dutch burn centers, at least 3 clinicians who had regular contact with children with burns (eg, nurse, nurse practitioner, occupational therapist, child life specialist) were selected as raters of the WeeFIM instrument. Selected raters read the Dutch translation of the WeeFIM manual and practiced assessing children. Experiences were shared with each other, and uncertainties were discussed. In September 2009, the obligatory credentialing WeeFIM examination was taken and passed by all raters, after which 9 raters started administering the WeeFIM instrument.
The instrument was administered to all consecutive children who were between 6 months and 16 years of age and admitted to 1 of the 3 centers for more than 24 hours with acute burns. Inclusion started in September 2009 and lasted 1 year. In one of the centers, due to internal communication problems, it was possible that some children were not assessed based on the extent of their injury. As a result, in the first months of the inclusion period, children with more severe burns were less likely to be included. A total of 16 children were included at that center. In the other 2 centers, 137 children were eligible for the study, and 118 children were included. In these centers, the WeeFIM instrument was not administered for the following reasons: children (and parents) were not communicating in a language the raters could understand (n=2), the child had an electrical injury (n=1), children were transferred to other hospitals (n=3), children (and parents) were not showing up at appointments (n=2), or there were logistical problems (n=11).
As shown in Table 1, a total of 134 children (16 from 1 burn center + 118 from the other 2 burn centers) were included. Furthermore, the table shows that most children admitted to the burn centers were very young and that many children (72%) were lost to follow-up. The fact that many children were unable to complete all assessments at all time points might influence the conclusions that can be drawn from the current data. In the present study, feasibility was partly evaluated by the assessment of inclusion rate. Therefore, inclusion rate was evaluated only on the basis of data gathered in the 2 centers where 137 children were found to be eligible for the study. The other feasibility or reliability measures assessed in this study are not thought to be influenced by the possible bias in the data.
Number of Children Seen at Different Periods Postburn per Age Groupa
Instruments
The WeeFIM instrument.
The WeeFIM instrument was used to examine basic daily living and functional skills in children. The instrument measures the level of functional independence, which has been defined as the child's consistent and usual performance, as well as the level of assistance needed by the child to perform daily living tasks effectively. The WeeFIM instrument was developed for use in children between the ages of 6 months and 7 years (but with application through adolescence) across health, developmental, educational, and community settings.21 The instrument can be administered by either direct observation or interview, can be used by multiple disciplines, and takes approximately 15 minutes to administer.16,22 The instrument is derived from the FIM, which is a functional independence measure for adults. It contains 18 items that cover 3 major domains: 8 items for self-care, 5 items for mobility, and 5 items for cognition. Each item is rated on a 7-point scale ranging from total assistance (1) to complete independence (7). American normative data, based on a sample of more than 500 children in good health and without disability, are available for the total score, as well as scores on the different domains.23 The WeeFIM instrument is reliable in both children with disabilities and those without disabilities.22,24,25 In children with developmental disabilities in the United States, the intraclass correlation coefficients (ICCs) for different subscales were greater than .90.25 Indexes of responsiveness in children with disabilities indicated reliable and statistically significant changes over time.26 Further testing of psychometric properties also has been done, suggesting distinct motor and cognitive scales and an age-specific item hierarchy.27,28
Translational process of the WeeFIM instrument.
Translation and cultural adaptation of the WeeFIM II Clinical Guide14 was done based on the “2-panel method.”29,30 The WeeFIM questionnaire and manual were translated into Dutch by 2 translators with varied but suitable profiles (M.K.N., H.A.R-M.). They informed themselves about the questionnaire (eg, underlying conceptual model, design, content) and subsequently translated the questionnaire and manual. They discussed and resolved inconsistencies, some after consultation with the publisher. Upon completion of the first draft, a person who was informed regarding both content and language abilities (A.S.N.) read through and corrected the whole document. These corrections then were checked against the original American text. In addition, one author (M.K.N.) translated the credentialing examination into Dutch. After 3 months, both authors completed the examination, in which 2 cases had to be scored. One translator (H.A.R-M.) completed the original examination with the American manual. The other author (M.K.N.) scored the translated cases with the Dutch manual. Both authors discussed their item scores, and if there were any differences in scoring, the original text and the Dutch translation were compared. Due to this procedure, at one place in the Dutch manual an “and/or term” was altered. This latest version was used by all raters.
