If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Most patients adhered to onabotulinumtoxinA, >5 sessions in 2 years for adherents.
•
Real-world evidence from the Adult Spasticity International Registry study can enhance spasticity patient care.
Abstract
Objective
To identify baseline characteristics and treatment-related variables that affect adherence to onabotulinumtoxinA treatment from the Adult Spasticity International Registry (ASPIRE) study.
Clinically meaningful thresholds used for treatment adherent (≥3 treatment sessions during 2-year study) and nonadherent (≤2 sessions). Data analyzed using logistic regression and presented as odds ratios (ORs) with 95% confidence intervals (CIs). Treatment-related variables assessed at sessions 1 and 2 only.
Results
Of the total population, 523 patients (71.6%) were treatment adherent with 5.3±1.6 sessions and 207 (28.4%) were nonadherent with 1.5±0.5 sessions. In the final model (n=626/730), 522 patients (83.4%) were treatment adherent and 104 (16.6%) were nonadherent. Baseline characteristics associated with adherence: treated in Europe (OR=1.84; CI, 1.06-3.21; P=.030) and use of orthotics (OR=1.88; CI, 1.15-3.08; P=.012). Baseline characteristics associated with nonadherence: history of diplopia (OR=0.28; CI, 0.09-0.89; P=.031) and use of assistive devices (OR=0.51; CI, 0.29-0.90; P=.021). Treatment-related variables associated with nonadherence: treatment interval ≥15 weeks (OR=0.43; CI, 0.26-0.72; P=.001) and clinician dissatisfaction with onabotulinumtoxinA to manage pain (OR=0.18; CI, 0.05-0.69; P=.012). Of the population with stroke (n=411), 288 patients (70.1%) were treatment adherent with 5.3±1.6 sessions and 123 (29.9%) were nonadherent with 1.5±0.5 session. In the final stroke model (n=346/411), 288 patients (83.2%) were treatment adherent and 58 (16.8%) were nonadherent. Baseline characteristics associated with adherence: treated in Europe (OR=2.99; CI, 1.39-6.44; P=.005) and use of orthotics (OR=3.18; CI, 1.57-6.45; P=.001). Treatment-related variables associated with nonadherence: treatment interval ≥15 weeks (OR=0.42; CI, 0.21-0.83; P=.013) and moderate/severe disability on upper limb Disability Assessment Scale pain subscale (OR=0.40; CI, 0.19-0.83; P=.015).
Conclusions
These ASPIRE analyses demonstrate real-world patient and clinical variables that affect adherence to onabotulinumtoxinA and provide insights to help optimize management strategies to improve patient care.
Spasticity is associated with several central nervous system disorders and can be defined as disordered sensorimotor control, stemming from an upper motor neuron lesion, which presents as intermittent or sustained involuntary activation of muscles.
is approved worldwide for the treatment of adult upper limb and lower limb spasticity. In combination with other therapies, onabotulinumtoxinA can mitigate the deleterious effects of spasticity, including limited dexterity and mobility, limb pain, impaired activities of daily living, and reduced quality of life.
Li S, Francisco GE. The use of botulinum toxin for treatment of spasticity. In: Whitcup SM, Hallett M, editors. Botulinum toxin therapy. Handbook of Experimental Pharmacology. Vol 263. Cham; Springer: 2019.
To successfully meet patients’ needs and goals, adherence to prescribed treatment is critical. However, little is known about the effect of patient and treatment-related variables on real-world adherence to onabotulinumtoxinA treatment for spasticity. An increased understanding of variables that affect treatment adherence is needed to inform clinical strategies to better manage spasticity and address knowledge gaps. The Adult Spasticity International Registry (ASPIRE) study describes real-world onabotulinumtoxinA use to treat adult spasticity across multiple etiologies and geographic regions over 2 years.
Individualized onabotulinumtoxinA treatment for upper limb spasticity resulted in high clinician- and patient-reported satisfaction: long-term observational results from the ASPIRE study.
High clinician- and patient-reported satisfaction with individualized onabotulinumtoxinA treatment for spasticity across several etiologies from the ASPIRE study.
The objective of this analysis was to identify baseline demographic and clinical characteristics and treatment-related variables that affect adherence to onabotulinumtoxinA treatment for spasticity from the ASPIRE study.
Methods
Full methodological details for ASPIRE, including study dates and size, inclusion/exclusion criteria, and data collected, have been published.
Individualized onabotulinumtoxinA treatment for upper limb spasticity resulted in high clinician- and patient-reported satisfaction: long-term observational results from the ASPIRE study.
High clinician- and patient-reported satisfaction with individualized onabotulinumtoxinA treatment for spasticity across several etiologies from the ASPIRE study.
OnabotulinumtoxinA treatments were administered according to country-specific regulations and standard clinical practices, without intervention from the study sponsor. Time to retreatment was not dictated by the sponsor, nor were the number of treatment sessions. Retreatment with onabotulinumtoxinA was anticipated to occur approximately every 12 weeks.
Financial support was not provided for any treatment or treatment-related costs. ASPIRE was conducted in agreement with all relevant regulatory requirements, including but not limited to the Guidelines for Good Pharmacoepidemiology Practices (issued by the International Society for Pharmacoepidemiology) and the Declaration of Helsinki.
Participants
Adult participants with spasticity related to upper motor neuron syndrome due to various etiologies were treated at the clinician's discretion with onabotulinumtoxinA during routine clinical practice. Participants were naïve (newly treated) or nonnaïve (previously treated) to botulinum toxin for spasticity. All participants provided written informed consent before study participation. Institutional review board approval was granted at each study site.
For this analysis, 2 patient populations (total and stroke) from ASPIRE were assessed. The total population included all participants who received ≥1 onabotulinumtoxinA treatment during the 2-year study. The population with stroke included all participants who received ≥1 onabotulinumtoxinA treatment during the 2-year study and identified stroke as their primary etiology at baseline.
