Patient Education and Counseling 97 (2014) 16–22

Contents lists available at ScienceDirect

Patient Education and Counseling

journal homepage: www.elsevier.com/locate/pateducou

Review

Video-assisted patient education to modify behavior:

A systematic review
Manar Abu Abed a, Wolfgang Himmel b, Stefan Vormfelde a, Janka Koschack b,*
a
Department of Clinical Pharmacology, University Medical Center, Göttingen, Germany
b
Department of General Practice, University Medical Center, Göttingen, Germany

A R T I C L E I N F O A B S T R A C T

Article history: Objective: To evaluate the efficacy of video-assisted patient education to modify behavior.
Received 3 February 2014 Methods: Fourteen databases were searched for articles published between January 1980 and October
Received in revised form 17 June 2014 2013, written in English or German. Behavioral change as main outcome had to be assessed by direct
Accepted 25 June 2014
measurement, objective rating, or laboratory data.
Results: Ten of the 20 reviewed studies reported successful behavioral modification in the treatment
Keywords: group. We discerned three different formats to present the information: didactic presentation (objective
Videotape recording
information given as verbal instruction with or without figures), practice presentation (real people
Audio–visual aid
filmed while engaged in a specific practice), narrative presentation (real people filmed while enacting
Teaching material
Health behavior scenes). Seven of the ten studies reporting a behavioral change applied a practice presentation or
Patient education narrative presentation format.
Systematic review Conclusion: The effectiveness of video-assisted patient education is a matter of presentation format.
Videos that only provide spoken or graphically presented health information are inappropriate tools to
modify patient behavior. Videos showing real people doing something are more effective.
Practice implications: If researchers wish to improve a skill, a model patient enacting the behavior seems
to be the best-suited presentation format. If researchers aim to modify a more complex behavior a
narrative presentation format seems to be most promising.
ß 2014 Elsevier Ireland Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1. Database and search terms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2. Selection criteria and procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.3. Quality assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.4. Subgroup analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.5. Statistical analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.1. Article selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2. Quality of selected studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.1. General description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.2. Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3. Efficacy of video-assisted patient education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.4. Subgroup analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.4.1. Comparison of asthma/COPD and diabetes mellitus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.4.2. Presentation format of the information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

* Corresponding author at: Department of General Practice, University Medical Center Göttingen, Humboldtallee 38, 37073 Göttingen, Germany. Tel.: +49 551 3914225;
fax: +49 551 399530.
E-mail address: [email protected] (J. Koschack).

http://dx.doi.org/10.1016/j.pec.2014.06.015
0738-3991/ß 2014 Elsevier Ireland Ltd. All rights reserved.
 M. Abu Abed et al. / Patient Education and Counseling 97 (2014) 16–22 17

4. Discussion and conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

4.1. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.1.1. A lid for every pot: didactic, practice or narrative presentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.1.2. Importance of adequate outcome parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.1.3. The persuasive power of narrative information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.2. Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.3. Practice implication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

