Authors: Rhonda M. Williams, PhD Aaron P. Turner, PhD Monica Green, PhD Daniel C. Norvell, PhD Alison W. Henderson, PhD Kevin N. Hakimi, MD Donna Jo Blake, MD Joseph M. Czerniecki, MD
Outcomes
ORIGINAL RESEARCH ARTICLE
Affiliations: From the VA Puget Sound Health Care System (RMW, APT, MG, AWH, KNH, JMC); Department of Rehabilitation Medicine, University of Washington, Seattle (RMW, APT, KNH, JMC); Spectrum Research, Inc, Tacoma, Washington (DCN); VA Eastern Colorado Health Care System (DJB); and University of Colorado, Denver (DJB).
Relationship Between Cognition and Functional Outcomes After Dysvascular Lower Extremity Amputation A Prospective Study
Correspondence: All correspondence and requests for reprints should be addressed to: Rhonda M. Williams, PhD, VA Puget Sound Health Care System, RCS-117, 1660 S. Columbian Way, Seattle, WA 98108.
Disclosures:
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal_s Web site (www.ajpmr.com).
www.ajpmr.com
cognitive screen and four neuropsychologic tests administered at both 6 wks and 4 mos after amputation and five functional outcomes measured 12 mos after lower extremity amputation.
Design: This study includes a prospective cohort from four medical centers. Participants were primarily male Veterans experiencing their first lower extremity amputation as a result of complications of diabetes mellitus or peripheral arterial disease. Of those eligible, 87 (64%) enrolled; 75 (86%) were retained at 12 mos. Measures included demographic/health information, four neuropsychologic measures, the Locomotor Capability IndexY5, the Gronigen Activity Restriction Scale, prosthetic use, community participation, and social integration.
Results: Better performance on the Short Portable Mental Status Question-
Editor’s Note:
DOI: 10.1097/PHM.0000000000000235
Williams RM, Turner AP, Green M, Norvell DC, Henderson AW, Hakimi KN, Blake DJ, Czerniecki JM: Relationship between cognition and functional outcomes after dysvascular lower extremity amputation: a prospective study. Am J Phys Med Rehabil 2015;94:707Y717.
Objective: The aim of this study was to examine associations between a
Supported by the United States Department of Veterans Affairs, Office of Research and Development, Rehabilitation Research and Development (Merit Review A41241 Joseph Czerniecki, principal investigator, and Career Development Award B4927W Aaron P. Turner, principal investigator). Financial disclosure statements have been obtained, and no conflicts of interest have been reported by the authors or by any individuals in control of the content of this article.
0894-9115/15/9409-0707 American Journal of Physical Medicine & Rehabilitation Copyright * 2014 Wolters Kluwer Health, Inc. All rights reserved.
ABSTRACT
naire at 4 mos was associated with greater 12-mo mobility and social integration. Better attention and working memory abilities 6 wks after amputation were associated with increased 12-mo prosthetic wear; and at 4 mos after amputation, with greater 12-mo mobility. Better verbal memory at 6 wks was associated with greater 12-mo social integration and community participation as well as increased prosthetic wear.
Conclusions: These findings highlight the potential value in including a brief, formal cognitive assessment in addition to a general mental status screen. Specific domains of cognitive function are differentially associated with functional outcomes and may inform amputation rehabilitation decisions. Key Words:
Cognition, Amputation, Outcome Assessment, Neuropsychologic Tests
Cognition and Postamputation Outcomes Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
707
A
daptation to limb loss in the context of chronic disease represents an ongoing process. After acute surgical recovery, individuals must avoid or manage residual limb wounds and revision surgeries, maintain the contralateral limb,1 learn to use gait aids or prosthetic equipment,2 maintain independent living,3 as well as manage their chronic disease. Vascular problems, particularly related to diabetes mellitus and peripheral arterial disease, are the most common etiology for lower extremity amputations (LEAs), accounting for 82% of cases annually.4 Because diabetes mellitus and peripheral arterial disease themselves, as well as factors with which they are often associated (i.e., pain, anxiety, medications), can adversely impact cognition, individuals undergoing amputation may be at risk for cognitive impairment.5,6 A recent review indicated that cognitive impairment was more prevalent among individuals with amputations than in the general population and was associated with poorer mobility, prosthetic use, and independent living outcomes.7 Although, historically, the most frequent method for quantifying cognition after amputation is noting the presence or absence of dementia,7Y9 several recent studies have used more specific neuropsychologic tests to evaluate the association between cognition and outcomes such as mobility, prosthetic use, and activities of daily living (ADL) independence. O’Neill and Evans10 found that visual-spatial learning and executive function abilities assessed 4 wks after amputation were associated with 8-mo mobility and hours of prosthetic wear, respectively. In contrast, Schoppen et al.11 found that neither a general cognitive screen, verbal memory test, nor executive function test administered 2 and 6 wks after amputation were predictive of mobility scores or prosthetic use 1 yr after amputation, although they were associated with ADLs at 1 yr after surgery. The authors’ previous work in this population suggests that cognition improves between before amputation and 6 wks after amputation and then stabilizes between 6 wks and 4 mos after amputation.12 Taken together, these findings suggest that the impact of cognition on outcome may vary depending on the timing of cognitive assessment, the domain of cognition assessed, and the specific outcome of interest. There is a need for prospective research to explore relationships between cognition and outcomes after amputation; such knowledge could inform prosthetic, therapy, and disposition decisions. Extant studies have been characterized by small sample sizes and methodological variance, particularly in selection of cognitive measures and timing
708
Williams et al.
of assessments.7 In addition, outcomes are often limited to mobility, ADLs, and prosthetic wear; little is known about the impact of cognition on social and community participation among individuals with LEA, although cognitive impairment has been associated with diminished social functioning in other populations with chronic disease,13 and social and community involvement is known to be an important goal as well as a factor that may bolster social support and mitigate the stress of adapting to limb loss among those with recent LEA.14Y16 The authors’ broad study goal was to identify useful ways to assess cognition that are associated with functional outcomes and can inform rehabilitation decision making. The first specific aim was to examine whether a general mental status (GMS) screen administered at both 6 wks and 4 mos after amputation is associated with five indicators of function at 12 mos after amputation: mobility, prosthetic use and ADLs, as well as novel indicators of community participation and social integration. The second aim was to determine whether the addition of four brief neuropsychologic tests (Digit Span, List Learning, List Recall, and Semantic Fluency) accounts for unique variance in functional outcomes above and beyond the GMS screen. The third aim was to examine whether cognition has different associations with 12-mo functional outcomes when it is assessed at 6 wks or 4 mos after amputation. These objectives address several gaps in the literature by using a range of neuropsychologic tests in addition to a GMS screen, an expanded range of outcome measures, and two possible assessment time points.
METHODS Study Design This study is part of a larger multisite prospective cohort study of individuals undergoing LEA as a result of peripheral arterial disease or diabetes mellitus at two VA medical centers, a university hospital, and a level I trauma center. Participants were assessed via in-person or telephone interview at 6 wks, 4 mos, and 12 mos after amputation. Study procedures were approved by the human subjects review boards at each study site.17,18
Participants Subjects were considered eligible if (1) they were 18 yrs or older; (2) they were awaiting (or underwent in the last 6 wks) a first major LEA, defined as transmetatarsal level or higher; and (3) the primary cause of amputation was diabetes mellitus or peripheral arterial disease. Subjects
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
were excluded if they had any of the following: (1) cognitive or language impairment that would preclude consent or participation, defined by more than four errors on the Short Portable Mental Status Questionnaire (SPMSQ)19; (2) previous amputations; (3) nonambulatory status for reasons unrelated to impairment of the extremity awaiting amputation; or (4) planned bilateral amputation. Of 239 individuals screened between 2005 and 2008, a total of 136 (57%) met study criteria and 87 (64%) enrolled. The primary reason for exclusion was previous amputation (n = 34, 38%); these individuals may also have screen failure of the SPMSQ, but this was not assessed once a candidate was excluded for other reasons. An additional 23 (22%) were excluded solely because of dementia diagnosis or failure of the SPMSQ. Six participants (7%) died, and 6 (7%) were lost to follow-up; 75 (86%) remained enrolled at the 12-mo follow-up (see Table 1 for participant characteristics).
Covariates Assessed at the Time of Neuropsychologic Testing Demographics Covariates were dichotomously coded as follows: (a) sex, male (0) vs. female (1); (b) race, white (1) vs. nonwhite (0); and (c) transtibial or lower amputation level (0) vs. transfemoral amputation level (1). Age was reported in years (Table 1).
Major Depressive Episode Because depressed mood can impact cognitive performance,20 presence of a major depressive episode (MDE) was assessed with the Depression module of the Patient Health Questionnaire,21 a nine-item screening measure. Participants who endorsed five
or more symptoms (one of which was depressed mood or anhedonia) on Bmore days than not[ in the past 2 wks were coded as having MDE (1), and those who did not were coded as having no MDE (0)21 (Table 2). Presence of MDE was assessed concurrently with the neuropsychologic assessment (i.e., at both 6 wks and 4 mos).
Self-rated Health Self-rated health (SRH) was assessed via a single item, with response options ranging from 1 (Bvery poor[) to 5 (Bvery good[)22 (Table 2). SRH was assessed concurrently with the neuropsychologic assessment (at both 6 wks and 4 mos).