Global Impression of Change.
The Global Impression of Change (GIC), a questionnaire with 2 items, was used to obtain a global impression of change considered important by patients and clinicians. The first item involved the severity of the condition, which could be scored on a 7-point scale from “normal” to “severe.” The second item involved the global change with regard to the start of treatment, ranging on a 7 point-scale from “complete recovery” to “much worse.”31 As an anchor for change, the GIC was rated by both the primary caregiver and the attending clinician. If the child was over the age of 12 years, he or she completed the GIC as well. The GIC was used to provide an external meaning for change.32
Procedure and Design
To test the feasibility of the WeeFIM instrument in Dutch burn centers, it was administered to all consecutive children 2 to 3 weeks postburn. If children were seen for follow-up on clinical indication at approximately 3 and 6 months postburn, they were reassessed with the WeeFIM instrument. Starting in April 2010, each time the WeeFIM instrument had to be administered, 2 WeeFIM raters, if possible, were scheduled for the appointment (n=52) or 2 appointments were planned within 1 week by the same rater (n=7) to get an indication of the interrater and intrarater reliability, respectively. If 2 people administered the WeeFIM instrument, they were in the same room but did not communicate about scores. The raters were randomly selected from those available. The GIC was assessed at 3 and 6 months postburn.
Data Analysis
All analyses were done using SPSS/PASW Statistics version 16 (SPSS Inc, Chicago, Illinois). An alpha level of .05 was adopted. Feasibility was evaluated by the assessment of inclusion rate, the frequency of missing answers per item, and administration time. Inclusion rate was evaluated based only on the information gathered in the 2 centers where no communication problems existed. A learning effect might be reflected in the time needed to administer the WeeFIM instrument. Therefore, weeks passed from the start of the measurements (September 2009) was examined as a predictor for assessment time.
Internal consistency reliability was determined using the Cronbach alpha for all items. Reliability was evaluated by the Spearman correlation between the scores of both raters. This correlation gave an impression of whether 2 raters agreed on ordering. Whether they also agreed on magnitude was indicated by ICCs. The ICC values measured the extent of consensus on use of the instrument by those who administered it. Although ICCs have primarily been used for interval data, they can be applied without distortion to ordinal data when intervals are assumed to be equivalent.33 In the present study, ICCs were calculated for the whole group. As the scores on the WeeFIM instrument are known to have a ceiling effect from the age of 7 years (resulting in high consensus between raters), ICCs also were calculated for the subgroup of children younger than 7 years.
As each child was rated by a different set of staff members randomly selected from those available, the one-way single-measures ICC (1,1) was calculated.34 An ICC between .60 and .69 was interpreted as substantial interrater reliability, an ICC larger than .70 was interpreted as acceptable, and an ICC higher than .80 was interpreted as outstanding.35
It was important to know whether a difference in individual scores was due to random error (eg, due to differences between raters) or that change was larger than expected by chance. For this purpose, the standard error of measurement (SEM) was calculated for the total test score. The SEM is an estimate of the unsystematic variance in an individual's score. The SEM was calculated as the standard deviation of the difference in test scores gathered by 2 raters at the same moment or gathered within a 1-week period. The 95% confidence interval of a single score X equals X ± 1.96 × SEM. As outliers can have a major impact on the standard deviation of the difference scores, the 95% limits of agreement also were determined.36
The minimal or least detectable difference (LDD) represents the minimal individual change that must be observed between 2 measurements before it can be concluded that a systematic change has occurred. Taking .05 as the significance level, the LDD equaled
Role of the Funding Source
This study was supported by a grant from the Dutch Burns Foundation.
Results
Feasibility
From September 2009 until October 2010, a total of 134 children were included. Due to communication problems in 1 of the 3 centers, in only 2 centers did personnel try to include all eligible children from the start of the study. The inclusion rate in those 2 centers was above 80%. In all record forms filled out (n=249), only one missing answer was observed: item 6 (toileting) had not been rated in a child who was fully independent in bladder and bowel management.
