This article was downloaded by: [Uppsala universitetsbibliotek]

On: 06 October 2014, At: 11:14

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Psychology, Health & Medicine

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/cphm20

Rehabilitation adherence and anterior

cruciate ligament reconstruction
outcome
a a a a
BW Brewer , AE Cornelius , JL Van Raalte , JC Brickner , JH
b b b b b
Sklar , JR Corsetti , MH Pohlman , TD Ditmar & K Emery
a
Springfield College , Springfield, MA, USA
b
New England Orthopedic Surgeons , Springfield, MA, USA
Published online: 23 Jan 2007.

To cite this article: BW Brewer , AE Cornelius , JL Van Raalte , JC Brickner , JH Sklar , JR

Corsetti , MH Pohlman , TD Ditmar & K Emery (2004) Rehabilitation adherence and anterior
cruciate ligament reconstruction outcome, Psychology, Health & Medicine, 9:2, 163-175, DOI:
10.1080/13548500410001670690

To link to this article: http://dx.doi.org/10.1080/13548500410001670690

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 PSYCHOLOGY, HEALTH & MEDICINE
VOL. 9, NO. 2, MAY 2004

Rehabilitation adherence and anterior

cruciate ligament reconstruction outcome

B. W. BREWER1, A. E. CORNELIUS1, J. L. VAN RAALTE1,

J. C. BRICKNER1, J. H. SKLAR2, J. R. CORSETTI2, M. H. POHLMAN2,
T. D. DITMAR2 & K. EMERY2
1
Springfield College, Springfield, MA, USA; 2New England Orthopedic Surgeons, Springfield, MA,
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

USA

Abstract This study examined the relationship between adherence to postoperative physical therapy
and outcome after surgical reconstruction of the anterior cruciate ligament (ACL) of the knee in a
sample of 72 men and 36 women. Indices of adherence were obtained during the first 6 weeks of
postsurgical rehabilitation. Outcome measures were administered before surgery and approximately 6
months after surgery. Results of a canonical correlation analysis indicated a statistically significant
association between the adherence variables and the outcome variables (r = 0.56, p 5 0.05). Greater
attendance at rehabilitation sessions and more favourable practitioner ratings of adherence during
rehabilitation sessions were associated with fewer self-reported knee symptoms and greater knee laxity.
The findings highlight the complex nature of the adherence – outcome relationship.

Introduction
Adherence to therapeutic regimens is considered a vital process in health care delivery.
Volumes of research have been devoted to documenting adherence rates, identifying factors
associated with adherence, and evaluating interventions designed to enhance adherence.
Implicit in the study of adherence is the assumption that people who adhere to medical
regimens experience better treatment outcomes than those who fail to adhere. This
assumption, however, may not always be warranted. Indeed, although positive associations
between treatment adherence and therapeutic outcome have been observed for many medical
regimens (Dunbar-Jacob & Schlenk, 1996), only 11 of 132 comparisons of adherent and
nonadherent individuals in a sample of 2125 adults with depression and/or one of four chronic
medical conditions showed statistically significant positive effects of adherence on health
outcome (Hays et al., 1994).
For medical treatments involving a substantial investment of time, effort, or money, the
presence or absence of a significant relationship between adherence and outcome may have
important implications for patient and practitioner behaviour. One such medical treatment
is rehabilitation following surgical reconstruction of the anterior cruciate ligament (ACL),

Address for correspondence: Britton W. Brewer, Center for Performance Enhancement and Applied Research,
Department of Psychology, Springfield College, Springfield, MA 01109, USA. Tel: + 1-413-748-3696. Fax: + 1-413-
748-3854. E-mail: bbrewer@spfldcol.edu

ISSN 1354–8506 print/ISSN 1465-3966 online/04/020163–13 # Taylor & Francis Ltd

DOI: 10.1080/13548500410001670690
 164 B. W. BREWER ET AL.

which is critical to the stability of the knee joint (Muller, 1983). It has been estimated that
approximately 80,000 Americans sustain acute tears of the ACL each year, often while
participating in physical activity, and that the health care costs associated with ACL tears are
nearly 1 billion dollars per annum (Griffin et al., 2000). ACL and other knee injuries during
adolescence and early adulthood are associated with a substantially elevated risk for
osteoarthritis in later adulthood (Gelber et al., 2000). Reconstructive surgery followed by a
course of physical therapy is generally recommended for young, active individuals who incur
ACL tears (Blair & Wills, 1991; DeCarlo et al., 1992; 1994; Marzo & Warren, 1991). In
response to clinical findings suggesting that adherence to the restrictions in activity
mandated in traditional conservative approaches to ACL rehabilitation was inversely related
to postoperative success, ‘accelerated’ postoperative physical therapy protocols have been
advocated for the past 15 years (Shelbourne et al., 1992; Shelbourne & Wilckens, 1990).
Accelerated rehabilitation programmes typically feature clinic- and home-based activities
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

