Simmons 

et al. BMC Medical Education (2022) 22:597

https://doi.org/10.1186/s12909-022-03653-8

RESEARCH Open Access

Testing an educational intervention

to improve health care providers’ preparedness
to care for victims of elder abuse: a mixed
method pilot study
Johanna Simmons1*, Atbin Motamedi2, Mikael Ludvigsson1,3 and Katarina Swahnberg4 

Aabstract 
Background:  Elder abuse is prevalent and associated with ill-health. However, health care providers often lack
education about elder abuse and older patients’ victimization often remains unknown to them. In this pilot study we
performed initial testing of an educational model aiming at improving health care providers’ preparedness to care
for older adults subjected to abuse, or more specifically their self-reported propensity to ask older patients questions
about abuse and perceived ability to manage the response.
Methods:  The educational model consisted of a full training day about elder abuse, including theory, group discus-
sions and forum theatre. Forum theatre is an interactive form of drama in which participants are not only observers,
but rather spect-actors, urged to participate in the scene. They are thereby given the opportunity to discuss and
practise difficult health care encounters. Medical interns (intervention group n = 16, control group n = 14) in Sweden
participated in the study and a mixed method convergent parallel design was used. Quantitative data was collected
at baseline and 6 months post-intervention using a questionnaire (the REAGERA-P). Qualitative interviews were
conducted with four of the participants in the intervention group and data was analysed using qualitative content
analysis.
Results:  The reported frequency of asking older patients questions about abuse increased in the intervention group
(p = 0.047), but not the control group (p = 0.38) post-intervention. Potential mediators for the improvement were an
increased awareness of elder abuse and higher self-efficacy for asking questions about elder abuse. Participants also
reported a higher perceived ability to manage cases of elder abuse, even though uncertainties concerning how to
provide the best possible care remained. The qualitative interviews indicated that learning from each other in group
discussions and forum theatre likely was an important contributor to the positive results.
Conclusion:  This pilot test indicated that the educational model may be effective in improving health care providers’
preparedness to care for older adults subjected to abuse. However, uncertainties about how to handle elder abuse
cases remained post-intervention. In a future full-scale test of the model more focus needs to be put on how to man-
age cases of elder abuse.

*Correspondence: johanna.simmons@liu.se
1
Department of Acute Internal Medicine and Geriatrics in Linköping,
and Department of Health, Medicine and Caring Sciences, Linköping
University, Linköping, Sweden
Full list of author information is available at the end of the article

© The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​
mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Simmons et al. BMC Medical Education (2022) 22:597 Page 2 of 16

Keywords:  Forum theatre, Forum play, Intimate partner violence, Medical education