Cognitive Measures General Mental Status Participants completed the ten-item SPMSQ,19 an assessment of orientation and GMS at each study time point, because GMS is a common method of measuring cognition in published studies.23,24 Individuals with more than four errors on the SPMSQ at the time of initial screening were excluded from this study because test norms demonstrate that more than four errors indicates moderate cognitive impairment.19
Neuropsychological Subtests Subtests from two neuropsychologic batteriesV Semantic Fluency, List Learning, and List Recall from the Repeatable Battery for the Assessment of Neuropsychological Status25 and Digit Span from the Wechsler Adult Intelligence Scale, Third Edition,26 were administered at 6 wks and 4 mos after surgery. Tests were chosen for their brevity; for their ease of administration; and to represent varying aspects of
TABLE 1 Sociodemographic, health, and outcome data among study participants with complete data at 12 mos (N = 75) Sociodemographic/Health (Assessed at 6 wks) Age, mean (SD), yrs Female, n (%) Race, n (%) White Other Amputation level, n (%) Below-knee (transmetatarsal and transtibial) Above-knee (transfemoral)
Range 61.54 (8.51) 6 (8) 61 (81.3) 14 (18.7) 68 (90.7) 7 (9.3)
Outcomes (Assessed at 12 mos)
Range
LCI-5, mean (SD) Hours of prosthetic wear per day, mean (SD) GARS, mean (SD) Community participation, mean (SD) Social integration, mean (SD)
www.ajpmr.com
47Y83
40.85 (15.54) 9.17 (5.49) 28.88 (12.97) 6.27 (2.28) 6.26 (2.68)
0Y56 0Y18 18Y72 0Y10 0Y12
Cognition and Postamputation Outcomes Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
709
710
Williams et al.
Note: summary based on final regression models (see Tables 4 and 5). For example, Digit Span and List Recall are associated with outcomes shown above and beyond variance explained by SPMSQ. Although not significant in any of the final models, executive functioning, the ability to initiate problem-solving behaviors, cognitive flexibility, and response inhibition, was assessed with the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) Semantic Fluency subtest. a P G 0.05. WAIS-III, Wechsler Adult Intelligence Scale, Third Edition.
Auditory-verbal memory; RBANS List Recall
a
a
a
a
a
Mobility Social integration Mobility Prosthetic use (hours of wear per day) Community participation Social integration Prosthetic use (hours of wear per day) a
4 mos
a
Ability to store and manipulate multiple pieces of information for a short time (i.e., G2 mins) The ability to retrieve verbal information presented orally after a delay
Mobility was assessed using the Locomotor Capability Index (LCI-5).28 Possible scores range from 0 to 56, with higher scores representing higher functional mobility.
Attention/working memory; WAIS-III Digit Span
Mobility
Orientation and basic awareness of current information
Outcomes Assessed 12 mos After Amputation
GMS; SPMSQ
cognitive functioning examined in other published studies, including auditory-verbal memory, attention, and executive function.12 To improve access, some participants (n = 24 [28%] at 6 wks; n = 26 [33%] at 4 mos) were assessed via telephone.27 Those tested via telephone reported higher levels of education than those tested in person (13.9 vs. 12.9 yrs, P G 0.05). There were no other significant differences between the two groups on any of the standardized neuropsychologic test scores at either time point, nor were there any significant differences on any other health or demographic factors. For each neuropsychologic test, raw scores were converted to standardized z-scores using norms that adjust for age and, in the cases of Wechsler Adult Intelligence Scale, Third Edition Digit Span, education and sex. In all cases, a z-score of 0 indicates function at the 50th percentile, whereas positive/ greater z-scores indicate better cognitive performance and negative/lower z-scores indicate poorer function. Each neuropsychologic test that was significant in the final models and the domain of cognitive function it assesses is described in Table 3. It should be noted that the Repeatable Battery for the Assessment of Neuropsychological Status offers two alternate forms to allow for repeated assessment; Form B was used at 6 wks and Form A was used at 4 mos. Details about test administration and test performance as well as data from visual-spatial neuropsychologic tests included in the broader study are included elsewhere.12
Test Administration Time
Median SRH score at 6 wks: 3; at 4 mos: 4.
TABLE 3 Summary of associations between cognitive domains and outcomes
a
6 wks
Met criteria for MDE, n (%) 15 (20.5) 15 (20.5) SRHa Very poor 2 (2.7%) 1 (1.4%) Poor 9 (12.3%) 9 (12.3%) Fair 29 (39.7%) 24 (32.9%) Good 27 (37.0%) 32 (43.8%) Very good 6 (8.2%) 7 (9.6%)
Test Description
4 mos
Cognitive Domain and Neuropsychologic Test Name
6 wks
12-mo Outcomes Associated with Domain
TABLE 2 Self-rated depression and health among study participants with complete data at 12 mos (N = 75)
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Prosthetic Use Participants who had been fit for a prosthetic limb were asked the mean number of hours per day that they wore their prosthesis and the number of hours they walked with their prosthesis.
Activities of Daily Living ADLs were assessed using the Gronigen Activity Restriction Scale (GARS).29 Total possible scores can range from 18 to 72, with higher scores indicating greater levels of impairment.29
Community Participation Because existing measures of community participation confound factors such as ability and desire to participate, necessity and availability of assistance, as well as actual behavioral engagement,30,31 the study staff (five rehabilitation professionals) adapted a well validated existing measure for this study population. The authors used three items from the Community Integration Questionnaire32 and created a fourth similarly formatted item to assess how often an individual engaged in community-based activities (e.g., shopping, leisure activities, community activities, and traveling) outside the home in the past month. Each item was scored categorically (e.g., 0, never; 1, one to four times; and 2, five or more times). Total possible sum scores ranged from 0 to 10, with higher scores indicating greater community participation. Although the modifications to this measure that were made for this study are not validated, initial examination suggests that this measure has construct validity; it is correlated at r = 0.45, P G 0.01, with concurrent (i.e., 12-mo) social integration, at r = 0.38, P G 0.01 with the LCI, and at r = j0.42, P G 0.01 with the GARS. The Cronbach > for this scale was 0.74, suggesting adequate internal consistency. Items for this measure and the social integration measure (below) are included in Appendix A (http://links.lww.com/PHM/A86).
Social Integration Social integration measured the extent to which participants were an active part of their social network, including frequency of social interactions and breadth of social network. The authors used one item from the Community Integration Questionnaire,32 used two items from the social integration subscale of the Craig Handicap Assessment and Reporting Technique,33 and developed two new items. The resulting five-item scale has possible sum scores ranging from 0 to 12, with higher scores indicating greater social participation. The www.ajpmr.com
total score is correlated at r = 0.43, P G 0.01 with concurrent social support, suggesting good construct validity. The Cronbach > for this scale was 0.76.
Data Analysis First, Pearson univariate correlations were computed to identify associations between neuropsychologic tests and each outcome measure. There were no significant univariate correlations between 6-wk neuropsychologic tests and ADLs as well as both 6-wk and 4-mo neuropsychologic tests and hours of walking with a prosthesis; hence, these models were not computed. Further, for parsimony, only neuropsychologic tests that were correlated at a univariate level with the outcome of interest at P e 0.10 were included. Multiple linear regression models were computed to examine the unique associations between neuropsychologic measures and outcome measures, controlling for health and demographic factors. Separate regression analyses were performed for each of the outcome variables with forced entry of the blocks of covariates. The first block in the regression models included demographic characteristics (age, sex, and race), the second included health factors (amputation level, SRH, and MDE), the third included GMS (SPMSQ), and the fourth included any neuropsychologic tests with a univariate correlation significant at P G 0.10. With the exception of specific neuropsychologic tests, all other variables (including SPMSQ) were entered into each model regardless of significance.
RESULTS Neuropsychologic Test Scores Overall, the participants in this study scored in the low average to average range on the neuropsychologic tests administered; the frequency of scores falling into the impaired range (less than the second percentile) ranged from 1% to 13%. Greater details about test scores are published elsewhere.12 The mean scores on the Reliable Digit Span, an embedded measure of task engagement, were consistently higher than the threshold commonly taken to indicate adequate effort (97), supporting the validity of these data.34
Mobility The mean LCI-5 mobility scores are presented in Table 2 and are discussed at greater length elsewhere.17 After controlling for health and demographic factors, neither GMS nor the specific neuropsychologic tests assessed at 6 wks after amputation were associated with mobility at 12 mos Cognition and Postamputation Outcomes
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
711
(Table 4). However, when these cognitive domains were measured at 4 mos after surgery, better GMS (A = 0.25, P e 0.01; Table 5) and higher scores for Digit Span, a measure of attention (A = 0.29, P e 0.01), were associated with greater mobility.
Prosthetic Use By 12 mos after amputation, 69 participants (92.0%) were fitted for a prosthesis.18 Among those fitted with a prosthesis, higher Digit Span (A = 0.35, P e 0.05) and List Recall (A = 0.35, P e 0.05) scores assessed at 6 wks were associated with increased hours of prosthetic wear at 12 mos. Specifically,
improving the z-score by 1 SD on either cognitive test was associated with wearing a prosthetic for approximately 2 additional hours per day. Although Digit Span and List Recall assessed at 4 mos were significantly and positively correlated with prosthetic use at a univariate level, these were not significantly associated with hours of prosthetic wear at 12 mos when controlling for other factors.
