CLINICAL RESEARCH
QUALITY OF LIFE IN AMYOTROPHIC LATERAL SCLEROSIS PATIENTS AND CAREGIVERS: IMPACT OF ASSISTIVE COMMUNICATION FROM EARLY STAGES ANA LONDRAL, MSc,1 ANABELA PINTO, MD, PhD,2 SUSANA PINTO, MD, PhD,1 LUIS AZEVEDO, PhD,3 and MAMEDE DE CARVALHO, MD, PhD4 Translational Clinical Physiology Unit, Instituto de Medicina Molecular, Avenida Professor Egas Moniz, University of Lisbon, 1649-028 Lisbon, Portugal 2 Department of Medical Rehabilitation, Hospital de Santa Maria–Centro Hospitalar Lisboa Norte (CHLN), Lisbon, Portugal 3 Center of Acquisition and Signal Processing, Instituto Superior T ecnico, University of Lisbon, Lisbon, Portugal 4 Department of Neurosciences, Hospital de Santa Maria–CHLN, Lisbon, Portugal 1
Accepted 16 March 2015 ABSTRACT: Introduction: In this study we performed a longitudinal investigation to assess the impact of early introduction of assistive communication devices (ACDs) on quality of life (QoL) in amyotrophic lateral sclerosis (ALS) patients and their caregivers. Methods: Patients were followed for 7–10 months (3 evaluation periods). Bulbar-onset ALS patients (N 5 27) and paired caregivers (N 5 17) were included. Fifteen randomly selected patients received early support in ACD use. Patients were assessed using the ALS Functional Rating Scale—revised (ALSFRS-R), the McGill QoL (MQoL), the Communication Effectiveness Index (CETI), and performance in writing; and caregivers were assessed with the MQoL and World Health Organization Quality of Life questionnaire (WHOQOL-BREF). Results: Patients with early support had higher MQoL Psychological and MQoL Existential well-being domains; caregivers had higher MQoL Support domain and their MQoL Psychological domain positively associated with patient CETI. Most patients could communicate using a touchscreen keyboard to write, even when handwriting and speech were not possible. Conclusion: Early intervention with an ACD seems to have a positive impact on QoL and gives patients the opportunity to improve skills for communication in later disease stages. Muscle Nerve 52: 933–941, 2015
Recent recommendations for clinical management in amyotrophic lateral sclerosis (ALS) indicate that patient autonomy and ability to communicate should be promoted.1,2 Accordingly, communication has been rated by ALS patients as a key domain for their independence.3 Severe dysarthria has major impact on the quality of life (QoL) of ALS patients, as most patients become unable to communicate through speech at Additional Supporting Information may be found in the online version of this article. Abbreviations: ACD, assistive communication device; ALS, amyotrophic lateral sclerosis; ALSFRS-R, ALS Functional Rating Scale—revised; ALSFRS-R-b, bulbar subscore of the ALSFRS-R; ALSFRS-R-ul, upper limbs subscore of the ALSFRS-R; CETI-m, Communication Effectiveness Index—modified; G1, group 1; G2, group 2; MQoL, McGill Quality of Life; QoL, quality of life; UL, upper limb; WHOQoL-BREF, World Health Organization Quality of Life questionnaire; wpm, words per minute Key words: amyotrophic lateral sclerosis; augmented and alternative communication; assistive device; caregiver; quality of life This work was supported by national funds provided by the Calouste Gulbenkian Foundation and Portuguese Association of ALS Patients (APELA). Correspondence to: A. Londral; e-mail: [email protected] C 2015 Wiley Periodicals, Inc. V
Published online 25 March 2015 in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/mus.24659
Communication and QoL in ALS
some stage of the disease in spite of speech therapy interventions.4,5 Early deterioration of speech intelligibility6 and later involvement of upper limbs (ULs) leads bulbar-onset patients to rapidly replace speech by handwriting (often a pen and a notepad for peerto-peer communication), until UL weakness demands new strategies for communication.7,8 As the disease progresses, most patients must replace speech communication by communication based on text input (handwriting, use of computer, or paper-based letter board). Modern tablet devices used as assistive communication devices (ACDs), with touchscreen input, can allow bulbar-onset patients to maintain autonomy in communication through simple applications with speech synthesizers. Nevertheless, ACD use is strongly dependent on the decisions and expectations of patients, their caregivers, and health professionals. As long as residual capacity for communicating through speech or handwriting is present, the use of communication devices may not be considered. When communication alternatives are insufficient (severe dysarthria associated with UL dysfunction), technologically more advanced methods need to be explored. Timely referral and economic burden remain critical issues in decisionmaking for communication support in ALS. However, late referral is a primary reason for low acceptance; indeed, adaptation of patients with marked generalized weakness to new strategies (e.g., eyetracking or even brain–computer interfaces) can be a difficult process, especially when there is lack of experience or support for continuous training in the use of these ACDs.7,9 When patient disability is severe, ACD use relies on caregiver support for daily setup and maintenance. For this reason, caregiver needs, skills, and expectations must be considered as important factors in clinical intervention with ACDs.9,10 Difficulty in understanding patient needs and feelings is the chief caregiver motivation for supporting intervention with an ACD.11 Furthermore, patient autonomy regarding communication has been MUSCLE & NERVE
December 2015
933
Table 1. Questionnaires and tests applied and their respective ranges. Test/questionnaire ALSFRS-R ALSFRS-R-b ALSFRS-R-ul WHOQoL-bref MQoL MQoL-SIS (single-item score) Communication Effectiveness Index—modified (CETI-m) Performance in typing Performance in handwriting Questionnaire for patients and caregivers: How familiar are you with computer devices?