Introduction facilitating care providers in their handling of elder abuse

The prevalence of elder abuse in community settings has cases, e.g. creating clear protocols on how to report and
been reported at 16% worldwide [1]. Elder abuse includes manage cases [9, 14, 19, 22]. In this study we consider
physical, emotional, sexual and financial abuse, as well barriers and facilitators for managing cases of elder abuse
as neglect, and it occurs at the hands of both profession- as part of the same continuum, e.g. lack of clear protocols
als and family members, e.g. adult children and intimate for managing cases would be considered a barrier while
partners. Elder abuse has been associated with psycho- the existence of such a protocol would be considered a
logical ill-health, disability, increased hospitalization facilitator.
and emergency department use, as well as admission to When providing education about elder abuse, it has
assisted living facilities [2–5]. Despite the negative health been found to be important to pay attention to local-
consequences, victims are often hesitant to ask for help ized needs and provide contact information for relevant
and the majority of elder abuse cases go unreported [6, local services for victims [12]. Also, the use of interac-
7]. It has therefore repeatedly been pointed out that the tive teaching techniques has been recommended [12,
health care system plays an important role in detecting 23, 24]. Using patient cases and hands-on active learning
and reporting cases of elder abuse [7, 8]. However, health with real or standardized patients has been well received
care providers are often unaware that their patients are [12, 24]. One type of interactive training is forum thea-
suffering from elder abuse and are often unsure how to tre, a form of interactive theatre developed by Augusto
manage cases. We previously reported that only half of Boal [25]. Forum theatre, and a version of it called forum
personnel at an acute internal medicine and geriatrics play, has previously been used in health care settings,
clinic in Sweden had ever talked about abuse with an both with staffs and students, e.g. to counteract abuse in
older patient and half of respondents were rather or very health care and to practice communication skills [26–29].
concerned about not being able to give victims a proper In conclusion, many health care providers report lack-
follow-up [9, 10]. A similar lack of awareness and knowl- ing education about elder abuse [9, 15] and only a few
edge about elder abuse has been found among health studies have investigated educational interventions about
care providers in other studies, both in Sweden and inter- elder abuse directed at health care providers [30]. A
nationally [11–13]. recent review about educational interventions for elder
One factor found to be associated with improved abuse in primary care suggests that education needs
recognition and management of elder abuse is hav- to focus solely on elder abuse and needs to be compre-
ing received education about elder abuse, which is also hensive and concise to realistically allow health care
often sought after by care providers [8, 9, 14, 15]. Edu- providers to attend sessions. Also, using multiple teach-
cation needs to address known barriers and facilitators ing methods is recommended, including interactive ele-
for asking about elder abuse and managing the response. ments, e.g. small-group discussions and role-play to
Barriers have been reported on a personal level, as well practice communication skills [30]. In this pilot study we
as on an organizational and system level [16]. On a per- test the effectiveness of a comprehensive one-day course
sonal level, health care providers are often unsure of about elder abuse, combining theory, group discussions
what constitutes abuse and when it should be reported and forum theatre. The latter has been suggested to be
[14, 17–19]. Many providers report feeling uneasy when an innovative educational method, stimulating reflection
addressing the issue, as well as lacking confidence in their and learning within health care [26]. Most previous edu-
ability to manage cases. Also, fear of retaliation from the cational interventions  concerning elder abuse in health
perpetrator and concern about negative reactions from care have used outcome measures that have not been
the patients are common among care providers, and validated, which is a threat to validity [30]. In this study
some express concerns about difficulties in ensuring pri- we used a mixed method approach, combining qualita-
vacy when asking about abuse [14, 16, 19–21]. Time con- tive interviews and quantitative data collected with a vali-
straints and unclarity about who has the responsibility to dated questionnaire [10].
care for victimized patients are often reported as barri-
ers on an organizational level [14, 16]. Also, care provid- Aim
ers do not always feel confident that the support systems The aim of this pilot study was to perform initial test-
can sufficiently meet the need of older adults subjected ing of an educational model aiming at improving health
to abuse [14]. Several studies report the importance of care providers’ preparedness to care for older adults
 Simmons et al. BMC Medical Education (2022) 22:597 Page 3 of 16