Activities of Daily Living Neither GMS nor the neuropsychologic tests administered at 6 wks were significantly associated
TABLE 4 Multiple regression analyses of 6-wk neuropsychologic test scores on 12-mo outcomes Step and Variables Mobility (LCI-5) 1. Demographics
Age Sex Ethnicity 2. Health factors Amputation level SRH at 6 wks MDE at 6 wks 3. Mental status at 6 wks SPMSQ 4. Cognition at 6 wks Digit Span Prosthetic use (hours of prosthetic wear) 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 6 wks MDE at 6 wks 3. Mental status at 6 wks SPMSQ 4. Cognition at 6 wks Digit Span List Recall Community participation 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 6 wks MDE at 6 wks 3. Mental status at 6 wks SPMSQ 4. Cognition at 6 wks List Recall Social integration 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 6 wks MDE at 6 wks 3. Mental status at 6 wks SPMSQ 4. Cognition at 6 wks List Recall
A
$ R2
$F
j0.16 j0.40a 0.25b j0.17 0.12 j0.19c j0.03 0.18
0.22
6.52a
0.07
2.19c
0.00 0.03
0.08 2.69
0.21 0.00 j0.07 j0.39a j0.22 j0.14 0.01 0.35b 0.35b
0.05
1.23
0.19
3.82b
0.00 0.16
0.01 3.75b
0.03
0.56
0.04
0.80
0.01 0.09
0.84 5.87b
0.07
1.42
0.06
1.22
0.04 0.10
2.78c 7.16b
0.03 0.01 0.17 j0.10 0.10 j0.14 j0.13 0.35b 0.11 j0.01 0.24c j0.01 0.23c j0.07 0.22c 0.40a
Beta values are standardized and show the A for the step in which that variable was added. There were no significant (P e 0.10) univariate correlations between any neuropsychologic test scores and GARS or hours walking with prosthetic scores. Thus, the models for ADLs and walking with prosthesis are not shown. a P e 0.01. b P e 0.05. c P e 0.10.
712
Williams et al.
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
TABLE 5 Multiple regression analyses of 4-mo neuropsychologic test scores on 12-mo outcomes Step and Variables Mobility (LCI-5) 1. Demographics
Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos Digit Span List Recall Semantic Fluency Prosthetic use (hours of prosthetic wear) 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos Digit Span List Recall ADLs (GARS) 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos Digit Span List Recall Community participation 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos Digit Span List Recall Social integration 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos List Recall
A
$ R2
$F
j0.11 j0.30a 0.25a j0.23a 0.37b j0.14 0.25a 0.29a j0.14 0.03
0.14
3.55a
0.27
9.52b
0.06 0.08
6.80b 3.25a
0.22 0.20 j0.04 j0.32a j0.36a j0.11 0.13 0.16 0.05
0.08
1.56
0.15
3.20a
0.02 0.02
0.95 0.73
0.30
0.47
0.16
4.21b
0.04 0.03
2.74c 1.20
j0.01 0.02 0.16 j0.12 0.36a j0.22 0.08 0.11 0.21
0.03
0.57
0.24
6.19b
0.01 0.05
0.43 1.92
0.13 0.01 0.25a j0.07 0.39b j0.05 0.25a 0.18
0.07
1.58
0.16
4.02b
0.06 0.02
4.69a 2.01
j0.12 0.04 j0.07 0.18 j0.27a 0.13 j0.19c j0.14 j0.10
Beta values are standardized and show the A for the step in which that variable was added. There were no significant (P e 0.10) univariate correlations between any neuropsychologic test scores and hours of walking with prosthetic scores. Thus, the model for walking with prosthesis is not shown. a P e 0.05. b P e 0.01. c P e 0.10.
with outcome measures at 12 mos, and hence, this regression model was not reported. Although Digit Span and List Recall assessed at 4 mos were correlated www.ajpmr.com
at a univariate level with 12-mo ADLs, these were not significantly associated with 12-mo ADLs after controlling for other factors in the model. Cognition and Postamputation Outcomes
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
713
Community Participation and Social Integration Higher GMS at 4 mos was associated with greater social integration (A = 0.25, P e 0.05). Higher List Recall scores at 6 wks were significantly associated with community participation (A = 0.35, P e 0.05) and social integration (A = 0.37, P e 0.01) at 12 mos after surgery.
DISCUSSION This study had several aims. First, it examined whether a brief screening of GMS after amputation would be associated with five 12-mo outcomes important to individuals with limb loss. Second, it examined whether the addition of four neuropsychologic tests provided additional insight into these outcomes above and beyond a brief screening. Finally, it examined whether the timing of assessment (6 wks vs. 4 mos) impacted the associations with outcomes. A GMS screen seems to be a valuable brief assessment and a useful predictor. Consistent with the literature, higher scores on a GMS screen were associated with increased mobility.9,35 The authors were able to extend previous findings to show that GMS is also associated with an individual’s engagement in a social network, which may have implications for an individual’s integration back into important relationships after rehabilitation. Support was also found for adding several brief neuropsychologic tests above and beyond a GMS screen. For example, social and community outcomes as well as hours of prosthetic wear were associated with verbal memory, an unsurprising finding given that these outcomes require the ability to learn and retain information as well as recognize and solve problems. For example, establishing and maintaining social relationships involve attending to and tracking pertinent information about another individual. Similarly, the donning and doffing of a prosthesis involves recalling several steps. That this was apparent in a sample in which individuals who failed the SPMSQ screen were excluded and mean memory performance at 6 wks fell in the average range suggests that memory might have a greater effect in a more diverse (i.e., impaired) sample. The prediction of mobility was improved by adding a measure of working memory and attention. This is also congruent with expectations because the ability to attend to limb positioning and use of gait aids would be expected to directly impact mobility. Schoppen et al. (2003) found verbal memory (measured by the 15-word test) to be uniquely
714
Williams et al.
associated with ADLs, a finding that was not replicated in this study. Although the measure of verbal memory in this study (List Recall) was similar to that of Schoppen et al. and significantly associated with ADLs at 12 mos at a univariate level, after controlling for health and demographic information, it was no longer significantly associated with ADLs. Verbal memory should plausibly contribute to success in ADLs because remembering a series of steps or instructions is critical in a variety of daily living tasks (e.g., cooking a meal). The sample in this study generally scored in the average range on verbal memory tasks. Schoppen et al. report that 90% of participants in their study scored at the 50th percentile or lower, but they do not describe the scores in further detail. Therefore, it may be that verbal memory is related to ADLs only when in an impaired range. In a related vein, it is possible that the verbal memory scores in this study are inflated because of practice effects or because of reduced time between the learning and recall trials. The fact that the authors did not find an association between executive function and prosthetic use, whereas O’Neill and Evans (2009) did, may be explained by differences in the timing of assessment (e.g., these authors assessed cognition at 2Y6 wks and prosthetic use at 8 mos) and by sampling differences (they include amputees with mixed etiologies). Although beyond the scope of this article, in a secondary analysis of a smaller sample of individuals fitted with a prosthesis by 4 mos, several neuropsychologic tests (Digit Span and Semantic Fluency) were associated with reported hours of walking with a prosthesis at 12 mos. Thus, cognition may play more of a role among those fitted earlier for a prosthesis and among those who ambulate more with their prosthesis (instead of wearing primarily for aesthetic purposes). This also highlights the importance of differentiating between wearing a prosthetic vs. ambulating with a prosthetic, a difference that is often not clearly delineated and not always consistent across studies.9,10,24,36 Future studies may help demarcate between prosthetic wear and prosthetic use in ambulation and better parse out predictors of each outcome. The impact of the timing of assessment on functional outcomes was multifarious. The authors cannot state that cognition should be assessed exclusively at 6 wks or at 4 mos because the impact of cognition varied with time assessed, even though cognitive performance was comparable at these two time points. For example, verbal memory assessed at 6 wks was significantly associated with prosthetic
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
wear, community participation, and social integration, but this association is weaker when verbal memory was assessed at 4 mos. It may be that the growing importance of the covariate SRH (not significant in any of the 6-wk models but significant in all the 4-mo models) overshadows the contributions of verbal memory on these outcomes. Similarly, associations between cognition and mobility also seem to shift. Although there is no obvious variable included that might mask the role of cognition in mobility at 6 wks, it may be that other more acute factors (i.e., pain, acclimation to discharge environment) may have resolved or evolved by 4 mos, allowing for more subtle effects of cognition to be visible. Thus, determining when to assess cognition may vary depending on the outcome of interest. Taken together, these findings suggest that, within a population of individuals with grossly intact overall cognitive functioning and undergoing an initial major LEA, aspects of attention and verbal memory are differentially associated with outcomes. Subtle impairment in cognition does not seem to impede ADL function after amputation, although, again, this may be an artifact of the relatively cognitively intact population of this study. However, even subtle impairment in the cognitive domains does seem to affect one’s ability to participate in meaningful social relationships and community activities as well as one’s ability to regain mobility or wear a prosthesis. The findings suggest that there is value in including more nuanced neuropsychologic measures in addition to a coarse cognitive screen, particularly if one is interested in outcomes other than mobility or ADL independence. The findings from this study have a variety of implications for rehabilitation care providers. Helping patients improve their focus and attention (by decreasing distractions and helping prioritize) may be associated with improved mobility, although future studies would be necessary to demonstrate that clinical intervention leads to improvement. Providing patients with compensatory memory skills may facilitate prosthesis use as well as maintenance of community participation and social integration, outcomes likely key in maintaining quality-of-life. Specifically, a rehabilitation team would be well served using repeatable/alternate versions of Digit Span and List Learning/List Recall subtests in addition to the SPMSQ at several time points. In addition, a visual attention and memory task may also be informative, particularly in a sample that has more varied cognitive abilities. There are a number of limitations that should be taken into consideration. To maximize retention, www.ajpmr.com
