Original Article

Healthc Inform Res. 2021 January;27(1):48-56.

https://doi.org/10.4258/hir.2021.27.1.48
pISSN 2093-3681 • eISSN 2093-369X

Effects of Self-Education on Patient Safety via

Smartphone Application for Self-Efficacy and
Safety Behaviors of Inpatients in Korea
Sumi Cho1, Eunjoo Lee2
1
Department of Nursing, Korea Nazarene University, Cheonan, Korea
2
College of Nursing, Research Institute of Nursing Science, Kyungpook National University, Daegu, Korea

Objectives: This study aimed to determine whether self-educational intervention on patient safety via a smartphone applica-
tion could improve the level of self-efficacy and safety behaviors of patients. In addition the effect of change in self-efficacy
on the improvement of safety behaviors after self-educational intervention was investigated. Methods: A one-group pre- and
post-test design and convenience sampling were implemented. Self-educational intervention via smartphone application was
provided to 94 participants in a tertiary university hospital in South Korea. The smartphone application included learning
contents on why the participation of patients is critical in preventing hospital-acquired infections and surgery-related adverse
events during hospitalization. Paired t-tests and hierarchical regression analysis were conducted to assess the effect of self-
educational intervention and self-efficacy on the improvement of safety behaviors of patients. Results: After the intervention,
the level of self-efficacy and safety behaviors significantly increased from 2.53 to 2.95 and from 2.00 to 2.62, respectively. In
the hierarchical regression analysis, the change in self-efficacy accounted for 35.4% of the variance in the improvement of
safety behaviors. Conclusions: The results of this study demonstrated that self-education on patient safety via a smartphone
application was an effective strategy to enhance patients’ self-efficacy and safety behaviors. This process could ultimately en-
hance patient safety by promoting patient involvement during hospitalization and preventing the occurrence of medical er-
rors.

Keywords: Patient Participation, Patient Safety, Self-Efficacy, Mobile Applications, Education

Submitted: June 24, 2020 I. Introduction

Revised: October 9, 2020
Accepted: December 22, 2020
Recently, interest in patient involvement in their own care
Corresponding Author has considerably increased to improve patient safety because
Eunjoo Lee patients could play a critical role in decreasing medical er-
College of Nursing, Research Institute of Nursing Science, Kyung-
rors [1]. Patients could participate in and contribute to the
pook National University, 680 Gukchaebosang-ro, Jung-gu, Daegu
41944, Korea. Tel: +82-53-420-4934, E-mail: [email protected] prevention of medical errors and adverse events in several
(https://orcid.org/0000-0002-6548-5593) stages of their care during hospitalization. Thus, many orga-
nizations for patient safety are leading campaigns on patient
This is an Open Access article distributed under the terms of the Creative Com-
mons Attribution Non-Commercial License (http://creativecommons.org/licenses/by- involvement, and researchers are developing interventions
nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduc-
tion in any medium, provided the original work is properly cited.
to educate patients so that they can play an active role in
improving their own safety. For example, the “Speak Up”
ⓒ 2021 The Korean Society of Medical Informatics
initiative of the US Joint Commission, the World Health
 Self-Education on Patient Safety