subjected to abuse. We used a mixed method approach during the Covid-19 pandemic, the number of partici-
to investigate: pants had to be limited. An invitation was e-mailed to all
the medical interns and the first 43 to respond (out of 58
1) How health care providers perceived the education employees in total) were given the possibility to partici-
and how it influenced a) their propensity to ask older pate. All interns employed on the other programme were
patients questions about abuse and b) their perceived invited to participate in the control group (n = 67).
ability to manage the response. Participants for the qualitative interviews were
2) Health care providers’ personal and organizational a) recruited from the intervention group. We attempted
sense of responsibility to identify victims, b) barriers to use purposeful sampling by using the participants’
and facilitators towards asking questions about elder follow-up evaluations to reach participants with various
abuse and managing the response, and how those opinions of the educational day. The possibility for this
were affected by the education was, however, limited due to the small number of partici-
pants, resulting in a combination of purposeful and con-
venient sampling.
Methods
Design Educational model
This study describes a non-randomized controlled cohort To facilitate transferal of the acquired competence into
pilot study of an educational model concerning elder practice, the educational model tested was grounded in
abuse, targeting health care providers. A mixed method participants’ own experiences, focused on active par-
convergent parallel design was used, i.e. both quantitative ticipation and contained a mix of different pedagogi-
and qualitative data were used and they were collected at cal methods, i.e. theory, group discussions and forum
the same time and given the same importance in analy- theatre.
sis [31]. This method was applied to collect different but
complementary data to understand the effects of the edu- Theory
cational model more comprehensively than when using The first part of the education (approximately 1 h 10 min)
either method alone. Using complementary data sources consisted of a lecture given by two of the authors (JS,
is especially beneficial for evaluation studies with limited ML). The lecture started by defining elder abuse and
sample sizes [32], such as ours. In the convergent parallel elaborating on its prevalence and health consequences.
design, quantitative and qualitative data are collected and Two short films illustrating elder abuse cases were inter-
analysed separately. Thereafter an interpretation is con- spersed in the lecture to increase awareness of what elder
ducted regarding how the two data sets converge, diverge abuse is and elicit emotions. One of the films illustrated
and relate to each other [31]. In accordance with this we a case of psychological abuse and abuse related to physi-
will present the quantitative and qualitative data sepa- cal dependence by an intimate partner while the other
rately in the methods and results sections, while inter- film showed a case of neglect by the victim’s son. There-
preting and relating the data together in the discussion. after followed some suggestions on how to talk about
elder abuse with older patients, a brief introduction to
Participants and setting trauma informed care and a presentation on the societal
In the Swedish health care system, medical interns are resources available for victims. Written material, includ-
physicians who have completed the medical education ing pamphlets about elder abuse directed at staff and
but are not yet licensed medical practitioners. They com- patients, was distributed. Also, a screening form that can
plete a structured program working in internal medicine, be used to identify older adults exposed to abuse (REAG-
surgery, psychiatry, emergency medicine and primary ERA-S) was introduced [33]. Finally, a brief introduction
care for 18 months. The current study was carried out in to motivational interviewing [34] and its applicability in
a region in southeast Sweden which holds two equivalent the context of elder abuse was given.
programmes for training medical interns. Participants for
the intervention group were recruited from one of the Case‑based group discussions
programmes while the other program provided partici- The second part of the education (approximately 1  h
pants for the control group. The education took place in 15 min) consisted of case-based group discussions. Two
October 2020 and was organized as part of the interns’ sets of two short films illustrating patient-health care
ordinary educational programme. Initially, all interns provider encounters were used as an introduction to the
employed on one of the two programmes were intended discussions. The first film illustrated a rather unsatisfac-
to be offered participation in the education. However, tory encounter in which an older woman told her health
because of restrictions concerning social gatherings care provider about psychological abuse she endured
Simmons et al. BMC Medical Education (2022) 22:597 Page 4 of 16