it was necessary to do a portion of assessments by telephone. This resulted in significant missing data for measures requiring motor responses and also shortened time delays on the verbal memory recall task, which may have resulted in nonstandardized administration and artificially elevated scores on the delayed auditory-verbal memory tasks. This also prevented the inclusion of visual-spatial tests in the analyses because these tasks could not be completed via telephone; this rendered the authors unable to compare the results of this study with those studies that included visual-spatial tasks. However, flexibility in administration allowed the authors to collect a larger sample size. The SPMSQ was used as both an exclusion criterion (to identify a sample with the potential for rehabilitation) and a factor in regression models. Variance in SPMSQ scores was, therefore, necessarily limited, reducing the ability to examine its associations with cognition across a full range of scores. In addition, the sample included only individuals undergoing their first amputation and included very few individuals with transfemoral amputations. It is probable that the role of cognition would become more important among those with amputation-related complications (e.g., revisions and contralateral amputations) or with higher amputation levels.37,38 Finally, the measures of community participation and social integration are not validated. These measures used in this study are based on established measures, with items added and/or modified to tailor the content to a new amputee population. Future studies are needed to further refine/validate these instruments, and ideally, they would include validated instruments against which to compare the novel tools developed here. In conclusion, this study suggests that both a GMS screen and brief neuropsychologic tests provide information that is useful in understanding 1-yr outcomes. Future directions for research include extending this study to a larger, more diverse sample. This would help to identify critical scores on neuropsychologic measures to determine whether there are useful cutoff points that are associated with functional outcomes. Such knowledge may be incorporated into a decision tree or prediction model when considering rehabilitation decisions. In addition, research can be extended to include selfmanagement outcomes, such as wound care and coping with complications. REFERENCES 1. Cruz CP, Eidt JF, Capps C, et al: Major lower extremity amputations at a Veterans Affairs hospital. Am J Surg 2003;186:449Y54
Cognition and Postamputation Outcomes Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
715
2. Marzoug EA, Landham TL, Dance C, et al: Better practical evaluation for lower limb amputees. Disabil Rehabil 2003;25:1071Y4 3. Dillingham TR, Yacub JN, Pezzin LE: Determinants of postacute care discharge destination after dysvascular lower limb amputation. PM R 2011;3:336Y44 4. Ziegler-Graham K, MacKenzie EJ, Ephraim PL, et al: Estimating the prevalence of limb loss in the United States: 2005 to 2050. Arch Phys Med Rehabil 2008;89: 422Y9 5. Waldstein SR, Tankard CF, Maier KJ, et al: Peripheral arterial disease and cognitive function. Psychosom Med 2003;65:757Y63 6. Phillips NA: Thinking on your feet: A neuropsychological review of peripheral vascular disease, in Elias SRWMF (ed): Neuropsychology of Cardiovascular Disease. Mahwah, NJ, Erlbaum, 2001, pp 121Y42 7. Coffey L, O’Keeffe F, Gallagher P, et al: Cognitive functioning in persons with lower limb amputations: A review. Disabil Rehabil 2012;34:1950Y64 8. Larner S, van Ross E, Hale C: Do psychological measures predict the ability of lower limb amputees to learn to use a prosthesis? Clin Rehabil 2003;17: 493Y8 9. Taylor SM, Kalbaugh CA, Blackhurst DW, et al: Preoperative clinical factors predict postoperative functional outcomes after major lower limb amputation: An analysis of 553 consecutive patients. J Vasc Surg 2005;42:227Y35 10. O’Neill BF, Evans JJ: Memory and executive function predict mobility rehabilitation outcome after lower-limb amputation. Disabil Rehabil 2009;31: 1083Y91 11. Schoppen T, Boonstra A, Groothoff JW, et al: Physical, mental, and social predictors of functional outcome in unilateral lower-limb amputees. Arch Phys Med Rehabil 2003;84:803Y11 12. Williams RM, Turner AP, Green M, et al: Changes in cognitive function from presurgery to 4 months postsurgery in individuals undergoing dysvascular amputation. Arch Phys Med Rehabil 2014;95:663Y9 13. Liddle PF: Cognitive impairment in schizophrenia: Its impact on social functioning. Acta Psychiatr Scand Suppl 2000;400:11Y6 14. Gallagher P, Maclachlan M: Adjustment to an artificial limb: A qualitative perspective. J Health Psychol 2001;6:85Y100 15. Rybarczyk B, Nicholas JJ, Nyenhuis D: Coping with leg amputation: Integrating research and clinical practice. Rehabil Psychol 1997;42:241Y56 16. Horgan O, MacLachlan M: Psychosocial adjustment to lower-limb amputation: A review. Disabil Rehabil 2004;26:837Y50 17. Norvell DC, Turner AP, Williams RM, et al: Defining successful mobility after lower extremity amputation for complications of peripheral vascular disease and diabetes. J Vasc Surg 2012;54:412Y9
716
Williams et al.
18. Webster JB, Hakimi KN, Williams RM, et al: Prosthetic fitting, use, and satisfaction following lowerlimb amputation: A prospective study. J Rehabil Res Dev 2012;49:1493Y504 19. Pfeiffer E: A Short Portable Mental Status Questionnaire for the assessment of organic brain deficit in elderly patients. J Am Geriatr Soc 1975;23: 433Y41 20. McDermott LM, Ebmeier KP: A meta-analysis of depression severity and cognitive function. J Affect Disord 2009;119:1Y8 21. Kroenke K, Spitzer RL, Williams JB: The PHQ-9: Validity of a brief depression severity measure. J Gen Intern Med 2001;16:606Y13 22. Verbrugge LM, Merrill SS, Liu X: Measuring disability with parsimony. Disabil Rehabil 1999;21:295Y306 23. Remes L, Isoaho R, Vahlberg T, et al: Predictors for institutionalization and prosthetic ambulation after major lower extremity amputation during an eight-year follow-up. Aging Clin Exp Res 2009;21: 129Y35 24. Bilodeau S, Hebert R, Desrosiers J: Lower limb prosthesis utilisation by elderly amputees. Prosthet Orthot Int 2000;24:126Y32 25. Randolph C, Tierney MC, Mohr E, et al: The Repeatable Battery for the Assessment of Neuropsychological Status (RBANS): Preliminary clinical validity. J Clin Exp Neuropsychol 1998;20:310Y9 26. Wechsler D: WAIS-III: Wechsler Adult Intelligence Scale. San Antonio, TX, The Psychological Corp, 1997 27. Fong TG, Fearing MA, Jones RN, et al: Telephone interview for cognitive status: Creating a crosswalk with the Mini-Mental State Examination. Alzheimers Dement 2009;5:492Y7 28. Grise MC, Gauthier-Gagnon C, Martineau GG: Prosthetic profile of people with lower extremity amputation: Conception and design of a follow-up questionnaire. Arch Phys Med Rehabil 1993;74: 862Y70 29. Kempen GI, Miedema I, Ormel J, et al: The assessment of disability with the Groningen Activity Restriction Scale. Conceptual framework and psychometric properties. Soc Sci Med 1996;43:1601Y10 30. Dijkers MP, Whiteneck G, El-Jaroudi R: Measures of social outcomes in disability research. Arch Phys Med Rehabil 2000;81(suppl):S63Y80 31. Zhang L, Abreu BC, Gonzales V, et al: Comparison of the Community Integration Questionnaire, the Craig Handicap Assessment and Reporting Technique, and the Disability Rating Scale in traumatic brain injury. J Head Trauma Rehabil 2002;17: 497Y509 32. Willer B, Allen LR: Community integration and barriers to integration for individuals with brain injury, in Finlayson M, Garner S (eds): Brain Injury Rehabilitation: Clinical Considerations. Baltimore, MD, Wiliams & Wilkins, 1993, pp 355Y75
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
33. Whiteneck GG, Charlifue SW, Gerhart KA, et al: Quantifying handicap: A new measure of long-term rehabilitation outcomes. Arch Phys Med Rehabil 1992;73:519Y26
36. Pohjolainen T, Alaranta H, Karkkainen M: Prosthetic use and functional and social outcome following major lower limb amputation. Prosthet Orthot Int 1990;14:75Y9
34. Strauss E, Sherman EMS, Spreen O: A Compendium of Neuropsychological Tests: Administration, Norms, and Commentary, ed 3. New York, NY, Oxford, 2006
37. Larsson J, Agardh CD, Apelqvist J, et al: Long-term prognosis after healed amputation in patients with diabetes. Clin Orthop 1998: 149Y58
35. Eijk MS, van der Linde H, Buijck BI, et al: Geriatric rehabilitation of lower limb amputees: A multicenter study. Disabil Rehabil 2012;34:145Y50
www.ajpmr.com
38. Izumi Y, Satterfield K, Lee S, et al: Risk of reamputation in diabetic patients stratified by limb and level of amputation: A 10-year observation. Diabetes Care 2006;29:566Y70
Cognition and Postamputation Outcomes Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
717
Outcomes
ORIGINAL RESEARCH ARTICLE
Affiliations: From the VA Puget Sound Health Care System (RMW, APT, MG, AWH, KNH, JMC); Department of Rehabilitation Medicine, University of Washington, Seattle (RMW, APT, KNH, JMC); Spectrum Research, Inc, Tacoma, Washington (DCN); VA Eastern Colorado Health Care System (DJB); and University of Colorado, Denver (DJB).