designed to promote rapid restoration of range of motion, quadriceps strength, and normal
gait and facilitate an early return to physical activity (Blair & Wills, 1991; DeCarlo et al.,
1994; Shelbourne et al., 1992; Shelbourne & Nitz, 1990). A consequence of the
implementation of accelerated protocols is an increased burden on patients to adhere to
the rigorous behavioural requirements of the prescribed rehabilitation regimen (Haynes et
al., 1979).
Although positive adherence – outcome associations have been documented for a variety
of orthopedic conditions treated with physical therapy, including low back pain (DiFabio et
al., 1995; Friedrich et al., 1996; Kolt & McEvoy, in press) and knee osteoarthritis (O’Reilly
et al., 1999; Rejeski et al., 1997), there is scant empirical evidence that better adherence to
rehabilitation is related to better outcomes following ACL reconstruction. Only three studies
have reported a positive adherence – outcome relationship for ACL rehabilitation. Derscheid
and Feiring (1987) documented a positive relationship between attendance at rehabilitation
sessions and clinical improvement following ACL reconstructive surgery. Treacy et al.
(1997) found that greater adherence to a clinic-based postoperative rehabilitation
programme was associated with better knee function, higher patient satisfaction, and a
faster return to physical activity. Similarly, Brewer, Van Raalte, Cornelius et al. (2000)
found that three adherence measures were significant predictors of functional ability (as
measured by the one-leg hop for distance test). Attendance at rehabilitation sessions and
practitioner ratings of adherence during rehabilitation sessions were positively correlated
with functional ability and home cryotherapy completion was negatively correlated with
functional ability.
Although certainly suggestive of a positive adherence – outcome relationship for
rehabilitation following ACL reconstruction, the investigations by Derscheid and Feiring,
Treacy et al., and Brewer et al. are limited in that none of the studies controlled for
preoperative levels of the outcome variables. It is possible that the participants whose knees
were more functional prior to surgery were more inclined or better able to adhere to the
rehabilitation protocol after surgery. Further, adherence to home-based rehabilitation activities
in the study by Brewer et al. was assessed exclusively through retrospective self-reports, which
are subject to both bias and inaccuracy. Consequently, the primary purpose of the present
study was to examine the relationship between rehabilitation adherence and rehabilitation
outcome following ACL reconstructive surgery while statistically controlling for preoperative
levels of the outcome variables and assessing adherence to home-based rehabilitation activities
concurrently with daily self-reports. Based on the findings of Derscheid and Feiring (1987),
Treacy et al. (1997), and Brewer et al. (2000), positive associations between adherence indices
and outcome measures were expected.
 ADHERENCE AND OUTCOME IN KNEE REHABILITATION 165

Methods
Participants
Participants were 108 consecutive patients (72 men and 36 women) with knee injuries who
had ACL reconstructive surgery from one of the three orthopedic surgeons on the project, who
were slated to receive physical therapy after surgery at one of three designated clinic locations,
and who agreed to participate in the study. In recruiting the sample, only one patient who
satisfied eligibility criteria decided not to participate in the study. The mean age of participants
was 29.38 (SD = 9.93) years, with a range of 14 – 54 years. The racial/ethnic breakdown of
participants was: 90% White, not of Hispanic Origin; 6% Hispanic; 4% Black, not of Hispanic
Origin; and 1% Asian/Pacific Islander.
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

Rehabilitation protocol
The accelerated rehabilitation protocol following ACL reconstruction developed by
Shelbourne and his colleagues (DeCarlo et al., 1992; Shelbourne et al., 1992; Shelbourne &
Nitz, 1990; Shelbourne & Wilckens, 1990) and subsequently updated (DeCarlo et al., 1994)
was prescribed by the orthopaedic surgeons and followed by the physical therapists affiliated
with the proposed study. This physical therapy protocol emphasizes early attainment of range
of motion (extension and flexion of the knee), quadriceps strength, and normal gait (DeCarlo
et al., 1994; Shelbourne et al., 1992; Shelbourne & Nitz, 1990). Under this protocol, patients
are instructed to complete daily home exercises. At the clinics affiliated with the study, home
exercise prescriptions were given to participants using Personalized Rehabilitation Programme
(PReP) videocassettes (EBI Medical Systems, Parsippany, NJ, USA) that were customized for
the accelerated physical therapy programme following ACL reconstruction. The videocassettes
provided detailed auditory and visual instructions on how to perform each rehabilitation
exercise as well as real-time information on number of repetitions and sets of each exercise to
perform, stretch hold times, range of motion, and rest times.

Measures
Demographic, injury-related, adherence, and rehabilitation outcome variables were measured
in this study.

Measures of demographic and injury-related variables

A questionnaire was used to obtain demographic and injury-related information from
participants. The questionnaire included items requesting information on participants’ age,
gender, race/ethnicity, and date of ACL injury. Participants’ orthopaedic surgeons identified
the type of graft used during surgery and the presence of knee injuries sustained in addition to
the ACL tear (e.g., meniscal damage).

Adherence measures
Adherence to the accelerated ACL rehabilitation protocol was assessed for both home- and
clinic-based activities. Adherence to the home exercise programme was measured through self-
report and objective methodologies. After their first physical therapy appointment following
ACL reconstruction, participants recorded on a daily basis for a six-week period the number of
times that they completed their home exercise regimen with and without the use of their PReP
 166 B. W. BREWER ET AL.

videocassette. Mean numbers of daily exercise completions with and without the use of the
PreP videocassette over the six-week period were calculated for each participant. A mean total
number of daily exercise completions was also calculated for each participant. An overall home
exercise completion variable, HOMEX, was computed by dividing the total number of daily
exercise completions reported by the number of days for which self-reports were obtained.
Although the self-report method of assessing adherence helped to guard against memory
bias in the recall of home rehabilitation behaviour, it did not eliminate the possibility of social
desirability bias in participants’ responses. Therefore, an objective measure of home exercise
completion was used to corroborate self-reported home exercise completion. Specifically,
using technology developed by the video rental industry, participants’ PreP videocassettes
contained a hidden electronic counter that recorded each instance in which the videocassettes
were played. This unobtrusive, objective method for assessing adherence to the home exercise
programme is similar to approaches that have been employed effectively to assess adherence to
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

