Fors et al.

BMC Medical Informatics and Decision Making (2019) 19:122

https://doi.org/10.1186/s12911-019-0847-3

RESEARCH ARTICLE Open Access

User evaluation of a novel SMS-based

reminder system for supporting post-stroke
rehabilitation
Uno Fors1* , Julius T. Kamwesiga2,3, Gunilla M. Eriksson3,4, Lena von Koch3,5 and Susanne Guidetti3

Abstract
Background: According to WHO stroke is a growing societal challenge and the third leading cause of global
disease-burden estimated using disability-adjusted life years. Rehabilitation after stroke is an area of mutual interest
for health care in many countries. Within the health care sector there is a growing emphasis on ICT services to
provide clients with easier access to information, self-evaluation, and self-management. ICT-supported care programs
possible to use in clients’ home environments are also recommended when there are long distances to the health care
specialists.
The aim of this study was to evaluate the technical usability of a SMS-based reminder system as well as user opinions
when using such a system to assist clients to remember to perform daily rehabilitation activities, to rate their
performance and to allow Occupational therapists (OT’s) to track and follow-up clients’ results over time.
Methods: Fifteen persons with stroke were invited to participate in the study and volunteered to receive
daily SMS-based reminders regarding three activities to perform on a daily basis as well as answer daily SMS-
based questions about their success rate during eight weeks. Clients, a number of family members, as well as
OTs were interviewed to evaluate their opinions of using the reminder system.
Results: All clients were positive to the reminder system and felt that it helped them to regain their abilities.
Their OTs agreed that the reminder and follow-up system was of benefit in the rehabilitation process. However,
some technical and other issues were limiting the use of the system for some clients. The issues were mostly linked to
the fact that the SMS system was based on a Swedish phone number, so that all messages needed to be sent
internationally.
Conclusion: In conclusion, it seems that this type of SMS-based reminder systems could be of good use in
the rehabilitation process after stroke, even in low income counties where few clients have access to Internet
or smart phones, and where access to healthcare services is limited. However, since the results are based on
clients’, OTs’ and family members’ expressed beliefs, we suggest that future research objectively investigate
the intervention’s beneficial effects on the clients’ physical and cognitive health.
Keywords: Stroke, Rehabilitation, Client-centred, Feedback, Occupational therapy, SMS-reminders, Africa

* Correspondence: [email protected]
1
Department of Computer and Systems Sciences (DSV), Stockholm University,
Stockholm, Sweden
Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 2 of 11

Background Reminder systems for clients

About stroke Computer based reminder systems have been tested and
According to the World Health Organization (WHO) found to be of good value in other domains than stroke.
stroke is a growing societal challenge and is the third For example, Dexter et al. [12] described a reminder sys-
leading cause of global disease-burden estimated using tem to increase the use of preventive care for hospital-
disability-adjusted life years [1]. Stroke causes impair- ized clients, which lead to a significant increase of such
ments, activity limitations and participation restrictions measures. Jangi et al. [13] performed a systematic review
[2] which often result in decreased functioning in every- on reminder systems used in physical therapy, which
day life. The increasing number of people having stroke showed that reminders in the form of SMSes, phone
leads to a growing global demand for rehabilitation ser- calls, letters or e-mails could have a good effect on im-
vices but is especially true in low- and middle-income proving clients’ adherence to physical therapy exercise
countries in which a significant number of people have programs. However, there have also been studies that in-
stroke [3]. However, the availability of rehabilitation ser- dicate that automated reminders might not be superior
vices is scarce in many regions of the world. Rehabilita- to paper based methods for clients with stroke [14].
tion after stroke is therefore an area of increasing
importance for healthcare in many countries, including ICT-based interventions for stroke
Sweden and Uganda. A number of different ICT-based systems have been
The increasing burden of stroke but limited access to proposed and tested to assist with the rehabilitation after
rehabilitation services creates a need for developing new stroke. For example, computer-support in terms of Tele-
strategies such as the use of Information and Communica- health solutions [15], Robotics [16], Virtual Reality [17, 18]
tion Technologies (ICT) like mobile phones for provision as well as Off-the-shelf computer games [19] have been
of healthcare services [4]. Within the healthcare sector tested. Also, more complex models with home-based ICT-
there is a growing emphasis on ICT-based services to platforms have been applied [20].
provide clients with easier access to information, self- However, most of these ICT-based solutions have been
evaluation, and self-management. ICT-supported pro- focused on a more direct training and rehabilitation of
grams in clients’ home environment are also recom- motor functions. Few have targeted the basic idea of
mended when there are long distances to the health care reminding the clients, and in a positive way, challenging
specialists [5]. the clients to perform their recommended daily activities
using ICT-tools. Even fewer of such studies have investi-
gated the use of such systems in developing, less
About rehabilitation of stroke and the need for reminders wealthy, regions of the world.
The goal for rehabilitation of people with stroke is de- Smart-phones have been used within rehabilitation for
fined as increased functioning and participation in life, example in India where modules with information about
(i.e. body function, activity and participation) and stroke, exercises to be performed at home, training of
well-being [6]. One way of reaching this goal is to functional skills as well as of ADL, and the use of assist-
focus on increasing the ability and independence in ive devices have been provided to persons with stroke
activities in daily living (ADL). Evidence is weak for and their family members. This information was pro-
various general rehabilitation interventions as inter- vided through text and video-clips on smart phones and
ventions for improved motor functioning. However, were field-tested and found feasible and acceptable by
there is strong evidence for task-specific training, the persons with stroke and their family members [21].
meaning activities or tasks that are relevant and pur- Interventions with the use of short message services
poseful for the individual [7], as well as for ADL (SMS) has been found effective to improve medication
interventions [2, 8]. Therefore, activities that are per- adherence after stroke [22], but studies which use SMSes
ceived as relevant and purposeful in everyday life for as reminders for performing activities as part of a re-
persons with stroke can be used as goals in an inter- habilitation program after stroke have not been found.
vention to improve the ADL functioning [8].
Interventions for compensation for the impact of cog- Stroke and rehabilitation in Uganda
nitive disabilities, with the goal to improve the perform- In Uganda, despite the large number of stroke cases,
ance of ADL is common within stroke rehabilitation [9, there are very few occupational therapists (OTs) and re-
10] and assistive devices with for example reminders habilitation resources available. Therefore, most clients
have contributed to improving the performance of activ- have so far been left to the care of their families and
ities among people with cognitive disabilities as memory with very limited professional assistance [23].
problems in the subacute phase as well as in the long Moreover, even if the use of the internet and mobile
term [9, 11]. phones have dramatically increased also in many low
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 3 of 11

