JMIR MHEALTH AND UHEALTH Kamal et al

Original Paper

Effect of 5-Minute Movies Shown via a Mobile Phone App on

Risk Factors and Mortality After Stroke in a Low- to Middle-
Income Country: Randomized Controlled Trial for the Stroke
Caregiver Dyad Education Intervention (Movies4Stroke)

Ayeesha Kamal1*, MD; Adeel Khoja1*, MSc; Bushra Usmani1, MD; Shahvaiz Magsi1, MD; Aresha Malani1,
PharmD; Zahra Peera1, MSc; Saadia Sattar1, MSc; Masood Ahmed Akram 1, PharmD; Sumaira Shahnawaz1, BA;
Maryam Zulfiqar1, MD; Abdul Muqeet2, BSc; Fabiha Zaidi2, BSc; Saleem Sayani2, MSc; Azmina Artani1, MSc;
Iqbal Azam3, MSc; Sarah Saleem3, MSc
1
Aga Khan University, Stroke Services and Research, Karachi, Pakistan
2
Aga Khan Development Network, Digital Health Resource Center, Karachi, Pakistan
3
Aga Khan University, Community Health Sciences, Karachi, Pakistan
*
these authors contributed equally

Corresponding Author:
Ayeesha Kamal, MD
Aga Khan University
Stroke Services and Research
Faculty Office Building, First Floor, Department of Medicine,
Stadium Road Campus
Karachi, 74800
Pakistan
Phone: 92 2134930051 ext 4559
Email: [email protected]

Abstract
Background: Pakistan is the sixth most populous nation in the world and has an estimated 4 million stroke survivors. Most
survivors are taken care of by community-based caregivers, and there are no inpatient rehabilitation facilities.
Objective: The objective of this study was to evaluate the effectiveness and safety of locally designed 5-min movies rolled out
in order of relevance that are thematically delivered in a 3-month program to deliver poststroke education to stroke survivor
and caregiver dyads returning to the community.
Methods: This study was a randomized controlled, outcome assessor–blinded, parallel group, single-center superiority trial in
which participants (stroke survivor-caregiver dyads) with first-ever stroke (both ischemic and hemorrhagic) incidence were
randomized within 48 hours of their stroke into either the video-based education intervention group or the control group. The
video-based education intervention group had health education delivered through short videos that were shown to the
participants and their caregivers at the time of admission, before discharge, and the first and third months of follow-up after
discharge. The control group had standardized care including predischarge education and counseling according to defined
protocols. All participants enrolled in the video education intervention and control groups were followed for 12 months after
discharge for outcome assessment in the outpatient stroke clinics. The primary outcome measures were the proportion of
participants achieving control of blood pressure, blood sugar, and blood cholesterol in the video intervention versus the control
group. Several predefined secondary outcomes were included in this study, of which we report the mortality and functional
disability in this paper. Analysis was by performed using the intention-to-treat principle.
Results: A total of 310 stroke survivors and their caregiver dyads (participant dyads) were recruited over a duration of 6
months. In total, 155 participant dyads were randomized into the intervention and control groups, each. The primary outcome
of control of three major risk factors revealed that at 12 months, there was a greater percentage of participants with a systolic
BP<125 mm Hg (18/54, 33% vs 11/52, 21%; P=.16), diastolic BP<85 mm Hg (44/54, 81% vs 37/52, 71%; P=.21), HbA1c
level<7% (36/55,
65% vs 30/40, 75%; P=.32), and low-density lipoprotein level<100 mg/dL (36/51, 70% vs 30/45, 67%; P=.68) in the
http://mhealth.jmir.org/2020/1/e12113/
JMIR Mhealth Uhealth 2020 | vol. 8 | iss. 1 | e12113 | p. 1

XSL•FO (page number not for citation purposes)

RenderX
 JMIR MHEALTH AND UHEALTH Kamal et al

intervention

http://mhealth.jmir.org/2020/1/e12113/
JMIR Mhealth Uhealth 2020 | vol. 8 | iss. 1 | e12113 | p. 2

XSL•FO (page number not for citation purposes)

RenderX
group than in the control group. The secondary outcome reported is the mortality among the stroke survivors because the
number of stroke-related complications was higher in the control group than in the intervention group (13/155, 8.4% vs 2/155,
1.3%), and this difference was statistically significant (P<.001).
Conclusions: The Movies4Stroke trial failed to achieve its primary specified outcome. However, secondary outcomes that
directly related to survival skills of stroke survivors demonstrated the effectiveness of the video-based intervention on
improving stroke-related mortality and survival without disability.
Trial Registration: ClinicalTrials.gov NCT02202330; https://www.clinicaltrials.gov/ct2/show/NCT02202330

(JMIR Mhealth Uhealth 2020;8(1):e12113) doi: 10.2196/12113

KEYWORDS
stroke; mobile health; noncommunicable diseases; adherence

Introduction Objective
Background In this study, we aimed to evaluate the effectiveness and
safety of locally designed 5-min movies rolled out in order of
Stroke is the second leading cause of death globally and the relevance that are thematically delivered in a 3-month
principal cause of acquired disability in adults. About two- program to deliver poststroke education to stroke survivor and
thirds of this burden is endured by the developing world [1]. caregiver dyads returning to the community. We hypothesized
Noncommunicable diseases (NCDs) are the biggest that the absence of trained personnel in the health community
contributors to the rising incidence of stroke. Around 90.5% could be mitigated by actually providing high-quality
of the global stroke burden is attributable to modifiable risk repetitive training using audio visual aids that served as a
factors, including 74.2% attributed to behavioral factors checklist for competency and survival skills to the stroke
(smoking, poor diet, and low physical activity). In addition, survivor and caregiver dyad [7,8].
hypertension, type 2 diabetes mellitus, and coronary artery
disease are important modifiable risk factors for stroke [2]. Methods
Pakistan also has a disproportionate burden of stroke and
NCD risk factors. At present, around 1 in 4 adult Pakistanis Study Design
has hypertension or diabetes, heart disease, or a stroke A randomized controlled, outcome assessor–blinded, parallel
equivalent, with most being unaware of their risks [3]. A local group, single-center superiority trial was conducted to assess
study investigated the prevalence of stroke in Pakistan among the efficacy of mobile phone video–based IT intervention for
adult Pashtun population and reported a prevalence of 4.8%, controlling 3 major risk factors–blood pressure [BP], blood
which is equal to 4 million persons affected in a country with glucose, and cholesterol—among adult stroke survivors.
a population of 180 million [3]. Important secondary outcomes included postdischarge
mortality attributable to stroke and measures of functional
Studies that describe the outcomes of stroke survivors in this
disability. Our detailed protocol has been published separately
setting report that at a median of 5.5 months after discharge,
[9].
12.3% of the patients had died, mostly from recurrent
vascular events or stroke complications. Poor functional Study Site
outcome, defined as Modified Rankin Scale (mRS) score>2, This trial was conducted in the Stroke unit, Neurology Ward,
was seen in 51% of the study participants, and cognitive Aga Khan University Hospital (AKUH), Karachi, Pakistan.
outcomes were poor in 42% of the survivors [4]. AKUH is an internationally recognized tertiary care
In a country of a population of 180 million, roughly only 23 institution, certified by Joint Commission International
centers exist to provide physical medicine or help with Accreditation, and caters to the needs of a large multiethnic
rehabilitation; most have not adopted a multidisciplinary urban population of 18 million people. Stroke care follows
approach toward patients, and none have inpatient services international protocols, with defined order sets and
[5]. Currently, there are no organized home care survival standardized pathways.
programs involving primary caretakers for stroke survivors in
Participants
Pakistan.
The sample population comprised adults (aged>18 years)
Despite these challenges, there is potential to leverage mobile admitted to AKUH with first-ever acute stroke and having a
technology to improve stroke outcomes. Pakistan has designated caregiver, meeting the eligibility criteria, and
widespread mobile connectivity, with a cellular density of giving informed consent.
77% [6]. These infrastructure enablers create distinct
opportunities for mobile health (mHealth). Our rationale was The criteria used while recruiting participants and caregivers at
to leverage information technology (IT)–based mHealth to the initial phase of the selection process are listed below.
provide a solution and knowledge and direct skills to the Eligibility Criteria
survivor and caregiver where the provision of chronic care is
rudimentary. The eligibility criteria have been presented in Textboxes 1 and
2.
Textbox 1. Inclusion criteria.
 Adult men and women aged 18 years of age
 Residents of Karachi and planning to live in Karachi till the follow-up period
 Able to understand Urdu (language of the videos) and the national language
 Admitted with first-ever stroke (ischemic or hemorrhagic)
 Modified Rankin Scale score4 (mild to moderate disability)
 Having at least one vascular risk factor that requires medical intervention
 Consenting to participate in the study and for follow-up visits, both stroke survivor and caregiver
 Have a designated caregiver at home who is responsible for appointments, follow-up, and overall care and are mobile phone literate, for
example, wife, daughters, daughter-in-law, and husband
 Stroke was medically stable, and participant was likely to return to the community after the in-hospital stay (thus actively treated strokes such
as decompressive surgeries, carotid endarterectomy, in-hospital sepsis, and ventilator complications that essentially preclude return to the
community settings were not offered in this chronic care support study).