home exercises using audio tapes (Hoelscher et al., 1984; 1986; Martin et al., 1981; Taylor et
al., 1983) and biofeedback equipment (Levitt et al., 1996). Participants brought their PReP
videocassette back to the clinic on an approximately weekly basis for the first six weeks after
their first physical therapy appointment so that the number of times that the videocassette had
been played could be recorded. The mean number of daily exercise completions registered by
the electronic counter was calculated for each participant.
Two measures of adherence to clinic-based aspects of the accelerated ACL rehabilitation
protocol were taken. First, patient attendance at rehabilitation sessions was monitored. For
each participant, a ratio of sessions attended to sessions scheduled was calculated. Second, at
each physical therapy appointment attended by participants, the practitioner (e.g., physical
therapist, athletic trainer) responsible for the rehabilitation of each participant on that day
completed the Sport Injury Rehabilitation Adherence Scale (SIRAS; Brewer et al., 2000). The
SIRAS is a three-item measure in which practitioners rate participants’ intensity of completion
of rehabilitation exercises, frequency of following practitioner instructions and advice, and
receptivity to changes in the physical therapy programme during that day’s appointment on 5-
point Likert-type scales. Scale anchors for the three items are ‘minimum effort’/‘maximum
effort,’ ‘never’/‘always,’ and ‘very unreceptive’/‘very receptive,’ respectively. Cronbach’s alpha
coefficients of 0.81 and 0.82 have been reported for the SIRAS and a test – retest reliability
coefficient of 0.77 has been obtained for the SIRAS over a one-week period. In support of the
criterion-related validity of the SIRAS, attendance at rehabilitation sessions has been positively
correlated with SIRAS scores in two samples (Brewer et al., 2000). High levels of interrater
agreement have been documented for the SIRAS (Brewer et al., 2002). A mean SIRAS score
across physical therapy appointments was computed for each participant.

Outcome measures
Consistent with previous investigations in which ACL rehabilitation outcome was evaluated
(e.g., DeCarlo et al., 1992; Shelbourne & Nitz, 1990; Shelbourne et al., 1990), multiple
measures of ACL rehabilitation outcome were taken. Specifically, laxity, functional ability, and
subjective symptoms were assessed.
Anterior – posterior laxity of the knee joint was assessed manually with a Lachman test
(Jonsson et al., 1982; Torg et al., 1976). Possible scores for the Lachman test are 0 (0 mm joint
play), 1 + (5 5 mm joint play), 2 + (5 – 10 mm joint play), and 3 + (4 10 mm joint play).
The one-leg hop for distance (Daniel et al., 1984) was used as a test of functional ability.
In this test, patients hopped for distance on one leg, taking off and landing with the same leg.
Hop distances were recorded in cm. Both the involved and uninvolved legs were tested three
 ADHERENCE AND OUTCOME IN KNEE REHABILITATION 167

times in the one-leg hop. The greatest distance hopped on the involved leg was divided by the
greatest distance hopped on the uninvolved leg. Support for the test – retest reliability of the
hop index (intraclass correlation coefficient = 0.81) has been obtained (Kramer et al., 1992),
and the index has been used effectively to evaluate the rehabilitation of ACL injuries (Tegner et
al., 1986).
Subjective symptoms were assessed with the Knee Outcomes Survey – Sports Activities
Scale (KOS-SAS; Borsa et al., 1998). The KOS-SAS is a self-report questionnaire with 6 items
measuring the extent to which various knee symptoms (i.e., pain, grinding or grating, stiffness,
swelling, partial giving way or slipping, complete giving way or slipping) affect respondents’
sports activity level and four items measuring the extent to which respondents’ knee affects
their ability to perform various sport tasks (i.e., running straight ahead, jumping and landing
on the involved leg, stopping and starting quickly, cutting and pivoting on the involved leg).
Item responses are given on 6-point Likert type scales scored from 0 to 5 and are summed and
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

multiplied by two to create a 100-point (maximum) scale. Scores on the KOS-SAS are highly
correlated (r s ranging from 0.67 to 0.87) with those on other major knee scoring systems
(Borsa et al., 1998). Cronbach’s alpha was 0.91 for the KOS-SAS in the current study.

Procedure
Patients meeting selection criteria were recruited from the office practices of the three
orthopaedic surgeons who are affiliated with this research. A research associate described the
purpose and procedures of the study to patients who expressed interest in participating in the
study. When the patients who agreed to participate in the study attended their preoperative
physical therapy appointment at one of the three participating physical therapy clinics
approximately one week before reconstructive surgery, they (and their parents or guardians
when appropriate) were asked to read and complete an informed consent form. At this time,
the research associate obtained from participants the addresses and telephone numbers of a
next of kin and a close friend (to facilitate tracking of participants over the course of the study)
and administered the questionnaire assessing demographic and injury-related variables to
participants. One of the rehabilitation practitioners affiliated with the project then conducted
an evaluation of participants in which all of the outcome variables except one-leg hop
performance (i.e., laxity and subjective symptoms) were assessed. It was not medically
advisable for participants to complete the one-leg hop test prior to surgery. Immediately
following reconstructive surgery, the orthopaedic surgeons recorded knee injuries sustained by
participants in addition to the ACL tear (e.g., meniscal damage).
At their first physical therapy appointment following reconstructive surgery, participants
received their PreP videocassette and instructions for the home exercise programme.
Participants were asked to bring their PreP videocassette back to the physical therapy clinic
on a weekly basis for the first six weeks after the initial postsurgical physical therapy
appointment to have the home exercise regimen updated to reflect progress in rehabilitation.
When videocassettes were returned, the rehabilitation practitioner updated the home exercise
regimen (by adding, subtracting, or modifying exercises as needed) and the research associate
covertly recorded the number displayed on an electronic device designed to read the value
registered by the electronic counter inside the videocassettes. Because informing participants
about the electronic counter in their PreP videocassette could have served as an intervention
(Taylor & May, 1996) and artificially inflated adherence to the home exercise regimen, it was
considered inappropriate to disclose the presence of the counter.
To facilitate self-reports of adherence to the home exercise programme, participants were
a given a six-week supply of data sheets and business reply envelopes addressed to the research
 168 B. W. BREWER ET AL.