income countries, a majority of the users in Uganda still Methods

use more basic, non-smart phones with very limited or Study setting
no Internet connection, leading to special challenges in This study was a sub-study of a larger overall clinical
using internet-based high tech solutions in healthcare study (F@ce), described elsewhere [28]. The study was a
services. collaborative project between Karolinska Institutet (KI)
in Stockholm (Division of Occupational Therapy), Sweden;
Stockholm University, Sweden (SMS-system development
About the overall F@ce project
and related services); and Uganda Allied Health Examina-
This project is a sub-study of the overall stroke rehabili-
tions Board (coordination of local OTs).
tation project “Participation in daily activities in every-
The overall study focusing on the rehabilitation of
day life after stroke - Developing and evaluating a model
post-stroke clients in Uganda was based on a set-up with
for a mobile phone supported and client-centered re-
four phases starting with 1. A training workshop for the
habilitation intervention in Uganda” led by researchers
OTs that should deliver the intervention according to
at Karolinska Institutet, Sweden.
F@ce™; 2. Gathering and informing subjects and local
The Medical Research Council’s guidance for develop-
OTs and collecting base-data from all clients; 3. The
ing complex interventions [24], has been used in previ-
F@ce™ intervention, including formulating three individ-
ous studies of a client-centred, activities in daily living
ual targets (goals) and planned strategies for recapture
(CADL) intervention, according to Bertilsson et al. [25].
the target activities; 4. Collecting final data including
Based on the theoretical base of an occupational and
interviewing clients and their relatives. This sub-study
phenomenological perspective as well as the rationale of
was focusing on the methodology of using SMS-based
CADL, with its different components, a further develop-
reminders to support clients to perform the daily train-
ment involving a qualitative study of how persons who
ing activities (goals).
have had a stroke used their mobile phones [26] as well
All activities in Uganda were coordinated by a local
as a culturally adapted Stroke Impact Scale (SIS) 3.0 has
OT (JTK), who was in direct contact with the other
previously been performed in Uganda [27]. This rehabili-
three OTs. The whole project was governed by a group
tation intervention (called F@ce™) has been further de-
of Swedish OT researchers. JTK was a PhD student at
veloped and refined in collaboration with experienced
KI during the time of the study.
practitioners, researchers and health informaticians.
F@ce stands for Face-to-face (F) with a collaboration
Participants
between the therapist and a client during all the different
Three OTs volunteering to participate in the overall
phases, including Assessment (@), Collaboration (C) and
study were trained in a workshop in Kampala, Uganda
Evaluation (E). The F@ce™ intervention integrates the
February 2016 to deliver the intervention according to
principles of client-centred practice with goal setting in-
F@ce™ and use an SMS-based reminder system to sup-
volving daily occupations that the person need and want
port the rehabilitation of their clients.
to do in everyday life during the entire rehabilitation
In the overall clinical study [28], thirty persons were
process and has in a feasibility study been found to im-
invited post-stroke by the local OT to, on a voluntary
prove primary outcomes (performance and satisfaction
basis, participate in the study. The overall study was set-
of valued daily activities in everyday life assessed using
up with a quasi-experimental pre-post design with an
the Canadian Occupational Performance Measure
intervention group (IG, n = 15) receiving the F@ce™
(COPM), as well as self-efficacy [28].
intervention using SMS-based reminders and a control
To support the work according to F@ce™, a SMS-based
group (CG, n = 15) where no SMS was supplied during
reminder and client monitoring tool was developed. This
the study period.
sub-study focus on the technical usability in terms of fea-
Since this actual sub-study was focusing on the tech-
tures and functions of the sub-systems for managing,
nical usability and user opinions of the SMS-reminders,
sending and receiving SMS-reminders as well as on the
only the 15 clients in the IG were included here.
user opinions of this health informatics solution.
All participants had been diagnosed with stroke and
were living in or adjacent to the capital of Uganda,
Aim Kampala.
The aim of this study was to evaluate the technical us- Eleven of the family members of the participants in
ability (as described above) and user opinions when the intervention group were also asked to participate
using a SMS-based reminder system to assist clients to in an open-ended interview on their experiences of
remember to perform daily rehabilitation activities, to living with a person with stroke and experiences of
rate their performance and to allow OT’s to track and the F@ce™ intervention and their opinions of using
follow-up clients’ results over time. the SMS-system.
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 4 of 11