Outcomes and data sources
To identify baseline clinical characteristics and treatment-related variables that affect adherence to onabotulinumtoxinA treatment in ASPIRE, clinically meaningful data-driven thresholds were established. Treatment adherent was defined as patients who received ≥3 treatment sessions with onabotulinumtoxinA during the 2-year study; treatment nonadherent was defined as patients who received ≤2 sessions. ASPIRE did not require a specific number of treatment sessions or specify time to retreatment; therefore, a patient could be labeled “nonadherent” according to our definition despite receiving their prescribed or desired number of treatments.
ASPIRE case report forms included original questionnaires developed through expert consensus (eg, clinician satisfaction) and published validated scales (eg, Disability Assessment Scale [DAS],
Comparing the validity of the Modified Modified Ashworth Scale (MMAS) and the Modified Tardieu Scale (MTS) in the assessment of wrist flexor spasticity in patients with stroke: protocol for a neurophysiological study.
). For this analysis, 12 baseline demographic and clinical characteristic categories and 7 treatment-related variable categories were assessed (table 1). To assess treatment-related variables, data from treatment sessions 1 and 2 were compared (ie, earliest sessions after enrollment for both adherent and nonadherent patients). If patients did not have data from treatment session 2 (eg, patient no longer required treatment or patient failed to complete the assessment), data from treatment session 1 were used. Patient satisfaction and numeric pain rating scale
A comparison of change in the 0-10 numeric rating scale to a pain relief scale and global medication performance scale in a short-term clinical trial of breakthrough pain intensity.
were collected in ASPIRE but were not included in this analysis because of the extent of missing or invalid data (ie, >50% of patients did not respond to the questionnaire at treatment session 1 and/or 2).
Table 1Logistic regression model variables
Baseline Demographics and Clinical Characteristics
Adverse event data were captured for up to 108 weeks in ASPIRE and were summarized using the Medical Dictionary for Regulatory Activities version 20.0 by system organ class and preferred term. Relationship to onabotulinumtoxinA treatment was adjudicated by a panel of safety clinicians.
Clinicians were asked a series of five questions to determine their satisfaction with onabotulinumtoxinA treatment for spasticity at each subsequent treatment session. For analysis, satisfaction was categorized into the following binary variables: extremely satisfied/satisfied/neither/not applicable or dissatisfied/extremely dissatisfied.
Functional impairment was assessed using the DAS.6 Four subscales in the upper limb (ie, dressing, hygiene, limb posture, pain) and 5 subscales in the lower limb (ie, dressing, hygiene, limb posture, pain, mobility) were scored on a 4-point scale from 0 (no disability) to 3 (severe disability [normal activities limited]) by the clinician at treatment session 1 and at each subsequent treatment session. For analysis, DAS scores were categorized into the following binary variables: no/mild disability or moderate/severe disability.
Functional impairment was assessed using the DAS.6 Four subscales in the upper limb (ie, dressing, hygiene, limb posture, pain) and 5 subscales in the lower limb (ie, dressing, hygiene, limb posture, pain, mobility) were scored on a 4-point scale from 0 (no disability) to 3 (severe disability [normal activities limited]) by the clinician at treatment session 1 and at each subsequent treatment session. For analysis, DAS scores were categorized into the following binary variables: no/mild disability or moderate/severe disability.
Severity of spasticity was assessed using the MMAS, a validated and reliable measure of the intensity of spasticity.4,5 At baseline, each clinical presentation was scored on a 5-point scale from 1 (no increase in tone) to 5 (limb rigid in flexion or extension) by the clinician. For analysis, the mean MMAS score for all presentations in the upper limb, lower limb, or both limbs was used.
Treatment-related variables were assessed at sessions 1 and 2 only.
† Adverse event data were captured for up to 108 weeks in ASPIRE and were summarized using the Medical Dictionary for Regulatory Activities version 20.0 by system organ class and preferred term. Relationship to onabotulinumtoxinA treatment was adjudicated by a panel of safety clinicians.
‡ Clinicians were asked a series of five questions to determine their satisfaction with onabotulinumtoxinA treatment for spasticity at each subsequent treatment session. For analysis, satisfaction was categorized into the following binary variables: extremely satisfied/satisfied/neither/not applicable or dissatisfied/extremely dissatisfied.
§ Europe includes France, Germany, Italy, Spain, and the United Kingdom.
‖ Functional impairment was assessed using the DAS.
Li S, Francisco GE. The use of botulinum toxin for treatment of spasticity. In: Whitcup SM, Hallett M, editors. Botulinum toxin therapy. Handbook of Experimental Pharmacology. Vol 263. Cham; Springer: 2019.
Four subscales in the upper limb (ie, dressing, hygiene, limb posture, pain) and 5 subscales in the lower limb (ie, dressing, hygiene, limb posture, pain, mobility) were scored on a 4-point scale from 0 (no disability) to 3 (severe disability [normal activities limited]) by the clinician at treatment session 1 and at each subsequent treatment session. For analysis, DAS scores were categorized into the following binary variables: no/mild disability or moderate/severe disability.
¶ Severity of spasticity was assessed using the MMAS, a validated and reliable measure of the intensity of spasticity.
At baseline, each clinical presentation was scored on a 5-point scale from 1 (no increase in tone) to 5 (limb rigid in flexion or extension) by the clinician. For analysis, the mean MMAS score for all presentations in the upper limb, lower limb, or both limbs was used.
Individualized onabotulinumtoxinA treatment for upper limb spasticity resulted in high clinician- and patient-reported satisfaction: long-term observational results from the ASPIRE study.
High clinician- and patient-reported satisfaction with individualized onabotulinumtoxinA treatment for spasticity across several etiologies from the ASPIRE study.
Specific to this analysis, ASPIRE was designed for high generalizability to real-world clinical practice and included patients who were naïve or nonnaïve to botulinum toxins. Because treatment history likely affects adherence to onabotulinumtoxinA, this variable was included in the analysis. Similarly, baseline severity scores (assessed via Modified Modified Ashworth Scale) and etiology were also included.
