Breaking the communication barrier using new technologies in aphasic patients post-stroke

Abstract

Objectives

To improve language in aphasic patients after a recent ischaemic stroke using computer applications for mobile devices (tablet) for 30 days and to establish whether length of use influences this improvement.

Method

Randomised, prospective, single-centre clinical trial with two intervention groups, one of which was a control group. Aphasic patients were included within a maximum of 7 days after an ischaemic stroke. The intervention group used tablet-specific software for 30 days. Language improvement was assessed using the Mississippi Aphasia Screening Test (MAST).

Results

The sample was 40 patients, mean age 73 years (± 10.2). Fifty percent were women. Patients were equally divided between the 2 groups. Mean device use was 311.8 (±334.6) minutes after the end of 30 days, with a final mean improvement in MAST of 46.5 (±22.3) points in the intervention group and 34.8 (±21.7) in the control group (p = .233). Adjusting for age, the results indicate that the patients who tend to improve the most are those under 65 years of age (43.8 (±30.5) points, p = .205). We were unable to demonstrate significant differences between the two groups.

Discussion and/or conclusions

The results obtained in the study suggest that new technologies can be an effective tool for language recovery, especially in young patients. Multicentre studies with a larger sample size and younger patients may be able to demonstrate the usefulness of this new tool.

Resumen

Objetivos

Establecer la mejoría del lenguaje en pacientes afásicos tras un ictus isquémico reciente, usando aplicaciones informáticas para dispositivos móviles (tablet) durante 30 días y establecer si el tiempo de uso influye en dicha mejoría.

Método

Ensayo clínico aleatorizado, prospectivo, unicéntrico con dos grupos de intervención, uno de ellos control. Se incluyeron pacientes afásicos en un plazo máximo de 7 días tras un ictus isquémico. El grupo de intervención utilizó un software específico para tablet durante 30 días. La mejoría del lenguaje se evaluó mediante el Mississipi Aphasia Screening Test (MAST).

Resultados

La muestra fue de 40 pacientes, edad media 73 años (±102). El 50% eran mujeres. Los pacientes se repartieron de forma equitativa entre los 2 grupos. El uso medio de los dispositivos fue de 311,8 (±334,6) minutos tras finalizar los 30 días, con una mejoría media final en el MAST de 46,5 (±22,3) puntos en el grupo de intervención y de 34,8 (±21,7) en el grupo control (p = 0,233). Al ajustar por edad, los resultados indican que los pacientes que más tienden a mejorar son los menores de 65 años (43,8 (±30,5) puntos, p = 0,205). No conseguimos demostrar diferencias significativas entre ambos grupos.

Discusión y/o conclusiones

Los resultados obtenidos en el estudio sugieren que las nuevas tecnologías pueden ser una herramienta eficaz para la recuperación del lenguaje sobre todo en pacientes jóvenes. La realización de estudios multicéntricos con un mayor tamaño muestral y pacientes más jóvenes tal vez podrían conseguir demostrar la utilidad de esta nueva herramienta.

Introduction

In Spain, stroke is the leading cause of death in women and the second leading cause of death in men. It is also a major cause of disability in adulthood and contributes to the development of impaired cognitive function that can lead to the onset of dementia.¹ Stroke can have various sequelae.² Language is one of the most frequently impaired higher functions, the most disabling disorder being aphasia.³

Aphasia is the loss or impairment of language due to a brain injury.³ It is not exclusively caused by stroke, it can also be caused by brain injury and other disorders that affect the language area.

Aphasia may be milder and there may be greater functional recovery depending on the extent of the injury.³

There are two main types of aphasia: motor aphasia, which affects Broca’s area and causes language processing disorders, and sensory aphasia, which affects Wernickes’ area and causes language comprehension disorders.⁴

Some types of language disorder have a better prognosis for recovery than others⁵ and allow us a certain degree of interaction with patients and understand them as we undertake procedures for their care.

Aphasia is present in 21%–38% of patients admitted to a stroke unit.⁶ The characteristics of aphasia mean that patients cannot communicate to express their needs and feelings. It is therefore essential to find tools to tackle these severe communication problems.⁷ Studies over the past 10 years have demonstrated the usefulness of technology in the process of language recovery,^8–12 and some have been conducted on the new portable technologies, such as computers, smartphones, and tablets.^13–15 Of particular note are those that have compared therapy using these devices versus speech therapy.^16–21 We conducted this study using specific language recovery software developed for tablets and administered by nursing staff to test its impact on improving language in aphasic patients after stroke.

The main objective of our study was to determine language improvement in aphasic patients after a recent ischaemic stroke (inclusion in the first 7 days after the event), using language recovery software applications developed for tablets for 30 days. Our secondary objective was to determine whether the length of time the patients used the software applications was related to this improvement.