The SMS reminder system The management system was used according to the
A Web-based system for managing client data, entering following procedure:
their and the OTs phone numbers, the three targeted
daily activities per person who have had a stroke, as well 1. The OT recruits a client, informs about the project
as the timing for reminders and other data collection, and, together with the client, decide on suitable
was developed using Node.js/PostgreSQL as a backend activities to reach the targets.
and HTML/CSS/JavaScript as a front-end. 2. The OT enters client’s phone number, his/her own
The central SMS-system was in its turn connected to phone number, the three activities, period of the
an international commercial SMS API service called reminders and time per day for sending out the
Twilio™ (Twilio.com). This was due to the fact that the SMSes with reminders of the activities, and time
project management could not find a local tele-operator per day to send out questions regarding the success
in Uganda who could give the same service. This rate of the activities in the system.
entailed that all SMS needed to be sent from and to a 3. Each morning, the individual client receives one
Swedish phone number. SMS reminder of the three target activities, see
The SMS-system allowed the local OT team to formu- Fig. 3. If the client could not manage the mobile
late individually targeted daily training activities that phone, a family member was to receive the SMS,
were designed to support the persons to improve their support the client to rate the performance, send
activities of everyday life according to the principles of scores by SMS and encourage the client to perform
F@ce™. Due to the limitations of the SMS-technology, the activities.
all reminders needed to be formulated very briefly, like 4. Each afternoon the client receives three different
“washing laundry”, “swipe the compound” etc. However, SMSes with a question per target activity, where
the local OTs could in the discussions with the clients the client is supposed to rate the level of success of
explain all three daily activities in more detail, if needed. the activity, see Fig. 3. The clients were instructed
The SMS-system was also designed to assist the OTs to to indicate that if he/she did not perform that
manage all client data, including mobile phone numbers specific activity they should answer with a 0
and individual daily activity targets and when to send (meaning “has not performed the activity”), or
out the morning SMS reminder and the evening follow- rate how successful they were (where 1 mean
up questions. Please refer to Fig. 1 for a screen-shot of “not so good”, and 5 “carried out the activity
the central system. very well”). Participants who rated 0 or who did
The system could also follow-up clients over time and not reply to the SMS reminder message,
display statistics of their responses to facilitate the moni- automatically launched a red flag on the OT’s
toring of the clients’ daily activities and possible success mobile phone. The OT would then call the
in doing them, see Fig. 2. The system could also display participant the following morning to find out
time-series graphs for each individual client or groups of what had happened.
clients. This feature could also be used for gathering 5. Steps 3 and 4 were repeated every weekday during
follow-up data for the research team. the study period (8 weeks).

Fig. 1 The SMS client management system main interface. Note that these are not real client data
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 5 of 11

Fig. 2 A screenshot of some of the statistics the central SMS client management system may present. Note that these are not real client data

6. After the study period, the local OT interviewed the post-intervention assessments of the overall study
the clients about their experiences of using the [28]. In this context and when that data collection was
SMS-system as a part of their rehabilitation. completed, the participants were interviewed about their
experiences of the F@ce™ intervention as well as the SMS-
The messages that were sent to the clients were based services. Both the clients and their family members were
on pure text SMS-messages due to the issue that most interviewed. Altogether 22 participants were interviewed
people in Uganda do not have a Smartphone. In our individually: 11 clients (two clients dropped out from the
study, only one participant of the 15 in the SMS-group intervention and two other could not be reached at the
had a Smartphone. time for the interviews) and 11 family members. The
Since we expected that most of the clients had an fi- questions were semi-structured and adhered to an inter-
nancial situation where sending and receiving daily view guide developed by the authors. A number of the
SMSes from and to Sweden was too expensive, the pro- questions in the interview guide focused on the technol-
ject bought “Air-time” for all participating clients so that ogy of the intervention, for example: During these 8 weeks
sending and receiving SMSes were free of charge during of intervention; What do you think worked well in the
the study period. intervention? Can you please tell us what you think about
During the first meeting, the local OT informed about the technology used, the SMS? Have you experienced any
stroke and provided advice to promote independent problem with the technology? How has it been for you
functioning in ADL. Some participants, also might have rating the goals using SMS? Each interview lasted roughly
received other rehabilitation services as needed, e.g. between 40 to 60 min. The participants were thus asked
physiotherapy and speech therapy. about their experiences regarding the intervention, its im-
The central SMS system automatically kept track of pact on their daily life and the SMS-service and its advan-
each clients’ activities and results. If a client response tages and challenges. Their answers were recorded
was a “0” in any of the three activity questions or if he/ digitally and then transcribed. Some answers were in
she did not reply at all, the associated OT automatically Luganda and those interviews were first transcribed and
received a SMS with instructions to contact that individ- then translated into English. The verbatims on the experi-
ual client and check what the problems might be. ences of the intervention and how it impacted their daily
life were then analysed using latent content analysis [29].
Interviews with participants The clinical part of the interviews will be presented in a
After the 8 weeks all participating clients were approached future publication. The part of the transcripts revealing
by the coordinating local OT (JTK) who also performed the experiences of the use of the mobile phone and the
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 6 of 11

In addition, the three assisting OTs and the local main

OT also answered a questionnaire. The questions asked
them to rate advantages and challenges involved in using
the SMS-service. See Additional file 1 for details.