Relationship Between Cognition and Functional Outcomes After Dysvascular Lower Extremity Amputation A Prospective Study
Correspondence: All correspondence and requests for reprints should be addressed to: Rhonda M. Williams, PhD, VA Puget Sound Health Care System, RCS-117, 1660 S. Columbian Way, Seattle, WA 98108.
Disclosures:
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal_s Web site (www.ajpmr.com).
www.ajpmr.com
cognitive screen and four neuropsychologic tests administered at both 6 wks and 4 mos after amputation and five functional outcomes measured 12 mos after lower extremity amputation.
Design: This study includes a prospective cohort from four medical centers. Participants were primarily male Veterans experiencing their first lower extremity amputation as a result of complications of diabetes mellitus or peripheral arterial disease. Of those eligible, 87 (64%) enrolled; 75 (86%) were retained at 12 mos. Measures included demographic/health information, four neuropsychologic measures, the Locomotor Capability IndexY5, the Gronigen Activity Restriction Scale, prosthetic use, community participation, and social integration.
Results: Better performance on the Short Portable Mental Status Question-
Editor’s Note:
DOI: 10.1097/PHM.0000000000000235
Williams RM, Turner AP, Green M, Norvell DC, Henderson AW, Hakimi KN, Blake DJ, Czerniecki JM: Relationship between cognition and functional outcomes after dysvascular lower extremity amputation: a prospective study. Am J Phys Med Rehabil 2015;94:707Y717.
Objective: The aim of this study was to examine associations between a
Supported by the United States Department of Veterans Affairs, Office of Research and Development, Rehabilitation Research and Development (Merit Review A41241 Joseph Czerniecki, principal investigator, and Career Development Award B4927W Aaron P. Turner, principal investigator). Financial disclosure statements have been obtained, and no conflicts of interest have been reported by the authors or by any individuals in control of the content of this article.
0894-9115/15/9409-0707 American Journal of Physical Medicine & Rehabilitation Copyright * 2014 Wolters Kluwer Health, Inc. All rights reserved.
ABSTRACT
naire at 4 mos was associated with greater 12-mo mobility and social integration. Better attention and working memory abilities 6 wks after amputation were associated with increased 12-mo prosthetic wear; and at 4 mos after amputation, with greater 12-mo mobility. Better verbal memory at 6 wks was associated with greater 12-mo social integration and community participation as well as increased prosthetic wear.
Conclusions: These findings highlight the potential value in including a brief, formal cognitive assessment in addition to a general mental status screen. Specific domains of cognitive function are differentially associated with functional outcomes and may inform amputation rehabilitation decisions. Key Words:
Cognition, Amputation, Outcome Assessment, Neuropsychologic Tests
Cognition and Postamputation Outcomes Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
707
A
daptation to limb loss in the context of chronic disease represents an ongoing process. After acute surgical recovery, individuals must avoid or manage residual limb wounds and revision surgeries, maintain the contralateral limb,1 learn to use gait aids or prosthetic equipment,2 maintain independent living,3 as well as manage their chronic disease. Vascular problems, particularly related to diabetes mellitus and peripheral arterial disease, are the most common etiology for lower extremity amputations (LEAs), accounting for 82% of cases annually.4 Because diabetes mellitus and peripheral arterial disease themselves, as well as factors with which they are often associated (i.e., pain, anxiety, medications), can adversely impact cognition, individuals undergoing amputation may be at risk for cognitive impairment.5,6 A recent review indicated that cognitive impairment was more prevalent among individuals with amputations than in the general population and was associated with poorer mobility, prosthetic use, and independent living outcomes.7 Although, historically, the most frequent method for quantifying cognition after amputation is noting the presence or absence of dementia,7Y9 several recent studies have used more specific neuropsychologic tests to evaluate the association between cognition and outcomes such as mobility, prosthetic use, and activities of daily living (ADL) independence. O’Neill and Evans10 found that visual-spatial learning and executive function abilities assessed 4 wks after amputation were associated with 8-mo mobility and hours of prosthetic wear, respectively. In contrast, Schoppen et al.11 found that neither a general cognitive screen, verbal memory test, nor executive function test administered 2 and 6 wks after amputation were predictive of mobility scores or prosthetic use 1 yr after amputation, although they were associated with ADLs at 1 yr after surgery. The authors’ previous work in this population suggests that cognition improves between before amputation and 6 wks after amputation and then stabilizes between 6 wks and 4 mos after amputation.12 Taken together, these findings suggest that the impact of cognition on outcome may vary depending on the timing of cognitive assessment, the domain of cognition assessed, and the specific outcome of interest. There is a need for prospective research to explore relationships between cognition and outcomes after amputation; such knowledge could inform prosthetic, therapy, and disposition decisions. Extant studies have been characterized by small sample sizes and methodological variance, particularly in selection of cognitive measures and timing
708
Williams et al.
of assessments.7 In addition, outcomes are often limited to mobility, ADLs, and prosthetic wear; little is known about the impact of cognition on social and community participation among individuals with LEA, although cognitive impairment has been associated with diminished social functioning in other populations with chronic disease,13 and social and community involvement is known to be an important goal as well as a factor that may bolster social support and mitigate the stress of adapting to limb loss among those with recent LEA.14Y16 The authors’ broad study goal was to identify useful ways to assess cognition that are associated with functional outcomes and can inform rehabilitation decision making. The first specific aim was to examine whether a general mental status (GMS) screen administered at both 6 wks and 4 mos after amputation is associated with five indicators of function at 12 mos after amputation: mobility, prosthetic use and ADLs, as well as novel indicators of community participation and social integration. The second aim was to determine whether the addition of four brief neuropsychologic tests (Digit Span, List Learning, List Recall, and Semantic Fluency) accounts for unique variance in functional outcomes above and beyond the GMS screen. The third aim was to examine whether cognition has different associations with 12-mo functional outcomes when it is assessed at 6 wks or 4 mos after amputation. These objectives address several gaps in the literature by using a range of neuropsychologic tests in addition to a GMS screen, an expanded range of outcome measures, and two possible assessment time points.
METHODS Study Design This study is part of a larger multisite prospective cohort study of individuals undergoing LEA as a result of peripheral arterial disease or diabetes mellitus at two VA medical centers, a university hospital, and a level I trauma center. Participants were assessed via in-person or telephone interview at 6 wks, 4 mos, and 12 mos after amputation. Study procedures were approved by the human subjects review boards at each study site.17,18
Participants Subjects were considered eligible if (1) they were 18 yrs or older; (2) they were awaiting (or underwent in the last 6 wks) a first major LEA, defined as transmetatarsal level or higher; and (3) the primary cause of amputation was diabetes mellitus or peripheral arterial disease. Subjects
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
were excluded if they had any of the following: (1) cognitive or language impairment that would preclude consent or participation, defined by more than four errors on the Short Portable Mental Status Questionnaire (SPMSQ)19; (2) previous amputations; (3) nonambulatory status for reasons unrelated to impairment of the extremity awaiting amputation; or (4) planned bilateral amputation. Of 239 individuals screened between 2005 and 2008, a total of 136 (57%) met study criteria and 87 (64%) enrolled. The primary reason for exclusion was previous amputation (n = 34, 38%); these individuals may also have screen failure of the SPMSQ, but this was not assessed once a candidate was excluded for other reasons. An additional 23 (22%) were excluded solely because of dementia diagnosis or failure of the SPMSQ. Six participants (7%) died, and 6 (7%) were lost to follow-up; 75 (86%) remained enrolled at the 12-mo follow-up (see Table 1 for participant characteristics).
Covariates Assessed at the Time of Neuropsychologic Testing Demographics Covariates were dichotomously coded as follows: (a) sex, male (0) vs. female (1); (b) race, white (1) vs. nonwhite (0); and (c) transtibial or lower amputation level (0) vs. transfemoral amputation level (1). Age was reported in years (Table 1).
Major Depressive Episode Because depressed mood can impact cognitive performance,20 presence of a major depressive episode (MDE) was assessed with the Depression module of the Patient Health Questionnaire,21 a nine-item screening measure. Participants who endorsed five
or more symptoms (one of which was depressed mood or anhedonia) on Bmore days than not[ in the past 2 wks were coded as having MDE (1), and those who did not were coded as having no MDE (0)21 (Table 2). Presence of MDE was assessed concurrently with the neuropsychologic assessment (i.e., at both 6 wks and 4 mos).
Self-rated Health Self-rated health (SRH) was assessed via a single item, with response options ranging from 1 (Bvery poor[) to 5 (Bvery good[)22 (Table 2). SRH was assessed concurrently with the neuropsychologic assessment (at both 6 wks and 4 mos).