Range
Time*
(maximum disability) to 48 (normal) (maximum disability) to 12 (no bulbar symptoms) (maximum disability) to 12 (no upper limitations) (poorest QoL) to 100 (max QoL) (lowest QoL) to 10 (highest QoL) 0 (very bad) to 10 (excellent) 0 (no communication) to 100 (normal) Words per minute (wpm) wpm 0 (I never used a computer) to 5 (I am an expert)
T0, T1, T2 T0, T1, T2 T0, T1, T2 T0, T2 T0, T1, T2
0 0 0 0 0
T0, T1, T2 T0, T1, T2 T0
*T0: at study entry; T1: 3–4 months after entry; T2: 7–10 months after entry.
reported as an essential family caregiver need that may influence the caregiver’s own QoL.12 The primary objective of this study was to explore the impact of early support with ACDs on QoL of ALS patients with bulbar onset and their caregivers. The score for QoL of both patients and caregivers was the main outcome. A secondary objective was to evaluate the efficiency of ACDs in supporting speech dysfunction. As secondary outcomes, we monitored patients’ functional scores, their self-perceived communication effectiveness, and upper limb function for performance of written communication (handwriting and typing on a keyboard). METHODS Patients. ALS patients were enrolled with the following inclusion criteria: (1) probable or definite disease according to the revised El Escorial criteria; (2) bulbar-onset disease, as defined by initial symptoms of dysarthria or dysphagia; (3) bulbar score for dysarthria of 3 or 2 (mild or moderate impairment), as given by the first question of the ALS Functional Rating Scale—revised (ALSFRS-R); and (4) informed consent. Patients with other medical conditions, respiratory symptoms, or clinical evidence of dementia were excluded. A main caregiver was identified by the patient as the person who provides daily support and accompanies the patient to clinical visits (frequently the husband/ wife or a sibling). The caregivers did not change over the duration of this study. The study was approved by the local ethics committee, and patients and caregivers gave written informed consent before participation.
Patients were assessed 3 times: at entry (T0); 3–4 months later (T1); and 7–10 months after entry (T2). Caregivers were assessed 2 times, at T0 and T2. Most assessments were done in our clinic, but some were done in the patient’s home. At each visit, the patient and caregiver were interviewed separately. The duration of the complete assessment was approximately 30 minutes. At T0,
Procedure.
934
Communication and QoL in ALS
patients were also randomized into 2 groups according to the intervention time [group 1 (G1): early intervention, patients received an ACD just after baseline assessment; group 2 (G2): late intervention, patients received communication devices according to the conventional approach at our center (referral when dysarthria score reaches 0 or 1 on the ALSFRS-R, which is anarthria or severe dysarthria)]. The ACD used was a small and light commercially available touchscreen-tablet device (approximately 200 3 130 mm). We used a software program based on a simple onscreen keyboard for text-to-speech communication with text prediction. This software was developed commercially for the specific purpose of text-to-speech communication and is available to download from the internet; for our purposes, an important feature was that the software could be used with different access strategies in addition to the touchscreen. Patients and caregivers were trained and supported in the use of the device during clinical appointments. Patients who had an internet connection at home were instructed to use e-mail communication through the ACD. E-mail communication was tested and stimulated by sending emails (demanding reply) periodically during the study. Patients were assessed with the following questionnaires: the McGill Quality of Life questionnaire (MQoL); the ALSFRS-R; and the modified Communication Effectiveness Index (CETI-m). Caregivers completed the MQoL and World Health Organization Quality of Life (WHOQoL-BREF) questionnaires, related to their own QoL. Both patients and caregivers answered a single question, rated on a 0–5 Likert-type scale, to assess their previous experience with computer devices (Table 1). of Life. MQoL Questionnaire. The MQoL13 consists of a questionnaire with 16 items, each rated on a Likert-type scale from 0 to 10. Five domains are calculated as the mean scores of 1–6 items: 2 health-related domains [physical symptoms Quality
MUSCLE & NERVE
December 2015
and physical well-being (MQoL-Phys)] and 3 non– health-related domains [psychological symptoms (MQoL-Psych); existential well-being (MQoL-Exist); support, related to the feeling of being cared for (MQoL-Sup)]. A total score (MQoL-Tot) was calculated as the mean of the scores in each domain (higher scores indicate higher QoL). In addition, this instrument includes a single-index score (MQoL-SIS), which measures overall quality of life, and a descriptive section (MQoL-part D), where respondents list what had the greatest impact (positive or negative) on their QoL in the past 2 days. MQoL is sensitive to psychological, supportive, and spiritual factors more than to health-related QoL measures.14 It has been used in various studies related to QoL in ALS patients.15–18 WHOQoL-BREF. The WHOQoL-BREF is a 26item questionnaire for measuring QoL in 4 domains: (1) physical health and well-being; (2) psychological health and well-being; (3) social relations; and (4) environment (related to financial resources, health and social services, home and physical environment, among others). Items are rated on a 1–5 Likert-type scale, and scores are calculated on a 100-point scale (higher scores correspond to better QoL). Ratings range from 1 (very poor) to 5 (very good).19–21 This questionnaire has been used extensively to investigate QoL of patients and caregivers in different chronic diseases, including ALS.22–24 Functional Measurements. The ALSFRS-R was applied at each visit. It is an instrument used for monitoring progression of functional disability in patients with ALS.25 Each question on the ALSFRSR (total of 12 questions) is rated from 4 (normal function) to 0, according