Results
All 15 clients agreed in collaboration with their OTs on
three daily target activities. The most common activities
were washing clothes, sorting beans, dressing self, but
many other activities were also targeted. None of the
clients complained about that the very brief SMS-
descriptions of the daily activities (e.g. “washing laun-
dry”) were difficult to follow.
Two clients in the intervention group dropped out be-
cause they moved out from the Kampala region and
could not be reached by the research team. The
remaining 13 clients continued to use the SMS through-
out the 8 weeks. However, at the time for the interviews,
two of the clients could not be reached, ending up in 11
clients interviewed totally.
The SMS system for reminders worked rather well
during the project, however due to unknown reasons
one client’s phone was unable to receive our SMSes dur-
ing a period. An unanticipated issue was also that some
of the participating clients were affected by their stroke
so that they could not operate their own phones, but
was relying on a family member to receive the reminders
and send the evening rated responses. Furthermore,
some clients did not use their phones on their own, but
Fig. 3 A screenshot from a hypothetical client’s phone showing the asked a family member to gather the SMSes and then
morning SMS with the three activities and two of the evening SMS tell the clients what was in the message, making the
questions of success rate. Note that these are not real client data SMS intervention somewhat complicated.

SMS-service were read through and the participants’ de- Interviews with clients and family members
scriptions were sorted according to whether they were At the follow-up after the intervention, 11 clients and 11
positive or negative to the SMS services. The content ana- family members (i.e., seven daughters, one son, one
lysis as well as the analysis of the narrated experiences of father, one mother and one niece) narrated their experi-
the SMS-service was performed by the second, third and ences of the SMS service. Almost all participants were
last author. positive to the SMS-support and the quotes below are il-
lustrating their descriptions/experiences in their narra-
tives. One client expressed it like this: “I liked it – it kept
Interviews with the OTs me busy, I finished the work I was supposed to do”. She
The local main OT in this project was coordinating the continued describing what receiving the SMS meant to
project and was also responsible to instruct and follow- her in her everyday life: “if this SMS had not, if they had
up the work by the three other OTs. All OTs were also not come … it would get me like that, I would not know
interviewed regarding their opinions regarding the tech- that I need to practice these things on my own”.
nical usability and clinical possibilities of using the re- Another client agreed and expressed it like this: “I
minder system. would like to have more SMS, yes [Clapping hands] so
Additionally, since the local main OT was in regular much so much so much so much so much. I tell you
contact with the other OTs, he received feedback from those, it kept me very busy”.
the OTs on a regular basis where they reported smaller Other clients mentioned positive effects like “Quicker
or larger issues of the SMS reminder system and the re- service and contacts”, “The SMS-service has helped me in
habilitation process. The OTs also filled in information my everyday life”, “We are working towards the goals, we
in their log books. were reminded”.
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 7 of 11

There were also clients mentioning that they would  The OTs reported that the clients stated that the
lack the SMS-reminders after the study period, like SMS system helped them to believe that someone
“what will I do now when the SMSes will not come any really cared about them. Especially the reminder
more”. According to the OTs, similar comments were function was seen as very positive sign of that
given by a majority of the clients. “someone” really cared about their health
The family members narrated about many different  OTs also found the SMS system to be a good way to
target activities in the intervention and how they were reach out to many clients regardless of the distance
assisting their family members (the person with stroke)  OTs also found the system to compel both the
in performing them. They also described the SMS- clients and the family members to adhere to the
reminders regarding which activities they had decided to rehabilitation process
focus on during the rehabilitation period. Most of the  It was also indicated by the OTs that the clients
family members were enthusiastic about the rehabilita- liked to monitor their own performance as they
tion and the use of the mobile phone as a tool in re- needed to send SMS rating their performances of
habilitation. One family, for example, was so inspired by the set targets
the opportunity to part in this intervention that they  The SMS system encouraged both clients and their
bought new phones so all of them could take part. The family members to work together for the common
same family expressed that they liked being a part of the rehabilitation goal
rehabilitation but that it sometimes was really demand-
ing, that the physical training they did in between was However, also a number of challenges were identified
“painful” and that “we practice at home because it is a by the OTs:
must”. This family member described the SMS-
reminders and follow-ups as difficult to handle and that  It was expensive to send and receive international
they were too many. She said:“ We reply, we do it be- SMSes
cause it is a must”.  Some clients struggled to send acceptable SMSes,
At the same time the same family member showed ap- since the system only accepted answers in the form
preciation over the SMS-reminders. The SMS kept the of Figs. 0–5, some clients tried to answer in text,
family active. She described it like this: “Sometimes we which did not work. When this did not work, they
were bored, but now she (the mother) is not bored. She could become discouraged
makes her life to be busy, good!”  Some elderly clients were not familiar with SMS at
Another family member expressed that the best with all, and were only used to use their phones to call
the reminder system was that it included follow up to someone up
see if it worked and that the family member benefitted  Using the older type of mobile phones with buttons
from it. She therefore appreciated the SMS-follow ups seemed to be challenging for some users, since they
and phrased it like this: …“ because in that way you show did not get an overview of what was sent and
that you are concerned about the recovery of the client received (which can be more easily visualized on
and follow-up. If you don’t follow-up you do not care. a Smart Phone)
When you don’t care they are not interested”.
Moreover, a daughter of another client also mentioned
that she felt that the F@ce™ program and the SMS- OT questionnaire
reminders supported her in her work with, and worries All four OTs filled in the questionnaire. The OTs were
about, her mother. The fact that the system automatic- in general very positive to the SMS reminder system.
ally should tell the OT to call her mother if anything did Please refer to Table 1 for details.
not work, strengthened the daughter.