1. Introduction 2. Methods

Making informed choices and taking an active role in their We performed a systematic review of the impact of audio–
health care is more challenging for patients today than it used to visual material on modifying patient behavior. Audio–visual
be: first, the biomedical progress leads to more choices that need material included the use of a videotape, a CD-ROM/DVD, or an
to be made because a greater variety of diagnostic and interactive website.
therapeutic options are available. Second, the self-care of
patients who suffer from a chronic illness becomes important 2.1. Database and search terms
[1]. For both reasons, health literacy is essential [2]. Clinical
research has revealed that poor health literacy is consistently We searched for clinical studies in 14 medical and psychological
associated with poorer outcome, e.g. with more hospitalizations, databases accessed through the German Institute of Medical
greater use of emergency care, poorer medication adherence, Documentation and Information (DIMDI): Cochrane Central
poorer ability to interpret labels and health messages [3]. These Register of Controlled Trials, Cochrane Database of Systematic
developments and study results are a call for the enhancement of Review, Database of Abstract of Reviews of Effects, EMBASE,
health literacy – and patient education may be the way to attain EMBASE Alert, gms, Health Technology Assessment Database,
this goal. MEDIKAT, Medline, PsychInfo, PSYNDEX, SciSearch, Social Sci-
There are many strategies of how to educate patients. They Search, SOMED. We limited our search to articles published
can be broadly divided into three categories according to the between January 1980 and October 2013, written in English or
presentation mode: (1) verbal instruction, (2) written material, German. We defined four categories of search terms: (1)
and (3) multimedia-based tools, including audio–visual inter- ‘education’ OR ‘teaching’ AND (2) ‘patient’ AND (3) ‘video’ OR
ventions presented on a data carrier (videotape, CD-ROM, DVD) ‘video assisted’ OR ‘audio–visual’ AND (4) ‘randomised’ OR
or on the internet. The multimedia-based or ‘video-assisted’ ‘randomized’ OR ‘controlled’. The articles were checked electroni-
patient education is held to have some advantages compared cally for duplicates.
with written or verbally presented education: Videos can be 2.2. Selection criteria and procedure
designed as a takeaway tool that allows more independent
application, away from the hospital clinic, at the patient’s own The studies retrieved from the databases were selected for the
pace and in the presence of friends or relatives [4]. Audio–visual final review in several steps (Fig. 1). First, article titles were checked
material can be entertaining, the medium is familiar and can also according to the categories described above. Three researchers
be used by those who have limited literacy. Moreover, (MAA; JK; WH) tested whether they would include or exclude the
information stored on data carriers has the advantage of being same articles from reading the title. Abstract reading and article
repeatable [5]. reading were the next two steps of the process (MAA; JK). The
Patient education is typically applied for at least three selection criteria were the following:
purposes: (1) enhancing knowledge to make informed choices,
i.e. providing patient decision-aids, (2) helping to cope with (1) Topic. Only studies were included that analyzed the efficacy of
negative feelings that can be developed in the forefront of
patient education programs to improve the handling of health
diagnostic or therapeutic procedures, (3) improving health
problems or diseases, such as dietary restrictions, medication
behavior, e.g. in the case of self-care activities such as regular
intake, exercise programs, and use of devices. We excluded
medication intake, lifestyle changes, or home-based disease
trials of decision-aids or information material that was
monitoring. While video-assisted patient education can be and is
designed to improve informed consent as well as trials of
used in all three areas, research shows ambiguous results as to
videos aiming to reduce pre-operative anxiety or concerns.
whether it is really effective. Videos designed to reduce pre-
(2) Study population. We included only studies with adult patients
procedural anxiety and improve coping skills seem to be
who suffered from a health-related problem.
effective [6]. There are also promising results that videos are
(3) Medium. We selected studies that evaluated the implementa-
effective in enhancing knowledge, especially in assisting deci-
tion of an audio–visual educational component.
sion-making for treatment options and informed consent [7]. In
(4) Intervention and study design. To properly assess the effect of
contrast, evidence for the efficacy of videos designed to improve
the audio–visual component, the selected study needed to
health behavior remains anecdotic, and a systematic evaluation
satisfy two criteria: a control condition was implemented and
is still lacking.
the audio–visual component was applied separately from other
In this paper we report the results of a systematic review that
interventions, such as talking with nurse or doctor after
evaluates the efficacy of video-assisted patient education in
watching the video. We only included studies without
modifying patient behavior. The ‘active ingredients’ of well-
randomization when an experimental design had implemented
designed studies are identified – studies that do not only modify
a control condition.
health-related patient behavior, but also demonstrate this effect (5) Outcome measurement: Included studies assessed the main
with a high level of evidence.
outcome by direct measurement, objective rating, or laboratory
18 M. Abu Abed et al. / Patient Education and Counseling 97 (2014) 16–22

(2) Video design. Videos designed to modify complex behavior were

accorded a higher quality when they were based on a
theoretical background or when their design was carefully
considered (1 point). Videos designed to optimize single
maneuvers do not necessarily need a theoretical framework;
they were considered non-written instruction manuals and
were also accorded 1 point. Studies without any scientific
rationale and description of the video development received 0
points.
(3) Video application. Studies were considered more valuable when
the watching of the video was monitored (2 points). Studies
describing the application without control or monitoring were
judged as second-best (1 point). When information about how
often and where patients watched the video was missing, 0
points were given.
(4) Reported statistical analysis. Studies that examine the effect of
an intervention have to compare at least two groups at two or
more points of times. Moreover, they often regard multiple
outcomes, such as rating scores or questionnaires. Therefore,
we required that a study had at least an adjustment of the alpha
error for multiple testing to consider the statistical analysis as
sufficient (1 point). Statistical analyses without adjustment
received 0 points, while 2 points indicated a sophisticated
multifactorial design with random and fixed effects.

In summary, 8 points represent the top score, 1 point indicates

the lowest quality.