team. Given the importance of having participants record their home adherence behaviour on
a daily basis (Stone et al., 1991), participants were instructed to mail each day’s self-report data
sheet back to the researchers the following morning. As an incentive to enhance compliance
with this procedure, participants were given one dollar for each completed data sheet that was
received and an additional seven dollars for each week of complete responding.
Measurement of adherence to the clinic-based portion of the accelerated rehabilitation
protocol also commenced at the first physical therapy appointment following reconstructive
surgery. At each participant’s physical therapy appointment, a research assistant documented
participant attendance/nonattendance and obtained SIRAS ratings for participant adherence
behaviour during that appointment (from the practitioner most involved with the patient’s
rehabilitation on that day).
Rehabilitation outcome measures were administered by a rehabilitation practitioner
affiliated with the project approximately six months after reconstructive surgery. To minimize
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

participant attrition, a participation incentive of $15.00 was offered for completion of each
assessment. Outcome assessments were coordinated through reminder letters, email messages,
and telephone calls to participants. When necessary, follow-up contacts were made with
participants through the relatives and friends identified by participants during preoperative
data collection. Participants were fully debriefed upon completion of their participation in the
study.

Statistical analyses
All statistical analyses were conducted with the SPSS for Windows (SPSS, Inc., Chicago, IL,
USA) statistical package.

Preliminary data screening and analyses

Descriptive statistics for all adherence and outcome variables were calculated. A one-way
multivariate analysis of variance (MANOVA) was performed to determine whether
participants with no postoperative outcome data differed from those with at least partial
postoperative outcome data for continuous variables measured prior to surgery (i.e., age,
Lachman test, KOS-SAS) and adherence variables. A series of one-way multivariate analyses
of covariance (MANCOVAs) with the postoperative outcome measures as dependent variables
and the preoperative outcome measures as covariates was conducted to assess possible
differences in the outcome measures as a function of participants’ gender, surgeon, graft type,
and presence/absence of meniscal damage. A repeated-measures MANOVA was performed on
the two outcome variables administered at both the preoperative and postoperative
assessments (i.e., Lachman test, KOS-SAS) to examine changes in outcome over the course
of surgery and rehabilitation. Intercorrelations were calculated among the adherence variables
and the preoperative and postoperative outcome variables.
To examine the correspondence between patient self-report of adherence and objective
assessment of adherence, a bivariate Pearson product-moment correlation coefficient was
calculated between the total number of self-reported completions of the home exercise routine
using the PReP videocassette and the number of times the PReP videocassette was played. The
possibility of systematic bias in the self-reports was examined through a within-subjects t-test
designed to detect a significant difference between the self-report and the objective
assessments of the number of times that the home exercise protocol was completed while
playing the PReP videocassette. To ensure that comparable self-report and videocassette data
were evaluated, only data from days for which participants returned their self-report booklet
 ADHERENCE AND OUTCOME IN KNEE REHABILITATION 169

and had a functional PReP videocassette in their possession were considered in the
correspondence and concordance analyses. Due to battery malfunction in the electronic
counter of several of the PReP videocassettes, only participants with at least one valid
videocassette play recorded were included in the correlation and t-test.

Main analysis
Canonical correlation analysis was used to evaluate the relationship between adherence to the
accelerated rehabilitation protocol (assessed by the set of three adherence measures) and
rehabilitation outcome following ACL reconstruction (assessed by the three outcome
measures). The influence of age, preoperative conditions, and the length of elapsed time
since surgery was controlled by using participants’ age, outcome scores from the preoperative
assessment, the number of days between surgery and the postoperative assessment as
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

covariates. The effect of these covariates was partialled out of the matrix of correlations
between the adherence measures and outcome measures prior to conducting the canonical
correlation analysis. Canonical loadings were examined to determine which variables
contributed to each variate in the canonical correlation analysis.

Results
Preliminary data screening and analyses
Descriptive statistics for and intercorrelations among adherence and preoperative and
postoperative outcome variables are presented in Table 1. In the MANOVA comparing
participants with missing adherence or outcome data to those without missing adherence or
outcome data, the multivariate effect was not statistically significant, indicating the presurgical
equivalence of the two groups of participants on the continuous variables assessed prior to
surgery. The series of one-way MANCOVAs with the postoperative outcome measures as
dependent variables and the preoperative outcome measures as covariates revealed no
significant multivariate effects of participants’ gender, surgeon, graft type, and presence/
absence of meniscal damage on postoperative outcome. Consequently, the data were collapsed
across participants’ gender, surgeon, graft type, and presence/absence of meniscal damage in
subsequent analyses.
In the repeated-measures MANOVA performed on the outcome variables, the
multivariate within-subjects time effect was statistically significant, Wilks’ lambda = 0.34,
F(2, 58) = 52.19, p 5 0.001. Statistically significant improvements were observed for scores on
both the Lachman test, F(1, 59) = 24.99, p 5 0.001, and the KOS-SAS, F(1, 59) = 103.17,
p 5 0.001. Intercorrelations among the adherence variables were generally weak, although the
correlations between attendance and SIRAS scores (r = 0.25, p 5 0.01) and attendance and
HOMEX scores (r = 7 0.29, p 5 0.005) were statistically significant. The correlation between
the two outcome measures administered at the preoperative assessment (i.e., Lachman test
and KOS-SAS) was weak and not statistically significant. Among the outcome measures
administered at the postoperative assessment, the correlation between KOS-SAS and one-leg
hop scores (r = 0.46, p 5 0.001) was significant and the correlation between KOS-SAS and
Lachman test scores (r = 7 0.22, p = 0.08) approached significance.
The correlation between the total number of self-reported completions of the home
exercise routine using the PReP videocassette and the number of times the PReP videocassette
was played was statistically significant, r = 0.58, p 5 0.001. The total number of self-reported
completions of the home exercise routine using the PReP videocassette was significantly
 170
B. W. BREWER ET AL.
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