Organization’s patient safety campaign, and the “Ask Me 3” to address safety issues, especially surgery-related adverse
educational program of the National Patient Safety Founda- events, hospital-acquired infections, and medication errors,
tion are representative examples [2,3]. because these issues are the most prevalent adverse events in
Although there is high interest in patient involvement in South Korea [19]. Therefore, this study aimed to determine
the enhancement of patient safety, few previous studies have whether the smartphone application to address patient safety
attempted to explain how to increase patient involvement issues could improve patients’ self-efficacy and safety behav-
and what factors influence the safety behaviors of patients iors. In addition, the effect of self-efficacy on the improve-
[4–7]. Moreover, only limited studies have tried to explain ment of safety behaviors of patients was also assessed.
patient involvement in the field of patient safety using
health-related theories [8–12]. According to previous stud- II. Methods
ies, the lack of confidence of patients is the main barrier to
their taking action to prevent medical errors; if patients feel 1. Design and Participants
they are unable to contribute to their own safety, they will A one-group pre- and post-test design and convenience sam-
not participate in securing their own safety during hospital- pling were implemented in this study. Ninety-four patients
ization [5,7]. from four medical–surgical units of a tertiary university
Luszczynska and Gunson [10] found that perceived behav- hospital in a large metropolitan city in South Korea were in-
ioral control was significantly associated with patients’ safety cluded. The sample size was calculated for a small effect size
behaviors, such as asking staff to wash their hands before a of 0.3, power of 0.8, and type I error probability of 0.05 for
procedure. Schwappach and Wernli [11] emphasized that the a one-sample pre- and post t-test. Using the G*Power pro-
perceived behavioral control of patients is the strongest pre- gram, the required sample size was 82. Therefore, the sample
dictor of patient intention to engage in medical error preven- size used had adequate power to test the hypothesis of this
tion. Behavioral control, which indicates what behaviors will study. The patient eligibility criteria were the following: (1)
be initiated to what extent and length regardless of obstacles, >20 years of age, (2) ability to read and speak the Korean
is determined by the level of self-efficacy. Therefore, the language, and (3) the ability to use a smartphone. The exclu-
self-efficacy of patients could be a significant determinant sion criteria were the following: (1) hearing and/or vision
of patient involvement in promoting his/her own safety and problem and (2) schedule for emergency surgery.
preventing medical errors. Previous studies have reported
that self-efficacy influences behavioral control of avoidance 2. Instrument
of alcohol consumption [13], smoking cessation [14], and To measure self-efficacy and safety behaviors, the Seniors
the regular performance of physical activities, such as walk- Empowerment and Advocacy in Patient Safety (SEAPS)
ing, jogging, and joining exercise classes [15,16]. Moreover, survey was used after permission was obtained from the
individuals who have higher self-efficacy perform better author [9]. The SEAPS survey was developed using a multi-
self-management, such as undergoing regular screening for step process with rigorous psychometric analysis. To develop
diseases and maintaining a healthy lifestyle for chronic dis- SEAPS, interviews with patients who had experienced a
ease management [17,18]. medical error or preventable harm during their healthcare
However, evidence of self-efficacy regarding the prevention were initiated with a review of government- and industry-
of medical errors, such as the adoption of safety behaviors published recommendations as well as medical and lay lit-
by patients, is currently extremely limited, even though it erature. Community focus group interviews were also imple-
can help enhance patient involvement in medical error pre- mented. It was developed based on the health belief model
vention. Additionally, data regarding interventions to im- and included four subscales: outcome efficacy, attitudes, self-
prove patients’ self-efficacy using smartphone applications efficacy, and behaviors. Originally, the SEAPS survey was
for patient safety are also lacking. No previous study has developed with 40 items; these were reduced to 21 items
determined whether patient safety-related self-educational after a pretest and pilot testing, which confirmed that 21
interventions provided using smartphone applications could items were in good agreement with the initial 40 items. With
improve patients’ self-efficacy and safety behaviors, thereby good validity, the Cronbach’s alphas of the subscales, namely,
enhancing patient involvement in patient safety and prevent- outcome efficacy, attitudes, self-efficacy, and behaviors were
ing medical errors during hospitalization. 0.91, 0.74, 0.91, and 0.79, respectively, in the study conduct-
In this study, a smartphone application was developed ed by Elder et al. [9].