Variables and statistical methods
Baseline demographic and clinical characteristics and treatment-related variables (see table 1) for the total population and population with stroke were analyzed using a series of logistic regression models (fig 1). Sample size limitations prevented etiologies other than stroke from being analyzed individually. Variables that achieved α level P<.2 in the univariate binary logistic regression models, as well as variables of clinical interest (eg, concomitant medications for spasticity), were combined into blocks of similar variables and analyzed using multivariate binary logistic regression. All variables that achieved α level P<.2 in the block models advanced to the final fully adjusted multivariate model. For the final model, statistical significance was accepted at P<.05 and clinically meaningful nonsignificant variables of interest at P<.1. Data are presented as odds ratios (ORs) with 95% confidence intervals (CIs). Missing data were minimal (<1%) and no imputation was performed. Statistical analyses were completed using IBM SPSS Statistics v24.0.a
Fig 1Logistic regression statistical analysis. Data for the total population and patient population with stroke were analyzed using a series of univariate and multivariate logistic regression models to obtain a final model, which idenfified variables that impacted adherence to onabotulinumtoxinA treatment from the ASPIRE study.
In ASPIRE, 730 patients received ≥1 onabotulinumtoxinA treatment for spasticity. Patients had a mean age of 53.6±15.4 years at enrollment, were nearly evenly distributed by sex (52.1% female, n=380/730), and were predominately White (77.0%, n=562/730); 36.8% of patients were naïve to botulinum toxin for spasticity. The most common etiology of spasticity was stroke (56.3%, n=411/730), followed by multiple sclerosis (MS; 16.3%, n=119/730).
Preliminary logistic regression models
Of the total population (N=730), 523 patients (71.6%) were categorized as treatment adherent and 207 patients (28.4%) as nonadherent (table 2). During the 2-year study, adherent patients had a mean ± SD of 5.3±1.6 treatment sessions, and nonadherent patients had 1.5±0.5 sessions. The mean ± SD treatment interval was 18.0±8.2 weeks for adherent patients and 22.9±15.4 weeks for nonadherent patients. The distribution of adherent and nonadherent patients across a range of treatment interval categories is shown in figure 2. Variables associated with adherence/nonadherence in the preliminary univariate models (supplemental table S1 and
Variables that met α level P<0.2 (indicated using italics) are shown in the table. Data presented as percent of patients (n) with OR and 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only.
Fig 2OnabotulinumtoxinA treatment interval for the total patient population. Data shown represent the distribution of total patients across treatment interval categories (ie, length of time between treatment sessions 1 and 2 in weeks) for adherent and nonadherent patients.
Variables associated with adherence/nonadherence in the final model
Of the total population (N=730), 626 patients had data for all variables and were included in the final model. Of those in the final model, 522 patients (83.4%) were categorized as treatment adherent and 104 patients (16.6%) as nonadherent. Adherent patients had a mean ± SD of 5.3±1.6 treatment sessions, whereas nonadherent patients had 2.0±0.0 sessions. All variables that achieved α level P<.2 in the block models (see supplemental table S3, available online only at http://www.archives-pmr.org/) were carried forward into the final model (fig 3). In the final total model, the following baseline clinical characteristics were associated with adherence: patient treated in Europe (OR=1.84; CI, 1.06-3.21; P=.030), etiology of MS (OR=2.06; CI, 0.97-4.35; P=.059), history of dysarthria (OR=2.28; CI, 0.98-5.33; P=.056), and use of orthotics (OR=1.88; CI, 1.15-3.08; P=.012). Baseline variables associated with nonadherence: history of diplopia (OR=0.28; CI, 0.09-0.89; P=.031), naïve to botulinum toxin for spasticity (OR=0.63; CI, 0.39-1.01; P=.056), and use of assistive devices (OR=0.51; CI, 0.29-0.90; P=.021). Treatment-related variables associated with adherence: treated for clenched fist (OR=1.64; CI, 0.95-2.83; P=.078). Treatment-related variables associated with nonadherence: treatment interval ≥15 weeks (OR=0.43; CI, 0.26-0.72; P=.001), moderate/severe disability on the upper limb DAS pain subscale (OR=0.56; CI, 0.30-1.03; P=.063), and clinician dissatisfaction with onabotulinumtoxinA to manage pain (OR=0.18; CI, 0.05-0.69; P=.012).
Fig 3Final model for the total patient population. Baseline demographic and clinical chracteristics and treatment-related variables that maintained α P<0.2 in the block models (see supplemental table S3, available online only at http://www.archives-pmr.org/) were carried forward into the final model shown here. Reference (abbreviated as “ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only. For interpretation of the figure, if both the upper and lower CIs are <1 (indicated with a dashed gray vertical line), the variable has a significant effect on treatment nonadherence. If both the upper and lower CIs are >1, the variable has a significant effect on treatment adherence. Statistical significance was accepted at *P<.05 and clinically meaningful nonsignificant variables of interest at #P<.1. LL, lower limb; Tx, treatment session; UL, upper limb.
In ASPIRE, 411 patients with spasticity resulting from stroke received ≥1 onabotulinumtoxinA treatment for spasticity. Patients were on average aged 58.7±14.1 years at enrollment, nearly evenly distributed by sex (50.6% male), and predominately White (75.2%); 39.4% were naïve to botulinum toxin for spasticity.
Preliminary logistic regression models
Of the population with stroke (N=411), 288 patients (70.1%) were categorized as treatment adherent and 123 patients (29.9%) as nonadherent (table 3). During the 2-year study, adherent patients had a mean ± SD of 5.3±1.6 treatment sessions, and nonadherent patients had 1.5±0.5 sessions. The mean ± SD treatment interval was 18.1±8.5 weeks for adherent patients and 23.6±16.0 weeks for nonadherent patients. The distribution of adherent and nonadherent patients across a range of treatment interval categories is shown in figure 4. As described above, variables associated with adherence/nonadherence in the preliminary univariate models (supplemental table S4 and supplemental table S5, available online only at http://www.archives-pmr.org/) and block models (supplemental table S6, available online only at http://www.archives-pmr.org/) are shown in the supplemental material.
Table 3Composition of the patient population with stroke by adherent and nonadherent status
Variables that met α level P<0.2 (indicated using italics) are shown in the table (note: if one variable in a multi-factor grouping achieved P<0.2, all variables within that grouping are shown). Data presented as percent of patients (n) with OR and 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. For etiology, the reference value is the sum of all other etiologies not being tested.