Textbox 2. Exclusion criteria.

 Serious aphasia, visual hemineglect, short-term memory loss in the stroke survivor precluding understanding, visualization, or retention of the
video material.
 Serious aphasia, visual hemineglect, short-term memory loss, dementia in the caregiver precluding understanding, visualization, or retention of
the video material.
 Iatrogenic stroke, that is, stroke due to nonatherosclerotic vascular disease and rare causes, for example, carotid dissections, gunshot to neck,
and coronary artery bypass surgery
 Stroke survivor/caregiver dyad continued poststroke care in a nursing-assisted, professional, or hospital setting and does not return to the
community after discharge
 Serious concurrent medical illnesses such as cancer, renal failure, acute liver disease in past 6 months (that precludes use of statins), and
chronic liver disease, which that exclude the use of stroke preventive medications or require nonstandardized therapy
 Any use of off-label, nonguideline medications, because of the stroke survivors’ unique comorbidities, that interferes with medication
compliance to antihypertensive, statins, antiplatelet, and antidiabetic agents

Randomization Process a thorough explanation regarding the nature of the study and
Stroke survivors and their caregivers (dyad) were assigned to the scheduled follow-up visits. A detailed face-to-face
either the intervention or the usual care (control) groups in a interview of stroke survivors and their caregivers was
parallel manner in a ratio of 1:1. Block randomization conducted to gather data on sociodemographic and medical
technique with a fixed block size of 10 was used. A history. A baseline clinical and functional assessment was
computer-generated randomization list was used to randomize performed, after which they were randomly assigned into the
participants into the intervention or control group. The intervention or control group by a trained research officer
randomization center was performed in a secure computer in who was not blinded to the assignment of the intervention.
the clinical trials unit (CTU), and the randomization list was Details regarding the proper functioning of mobile app and its
generated by CTU staff not involved in recruitment, outcome installation were taught to the participants in the intervention
ascertainment, or any aspect of the study. group. A memory card containing Movies4Stroke app was
installed in the participant’s Android phone along with the
Allocation Concealment delivery of the first set of 5-min videos. Videos were shown at
The randomization list was centralized and thus not the time of enrollment into the study, at discharge, and at the
predictable. No one from the research team had any access to first and third month after discharge from the hospital. The
randomization list, randomization envelopes, and block size app was designed to provide access to the videos in a
or code. Envelopes were sealed and opaque, and it was scheduled manner. To maintain contact and follow-up, a
impossible to view the sequence even if held against bright Stroke Helpline number was provided to participants in both
sunlight. Randomization list and opaque envelopes containing the groups. The helpline number was active 24/7, and the
the randomization sequence were always kept inside the person operating at the helpline number was trained to receive
premises of CTU under lock and key. calls with the most frequently anticipated questions and to
answer them accordingly. The operator had access to a stroke
Identification and Enrollment of the Study specialist at all times for support. If the participant was
Participants allocated to the usual care group, he/she was informed about
Informed consent was obtained from eligible participants who the details regarding discharge, follow-up
volunteered to be a part of this study after they were provided
appointments at the clinic, and access to free lab vouchers at messages were sent through a Web-based, programmed,
the 6th and 12th months. open-access software entitled Frontline by a trained IT
Sampling Technique professional.
A purposive sample was selected from adult stroke survivors Intervention Group
admitted into the Stroke unit, Neurology Ward, AKUH, In addition to the usual care, the intervention group at the
Karachi, after the assessment of the eligibility criteria and time of admission received the introductory teaching session
obtaining informed consent. with installation of the app and the first set of 5-min videos on
Technical Part of the Intervention various stroke-related topics as described below. In the first
session, different skills such as swallowing exercises,
The Movies4Stroke app was developed by biomedical and
different rehabilitation exercises, and nasogastric tube feeding
software engineers of Aga Khan Development Network
were taught to the caregivers. The second session was
Electronic Health (eHealth) Resource Center in collaboration
delivered at the time of discharge, which included videos on
with stroke specialists, rehabilitation and swallowing experts,
emergency preparedness, such as cardiopulmonary
and epidemiologists. The intervention was first pilot tested on
resuscitation, seizures, heart attack, and hypoglycemia, while
study team members’ Android cell phones. Any discrepancies
simultaneously discussing and answering any queries the
and bugs were removed from the app. The intervention was
participants had after watching each set of videos. The third
then launched on tablets, specially purchased for showing
session was delivered at the first month of follow-up after
movies to the stroke survivors and their caregivers in clinical
discharge and included videos on frequently used medications
ward settings. Memory chips were also purchased, so that the
by stroke survivors, such as anticoagulants, antihypertensive,
Movies4Stroke app could be transferred into the participants’
and lipid-lowering drugs. The fourth session was delivered at
cell phones.
the third-month of follow-up after discharge and included
Participants in the intervention group were sent weekly videos on secondary stroke prevention (recurrent attack)—
messages twice as a reminder to watch the movies at home. exercise, physical activity, recognition of depression, diet
These modification, and accurate measurement of BP and blood
sugar levels (Figure 1).
Figure 1. Movies4Stroke collage. CPR: cardiopulmonary resuscitation.

Control Group Participants in the control group received the standard of care
that is provided to stroke patients at AKUH. Stroke survivors
were given instructions before discharge regarding diet, the to the videos as mentioned in the protocol [9]. There were
need for rehabilitation, possible complications, and trained research officers to ensure compliance of the
medication use; information booklets were also handed out. A intervention group participants at each video delivery;
multidisciplinary team comprising a neurophysician, stroke moreover, constant SMS reminders were sent to the study
nurse, dietitian, and physiotherapist imparted the information. participants in the intervention group (as a measure of
Verbal instructions were given to stroke survivors and their reinforcement) to watch these thematic movies in a relaxed
caregivers. On the day of discharge, or 24 hours before home environment, and they were also reminded about their
discharge, a discharge coordinator provided the researchers scheduled follow-up visits.
details about the skills learned and ensured that the medical,
social, and rehabilitation requirements were in place before Follow-Up
going out of the hospital. All the study participants were Follow-up visit for each stroke survivor-caregiver dyad was
provided follow-up appointments at the clinic. A detailed organized at 1, 3, 6, 9, and 12 months after discharge in the
written discharge summary was handed over to the caregiver, neurology clinic for outcome ascertainment. Stroke survivors
detailing all aspects of care, including follow-up visit, and their caregivers were given a handout with instructions
medications, lab investigations, and serious alerts. The and basic information about their subsequent follow-up visit.
control group compliance to this standard of care was Caregivers were also explained verbally about the importance
ensured, as all staff follow the abovementioned discharge of their follow-up visit. They were asked to contact the study
protocol and document education. There is designated staff team through the Stroke Helpline for any queries regarding
education and a safe discharge coordinator dedicated for all their health.
admissions at this center. The center at which this study was
performed is an internationally accredited center, and The follow-up rates were maximized by sending SMS
performance and documentation of these quality of care reminders to all the study participants about their respective
standards are a part of the standard of care protocol. This follow-up visits at least a day before their scheduled follow-
standard of care was followed for all participants in the up visit through our Stroke Helpline number and by also
control group because of these regulations that are in place to allowing an approximate 14-day grace period to the stroke
maintain accreditation and auditable quality of care. The survivor and caregiver who were unable to report as per their
control group did not receive the additional visual teaching of scheduled follow-up visit. Those participants who did not
the video-based intervention. This standard of care was appear for their scheduled follow-up visit were contacted
followed for all participants including those who received the through phone or approached through indirect means, such as
video-based intervention. contacting them when they came to the AKUH for any other
clinic or physiotherapy visit or lab investigations. Details of
Compliance During the Administration of the stroke survivors’ visit to the AKUH, other than the neurology
Intervention clinic, was obtained through telecommunication with the
A study officer who was not blinded to the intervention group caregiver or tracking the stroke survivor through the
took several measures to ensure compliance of the participants synchronized electronic medical record system of the AKUH
(Figure 2).
Figure 2. Study flow (enrollment to follow-up).