2.4. Subgroup analysis

Presentation format. In the course of the review process, a study

characteristic evolved that we did not define a priori: the
presentation format of the information. We identified three
presentation formats:

(1) Didactic presentation, i.e. objective information is given as

verbal instruction with or without figures, e.g. a doctor or
scientist presenting information and showing figures on a
blackboard.
Fig. 1. Flowchart of inclusion procedure.
(2) Practice presentation, i.e. real people are filmed while engaged
data. Studies were excluded when the main outcome was in a specific practice, e.g. a patient with asthma is filmed by
assessed by self-report or questionnaire data alone. using the inhaler, first in an incorrect, after instruction in a
correct manner.
During the last step of article selection, cited references were (3) Narrative presentation, i.e. real people are filmed while acting
cross-checked to identify studies not found in the electronic out scenes, e.g. while sitting in the staff canteen, a young
search. mechanic told his colleague about his feelings, expectations
and experiences about visiting a public STD clinic for gonorrhea
2.3. Quality assessment treatment.

At first, we had aimed to use an established protocol to assess In a subgroup analysis, we compared the outcome of trials that
the quality of the included articles, e.g. the WIDER recommenda- directly assessed behavior according to these three presentation
tions [8]. However, given the heterogeneous nature of the included formats.
studies as well as the specifics of the intervention, i.e. videos, we Condition-related analysis. In the course of the analysis, it
developed a rationale that allowed us (a) to describe the became apparent that the efficacy of video-assisted patient
intervention in a standardized manner, e.g. mode of the video education totally differed in two health conditions, namely
application and (b) to include established quality criteria, e.g. asthma/COPD and diabetes mellitus. This is important because
appropriateness of the biometric analysis. We assessed the quality these are very similar diseases, i.e. both are chronic conditions that
of the included articles on five dimensions. Scores were accorded require effective self-management regarding drug intake, self-
as follows: monitoring etc. Therefore, we performed a subgroup-analysis of
trials that addressed one of these two conditions.
(1) Study design. Randomized-controlled trials (RCTs) with an
elaborate control condition, i.e. not merely standard care, 2.5. Statistical analysis
received 3 points, followed by RCTs with a standard control
condition (2 points). Experimental designs defined as studies Regarding the small number of included studies and the
with a control condition but without reliable randomization nominal or ordinal level of the assessed variables, we applied non-
received 1 point. parametric tests: a Mann–Whitney-U-Test to analyze differences
 M. Abu Abed et al. / Patient Education and Counseling 97 (2014) 16–22 19