Administration time often was missing (31% of the 249 record forms). The average time recorded as needed for the interview and for filling out the WeeFIM record form was 13 minutes (SD=6.8); 83% of the interviews were completed within 15 minutes (range=1–60 minutes; n=170 measurements). Time elapsed since the beginning of the study showed that more experience administering the WeeFIM instrument resulted in less time needed to complete the test (F1,167=9.69, P=.002). Administration time was not influenced by the age of the child (F16,152=1.20, P=.27, not significant). Longer administration times (>40 minutes) were observed if scores were obtained by observation instead of interviewing the primary caregivers (n=4).
Reliability
The Cronbach alpha for internal consistency was .981 over 18 items measured in 134 children a few weeks after the burn accident. Over all measurements, including those at 3 or 6 months postburn, the Cronbach alpha was .984. A value of .80 and higher was considered good for confirmatory purposes.37
The WeeFIM instrument was administered by 2 raters on 52 occasions. The Spearman rank correlation coefficients were .98 (total score), .95 (self-care), .99 (mobility), and .94 (cognition). All ICC values for interrater reliability per item, subscale score, and total score for the total group were above .80 (Tab. 2), which is outstanding.35 For the younger group of children (aged below 7 years), the interrater ICCs were nearly all outstanding, except for eating, transfer to shower or tub, and 3 cognition items in which the lower bound of the 95% confidence interval was below .80.
For 7 children, one rater administered the WeeFIM twice. The Spearman rank coefficients for these ratings were higher than those found for different raters scoring at the same time: .99 (total score), .99 (self-care), .97 (mobility), and .95 (cognition). The intrarater ICCs also were outstanding (>.80).
Agreement
The SEM for the WeeFIM total score was 3.7 points. The 95% confidence interval based on the SEM was (1.96 × SEM) 7.25, which is comparable to the 8 points for the 95% limits of agreement.36 In 25% of all cases (total n=59), WeeFIM total scores were exactly the same.
The Figure shows that difference between observers was not related to the degree of independence (total score). The correlation between raters' differences on the WeeFIM instrument and age in months was .15, which was interpreted as low.
Scatterplot (Blant-Altman plot) of difference in WeeFIM scores between observers and total WeeFIM score.
The LDD on the total score between 2 raters was
Reliability Measures for Test Items of the WeeFIM Instrument: Spearman Rank Correlation Coefficients (rs) With Age in Months and Between the Ratings of 2 Staff Members and Interrater Reliability Intraclass Correlation Coefficients (ICCs) and 95% Confidence Intervals (95% CI) for the Whole Group (n=52) and for Children Younger Than 7 Years of Age (n=34)
In addition to the LDD, the scores on the GIC were used to get an impression concerning the MIC.31 However, filling out the GIC took extra time and effort. Therefore, an interim analysis on the GIC data (n=27) was performed. After 3 or 6 months, change scores on the WeeFIM instrument were not differentiating between groups perceived as more or less ill or recovered. For most children, both parents and clinicians regarded the severity of the illness as normal and the child as much recovered after 3 months. The WeeFIM scores of the children who were perceived as not ill by the clinician changed between −3 and +48 (n=26). The score of the child perceived as minimally ill improved by 30 points and fell within the change scores obtained from the other children (−3 and +48 points). Only 4 parents regarded their child as minimally to moderately ill. The change scores of these children also fell within the range of change obtained by the other children.
Discussion
The goal of this study was to find out whether the WeeFIM instrument is feasible and reliable for use in children with burns admitted to Dutch burn centers. Interviewing a child and the parents is considered feasible and takes approximately 10 to 15 minutes. Moreover, the reliability of the WeeFIM scores is good. A standard measurement error of 4 points has to be taken into account by the clinician who wants to monitor the individual child. This standard measurement error means that the true score of a child with, for example, a total score of 50 lies with a 95% probability between 42 and 58. This child has significantly improved between 2 consecutive measurement occasions if the score changes 11 points or more, thus from 50 to more than 61 points. For the clinician, this LDD of 11 points allows a statistical interpretation of improvement. An indication for the MIC could not be derived from the GIC due to less variance in the observed changes.