Cognitive Measures General Mental Status Participants completed the ten-item SPMSQ,19 an assessment of orientation and GMS at each study time point, because GMS is a common method of measuring cognition in published studies.23,24 Individuals with more than four errors on the SPMSQ at the time of initial screening were excluded from this study because test norms demonstrate that more than four errors indicates moderate cognitive impairment.19
Neuropsychological Subtests Subtests from two neuropsychologic batteriesV Semantic Fluency, List Learning, and List Recall from the Repeatable Battery for the Assessment of Neuropsychological Status25 and Digit Span from the Wechsler Adult Intelligence Scale, Third Edition,26 were administered at 6 wks and 4 mos after surgery. Tests were chosen for their brevity; for their ease of administration; and to represent varying aspects of
TABLE 1 Sociodemographic, health, and outcome data among study participants with complete data at 12 mos (N = 75) Sociodemographic/Health (Assessed at 6 wks) Age, mean (SD), yrs Female, n (%) Race, n (%) White Other Amputation level, n (%) Below-knee (transmetatarsal and transtibial) Above-knee (transfemoral)
Range 61.54 (8.51) 6 (8) 61 (81.3) 14 (18.7) 68 (90.7) 7 (9.3)
Outcomes (Assessed at 12 mos)
Range
LCI-5, mean (SD) Hours of prosthetic wear per day, mean (SD) GARS, mean (SD) Community participation, mean (SD) Social integration, mean (SD)
www.ajpmr.com
47Y83
40.85 (15.54) 9.17 (5.49) 28.88 (12.97) 6.27 (2.28) 6.26 (2.68)
0Y56 0Y18 18Y72 0Y10 0Y12
Cognition and Postamputation Outcomes Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
709
710
Williams et al.
Note: summary based on final regression models (see Tables 4 and 5). For example, Digit Span and List Recall are associated with outcomes shown above and beyond variance explained by SPMSQ. Although not significant in any of the final models, executive functioning, the ability to initiate problem-solving behaviors, cognitive flexibility, and response inhibition, was assessed with the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) Semantic Fluency subtest. a P G 0.05. WAIS-III, Wechsler Adult Intelligence Scale, Third Edition.
Auditory-verbal memory; RBANS List Recall
a
a
a
a
a
Mobility Social integration Mobility Prosthetic use (hours of wear per day) Community participation Social integration Prosthetic use (hours of wear per day) a
4 mos
a
Ability to store and manipulate multiple pieces of information for a short time (i.e., G2 mins) The ability to retrieve verbal information presented orally after a delay
Mobility was assessed using the Locomotor Capability Index (LCI-5).28 Possible scores range from 0 to 56, with higher scores representing higher functional mobility.
Attention/working memory; WAIS-III Digit Span
Mobility
Orientation and basic awareness of current information
Outcomes Assessed 12 mos After Amputation
GMS; SPMSQ
cognitive functioning examined in other published studies, including auditory-verbal memory, attention, and executive function.12 To improve access, some participants (n = 24 [28%] at 6 wks; n = 26 [33%] at 4 mos) were assessed via telephone.27 Those tested via telephone reported higher levels of education than those tested in person (13.9 vs. 12.9 yrs, P G 0.05). There were no other significant differences between the two groups on any of the standardized neuropsychologic test scores at either time point, nor were there any significant differences on any other health or demographic factors. For each neuropsychologic test, raw scores were converted to standardized z-scores using norms that adjust for age and, in the cases of Wechsler Adult Intelligence Scale, Third Edition Digit Span, education and sex. In all cases, a z-score of 0 indicates function at the 50th percentile, whereas positive/ greater z-scores indicate better cognitive performance and negative/lower z-scores indicate poorer function. Each neuropsychologic test that was significant in the final models and the domain of cognitive function it assesses is described in Table 3. It should be noted that the Repeatable Battery for the Assessment of Neuropsychological Status offers two alternate forms to allow for repeated assessment; Form B was used at 6 wks and Form A was used at 4 mos. Details about test administration and test performance as well as data from visual-spatial neuropsychologic tests included in the broader study are included elsewhere.12
Test Administration Time
Median SRH score at 6 wks: 3; at 4 mos: 4.
TABLE 3 Summary of associations between cognitive domains and outcomes
a
6 wks
Met criteria for MDE, n (%) 15 (20.5) 15 (20.5) SRHa Very poor 2 (2.7%) 1 (1.4%) Poor 9 (12.3%) 9 (12.3%) Fair 29 (39.7%) 24 (32.9%) Good 27 (37.0%) 32 (43.8%) Very good 6 (8.2%) 7 (9.6%)
Test Description
4 mos
Cognitive Domain and Neuropsychologic Test Name
6 wks
12-mo Outcomes Associated with Domain
TABLE 2 Self-rated depression and health among study participants with complete data at 12 mos (N = 75)
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Prosthetic Use Participants who had been fit for a prosthetic limb were asked the mean number of hours per day that they wore their prosthesis and the number of hours they walked with their prosthesis.
Activities of Daily Living ADLs were assessed using the Gronigen Activity Restriction Scale (GARS).29 Total possible scores can range from 18 to 72, with higher scores indicating greater levels of impairment.29
Community Participation Because existing measures of community participation confound factors such as ability and desire to participate, necessity and availability of assistance, as well as actual behavioral engagement,30,31 the study staff (five rehabilitation professionals) adapted a well validated existing measure for this study population. The authors used three items from the Community Integration Questionnaire32 and created a fourth similarly formatted item to assess how often an individual engaged in community-based activities (e.g., shopping, leisure activities, community activities, and traveling) outside the home in the past month. Each item was scored categorically (e.g., 0, never; 1, one to four times; and 2, five or more times). Total possible sum scores ranged from 0 to 10, with higher scores indicating greater community participation. Although the modifications to this measure that were made for this study are not validated, initial examination suggests that this measure has construct validity; it is correlated at r = 0.45, P G 0.01, with concurrent (i.e., 12-mo) social integration, at r = 0.38, P G 0.01 with the LCI, and at r = j0.42, P G 0.01 with the GARS. The Cronbach > for this scale was 0.74, suggesting adequate internal consistency. Items for this measure and the social integration measure (below) are included in Appendix A (http://links.lww.com/PHM/A86).
Social Integration Social integration measured the extent to which participants were an active part of their social network, including frequency of social interactions and breadth of social network. The authors used one item from the Community Integration Questionnaire,32 used two items from the social integration subscale of the Craig Handicap Assessment and Reporting Technique,33 and developed two new items. The resulting five-item scale has possible sum scores ranging from 0 to 12, with higher scores indicating greater social participation. The www.ajpmr.com
total score is correlated at r = 0.43, P G 0.01 with concurrent social support, suggesting good construct validity. The Cronbach > for this scale was 0.76.
Data Analysis First, Pearson univariate correlations were computed to identify associations between neuropsychologic tests and each outcome measure. There were no significant univariate correlations between 6-wk neuropsychologic tests and ADLs as well as both 6-wk and 4-mo neuropsychologic tests and hours of walking with a prosthesis; hence, these models were not computed. Further, for parsimony, only neuropsychologic tests that were correlated at a univariate level with the outcome of interest at P e 0.10 were included. Multiple linear regression models were computed to examine the unique associations between neuropsychologic measures and outcome measures, controlling for health and demographic factors. Separate regression analyses were performed for each of the outcome variables with forced entry of the blocks of covariates. The first block in the regression models included demographic characteristics (age, sex, and race), the second included health factors (amputation level, SRH, and MDE), the third included GMS (SPMSQ), and the fourth included any neuropsychologic tests with a univariate correlation significant at P G 0.10. With the exception of specific neuropsychologic tests, all other variables (including SPMSQ) were entered into each model regardless of significance.
RESULTS Neuropsychologic Test Scores Overall, the participants in this study scored in the low average to average range on the neuropsychologic tests administered; the frequency of scores falling into the impaired range (less than the second percentile) ranged from 1% to 13%. Greater details about test scores are published elsewhere.12 The mean scores on the Reliable Digit Span, an embedded measure of task engagement, were consistently higher than the threshold commonly taken to indicate adequate effort (97), supporting the validity of these data.34
Mobility The mean LCI-5 mobility scores are presented in Table 2 and are discussed at greater length elsewhere.17 After controlling for health and demographic factors, neither GMS nor the specific neuropsychologic tests assessed at 6 wks after amputation were associated with mobility at 12 mos Cognition and Postamputation Outcomes
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
711
(Table 4). However, when these cognitive domains were measured at 4 mos after surgery, better GMS (A = 0.25, P e 0.01; Table 5) and higher scores for Digit Span, a measure of attention (A = 0.29, P e 0.01), were associated with greater mobility.
Prosthetic Use By 12 mos after amputation, 69 participants (92.0%) were fitted for a prosthesis.18 Among those fitted with a prosthesis, higher Digit Span (A = 0.35, P e 0.05) and List Recall (A = 0.35, P e 0.05) scores assessed at 6 wks were associated with increased hours of prosthetic wear at 12 mos. Specifically,
improving the z-score by 1 SD on either cognitive test was associated with wearing a prosthetic for approximately 2 additional hours per day. Although Digit Span and List Recall assessed at 4 mos were significantly and positively correlated with prosthetic use at a univariate level, these were not significantly associated with hours of prosthetic wear at 12 mos when controlling for other factors.