Table 1. Means, standard deviations, intercorrelations, and canonical loadings of adherence and preoperative and postoperative outcome measures

Intercorrelations and canonical loadings

Canonical
Variable M SD 2 3 4 5 6 7 8 loading

1 HOMEX 2.89 1.43 7 0.29** 7 0.18 0.12 7 0.09 0.15 7 0.05 7 0.10 0.22
2 Attendance 0.91 0.14 0.25* 7 0.12 0.18 0.20 7 0.03 0.09 0.58
3 SIRAS 13.72 0.89 0.02 0.03 0.33** 0.08 7 0.07 0.86
4 Lachman test (pre) 1.96 0.68 7 0.06 0.26* 7 0.02 7 0.13
5 KOS-SAS (pre) 39.23 24.98 7 0.08 0.19 0.31*
6 Lachman test (post) 1.38 0.66 7 0.22 7 0.03 0.74
7 KOS-SAS (post) 76.80 19.48 0.46** 0.63
8 One-leg hop (post) 0.83 0.14 0.07

Note. n = 63 – 107. HOMEX = home exercise completion; SIRAS = Sport Injury Rehabilitation Adherence Scale; KOS-SAS = Knee Outcomes Survey—Sports Activities Scale.
*p 5 0.05.
**
p 5 0.005.
 ADHERENCE AND OUTCOME IN KNEE REHABILITATION 171

greater than the number of times the PReP videocassette was played, t(63) = 7.92, p 5 0.001,
although the mean difference between the two values was only 0.41 (SD = 0.42) counts per
day.

Main analysis
In the canonical correlation analysis performed between the set of three adherence variables
and the three outcome measures (controlling for preoperative conditions and the length of
elapsed time since surgery), the omnibus multivariate test was statistically significant, Wilks’
lambda = 0.65, Chi2(9) = 21.11, p 5 0.05, with only the first canonical correlation contribut-
ing significantly to the relationship (r = 0.56), accounting for 31% of the relationship between
the two variates in the canonical correlation.
To interpret which variables contributed significantly to each of the variates of the
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

canonical correlation, only loadings greater than 0.30 were considered (Pedhazur, 1982). For
the adherence variate, attendance and SIRAS scores had substantial positive loadings. For the
outcome variate, Lachman test and KOS-SAS scores had substantial positive loadings. These
findings indicate that greater attendance and higher SIRAS scores were associated with higher
levels of Lachman test and KOS-SAS scores. Thus, although better adherence to clinic-based
activities was related to a better outcome in terms of self-reported physical symptoms (KOS-
SAS), it was also related to greater knee laxity (i.e., higher scores on the Lachman test). The
loadings for HOMEX and the one-leg hop were not significant contributors to their respective
variates.

Discussion
The current study improved upon previous investigations of the adherence – outcome
relationship in ACL rehabilitation (Brewer et al., 2000; Derscheid & Feiring, 1987; Treacy
et al., 1997) in several key respects. Administering outcome measures both before and after
surgery enabled participants to serve as their own controls, allowed for the accounting of
approximately 10% of error variance in the outcome measures, and facilitated documentation
of participants’ improvements in knee laxity and self-reported physical symptoms over the
course of surgery and rehabilitation. Further, obtaining a concurrent daily record of
participants’ adherence to home-based rehabilitation activities is likely to have enhanced the
accuracy of their self-reports, the reliability of which was corroborated in an unbiased manner
through objective measurement. Although there was a tendency for participants to overreport
the number of times they did their home rehabilitation exercises while watching their PReP
videocassette each day, the magnitude of the overreporting was small and relatively constant
across levels of adherence (as indicated by the 0.58 correlation between self-report and
electronic counter values).
The results indicated that adherence to an accelerated rehabilitation protocol
following ACL reconstructive surgery accounted for a substantial portion of the variance
in rehabilitation outcome. Consistent with previous ACL research (Brewer et al., 2000;
Derscheid & Feiring, 1987; Treacy et al., 1997), adherence to clinic-based rehabilitation
activities was associated with knee rehabilitation outcomes. The hypothesized positive
adherence – outcome relationship was documented for one of the three outcome
variables—self-reported physical symptoms. Thus, the participants who reported experien-
cing fewer adverse knee symptoms 6 months after surgery were those who attended a
higher percentage of their scheduled rehabilitation sessions and were rated by their
rehabilitation practitioners as giving greater effort, following instructions better, and
 172 B. W. BREWER ET AL.