After a few weeks in which each rater used the WeeFIM instrument at least once, the instrument was administered, on average, in 10 to 15 minutes. This time interval also is mentioned in the WeeFIM manual and was reported by other authors.16,22,38 This is an important finding, as administration time was one of the criteria to explore in evaluating the use of the WeeFIM instrument instead of the PEDI. Raters found it difficult to report how much time they spent administering the WeeFIM instrument, which resulted in 31% missing data on administration time. Often the raters interviewed the parents or child during the changes of wound dressing and other regular care activities. Under such circumstances, they did not record time, as they felt that the extra time needed was “only some seconds” to write down 18 numbers on the record form. Another aspect of feasibility is the number of missing answers. In our study, only one answer is missing. A high answer rate is to be expected, as the guidelines prescribe that if a child does not perform a functional activity, even though the assessor knows the child can do it, that item should be rated level 1 (total assistance).
The Cronbach alpha shows that the internal consistency of the Dutch translation of the WeeFIM instrument is high when used in Dutch children with burns. The reliability of the instrument expressed in ICC values also is good, even though the raters experienced difficulties scoring 2- and 3-year-old children and the cognitive items in general. We believe that misinterpretation of capability and actual performance, as described by Serghiou et al,20 can easily occur in this age group and these items. The relatively low ICC values for some of the cognitive items in children younger than 7 years of age might reflect the difficulty the WeeFIM raters in the Dutch burn centers experienced. Recently, the WeeFIM 0–3 module for children between 0 and 3 years of age has become available,39 which might solve these problems. Nevertheless, this study shows that it is possible to measure functional outcome in Dutch children with burns in a reliable way with the WeeFIM instrument. Moreover, the differences between raters are not related to the level of independence or age of the child, whereas, as expected, independence levels do correlate with age.
A strength of this study is that high ICC values were found even though the one-way single-measures ICC (1,1) often is lower than all other kinds of ICCs.34 The interrater reliability values confirm that a child considered as very dependent by one clinician often was thought to be very dependent by a colleague as well, and vice versa. The intrarater reliability values were even better. In other countries where the psychometric properties of the WeeFIM instrument have been studied, high ICC values also have been found.38,40 In addition, the interrater reliability values were not based on a fixed set of clinicians. They were based on a set of changing clinicians who were available on the day the researcher wanted the WeeFIM instrument to be administered. Therefore, the results of this study indicate that the WeeFIM instrument's subscale measures can be generalized to other clinicians in the Netherlands who want to use the translated version of the WeeFIM instrument.
A limitation of our study, however, is that only 7 test-retest measurements were available to calculate the intrarater reliability. At the burn centers, it turns out to be easier to schedule 2 different examiners at the same time than the same rater to administer the instrument again within a short period. For the interpretation of the intrarater reliability measurements, we feel that test effects are visible when administering the WeeFIM instrument within a 1-week period. In some cases, for example, the interview leads to another perception of parents about their child's capabilities, which leads to other responses within a few days (memory of the previous interview leads to additions or a more realistic view about the performances of the child). Nevertheless, both intrarater and interrater ICC values were more than acceptable. The intrarater values were better than the interrater values, most likely because variance of tester did not exist. In clinical practice, however, high interrater reliability is especially important, as patients are observed by the available clinician. This study shows that the WeeFIM values are reliable, independent of the rater.
A strength of this study is that we tried to measure the MIC and the LDD. In many studies, the terms “reliability” and “agreement” are used interchangeably. However, they are conceptually different, although both provide information about the quality of measurements.19 Clinical test theory is used to describe the measurement error. Based on the SEM of 3.7 points, a difference between measurements should be at least 11 points before a clinician can state that an individual child has improved. This LDD is lower than but comparable to the 13 points needed to state progress in independence with the FIM instrument that is used in adults.24 The change in scores does not seem to be associated with opinions about the illness or recovery, as expressed in the GIC questionnaire. Clinicians perceive most children with burns as much recovered and not ill after a few months. Parents, however, perceive their children as not totally recovered. Some parents explained to us that scars were present, and they, therefore, perceived their child as not recovered. They hoped that full recovery would imply no lasting signs of burn injury. Regardless of the impression of change, WeeFIM change scores show much variance within a 3-month period. An improvement in WeeFIM scores, however, is normal in children with typical development as functional independence develops. Regular development, recovery, difference in independent behavior due to hospital stay, and measurement error are all factors that can lead to better WeeFIM scores a few months after burn injury. It is important to realize that clinical interpretation is never so straightforward that a clinician can decide based on one test outcome. An experienced clinician observing nonsignificant improvement (less than 11 points) on the WeeFIM instrument can still regard it as “real” improvement when changes in other outcome measures point in the same direction.