between studies that did or did not report a behavioral change, and was impossible to control whether it was used properly.
a Spearman rank correlation test to analyze relations between Studies monitoring the video use (n = 6) were mainly those that
variables. All tests were two-tailed with a = 0.05, unless otherwise included video-watching at the study site. One study with
declared. The statistical analyses of the review were explorative video-watching at home evaluated the intervention engage-
rather than inferential; thus, we did not test against a hypothesis, ment through computer-assisted log-file analysis [23].
so that power calculations were not necessary. (4) Reported statistical analysis. We rated the reported statistical
analyses as insufficient in 9 of the 20 articles, all with multiple
3. Results testing of within-group and between-group differences. Eight
studies with adequate statistics tested the effect of the video by
3.1. Article selection analysis of variance (ANOVA) for repeated measurements with
time  group-interaction. Three study groups discussed the
Fig. 1 shows the results of the selection process. After reading ANOVA for repeated measurement as inferior compared with
the titles of 1377 articles that were the result of the electronic general estimation equation models (GEE) for dealing with
database search, the abstracts of the remaining 248 articles were longitudinal data [21,23,27]. The GEE approach accounts for
read. We excluded another 197 articles; the most common reason missing data without list-wise deletion of data. However, in
for exclusion was ‘inadequate outcome measure’, especially in one study using ANOVA to test group differences at follow-up,
cases where knowledge or attitudes was the only outcome the problem of missing data was discussed and an intention-to-
measure. A total of 51 articles were read in detail. The main treat analysis with last observation carried forward was
reasons for exclusion on this level of the selection process were additionally conducted [14].
either ‘inadequate outcome measure’, i.e. the main outcome was
assessed by self-ratings or questionnaires, or ‘inadequate inter-
vention’, i.e. the effect of the video-assisted education could not be 3.3. Efficacy of video-assisted patient education
separated from other interventions. A total of 20 articles were
included in the final analysis. Of the 20 studies included in the systematic review, 13 reported
a difference between experimental/treatment condition versus
3.2. Quality of selected studies control condition (cp. Table 1). However, only 10 of these 20
studies showed a difference in the outcome of interest, i.e. an
3.2.1. General description outcome that directly assessed the addressed behavior or the
The 20 articles covered the period of time from 1983 to 2011. consequences of its modification by objective ratings or surrogate
While in the period 1983–2000 long intervals up to the point of parameters. There was no statistically significant difference in the
years between publications existed, from the year 2000 onward, overall score (5.8  1.1 versus 5.1  1.9; Mann–Whitney-U-Test:
there was at least one article per year. The articles addressed 12 p = 0.631) between the two groups, i.e. studies that did report a
different diseases or health-related problems. Five articles dealt with behavioral change compared with studies that did not report a
diabetes mellitus type 2 [9–13] and four with asthma/chronic change. We also compared the studies by the years elapsed since
obstructive pulmonary disease (COPD) [14–17]. Sexually transmit- publication. The mean of years elapsed since publication of the 10
ted diseases were addressed in two articles [18,19]. Nine conditions studies reporting a behavioral change was 10.4  10.0 and of the 10
were only addressed in one article, including ankle sprain [20], studies not reporting any behavioral change the mean was
anticoagulation [21], cancer [22], hypertension [23], medication 12.7  10.1. This difference was statistically not significant (Mann–
adherence (different drugs/diseases) [24], sleep apnea [25], and Whitney-U-Test: p = 0.353). To consider possible modifying effects of
surgery [26]. recent developments in design and statistical analysis, we correlated
The used videos were very different in length, content, and the years elapsed since publication and the study quality (sum score)
structure. Some were best described as interactive patient over all 20 studies. The correlation was nearly zero (Spearman rank
textbooks, structured by chapters that could be chosen. Others correlation: r = 0.090; p = 0.707).
were more like instruction videos to improve certain movements However, analyzing all variables presented in Table 1 in a more
known from other areas like sports. Some video resembled short qualitative way, we found some remarkable differences: authors of
film sequences, following a story board and showing real people studies with a positive study result based their video-design on a
acting and talking. scientific background or described the maneuvers that should be
optimized by the video more often and more detailed than did
3.2.2. Evaluation authors of studies with a negative result (9/10 versus 5/10). Half of
Table 1 shows the characteristics of the 20 studies in detail. the positive studies monitored the application of the video, but
none of the negative studies did.
(1) Study design. All studies were RCTs, but only 12 had tested
against an elaborated control condition, i.e. more than merely 3.4. Subgroup analyses
standard care.
(2) Video-design. We found no information about the development 3.4.1. Comparison of asthma/COPD and diabetes mellitus
of the video in six articles. In six articles, the researcher used In the 20 studies included, two conditions, diabetes mellitus
the video to optimize a single maneuver [14–17,22,27], such as and asthma/COPD, were most prominent (five and four studies,
the correct use of an inhaler in asthma therapy. Their videos respectively). Three of the four asthma/COPD studies dealt with
were judged as non-written instruction manuals that do not optimizing the inhaler technique [15–17]. These three studies used
have to be well-founded. In eight articles, we found detailed a video-design that was manual-based, i.e. characteristics of the
information about the development of the video, with a mix of correct inhaler technique were described and realized in the video
theoretical and empirical rationales [10,18–21,23,26,28]. with a model patient showing the correct use. They measured the
(3) Video application. Two articles provided no information about behavioral change directly by comparing the correct inhaler use at
the application of the video, more than half (12/20) described baseline and after education, and all of them showed improvement
the application, but did not monitor the video use – because the (objective ratings by an experienced medical professional not privy
video was intended to be watched at home. This meant that it to the treatment allocation).
 20
Table 1
Detailed evaluation of the articles included in the systematic review (n = 34) (in alphabetic order).