Vol. 27 • No. 1 • January 2021 www.e-hir.org 49

Sumi Cho and Eunjoo Lee

The SEAPS survey was adopted in this work for several and provided instructions on how to use the application.
reasons. First, it was developed based on the direct input of First, the participants were asked to select the “Yes” option
patients who had experienced medical errors and prevent- in the smartphone application if they wanted to participate
able harm during their healthcare. Second, this tool also in the study. Then, they were asked to complete the online
includes items regarding safety behaviors that patients can pretest questionnaire via the smartphone application, which
use as strategies to improve their safety and quality of care. included (1) demographic characteristics, (2) items on self-
Therefore, this has a good match with the purpose of this efficacy in relation to patient safety, and (3) items on safety
study. All items were measured using a 4-point Likert scale, behaviors. After using the educational smartphone appli-
with higher scores indicating higher levels of self-efficacy cation for 3 days, the participants were asked to complete
with regard to patient safety and higher frequency of safety the posttest questionnaire, which included the same items,
behaviors. except the questions on demographics. The participants did
Self-efficacy was measured using five questions on “how not receive any compensation for their participation.
confident patients are actually doing these five tasks,” and
safety behavior was measured using six questions on “how 5. Intervention
often patients are doing these six tasks.” The Cronbach’s al- The smartphone application used in this study included
phas of self-efficacy and safety behavior were 0.81 and 0.89, learning contents on the importance of patient involvement
respectively. in patient safety and what patients have to do to prevent ad-
The demographic variables measured were age, sex, educa- verse events related to hospital-acquired infections, surgery-
tion level, and number of previous hospitalizations. The edu- related adverse events, medication errors, falls, and so forth.
cation level was collapsed into four categories: elementary The developmental process of the smartphone application
school or lower, middle school, high school, and college or for patient safety education has been previously published
higher. When the participants answered “Yes” to the ques- [20].
tion on previous hospitalization, they were asked about the Patients who were willing to use the smartphone applica-
number of previous hospitalizations. tion and agreed to participate in the study by clicking the
“Yes” button in the smartphone application were asked to
3. Ethical Considerations complete the pretest questionnaire. The participants received
The entire content of the smartphone application and re- instructions from the research assistants on how to use the
search process and the participants’ information statement smartphone application to acquire information on patient
and informed consent were reviewed and approved by the safety issues. They used the application for 3 days during
Kyungpook National University Ethics Committee (No. their hospitalization. The educational content on patient
2014-0026). All specific identifiers of the participants were safety is shown in Table 1.
removed. Data were maintained anonymously and securely On the first day, the participants were encouraged to ac-
stored in locked files in password-protected computers. cess the introduction to patient safety, an explanation of the
importance of patient involvement in patient safety, and
4. Data Collection ten tips to improve patient safety. Media reports on medical
The data were collected from March to June 2015. After ap- errors substantiated with video clips and animations were
proval was obtained from the Institutional Review Board, the provided via the smartphone application to capture the par-
researcher contacted the hospital manager and visited the ticipants’ attention and increase their awareness of the risk of
hospital. The researcher explained the purpose of the study adverse events during hospitalization.
to the head nurses in the four medical–surgical units and On the second day, the participants were encouraged to
obtained permission from the head nurse for data collection review the contents on how patients could participate in
from patients admitted in their units. The research assistant and cooperate with clinicians in preventing adverse events.
met with the patients admitted to these units and explained The content included what patients should ask their health-
the purpose and process of this study to them. The partici- care providers during hospitalizations to enhance their own
pants were again informed of the specific purpose of this safety and what they should know about their condition.
study, procedures, and their right to withdraw at any time For example, the smartphone application provided ques-
during the research process. The research assistants installed tions that patients should ask their clinicians and what they
the smartphone application on the participants’ smartphones should know about their medications. Moreover, the partici-

50 www.e-hir.org https://doi.org/10.4258/hir.2021.27.1.48
 Self-Education on Patient Safety

Table 1. Self-education contents on patient safety of smartphone application

Time Self-education contents on patient safety of smartphone application

Pre-test Demographic characteristics
Self-efficacy in relation to patient safety
Safety behaviors
1st day Introduction of patient safety
Importance of patient involvement in patient safety
Ten tips to improve patient safety
Major adverse events in healthcare settings
Media reports on medical errors
2nd day Information on what patients should do and what clinicians are doing to prevent medical errors
What patients should ask their healthcare providers during hospitalizations
Customized information according to their condition
“Surgery”, “medication”, “mobility”, and “catheter care” for more detailed information
3rd day Review of the previous day’s learning contents
True/false quiz questions to reinforce learning
Post-test Self-efficacy in relation to patient safety
Safety behaviors

What to ask clinicians

Patient safety app What to know at discharge
What to do before surgery
Admission discharge
& surgery What to know about medication
What to know about medication list
Medication About antibiotics medication

Mobility Prevention of decubitus ulcer Figure 1. Captured image of cus-

Prevention of fall
Catheter care tomized information icons
Types of catheter-associated infections on the smartphone appli­
How to prevent catheter-associated infections
cation.