Non-Adherent% (n)
Adherent% (n)
OR (95% CI)
P Value
Country/region
Europe
20.6 ( 51)
79.4 (196)
1.49 (1.03, 2.17)
0.034
Taiwan
80.0 ( 32)
20.0 ( 8)
0.10 (0.04, 0.22)
<0.001
USA (ref.)
28.0 (124)
72.0 (319)
Caregiver relation
Spouse
24.9 ( 56)
75.1 (169)
1.05 (0.70, 1.56)
0.824
Other family
34.4 ( 43)
65.6 ( 82)
0.66 (0.42, 1.04)
0.071
Non-family
39.0 ( 30)
61.0 ( 47)
0.54 (0.32, 0.92)
0.023
No caregiver (ref.)
25.7 ( 78)
74.3 (225)
Treatment history
Naïve
36.4 ( 98)
63.6 (171)
0.54 (0.39, 0.75)
<0.001
Non-naïve (ref.)
23.6 (109)
76.4 (352)
Etiology - Stroke
Yes
29.9 (123)
70.1 (288)
0.84 (0.60, 1.16)
0.285
No (ref.)
26.3 ( 84)
73.7 (235)
Etiology - Multiple sclerosis
Yes
22.7 ( 27)
77.3 ( 92)
1.42 (0.90, 2.26)
0.135
No (ref.)
29.5 (180)
70.5 (431)
Etiology - Cerebral palsy
Yes
19.5 ( 15)
80.5 ( 62)
1.72 (0.96, 3.10)
0.070
No (ref.)
29.4 (192)
70.6 (461)
Etiology - Traumatic brain injury
Yes
42.2 ( 19)
57.8 ( 26)
0.52 (0.28, 0.96)
0.036
No (ref.)
27.4 (188)
72.6 (497)
Etiology - Spinal cord injury
Yes
28.6 ( 12)
71.4 ( 30)
0.99 (0.50, 1.97)
0.989
No (ref.)
28.3 (195)
71.7 (493)
Treatment modality - Orthotics
Yes
25.7 ( 97)
74.3 (280)
1.31 (0.95, 1.81)
0.104
Not mentioned (ref.)
31.2 (110)
68.8 (243)
Treatment modality - Surgeries/procedures
Yes
20.4 ( 19)
79.6 ( 74)
1.63 (0.96, 2.78)
0.072
Not mentioned (ref.)
29.5 (188)
70.5 (449)
Medical history - Diabetes
Yes
37.2 ( 35)
62.8 ( 59)
0.62 (0.40, 0.98)
0.042
Not mentioned (ref.)
27.0 (172)
73.0 (464)
Medical history - Diplopia
Yes
42.9 ( 9)
57.1 ( 12)
0.52 (0.21, 1.25)
0.141
Not mentioned (ref.)
27.9 (198)
72.1 (511)
Medical history - Dysarthria
Yes
19.5 ( 17)
80.5 ( 70)
1.73 (0.99, 3.01)
0.054
Not mentioned (ref.)
29.5 (190)
70.5 (453)
n, sample size; Ref, reference
Variables that met α level P<0.2 (indicated using italics) are shown in the table (note: if one variable in a multi-factor grouping achieved P<0.2, all variables within that grouping are shown). Data presented as percent of patients (n) with OR and 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. For etiology, the reference value is the sum of all other etiologies not being tested.
Variables that met α level P<0.2 (indicated using italics) are shown in the table. Data presented as percent of patients (n) with OR and 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only.
Clinician satisfaction - OnabotA helped PT/OT or exercise
Dissatisfied/extremely dissatisfied
36.4 ( 4)
63.6 ( 7)
0.34 (0.10, 1.18)
0.089
Extremely satisfied/satisfied/neither/NA (ref.)
16.2 (100)
83.8 (516)
Clinician satisfaction - Continue to use onabotA for spasticity
Probably not/definitely not
66.7 ( 2)
33.3 ( 1)
0.10 (0.01, 1.09)
0.059
Yes, Definitely/probably yes/undecided (ref.)
16.3 (102)
83.7 (522)
n, sample size; NA, not applicable; onabotA, onabotulinumtoxinA; OT, occupational therapy; PT, physical therapy; Ref, reference
Variables that met α level P<0.2 (indicated using italics) are shown in the table. Data presented as percent of patients (n) with OR and 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only.
Data presented as OR with 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only. P-values are italicized within the table if the variable achieved α level P<0.2.