from her husband. The provider reacted by simply tell- e.g. what happens when the provider talks to the older
ing the woman what to do (telling the woman to leave woman and her son separately? How can the provider
her husband), but the provider did not try to understand gain the older woman’s trust and find ways of helping
the complexity of the woman’s situation nor her per- her? In addition to the aforementioned scene, another
sonal preferences. This was followed by group discussion one, focusing on how to ask questions about abuse, had
about challenges in the encounter and how it could be been prepared in advanced. Also, participants contrib-
improved. Thereafter a second short film illustrating the uted scenes based on their own experiences, which the
same encounter was shown, but this time the health care drama teachers acted out as improvisations. Due to time
provider asked open questions and let the older woman restrictions all scenes could not be played out, but they
talk about the challenges of her situation. Together the stimulated reflection and discussions among participants
provider and older woman started finding strategies to about all forms of elder abuse, including difficulties con-
handle her difficulties. Afterwards, group discussions cerning how to identify cases when there are no obvious
focused on how different strategies taken by the health physical signs of abuse.
care providers in the two films changed the outcome of
the patient encounter. The same procedure, i.e. two short Measurement
films with different endings combined with group discus- The Responding to Elder Abuse in GERiAtric care –
sions, was thereafter repeated once again. The topic of Provider questionnaire (The REAGERA-P) was used
the other set of short films was how motivational inter- to evaluate the education quantitatively [10]. It con-
viewing techniques can be used to help decision making tains questions about personal experiences of talking to
about life changes and whether to seek help for abusive patients about abuse, self-efficacy for asking questions
experiences. about abuse and managing the response, cause for con-
cern as well as known barriers and facilitators for asking
Forum theatre questions about abuse. Self-efficacy pertains to a person’s
The third part of the education (approximately 2  h perceived ability to conduct a certain task and theo-
30  min) consisted of forum theatre, led by three drama retically a person with high self-efficacy is more likely to
teachers. In forum theatre the spectators are not only perform that particular task successfully [35]. The devel-
observers, but rather spect-actors, urged to participate opment and validation of REAGERA-P is described in
in the scene [25, 26]. The forum theatre started with the detail elsewhere [10]. In brief it has been tested for face
drama teachers acting out a scene showing a problem- validity and comprehensibility through cognitive inter-
atic health care encounter. The first scene portrayed an views, while construct and convergent validity was tested
older woman about to be discharged from the emergency in a sample of 154 health care providers. Cronbach’s
room after being treated for a broken arm. Plans had alpha for the self-efficacy scales was satisfactory (asking
been made to initiate home care but when her son, who questions = 0.75 and managing the response = 0.87) [10].
lived with her, heard about that he became aggressive In this study some new items about sense of responsibil-
towards both the health care provider and towards his ity as well as personal and organizational barriers were
mother. The health care provider started asking the older added to the REAGERA-P (items c and f below). This was
woman questions about abuse (with the son present) but done to expand the items that could potentially mediate
she denied such experiences and before she left together the effect of the educational model. Comprehensibil-
with her son, she said that home care was not needed ity of those items was assured by conducting cognitive
after all. Consequently, the mother’s need for home care interviews with five health care providers. All the items
was neglected, and questions about abuse were inappro- in REAGERA-P used for this study are described below
priately posed in the presence of the aggressive son. In and can be found in their entirety as additional file  1.
the next step of the forum theatre, the same scene was Responses were given on ordinal scales that are pre-
played out again, but now the participants were invited sented together with the results in Table 2.
to pause the scene at any time and suggest alternative
ways for the health care provider to act. They were also Items in REAGERA‑P concerning study aim 1
encouraged to take the role of the health care provider in
the play themselves, thereby testing the consequences of a) Propensity to ask questions: How many times have
alternative ways of acting. Discussions between partici- you asked older patients questions about abuse in the
pants about the situation and ways to manage it were also past six months?
initiated. The scene was repeated several times, and in b) Perceived ability to manage the response: A five item
this way, participants together explored how their actions self-efficacy scale in which respondents were asked to
and reactions could improve the health care encounter, rate their perceived ability to manage different tasks
 Simmons et al. BMC Medical Education (2022) 22:597 Page 5 of 16