to progressive functional impairment. We calculated the total score (ALSFRS-R-tot), which ranges from 48 (normal) to 0, and 2 subscores: bulbar function (ALSFRS-Rb)26 and upper limb function (ALSFRS-R-ul).27 Subscores range from 12 to 0 and were calculated by summing the first 3 questions and the fourth to sixth questions of the ALSFRS-R, respectively. The ALSFRS-R-b rates speech, salivation, and swallowing functions; the ALSFRS-R-ul rates UL function related to handwriting, handling utensils in daily activities, and dressing. The CETI-m28,29 is an adaptation of the original CETI designed for patients with aphasia. It is a valid measure of functional communication and is sensitive to changes in performance over time.30 The CETI-m rates self-perceived limitation of a person when communicating; it consists of 10 questions related to communication in different contextual situations rated on a visual Likert-type scale. The total score is given by the sum of the individual scores, presented on a 100-point scale, Communication and QoL in ALS
representing the minimum (0) to maximum (100) communication effectiveness. Performance Measures. To monitor performance of the ULs in written communication at each evaluation session, patients were asked to write, sequentially, 3 different sentences with 2 different modalities: handwriting and using a keyboard. Both performance of handwriting and typing on a keyboard were calculated as the mean rate of the writing tasks, as measured by words per minute (wpm). Statistical Analysis. Primary outcomes were scores from the MQoL (single-index item, total score, physical, psychological health, existential wellbeing, and support) and WHOQoL-BREF scores (4 domains: physical health and well-being; psychological health and well-being; social relations; and environment). Secondary outcomes were ALSFRSR-tot, ALSFRS-R-b, ALSFRS-R-ul, CETI-m, typing rate, and handwriting rate. Descriptive statistics were used to describe participants and the results of the single question regarding previous computer use. Pearson and Spearman correlation coefficients were used to examine associations among different variables for normal and non-normal data distributions, respectively. Changes in outcome variables, ranging from T0 to T2, were analyzed by paired-sample t-test and Wilcoxon signed-rank test. The t-test and Mann– Whitney U-test for independent samples were used to test differences between the 2 different groups (G1 vs. G2). P < 0.05 was considered statistically significant. For each of the statistical analyses, we used Bonferroni adjustment of the P-value for multiple comparisons. However, for the exploratory purpose of this study, we considered P < 0.05 as a trend that was worth reporting. RESULTS
Twenty-seven ALS patients, mean age 64.8 (SD 10.2) years (range 39–83 years), were included in the study. Among this group, 1 declined immediately after baseline assessment, and 5 were evaluated only twice due to rapid disease progression (Fig. 1). Only 17 caregivers were interviewed (4 patients came to clinical visits alone or with different caregivers, and 6 caregivers declined to fill in the questionnaires). Most of the patients were women (81.5%), and previous experience in the use of computers, rated by patients from 0 to 5, was generally low (median 5 2). Low school education and poor previous experience with technology of most of the participants coincided with the demographic characteristics of the Portuguese population within this age group. Table 2 describes the demographic data of participants. MUSCLE & NERVE
December 2015
935
FIGURE 1. Flowchart of the recruitment process. N, number of patients; Nc, number of caregivers.
Patients in G1 started to use an ACD after baseline assessment. There was no record of spontaneous use of assistive communication devices in patients randomized to G2. For this exploratory analysis of the impact of communication in QoL, we analyzed each domain of QoL questionnaires and compared scores between different assessments and groups of patients and caregivers. As many variables were compared, no significant result was found when considering adjusted P-values for multiple comparisons. However, taking into account the exploratory aim of this study, we considered P < 0.05 to be a trend with regard to the influence of communication on QoL. Mean scores of all outcome variables and results of all statistical comparisons are detailed in Tables S1 and S2 (refer to Supplementary Material, available online). QoL. Patients. Mean values calculated for all patients (through MQoL) showed a decrease in total, physical, and existential domains of QoL (P < 0.05). Single-index score, psychological symptoms, and support domains did not change with time (P > 0.05). In descriptive parts of the MQoL (physical symptoms and part D), 23 patients (88.5%) referred in at least 1 assessment to speech problems or difficulty being understood by others as negative factors in QoL; 2 patients in G1 referred to communication device as having a positive impact in QoL. Patients with early intervention with their ACD had higher MQoL scores than the other patients, particularly for the existential well-being domain at T1 (P 5 0.045) and the psychological symptoms domain, both at T1 (P 5 0.047) and T2 (P 5 0.032) (Table 3). 936
Communication and QoL in ALS
Caregivers. The mean scores for caregiver QoL (MQoL) decreased from T0 to T2, although only the single-index score (MQoL-SIS) attained a difference with P < 0.05. The mean scores of the WHOQoL-BREF showed a decrease (P < 0.05) in all domains, from T0 to T2, except for the social relations domain (P 5 0.749). In the descriptive item of the MQoL (part D), 8 (47%) caregivers referred to patient speech problems as having a negative impact on their own QoL; 5 of 11 (46%) caregivers of patients in G1 believed the ACD had a positive impact on their QoL. When comparing results from caregivers of patients in G1 vs. G2, we identified a higher QoL in the former group in the domain of support (P 5 0.016) (Table 3). No differences between the
Table 2. Characteristics of all participants at baseline (T0): patients and caregivers.