OR (95% CI)
P Value
Socio-demographics
Country/region - Europe (ref. USA)
1.41 (0.97, 2.06)
0.072
Country/region - Taiwan (ref. USA)
0.08 (0.04, 0.19)
<0.001
Age at enrollment (years)
0.99 (0.98, 1.00)
0.072
Caregiver - Spouse (ref. no caregiver)
1.56 (0.99, 2.43)
0.053
Caregiver - Other family (ref. no caregiver)
0.77 (0.49, 1.23)
0.282
Caregiver - Non-family (ref. no caregiver)
0.84 (0.47, 1.49)
0.544
Etiology
Cerebral palsy
1.78 (0.98, 3.22)
0.057
Multiple sclerosis
1.44 (0.90, 2.31)
0.127
Traumatic brain injury
0.57 (0.31, 1.06)
0.074
Prior treatments
Naïve to botulinum toxin(s) for spasticity
0.55 (0.40, 0.77)
<0.001
≥1 concomitant medication for spasticity
1.15 (0.82, 1.62)
0.421
Use of assistive devices
0.71 (0.49, 1.04)
0.082
Use of intrathecal therapy
0.93 (0.53, 1.62)
0.790
Use of orthotics
1.41 (1.00, 2.00)
0.050
Use of physio/occupational therapy
0.94 (0.62, 1.42)
0.755
Prior surgeries/procedures
1.51 (0.88, 2.59)
0.135
Medical history
Diabetes
0.62 (0.39, 0.97)
0.038
Diplopia
0.41 (0.17, 1.02)
0.056
Dysarthria
1.90 (1.07, 3.38)
0.028
Treatment interval
≥15 weeks
0.58 (0.37, 0.89)
0.012
Treated clinical presentations
Upper limb - Clenched fist
1.68 (1.16, 2.42)
0.006
Upper limb - Thumb-in-palm
0.54 (0.33, 0.88)
0.014
Lower limb - Equinovarus foot
1.22 (0.86, 1.71)
0.265
Lower limb - Flexed hip
2.79 (0.95, 8.18)
0.062
Lower limb - Flexed toes
1.41 (0.82, 2.44)
0.218
DAS
Upper limb
Moderate/severe disability - Dressing
0.98 (0.56, 1.69)
0.938
Moderate/severe disability - Hygiene
0.94 (0.55, 1.62)
0.828
Moderate/severe disability - Limb posture
1.25 (0.80, 1.94)
0.332
Moderate/severe disability - Pain
0.37 (0.23, 0.59)
<0.001
Lower limb
Moderate/severe disability - Dressing
1.00 (0.56, 1.79)
0.990
Moderate/severe disability - Hygiene
0.95 (0.52, 1.72)
0.857
Moderate/severe disability - Limb posture
1.22 (0.75, 1.98)
0.420
Moderate/severe disability - Pain
0.60 (0.38, 0.94)
0.026
Moderate/severe disability - Mobility
1.28 (0.79, 2.07)
0.314
Clinician satisfaction
Dissatisfied with onabotA to manage spasticity
1.11 (0.17, 7.17)
0.911
Dissatisfied with onabotA to manage pain
0.20 (0.04, 0.90)
0.037
Dissatisfied with sustained benefit of onabotA treatment
1.21 (0.30, 4.89)
0.793
Dissatisfied with onabotA to help PT/OT or exercise
Data presented as OR with 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only. P-values are italicized within the table if the variable achieved α level P<0.2.
Fig 4OnabotulinumtoxinA treatment interval for the patient population with stroke. Data shown represent the distribution of patients with stroke across treatment interval categories (ie, length of time between treatment sessions 1 and 2 in weeks) for adherent and nonadherent patients.
Variables that met α level P<0.2 (indicated using italics) are shown in the table (note: if one variable in a multi-factor grouping achieved P<0.2, all variables within that grouping are shown). Data presented as percent of patients (n), unless otherwise indicated, with OR and 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis.
Non-Adherent% (n)
Adherent% (n)
OR (95% CI)
P Value
Country/region
Europe
18.8 ( 26)
81.2 (112)
1.89 (1.14, 3.13)
0.014
Taiwan
76.7 ( 23)
23.3 ( 7)
0.13 (0.05, 0.32)
<0.001
USA (ref.)
30.5 ( 74)
69.5 (169)
Age at enrollment (years), mean (SD)
60.5 (15.2)
57.9 (13.5)
0.99 (0.97, 1.02)
0.090
Caregiver relation
Spouse
24.8 ( 41)
75.2 (124)
1.10 (0.65, 1.85)
0.720
Other family
39.4 ( 28)
60.6 ( 43)
0.56 (0.30, 1.03)
0.061
Non-family
45.0 ( 18)
55.0 ( 22)
0.44 (0.21, 0.92)
0.030
No caregiver (ref.)
26.7 ( 36)
73.3 ( 99)
Employment status
Employed full- or part-time
20.5 ( 9)
79.5 ( 35)
1.75 (0.82, 3.77)
0.151
Not employed (ref.)
31.1 (114)
68.9 (253)
Treatment history
Naïve
35.8 ( 58)
64.2 (104)
0.63 (0.41, 0.97)
0.037
Non-naïve (ref.)
26.1 ( 65)
73.9 (184)
Treatment modality - Orthotics
Yes
24.9 ( 60)
75.1 (181)
1.78 (1.16, 2.72)
0.008
Not mentioned (ref.)
37.1 ( 63)
62.9 (107)
Treatment modality - Surgeries/procedures
Yes
16.2 ( 6)
83.8 ( 31)
2.35 (0.96, 5.79)
0.063
Not mentioned (ref.)
31.3 (117)
68.7 (257)
Medical history - Diabetes
Yes
39.5 ( 32)
60.5 ( 49)
0.58 (0.35, 0.97)
0.037
Not mentioned (ref.)
27.6 ( 91)
72.4 (239)
Medical history - Dysarthria
Yes
21.1 ( 12)
78.9 ( 45)
1.71 (0.87, 3.37)
0.118
Not mentioned (ref.)
31.4 (111)
68.6 (243)
n, sample size; Ref, reference
Variables that met α level P<0.2 (indicated using italics) are shown in the table (note: if one variable in a multi-factor grouping achieved P<0.2, all variables within that grouping are shown). Data presented as percent of patients (n), unless otherwise indicated, with OR and 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis.
Variables that met α level P<0.2 (indicated using italics) are shown in the table. Data presented as percent of patients (n), with OR and 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only.
n, sample size; NA, not applicable; onabotA, onabotulinumtoxinA; Ref, reference
Variables that met α level P<0.2 (indicated using italics) are shown in the table. Data presented as percent of patients (n), with OR and 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only.
Data presented as OR with 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only. P-values are italicized within the table if the variable achieved α level P<0.2.