showing greater receptiveness to changes in the rehabilitation protocol during rehabilita-

tion sessions.
In contrast to the findings for self-reported physical symptoms, a negative adherence –
outcome relationship was obtained for a second outcome variable—knee laxity. Counter to the
hypothesized pattern of results, the findings for knee laxity are puzzling in light of the
significant improvements within the sample in both laxity and self-reported symptoms after
surgery and rehabilitation. The results suggest that accelerated ACL rehabilitation protocols
have potential costs and potential benefits, and that greater knee laxity may be a potential cost
for highly adherent individuals. In a recent study (Beynnon et al., 2002), individuals who
completed an accelerated ACL rehabilitation protocol had greater knee laxity at 24 months
postsurgery than those who completed a nonaccelerated (but not conservative) rehabilitation
protocol.
The negative correlation between knee laxity and self-reported physical symptoms is
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

indicative of the general lack of correspondence among measures of ACL rehabilitation

outcome (e.g., Brewer et al., 2000; Sekiya et al., 1998; Snyder-Mackler et al., 1997) and may
in part explain the divergent results for knee laxity and self-reported physical symptoms.
Another potential explanation is that participants with greater laxity following surgery
perceived their knees as ‘loose’ and recognized the need for increased adherence to clinic-
based rehabilitation activities. It is also possible that the rehabilitation practitioners of
participants with loose knees after surgery noted the laxity and elicited greater adherence
from the participants by communicating stronger expectations to adhere (Taylor & May,
1995). Further research in which additional assessments of knee laxity are taken in the
period between the preoperative and 6-month postoperative data collection episodes is
needed to investigate these possibilities.
The results of this study affirm the importance of adhering to clinic-based rehabilitation
activities to maximize postoperative ACL rehabilitation outcomes and suggest that adherence
to home-based rehabilitation activities may be less critical to achieving desired outcomes. It
should be noted that these findings were obtained in the context of a rehabilitation protocol
consisting of 2 to 3 clinic visits per week for 2 to 4 months. Adherence to home-based
rehabilitation activities may be more important when the rehabilitation protocol calls for fewer
clinic visits (Beard & Dodd, 1998; Fischer et al., 1998; Schenck et al., 1997; Treacy et al.,
1997). Interestingly, it appears that participants in the current study chose to emphasize either
clinic-based aspects of rehabilitation or home-based aspects of rehabilitation. Home exercise
completion was negatively related to both attendance at rehabilitation sessions (p 5 0.005) and
practitioner ratings of adherence during rehabilitation sessions (p 5 0.08).
Combined with previous findings on the adherence – outcome relationship for
postsurgical ACL rehabilitation (Brewer et al., 2000; Derscheid & Feiring, 1987; Treacy
et al., 1997), the current results suggest that interventions designed to boost treatment
adherence (Fisher et al., 1993; Worrell, 1992) might have a favourable effect on
rehabilitation outcomes. Given that the data most strongly support links between adherence
to clinic- (vs. home-) based aspects of physical therapy and rehabilitation outcomes, the
logical primary targets of adherence enhancement programmes are attendance at and
behaviour during rehabilitation sessions. As ACL rehabilitation protocols become
increasingly reliant on home-based rehabilitation activities (Beard & Dodd, 1998; Fischer
et al., 1998; Schenck et al., 1997; Treacy et al., 1997), however, adherence enhancement
efforts should follow suit and encompass both clinic- and home-based rehabilitation
components. If developed, implemented, and evaluated systematically, behavioural
interventions have the potential to influence both the quality and rate of rehabilitation
after ACL reconstruction.
 ADHERENCE AND OUTCOME IN KNEE REHABILITATION 173

Acknowledgements
This article was supported in part by grant number R29 AR44484 from the National Institute
of Arthritis and Musculoskeletal and Skin Diseases. Its contents are solely the responsibility of
the authors and do not represent the official views of the National Institute of Arthritis and
Musculoskeletal and Skin Diseases. We thank Howard Tennen for his helpful comments on
an earlier draft and Josh Avondoglio, Lisa Benjamin, Jeff Benoit, Kathy Bernardini, Jim Biron,
Ruth Brennan, Matt Buman, Judy Catalano, Tarra Cemborski, Kim Cochrane, Rosa Correa,
Bob Crawford, Candi Danielle, Amie Dillman, Joann Golden, Reubin Gonzalez, Bryan Gross,
Stephanie Habif, Whitney Hartmann, Jay Hatten, Carter Hunt, Jean Hutchinson, Angie
Jensen, Aaron Kopish, Dawn Kresge, Ellie Laino, Amy Lowery, Kevin McAllister, Thomas
Melvin, Alexa Mignano, Joe Monserrat, Liz Montemagni, Jonna Mullane, Rob Olenchak, Jens
Omli, Scott Quarforth, Alice Robitaille, Raylene Ross, Carrie Scherzer, Josie Scibelli, Wendy
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

Sewack, Ben Shachar, Sumiyo Shiina, and Marie Trombley for their assistance in data
collection.