Total (N) Gender (W/M) Age, in years [mean 6 SD (range)] Education Primary school Elementary/high school University Relation of caregiver Husband/wife Sibling Experience with computers* Median (range) Internet access at home
Patients
Caregivers
27 22/5 64.8 6 10.2 (39–83)
17 6/11 57.52 6 13.9 (26–81)
12 (44.4%) 10 (37%) 5 (18.5%)
5 (22.7%) 6 (27.3%) 6 (27.3%) 10 7
2 (0–4) 12 (44%)
3 (0–5)
*Experience in using computers (minimum 5 0, maximum 5 5).
MUSCLE & NERVE
December 2015
Table 3. Mean values of dependent variables tested in each evaluation period (T0, T1, T2) comparing G1 and G2 results. T0
CETI-m MQoL-psych (patient) MQoL-exist (patient) MQoL-sup (caregiver)
T1
T2
G1
G2
G1
G2
G1
G2
64.67 6 27.26 5.34 6 2.21 7.25 6 1.60 6.30 6 1.44
71.33 6 31.92 4.68 6 1.57 7.00 6 2.03 4.42 6 1.83
44.53 6 20.63 5.61 6 2.20 7.52 6 1.37 NA
47.11 6 29.31 4.01 6 1.39* 6.40 6 1.24* NA
48.83 6 16.24 4.93 6 1.59 6.53 6 1.65 6.71 6 1.41
17.20 6 11.80* 3.18 6 1.10* 5.02 6 1.15* 3.92 6 2.13*
Comparisons are shown where there was a trend (P < 0.05) for the influence of the ACD on the quality of life for patients and caregivers. There were no statistically significant differences between the 2 groups when adjusting for multiple comparisons (Bonferroni method). NA, not applicable; CETI-m, Communication Effectiveness Index—modified (from 0 to 100); MQoL-psych, MQoL psychological symptoms subscore (lowest 5 0, highest 5 10); MQoL-exist, MQoL existential well-being subscore (lowest 5 0, highest 5 10); MQoL-sup, MQoL support subscore (lowest 5 0, highest 5 10). *P < 0.05 for comparison between G1 and G2 (independent-samples t-test).
2 groups (P > 0.05) were found in any of the WHOQoL-BREF domains. Caregiver MQoL scores and patient MQoL scores were not correlated in any of the periods studied (P > 0.05). Functional Measurements. As expected, evaluation of bulbar function (ALSFRS-R-b score) decreased significantly over time (Table 4). The decline was linear and showed no difference in percentage change between T0–T1 and T1–T2. We found no significant difference between the 2 patient groups in bulbar score decline. The higher the bulbar impairment (lower ALSFRS-R-b scores), the lower the self-perceived communication effectiveness index (CETI-m) for patients in G1 (r 5 0.419, P < 0.001) and G2 (r 5 0.809, P < 0.001). However, when considering data from patients in G1, after they started to use the ACD (T1 and T2), patients with lower bulbar functional scores did not necessarily have lower self-perceived communication (r 5 0.101, P 5 0.629). In general, we found positive correlations between patient QoL, bulbar function, and selfperceived communication; the higher the bulbar dysfunction (lower ALSFRS-b score), the lower the QoL, both in total score (r 5 0.297, P 5 0.019) and in single-item score (r 5 0.280, P 5 0.028). Patients with higher scores in self-perceived communication effectiveness had higher QoL, both in total score
(r 5 0.268, P 5 0.027) and in the psychological symptoms domain (r 5 0.265, P 5 0.029). No correlations were found between caregiver QoL scores and patient bulbar function (ALSFRSR-b) or communication (CETI-m) scores, for the whole group of caregivers. However, when analyzing caregivers of patients who received early ACD intervention, we found that patients with higher self-perceived communication had caregivers with higher QoL scores for the psychological symptoms domain of MQoL (r 5 0.515, P 5 0.011). Performance Measures. Handwriting performance showed a mean of 17.5 (SD 8.1) wpm (range 7.6– 36.4) at T0, decreasing significantly (P 5 0.002) from baseline assessment to T2 [mean 12.4 (SD 10) wpm, range 0–27.4 wpm] (Table 4). The lower the UL function (ALSFRS-R-ul), the lower the measured handwriting performance (r 5 0.693, P < 0.001). Due to symptoms of disease progression in ULs, 5 patients were unable to perform handwriting at T1 or T2, although they could still use the ACD for typing. These patients had difficulty grasping a pen, but were able to use 1 finger on the keyboard or touchscreen. Performance using a keyboard varied widely among patients due to previous experience with using computer devices [mean 9.2 (SD 5.2), range 2.4–18.7 wpm)] at baseline assessment. During the period of study, patient typing performance did
Table 4. Mean values for secondary outcome variables tested in this study during each assessment period (T0, T1, T2).