pants could select a customized information menu accord- analyze the differences between the baseline and posttest
ing to their condition with submenus: “surgery”, “medication”, scores. Hierarchical regression analysis was used to identify
“mobility”, and “catheter care” for more detailed information the factors influencing improvement in safety behaviors. For
(Figure 1). the assumptions of the multiple linear regression analysis,
On the third day, the participants reviewed the previous residual autocorrelation was tested with the Durbin–Watson
day’s learning content and took a true/false quiz to reinforce statistic, and no autocorrelation existed. A p-value <0.05
learning. The research assistants asked the participants if was considered to be statistically significant. Analyses were
they had reviewed the daily learning content of the smart- conducted using the SPSS version 20.0 (IBM, Armonk, NY,
phone application and encouraged them to review the con- USA) software for Windows.
tents. The amount of time each participant spent using the
smartphone application for self-education was not limited. III. Results
After using the self-education smartphone application for 3
days, the participants completed the posttest questionnaire 1. Participants
with the help of the research assistants. A total of 110 patients admitted to the medical–surgical
units expressed interest in participating in this study, of
6. Data Analysis whom 99 participants consented to participate by signing
The participants’ demographic data were analyzed in terms the written consent form and completing the questionnaire.
of percentages and frequencies. Paired t-tests were used to Two participants discontinued intervention, and three par-

Vol. 27 • No. 1 • January 2021 www.e-hir.org 51

Sumi Cho and Eunjoo Lee

ticipants could not be complete the post-test because they 42.6% had completed high school. A total of 58.5% of the
were discharged. Eventually, the data from 94 participants participants were female; 41.5% had never been hospitalized,
were analyzed. The average age of the participants was 45 ± whereas the rest had previously been admitted to the hospi-
12.8 years. Almost half of them (43.6%) had completed col- tal one or more times. These findings are shown in Table 2.
lege education or education higher than college level, and
2. D
 ifferences in Self-Efficacy Level and Safety
Behaviors Before and After Intervention
Table 2. Demographic characteristics of the participants (n = 94) The mean level of self-efficacy increased from 2.53 ± 0.49 to
Variable Value 2.95 ± 0.61 after self-education intervention on patient safety
Age (yr) 45.0 ± 12.8 using the smartphone application (t = −7.28, p < 0.001). As
shown in Table 3, the scores for all subscales in self-efficacy
<40 32 (34.0)
showed improvement, and there were statistically significant
40–59 22 (23.4)
differences between the pre- and posttest scores.
60–69 29 (30.9)
Moreover, the result showed that the mean score of safety
≥70 11 (11.7)
behaviors also significantly increased from 2.00 ± 0.67 to
Sex 2.62 ± 0.76, and there were significant improvements in all
Female 55 (58.5) subscales of safety behaviors after self-education on patient
Male 39 (41.5) safety via the smartphone application (t = −8.62, p < 0.001)
Education (Table 4).
Elementary 2 (2.1)
Middle 6 (6.4) 3. Factors Influencing Improvement in Safety Behaviors
High 40 (42.6) Demographic characteristics, including age, sex, education
≥College 41 (43.6) level, and the number of hospitalizations, as well as change
in self-efficacy score, were evaluated using hierarchical
No answer 5 (6.3)
regression analysis to investigate the influence on improve-
Number of previous hospitalizations
ment in safety behaviors (Table 5). The variables’ categories
0 39 (41.5)
of sex and education level were converted into dummy vari-
1 8 (8.5)
ables. The Durbin–Watson statistic was equal to 1.71, and it
2 30 (31.9) indicated that there was no autocorrelation problem between
≥3 13 (13.8) the residual amounts. The variance inflation factors for inde-
No answer 4 (4.3) pendent variables were calculated for collinearity diagnosis,
Values are presented as mean ± standard deviation or number (%). and they were between 1.01 and 1.50, which indicated that

Table 3. Differences in self-efficacy levels before and after intervention

Questions Before After t p-value

1. How confident are you that you could teach yourself about your own 2.40 ± 0.78 2.80 ± 0.82 −4.58 <0.001
health problems and medications?
2. How confident are you that you could call the doctor’s office if you 2.68 ± 0.82 3.02 ± 0.80 −3.23 0.002
have not received the results of laboratory tests or X-rays?
3. How confident are you that you could obtain a second opinion from 2.50 ± 0.85 2.94 ± 0.89 −4.97 <0.001
another doctor if you think it is needed?
4. How confident are you that you could provide your doctors a com- 2.63 ± 0.77 3.04 ± 0.73 −4.67 <0.001
plete and thorough summary of your health problems?
5. How confident are you that you could ask a friend or family member 2.45 ± 0.81 2.94 ± 0.84 −5.06 <0.001
to come with you to doctor visits?
Mean of the five questions 2.53 ± 0.49 2.95 ± 0.61 −7.28 <0.001
Values are presented as mean ± standard deviation.