Interviews with the participating occupational therapists Discussion

The four OTs participating in this project were all posi- In this pilot study, a SMS-based reminder system for
tive to the SMS-based reminder system. A number of supporting the rehabilitation process of post-stroke cli-
advantages of the SMS-reminder system were identified ents in Uganda was developed, implemented and evalu-
from the answers like (based on quotes from the ated regarding technical usability and user opinions. A
questionnaires): special focus was to investigate if a relatively simple re-
minder system, based on pure text SMS-messages, could
 An SMS service like the one tested was seen as a be developed and used, to overcome the challenge of
good way to remind clients to perform daily activities that most clients in Uganda lack both access to modern
to reach rehabilitation targets Smartphones as well as to the Internet.
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 8 of 11

Table 1 Answers from the OTs on the questionnaire

Question Very much/ Much/Agree A little/To Not at all Comments
Completely agree some extent
Do you think the SMS-service has helped 2 2
your clients in their rehabilitation?
Do you think the SMS-service has helped 1 3
your clients in their everyday life?
Did you run into any technical problems 1 3 Clients could not
with the SMS service? always receive
messages
Using the SMS-service was an asset in the 2 2
rehabilitation of the clients?
Do you think the SMS-service has helped 3 1
you in the rehabilitation with your clients?
Would you recommend this SMS-service for 4
others to use in the rehabilitation of clients?
What was the best thing with the SMS system? It enables clients to work out their objectives. They get to connect to therapists too.
On spot. Gives a quick feedback.
The time was good for reminding. It helped the clients to remember.
To get the clients to work against the set goals.
What was the most troublesome thing with When they do not get feedback from the server (messages not going through)
the SMS-reminder system? Sometimes, inconsistence: the therapists did not know exactly how the SMS came
and looked like in order to advice the client better in how to use them
The reply format on the afternoon questions was not easily understood

SMS reminder system issues A third issue was that some clients were unable to
It was found that a SMS-based reminder system could read and answer the SMSes sent out. This could be due
be developed and implemented rather straightforward to their post-stroke problems or to other reasons. In
and easy. However, a number of technical issues with many cases, this challenge was solved via support from
this pilot system were also revealed. One was that SMS- the clients’ family members, but this solution is not opti-
messages only can handle 150 characters, and since mal. In the future, this problem might be eased by using
there were three daily activities for each client, the re- tablets instead, where the text can be displayed in larger
minder texts needed to be short. However, since none of size and the “answer buttons” (soft touch-screen buttons
the clients complained about the short messages, we be- on a tablet) can be made bigger. However, this will prob-
lieve that this mode of communication was acceptable. ably be an issue to look into in future studies. Post-
In the future, where more persons probably will have stroke clients in any country often have both limited
access to smart phones also in low income countries, motor and cognitive abilities, and new smart solutions
this issue will be less prominent. need to be developed in the future to overcome such
The issue with that no local telecom operator support- problems.
ing computer-based SMS-messaging API services could Regardless of these challenges, we see it evident that a
be identified in Uganda, led to the drawback that all mobile phone-based reminder system can be imple-
messages needed to go from Sweden to the clients in mented with rather small resources even in low income
Uganda, and the same with their responses on the even- countries like Uganda. To the best of our knowledge
ing surveys that were needed to be sent back to Sweden there are no similar SMS-based reminder systems in
from the clients’ local phones in Uganda. This might Europe, which makes us believe that similar systems
also been the reason for that one of the clients did not could be of use also in high income countries.
receive the SMS-messages over a period of time. Even if
the financial drawback of sending and receiving SMSes The central SMS server system
to and from Sweden was solved via pre-paying “air-time” As was shown in Figs. 1 and 2, the central SMS server
for the participants, it is obviously not a perfect solution. system was used to both enter participant data, their re-
This might also been the reason for that some clients lated daily activities, record the client answers, automat-
had some other related issues (but which were solved). ically tell the local OT if a client was not answering, as
In a future reminder application, a local SMS API- well as to store and display follow-up data. This feature,
service need to be set-up and connected to a local phone even if not fully used in this pilot study, can probably be
number and a local server. As far as we understand it, used in future similar projects to help clinicians to
today this is possible also within Uganda. follow-up clients over longer terms.
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 9 of 11