In conclusion, the clinicians in Dutch burn centers are able to measure functional independence with the WeeFIM instrument in a reliable way. The WeeFIM instrument is feasible and reliable to be administered during consultation hours. For the evaluation of individual improvement, an LDD of 11 points has to be taken into account. The results are similar to those found in other populations.
Further research on the data gathered in this study is necessary to determine whether a license will be granted by the Uniform Data System for Medical Rehabilitation for use of the WeeFIM instrument within the burn centers. The fact that the WeeFIM instrument can be used in a reliable way in Dutch burn centers does not mean that the test is valid for use in this population. A high level of consistency among items was found, indicating high internal validity. However, whether the instrument and available norms are valid to determine and follow up on the level of independent functioning in the children admitted to Dutch burn centers needs to be examined as well. Nevertheless, the results of the present study are generally favorable for adding the WeeFIM instrument to our standard core set for the measurement and reporting of functional outcome of children with burns.
The Bottom Line
What do we already know about this topic?
Burns might have a major impact on population health, because they frequently occur in young children 0 to 4 years of age. Many of these children have permanent consequences of their injuries from childhood to adolescence.
What new information does this study offer?
This study shows that the WeeFIM instrument, which is frequently used to measure functional independence in children, is a feasible and reliable instrument for use with children with burns.
If you're a caregiver, what might these findings mean for you?
Your child's clinician needs to find an improvement of more than 11 points on the WeeFIM instrument before he or she can state that the child is functioning more independently.
Footnotes
-
Dr Niemeijer, Dr Reinders-Messelink, and Dr Nieuwenhuis provided concept/idea/research design and writing. Ms Disseldorp provided data collection. Dr Niemeijer provided data analysis. Dr Niemeijer and Dr Nieuwenhuis provided project management and participants. Dr Nieuwenhuis provided fund procurement and facilities/equipment. Dr Reinders-Messelink, Ms Disseldorp, and Dr Nieuwenhuis provided consultation (including review of manuscript before submission).
-
The authors thank the following people for their dedication to this study: Margriet van Baar, Anita Boekelaar, Ina Boerma, Anneke Dumans, Helma Hofland, Paula Gieles, Ina van Ingen Schenau, Hennie Schouten, Anneke van de Steenoven, Kitty Stoker, Miranda Venema, Martijn van der Wal, and Jan-Kees Zuiker. They also thank the members of the Centre for Functional Assessment Research of the Uniform Data System for Medical Rehabilitation (UDSmr), State University of New York at Buffalo, for making it possible to use the WeeFIM instrument in the Netherlands and for their feedback.
-
This study was approved by the medical ethical committees of the Maasstad Hospital in Rotterdam, the Red Cross Hospital in Beverwijk, and the Martini Hospital in Groningen.
-
An oral presentation of this research was given at the 14th European Burns Association Congress; September 14–17, 2011; The Hague, the Netherlands.
-
This study was supported by a grant from the Dutch Burns Foundation.
-
The use of the WeeFIM instrument to collect data for this research study was authorized and conducted in accordance with the terms of a special purpose license granted to the licensee by the Uniform Data System for Medical Rehabilitation (UDSmr), a division of U. B. Foundation Activities Inc. The licensee has not been trained by UDSmr in the use of the WeeFIM instrument, and the patient data collected during the course of this research study have not been submitted to or processed by UDSmr. No implication is intended that such data have been or will be subjected to UDSmr's standard data processing procedures or that the data are otherwise comparable to data processed by UDSmr. WeeFIM and FIM are trademarks of UDSmr.
- Received November 17, 2011.
- Accepted March 21, 2012.
- © 2012 American Physical Therapy Association
References
Issue highlights