References
BEARD, D. J. & DODD, C. A. F. (1998). Home or supervised rehabilitation following anterior cruciate ligament
reconstruction: A randomized controlled trial. Journal of Orthopaedic and Sports Physical Therapy, 27, 134 – 143.
BEYNNON, B. D., UH, B., FLEMING, B., POOLE, R. & JOHNSON, R. J. (2002). Rehabilitation following ACL reconstruction.
Paper presented at the annual meeting of the American Orthopaedic Society for Sports Medicine, Orlando, FL, July
2002.
BLAIR, D. F. & WILLS, R. P. (1991). Rapid rehabilitation following anterior cruciate ligament reconstruction. Athletic
Training, 26, 32 – 43.
BORSA, P. A., LEPHART, S. M. & IRRGANG, J. J. (1998). Sport-specificity of knee scoring systems to assess disability in
anterior cruciate ligament-deficient athletes. Journal of Sport Rehabilitation, 7, 44 – 60.
BREWER, B. W., AVONDOGLIO, J. B., CORNELIUS, A. E., VAN RAALTE, J. L., BRICKNER, J. C., PETITPAS, A. J., PIZZARI, T.,
SCHOO, A. M. M., EMERY, K. & HATTEN, S. J. (2002). Construct validity and interrater agreement of the Sport Injury
Rehabilitation Adherence Scale. Journal of Sport Rehabilitation, 11, 170 – 178.
BREWER, B. W., VAN RAALTE, J. L., CORNELIUS, A. E., PETITPAS, A. J., SKLAR, J. H., POHLMAN, M. H., KRUSHELL, R. J.
& DITMAR, T. D. (2000). Psychological factors, rehabilitation adherence, and rehabilitation outcome after anterior
cruciate ligament reconstruction. Rehabilitation Psychology, 45, 20 – 37.
BREWER, B. W., VAN RAALTE, J. L., PETITPAS, A. J., SKLAR, J. H., POHLMAN, M. H., KRUSHELL, R. J., DITMAR, T. D.,
DALY, J. M. & WEINSTOCK, J. (2000). Preliminary psychometric evaluation of a measure of adherence to clinic-based
sport injury rehabilitation. Physical Therapy in Sport, 1, 68 – 74.
DANIEL, D. M., STONE, M. L., RIEHL, B. & MOORE, M. R. (1984). A measurement of lower limb function. The one leg
hop for distance. American Journal of Knee Surgery, 4, 212 – 214.
DECARLO, M. S., SELL, D. E., SHELBOURNE, K. D. & KLOOTWYK, T. E. (1994). Current concepts on accelerated ACL
rehabilitation. Journal of Sport Rehabilitation, 3, 304 – 318.
DECARLO, M. S., SHELBOURNE, K. D., MCCARROLL, J. R. & RETTIG, A. C. (1992). Traditional versus accelerated
rehabilitation following ACL reconstruction: A one-year follow-up. Journal of Orthopaedic and Sports Physical Therapy,
15, 309 – 316.
DERSCHEID, G. L. & FEIRING, D. C. (1987). A statistical analysis to characterize treatment adherence of the 18 most
common diagnoses seen at a sports medicine clinic. Journal of Orthopaedic and Sports Physical Therapy, 9, 40 – 46.
DIFABIO, R. P., MACKEY, G. & HOLTE, J. B. (1995). Disability and functional status in patients with low back pain
receiving workers’ compensation: A descriptive study with implications for the efficacy of physical therapy. Physical
Therapy, 75, 180 – 193.
DUNBAR-JACOB, J. & SCHLENK, E. (1996). Treatment adherence and clinical outcome: Can we make a difference? In: R.
J. RESNICK & R. H. ROZENSKY (Eds), Health psychology through the life span: Practice and research opportunities
(pp. 323 – 343). Washington, DC: American Psychological Association.
FISCHER, D. A., TEWES, D. P., BOYD, J. L., SMITH, J. P. & QUICK, D. C. (1998). Home based rehabilitation for anterior
cruciate ligament reconstruction. Clinical Orthopaedics and Related Research, 347, 194 – 199.
FISHER, A. C., SCRIBER, K. C., MATHENY, M. L., ALDERMAN, M. H. & BITTING, L. A. (1993). Enhancing rehabilitation
adherence. Journal of Athletic Training, 28, 312 – 318.
 174 B. W. BREWER ET AL.

FRIEDRICH, M., CERMAK, T. & MADERBACHER, P. (1996). The effect of brochure use versus therapist teaching on
patients performing therapeutic exercise and on changes in impairment status. Physical Therapy, 76, 1082 – 1088.
GELBER, A. C., HOCHBERG, M. C., MEAD, L. A., WANG, N. Y., WIGLEY, F. M. & KLAG, M. J. (2000). Joint injury in
young adults and the risk for subsequent knee and hip osteoarthritis. Annals of Internal Medicine, 133, 321 – 328.
GRIFFIN, L. Y., AGEL, J., ALBOHM, M. J., ARENDT, E. A., DICK, R. W., GARRETT, W. E., GARRICK, J. G., HEWETT, T.
E., HUSTON, L., IRELAND, M. L., JOHNSON, R. J., KIBLER, W. B., LEPHART, S., LEWIS, J. L., LINDENFELD, T. N.,
MANDELBAUM, B. R., MARCHAK, P., TEITZ, C. C. & WOJTYS, E. M. (2000). Noncontact anterior cruciate ligament
injuries: Risk factors and prevention strategies. Journal of the American Academy of Orthopaedic Surgeons, 18, 141 – 150.
HAYNES, R., TAYLOR, D., SNOW, J. & SACKETT, D. (1979). Annotated and indexed bibliography on compliance with
therapeutic and preventive regimens. In R. HAYNES, D. TAYLOR, & D. SACKETT (Eds), Compliance in health care
(pp. 337 – 474). Baltimore: Johns Hopkins Press.
HAYS, R. D., KRAVITZ, R. L., MAZEL, R. M., SHERBOURNE, C. D., DIMATTEO, M. R., ROGERS, W. H. & GREENFIELD, S.
(1994). The impact of patient adherence on health outcomes for patients with chronic disease in the Medical
Outcomes Study. Journal of Behavioral Medicine, 17, 347 – 360.
HOELSCHER, T. J., LICHSTEIN, K. L. & ROSENTHAL, T. L. (1984). Objective vs. subjective assessment of relaxation
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014