Additionally, such a central system can also serve re- as well as in line with findings by Perri-Moore et al. [5]
searchers with aggregated data for both individuals and who mentioned that “Automated technology may reliably
groups of clients. assist clients to adhere to their health regimen, increase
attendance rates, supplement discharge instructions, de-
Clients and family members crease readmission rates, and potentially reduce clinic
When asking the clients and their family members about costs”.
their opinions of the SMS-reminder system, all were The OTs also noted similar challenges as described
quite positive and indicated that this system made the above, regarding a number of technical issues. These in-
rehabilitation quicker and better. The finding that some cluded sending/receiving international calls, connectivity
clients did not read and send back their responses on issues and problems reading the small screen etc. The
the SMS-messages themselves, but relied on that a fam- fact that some clients were reported to have troubles
ily member assisted them, is somewhat challenging. It with answering the evening SMS in Figs. 0–5 (and not
could be interpreted as a sign of that the SMS- in text) can be seen as an indication of that the instruc-
reminders were difficult to handle. But on the other tions to the clients should have been more clear. But
hand it can also be a sign of the culture to involve the again, as have noted, a SMS-based system only allows
family and that the family wanted to actively participate short messages of 150 characters, so if the instructions
in the rehabilitation of the family member who had had should be longer, another type of communication mode
a stroke. Moreover, we interpret the positive attitudes to might be needed like SmartPhone or Tablet apps or
the SMS-reminders as an indication that the mode of similar that allows longer texts. But such solutions also
communication through SMS was well received, even if require access to Internet, which is not a standard for
this was a completely new way of communication be- many people in low income regions. Moreover, the OTs
tween clients and clinicians for all participants. interviewed also indicated that they would recommend
However, since this was a very limited study during a this type of SMS-based system to others to use in the re-
limited period of time, we have no hard evidence on that habilitation of clients, and that they believed that the re-
the rehabilitation results were really improved by the minder system helped their clients to improve.
SMS-intervention. In the overall study [28], there was a
significant difference detected between the IG and CG Overall results
in changes between baseline and follow-up for the primary Thus, taken together the findings support each other
outcomes COPM performance component and self- and are not just the opinions, but rather the common
efficacy, in favour of the F@ce™ intervention [27, 28]. view of most participants involved in this study. Both
Moreover, since rehabilitation measures after stroke often the therapists and the clients and their family members
are based on that the clients themselves should feel that were supportive of this type of mobile phone supported
they make progress, we believe that the SMS-reminder rehabilitation processes. However, as mentioned above,
system supported the clients in this way anyhow. Add- we cannot guarantee that the SMS-reminders were
itionally, since more than one client expressed worries resulting in objectively improved physical and cognitive
what should happen now, when they did not receive the health of the clients. Other limitations of the study is
reminders, we think that there was a real positive effect by that we only targeted Uganda and post-stroke clients
the reminders. However, since the results are only based linked to one hospital only. However, as many low-
on clients’, OTs’ and family members’ expressed beliefs, we income countries often face similar issues with regard to
cannot guarantee that the SMS-reminders were resulting limited access to technology, restricted financial capabil-
in objectively improved physical and cognitive health of ities as well as limited access to health care profes-
the clients. Future studies should look into the possibility sionals, we believe that this study is a good example of
to more in detail assess the use of this type of re- what can be done in such areas and thus, we believe that
minder systems and their possible impact on the the results should be applicable for also other low in-
health of the clients. come regions of the world.

Occupational therapists Conclusions

When interviewing the OTs participating in this study, This study, even if it was a limited pilot study, resulted
their opinions were very positive as well. All four of in a number of conclusions, including that it was pos-
them believed the SMS-service to be supportive to their sible to develop and implement a SMS-based system for
clients in the rehabilitation process. All of them also in- reminding clients post-stroke about the daily target ac-
dicated that the SMS reminder system supported their tivities, as well as to survey the clients’ for their estima-
clients in their everyday life. This is to our mind good tion of the result of the activities. The system could also
signs of a potential use of similar systems in the future, be used to automatically warn OTs about clients that
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 10 of 11

were not able to perform the activities or who did not Consent for publication
answer the survey questions. All clients seemed to ap- Not applicable.