in their work (e.g. helping an older patient subjected The Pearson’s chi square test, or when appropriate
to abuse to reach the right body in health care or the Fisher’s exact test, was used to test for differences in
right support function in society) on a scale from 0 background characteristics of the intervention and con-
(= able to manage the task very poorly) to 10 (= able trol group, as well as for differences between the group of
to manage the task very well). respondents lost to follow-up and retained in the inter-
vention and control group respectively.
Due to the low number of participants, it was not pos-
sible to use multivariate statistics or ANOVA to make a
Items in REAGERA‑P concerning study aim 2 comparison between changes in intervention and control
group while controlling for other variables. Instead, we
iii) Sense of responsibility: To what extent do you feel tested for univariate differences at baseline and 6-month
that a) healthcare services and b) you, in your pro- follow-up.
fessional role, have a responsibility to identify older
patients who are or have previously been subjected to Aim 1, Propensity to ask questions
abuse? Responses were given on an ordinal scale (none, 1 time,
iv) Cause for concern (personal barriers): How con- 2–4 times, 5 or more) and two analyses were performed.
cerned are you about the following things when First, answers were kept on the ordinal level and Wilcox-
it comes to asking older patients questions about on’s signed rank test for paired samples was used to com-
abuse? a) That the patient reacts negatively if I ask pare within-group frequency of asking questions about
questions; b) That the patient-care provider relation- abuse at baseline and follow-up. Second, data was dichot-
ship will be negatively impacted if I ask questions; c) omized and the McNemar test for paired data was used
That I will not be able to offer the patient a good fol- to compare the proportion of respondents who reported
low up. having asked older patients about abuse in the previous
v) Self-efficacy for asking questions (personal facili- 6 months with those who had not or did not remember
tator): A three item self-efficacy scale in which doing so.
respondents were asked to rate their perceived ability
to perform different tasks in their work (e.g. asking
Aim 1, Perceived ability to manage the response
questions about abuse to an older patient who has no
A sum-score for the five items that constitute self-efficacy
clear indication of now being, or having previously
for managing the response was created for each partici-
been, subjected to abuse) on a scale from 0 (= able to
pant and a mean score was calculated for the interven-
manage the task very poorly) to 10 (= able to manage
tion and control group respectively. A paired t-test was
the task very well).
used to investigate within group changes on the self-effi-
vi) Personal and organizational barriers: To what extent
cacy scale between baseline and follow-up and an inde-
do you think that, at your workplace, the follow-
pendent sample t-test were used to compare difference
ing factors prevent you from asking older patients
between the intervention and control group.
questions about abuse? a) Lack of time; b) My own
insufficient awareness of the problem; c) Inadequate
Aim 2, Sense of responsibility, personal and organizational
routines at the workplace for asking questions; d)
barriers and facilitators
Inadequate routines at the workplace for handling
A sum-score for the three items that constitute self-effi-
the answer.
cacy for asking questions was calculated and a paired
t-test was used to compare within group changes in mean
between baseline and follow-up while an independent
Quantitative data collection and analysis t-test was used to compare differences between the inter-
Data was collected at baseline and at 6-month follow- vention and control group. For ordinal scales, Wilcoxon’s
up. Participants in the intervention group filled out a signed rank test for paired samples was used to analyse
web-based version of the REAGERA-P as the first part changes between baseline and follow-up.
of the educational day (October 2020), while the survey
was e-mailed to participants in the control group a cou- Qualitative data collection and analysis
ple of weeks later. The follow-up survey was e-mailed to Semi-structured face-to-face interviews with open-ended
both the intervention and control group in April 2021, i.e. questions were conducted by the second author (AM)
6  months post-intervention. Three reminders were sent within five months following the education. The inter-
at baseline and four at follow-up. viewer had not been involved with developing or imple-
menting the educational model. He was, at the time, part
Simmons et al. BMC Medical Education (2022) 22:597 Page 6 of 16