# First Publication Health issue Outcomes and Study Video design Video Statistical Overall Difference
author year their assessmenta design application analysis scoreb between
groupsc

1 Bassett SF [20] 2010 Ankle sprain Adherence to physiotherapy by attendance RCT+ Manual/well-founded Described Multifactorial 7 In other outcomes
rates ankle function by objective ratings
2 Doering S [26] 2001 Surgery Post-OP mobility by objective ratings RCT Manual/well-founded Monitored Insufficient 5 In main outcomes
3 Dyson PA [9] 2010 Diabetes Medication adherence by HbA1c level RCT Not specified Described Insufficient 3 No difference
physical activity by pedometer
4 Gerber BS [10] 2005 Diabetes Diabetes self-management skills by HbA1c RCT+ Manual/well-founded Monitored Multifactorial 8 No difference
level, BMI, blood pressure level and by
questionnaire data
5 Hagan LD [28] 1983 Psychiatric Therapy persistence by attendance rates RCT+ Manual/well-founded Monitored Insufficient 6 In main outcomes
disorders
6 Haines TP [27] 2009 High risk of falls Mobility after discharge by fall rates, RCT Manual/well-founded Described Multifactorial 6 No difference
objective ratings, self-reports

M. Abu Abed et al. / Patient Education and Counseling 97 (2014) 16–22

7 Houston TK [23] 2011 Hypertension Hypertension control by blood pressure RCT+ Manual/well-founded Monitored Multifactorial 8 In main outcomes
level
8 Huang JP [11] 2009 Diabetes Diabetes self-management skills by HbA1c RCT Not specified Described Insufficient 3 No difference
level, self-reports
9 Kinnane N [22] 2008 Cancer Self-management of chemotherapy side RCT Manual/well-founded Monitored Insufficient 5 In main outcomes
effects by objective rating of symptom
awareness
10 Mazor KM [21] 2006 Anti-coagulation Adherence to laboratory monitoring by RCT Manual/well-founded Described Multifactorial 6 In other outcomes
attendance rates knowledge and beliefs by
questionnaire data
11 McCulloch DK [12] 1983 Diabetes Glycemic control by HbA1c level RCT Not specified Described Insufficient 3 No difference
knowledge of and adherence to dietary
restrictions by questionnaire and self-
report
12 Mulrow C [13] 1986 Diabetes Sustained glycemic control by HbA1c level RCT+ Not specified Described Multifactorial 5 No difference
sustained weight control by weight
measurement
13 O’Donnell CR [18] 1997 STD Condom use by infection rates RCT Manual/well-founded Not specified Insufficient 3 In main outcomes
14 Opat AJ [14] 2000 Asthma/COPD Asthma control by peak expiratory flow and RCT+ Manual/well-founded Described Multifactorial 7 In other outcomes
by self-reports of inhaled reliever
medication intake, symptoms
15 Powell KM [24] 1995 Medication Medication intake by medication RCT Not specified Described Insufficient 3 No difference
adherence possession rates
16 Savage I [15] 2003 Asthma/COPD Quality of inhaler technique by objective RCT Manual/well-founded Monitored Insufficient 5 In main outcomes
ratings
17 Self TH [16] 1983 Asthma/COPD Quality of inhaler technique by objective RCT+ Manual/well-founded Described Multifactorial 7 In main outcomes
ratings
18 Solomon MZ [19] 1988 STD Adherence to follow-up examination by RCT Manual/well-founded Not specified Multifactorial 5 In main outcome
attendance rates informing sexual contacts
by attendance rates knowledge by
questionnaire
19 van der Palen J [17] 1997 Asthma/COPD Quality of inhaler technique by objective RCT+ Manual/well-founded Described Multifactorial 7 In main outcome
ratings
20 Wiese HJ [25] 2005 Sleep apnoea Adherence to CHAP by attendance rates RCT Not specified Described Multifactorial 5 In main outcome
knowledge by questionnaires

RCT randomized controlled trial; +with elaborated control condition

a
Outcome measures of primary interest are underlined.
b
Maximum score is 8 with study design [RCT = 2, RCT+ = 3], video design [not specified = 0, manual/well-founded = 1], video application [not specified = 0, described = 1, monitored = 2], statistical analysis [insufficient = 0,
adjusted = 1, multifactorial = 2].
c
Differences in all studies favored the treatment group.
 M. Abu Abed et al. / Patient Education and Counseling 97 (2014) 16–22 21