52 www.e-hir.org https://doi.org/10.4258/hir.2021.27.1.48
 Self-Education on Patient Safety

Table 4. Differences in safety behaviors before and after intervention

Questions Before After t p-value

1. How often do you teach yourself about your own health problems and medicines? 2.19 ± 0.88 2.72 ± 0.80 −5.58 <0.001
2. How often do you ask your doctors questions about your health problems, labo- 2.27 ± 0.83 2.77 ± 0.89 −5.60 <0.001
ratory tests, and medications?
3. How often do you keep an updated list of all your medicines, including those 1.65 ± 0.97 2.39 ± 1.03 −6.83 <0.001
from the drug store and health food store?
4. How often do you call the doctor’s office if you have not received the results of 1.86 ± 0.91 2.59 ± 0.97 −8.27 <0.001
laboratory tests or X-rays?
5. How often do you obtain a second opinion from another doctor when needed? 2.11 ± 1.00 2.60 ± 0.99 −4.78 <0.001
6. How often do you ask a friend or family member to come with you to doctors’ visits? 1.97 ± 0.93 2.65 ± 0.99 −6.17 <0.001
Mean of the six questions 2.00 ± 0.67 2.62 ± 0.76 −8.62 <0.001
Values are presented as mean ± standard deviation.

Table 5. Factors influencing improvement in safety behaviors

Unstandardized Stand coefficient

Model Variable Stand error t (p-value)
coefficient B beta
Model 1 Constant 0.76 0.31 - 2.43 (0.017)
Age (yr) 0.00 0.01 0.02 0.14 (0.886)
Sex (male = 1) −0.73 0.17 −0.05 −0.44 (0.662)
Education (college or higher = 1)
Elementary or none −0.10 0.58 −0.02 −0.16 (0.874)
Middle school 0.11 0.34 0.04 0.31 (0.757)
High school 0.21 0.18 −0.15 −1.15 (0.253)
Number of previous hospitalizations −0.06 0.07 −0.10 −0.90 (0.371)
2 2
F = 0.44, p = 0.85, R = 0.03, adjusted R = −0.04
Model 2 Constant 0.50 0.25 - 1.90 (0.061)
Age (yr) −0.00 0.01 −0.07 −0.64 (0.523)
Sex (male = 1) −0.12 0.13 −0.08 −0.90 (0.373)
Education (college or higher = 1)
Elementary or none −0.05 0.45 −0.01 −0.10 (0.922)
Middle school 0.02 0.27 0.01 0.08 (0.938)
High school 0.05 0.15 0.04 0.36 (0.723)
Number of previous hospitalizations −0.00 0.06 −0.01 −0.05 (0.960)
Change in self-efficacy 0.83 0.12 0.65 7.15 (<0.001)
2 2
F = 7.90, p < 0.001, R = 0.41, adjusted R = 0.35

multicollinearity was negligible. The first model including adjusted R2 = 0.354) of the variance in the improvement of
demographic variables, age, sex, education level, and number safety behaviors.
of previous hospitalizations, was not significant. When the
change in self-efficacy score between the baseline and post- IV. Discussion
test scores was additionally included, the result showed that
the prediction model was significant (F = 7.929, p < 0.001). This study assessed the effects of self-education on patient
The change in self-efficacy accounted for 35.4% (R2 = 0.406, safety issues via a smartphone application on the improve-

Vol. 27 • No. 1 • January 2021 www.e-hir.org 53

Sumi Cho and Eunjoo Lee

ment of patients’ self-efficacy and safety behaviors and in- individuals.