ALSFRS-R ALSFRS-R-b ALSFRS-R-ul Handwriting rate (wpm) Typing rate (wpm) CETI-m
T0
T1
T2
P(T0,T2)
27.18 6 8.54 6.32 6 2.71 9.45 6 3.04 16.2 6 6.6 9.2 6 5.2 67.17 6 28.6
22.15 6 8.96 4.75 6 2.47 8.25 6 4 14.1 6 9.8 8.5 6 5.4 45.50 6 23.65
15.50 6 7.88 2.75 6 2.05 6.44 6 4.34 10.7 6 9.8 6.7 6 4.9 39.53 6 20.9
QUALITY OF LIFE IN AMYOTROPHIC LATERAL SCLEROSIS PATIENTS AND CAREGIVERS: IMPACT OF ASSISTIVE COMMUNICATION FROM EARLY STAGES ANA LONDRAL, MSc,1 ANABELA PINTO, MD, PhD,2 SUSANA PINTO, MD, PhD,1 LUIS AZEVEDO, PhD,3 and MAMEDE DE CARVALHO, MD, PhD4 Translational Clinical Physiology Unit, Instituto de Medicina Molecular, Avenida Professor Egas Moniz, University of Lisbon, 1649-028 Lisbon, Portugal 2 Department of Medical Rehabilitation, Hospital de Santa Maria–Centro Hospitalar Lisboa Norte (CHLN), Lisbon, Portugal 3 Center of Acquisition and Signal Processing, Instituto Superior T ecnico, University of Lisbon, Lisbon, Portugal 4 Department of Neurosciences, Hospital de Santa Maria–CHLN, Lisbon, Portugal 1
Accepted 16 March 2015 ABSTRACT: Introduction: In this study we performed a longitudinal investigation to assess the impact of early introduction of assistive communication devices (ACDs) on quality of life (QoL) in amyotrophic lateral sclerosis (ALS) patients and their caregivers. Methods: Patients were followed for 7–10 months (3 evaluation periods). Bulbar-onset ALS patients (N 5 27) and paired caregivers (N 5 17) were included. Fifteen randomly selected patients received early support in ACD use. Patients were assessed using the ALS Functional Rating Scale—revised (ALSFRS-R), the McGill QoL (MQoL), the Communication Effectiveness Index (CETI), and performance in writing; and caregivers were assessed with the MQoL and World Health Organization Quality of Life questionnaire (WHOQOL-BREF). Results: Patients with early support had higher MQoL Psychological and MQoL Existential well-being domains; caregivers had higher MQoL Support domain and their MQoL Psychological domain positively associated with patient CETI. Most patients could communicate using a touchscreen keyboard to write, even when handwriting and speech were not possible. Conclusion: Early intervention with an ACD seems to have a positive impact on QoL and gives patients the opportunity to improve skills for communication in later disease stages. Muscle Nerve 52: 933–941, 2015
Recent recommendations for clinical management in amyotrophic lateral sclerosis (ALS) indicate that patient autonomy and ability to communicate should be promoted.1,2 Accordingly, communication has been rated by ALS patients as a key domain for their independence.3 Severe dysarthria has major impact on the quality of life (QoL) of ALS patients, as most patients become unable to communicate through speech at Additional Supporting Information may be found in the online version of this article. Abbreviations: ACD, assistive communication device; ALS, amyotrophic lateral sclerosis; ALSFRS-R, ALS Functional Rating Scale—revised; ALSFRS-R-b, bulbar subscore of the ALSFRS-R; ALSFRS-R-ul, upper limbs subscore of the ALSFRS-R; CETI-m, Communication Effectiveness Index—modified; G1, group 1; G2, group 2; MQoL, McGill Quality of Life; QoL, quality of life; UL, upper limb; WHOQoL-BREF, World Health Organization Quality of Life questionnaire; wpm, words per minute Key words: amyotrophic lateral sclerosis; augmented and alternative communication; assistive device; caregiver; quality of life This work was supported by national funds provided by the Calouste Gulbenkian Foundation and Portuguese Association of ALS Patients (APELA). Correspondence to: A. Londral; e-mail: [email protected] C 2015 Wiley Periodicals, Inc. V
Published online 25 March 2015 in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/mus.24659
Communication and QoL in ALS
some stage of the disease in spite of speech therapy interventions.4,5 Early deterioration of speech intelligibility6 and later involvement of upper limbs (ULs) leads bulbar-onset patients to rapidly replace speech by handwriting (often a pen and a notepad for peerto-peer communication), until UL weakness demands new strategies for communication.7,8 As the disease progresses, most patients must replace speech communication by communication based on text input (handwriting, use of computer, or paper-based letter board). Modern tablet devices used as assistive communication devices (ACDs), with touchscreen input, can allow bulbar-onset patients to maintain autonomy in communication through simple applications with speech synthesizers. Nevertheless, ACD use is strongly dependent on the decisions and expectations of patients, their caregivers, and health professionals. As long as residual capacity for communicating through speech or handwriting is present, the use of communication devices may not be considered. When communication alternatives are insufficient (severe dysarthria associated with UL dysfunction), technologically more advanced methods need to be explored. Timely referral and economic burden remain critical issues in decisionmaking for communication support in ALS. However, late referral is a primary reason for low acceptance; indeed, adaptation of patients with marked generalized weakness to new strategies (e.g., eyetracking or even brain–computer interfaces) can be a difficult process, especially when there is lack of experience or support for continuous training in the use of these ACDs.7,9 When patient disability is severe, ACD use relies on caregiver support for daily setup and maintenance. For this reason, caregiver needs, skills, and expectations must be considered as important factors in clinical intervention with ACDs.9,10 Difficulty in understanding patient needs and feelings is the chief caregiver motivation for supporting intervention with an ACD.11 Furthermore, patient autonomy regarding communication has been MUSCLE & NERVE
December 2015
933
Table 1. Questionnaires and tests applied and their respective ranges. Test/questionnaire ALSFRS-R ALSFRS-R-b ALSFRS-R-ul WHOQoL-bref MQoL MQoL-SIS (single-item score) Communication Effectiveness Index—modified (CETI-m) Performance in typing Performance in handwriting Questionnaire for patients and caregivers: How familiar are you with computer devices?