OR (95% CI)
P Value
Socio-demographics
Country/region - Europe (ref. USA)
1.87 (1.11, 3.15)
0.018
Country/region - Taiwan (ref. USA)
0.09 (0.04, 0.23)
<0.001
Age at enrollment (years)
0.98 (0.96, 0.99)
0.013
Caregiver - Spouse (ref. no caregiver)
1.94 (1.07, 3.51)
0.029
Caregiver - Other family (ref. no caregiver)
0.72 (0.38, 1.38)
0.323
Caregiver - Non-family (ref. no caregiver)
0.84 (0.37, 1.90)
0.673
Employed full- or part-time (ref. not employed)
1.33 (0.57, 3.13)
0.511
Prior treatments
Naïve to botulinum toxin(s) for spasticity
0.68 (0.44, 1.07)
0.094
≥1 concomitant medication for spasticity
1.18 (0.75, 1.87)
0.482
Use of assistive devices
0.58 (0.34, 0.98)
0.042
Use of intrathecal therapy
0.87 (0.40, 1.91)
0.731
Use of orthotics
2.05 (1.28, 3.28)
0.003
Use of physio/occupational therapy
1.01 (0.56, 1.80)
0.987
Prior surgeries/procedures
2.16 (0.86, 5.42)
0.102
Medical history
Diabetes
0.58 (0.35, 0.97)
0.037
Dysarthria
1.72 (0.87, 3.38)
0.118
Treatment interval
≥15 weeks
0.58 (0.33, 1.05)
0.070
Treated clinical presentations
Upper limb - Clenched fist
1.92 (1.18, 3.13)
0.009
Upper limb - Thumb-in-palm
0.51 (0.29, 0.88)
0.015
Lower limb - Equinovarus foot
1.38 (0.87, 2.20)
0.170
Lower limb - Flexed hip
3.81 (0.46, 31.41)
0.214
Lower limb - Flexed toes
1.44 (0.68, 3.08)
0.341
DAS
Upper limb
Moderate/severe disability - Dressing
0.67 (0.35, 1.29)
0.231
Moderate/severe disability - Hygiene
1.01 (0.54, 1.90)
0.978
Moderate/severe disability - Limb posture
1.27 (0.73, 2.21)
0.406
Moderate/severe disability - Pain
0.33 (0.19, 0.59)
<0.001
Lower limb
Moderate/severe disability - Dressing
1.20 (0.55, 2.65)
0.646
Moderate/severe disability - Hygiene
1.48 (0.64, 3.43)
0.362
Moderate/severe disability - Limb posture
0.97 (0.49, 1.92)
0.931
Moderate/severe disability - Pain
0.45 (0.23, 0.89)
0.021
Moderate/severe disability - Mobility
1.55 (0.82, 2.90)
0.175
Clinician satisfaction
Dissatisfied with onabotA to manage spasticity
4.21 (0.21, 82.87)
0.344
Dissatisfied with onabotA to manage pain
0.15 (0.02, 1.05)
0.056
Dissatisfied with sustained benefit of onabotA treatment
1.32 (0.25, 6.98)
0.742
Dissatisfied with onabotA to help PT/OT or exercise
Data presented as OR with 95% CI. Reference (abbreviated as "ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only. P-values are italicized within the table if the variable achieved α level P<0.2.
Variables associated with adherence/nonadherence in the final model
Of the population with stroke (N=411), 346 patients had data for all variables and were included in the final model. Of those in the final model, 288 patients (83.2%) were categorized as treatment adherent and 58 patients (16.8%) as nonadherent. Adherent patients had a mean ± SD of 5.3±1.6 treatment sessions, and nonadherent patients had 2.0± 0.0 sessions. All variables that achieved α level P<.2 in the block models (see supplemental table S6, available online only at http://www.archives-pmr.org/) were carried forward into the final model (fig 5). In the final stroke model, the following baseline clinical characteristics were associated with adherence: patient treated in Europe (OR=2.99; CI, 1.39-6.44; P=.005), use of orthotics (OR=3.18; CI, 1.57-6.45; P=.001), and prior surgeries/procedures (OR=3.25; CI, 0.93-11.33; P=.064). Baseline characteristics associated with nonadherence: higher age at enrollment (OR=0.98; CI, 0.95-1.00; P=.097) and use of assistive devices (OR=0.46; CI, 0.20-1.03; P=.058). Treatment-related variables associated with nonadherence: treatment interval ≥15 weeks (OR=0.42; CI, 0.21-0.83; P=.013), patient treated for thumb-in-palm (OR=0.48; CI, 0.21-1.07; P=.072), and moderate/severe disability on the upper limb DAS pain subscale (OR=0.40; CI, 0.19-0.83; P=.015).
Fig 5Final model for the patient population with stroke. Baseline demographic and clinical chracteristics and treatment-related variables that maintained α P<0.2 in the block models (see supplemental table S6, available online only at http://www.archives-pmr.org/) were carried forward into the final model shown here. Reference (abbreviated as “ref.”) indicates the comparator value used for analysis. Treatment-related variables were assessed at sessions 1 and 2 only. For interpretation of the figure, if both the upper and lower CIs are <1 (indicated with a dashed gray vertical line), the variable has a significant effect on treatment nonadherence. If both the upper and lower CIs are >1, the variable has a significant effect on treatment adherence. Statistical significance was accepted at *P<.05 and clinically meaningful nonsignificant variables of interest at #P<.1. LL, lower limb; Tx, treatment session; UL, upper limb.
Individualized onabotulinumtoxinA treatment for upper limb spasticity resulted in high clinician- and patient-reported satisfaction: long-term observational results from the ASPIRE study.
ASPIRE data have increased generalizability to clinical settings and build upon evidence from previous controlled trials, in part owing to the real-world study design (ie, noninterventional, observational) and patient etiologies examined (ie, stroke, MS, cerebral palsy, traumatic brain injury, spinal cord injury). ASPIRE offers a unique opportunity to gain clinical insights into variables that can affect adherence to onabotulinumtoxinA treatment. The objective of this analysis was to identify baseline demographic and clinical characteristics and treatment-related variables that affect adherence to onabotulinumtoxinA treatment for spasticity from the ASPIRE study.
Previous publications have explored adherence to spasticity treatments.
However, to our knowledge (see supplemental table S7, available online only at http://www.archives-pmr.org/, for search terms), this is one of the first publications to assess real-world adherence to botulinum toxin treatment for spasticity
and the first to assess adherence to onabotulinumtoxinA specifically across multiple etiologies. Variables associated with adherence and nonadherence to onabotulinumtoxinA treatment in the total and stroke logistic regression models from this analysis of ASPIRE are discussed below. We propose hypotheses for each variable based on our clinical experience and, where available, published literature. Any apparent literature gaps reveal a need for increased discussion in the medical field, because these variables are likely important for care pathways.