Participants’ Timeline Of the secondary outcomes, two are discussed in this paper
After recruiting stroke survivors along with their primary with respect to Movies4Stroke trial protocol; others will be
caregivers from the Stroke unit at the AKUH, the participant discussed in the subsequent paper.
dyads were not expected to come in for any additional visits
for the study purpose other than the scheduled 5 follow-up Stroke-Related Mortality Among Stroke
visits at the stroke clinic. Our study started enrolling Survivors Was Ascertained at 12 Months Post
participants from January 19, 2015, and the last participant Discharge
was recruited on May 15, 2015. The last follow-up was Information on mortality among stroke survivors after
completed on June 29, 2016. discharge was ascertained through a precoded and validated
verbal autopsy scale. In addition, we correlated all mortality
Trial Outcomes
with hospital records. Mortality was further categorized as per
The primary outcome measure reported was as follows: the analysis criteria into 3 categories: stroke-related mortality,
control of 3 major risk factors—BP, blood sugar, and lipids— mortality (because of nonstroke complications), and out-of-
measured via standardized methods in the central laboratory hospital mortality.
was ascertained at baseline and 6 and 12 months after
discharge. All the 3 risk factors were assessed, as the Included
proportion of participants achieving BP control (<125/85 mm Mortality after discharge from the AKUH because of stroke-
Hg), blood sugar (glycosylated hemoglobin A1c or related complications was assessed by a review of medical
HbA1c<7%), and blood cholesterol (low-density lipoprotein records, interviews from primary caregivers of the patients,
[LDL] level<100 mg/dL). and the verbal autopsy standard procedure.
Censored Ethics and Human Subject Protection
Mortality after discharge from the hospital because of Written informed consent, in both English and Urdu, was
competing risk, that is, cause of death not related to stroke obtained from all the study participants at the time of
because of, for example, head trauma, gun shot, and cancer, recruitment. The confidentiality and privacy of the
was assessed. participants was maintained by deidentification of the subject
Patients who were alive after 1 year of follow-up after information. Only research staff was authorized entry into the
discharge (censored because of lack of outcome of interest or hospital system on the computers that were used for data
statistical considerations) were assessed. storage. All source documents were maintained in locked files
in locked room. Fingerprint encryption was added to all
Excluded sensitive data, for example, mobile numbers, app logs, and
In-hospital mortality before patients were discharged from the error logs. The Ethical Review Committee (ERC) of Aga
hospital was a result of complications arising from an index Khan University, Karachi (ERC number 2890-Med-ERC-14),
stroke occurring before discharge into the community in a approved the study.
stable state, for example, iatrogenic complications, sepsis, and
progression of index stroke during admission. Results
Moreover, the categorization of all these deaths was further Overall Trial Flow
validated by our team of experts including stroke specialist,
epidemiologist, statisticians, and research supervisor. A total of 310 stroke survivors and their caregiver dyads, ie,
620 individuals (participant dyads), were recruited over a
Stroke Disability and Neurological Deficits Among duration of 6 months. As this clinical trial had a fixed block
Stroke Survivors Was Ascertained at Baseline and 6 design, 155 participant dyads were randomized in each of the
and 12 Months After Discharge intervention and control group (310 in each group). We
A total of 3 different neurological functional assessment tools screened 400 participant dyads to assess eligibility; of these,
were used in this clinical trial, as each of them 50 were not eligible, and 40 participant dyads refused to
captures/measures a different parameter with respect to the participate in the study (30% were excluded; Figure 3). The
functional status of stroke survivors after acute stroke: reasons for exclusions were mRS>4 (n=25), travel plans
(n=15), and non-Karachi residents (n=10). The reasons for
 mRS tool is widely used by neuro physicians globally to refusal were mainly the lack of ability to return for follow-up
assess functional disability after acute stroke [10-12]. and personal interest of the stroke survivor or caregiver to
 The National Institutes of Health Stroke Scale (NIHSS) participate in the study. We were able to complete information
tool is widely used by neurophysicians globally to on 141 participant dyads in the intervention and 137 in the
objectively quantify the impairment caused by stroke control group at the end of 1-year postdischarge follow-up.
[12,13]. From the intervention group, 11 participant dyads were lost to
 The Barthel Index (BI) tool is used globally by health follow-up at 1 year postdischarge and 3 stroke survivors died
care providers to assess the level of dependency among because of in-hospital mortality (before being discharged)
stroke survivors after an acute episode of stroke [12,14]. resulting from inpatient complications from the index stroke,
as compared with 15 participant dyads who were lost to
The abovementioned tools were used to assess the functional follow-up and 3 who died because of in-hospital mortality in
status of stroke survivors at baseline and 6- and 12-month the control group. There was one protocol violation in the
follow-ups. control group that was excluded from the final analysis.
Figure 3. Trial flow diagram. mRS: Modified Rankin Scale.

Baseline Characteristics of the Study Participants intervention group as compared to the control group
Mean age of stroke survivors in the intervention group was (124/155, 80.0%), and more than half of the stroke survivors
60.6 (SD 12.0) years, whereas it was 59.7 (SD 14.3) years in were living in a joint family system (93/155, 60.0%, in the
the control group. The caregivers were relatively younger, intervention group vs 107/155, 69.0%, in the control group).
with the mean age of 38.7 (SD 11.7) years in the intervention Median (interquartile range) number of household members
group and living with stroke survivors was 6 (range: 5-8) in the
39.8 (SD 14) years in the control group. In our trial, most of intervention group as compared with 7 (range: 5-10) in the
the stroke survivors were males (109/155, 70.3%, in the control group. Median household monthly income of the
intervention group vs 100/155, 65.0%, in the control group). stroke survivors was Rs 50,000 (US $416) in both the groups.
More than two-thirds of the stroke survivors in our study had Approximately, four-fifths of our study participants were
more than 5 years of education (114/155, 73.5%, in the Android mobile phone users in the intervention group
intervention group vs 107/155, 69.0%, in the control group), (120/155, 77.4%) as compared with around two-third in the
with 40 of 155 (25.8%) patients being employed in the control group (99/155, 63.8%). Most variables were
intervention group and 39 of 155 (25.1%) in the control uniformly distributed between the two groups and not
group. More than four-fifths of our study participants were statistically significant at baseline (refer to Table 1).
married (134/155, 86.5%) in the
Table 1. Baseline characteristics of study participants according to their group allocation (310 participant dyads)

Baseline characteristics Intervention group (N=155) Control group (N=155) P value

Age (years) of the patient, mean (SD) 60.6 (12.0) 59.7 (14.3) .17
Patient’s gender, n (%) .35
Male 109 (70.0) 100 (65.0)
Female 46 (30.0) 55 (35.0)
Age (years) of the caregiver, mean (SD) 38.7 (11.7) 39.8 (14.0)
Source of the patient, n (%) .26
Direct 90 (58.0) 81 (52.2)
Referral 65 (42.0) 74 (47.8)
Type of operating system, n (%) .01
None 14 (9.0) 30 (19.3)
Android 127 (82.0) 104 (67.1)
Windows 6 (3.9) 13 (8.4)
IOS 8 (5.1) 8 (5.2)
Anyone at home with Android, n (%) <.001
Yes 120 (77.4) 99 (63.8)
No 35 (22.6) 56 (36.2)
Desktop PC at home, n (%) .02
Yes 143 (92.3) 130 (83.8)
No 12 (7.7) 25 (16.2)
Patient education, n (%) .22
Illiterate 17 (11.0) 31 (20.0)
Primary education 24 (15.5) 17 (11.0)
Secondary education 36 (23.2) 34 (22.0)
Higher secondary education 21 (13.5) 18 (11.6)
Above intermediate 57 (36.8) 55 (35.4)
Patient marital status, n (%) .14
Single 3 (1.9) 9 (5.8)
Married 134 (86.5) 124 (80.0)
Widowed 18 (11.6) 22 (14.2)
Patient family status, n (%) .12
Joint family 93 (60.0) 107 (69.0)
Nuclear family 62 (40.0) 48 (31.0)
Monthly family income (Pakistani rupee), median (IQR) 50,000 (30,000-100,000) 50,000 (30,000-70,000)
Patient employment status, n (%) .45
Employed 40 (25.8) 39 (25.1)
Unemployed 5 (3.3) 10 (6.5)
Retired 27 (17.4) 27 (17.4)
Housewife 43 (27.7) 50 (32.2)
Others 40 (25.8) 29 (18.7)
Family land ownership (acre), n (%) .92
None 47 (30.3) 47 (30.3)
<1 92 (59.4) 89 (57.4)
 Baseline characteristics Intervention group (N=155) Control group (N=155) P value
Between 1 and 10 10 (6.4) 13 (8.4)
>10 6 (3.9) 6 (3.9)
Total household members, median (IQR) 6 (5-8) 7 (5-10) .52
Length of hospital stay, median (IQR) 4 (3-4) 3 (3-5) .71
Tissue plasminogen activator, n (%) .98
Yes 7 (4.5) 7 (4.5)
No 148 (95.5) 148 (95.5)

Primary Outcome mm Hg was similar in the intervention group and control

group (36/109, 50.0% vs 36/99, 50.0%; risk ratio [RR] 0.91,
Systolic Blood Pressure 95% CI
At baseline, there was a uniform distribution of study 0.62-1.32). At the final follow-up visit (at 12 months), there
participants with the two categories of systolic BP (<125 mm was a greater percentage of participants with a systolic BP of
Hg and >125 mm Hg) in the intervention group and the <125 mm Hg in the intervention group than in the control
control group (49/155, 50.0%, vs 48/155, 50.0%). At the 6- group (18/54, 62.0% vs 11/52, 38.0%; RR 1.58, 95% CI
month visit, the distribution of study participants with a 0.83-2.98).
systolic BP of <125 However, none of these results were statistically significant
(refer to Table 2).