of the research team but also a medical intern himself, However, the attrition analysis revealed that in the inter-
and was a participatory observer during the education, vention group a higher proportion of those retained at
i.e. he took part in group discussions and forum theatre follow-up reported education about violence in a close
but was not included in the quantitative data collection. relationship at baseline, compared to those lost to follow-
An interview guide with open-ended questions was used up (p = 0.03) (Table 1).
(additional file 2). The interview focused on participants’
experiences regarding the education, as well as the kind Aim 1
of support they felt would help them to ask older patients Propensity to ask questions
questions about abuse in their clinical practice. All the We found a significant increase in the frequency of ask-
interviews took place in a secluded room on a university ing questions in the intervention group at follow-up
campus, lasted 40–60  min and were recorded and tran- (p = 0.047), i.e. more respondents reported having asked
scribed verbatim. questions about abuse on several occasions at follow-up.
Data was analysed using qualitative content analy- The same pattern was not found for the control group
sis as described by Graneheim and Lundman [36]. First, (p = 0.38) (Table  2). There was no significant change
two of the authors (AM, KS) read through the tran- concerning the proportion of participants who reported
scripts repeatedly to get a sense of the whole. Thereaf- asking patients questions about abuse, but the trend was
ter, AM extracted the text into meaning units related to towards an increase in the intervention group (baseline
the study’s aim. The meaning units were then labelled n = 5; 31%; follow-up n = 9; 56% p = 0.13) and towards a
with codes and sorted into subcategories. An inductive decrease in the control group (baseline n = 9, 64%; fol-
approach was applied, meaning the codes were com- low-up n = 6, 43%, p = 0.25) (Table 1).
pared based on similarities and differences before being
further sorted into subcategories [36, 37]. The subcatego- Perceived ability to manage the response
ries were then organized into a smaller number of cate- We found a significant increase in self-efficacy between
gories. The categories and subcategories were discussed baseline and follow-up for managing the response
and revised by the same two authors (AM, KS) continu- (p = 0.04) in the intervention group, but not the control
ously throughout the process. Thereafter, the results group (p = 0.14). The mean difference between baseline
were also discussed with the first author (JS), resulting in and follow up self-efficacy score was 3.8 for the interven-
agreement on the final four categories. The continuous tion group and 2.8 for the control group, this one-point
discussion within the research group aimed at strength- difference between the groups was not statistically differ-
ening the validity of the study, not by reaching identical ent (p = 0.7) (Table 3).
statements, but rather by increasing reflexivity when the
authors contested each other’s thoughts and interpreta- Aim 2
tion of results [38]. Sense of responsibility and internal barriers and facilitators
for asking questions about abuse and managing
Results the response
Quantitative results Self-efficacy for asking questions was significantly
In total, 43 medical interns signed up for the educational increased between baseline and follow-up in the inter-
day but only 39 attended. Also, three participants came vention group (p = 0.04), but not in the control group
late or had technical problems answering the question- (p = 0.84). The mean self-efficacy score increased by
naire and were therefore excluded from the study, i.e. 36 2.9 in the intervention group and decreased by 0.3 in
interns were eligible for inclusion. One person declined the control group between baseline and follow up. This
participation, leaving a sample of 35 who answered the in-between group difference in mean of 3.2 points was
baseline questionnaire (response rate 97%). Nineteen however not significant (p = 0.10) (Table  3). We found
participants (54%) in the intervention group were lost to no significant changes in either the intervention or con-
follow-up, leaving a sample of 16 (46%) who participated trol group concerning estimation of own responsibil-
at both measurement points. Of the 67 medical interns ity for asking questions at follow up. However, in both
asked to participate in the control group, 20 answered the intervention and control group most respondents
the baseline survey (response rate = 30%). Six partici- reported a high sense of responsibility already at baseline
pants in the control group (30%) were lost to follow-up, (Table  2). Respondents in the intervention group attrib-
leaving a sample of 14 (70%) who participated at both uted higher responsibility to the health care organization
measurement points. There was no significant difference to ask questions about abuse at follow-up compared to
in background characteristics between the intervention baseline (p = 0.046), which was not seen in the control
and control group in the sample retained at follow-up. group (p = 0.16). Respondents in the intervention group
 Simmons et al. BMC Medical Education (2022) 22:597 Page 7 of 16

Table 1  Background characteristics of participants (intervention group n = 16, control group n = 14) and attrition analysis, significant
differences at the p < 0.05 level are marked as bold
Sample Attrition analysis