All five diabetes studies dealt with improving the glycemic and O’Donnell et al. [18] showed increased attendance rates in
control assessed by HbA1c levels [9–13]. Better HbA1c levels were patients with sexually transmitted diseases, Houston et al. [23]
interpreted as being the consequences of behavioral modifications revealed better blood pressure control in hypertensive patients. All
regarding lifestyle changes such as food intake and physical three studies used highly sophisticated videos with a screenplay
activity or better medication adherence. In four of the five studies, that presented actors or even real patients discussing their
the video in the treatment condition aimed at improving decisions, their problems and their coping strategies.
knowledge by giving dietary advice and information about However, the narrative presentation of educational information
physical activity [9,11–13]. None of the five studies showed a is possibly not always the best choice. Video-assisted patient
difference between treatment group and control group. education can also be effective when a model patient demonstrates
the best practice. This assumption is supported by Bandura’s
3.4.2. Presentation format of the information ‘Social Cognitive Theory’, which places observational learning at
In the course of the analysis, we became aware that mainly the center of behavioral modelling [30]. The findings of effective
three different presentation formats were used to relay the patient education in asthma/COPD that aims to optimize the
information. Videos that provided objective information by verbal inhaler technique are in line with this idea. This means that the
or graphical presentation were used in nine studies [9,11– practice presentation is a good choice to optimize a specific
13,18,20,22,24,28]. Five studies used a video that showed real technique or a single maneuver.
people engaged in a specific practice [14,15–17,27]. Real people
acting and talking in scenes were used in six studies 4.1.2. Importance of adequate outcome parameters
[10,19,21,23,25,26]. Seven of the ten studies reporting a behavioral A main strength of our review is the selective and clear focus on the
change seemed to favor the practice presentation or narrative impact of video-assisted patient education on health-related behavior,
presentation format [15–17,19,23,25,26], whereas only four of the which is an outcome that really matters in healthcare. While reviews
ten studies with a negative result used these formats [10,14,21,27]. have shown that videos can effectively enhance knowledge, especially
in assisting decision-making for treatment options [5] and in the
4. Discussion and conclusion context of informed consent [7] or in reducing pre-procedural anxiety
and improving coping [6], it was vital to investigate the efficacy of
In this systematic review, we found no clear evidence for the video-assisted patient education to improve health-related behavior.
efficacy of video-assisted patient education in modifying behavior. Many educational programs have been designed and conducted in
Of the 20 articles included in the review, 10 reported a difference this context, but systematic evidence was still lacking.
between experimental/treatment condition versus control condi- However, focusing on behavioral modification as the primary
tion in the expected direction. That is neither a clear demonstration outcome revealed a decisional conflict: while studies that only
of, nor a trend towards, efficacy. However, we became aware of two aimed to improve knowledge or alter attitudes and beliefs were
important cues and their interplay, which may be crucial for the easily identified and rigorously excluded, the direct measurement of
efficacy of video-assisted patient education: the format in which behavioral modification as the primary outcome was a very strict
the educational information is presented and the complexity of the criterion to include a study. An example of a direct assessment of the
addressed behavior. addressed behavior would be measuring attendance rates after
education about the importance of follow-up consultation in
4.1. Discussion sexually transmitted diseases. However, direct measurement of
behavior is not always possible; studies using meaningful and well-
4.1.1. A lid for every pot: didactic, practice or narrative presentation founded surrogate markers and objective ratings of the addressed
In the course of the analysis of the 20 studies, three different behavior were also included. An example of this is the improvement
presentation formats for the educational information proved to be of self-management skills in diabetes as the behavioral outcome,
important: (1) providing objective information verbally or with HbA1c levels as a surrogate marker.
graphically (‘didactic presentation’), (2) real people engaged in a
specific practice (‘practice presentation’), and (3) real people 4.1.3. The persuasive power of narrative information
enacting scenes (‘narrative presentation’). Seven of the ten studies Since most patient education programs aim to modify/optimize
that reported a significant behavioral change seemed to favor the complex behavior and not only specific techniques and maneuvers,
practice or narrative presentation, while only four of the ten we now discuss the narrative presentation format in more detail,
studies with a negative result used these formats. referring to the overview on narrative information in health
Consideration of common psychological concepts and models communication of Matthew Kreuter and colleagues [31]. Although
may elucidate why the presentation format has an impact on the they focused on communication about cancer, their definitions,
efficacy of video-assisted patient education. It is known that taxonomic framework and conclusions can be generalized for our
knowledge alters attitude, but there is no straightforward clear purpose. They defined narrative information as ‘‘a representation of
association between attitude and behavior, as has been shown connected events and characters that has an identifiable structure, is
previously [29]. Thus, enhancing the knowledge of patients by bounded in space and time, and contains implicit or explicit
verbally or graphically presented didactic information is definitely messages about the topic being addressed’’ (p. 222). They described
not the most effective way to modify patient behavior. The four different capabilities of narratives: (1) overcoming resistance,
negative results of the five studies that aimed to improve glycemic (2) facilitating information processing, (3) providing surrogate social
control in patients with diabetes mellitus seem to support this connections, and (4) addressing emotional and existential issues.
assumption: glycemic control is easy to assess by the HbA1c level, Thus, educational videos that focus on the modification of complex
but is the result of a very complex behavioral construct, i.e. behavior, such as adherence to lifestyle changes, should be designed
adherence to lifestyle changes and medication regime. Four of the in a narrative instead of in a didactic format.
five diabetes studies preferred a didactic presentation.
Obviously, knowledge improvement is not sufficient to modify 4.2. Conclusion
adherence to lifestyle changes and medication regime. The results
of the three other studies that aimed to modify complex behavior The main conclusion we draw is that the effectiveness of video-
indirectly confirmed this interpretation: Solomon and Dejon [19] assisted patient education is a matter of presentation format. Using
22 M. Abu Abed et al. / Patient Education and Counseling 97 (2014) 16–22