Table 2. Systolic blood pressure.

Systolic blood pressure (mm Hg)
Intervention group, n (%) Control group, n (%) Risk ratio (95% CI) P value (overall)
a
Baseline results (N=310) .96
<125 49 (50.0) 48 (50.0) 1.00 (0.72-1.40)
>125 106 (50.0) 107 (50.0) 1.00 (0.72-1.40)

b
6-month results (N=208) .61
<125 36 (50.0) 36 (50.0) 0.91 (0.62-1.32)
>125 73 (54.0) 63 (46.0) 0.91 (0.62-1.32)

12-month results (N=106)

c .16

<125 18 (62.0) 11 (38.0) 1.58 (0.83-2.98)

>125 36 (47.0) 41 (53.0) 1.58 (0.83-2.98)

a
Baseline: intervention group (n=155) and control group (n=155).
b
6-month results: intervention group (n=109) and control group (n=99).
c
12-month results: intervention group (n=54) and control group (n=52).

group (82/109, 55.0% vs 68/99, 45.0%; RR 1.10, 95% CI

Diastolic Blood Pressure 0.92-1.30). Similar to this, at the final visit at 12 months, there
At baseline, there were slightly more participants with a was a greater percentage of participants with a diastolic BP of
diastolic BP of <85 mm Hg in the intervention group than in <85 mm Hg in the intervention group than in the control
the control group (109/155, 51.0% vs 106/155, 49.0%). At 6 group (44/54, 55.0% vs 37/52, 45.0%; RR 1.15, 95% CI
months, there was a greater percentage of stroke survivors 0.92-1.42).
with a diastolic BP of <85 mm Hg in the intervention group These results failed to reach a statistically significant level
than in the control (refer to Table 3).
Table 3. Diastolic blood pressure.
Diastolic blood pressure
Intervention group, n (%) Control group, n (%) Risk ratio (95% CI) P value (overall)
(mmHg) Baseline results .84
a
(N=310)
<85 109 (51.0) 106 (49.0) 1.02 (0.88-1.18)
>85 46 (49.0) 49 (51.0) 1.02 (0.88-1.18)
 b
6-month results (N=208) .29
<85 82 (55.0) 68 (45.0) 1.10 (0.92-1.30)
>85 27 (46.0) 31 (54.0) 1.10 (0.92-1.30)
c
12-month results (N=106) .21
<85 44 (55.0) 37 (45.0) 1.15 (0.92-1.42)
>85 10 (40.0) 15 (60.0) 1.15 (0.92-1.42)

a
Baseline: intervention group (n=155) and control group (n=155).
b
6-month results: intervention group (n=109) and control group (n=99).
c
12-month results: intervention group (n=54) and control group (n=52).

group, (81/105, 52.0% vs 74/91, 48.0%; RR 0.95, 95% CI

Glycosylated Hemoglobin A1c 0.82-1.10). Similarly, at the final visit at 12 months, there was
At baseline, there was a smaller percentage of participants a greater percentage of participants with an HbA 1c level <7%
with an HbA1c level <7% in the intervention group than in the in the intervention group than in the control group (36/55,
control group (88/155, 45.0% vs 105/155, 55.0%). At the 6- 55.0% vs 30/40, 45.0%; RR 0.87, with 95% CI 0.66-1.14).
month visit, there was a slightly higher percentage of stroke However,
survivors with a HbA1c level <7% in the intervention group these results failed to demonstrate statistical significance
than in the control (refer to Table 4).

Table 4. Glycosylated hemoglobin A1c.

Glycosylated hemoglobin A1c

(%) Intervention group, n (%) Control group, n (%) Risk ratio (95% CI) P value (overall)

a .03
Baseline results (N=310)
<7 85 (45.0) 105 (55.0) 0.81 (0.68-0.98)
>7 70 (58.0) 50 (42.0) 0.81 (0.68-0.98)
b
6-month results (N=196) .47
<7 81 (52.0) 74 (48.0) 0.95 (0.82-1.10)
>7 24 (58.0) 17 (42.0) 0.95 (0.82-1.10)

12-month results (N=95)

c .32

<7 36 (55.0) 30 (45.0.0) 0.87 (0.66-1.14)

>7 19 (65.0) 10 (35.0) 0.87 (0.66-1.14)

a
Baseline: intervention group (n=155) and control group (n=155).
b
6-month results: intervention group (n=105) and control group (n=91).
c
12-month results: intervention group (n=55) and control group (n=40).

(73/106, 57.0% vs 56/90, 43.0%; RR 1.10, 95% CI 0.90-1.36).

Low-Density Lipoprotein Again, a very similar trend was seen at the fifth follow-up
At baseline, there was a greater percentage of study visit: There was a greater percentage of participants with an
participants with an LDL level<100 mg/dL in the intervention LDL level <100 mg/dL in the intervention group than in the
group than in the control group (92/155, 59.0% vs 64/155, control group (36/51, 55.0%, vs 30/45, 45.0%; RR 1.06,
41.0%). A similar trend was seen at the third follow-up visit: 95% CI
There was a higher percentage of stroke survivors with an 0.81-1.39). However, these results failed to demonstrate
LDL level<100 mg/dL in the intervention group than in the statistical significance (refer to Table 5).
control group
Table 5. Low-density lipoprotein.
Low-density lipoprotein (mg/dL)
Intervention group, n (%) Control group, n (%) Risk ratio (95% CI) P value (overall)
a <.001
Baseline results (N=310)
<100 92 (59.0) 64 (41.0) 1.42 (1.13-1.78)
>100 63 (41.0) 91 (59.0) 1.42 (1.13-1.78)

b
6-month results (N=196) .33
 <100 73 (57.0) 56 (43.0) 1.10 (0.90-1.36)
>100 33 (49.0) 34 (51.0) 1.10 (0.90-1.36)
c
12-month results (N=96) .68
<100 36 (55.0) 30 (45.0) 1.06 (0.81-1.39)
>100 15 (50.0) 15 (50.0) 1.06 (0.81-1.39)

a
Baseline: intervention group (n=155) and control group (n=155).
b
6-month results: intervention group (n=106) and control group (n=90).
c
12-month results: intervention group (n=51) and control group (n=45).

Censored deaths in both the groups were caused by non–

Secondary Outcomes stroke-related complications and were therefore not
Mortality and the Number Needed to Treat considered a part of the included group. The results were
found to be highly significant (P<.001). Absolute risk
Overall, 35 deaths were reported over the course of 1-year
reduction (ARR) of mortality related to stroke-related
follow-up in the Movies4Stroke trial (Multimedia Appendix
complications was 7%, which yielded a number needed to
1). Mortality among stroke survivors because of stroke-
treat (NNT) of 15. It meant that we needed to show the video-
related complications (included group) was higher in the
based intervention to 15 stroke survivors to prevent 1 death
control group than in the intervention group (13/155, 8.4% vs
from stroke-related complications (refer to Table 6 and Figure
2/155, 1.3%). The most common cause of mortality was
4).
aspiration pneumonia.

Table 6. Categorization of mortality, ARR, and NNT.

Mortality categorized
Intervention group Control group Absolute risk Number needed to P value
(N=155), n (%) (N=155), n (%) reduction treat
Patient is alive
140 (90.3) 135 (87.1) N/A
Included N/A
a N/A
2 (1.3) 13 (8.4) <.001
Censored 7.1 b
15
10 (6.5) 4 (2.6) N/A
Excluded N/A N/A
3 (1.9) 3 (1.9) N/A
a N/A N/A
Not applicable.
b
Number needed to treat of 15 has been rounded off to a whole number.
Figure 4. The Kaplan-Meier survival estimates in the first year after stroke in the intervention versus control group because of stroke-related avoidable
mortality.