Range
Time*
(maximum disability) to 48 (normal) (maximum disability) to 12 (no bulbar symptoms) (maximum disability) to 12 (no upper limitations) (poorest QoL) to 100 (max QoL) (lowest QoL) to 10 (highest QoL) 0 (very bad) to 10 (excellent) 0 (no communication) to 100 (normal) Words per minute (wpm) wpm 0 (I never used a computer) to 5 (I am an expert)
T0, T1, T2 T0, T1, T2 T0, T1, T2 T0, T2 T0, T1, T2
0 0 0 0 0
T0, T1, T2 T0, T1, T2 T0
*T0: at study entry; T1: 3–4 months after entry; T2: 7–10 months after entry.
reported as an essential family caregiver need that may influence the caregiver’s own QoL.12 The primary objective of this study was to explore the impact of early support with ACDs on QoL of ALS patients with bulbar onset and their caregivers. The score for QoL of both patients and caregivers was the main outcome. A secondary objective was to evaluate the efficiency of ACDs in supporting speech dysfunction. As secondary outcomes, we monitored patients’ functional scores, their self-perceived communication effectiveness, and upper limb function for performance of written communication (handwriting and typing on a keyboard). METHODS Patients. ALS patients were enrolled with the following inclusion criteria: (1) probable or definite disease according to the revised El Escorial criteria; (2) bulbar-onset disease, as defined by initial symptoms of dysarthria or dysphagia; (3) bulbar score for dysarthria of 3 or 2 (mild or moderate impairment), as given by the first question of the ALS Functional Rating Scale—revised (ALSFRS-R); and (4) informed consent. Patients with other medical conditions, respiratory symptoms, or clinical evidence of dementia were excluded. A main caregiver was identified by the patient as the person who provides daily support and accompanies the patient to clinical visits (frequently the husband/ wife or a sibling). The caregivers did not change over the duration of this study. The study was approved by the local ethics committee, and patients and caregivers gave written informed consent before participation.
Patients were assessed 3 times: at entry (T0); 3–4 months later (T1); and 7–10 months after entry (T2). Caregivers were assessed 2 times, at T0 and T2. Most assessments were done in our clinic, but some were done in the patient’s home. At each visit, the patient and caregiver were interviewed separately. The duration of the complete assessment was approximately 30 minutes. At T0,
Procedure.
934
Communication and QoL in ALS
patients were also randomized into 2 groups according to the intervention time [group 1 (G1): early intervention, patients received an ACD just after baseline assessment; group 2 (G2): late intervention, patients received communication devices according to the conventional approach at our center (referral when dysarthria score reaches 0 or 1 on the ALSFRS-R, which is anarthria or severe dysarthria)]. The ACD used was a small and light commercially available touchscreen-tablet device (approximately 200 3 130 mm). We used a software program based on a simple onscreen keyboard for text-to-speech communication with text prediction. This software was developed commercially for the specific purpose of text-to-speech communication and is available to download from the internet; for our purposes, an important feature was that the software could be used with different access strategies in addition to the touchscreen. Patients and caregivers were trained and supported in the use of the device during clinical appointments. Patients who had an internet connection at home were instructed to use e-mail communication through the ACD. E-mail communication was tested and stimulated by sending emails (demanding reply) periodically during the study. Patients were assessed with the following questionnaires: the McGill Quality of Life questionnaire (MQoL); the ALSFRS-R; and the modified Communication Effectiveness Index (CETI-m). Caregivers completed the MQoL and World Health Organization Quality of Life (WHOQoL-BREF) questionnaires, related to their own QoL. Both patients and caregivers answered a single question, rated on a 0–5 Likert-type scale, to assess their previous experience with computer devices (Table 1). of Life. MQoL Questionnaire. The MQoL13 consists of a questionnaire with 16 items, each rated on a Likert-type scale from 0 to 10. Five domains are calculated as the mean scores of 1–6 items: 2 health-related domains [physical symptoms Quality
MUSCLE & NERVE
December 2015
and physical well-being (MQoL-Phys)] and 3 non– health-related domains [psychological symptoms (MQoL-Psych); existential well-being (MQoL-Exist); support, related to the feeling of being cared for (MQoL-Sup)]. A total score (MQoL-Tot) was calculated as the mean of the scores in each domain (higher scores indicate higher QoL). In addition, this instrument includes a single-index score (MQoL-SIS), which measures overall quality of life, and a descriptive section (MQoL-part D), where respondents list what had the greatest impact (positive or negative) on their QoL in the past 2 days. MQoL is sensitive to psychological, supportive, and spiritual factors more than to health-related QoL measures.14 It has been used in various studies related to QoL in ALS patients.15–18 WHOQoL-BREF. The WHOQoL-BREF is a 26item questionnaire for measuring QoL in 4 domains: (1) physical health and well-being; (2) psychological health and well-being; (3) social relations; and (4) environment (related to financial resources, health and social services, home and physical environment, among others). Items are rated on a 1–5 Likert-type scale, and scores are calculated on a 100-point scale (higher scores correspond to better QoL). Ratings range from 1 (very poor) to 5 (very good).19–21 This questionnaire has been used extensively to investigate QoL of patients and caregivers in different chronic diseases, including ALS.22–24 Functional Measurements. The ALSFRS-R was applied at each visit. It is an instrument used for monitoring progression of functional disability in patients with ALS.25 Each question on the ALSFRSR (total of 12 questions) is rated from 4 (normal function) to 0, according