compliance among anxious individuals. Behaviour Research and Therapy, 22, 187 – 193.
HOELSCHER, T. J., LICHSTEIN, K. L. & ROSENTHAL, T. L. (1986). Home relaxation practice in hypertension treatment:
Objective assessment and compliance induction. Journal of Consulting and Clinical Psychology, 54, 217 – 221.
JONSSON, T., ALTHOFF, B., PETERSON, L. & RENSTROM, P. (1982). Clinical diagnosis of ruptures of the anterior cruciate
ligament: A comparative study of the Lachman test and the anterior drawer sign. American Journal of Sports Medicine,
10, 100 – 102.
KOLT, G. S. & MCEVOY, J. F. (2003). Adherence to rehabilitation in patients with low back pain. Manual Therapy, 8,
110 – 116.
KRAMER, J. F., NUSCA, D., FOWLER, P. & WEBSTER-BOGAERT, S. (1992). Test-retest reliability of the one-leg hop test
following ACL reconstruction. Clinical Journal of Sport Medicine, 2, 240 – 243.
LEVITT, R., DEISINGER, J. A., WALL, J. R., FORD, L. & CASSISI, J. E. (1996). EMG feedback-assisted postoperative
rehabilitation of minor arthroscopic knee surgeries. Journal of Sports Medicine and Physical Fitness, 35, 218 – 223.
MARTIN, J. E., COLLINS, F. L., HILLENBERG, J. B., ZABIN, M. A. & KATELL, A. D. (1981). Assessing compliance to
home practice: A simple technology of a critical problem. Journal of Behavioral Assessment, 3, 193 – 198.
MARZO, J. M. & WARREN, R. F. (1991). Results of nonoperative treatment of anterior cruciate ligament injury:
Changing perspectives. Advances in Orthopaedic Surgery, 15, 59 – 69.
MULLER, W. (1983). The knee: Form, function, and ligament reconstruction. New York: Springer-Verlag.
O’REILLY, S. C., MUIR, K. R. & DOHERTY, M. (1999). Effectiveness of home exercise on pain and disability from
osteoarthritis of the knee: A randomised controlled trial. Annals of the Rheumatic Diseases, 58, 15 – 19.
PEDHAZUR, E. J. (1982). Multiple regression in behavioral research. (2nd edn). New York: Holt, Rinehart and Winston.
REJESKI, W. J., BRAWLEY, L. R., ETTINGER, W., MORGAN, T. & THOMPSON, C. (1997). Compliance to exercise therapy
in older participants with knee osteoarthritis: Implications for treating disability. Medicine and Science in Sports and
Exercise, 29, 977 – 985.
SCHENCK, R. C., JR., BLASCHAK, M. J., LANCE, E. D., TURTURRO, T. C. & HOLMES, C. F. (1997). A prospective
outcome study of rehabilitation programs and anterior cruciate ligament reconstruction. Arthroscopy: The Journal of
Arthroscopic and Related Surgery, 13, 285 – 290.
SEKIYA, I., MUNETA, T., OGIUCHI, T., YAGISHITA, K. & YAMAMOTO, H. (1998). Significance of the single-legged hop
test to the anterior cruciate ligament-reconstructed knee in relation to muscle strength and anterior laxity. American
Journal of Sports Medicine, 26, 384 – 388.
SHELBOURNE, K. D., KLOOTWYK, T. E. & DECARLO, M. S. (1992). Update on accelerated rehabilitation after anterior
cruciate ligament reconstruction. Journal of Orthopaedic and Sports Physical Therapy, 15, 303 – 308.
SHELBOURNE, K. D. & NITZ, P. (1990). Accelerated rehabilitation after anterior cruciate ligament reconstruction.
American Journal of Sports Medicine, 18, 292 – 299.
SHELBOURNE, K. D. & WILCKENS, J. H. (1990). Current concepts in anterior cruciate ligament rehabilitation.
Orthopaedic Review, 19, 957 – 964.
SNYDER-MACKLER, L., FITZGERALD, G. K., BARTOLOZZI, A. R. & CICCOTTI, M. G. (1997). Passive joint laxity and
functional outcome after anterior cruciate ligament injury. American Journal of Sports Medicine, 25, 191 – 195.
STONE, A., KESSER, R. & HAYTHORNTHWAITE, J. (1991). Measuring daily events and experiences: Decisions for the
researcher. Journal of Personality, 59, 575 – 608.
TAYLOR, A. H. & MAY, S. (1995). Physiotherapist’s expectations and their influence on compliance to sports injury
rehabilitation. In R. VANFRAECHEM-RAWAY & Y. VANDEN AUWEELE (Eds), IXth European Congress on Sport Psychology
proceedings: Part II (pp. 619 – 625). Brussels: European Federation of Sports Psychology.
 ADHERENCE AND OUTCOME IN KNEE REHABILITATION 175

TAYLOR, A. H. & MAY, S. (1996). Threat and coping appraisal as determinants of compliance to sports injury
rehabilitation: An application of protection motivation theory. Journal of Sports Sciences, 14, 471 – 482.
TAYLOR, C. B., AGRAS, W. S., SCHNEIDER, J. A. & ALLEN, R. A. (1983). Adherence to instructions to practice relaxation
exercises. Journal of Consulting and Clinical Psychology, 51, 952 – 953.
TEGNER, Y., LYSHOLM, J., LYSHOLM, M. & GILLQUIST, J. (1986). A performance test to monitor rehabilitation and
evaluate anterior cruciate ligament injuries. American Journal of Sports Medicine, 14, 156 – 159.
TORG, J. S., CONRAD, W. & KALEN, V. (1976). Clinical diagnosis of anterior cruciate ligament instability in the athlete.
American Journal of Sports Medicine, 4, 84 – 93.
TREACY, S. H., BARRON, O. A., BRUNET, M. E. & BARRACK, R. (1997). Assessing the need for extensive supervised
rehabilitation following arthroscopic ACL reconstruction. The American Journal of Orthopedics, 26, 25 – 29.
WORRELL, T. W. (1992). The use of behavioral and cognitive techniques to facilitate achievement of rehabilitation
goals. Journal of Sport Rehabilitation, 1, 69 – 75.
Downloaded by [Uppsala universitetsbibliotek] at 11:14 06 October 2014