Change in Functional Status (Disability and higher percentage of stroke survivors with moderate to severe
Severity) Among Stroke Survivors disability were present in the intervention group than in the
control group (24/135, 17.8% vs 18/129, 14.0%; OR 1.28,
Modified Rankin Scale
95% CI 0.65-2.55). However, at 12 months, a higher
Table 7 and Figures 5 and 6 show that at baseline, as assessed percentage of stroke survivors had minimal to no disability in
by mRS, a higher percentage of stroke survivors with the intervention group than in the control group (91/128,
moderate to severe disability were present in the intervention 71.1% vs 71/120, 59.2%). At 12 months postdischarge, NNT
group than in the control group (46/155, 29.7% vs 36/155, was 9, as assessed by the mRS. This meant that we needed to
23.2%; odds ratio [OR] 1.18, 95% CI 0.58-2.39). Similarly, show Movies4Stroke to 9 stroke survivors to achieve minimal
at 6 months, a to no disability caused by stroke after a year of follow-up.
Table 7. Modified Rankin Scale.
Modified Rankin Scale score
Intervention group, n (%) Control group, n (%) Odds ratio (95% CI) P value (overall)
a .40
Baseline results (N=310)

0-1 25 (16.1) 25 (16.1) Reference

2-3 84 (54.2) 94 (60.7) 0.82 (0.44-1.55)
4-5 46 (29.7) 36 (23.2) 1.18 (0.58-2.39)

6-month results (N=264)

b .40

0-1 82 (60.7) 79 (61.2) Reference

2-3 29 (21.5) 28 (21.7) 0.99 (0.55-1.83)
4-5 24 (17.8) 18 (14.0) 1.28 (0.65-2.55)
Death 0 (0.0) 4 (3.1)
—c
d
12-month results (N=248) .07
0-1 91 (71.1) 71 (59.2) Reference
2-3 18 (14.1) 30 (25.0) 0.47 (0.24-0.91)
4-5 19 (14.8) 19 (15.8) 0.78 (0.38-1.58)
a
Baseline: intervention group (n=155) and control group (n=155).
b
6-month results: intervention group (n=135) and control group (n=129).
c
Odds ratio with their 95% CI could not be estimated because of the empty cell count in the intervention group.
d
12-month results: intervention group (n=128) and control group (n=120).

Figure 5. Functional status of stroke survivors—6 months. NIHSS: National Institutes of Health Stroke Scale.
Figure 6. Functional status of stroke survivors—baseline and 12 months assessment (comparison).

National Institutes of Health Stroke Scale 25/129, 19.4%; OR 0.98, 95% CI 0.50-1.92). At 12 months, a
Table 8 and Figures 5 and 6 show that at baseline, as assessed higher percentage of stroke survivors had minimal to no
by the NIHSS, a higher percentage of stroke survivors had impairment caused by stroke in the intervention group than in
severe impairment caused by stroke in the control group than the control group (65/128, 50.8% vs 55/120, 45.8%). At 12
in the intervention group (72/155, 46.4% vs 61/155, 39.4%; months postdischarge, NNT was 20 as per the NIHSS
OR 0.82, 95% CI 0.39-1.74). At 6 months, a smaller assessment. This meant that we needed to show
percentage of stroke survivors had severe impairment caused Movies4Stroke to 20 stroke survivors to have minimal to no
by stroke in the intervention group than in the control group disability caused by stroke after a year of follow-up.
(23/135, 17.0% vs
Table 8. National Institutes of Health Stroke Scale.
National Institutes of
Health Stroke Scale score Intervention group, n (%) Control group, n (%) Odds ratio (95% CI) P value
(overall)
a
Baseline results (N=310) .52
0-1
18 (11.6) 17 (11.0) Reference
2-8
76 (49.0) 66 (42.6) 1.09 (0.51-2.28)
>9
61 (39.4) 72 (46.4) 0.82 (0.39-1.74)
b
6-month results (N=264) .08
0-1 58 (43.0) 62 (48.0) Reference
2-8 54 (40.0) 38 (29.5) 1.52 (0.88-2.63)
>9 23 (17.0) 25 (19.4) 0.98 (0.50-1.92)
Death 0 (0.0) 4 (3.1)
—c
d
12-month results (N=248) .73
0-1 65 (50.8) 55 (45.8) Reference
2-8 42 (32.8) 44 (36.7) 0.81 (0.46-1.41)
>9 21 (16.4) 21 (17.5) 0.85 (0.42-1.71)

a
Baseline: intervention group (n=155) and control group (n=155).
b
6-month results: intervention group (n=135) and control group (n=129).
c
Odds ratio with their 95% CI could not be estimated because of the empty cell count in the intervention group.
d
12-month results: intervention group (n=128) and control group (n=120).

As evident in Figure 7 and according to the mRS, when

Barthel Index comparing the baseline with the 12-month follow-up visit,
Table 9 and Figures 5 and 6 show that at baseline, as assessed more survivors in the intervention group had seen the videos
by the BI, an equal percentage of stroke survivors had total to and had minimal to no disability at the end of the 12 months
severe dependency in the intervention group as compared as compared with the control group. Similarly, according to
with the control group (78/155, 50.3% vs 78/155, 50.3%; OR NIHSS, when comparing baseline with 12-month follow-up
1.08, 95% CI 0.63-1.86). At 6 months, a smaller percentage visits, there was a higher percentage of survivors with
of stroke survivors had total to severe dependency in the minimal neurologic deficit at 12 months in the intervention
intervention group than in the control group (22/135, 16.3% group as compared with the control group. Similarly, for BI,
vs 23/129, 17.8%; OR 0.93, 95% CI 0.48-1.81). At 12 when comparing baseline with 12-month follow-up visits,
months, a higher there was a higher percentage of survivors with had minimal
percentage of stroke survivors with minimal to no to no dependency at 12 months in the intervention group as
dependency were present in the intervention group than in the compared with the control group.
control group (87/128, 68.0% vs 71/120, 59.2%). At 12
months postdischarge, NNT was 12, as per the BI assessment.
Table 9. Barthel Index.
Barthel Index score
Intervention group, n (%) Control group, n (%) Odds ratio (95% CI) P value (overall)
a .94
Baseline results (N=310)
95-100
38 (24.5) 40 (25.8) Reference
55-90
39 (25.2) 37 (23.9) 1.11 (0.59-2.09)
0-50
78 (50.3) 78 (50.3) 1.08 (0.63-1.86)
b
6-month results (N=264) .16
95-100 75 (55.6) 73 (56.6) Reference
55-90 38 (28.1) 29 (22.5) 1.28 (0.71-2.28)
0-50 22 (16.3) 23 (17.8) 0.93 (0.48-1.81)
Death 0 (0.0) 4 (3.1)
—c
d
12-month results (N=248)
 .35
95-100
87 (68.0) 71 (59.2) Reference
55-90
23 (18.0) 27 (22.5) 0.70 (0.37-1.32)
0-50
18 (14.0) 22 (18.3) 0.67 (0.33-1.34)

a
Baseline: intervention group (n=155) and control group (n=155).
b
6-month results: intervention group (n=135) and control group (n=129).
c
Odds ratio with their 95% CI could not be estimated because of the empty cell count in the intervention group.
d
12-month results: intervention group (n=128) and control group (n=120).

Figure 7. Functional status of stroke survivors—baseline and 12 months assessment (comparison).