preciate the SMS-reminders as well as the evening sur-

Competing interests
vey about their success rate. Family members also The authors declare that they have no competing interests.
seemed to see the SMS-reminder system as valuable. In-
volved OTs indicated the SMS-reminder system as a Author details
1
Department of Computer and Systems Sciences (DSV), Stockholm University,
good tool to support rehabilitation after stroke saw great Stockholm, Sweden. 2Uganda Allied Health Examinations Board, Kampala,
clinical possibilities of using the reminder system. The Uganda. 3Division of Occupational Therapy, Department of Neurobiology
text-only based reminders was a good way to solve is- Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden.
4
Department of Neuroscience, Rehabilitation Medicine, Uppsala University,
sues of that nearly all clients lacked smart phones and Uppsala, Sweden. 5Theme Neuro, Karolinska University Hospital, Stockholm,
internet connections. A number of technical issues were Sweden.
identified, most of them due to the need to use inter-
Received: 17 August 2018 Accepted: 20 June 2019
national calls for the SMS-messages. Finally, if a similar
system should be implemented in the future, all mes-
sages should be based on local communication set-ups. References
1. Mackay J, Mensah G. The atlas of heart disease and stroke. Geneva: WHO;
2004. https://www.who.int/cardiovascular_diseases/resources/atlas/en/.
Additional file Accessed 28 June 2019.
2. Legg L, Drummond A, Leonardi-Bee J, Gladman J, Corr S, Donkervoort M, et
Additional file 1: The Questionnaire to the OTs. (DOCX 14 kb) al. Occupational therapy for clients with problems in personal activities of
daily living after stroke: systematic review of randomised trials. BMJ. 2007;
335(7626):922. https://doi.org/10.1136/bmj.39343.466863.55.
Acknowledgements 3. Ferri CP, Schoenborn C, Kaira L, et al. Prevalence of stroke and related
We are thankful for the support from the Swedish Association of Occupational burden among older people living in Latin America, India and China. J
Therapists and the HELD research group at Karolinska Institutet, Sweden. We Neurol Neurosurg Psychiatry. 2011;82:1074–82.
also thank the clients, their family members and the local OTs for participating 4. Mechael PN. The case for mHealth in developing countries. Innovations.
in the project. 2009;4:103–18.
5. Perri-Moore S, Kapsandoy S, Doyon K, Hill B, Archer M, Shane-McWhorter L,
Authors’ contributions Bray BE, Zeng-Treitler Q. Automated Alerts and Reminders Targeting Clients:
JTK is living and working in Uganda and this research formed a part of his A Review of the Literature. Client Educ Couns. 2016;99(6):953–9. https://doi.
doctoral studies. This research was supervised by SG, GME and LvK who all org/10.1016/j.pec.2015.12.010.
contributed to the study design. JTK, SG, and GME organised and conducted 6. von Groote PM, Bickenbach JE, Gutenbrunner C. The world report on
the workshops. UF was responsible for the technical part of the study. All disability - implications, perspectives and opportunities for physical and
authors took part in the interpretation of the results and made significant rehabilitation medicine (PRM). J Rehabil Med. 2011;43(10):869–75. https://
contribution to the manuscript, including the drafting and the revising of doi.org/10.2340/16501977-0872.
the manuscript for intellectual content. All authors read and approved the 7. National Board of Health and Welfare (Socialstyrelsen): National guidelines
final manuscript. for stroke care. Stockholm: The National Board of Health and Welfare. 2009.
8. Steultjens E, Deeker J, Bouter L, van de Nes J, Cup E, van den Ende C.
Occupational therapy for stroke clients. A systematic review. Stroke. 2003;34:
Funding
676–87.
This study was supported by grants from the Swedish Research Council,
9. Boman IL, Tham K, Granqvist A, Bartfai A, Hemmingsson H. Using electronic
Developmental Research (grant number 2014–28-63). The funding body did
aids to daily living after acquired brain injury: a study of the learning
not influence the design of the study, nor the collection, analysis, and
process and the usability. Disability and Rehabilitation: Assistive Technology.
interpretation of data, nor the writing of the manuscript.
2007;2(1):23–33.
10. Cicerone K, Dahlberg C, Malec J, Langenbahn D, Feliceti T, Kneipp S, et al.
Availability of data and materials Evidence-based cognitive rehabilitation: updated review of the literature
If our manuscript will be accepted for publication we are unfortunately from 1998 through 2002. Arch Phys Med Rehabil. 2005;86:1681–92.
unable to make the client data available publicly due to our responsibility to 11. Arthanat S, Nochajski SM, Stone J. The international classification of
protect the confidentiality of the participants. This would also violate the ethical functioning, disability and health and its application to cognitive
permit. disorders. Disabil Rehabil. 2004;26(4):235–45.
As our OT participants were selected from a small total population of 12. Dexter PR, Perkins S, Overhage JM, Maharry K, Kohler RB, McDonald CJ. A
occupational therapists in Uganda, in making this information public we computerized reminder system to increase the use of preventive care for
would be unable to ensure their anonymity, and would violate the conditions hospitalized clients. N Engl J Med. 2001;345(13):965–70.
under which they agreed to participate in the study. 13. Jangi M, Ferandez-de-las-Penas C, Tara M, Moghbeli F, Ghaderi F, Javanshir
K. A systematic review on reminder systems in physical therapy. Caspian J
Ethics approval and consent to participate Intern Med. 2018;9(1):7–15. https://doi.org/10.22088/cjim.9.1.7.
This study has been approved by the ethical review committee of the 14. Emmerson KB, Harding KE, Taylor NF. Home exercise programmes
Uganda National Council for Science and Technology no. HS 703 and by supported by video and automated reminders compared with standard
Mulago hospital’s ethical board (date: 12/1–2011). All participant data paper-based home exercise programmes in clients with stroke: a
has been aggregated and analysed anonymously. All participants were randomized controlled trial. Clin Rehabil. 2017;31(8):1068–77. https://doi.org/
invited on a voluntary basis and gave their written informed consent to 10.1177/02692155166808566.
participate. The clients were informed that they could withdraw at any 15. Joubert J, Joubert L, Medeiros de Bustos E, Ware D, Jackson D, Harrison T,
moment without any consequences for them and that all data would Cadilhac D. Telestroke in stroke survivors. Cerebrovasc Dis. 2009;27(suppl 4):
be kept anonymous. 28–35.
The clients received no compensation other than that they for free could 16. Lo AC, Guarino P, Krebs HI, Volpe BT, Bever CT, Duncan PW, Ringer RJ,
send and receive the SMSs via a pre-paid service that the project gave them Wagner TH, Richards LG, Bravata DM, Haselkorn JK, Wittenberg GF,
access to during the duration of the project. Federman DG, Corn BH, Maffucci AD, Peduzzi P. Multicenter randomized trial
Fors et al. BMC Medical Informatics and Decision Making (2019) 19:122 Page 11 of 11