to progressive functional impairment. We calculated the total score (ALSFRS-R-tot), which ranges from 48 (normal) to 0, and 2 subscores: bulbar function (ALSFRS-Rb)26 and upper limb function (ALSFRS-R-ul).27 Subscores range from 12 to 0 and were calculated by summing the first 3 questions and the fourth to sixth questions of the ALSFRS-R, respectively. The ALSFRS-R-b rates speech, salivation, and swallowing functions; the ALSFRS-R-ul rates UL function related to handwriting, handling utensils in daily activities, and dressing. The CETI-m28,29 is an adaptation of the original CETI designed for patients with aphasia. It is a valid measure of functional communication and is sensitive to changes in performance over time.30 The CETI-m rates self-perceived limitation of a person when communicating; it consists of 10 questions related to communication in different contextual situations rated on a visual Likert-type scale. The total score is given by the sum of the individual scores, presented on a 100-point scale, Communication and QoL in ALS
representing the minimum (0) to maximum (100) communication effectiveness. Performance Measures. To monitor performance of the ULs in written communication at each evaluation session, patients were asked to write, sequentially, 3 different sentences with 2 different modalities: handwriting and using a keyboard. Both performance of handwriting and typing on a keyboard were calculated as the mean rate of the writing tasks, as measured by words per minute (wpm). Statistical Analysis. Primary outcomes were scores from the MQoL (single-index item, total score, physical, psychological health, existential wellbeing, and support) and WHOQoL-BREF scores (4 domains: physical health and well-being; psychological health and well-being; social relations; and environment). Secondary outcomes were ALSFRSR-tot, ALSFRS-R-b, ALSFRS-R-ul, CETI-m, typing rate, and handwriting rate. Descriptive statistics were used to describe participants and the results of the single question regarding previous computer use. Pearson and Spearman correlation coefficients were used to examine associations among different variables for normal and non-normal data distributions, respectively. Changes in outcome variables, ranging from T0 to T2, were analyzed by paired-sample t-test and Wilcoxon signed-rank test. The t-test and Mann– Whitney U-test for independent samples were used to test differences between the 2 different groups (G1 vs. G2). P < 0.05 was considered statistically significant. For each of the statistical analyses, we used Bonferroni adjustment of the P-value for multiple comparisons. However, for the exploratory purpose of this study, we considered P < 0.05 as a trend that was worth reporting. RESULTS
Twenty-seven ALS patients, mean age 64.8 (SD 10.2) years (range 39–83 years), were included in the study. Among this group, 1 declined immediately after baseline assessment, and 5 were evaluated only twice due to rapid disease progression (Fig. 1). Only 17 caregivers were interviewed (4 patients came to clinical visits alone or with different caregivers, and 6 caregivers declined to fill in the questionnaires). Most of the patients were women (81.5%), and previous experience in the use of computers, rated by patients from 0 to 5, was generally low (median 5 2). Low school education and poor previous experience with technology of most of the participants coincided with the demographic characteristics of the Portuguese population within this age group. Table 2 describes the demographic data of participants. MUSCLE & NERVE
December 2015
935
FIGURE 1. Flowchart of the recruitment process. N, number of patients; Nc, number of caregivers.
Patients in G1 started to use an ACD after baseline assessment. There was no record of spontaneous use of assistive communication devices in patients randomized to G2. For this exploratory analysis of the impact of communication in QoL, we analyzed each domain of QoL questionnaires and compared scores between different assessments and groups of patients and caregivers. As many variables were compared, no significant result was found when considering adjusted P-values for multiple comparisons. However, taking into account the exploratory aim of this study, we considered P < 0.05 to be a trend with regard to the influence of communication on QoL. Mean scores of all outcome variables and results of all statistical comparisons are detailed in Tables S1 and S2 (refer to Supplementary Material, available online). QoL. Patients. Mean values calculated for all patients (through MQoL) showed a decrease in total, physical, and existential domains of QoL (P < 0.05). Single-index score, psychological symptoms, and support domains did not change with time (P > 0.05). In descriptive parts of the MQoL (physical symptoms and part D), 23 patients (88.5%) referred in at least 1 assessment to speech problems or difficulty being understood by others as negative factors in QoL; 2 patients in G1 referred to communication device as having a positive impact in QoL. Patients with early intervention with their ACD had higher MQoL scores than the other patients, particularly for the existential well-being domain at T1 (P 5 0.045) and the psychological symptoms domain, both at T1 (P 5 0.047) and T2 (P 5 0.032) (Table 3). 936
Communication and QoL in ALS
Caregivers. The mean scores for caregiver QoL (MQoL) decreased from T0 to T2, although only the single-index score (MQoL-SIS) attained a difference with P < 0.05. The mean scores of the WHOQoL-BREF showed a decrease (P < 0.05) in all domains, from T0 to T2, except for the social relations domain (P 5 0.749). In the descriptive item of the MQoL (part D), 8 (47%) caregivers referred to patient speech problems as having a negative impact on their own QoL; 5 of 11 (46%) caregivers of patients in G1 believed the ACD had a positive impact on their QoL. When comparing results from caregivers of patients in G1 vs. G2, we identified a higher QoL in the former group in the domain of support (P 5 0.016) (Table 3). No differences between the
Table 2. Characteristics of all participants at baseline (T0): patients and caregivers.