Activities of Daily Living Neither GMS nor the neuropsychologic tests administered at 6 wks were significantly associated
TABLE 4 Multiple regression analyses of 6-wk neuropsychologic test scores on 12-mo outcomes Step and Variables Mobility (LCI-5) 1. Demographics
Age Sex Ethnicity 2. Health factors Amputation level SRH at 6 wks MDE at 6 wks 3. Mental status at 6 wks SPMSQ 4. Cognition at 6 wks Digit Span Prosthetic use (hours of prosthetic wear) 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 6 wks MDE at 6 wks 3. Mental status at 6 wks SPMSQ 4. Cognition at 6 wks Digit Span List Recall Community participation 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 6 wks MDE at 6 wks 3. Mental status at 6 wks SPMSQ 4. Cognition at 6 wks List Recall Social integration 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 6 wks MDE at 6 wks 3. Mental status at 6 wks SPMSQ 4. Cognition at 6 wks List Recall
A
$ R2
$F
j0.16 j0.40a 0.25b j0.17 0.12 j0.19c j0.03 0.18
0.22
6.52a
0.07
2.19c
0.00 0.03
0.08 2.69
0.21 0.00 j0.07 j0.39a j0.22 j0.14 0.01 0.35b 0.35b
0.05
1.23
0.19
3.82b
0.00 0.16
0.01 3.75b
0.03
0.56
0.04
0.80
0.01 0.09
0.84 5.87b
0.07
1.42
0.06
1.22
0.04 0.10
2.78c 7.16b
0.03 0.01 0.17 j0.10 0.10 j0.14 j0.13 0.35b 0.11 j0.01 0.24c j0.01 0.23c j0.07 0.22c 0.40a
Beta values are standardized and show the A for the step in which that variable was added. There were no significant (P e 0.10) univariate correlations between any neuropsychologic test scores and GARS or hours walking with prosthetic scores. Thus, the models for ADLs and walking with prosthesis are not shown. a P e 0.01. b P e 0.05. c P e 0.10.
712
Williams et al.
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
TABLE 5 Multiple regression analyses of 4-mo neuropsychologic test scores on 12-mo outcomes Step and Variables Mobility (LCI-5) 1. Demographics
Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos Digit Span List Recall Semantic Fluency Prosthetic use (hours of prosthetic wear) 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos Digit Span List Recall ADLs (GARS) 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos Digit Span List Recall Community participation 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos Digit Span List Recall Social integration 1. Demographics Age Sex Ethnicity 2. Health factors Amputation level SRH at 4 mos MDE at 4 mos 3. Mental status at 4 mos SPMSQ 4. Cognition at 4 mos List Recall
A
$ R2
$F
j0.11 j0.30a 0.25a j0.23a 0.37b j0.14 0.25a 0.29a j0.14 0.03
0.14
3.55a
0.27
9.52b
0.06 0.08
6.80b 3.25a
0.22 0.20 j0.04 j0.32a j0.36a j0.11 0.13 0.16 0.05
0.08
1.56
0.15
3.20a
0.02 0.02
0.95 0.73
0.30
0.47
0.16
4.21b
0.04 0.03
2.74c 1.20
j0.01 0.02 0.16 j0.12 0.36a j0.22 0.08 0.11 0.21
0.03
0.57
0.24
6.19b
0.01 0.05
0.43 1.92
0.13 0.01 0.25a j0.07 0.39b j0.05 0.25a 0.18
0.07
1.58
0.16
4.02b
0.06 0.02
4.69a 2.01
j0.12 0.04 j0.07 0.18 j0.27a 0.13 j0.19c j0.14 j0.10
Beta values are standardized and show the A for the step in which that variable was added. There were no significant (P e 0.10) univariate correlations between any neuropsychologic test scores and hours of walking with prosthetic scores. Thus, the model for walking with prosthesis is not shown. a P e 0.05. b P e 0.01. c P e 0.10.
with outcome measures at 12 mos, and hence, this regression model was not reported. Although Digit Span and List Recall assessed at 4 mos were correlated www.ajpmr.com
at a univariate level with 12-mo ADLs, these were not significantly associated with 12-mo ADLs after controlling for other factors in the model. Cognition and Postamputation Outcomes
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
713
Community Participation and Social Integration Higher GMS at 4 mos was associated with greater social integration (A = 0.25, P e 0.05). Higher List Recall scores at 6 wks were significantly associated with community participation (A = 0.35, P e 0.05) and social integration (A = 0.37, P e 0.01) at 12 mos after surgery.
DISCUSSION This study had several aims. First, it examined whether a brief screening of GMS after amputation would be associated with five 12-mo outcomes important to individuals with limb loss. Second, it examined whether the addition of four neuropsychologic tests provided additional insight into these outcomes above and beyond a brief screening. Finally, it examined whether the timing of assessment (6 wks vs. 4 mos) impacted the associations with outcomes. A GMS screen seems to be a valuable brief assessment and a useful predictor. Consistent with the literature, higher scores on a GMS screen were associated with increased mobility.9,35 The authors were able to extend previous findings to show that GMS is also associated with an individual’s engagement in a social network, which may have implications for an individual’s integration back into important relationships after rehabilitation. Support was also found for adding several brief neuropsychologic tests above and beyond a GMS screen. For example, social and community outcomes as well as hours of prosthetic wear were associated with verbal memory, an unsurprising finding given that these outcomes require the ability to learn and retain information as well as recognize and solve problems. For example, establishing and maintaining social relationships involve attending to and tracking pertinent information about another individual. Similarly, the donning and doffing of a prosthesis involves recalling several steps. That this was apparent in a sample in which individuals who failed the SPMSQ screen were excluded and mean memory performance at 6 wks fell in the average range suggests that memory might have a greater effect in a more diverse (i.e., impaired) sample. The prediction of mobility was improved by adding a measure of working memory and attention. This is also congruent with expectations because the ability to attend to limb positioning and use of gait aids would be expected to directly impact mobility. Schoppen et al. (2003) found verbal memory (measured by the 15-word test) to be uniquely
714
Williams et al.
associated with ADLs, a finding that was not replicated in this study. Although the measure of verbal memory in this study (List Recall) was similar to that of Schoppen et al. and significantly associated with ADLs at 12 mos at a univariate level, after controlling for health and demographic information, it was no longer significantly associated with ADLs. Verbal memory should plausibly contribute to success in ADLs because remembering a series of steps or instructions is critical in a variety of daily living tasks (e.g., cooking a meal). The sample in this study generally scored in the average range on verbal memory tasks. Schoppen et al. report that 90% of participants in their study scored at the 50th percentile or lower, but they do not describe the scores in further detail. Therefore, it may be that verbal memory is related to ADLs only when in an impaired range. In a related vein, it is possible that the verbal memory scores in this study are inflated because of practice effects or because of reduced time between the learning and recall trials. The fact that the authors did not find an association between executive function and prosthetic use, whereas O’Neill and Evans (2009) did, may be explained by differences in the timing of assessment (e.g., these authors assessed cognition at 2Y6 wks and prosthetic use at 8 mos) and by sampling differences (they include amputees with mixed etiologies). Although beyond the scope of this article, in a secondary analysis of a smaller sample of individuals fitted with a prosthesis by 4 mos, several neuropsychologic tests (Digit Span and Semantic Fluency) were associated with reported hours of walking with a prosthesis at 12 mos. Thus, cognition may play more of a role among those fitted earlier for a prosthesis and among those who ambulate more with their prosthesis (instead of wearing primarily for aesthetic purposes). This also highlights the importance of differentiating between wearing a prosthetic vs. ambulating with a prosthetic, a difference that is often not clearly delineated and not always consistent across studies.9,10,24,36 Future studies may help demarcate between prosthetic wear and prosthetic use in ambulation and better parse out predictors of each outcome. The impact of the timing of assessment on functional outcomes was multifarious. The authors cannot state that cognition should be assessed exclusively at 6 wks or at 4 mos because the impact of cognition varied with time assessed, even though cognitive performance was comparable at these two time points. For example, verbal memory assessed at 6 wks was significantly associated with prosthetic
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
wear, community participation, and social integration, but this association is weaker when verbal memory was assessed at 4 mos. It may be that the growing importance of the covariate SRH (not significant in any of the 6-wk models but significant in all the 4-mo models) overshadows the contributions of verbal memory on these outcomes. Similarly, associations between cognition and mobility also seem to shift. Although there is no obvious variable included that might mask the role of cognition in mobility at 6 wks, it may be that other more acute factors (i.e., pain, acclimation to discharge environment) may have resolved or evolved by 4 mos, allowing for more subtle effects of cognition to be visible. Thus, determining when to assess cognition may vary depending on the outcome of interest. Taken together, these findings suggest that, within a population of individuals with grossly intact overall cognitive functioning and undergoing an initial major LEA, aspects of attention and verbal memory are differentially associated with outcomes. Subtle impairment in cognition does not seem to impede ADL function after amputation, although, again, this may be an artifact of the relatively cognitively intact population of this study. However, even subtle impairment in the cognitive domains does seem to affect one’s ability to participate in meaningful social relationships and community activities as well as one’s ability to regain mobility or wear a prosthesis. The findings suggest that there is value in including more nuanced neuropsychologic measures in addition to a coarse cognitive screen, particularly if one is interested in outcomes other than mobility or ADL independence. The findings from this study have a variety of implications for rehabilitation care providers. Helping patients improve their focus and attention (by decreasing distractions and helping prioritize) may be associated with improved mobility, although future studies would be necessary to demonstrate that clinical intervention leads to improvement. Providing patients with compensatory memory skills may facilitate prosthesis use as well as maintenance of community participation and social integration, outcomes likely key in maintaining quality-of-life. Specifically, a rehabilitation team would be well served using repeatable/alternate versions of Digit Span and List Learning/List Recall subtests in addition to the SPMSQ at several time points. In addition, a visual attention and memory task may also be informative, particularly in a sample that has more varied cognitive abilities. There are a number of limitations that should be taken into consideration. To maximize retention, www.ajpmr.com