of robot-assisted rehabilitation for chronic stroke: methods and entry

characteristics for VA ROBOTICS. Neurorehabil Neural Repair. 2009;23(8):
775–83.
17. Broeren J, Bjorkdahl A, Claesson L, Goude D, Lundgren-Nilsson Å, Samuelsson
H, Blomstrand C, Sunnerhagen KS, Rydmark M. Virtual Rehabilitation after
Stroke. Proceedings of MIE 2008: Stud Health Technol Inform. 2008;136:77–82.
18. Saposnik G, Levin M, and for the Stroke Outcome Research Canada
(SORCan) Working Group. Virtual Reality in Stroke Rehabilitation. Stroke.
2011;42(5). https://doi.org/10.1161/STROKEAHA.110.605451.
19. Yong Joo L, Soon Yin T, Xu D, Thia E, Fen Chia P, Wee Keong Kuah C, Keng
He K. A Feasibility Study Using Interactive Commercial Off-Theshelf
Computer Gaming In Upper Limb Rehabilitation In Clients After Stroke.
J Rehabil Med. 2010;42:437–41.
20. Zheng H, Davies R, Stone T, Wilson S, Hammerton J, Mawson SJ, Ware PM,
Black ND, Harris ND, Eccleston C, Hu H, Zhou H, Mountain GA. SMART
Rehabilitation: Implementation of ICT Platform to Support Home-Based
Stroke Rehabilitation. In: Stephanidis C (ed) Universal Acess in Human
Computer Interaction. Coping with Diversity. UAHCI 2007. Lecture notes in
computer science, vol 4554. Springer, Berlin, Heidelberg doi:https://doi.org/
10.1007/978-3-540-73279-2_9.
21. Sureshkumar K, Murthy GVS, Natarajan S, et al. Evaluation of the feasibility
and acceptability of the ‘Care for Stroke’ intervention in India, a
smartphone-enabled, carersupported, educational intervention for
management of disability following stroke. BMJ Open. 2016;6:e009243.
https://doi.org/10.1136/bmjopen-2015-009243.
22. Kamran Kamal A, Shaikh Q, Pasha O, Azam I, Islam M, Memon AA, Rehman
H, Akram MA, Affan M, Nazir S, Aziz S, Jan M, Andani A, Muqeet A, Ahmed
B, Khoja S. A randomized controlled behavioral intervention trial to improve
medication adherence in adult stroke clients with prescription tailored short
messaging service (SMS)-SMS4Stroke study. BMC Neurol. 2015;15:212.
https://doi.org/10.1186/s12883-015-0471-5.
23. Chin JH. Stroke in sub-Saharan Africa: an urgent call for prevention.
Neurology. 2012;78(13):1007–8.
24. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M.
Developing and evaluating complex interventions: the new Medical
Research Council guidance. Int J Nurs Stud. 2013;50(5):587.
25. Bertilsson AS, Ranner M, von Koch L, Eriksson G, Johansson U, Ytterberg C,
Guidetti S, Tham K. A client-centred ADL intervention: three-month follow-
up of a randomized controlled trial. Scand J Occup Ther. 2014;21(5):377–91.
https://doi.org/10.3109/11038128.2014.880126.
26. Kamwesiga J, Tham K, Guidetti S. Experiences of using mobile phones in
everyday life among persons with stroke and their families in Uganda – a
qualitative study. Disabil Rehabil. 2017;39(5):438–49. https://doi.org/10.3109/
09638288.2016.1146354.
27. Kamwesiga J, Koch L, Kottorp A, Guidetti S. Cultural adaptation and
validation of Stroke Impact Scale 3.0 version in Uganda- a feasibility study.
SAGE Open Med. 2016;29:4. https://doi.org/10.1177/2050312116671859.
28. Kamwesiga TJ, Eriksson G, Tham K, Fors U, Ndiwalana A, von Koch L,
Guidetti S. A feasibility study of a mobile phone supported family-centred
ADL intervention, F@ce™, after stroke in Uganda. Glob Health. 2018;14:82
https://doi.org/10.1186/s12992-018-0400-7.
29. Graneheim UH, Lundman B. Qualitative content analysis in nursing research:
concepts, procedures and measures to achieve trustworthiness. Nurse Educ
Today. 2004;24(2):105–12.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.