Discussion understanding in a literacy-challenged population that,

however, had access to a cell phone. None of the trial
Principal Findings and Their Contextual Relevance participants were excluded because of the lack of a cell phone.
This mHealth trial failed to reach its primary outcome measure
Movies4Stroke is a randomized controlled mHealth trial
of the control of hypertension, LDL cholesterol, and HbA 1c;
evaluating the effectiveness and safety of a phone-based
intervention that showed thematically designed videos to however, the prespecified secondary variables of improved
assist the stroke survivor and caregiver dyad to get the functional outcomes and reduced mortality were improved in
knowledge, skills, and confidence needed to improve the intervention group in this study because of stroke-related
poststroke risk factor control, survival, and functional complications. This study demonstrated that showing the video
outcomes. The rationale of this trial was to provide repetitive to a literacy-challenged population in a country without any
high-quality survival and training to stroke survivor and chronic care health facility saved lives. The NNT was only 15,
caregiver dyads returning to a low- to middle-income to save 1 life, because of avoidable stroke-related
community where rehabilitation and chronic care systems are complications and death. This trial provides early data on
underdeveloped. The mode of video-based IT intervention, nonpharmacological intervention in stroke survivors. There are
along with competency and understanding checks and the few interventions of this nature identified in a recent meta-
ability to repeat and see a video, was used to assist the analysis and review and none that employs mHealth in a
complex, challenging, and resource-strapped setting (refer to
Multimedia Appendix 2) [15-21]. with the later stages of stroke. Hence at admission, the
Primary and Secondary Outcome Results: survivor and caregiver dyad were provided with skills and
Mechanistic Explanations training to recognize life-threatening emergencies and
respond to them accordingly. At the time of discharge, each
Movies4Stroke failed to achieve its primary outcome variable dyad was taught rehabilitation and safe swallowing. Thematic
of the control of primary risk variables of hypertension, LDL movies on adherence and medication safety module were
cholesterol, or HbA1c in the intervention. There could be taught in the first month; thus, these preventive aspects were
several reasons for these findings. First, the baseline dealt with much later, and this might in itself be associated
adherence in this population was already substantially better with learning fatigue, or perhaps, these movies were taught by
than the general population, possibly because of the the time the risk factor control had already maximized in the
seriousness of stroke and exposure to specialist care and participants in the study.
counseling techniques; thus, it became difficult to further
improve adherence and the related improvement in risk Finally, it is difficult to design measures that capture pill
factors. This could be explained by the Health Belief Model compliance in the stroke population; stroke patients have
demonstrating that better outcomes are found in patients who diverse prescriptions that vary in the type of drug classes,
understand the severity of their disease and the benefits that number, and frequency of dosage, and no single biomarker
changing their actions would yield [22,23]. would be applicable to all the study participants nor could
electronic pill boxes capture the actual drug usage to achieve
Second, in this study, to maintain long-term follow-up, the BP, glucose, or cholesterol control, as they record the number
control group also had access to a 24/7 helpline, and of times a box is opened. Moreover, it cannot be assumed that
interaction with health care personnel who assisted and the participants have consumed all the pills for that dose when
resolved queries may have improved behavior in this group as they open a box. However, we can see that the harder
well. The presence of a centralized helpline has shown to help outcome measure of the control of BP, cholesterol, or blood
address queries and patients cope with difficult situations sugar was not achieved.
[24].
In addition, these negative findings on risk factor control shed
Third, this was a sequential and thematic intervention where light on the importance of health theory and behavioral
these 5-min video lessons were rolled out in order of what science designing an intervention, rather than considering just
skill would be important for survival in the early stages as simple knowledge transfer while designing an intervention.
compared This resonates with previous studies concluding that to
achieve effective change in health systems and patient care,
knowledge transfer alone is insufficient and has to be
supplemented with other forms of intervention [25]. In a
single-center study [26], showing just one video in an in-
patient setting increased knowledge; however, it did not
translate into BP monitoring in a home environment or
increase physician follow-up visits. The sustained
intervention in our study revealed that, at least the skills to
prevent complications might have been acquired.
Nonadherence to medication has two components—
intentional and unintentional nonadherence [27]. Intentional
nonadherence refers to nonadherence that is deliberate and
may be because of motivational factors that may be directly
related to the perceived efficacy of the medicine, distrust, and
the lack of knowledge, whereas unintentional nonadherence is
nonadherence that is largely related to either forgetting or the
lack of capacity to take the medicine. In our study, we focused
on improving knowledge, empowerment, and self-efficacy, but
we did not send repeated reminders to improve unintentional
nonadherence. Similar results were obtained in a study
implemented in Sydney, Australia, focusing on participants
with coronary heart disease in which repeated lifestyle
modification messages/reminders showed significant results
with respect to harder outcomes such BP and cholesterol [28].
This highlights the importance of mHealth design to be
multifaceted and used as a platform for re-enforcement to
address behavior change with regard to adherence. It also
highlights the potential of mHealth to be harnessed in
reducing the burden of NCDs when interventions are
designed using health theory, and they can inform
effectiveness, thus achieving the millennium development
goals [29,30].
Interestingly, each of our prespecified secondary outcome encouraged participants to continue with the exercises. The
variables improved in this study. Secondary outcomes of added feature of being able to replay and rewatch these videos
Movies4Stroke trial aimed at improving functional outcomes might have been another factor that aided their
and reducing mortality because of stroke-related
complications, and the results are very promising, given the
difficult context of the intervention. The mortality analysis
showed that in the control group, there were 9 of 13 cases of
massive aspiration pneumonias that resulted in mortality,
which could have been avoided by aspiration precautions and
learning tube feeding safety measures. In addition,
recognizing aspiration and reporting early could have resulted
in saving lives. There were 2 mortalities in the intervention
group, one because of a massive recurrent stroke and the other
because of aspiration. The other mortalities were not related
directly to stroke-mediated complications. Given the limited
nature of these observations, these findings are worthy of
attention but need to be confirmed in larger cohorts with
prolonged follow-up periods.
Skills to help identify, prevent, and respond to poststroke
complications were taught through 5-min videos that had
been rolled out in an orderly manner, with preventable early
complications addressed first, followed by late complications.
This method of interactive, repeated teaching increased the
skill and confidence of the caregivers and thus resulted in
saving lives and improving the functional outcomes and
reducing disability.
Other studies have similarly reported that timely recognition
and reporting of complications resulted in low mortality and
improved outcomes among stroke survivors [31,32]. Through
these movies, knowledge transfer helped change the skills of
the caregivers rather than the adherence of stroke survivors.
Along with this, Movies4Stroke might have influenced health
beliefs of stroke survivors and their caregivers by helping
them understand their increased susceptibility and therefore
the importance of prompt action to prevent death because of
complications [33,34]. Moreover, Movies4Stroke also had
videos aimed at providing psychoeducational intervention and
support through the helpline to the caregivers along with the
procedural knowledge. These three components have been
pivotal in helping prevent stress and strain in informal
caregivers [19], which may also explain the active
involvement and effective skill knowledge of caregivers in
this study.
Owing to the varying definitions of successful stroke
recovery, it is difficult to find a standard measure of recovery,
which is why we used three different scales [35]. The
improvement observed in functional outcomes in the
intervention group might have been because of the learning
model that these videos were based upon. Rehabilitation
measures and approaches were described and taught in the
videos. However, the demonstration of each, by registered
professional physiotherapists, nurses, stroke survivors with
significant disability, and speech and swallowing experts,
helped augment the learning process and made it easy for
patients and their caregivers to mimic and follow each step
taught effectively. One of the most important aspects of these
movies was watching real stroke survivors perform the
exercises themselves; it acted as a morale booster and further
learning process, resulting in significant improvement in randomized controlled trial design at a center where an
functional status. internationally certified care model is followed with an
algorithmic approach; hence, the results generated can be
These results are encouraging in several ways. First, as the
attributed to the effect of intervention.
extent of improvement in the functionality of stroke
survivors demonstrates the ability to return to an improved The risk of bias was minimized by paying attention to
functionality level with the help of only these movies and domains that defined trial quality [19,40,41]. Selection bias
their primary caregivers without the need of a skilled nurse was reduced by random centralized computer-based sequence
or a standard rehabilitation center. This is similar to the generation; allocation was concealed by opaque sequentially
results of a trial that compared and concluded equivalent numbered envelopes that were dispatched by CTU, which
improvements in patients who received home-based included university-based staff, which was separate from the
rehabilitation therapy with a caregiver and others who research team. Performance and detection bias was reduced
received therapy with a professional therapist, thereby by blinded outcome assessment and cross-checks of
establishing home-based rehabilitation as equally effective, contamination during the course of the trial and uniform
as shown with Movies4Stroke [36]. These video lessons training of staff on study. Attrition bias was addressed by
filled the niche that the lack of these interventions might multiple means of follow-up—maintaining good level of
have created and bridged the gap between learning and understanding with the participants, sending SMS reminder
implementation process [37]. Although there have been for follow-ups, and preserving communication through Stroke
several discussions calling for attention to NCDs at national Helpline number. We were successful in keeping an overall
and international level to help meet the MDGs, there has also dropout rate of less than 9%, with sensitivity analysis
been concern regarding the high costs of treatment and showing that were no significant differences between the
recurrence of debilitating events such as cardiovascular characteristics of dropouts/those lost to follow-up and those
death. These have prompted debate among the policy makers who continued the study in both the groups (Multimedia
to establish ways for primary prevention, affordable Appendix 3). Reporting bias was reduced by reporting all
treatment measures, and monitoring of NCDs [38,39]. possible outcomes of interest as defined in the protocol.
Movies4Stroke is a step in this direction, aimed at providing Intention-to-treat analysis has been used to report outcomes.
better health care, and prevention through diet and healthy
activity, thus curbing the alarming burden of NCDs in To ensure compliance to the intervention, the first time, the
LMICs videos were viewed in the presence of a research officer, and
intervention was delivered at 100% rate, with complete
Trial Strengths verbalized understanding and unhurried visualization. To
The major strengths of this trial were the use of a track