vestigated the factors influencing the improvement of safety In this study, the predictive power of the change in self-ef-
behaviors. The findings of this study indicate that patients’ ficacy in relation to the improvement of safety behaviors was
self-efficacy and safety behaviors could be enhanced by self- 35.4%. Previous studies also reported that self-efficacy could
education on patient safety using a smartphone application. influence safety behaviors and resulted in increased patient
This study also revealed that change in self-efficacy was the involvement and prevention of medical errors during hospi-
most important attribute of safety behavior improvement. talization [8,23]. However, these studies did not examine the
Schwappach et al. [21] provided advice on the admission effect of education on safety efficacy and safety behavior im-
of patients and asked them to follow the recommendations. provement. This study controlled the baseline of self-efficacy
Additionally, healthcare workers and clinical staff were asked and examined the variables associated with changes in self-
to participate in lessons and information meetings to pre- efficacy from baseline to post-intervention. The important
pare them for patient’s questions on safety issues. However, finding of this study is that even patients who have low self-
in the study conducted by Schwappach et al. [21], the perfor- efficacy on patient safety could improve their patient safety
mance of safety behaviors of patients was not improved after behaviors through self-education using their smartphones by
the safety advisory interventions by clinicians. This study enhancing their self-efficacy. Therefore, effective educational
provided real media reports on medical errors and major ad- interventions to enhance the self-efficacy of patients is criti-
verse events occurring during hospitalization, which might cally needed to encourage patient involvement to prevent
be used as triggers to accept recommended safety behaviors medical errors during hospitalization.
by the participants. Real stories of medical errors and ad- To increase awareness of patient safety during hospitaliza-
verse events could gain more attention of patients on safety tion, many patient education strategies are implemented us-
behaviors to prevent medical errors. ing information leaflets and posters. However, according to a
The health belief model posits that a “cue to action” is nec- previous study, patients often fail to acquire new information
essary to prompt engagement in health behaviors [22]. Par- on patient safety, and between 40% and 80% of the informa-
ticipants in this study were presented medical errors from tion is immediately forgotten because the primary concern
media reports via the smartphone application, which might of patients is their illness, not safety-related issues [24].
function as external cues to safety behavior and could stimu- Thus, development of different strategies, such as the use of
late participants to take action and influenced their self- a smartphone application, is needed to prevent patients from
efficacy [8]. Moreover, providing the educational content, getting lost in the myriad of information made available by
including what to do and what to know to prevent adverse healthcare providers.
events, which might occur in patients, could improve par- Smartphones are easily accessible and can provide the re-
ticipants’ self-efficacy. quired information whenever and wherever necessary dur-
In this study, the mean scores of both self-efficacy and safe- ing hospitalization [25]. As patients’ awareness of potential
ty behaviors showed significant increases from 2.53 to 2.95 risks increases, patients could be more involved and try to
and from 2.00 to 2.62, respectively. Therefore, self-education engage in safety behaviors to prevent potential medical er-
on patient safety issues via the smartphone application could rors. Therefore, nurses and healthcare professionals need to
be an effective and efficient strategy for enhancing the self- provide appropriate information to patients on safety issues
efficacy and safety behaviors of patients. Self-efficacy on pa- during hospitalization and the roles every patient can play in
tient safety was defined as an individual’s perception of his/ preventing medical errors [26].
her competence to successfully improve patient safety. Schwappach and Wernli [11] analyzed patients receiving
In this study, the level of self-efficacy was comparatively chemotherapy in a large regional hospital for their attitudes,
higher than that reported in the study by Elder et al. [9], who norms, behavioral control, and intentions to engage in safety
had adopted the same instrument. In their study, the mean behavior. In their study, the mean scores of instrumental
self-efficacy score was 1.6. This discrepancy may have been and experimental attitudes were 5.49 and 4.11, respectively,
caused by differences in participants’ ages and the settings and the mean scores of behavioral control and intention to
adopted for data collection. The mean age in the present prevent medical errors were 5.24 and 5.36, respectively, as
study was 45 years, and the study participants were hospital- measured on a 7-point Likert scale, and the mean age of par-
ized, whereas the mean age in the study by Elder et al. [9] ticipants was 61 years. These scores were higher than those
was 70 years, and participants were community-dwelling observed in this study. This may be attributed to the fact