it was necessary to do a portion of assessments by telephone. This resulted in significant missing data for measures requiring motor responses and also shortened time delays on the verbal memory recall task, which may have resulted in nonstandardized administration and artificially elevated scores on the delayed auditory-verbal memory tasks. This also prevented the inclusion of visual-spatial tests in the analyses because these tasks could not be completed via telephone; this rendered the authors unable to compare the results of this study with those studies that included visual-spatial tasks. However, flexibility in administration allowed the authors to collect a larger sample size. The SPMSQ was used as both an exclusion criterion (to identify a sample with the potential for rehabilitation) and a factor in regression models. Variance in SPMSQ scores was, therefore, necessarily limited, reducing the ability to examine its associations with cognition across a full range of scores. In addition, the sample included only individuals undergoing their first amputation and included very few individuals with transfemoral amputations. It is probable that the role of cognition would become more important among those with amputation-related complications (e.g., revisions and contralateral amputations) or with higher amputation levels.37,38 Finally, the measures of community participation and social integration are not validated. These measures used in this study are based on established measures, with items added and/or modified to tailor the content to a new amputee population. Future studies are needed to further refine/validate these instruments, and ideally, they would include validated instruments against which to compare the novel tools developed here. In conclusion, this study suggests that both a GMS screen and brief neuropsychologic tests provide information that is useful in understanding 1-yr outcomes. Future directions for research include extending this study to a larger, more diverse sample. This would help to identify critical scores on neuropsychologic measures to determine whether there are useful cutoff points that are associated with functional outcomes. Such knowledge may be incorporated into a decision tree or prediction model when considering rehabilitation decisions. In addition, research can be extended to include selfmanagement outcomes, such as wound care and coping with complications. REFERENCES 1. Cruz CP, Eidt JF, Capps C, et al: Major lower extremity amputations at a Veterans Affairs hospital. Am J Surg 2003;186:449Y54
Cognition and Postamputation Outcomes Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
715
2. Marzoug EA, Landham TL, Dance C, et al: Better practical evaluation for lower limb amputees. Disabil Rehabil 2003;25:1071Y4 3. Dillingham TR, Yacub JN, Pezzin LE: Determinants of postacute care discharge destination after dysvascular lower limb amputation. PM R 2011;3:336Y44 4. Ziegler-Graham K, MacKenzie EJ, Ephraim PL, et al: Estimating the prevalence of limb loss in the United States: 2005 to 2050. Arch Phys Med Rehabil 2008;89: 422Y9 5. Waldstein SR, Tankard CF, Maier KJ, et al: Peripheral arterial disease and cognitive function. Psychosom Med 2003;65:757Y63 6. Phillips NA: Thinking on your feet: A neuropsychological review of peripheral vascular disease, in Elias SRWMF (ed): Neuropsychology of Cardiovascular Disease. Mahwah, NJ, Erlbaum, 2001, pp 121Y42 7. Coffey L, O’Keeffe F, Gallagher P, et al: Cognitive functioning in persons with lower limb amputations: A review. Disabil Rehabil 2012;34:1950Y64 8. Larner S, van Ross E, Hale C: Do psychological measures predict the ability of lower limb amputees to learn to use a prosthesis? Clin Rehabil 2003;17: 493Y8 9. Taylor SM, Kalbaugh CA, Blackhurst DW, et al: Preoperative clinical factors predict postoperative functional outcomes after major lower limb amputation: An analysis of 553 consecutive patients. J Vasc Surg 2005;42:227Y35 10. O’Neill BF, Evans JJ: Memory and executive function predict mobility rehabilitation outcome after lower-limb amputation. Disabil Rehabil 2009;31: 1083Y91 11. Schoppen T, Boonstra A, Groothoff JW, et al: Physical, mental, and social predictors of functional outcome in unilateral lower-limb amputees. Arch Phys Med Rehabil 2003;84:803Y11 12. Williams RM, Turner AP, Green M, et al: Changes in cognitive function from presurgery to 4 months postsurgery in individuals undergoing dysvascular amputation. Arch Phys Med Rehabil 2014;95:663Y9 13. Liddle PF: Cognitive impairment in schizophrenia: Its impact on social functioning. Acta Psychiatr Scand Suppl 2000;400:11Y6 14. Gallagher P, Maclachlan M: Adjustment to an artificial limb: A qualitative perspective. J Health Psychol 2001;6:85Y100 15. Rybarczyk B, Nicholas JJ, Nyenhuis D: Coping with leg amputation: Integrating research and clinical practice. Rehabil Psychol 1997;42:241Y56 16. Horgan O, MacLachlan M: Psychosocial adjustment to lower-limb amputation: A review. Disabil Rehabil 2004;26:837Y50 17. Norvell DC, Turner AP, Williams RM, et al: Defining successful mobility after lower extremity amputation for complications of peripheral vascular disease and diabetes. J Vasc Surg 2012;54:412Y9
716
Williams et al.
18. Webster JB, Hakimi KN, Williams RM, et al: Prosthetic fitting, use, and satisfaction following lowerlimb amputation: A prospective study. J Rehabil Res Dev 2012;49:1493Y504 19. Pfeiffer E: A Short Portable Mental Status Questionnaire for the assessment of organic brain deficit in elderly patients. J Am Geriatr Soc 1975;23: 433Y41 20. McDermott LM, Ebmeier KP: A meta-analysis of depression severity and cognitive function. J Affect Disord 2009;119:1Y8 21. Kroenke K, Spitzer RL, Williams JB: The PHQ-9: Validity of a brief depression severity measure. J Gen Intern Med 2001;16:606Y13 22. Verbrugge LM, Merrill SS, Liu X: Measuring disability with parsimony. Disabil Rehabil 1999;21:295Y306 23. Remes L, Isoaho R, Vahlberg T, et al: Predictors for institutionalization and prosthetic ambulation after major lower extremity amputation during an eight-year follow-up. Aging Clin Exp Res 2009;21: 129Y35 24. Bilodeau S, Hebert R, Desrosiers J: Lower limb prosthesis utilisation by elderly amputees. Prosthet Orthot Int 2000;24:126Y32 25. Randolph C, Tierney MC, Mohr E, et al: The Repeatable Battery for the Assessment of Neuropsychological Status (RBANS): Preliminary clinical validity. J Clin Exp Neuropsychol 1998;20:310Y9 26. Wechsler D: WAIS-III: Wechsler Adult Intelligence Scale. San Antonio, TX, The Psychological Corp, 1997 27. Fong TG, Fearing MA, Jones RN, et al: Telephone interview for cognitive status: Creating a crosswalk with the Mini-Mental State Examination. Alzheimers Dement 2009;5:492Y7 28. Grise MC, Gauthier-Gagnon C, Martineau GG: Prosthetic profile of people with lower extremity amputation: Conception and design of a follow-up questionnaire. Arch Phys Med Rehabil 1993;74: 862Y70 29. Kempen GI, Miedema I, Ormel J, et al: The assessment of disability with the Groningen Activity Restriction Scale. Conceptual framework and psychometric properties. Soc Sci Med 1996;43:1601Y10 30. Dijkers MP, Whiteneck G, El-Jaroudi R: Measures of social outcomes in disability research. Arch Phys Med Rehabil 2000;81(suppl):S63Y80 31. Zhang L, Abreu BC, Gonzales V, et al: Comparison of the Community Integration Questionnaire, the Craig Handicap Assessment and Reporting Technique, and the Disability Rating Scale in traumatic brain injury. J Head Trauma Rehabil 2002;17: 497Y509 32. Willer B, Allen LR: Community integration and barriers to integration for individuals with brain injury, in Finlayson M, Garner S (eds): Brain Injury Rehabilitation: Clinical Considerations. Baltimore, MD, Wiliams & Wilkins, 1993, pp 355Y75
Am. J. Phys. Med. Rehabil. & Vol. 94, No. 9, September 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
33. Whiteneck GG, Charlifue SW, Gerhart KA, et al: Quantifying handicap: A new measure of long-term rehabilitation outcomes. Arch Phys Med Rehabil 1992;73:519Y26
36. Pohjolainen T, Alaranta H, Karkkainen M: Prosthetic use and functional and social outcome following major lower limb amputation. Prosthet Orthot Int 1990;14:75Y9
34. Strauss E, Sherman EMS, Spreen O: A Compendium of Neuropsychological Tests: Administration, Norms, and Commentary, ed 3. New York, NY, Oxford, 2006
37. Larsson J, Agardh CD, Apelqvist J, et al: Long-term prognosis after healed amputation in patients with diabetes. Clin Orthop 1998: 149Y58
35. Eijk MS, van der Linde H, Buijck BI, et al: Geriatric rehabilitation of lower limb amputees: A multicenter study. Disabil Rehabil 2012;34:145Y50
www.ajpmr.com
38. Izumi Y, Satterfield K, Lee S, et al: Risk of reamputation in diabetic patients stratified by limb and level of amputation: A 10-year observation. Diabetes Care 2006;29:566Y70
Cognition and Postamputation Outcomes Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
717