usage patterns at home, we remotely monitored the access and however, most stroke survivors were on different schedules for
use of our app that had been installed in the mobile phone. follow-up visits and rehabilitation times, and so, we expected
Early results show that the rehabilitation and tube feeding less contamination than in areas where a lot of time is spent
videos were the highly accessed videos at home. These together by families.
analyses are still ongoing.
Trial Limitations
The major limitation of our clinical trial is that it is a single-
center study, chosen because of the fact that this site provided
a standard of care that is algorithmic and replicable; thus,
results generated can be attributed to the effect of
intervention. In this study, we reported its efficacy, but the
performance and potential effect size in different sites may be
variable and potentially be more effective, given the usual
care standards in even more resource-strapped comparative
health systems, for example, the government sector, and thus,
external validity is limited. Furthermore, those with more
severe stroke or better health access because of
socioeconomic status are more likely to visit the health
systems and volunteer to participate in an educational training
intervention, thus limiting direct external validity. We would
definitely include nonresponse analysis in futures studies to
further identify characteristics of those most likely to adopt
the intervention. Another significant limitation is of
contamination bias in an educational intervention. Care was
taken to avoid contamination of the nonintervention group
with the intervention. To ensure this, videos were shown in a
separate room and not at the bedside. Given the fact that
families share information, contamination was possible;
Obviously, it is an inherent limitation of an educational interventions are safe and feasible, despite their complexity.
intervention that blinding of participants is not possible. This Our results demonstrate the key importance of health theory
intervention in a study setting required human resources in in designing these complex health interventions for
terms of a study officer and IT back up, development replicability and informing further interventions.
required human resource, and deployment required at least Interventions that target compliance must have repetitive
three staff in a study setting. To predict feasibility, we need reminders for nonintentional adherence, and those that target
to further analyze cost-effectiveness and realistic compliance knowledge and skills transfer must have the capacity to repeat
in a clinical setting with the current patient volumes, so this and bolster confidence as well as provide the user the
limits directly recommending applicability. This study has opportunity to model themselves from the materials taught,
collected data on cost-effectiveness, which is under analysis despite the lack of literacy skills to be safe. Complex
at this time. interventions targeting these settings need to have a design
theory in place to deliver both these aspects to be effective.
Way Forward
In conclusion, we demonstrate the potential of mHealth Please refer to Consolidated Standards of Reporting Trials-
interventions to save lives and reduce disability in low- to eHealth checklist for details regarding this Movies4Stroke trial
middle-income country settings; we also show that these [42].

Acknowledgments
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the paper. The
content is solely the responsibility of the authors and does not necessarily represent the official views of the Fogarty
International Center (National Institutes of Health or NIH), National Institute of Neurological Disorders and Stroke (NIH), and
University Research Council of Aga Khan University.

Conflicts of Interest
None declared.

Multimedia Appendix 1
Detailed Mortality Analysis.
[DOCX File , 24 KB-Multimedia Appendix 1]

Multimedia Appendix 2
Non-pharmacological interventions targeting stroke survivors and their caregivers.
[DOCX File , 18 KB-Multimedia Appendix 2]

Multimedia Appendix 3
Analysis of Lost to Follow Up and Mortality.
[DOCX File , 14 KB-Multimedia Appendix 3]