Variables associated with adherence and nonadherence to onabotulinumtoxinA treatment in the final total model and final stroke model
In both models, onabotulinumtoxinA treatment in Europe was associated with adherence, possibly because of different health care models than in the United States. In Europe, onabotulinumtoxinA treatment costs are often fully covered by medical insurance, reducing the financial and logistical burdens for patients and clinicians, which may ultimately improve access to care and treatment persistence. The use of orthotics was also associated with adherence, which may indicate a desire by patients to reduce their dependency on, or need for, a splint or brace.
In contrast, the use of assistive devices at baseline was associated with nonadherence in both models. The use of assistive devices may indicate patients with more severe spasticity,
for which onabotulinumtoxinA treatment alone may not be sufficient, leading to reduced adherence. However, it should be noted that severe spasticity can also interfere with, or prevent the use of, assistive devices. Spasticity-related pain in the upper limb, as assessed by DAS,
The efficacy of botulinum toxin A for spasticity and pain in adults: a systematic review and meta-analysis using the Grades of Recommendation, Assessment, Development and Evaluation approach.
suggesting that pain relief may be an appropriate secondary goal of onabotulinumtoxinA treatment. A treatment interval ≥15 weeks between sessions 1 and 2 was associated with nonadherence. According to the package insert, onabotulinumtoxinA should be administered when the effect of the previous injection has diminished and is anticipated to occur approximately every 12 weeks.
Longer treatment intervals could be due to a patient's lack of logistical support to participate in treatment (eg, owing to burdened caregivers) and/or other barriers to care (eg, medical complications or mental health factors) that make it difficult to adhere to, or participate in, treatment.
A previous study found that the second most common determinant for discontinuation in patients treated with botulinum toxin for MS-related spasticity was “logistic problems or barriers to reach the structure [MS center].”
(p1844) Lee et al support this finding, suggesting that “incapability to return to the clinic owning to organizational issues (eg, transportation, especially for more disabling disorders such as SPAS [spasticity])” may have negatively affected long-term adherence to botulinum toxin treatment in their study.
(p24) Alternatively, longer treatment intervals may be consistent with the patient's prescribed treatment regimen.
Variables associated with adherence and nonadherence to onabotulinumtoxinA treatment in the final total model only
In the total model, which includes patients with stroke, MS, cerebral palsy, traumatic brain injury, and spinal cord injury, history of dysarthria was associated with adherence. Dysarthria could be indicative of medullary involvement leading to greater motor dysfunction,
which may be more responsive to botulinum toxin treatment for spasticity, leading to higher adherence. MS as the primary etiology of spasticity was also associated with adherence in the total model. Owing to the early age of onset and the nature of their disease being chronic, as well as often progressive if not treated effectively,
patients with MS may have higher motivation to adhere to prescribed treatment compared with the other etiologies in ASPIRE, especially because older age and longer duration of MS have been associated with higher severity,
was also associated with adherence in the total model.
In contrast, history of diplopia was associated with nonadherence in the total model, which could indicate a lesion involving the midbrain leading to ataxic movement disorders,
which may not be as responsive to spasticity treatments. Clinician dissatisfaction with onabotulinumtoxinA to manage pain was associated with nonadherence in the total model, reinforcing that pain relief may be more appropriate as a secondary goal of onabotulinumtoxinA treatment for spasticity. Being naïve to botulinum toxin for spasticity was associated with nonadherence, which emphasizes the need for early patient education concerning onabotulinumtoxinA treatment goals and expectations.
Unrealistic expectations from patients, family members, and/or caregivers have been cited as one of the most common reasons for poor response to botulinum toxin therapy for spasticity management.
Li S, Francisco GE. The use of botulinum toxin for treatment of spasticity. In: Whitcup SM, Hallett M, editors. Botulinum toxin therapy. Handbook of Experimental Pharmacology. Vol 263. Cham; Springer: 2019.
Variables associated with adherence and nonadherence to onabotulinumtoxinA treatment in the final stroke model only
Prior surgeries or procedures were associated with adherence in the stroke model, which may indicate a patient's greater involvement with multimodal spasticity management.
was associated with nonadherence. Lee et al postulated that older age and disease progression (eg, additional strokes or other comorbidities) could be contributing factors to reduced long-term adherence to botulinum toxin treatment.
Patients being treated for thumb-in-palm, which can be a difficult clinical presentation to treat because of inaccessibility of the target muscles (especially if accompanied by clenched fist), was also associated with nonadherence in the stroke model.
Study limitations
Limitations common to real-world observational studies were discussed in previous ASPIRE publicatons,
Individualized onabotulinumtoxinA treatment for upper limb spasticity resulted in high clinician- and patient-reported satisfaction: long-term observational results from the ASPIRE study.
High clinician- and patient-reported satisfaction with individualized onabotulinumtoxinA treatment for spasticity across several etiologies from the ASPIRE study.
including the lack of control over study elements, patient dropout due to study length, and the effect of confounding factors on data analysis and interpretation. Specific to this analysis, treatment-related variables were assessed at treatment sessions 1 and 2 only based on the treatment adherence/nonadherence definitions. Data gathered during this time frame may not fully represent treatment outcomes at later time points. In addition, ASPIRE was designed to include approximately one-third of patients who were naïve to botulinum toxin for spasticity and two-thirds who were nonnaïve/continuing botulinum toxin treatment,
which may have skewed the patient population in favor of those adherent to onabotulinumtoxinA treatment. Owing to sample size limitations, separate analyses for specific etiologies other than stroke were not done. Lastly, a less stringent threshold was applied at the univariate and multivariate block phases of the analysis (P<.2) to ensure that potentially important variables were not prematurely removed from the model because of low sample size or heterogeneity in the data set. Importantly, the more stringent P<.05 was applied at the final model stage to ensure the robust identification of variables that affected onabotulinumtoxinA treatment adherence in this study.
Conclusions
In ASPIRE, most patients adhered to onabotulinumtoxinA treatment for spasticity, with adherent patients having an average >5 treatment sessions during the 2-year study. These analyses provide real-world insights to improve adherence to onabotulinumtoxinA treatment, including use of orthotics and treatment in Europe. In addition, these analyses further elucidate variables associated with nonadherence, including a retreatment interval ≥15 weeks, use of assistive devices, and moderate/severe disability on the upper limb DAS pain subscale, for which clinicians should pay particular attention to better support their patients. Increased knowledge of variables that affect onabotulinumtoxinA treatment adherence can help to optimize spasticity management strategies to improve patient care.