Method

A randomised, prospective, single-centre clinical trial conducted at the Hospital Universitario Virgen Macarena in Seville. The participants met the inclusion and exclusion criteria (Table 1) and were randomised using a randomised list, stratified by sex and age. Age groups were defined as those over 65 years and those under 65 years.

Table 1.

Inclusion and exclusion criteria.

Inclusion criteria

Age 18 years or older

Ischaemic stroke confirmed by brain imaging test

Inclusion in the study within the first 7 days following onset of symptoms

NIHSS core of 1 or more in item 9

NIHSS score of 15 or under

Rankin score between 0 and 3

Ability to pass the test on the use of a computing device

Availability of the caregiver

Exclusion criteria

Dementia or cognitive impairment prior to the stroke

Characteristics that in the opinion of the investigator demonstrate that compliance with the study will not be adequate

The study intervention had two arms. In the first arm, patients used a computer programme to express their needs as a communicator and as a system of exercises for language rehabilitation. The second arm was the control group, in which the patients received no intervention.

In all cases the patient received concomitant speech therapy for the rehabilitation of language disorders.

The duration of the intervention was 30 days for the patients who received therapy with the device.

Assessments and visits were conducted 30, 60, and 90 days after the intervention.

Tools used

The Mississippi Aphasia Screening Test (MAST)²² was used to assess aphasia. This scale assesses the areas of reading, writing, comprehension, and repetition of language by scoring individual items that add up to an overall score between 0 and 100 points. This determines the patient’s degree of impairment. The scale was administered by the nursing staff and the score reviewed by a speech therapist, both blinded. Let me Talk® from Appnotize UG was used as a communicator.

Puedo Hablar® from Las Palabras Perdidas was the programme used as a set of language rehabilitation exercises. It lists the number of language recovery exercises that are well performed and points out the areas where there are deficiencies.

The scale was administered at each visit and therefore we can compare the baseline situation with the evolution of the different visits at the end of the intervention.

The patients kept a diary in which they recorded the number of minutes per day they used the devices, which was checked against the record measured by the applications, to ensure as reliable an account as possible of the time spent using the devices. We divided the sample into three age groups: under 65 years (group 1), 65−80 years (group 2), and over 80 years (group 3).

Data analysis

For statistical analysis we used an ANOVA (one-way analysis of variance) test with IBM SPSS® statistical software SPSS® version 27.0 (9 April 2019) to compare means.

We used MAST scores to obtain the data that we compared, taking as a reference the values obtained at the baseline visit (initial status of the patient after the stroke) and comparing them with those obtained at the last visit of the study. The difference showed how much each patient's language had improved quantitatively. The arithmetic means of the improvements in MAST scores were calculated by grouping the patients by sex and age. These same means were used for the ANOVA test.

For the minutes that the devices were used we recorded the minutes each patient used them per day and the minutes they used them in total over the 30 days that the intervention with the device lasted. When distributing the patients by sex and age, we calculated the arithmetic means of the minutes of use during that time for each group to compare them with each other using the ANOVA test.

Ethical considerations

The patients participated in the study after reading and understanding, then voluntarily signing the informed consent form provided.

The Andalusian Biomedical Research Ethics Portal (PEIBA, act n. 08/2018) approved the present study to be conducted at the Hospital Virgen Macarena in Seville, in compliance with Law 3/2018 on data protection.

Results

Forty patients were included in the study with a mean age of 73 years (±10.2) (42–88 years). Fifty percent of them were female. Thirty patients were aged over 65 years (75%) and there were 10 patients in the under 65 age group (25%). Of the total number of patients, 7 were assigned to the control group (17.5%), 6 died during the study (15%), 3 dropped out of the study (7.5%), and there was one inclusion failure. Twenty-three patients received the intervention and completed the study (57.5%) (Fig. 1).

Figure 1.

Flow chart of the distribution of the sample.

The baseline characteristics of the population were similar between the two groups and no statistically significant differences were found (Table 2).

Table 2.

Baseline description of the sample.

	Intervention arm (n = 30)	Control arm (n = 10)	Statistical significance (p)
Mean age, years (SD)	73.20 (±9.53)	72.40 (±2.79)	.834
Male sex, n (%)	15 (50%)	5 (50%)	1.000
Hypertension, n (%)	23 (76,6%)	7 (70%)	.689
Dyslipidaemia, n (%)	10 (30%)	4 (40%)	.718
Diabetes, n (%)	10 (30%)	2 (20%)	.693
Smoker, n (%)	5 (16.6%)	4 (40%)	.190
Previous stroke, n (%)	7 (23.3%)	0 (0%)	.161
Alcohol consumption, n (%)	2 (6.6%)	1 (10%)	1.000

Measurements using the MAST test at 90 days showed an overall mean improvement of 46.5 (±22.3) points in the intervention group versus an improvement of 34.8 (±21.7) points on average in the control group with respect to the patients' baseline assessments (p = .233). No significant difference in improvement was found.