Total (N) Gender (W/M) Age, in years [mean 6 SD (range)] Education Primary school Elementary/high school University Relation of caregiver Husband/wife Sibling Experience with computers* Median (range) Internet access at home
Patients
Caregivers
27 22/5 64.8 6 10.2 (39–83)
17 6/11 57.52 6 13.9 (26–81)
12 (44.4%) 10 (37%) 5 (18.5%)
5 (22.7%) 6 (27.3%) 6 (27.3%) 10 7
2 (0–4) 12 (44%)
3 (0–5)
*Experience in using computers (minimum 5 0, maximum 5 5).
MUSCLE & NERVE
December 2015
Table 3. Mean values of dependent variables tested in each evaluation period (T0, T1, T2) comparing G1 and G2 results. T0
CETI-m MQoL-psych (patient) MQoL-exist (patient) MQoL-sup (caregiver)
T1
T2
G1
G2
G1
G2
G1
G2
64.67 6 27.26 5.34 6 2.21 7.25 6 1.60 6.30 6 1.44
71.33 6 31.92 4.68 6 1.57 7.00 6 2.03 4.42 6 1.83
44.53 6 20.63 5.61 6 2.20 7.52 6 1.37 NA
47.11 6 29.31 4.01 6 1.39* 6.40 6 1.24* NA
48.83 6 16.24 4.93 6 1.59 6.53 6 1.65 6.71 6 1.41
17.20 6 11.80* 3.18 6 1.10* 5.02 6 1.15* 3.92 6 2.13*
Comparisons are shown where there was a trend (P < 0.05) for the influence of the ACD on the quality of life for patients and caregivers. There were no statistically significant differences between the 2 groups when adjusting for multiple comparisons (Bonferroni method). NA, not applicable; CETI-m, Communication Effectiveness Index—modified (from 0 to 100); MQoL-psych, MQoL psychological symptoms subscore (lowest 5 0, highest 5 10); MQoL-exist, MQoL existential well-being subscore (lowest 5 0, highest 5 10); MQoL-sup, MQoL support subscore (lowest 5 0, highest 5 10). *P < 0.05 for comparison between G1 and G2 (independent-samples t-test).
2 groups (P > 0.05) were found in any of the WHOQoL-BREF domains. Caregiver MQoL scores and patient MQoL scores were not correlated in any of the periods studied (P > 0.05). Functional Measurements. As expected, evaluation of bulbar function (ALSFRS-R-b score) decreased significantly over time (Table 4). The decline was linear and showed no difference in percentage change between T0–T1 and T1–T2. We found no significant difference between the 2 patient groups in bulbar score decline. The higher the bulbar impairment (lower ALSFRS-R-b scores), the lower the self-perceived communication effectiveness index (CETI-m) for patients in G1 (r 5 0.419, P < 0.001) and G2 (r 5 0.809, P < 0.001). However, when considering data from patients in G1, after they started to use the ACD (T1 and T2), patients with lower bulbar functional scores did not necessarily have lower self-perceived communication (r 5 0.101, P 5 0.629). In general, we found positive correlations between patient QoL, bulbar function, and selfperceived communication; the higher the bulbar dysfunction (lower ALSFRS-b score), the lower the QoL, both in total score (r 5 0.297, P 5 0.019) and in single-item score (r 5 0.280, P 5 0.028). Patients with higher scores in self-perceived communication effectiveness had higher QoL, both in total score
(r 5 0.268, P 5 0.027) and in the psychological symptoms domain (r 5 0.265, P 5 0.029). No correlations were found between caregiver QoL scores and patient bulbar function (ALSFRSR-b) or communication (CETI-m) scores, for the whole group of caregivers. However, when analyzing caregivers of patients who received early ACD intervention, we found that patients with higher self-perceived communication had caregivers with higher QoL scores for the psychological symptoms domain of MQoL (r 5 0.515, P 5 0.011). Performance Measures. Handwriting performance showed a mean of 17.5 (SD 8.1) wpm (range 7.6– 36.4) at T0, decreasing significantly (P 5 0.002) from baseline assessment to T2 [mean 12.4 (SD 10) wpm, range 0–27.4 wpm] (Table 4). The lower the UL function (ALSFRS-R-ul), the lower the measured handwriting performance (r 5 0.693, P < 0.001). Due to symptoms of disease progression in ULs, 5 patients were unable to perform handwriting at T1 or T2, although they could still use the ACD for typing. These patients had difficulty grasping a pen, but were able to use 1 finger on the keyboard or touchscreen. Performance using a keyboard varied widely among patients due to previous experience with using computer devices [mean 9.2 (SD 5.2), range 2.4–18.7 wpm)] at baseline assessment. During the period of study, patient typing performance did
Table 4. Mean values for secondary outcome variables tested in this study during each assessment period (T0, T1, T2).
ALSFRS-R ALSFRS-R-b ALSFRS-R-ul Handwriting rate (wpm) Typing rate (wpm) CETI-m
T0
T1
T2
P(T0,T2)
27.18 6 8.54 6.32 6 2.71 9.45 6 3.04 16.2 6 6.6 9.2 6 5.2 67.17 6 28.6
22.15 6 8.96 4.75 6 2.47 8.25 6 4 14.1 6 9.8 8.5 6 5.4 45.50 6 23.65
15.50 6 7.88 2.75 6 2.05 6.44 6 4.34 10.7 6 9.8 6.7 6 4.9 39.53 6 20.9