Suppliers
a
IBM SPSS Statistics, version 24.0, IBM.
Acknowledgments
We thank the investigators who participated in this study. Writing and editorial assistance were provided by Monica R.P. Elmore, PhD and K.M. Hirahatake, PhD, of AbbVie.
References
Pandyan AD
Gregoric M
Barnes MP
et al.
Spasticity: clinical perceptions, neurological realities and meaningful measurement.
Li S, Francisco GE. The use of botulinum toxin for treatment of spasticity. In: Whitcup SM, Hallett M, editors. Botulinum toxin therapy. Handbook of Experimental Pharmacology. Vol 263. Cham; Springer: 2019.
Individualized onabotulinumtoxinA treatment for upper limb spasticity resulted in high clinician- and patient-reported satisfaction: long-term observational results from the ASPIRE study.
High clinician- and patient-reported satisfaction with individualized onabotulinumtoxinA treatment for spasticity across several etiologies from the ASPIRE study.
Comparing the validity of the Modified Modified Ashworth Scale (MMAS) and the Modified Tardieu Scale (MTS) in the assessment of wrist flexor spasticity in patients with stroke: protocol for a neurophysiological study.
A comparison of change in the 0-10 numeric rating scale to a pain relief scale and global medication performance scale in a short-term clinical trial of breakthrough pain intensity.
The efficacy of botulinum toxin A for spasticity and pain in adults: a systematic review and meta-analysis using the Grades of Recommendation, Assessment, Development and Evaluation approach.
Current affiliation for Wuwei Fang, Duke University School of Medicine, Durham, NC. Current affiliation for George F. Wittenberg, University of Pittsburgh, Pittsburgh, PA.
Presented to the American Congress of Rehabilitation Medicine (ACRM), November 5–8, 2019, Chicago, IL; the International Congress of Parkinson's Disease and Movement Disorders (MDS), September 22–26, 2019, Nice, France; the American Academy of Physical Medicine and Rehabilitation (AAPM&R), November 14–17, 2019, San Antonio, TX; the Rehabilitation Medicine Society of Australia and New Zealand (RMSANZ), October 20–23, 2019, Adelaide, Australia; the UK Stroke Forum (UKSF), December 3–5, 2019, Telford, United Kingdom; the International Society of Physical and Rehabilitation Medicine and Association of Academic Physiatrists (ISPRM/AAP), March 4–9, 2020, Orlando, FL; the American Academy of Neurology (AAN), April 25–May 1, 2020, Toronto, Canada; the MDS, September 12–16, 2020, virtual; the ACRM, October 21–24, 2020, virtual; the AAPM&R, November 8–December 7, 2020, virutal; the European Stroke Organisation and World Stroke Organization (ESO-WSO), November 7–9, 2020, virutal; the Americas Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS)–European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS), September 11–13, 2020, virtual; the World Congress for Neurorehabilitation and the French Society of Physical and Rehabilitation Medicine (WFNR-SOFMER), October 7–10, 2020, virtual; and the International Neurotoxin Association, June 12–14, 2020, virtual.
Supported by Allergan, prior to its acquisition by AbbVie.
Clinical Trial Registration No.: NCT01930786.
AbbVie is committed to responsible data sharing regarding the clinical trials we sponsor. This includes access to anonymized, individual, and trial-level data (analysis data sets), as well as other information (eg, protocols and clinical study reports), as long as the trials are not part of an ongoing or planned regulatory submission. This includes requests for clinical trial data for unlicensed products and indications.
These clinical trial data can be requested by any qualified researchers who engage in rigorous, independent scientific research and will be provided following review and approval of a research proposal and Statistical Analysis Plan (SAP) and execution of a Data Sharing Agreement (DSA). Data requests can be submitted at any time and the data will be accessible for 12 months, with possible extensions considered. For more information on the process, or to submit a request, visit the following link: https://www.abbvie.com/our-science/clinical-trials/clinical-trials-data-and-information-sharing/data-and-information-sharing-with-qualified-researchers.html.
Disclosures: Financial arrangements of the authors with companies whose products may be related to the present publication are listed, as declared by the authors. Alberto Esquenazi consulted for Allergan, an AbbVie company; Ipsen; and Merz and received research grants from Allergan, an AbbVie company, and Ipsen. Gerard E. Francisco consulted for, and received research grants from, Allergan, an AbbVie company; Ipsen; Merz; Microtransponder; Ottobock; Parker Hannifin; Revance; Re-Walk; Saol Therapeutics; Shionogi; and Sword Health and received research grants from the National Institutes of Health–National Institute of Neurological Disorders and Stroke; the National Institute on Disability, Independent Living, and Rehabilitation Research; Mission Connect, a Project of TIRR Foundation; and the Memorial Hermann Foundation. Wuwei Feng received research grants from the National Institutes of Health (NIH) and American Heart Association/American Stroke Association (AHA/ASA). Alessio Baricich consulted for, and received research grants from, Allergan, an AbbVie company; Ipsen; and Merz. Philippe Gallien received research grants from Allergan, an AbbVie company; Merz; and Ipsen. Kristina Fanning is an employee of Vedanta Research, which has received support funded by Allergan, an AbbVie company; Dr Reddy's Laboratories/Promius Pharma; Amgen; and Mallinckrodt. Aleksej Zuzek is a full-time employee of AbbVie. Daniel S. Bandari is a consultant and/or speaker for Biogen, Genentech, Genzyme, and EMD-Serono and received research support from Biogen, Genentech, and Med-day. George F. Wittenberg was on the Clinical Advisory Board of the Battelle Memorial Foundation.
Treatment-related variables were assessed at sessions 1 and 2 only.
00497-4&id=gr1.jpg){kind=link}
00497-4&id=gr2.jpg){kind=link}
00497-4&id=gr3.jpg){kind=link}
00497-4&id=gr4.jpg){kind=link}
00497-4&id=gr5.jpg){kind=link}