didactic information may increase health literacy, but is not the Workgroup for Intervention Development and Evaluation Research (WID-
ER) recommendations. Implement Sci 2013;8:52.
sufficient to modify health-related behavior. Videos that only [9] Dyson PA, Beatty S, Matthews DR. An assessment of lifestyle video education
provide spoken or graphically presented health information are for people newly diagnosed with type 2 diabetes. J Hum Nutr Dietet
inappropriate tools to modify patient behavior. Instead, videos 2010;23:353–9.
[10] Gerber BS, Brodsky IG, Lawless KA, Smolin LI, Arozullah AM, Smith EV, et al.
with a narrative format seem to be a powerful education tool. Implementation and evaluation of a low-literacy diabetes education computer
However, regarding the persuasive character and the problem of multimedia application. Diabetes Care 2005;28:1574–80.
subjectivity of personal experience against the objectivity of [11] Huang J, Chen H, Yeh M. A comparison of diabetes learning with and without
interactive multimedia to improve knowledge, control, and self-care among
epidemiological data, future research will have to consider the people with diabetes in Taiwan. Public Health Nurs 2009;26:317–28.
ethical challenges for patient education [32,33]. [12] McCulloch DK, Mitchell RD, Ambler J, Tattersal RB. Influence of imaginative
teaching of diet on compliance and metabolic control in insulin dependent
diabetes. Brit Med J 1983;287:1858–61.
4.3. Practice implication
[13] Mulrow C, Bailey S, Sönksen PH, Slavin B. Evaluation of an audiovisual
diabetes education program: negative results of a randomized trial of
In addition to the carefully chosen behavioral outcome patients with non-insulin-dependent diabetes mellitus. J Gen Intern Med
parameter, researchers need to consider the behavior they wish 1987;2:215–9.
[14] Opat AJ, Cohen MM, Bailey MJ, Abramson MJ. A clinical trial of the Buteyko
to modify. If they wish to improve a skill, a model patient enacting Breathing Technique in asthma as taught by a video. J Asthma 2000;37:
the behavior seems to be the best-suited presentation format. If 557–64.
they aim to modify a more complex behavior such as coping [15] Savage I, Goodyer L. Providing information on metered dose inhaler technique:
is multimedia as effective as print? Fam Pract 2003;20:552–7.
strategies or lifestyle changes, a narrative presentation format [16] Self TH, Brooks JB, Lieberman P, Ryan MR. The value of demonstration and role
seems to be most promising. of the pharmacist in teaching the correct use of pressurized bronchodilators.
Can Med Assoc J 1983;128:129–31.
[17] van der Palen J, Klein JJ, Kerkhoff AHM, van Herwaarden CLA, Seydel ER.
Authorship statement Evaluation of the long-term effectiveness of three instruction modes for
inhaling medicines. Patient Educ Couns 1997;32:S87.
All authors of this paper fulfill the criteria of authorship. In [18] O’Donnell CR, O’Donnell L, San Doval A, Duran R, Labes K. Reduction in STD
infections subsequent to an STD clinic visit. Using video-based patient
addition, we assert that there is no one else who fulfils the criteria education to supplement provider interactions. Sex Transm Dis 1998;25:
but has not been included as author. 161–8.
[19] Solomon MZ, Dejon W. The impact of a clinic-based educational videotape on
knowledge and treatment behavior of men with gonorrhea. Sex Transm Dis
Ethical approval
1988;15:127–32.
[20] Bassett SF, Prapavessis H. A test of an adherence-enhancing adjunct to
Ethical approval was not required. physiotherapy steeped in the protection motivation theory. Physiother Theory
Pract 2011;27:360–72.
[21] Mazor KM, Baril J, Dugan E, Spencer F, Burgwinkle P, Gurwitz JH. Patient
Funding education about anticoagulant medication. Is narrative evidence or statistical