54 www.e-hir.org https://doi.org/10.4258/hir.2021.27.1.48
 Self-Education on Patient Safety

that chemotherapy requires increased attention of patients, important predictor for the promotion of health behaviors,
and patients should be aware of the potential risks induced the evidence of self-efficacy with regard to the implementa-
by chemotherapy because its outcome could occasionally be tion of safety behaviors to prevent medical errors by patients
life-threatening. is extremely limited. Therefore, the importance of patients’
Previous studies have shown that younger patients are more self-efficacy should be considered by both healthcare profes-
willing to be involved in safety issues than older patients [27], sionals and nurses, and diverse strategies to enhance the self-
and patients who have higher education levels are more ac- efficacy of patients should be developed in future studies.
tive than those with lower education levels [8,27]. However,
in this study, there were no significant associations between Conflict of Interest
the improvement in safety behaviors and participants’ so-
ciodemographic factors, such as sex, education level, age, No potential conflict of interest relevant to this article was
and previous hospitalization. This can be attributed to the reported.
high education level of the participants in this study; specifi-
cally, almost half of the patients (43.6%) had education levels Acknowledgments
of college or higher and comparatively young age. Therefore,
using a more controlled study design, further investigation is This work was supported by the National Research Founda-
needed to understand why sociodemographic factors, such tion of Korea of the Ministry of Education of the Republic of
as sex, educational level, and age, were not related to safety Korea (No. NRF-2015S1A5A2A01009760).
behaviors in this study.
This study had several limitations. First, convenience ORCID
sampling was used to recruit participants from only one
academically affiliated tertiary hospital in South Korea, lim- Sumi Cho (https://orcid.org/0000-0001-6622-243X)
iting the generalizability of the results beyond this setting. Eunjoo Lee (https://orcid.org/0000-0002-6548-5593)
Thus, further research is required to confirm the findings of
this study. Second, the instrument used in the study to test References
patients’ self-efficacy in relation to patient safety and safety
behaviors was developed for senior citizens in a Western 1. Vincent CA, Coulter A. Patient safety: what about the
country; thus, this may limit the validity of the study. Third, patient? BMJ Qual Saf 2002;11(1):76-80.
this study only examined self-efficacy; it did not include oth- 2. Entwistle VA, Mello MM, Brennan TA. Advising pa-
er possible factors, such as perceived susceptibility, perceived tients about patient safety: current initiatives risk shift-
barriers, and cues to safety behaviors. In further studies, ing responsibility. Jt Comm J Qual Patient Saf 2005;
these variables need to be included to identify the specific 31(9):483-94.
impact of self-efficacy and other factors and effects of inter- 3. Longtin Y, Sax H, Leape LL, Sheridan SE, Donaldson L,
action among these variables. Finally, this study only mea- Pittet D. Patient participation: current knowledge and
sured the self-efficacy and safety behaviors of participants applicability to patient safety. Mayo Clin Proc 2010;
after 3 consecutive days; therefore, it is strongly recommend- 85(1):53-62.
ed to examine the changes in safety behaviors of participants 4. Buetow S, Davis R, Callaghan K, Dovey S. What attri-
after 3 or 6 months because actual changes in behaviors need butes of patients affect their involvement in safety? A
more time. key opinion leaders’ perspective. BMJ Open 2013;3(8):
In conclusion, the results of this study showed that self- e003104.
educational intervention via a smartphone application is an 5. Davis RE, Jacklin R, Sevdalis N, Vincent CA. Patient
effective strategy to enhance patients’ self-efficacy and safety involvement in patient safety: what factors influence
behaviors. This process could ultimately enhance patient patient participation and engagement? Health Expect
safety by promoting patient involvement during hospitaliza- 2007;10(3):259-67.
tion and preventing the occurrence of medical errors. The 6. Doherty C, Stavropoulou C. Patients’ willingness and
current international policy suggests that increased patient ability to participate actively in the reduction of clinical
involvement in safety issues could help prevent medical er- errors: a systematic literature review. Soc Sci Med 2012;
rors. Even though self-efficacy has been reported to be an 75(2):257-63.