In terms of age group, we can observe that the patients in group 1 tended to improve more than those in the intervention group (43.8 [±30.59] vs. 12 [±5.65] points, p = .205), with no significant differences. In group 2, no differences were observed between the intervention group and the control group (49.4 [±22.11] vs 49.5 [±15.71] points, p = .994). Finally, in group 3 we again found differences in favour of the intervention group (45 [±13.08] vs. 22 [±0] points, p = .165), although they were not significant (Table 3).

Table 3.

Mean difference in MAST score from baseline and mean minutes of device use.

	Group 1 (<65 years) (n = 9)	Group 2 (65−80 years) (n = 14)	Group 3 (>80 years) (n = 7)
MAST score
Controls	12 (±5.65)	49.5 (±15,71)	22 (±0)
Cases	43.8 (±30.59)	49.4 (±22.11)	45 (±13.08)
P-value	.205	.994	.165
Minutes of use	367.4 (±279.36)	408.3 (±413.48)	86.1 (±86.30)

The mean number of minutes the programme was used was 311.8 (±334.6) throughout the intervention (in the first 30 days of the study) in the patients who finished the study (n = 23). In terms of age group, the mean use was 367.4 min in the group 1 (±279.3), 408.3 min in group 2 (±413.48), and 86.1 min in group 3 (±86.3) (p = .154).

Discussion

Speech therapy is currently used to treat language disorders. Some nursing units have pictogram charts that allow basic communication with this type of patient. However, this resource is insufficient for patients to clearly express their needs. Nursing staff could use these resources and play a more active role in language recovery.

Our results show that the patients who received some type of intervention had better MAST test scores than the patients in the control group overall, without these differences being statistically significant, which contrasts with the results obtained in other studies^13,14,17,21 positioned along the same lines and which have obtained similar results, in which aphasic patients who used language recovery programmes improved compared to those who did not. The youngest patients, under 65 years, showed the greatest improvement. This improvement decreases as the age of the patients increases, although it increases again in the group of the longest-lived patients. Comparing our results with those found in the literature reviewed, we observe that other studies have failed to find a relationship between age and recovery from aphasia.¹⁰

If we look at the time the devices were used, we can infer that, in the group under 65 years of age, where there was effective use of the device, we found better results in terms of improvement in the MAST test. However, this is not the case for the other groups, and therefore other factors must be influencing this. In other studies reviewed, evidence has been found that there are patients who improve with speech therapy, even if they do not use the devices with specific software,⁸ which shows that the speech therapy given as standard clinical practice and the natural evolution of the stroke itself influence the ability to recover language.

Initially in our study we estimated that patients can gain optimal benefit using the device for 30 min day i.e., around 900 min by the end of the intervention.⁸ We observed that the mean time the devices were used was much lower overall in our sample. This would imply that if programmes were used more, we would find better MAST scores. However, to do so we would have to consider modifications to the protocol and target the study to a younger population, and add training strategies beforehand or amend the software to adapt it better to an older population.

One of the limitations we found when conducting the study was the advanced age of the patients and their level of prior knowledge using computer programmes on digital devices such as tablets or smartphones. Although all the patients who took part in the study were helped by younger family members, with better command of current technology, we noticed that there was fatigue and a lack of interest, especially in the older patients, which resulted in poor compliance with the therapy, as is reflected in the results.

Patient recruitment was another of the principal limitations of the study, made difficult by the COVID-19 pandemic. It was difficult to include patients due to mobility limitations caused by the lockdown and travel restrictions, and their fear of coming to the hospital for the study visits. However, this type of tool, which can be used in the patient's home, is very useful to avoid frequent visits to health centres for rehabilitation.

The study results are not statistically significant to establish whether improvement in the language skills of aphasic patients after an ischaemic stroke is related to the use of digital devices with specific software. However, we found no statistically significant differences to show that the length of time these devices are used influences improvement in language. However, our results suggest that new technologies could be useful in language recovery, especially in younger patients. Future studies with larger sample sizes may confirm the trends found in our study.

Nurses can play a significant role in recovering the ability to communicate by supporting patients using new technologies to improve compliance and adherence to these new therapies.

This is an area where further research and studies are needed in the use of technology to meet patients’ communication needs with nursing staff and their caregivers.

Funding

The funding entity of the Spanish Society of Neurological Nursing (SEDENE) awarded 1500 Euros.

Conflict of interests

The conflict of interest is that the study was awarded the abovementioned sum of 1500 Euros by SEDENE.