evidence more effective? Patient Educ Couns 2007;69:145–57.
The study was not funded. [22] Kinnane N, Stuart E, Thompson L, Evans K, Schneider-Kolsky M. Evaluation of
the addition of video-based education for patients receiving standard pre-
chemotherapy education. Eur J Cancer Care 2008;17:328–39.
Conflict of interest statement [23] Houston TK, Allison JJ, Sussman M, Horn W, Holt CL, Trobaugh J, et al.
Culturally appropriate storytelling to improve blood pressure. A randomized
trial. Ann Intern Med 2011;154:77–84.
No financial disclosures. [24] Powell KM, Edgren B. Failure of educational videotapes to improve medication
compliance in a health maintenance organization. Amer J Health Syst Pharm
References 1995;52:2195–9.
[25] Wiese HJ, Boehtel C, Phillips B, Wilson JF, Peters J, Viggiano T. CPAP compli-
[1] Wilson PM, Mayor V. Long-term conditions. 2: supporting and enabling self- ance: video education may help! Sleep Med 2005;6:171–4.
care. Brit J Community Nurs 2006;11:6–10. [26] Doering S, Behensky H, Rumpold G, Schatz DS, Rössler S, Hofstötter B, et al.
[2] Schulz PJ, Nakamoto K. Health literacy and patient empowerment in health Video-Vorbereitung bei Hüftgelenksersatz-Operation verbessert die Mobilität
communication: the importance of separating conjoined twins. Patient Educ nach drei Monaten. Z Psychosom Med Psychother 2001;47:140–52.
Couns 2013;90:4–11. [27] Haines TP, Russell T, Brauer SG, Erwin S, Lane P, Urry S, et al. Effectiveness of a
[3] Berkman ND. Low health literacy and health outcomes: an updated systematic video-based exercise programme to reduce falls and improve health-related
review. Ann Intern Med 2011;155:97. quality of life among older adults discharged from hospital: a pilot randomized
[4] Gysels M, Higginson IJ. Interactive technologies and videotapes for patient controlled trial. Clin Rehabil 2009;23:973–85.
education in cancer care: systematic review and meta-analysis of randomised [28] Hagan LD, Beck NC, Kunce JT, Heisler GH. Facilitating psychiatric patient
trials. Support Care Cancer 2007;15:7–20. follow-up: a study of transfer attrition. J Clin Psychol 1983;39:494–9.
[5] Green MJ. Effect of a computer-based decision aid on knowledge, perceptions, [29] Fishbein M, Ajzen I. Belief, attitude, intention, and behavior: an introduction to
and intentions about genetic testing for breast cancer susceptibility: a ran- theory and research. Reading, MA: Addison-Wesley Pub. Co.; 1975.
domized controlled trial. J Amer Med Assoc 2004;292:442–52. [30] Bandura A. Self-efficacy: toward a unifying theory of behavioral change.
[6] Schofield P, Jefford M, Carey M, Thomson K, Evans M, Baravelli C, et al. Psychol Rev 1977;84:191–215.
Preparing patients for threatening medical treatments: effects of a chemo- [31] Kreuter MW, Green MC, Cappella JN, Slater MD, Wise ME, Storey D, et al.
therapy educational DVD on anxiety, unmet needs, and self-efficacy. Support Narrative communication in cancer prevention and control: a framework to
Care Cancer 2008;16:37–45. guide research and application. Ann Behav Med 2007;33:221–35.
[7] Ryan RE, Prictor MJ, McLaughlin KJ, Hill SJ. Audio–visual presentation of [32] Hinyard LJ, Kreuter MW. Using narrative communication as a tool for health
information for informed consent for participation in clinical trials. Cochrane behavior change: a conceptual, theoretical, and empirical overview. Health
Database Syst Rev 2008;CD003717. Educ Behav 2007;34:777–92.
[8] Albrecht L, Archibald M, Arseneau D, Scott SD. Development of a checklist to [33] Myers KR, Green MJ. Storytelling: a novel intervention for hypertension. Ann
assess the quality of reporting of knowledge translation interventions using Intern Med 2011;154:129–30.