Vol. 27 • No. 1 • January 2021 www.e-hir.org 55

Sumi Cho and Eunjoo Lee

7. Hall J, Peat M, Birks Y, Golder S, Entwistle V, Gilbody S, 17. Shin DS, Kim CJ, Choi YJ. Effects of an empowerment
et al. Effectiveness of interventions designed to promote program for self-management among rural older adults
patient involvement to enhance safety: a systematic re- with hypertension in South Korea. Aust J Rural Health
view. Qual Saf Health Care 2010;19(5):e10. 2016;24(3):213-9.
8. Bishop AC, Baker GR, Boyle TA, MacKinnon NJ. Using 18. Wichit N, Mnatzaganian G, Courtney M, Schulz P,
the Health Belief Model to explain patient involvement Johnson M. Randomized controlled trial of a family-
in patient safety. Health Expect 2015;18(6):3019-33. oriented self-management program to improve self-
9. Elder NC, Regan SL, Pallerla H, Levin L, Post D, Cegela efficacy, glycemic control and quality of life among Thai
DJ. Development of an instrument to measure seniors’ individuals with type 2 diabetes. Diabetes Res Clin Pract
patient safety health beliefs: the Seniors Empowerment 2017;123:37-48.
and Advocacy in Patient Safety (SEAPS) survey. Patient 19. Kim NY. Analysis of patient safety incident in Korea. J
Educ Couns 2007;69(1-3):100-7. Korean Acad Nurs Adm 2016;26(2):151-9.
10. Luszczynska A, Gunson KS. Predictors of asking medi- 20. Cho S, Lee E. Effects of the smartphone application “Safe
cal personnel about handwashing: the moderating role Patients” on knowledge of patient safety issues among
of patients’ age and MRSA infection status. Patient Educ surgical patients. Comput Inform Nurs 2017;35(12):639-
Couns 2007;68(1):79-85. 46.
11. Schwappach DL, Wernli M. Predictors of chemotherapy 21. Schwappach DL, Frank O, Buschmann U, Babst R. Ef-
patients’ intentions to engage in medical error preven- fects of an educational patient safety campaign on pa-
tion. Oncologist 2010;15(8):903-12. tients’ safety behaviours and adverse events. J Eval Clin
12. Schwappach DL. Engaging patients as vigilant part- Pract 2013;19(2):285-91.
ners in safety: a systematic review. Med Care Res Rev 22. Carpenter CJ. A meta-analysis of the effectiveness of
2010;67(2):119-48. health belief model variables in predicting behavior.
13. Oei TP, Burrow T. Alcohol expectancy and drinking Health Commun 2010;25(8):661-9.
refusal self-efficacy: a test of specificity theory. Addict 23. Hibbard JH, Peters E, Slovic P, Tusler M. Can patients be
Behav 2000;25(4):499-507. part of the solution? Views on their role in preventing
14. Baldwin AS, Rothman AJ, Hertel AW, Linde JA, Jeffery medical errors. Med Care Res Rev 2005;62(5):601-16.
RW, Finch EA, et al. Specifying the determinants of 24. Shankar J. Patients’ memory for medical information. J
the initiation and maintenance of behavior change: an R Soc Med 2003;96(10):520.
examination of self-efficacy, satisfaction, and smoking 25. Dennison L, Morrison L, Conway G, Yardley L. Oppor-
cessation. Health Psychol 2006;25(5):626-34. tunities and challenges for smartphone applications in
15. Shafieian M, Kazemi A. A randomized trial to promote supporting health behavior change: qualitative study. J
physical activity during pregnancy based on health be- Med Internet Res 2013;15(4):e86.
lief model. J Educ Health Promot 2017;6:40. 26. Vaismoradi M, Jordan S, Kangasniemi M. Patient par-
16. Shao C, Wang J, Liu J, Tian F, Li H. Effect of a Health ticipation in patient safety and nursing input: a system-
Belief Model-based education program on patients’ be- atic review. J Clin Nurs 2015;24(5-6):627-39.
lief, physical activity, and serum uric acid: a randomized 27. Arora NK, McHorney CA. Patient preferences for medi-
controlled trial. Patient Prefer Adherence 2018;12:1239- cal decision making: who really wants to participate?
45. Med Care 2000;38(3):335-41.

56 www.e-hir.org https://doi.org/10.4258/hir.2021.27.1.48