Multimedia Appendix 4
CONSORT-EHEALTH checklist V1.6.1.
[PDF File (Adobe PDF File), 2557 KB-Multimedia Appendix 4]
References
1. Feigin VL, Lawes CM, Bennett DA, Barker-Collo SL, Parag V. Worldwide stroke incidence and early case fatality
reported in 56 population-based studies: a systematic review. Lancet Neurol 2009 Apr;8(4):355-369. [doi:
10.1016/S1474-4422(09)70025-0] [Medline: 19233729]
2. Feigin VL, Roth GA, Naghavi M, Parmar P, Krishnamurthi R, Chugh S, Global Burden of Diseases‚ Injuries and Risk
Factors Study 2013 and Stroke Experts Writing Group. Global burden of stroke and risk factors in 188 countries, during
1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet Neurol 2016 Aug;15(9):913-
924. [doi: 10.1016/S1474-4422(16)30073-4] [Medline: 27291521]
3. Jafar TH. Blood pressure, diabetes, and increased dietary salt associated with stroke--results from a community-based
study in Pakistan. J Hum Hypertens 2006 Jan;20(1):83-85. [doi: 10.1038/sj.jhh.1001929] [Medline: 16151442]
4. Khan M, Ahmed B, Ahmed M, Najeeb M, Raza E, Khan F, et al. Functional, cognitive and psychological outcomes,
and recurrent vascular events in Pakistani stroke survivors: a cross sectional study. BMC Res Notes 2012 Feb 9;5:89
[FREE Full text] [doi: 10.1186/1756-0500-5-89] [Medline: 22321339]
5. Rathore FA, New PW, Iftikhar A. A report on disability and rehabilitation medicine in Pakistan: past, present, and
future directions. Arch Phys Med Rehabil 2011 Jan;92(1):161-166. [doi: 10.1016/j.apmr.2010.10.004] [Medline:
21187218]
6. Liu G, Liu X, Li H, Du Q, Wang F. Optical coherence tomographic analysis of retina in retinitis pigmentosa patients.
Ophthalmic Res 2016;56(3):111-122 [FREE Full text] [doi: 10.1159/000445063] [Medline: 27352292]
7. Maiman LA, Becker MH. The health belief model: origins and correlates in psychological theory. Health Educ
Monogr 1974;2(4):336-353. [doi: 10.1177/109019817400200404]
8. Bandura A. Health promotion from the perspective of social cognitive theory. Psychol Health 1998;13(4):623-649.
[doi: 10.1080/08870449808407422]
9. Kamal AK, Khoja A, Usmani B, Muqeet A, Zaidi F, Ahmed M, et al. Translating knowledge for action against stroke--
using 5-minute videos for stroke survivors and caregivers to improve post-stroke outcomes: study protocol for a
randomized controlled trial (Movies4Stroke). Trials 2016 Jan 27;17:52 [FREE Full text] [doi: 10.1186/s13063-016-
1175-x] [Medline: 26818913]
10. Bonita R, Beaglehole R. Recovery of motor function after stroke. Stroke 1988 Dec;19(12):1497-1500.
[doi: 10.1161/01.str.19.12.1497] [Medline: 3201508]
11. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of
handicap in stroke patients. Stroke 1988 May;19(5):604-607. [doi: 10.1161/01.str.19.5.604] [Medline: 3363593]
12. Kasner SE. Clinical interpretation and use of stroke scales. Lancet Neurol 2006;5(7):603-612. [doi:
10.1016/s1474-4422(06)70495-1]
13. Lyden P, Claesson L, Havstad S, Ashwood T, Lu M. Factor analysis of the National Institutes of Health Stroke Scale
in patients with large strokes. Arch Neurol 2004 Nov;61(11):1677-1680. [doi: 10.1001/archneur.61.11.1677]
[Medline: 15534178]
14. Schiemanck SK, Post MW, Witkamp TD, Kappelle LJ, Prevo AJ. Relationship between ischemic lesion volume and
functional status in the 2nd week after middle cerebral artery stroke. Neurorehabil Neural Repair 2005 Jun;19(2):133-
138. [doi: 10.1177/154596830501900207] [Medline: 15883357]
15. ATTEND Collaborative Group. Family-led rehabilitation after stroke in India (ATTEND): a randomised controlled
trial. Lancet 2017 Aug 5;390(10094):588-599. [doi: 10.1016/S0140-6736(17)31447-2] [Medline: 28666682]
16. Forster A, Dickerson J, Young J, Patel A, Kalra L, Nixon J, TRACS Trial Collaboration. A structured training
programme for caregivers of inpatients after stroke (TRACS): a cluster randomised controlled trial and cost-
effectiveness analysis. Lancet 2013 Dec 21;382(9910):2069-2076. [doi: 10.1016/S0140-6736(13)61603-7] [Medline:
24054816]
17. Markle-Reid M, Orridge C, Weir R, Browne G, Gafni A, Lewis M, et al. Interprofessional stroke rehabilitation for
stroke survivors using home care. Can J Neurol Sci 2011 Mar;38(2):317-334. [doi: 10.1017/s0317167100011537]
[Medline: 21320840]
18. Cameron JI, Naglie G, Green TL, Gignac MA, Bayley M, Huijbregts M, et al. A feasibility and pilot randomized
controlled trial of the 'Timing it Right Stroke Family Support Program'. Clin Rehabil 2015 Nov;29(11):1129-1140. [doi:
10.1177/0269215514564897] [Medline: 25552525]
19. Legg LA, Quinn TJ, Mahmood F, Weir CJ, Tierney J, Stott D, et al. Non-pharmacological interventions for caregivers
of stroke survivors. Cochrane Database Syst Rev 2011 Oct 5(10):CD008179. [doi: 10.1002/14651858.CD008179.pub2]
[Medline: 21975778]
20. Chumbler NR, Li X, Quigley P, Morey MC, Rose D, Griffiths P, et al. A randomized controlled trial on Stroke
telerehabilitation: The effects on falls self-efficacy and satisfaction with care. J Telemed Telecare 2015;21(3):139-
143. [doi: 10.1177/1357633x15571995]
21. Mackay-Lyons M, Thornton M, Ruggles T, Che M. Non-pharmacological interventions for preventing secondary
vascular events after stroke or transient ischemic attack. Cochrane Database Syst Rev 2013(3):CD008656. [doi:
10.1002/14651858.CD008656.pub2] [Medline: 23543566]
22. Rosenstock IM, Strecher VJ, Becker MH. Social learning theory and the Health Belief Model. Health Educ
Q 1988;15(2):175-183. [doi: 10.1177/109019818801500203] [Medline: 3378902]
23. Sharma M, Romas JA. Theoretical Foundations of Health Education and Health Promotion. Burlington, Massachusetts:
Jones & Bartlett Publishers; 2011.
24. Lundy YD. Abstract T P286: telephone support program promotes stroke survivor knowledge, self-care, and
satisfaction. Stroke 2015;46(Suppl 1):ATP286 [FREE Full text]
25. Kitson AL. The need for systems change: reflections on knowledge translation and organizational change. J Adv Nurs
2009 Jan;65(1):217-228. [doi: 10.1111/j.1365-2648.2008.04864.x] [Medline: 19032518]
26. Denny MC, Vahidy F, Vu KY, Sharrief AZ, Savitz SI. Video-based educational intervention associated with
improved stroke literacy, self-efficacy, and patient satisfaction. PLoS One 2017;12(3):e0171952 [FREE Full text]
[doi: 10.1371/journal.pone.0171952] [Medline: 28333925]
27. Molloy GJ, Messerli-Bürgy N, Hutton G, Wikman A, Perkins-Porras L, Steptoe A. Intentional and unintentional
non-adherence to medications following an acute coronary syndrome: a longitudinal study. J Psychosom Res 2014
May;76(5):430-432 [FREE Full text] [doi: 10.1016/j.jpsychores.2014.02.007] [Medline: 24745787]
28. Chow CK, Redfern J, Hillis GS, Thakkar J, Santo K, Hackett ML, et al. Effect of lifestyle-focused text messaging on
risk factor modification in patients with coronary heart disease: a randomized clinical trial. J Am Med Assoc
2015;314(12):1255-1263. [doi: 10.1001/jama.2015.10945] [Medline: 26393848]
29. Schmidt H, Barnhill A. Equity and noncommunicable disease reduction under the sustainable development goals.
PLoS Med 2015 Sep;12(9):e1001872 [FREE Full text] [doi: 10.1371/journal.pmed.1001872] [Medline: 26349050]
30. Dalal P, Bhattacharjee M, Vairale J, Bhat P. UN millennium development goals: Can we halt the stroke epidemic in
India? Ann Indian Acad Neurol 2007;10(3):130. [doi: 10.4103/0972-2327.34791]
31. Janus-Laszuk B, Mirowska-Guzel D, Sarzynska-Dlugosz I, Czlonkowska A. Effect of medical complications on
the after-stroke rehabilitation outcome. NeuroRehabil 2017;40(2):223-232. [doi: 10.3233/nre-161407]
32. Domka E, Myjkowska E, Kwolek A. [Incidence of neuromedical complications during rehabilitation after stroke].
Neurol Neurochir Pol 2005;39(4):300-309. [Medline: 16096935]
33. Hochbaum G, Rosenstock I. United States Public Health Service. Kegels S: Health belief model URL:
https://academia. edu/
34. Green EC, Murphy E. Wiley Online Library. 2014. Health Belief Model URL:
https://onlinelibrary.wiley.com/doi/abs/ 10.1002/9781118410868.wbehibs410 [accessed 2019-11-04]
35. Duncan PW, Lai SM, Keighley J. Defining post-stroke recovery: implications for design and interpretation of drug
trials. Neuropharmacol 2000;39(5):835-841. [doi: 10.1016/s0028-3908(00)00003-4]
36. Barzel A, Ketels G, Stark A, Tetzlaff B, Daubmann A, Wegscheider K, et al. Home-based constraint-induced movement
therapy for patients with upper limb dysfunction after stroke (HOMECIMT): a cluster-randomised, controlled trial.
Lancet Neurol 2015 Sep;14(9):893-902. [doi: 10.1016/S1474-4422(15)00147-7] [Medline: 26231624]
37. Winstein CJ, Stein J, Arena R, Bates B, Cherney LR, Cramer SC, American Heart Association Stroke Council‚ Council
on Cardiovascular and Stroke Nursing‚ Council on Clinical Cardiology‚Council on Quality of Care and Outcomes
Research. Guidelines for adult stroke rehabilitation and recovery: a guideline for healthcare professionals from the
American Heart Association/American Stroke Association. Stroke 2016 Jun;47(6):e98-169. [doi:
10.1161/STR.0000000000000098] [Medline: 27145936]
38. Beaglehole R, Bonita R, Horton R, Adams C, Alleyne G, Asaria P, Lancet NCD Action Group, NCD Alliance.
Priority actions for the non-communicable disease crisis. Lancet 2011 Apr 23;377(9775):1438-1447. [doi:
10.1016/S0140-6736(11)60393-0] [Medline: 21474174]
39. Alleyne G, Binagwaho A, Haines A, Jahan S, Nugent R, Rojhani A, Lancet NCD Action Group. Embedding
non-communicable diseases in the post-2015 development agenda. Lancet 2013 Feb 16;381(9866):566-574.
[doi: 10.1016/S0140-6736(12)61806-6] [Medline: 23410606]
40. Boutron I, Moher D, Altman DG, Schulz KF, Ravaud P, CONSORT Group. Extending the CONSORT statement to
randomized trials of nonpharmacologic treatment: explanation and elaboration. Ann Intern Med 2008 Feb 19;148(4):295-
309. [doi: 10.7326/0003-4819-148-4-200802190-00008] [Medline: 18283207]
41. Chan A, Tetzlaff J, Altman D, Laupacis A, Gøtzsche PC, Krleža-Jeri K, et al. SPIRIT 2013 statement: defining standard
protocol items for clinical trials. Ann Intern Med 2013 Feb 5;158(3):200-207 [FREE Full text] [doi:
10.7326/0003-4819-158-3-201302050-00583] [Medline: 23295957]
42. Eysenbach G, CONSORT-EHEALTH Group. CONSORT-EHEALTH: improving and standardizing evaluation reports of
Web-based and mobile health interventions. J Med Internet Res 2011 Dec 31;13(4):e126 [FREE Full text] [doi:
10.2196/jmir.1923] [Medline: 22209829]

Abbreviations
AKUH: Aga Khan University Hospital
ARR: absolute risk reduction
 BI: Barthel Index
BP: blood pressure
CTU: clinical trials unit
eHealth: electronic health
ERC: Ethical Review Committee
HbA1c: glycosylated hemoglobin A1c
IT: information technology
LDL: low-density lipoprotein
mHealth: mobile health
mRS: Modified Rankin Scale
NCD: noncommunicable disease
NIHSS: National Institutes of Health Stroke Scale
NNT: number needed to treat
OR: odds ratio
RR: risk ratio

Edited by G Eysenbach; submitted 05.09.18; peer-reviewed by B Davis, J Wang; comments to author 07.01.19; revised version
received 01.03.19; accepted 19.07.19; published 28.01.20
Please cite as:
Kamal A, Khoja A, Usmani B, Magsi S, Malani A, Peera Z, Sattar S, Ahmed Akram M, Shahnawaz S, Zulfiqar M, Muqeet A, Zaidi
F, Sayani S, Artani A, Azam I, Saleem S
Effect of 5-Minute Movies Shown via a Mobile Phone App on Risk Factors and Mortality After Stroke in a Low- to Middle-Income
Country: Randomized Controlled Trial for the Stroke Caregiver Dyad Education Intervention (Movies4Stroke)
JMIR Mhealth Uhealth 2020;8(1):e12113
URL: http://mhealth.jmir.org/2020/1/e12113/
doi: 10.2196/12113
PMID:

©Ayeesha Kamal, Adeel Khoja, Bushra Usmani, Shahvaiz Magsi, Aresha Malani, Zahra Peera, Saadia Sattar, Masood Ahmed
Akram, Sumaira Shahnawaz, Maryam Zulfiqar, Abdul Muqeet, Fabiha Zaidi, Saleem Sayani, Azmina Artani, Iqbal Azam,
Sarah Saleem. Originally published in JMIR mHealth and uHealth (http://mhealth.jmir.org), 28.01.2020. This is an open-
access article distributed under the terms of the Creative Commons Attribution License
(https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium,
provided the original work, first published in JMIR mHealth and uHealth, is properly cited. The complete bibliographic
information, a link to the original publication on http://mhealth.jmir.org/, as well as this copyright and license information must
be included.