Intervention Control Intervention Control

n = 16 n = 14 Lost to Retained Lost to follow Retained

follow-up n = 16 up n = 14
n = 19 n = 6
n % n % N % n % n % n %

Sex
Female 10 62.5 10 71.4 14 73.7 10 62.5 3 50.0 10 71.4
  Male 6 37.5 4 28.6 5 26.3 6 37.5 3 50.0 4 28.6
Age
  ≤ 34 years 12 75.0 14 100 15 78.9 12 75.0 6 100 14 100
  35–49 years 4 25.0 - - 4 21.1 4 25.0 - - - -
Medical school training about abuse at baseline
  No, Do not remember 4 25.0 1 7.1 2 10.5 4 25.0 2 33.3 1 7.1
  Yes, violence in close relationships 12 75.0 13 92.9 17 89.5 12 75.0 4 66.7 13 92.9
  Yes, elder abuse 2 12.5 4 28.6 4 21.1 2 12.5 1 16.7 4 28.6
Other training about abuse at baseline
  No, Do not remember 8 50.0 5 35.7 15 78.9 8 50.0 2 33.3 5 35.7
  Yes violence in close relationships 8 50.0 9 64.3 3 15.8 8 50.0 4 66.7 9 64.3
  Yes, elder abuse 2 12.5 3 21.4 1 5.3 2 12.5 2 33.3 3 21.4
Asked questions about abuse at baseline
  No, Do not remember 11 68.8 5 35.7 15 78.9 11 68.8 3 50.0 5 35.7
  Yes 5 31.3 9 64.3 4 21.1 5 31.3 3 50.0 9 64.3
Asked questions about abuse at follow up
  No, Do not remember 7 43.8 8 57.1
  Yes 9 56.3 6 42.9
Pearson’s chi square test, or when appropriate Fisher’s exact test, were used to compare differences in background characteristics between intervention and control
group at baseline as well as differences between those lost to follow up and those retained in the intervention and control group respectively

were less likely to report their own lack of awareness as baseline and follow-up. However, fewer respondents in
a barrier at follow-up (p = 0.04), while there was no such the control group reported that a lack of routines for ask-
difference in the control group (p = 0.16) (Table 2). There ing questions (p = 0.03) and a lack of routines for manag-
were no significant changes between baseline and follow- ing the response (borderline significant, p = 0.052) were
up in either the intervention or control group concern- barriers to asking questions at follow-up. No significant
ing any of the causes for concern when asking questions difference was seen in the intervention group for the
about abuse. The higher levels of concern (rather worried same two variables (p = 0.32 and p = 0.37 respectively).
and very worried) were commonly reported for concern No changes were seen for time restraints as a barrier in
about not being able to provide a proper follow-up, while either the intervention or control group (Table 2).
concerns about negative reactions or negative effects
on the patient-provider relationship were commonly Qualitative results
reported at the lower levels of concern (not at all or lit- Four participants were interviewed, all of whom were
tle worried) at both measurement points in both groups female. Analysis of the interviews resulted in four cat-
(Table 2). egories: Internal processes and new perspectives; Moti-
vational processes; Area of responsibility; Feelings of
Organizational barriers to asking questions about abuse insecurity and challenges in responding to elder abuse.
and managing the response
A majority of respondents in both the intervention and Internal processes and new perspectives
control group reported a lack of routines for manag- The participants all described emotional reactions to
ing cases as a barrier to some or a large extent at both the content of the educational day, e.g. frustration,
Simmons et al. BMC Medical Education (2022) 22:597 Page 8 of 16

Table 2  Frequency of asking questions about elder abuse, sense of responsibility and barriers for asking questions
Intervention group (n = 16) Control group (n = 14)

Base-line 6 m Rank (n) p Base-line 6 m Rank p

Rank (n) Neg Pos Ties Neg Pos Ties

Asked questions about abuse previous 6 months 1 6 9 0.047 4 3 7 0.38

  No, Do not remember 11 7 5 8
  One time 1 1 5 3
  2–4 times 3 6 3 1
  5 times or more 1 2 1 2
Sense of responsibility
  Own responsibility 1 5 10 0.10
  None - - - 3 3 8 1
  Small extent 1 - 1 1
  Some extent 5 3 6 6
  Large extent 10 13 7 7
Health care responsibility 0 4 12 0.046 0 2 12 0.16
  None - - - -
  Small extent 1 - - -
  Some extent 5 3 8 6
  Large extent 10 13 6 8
Perceived Individual level barriers
  Own lack of awareness 2 9 5 0.04 1 4 8 0.16
  Large extent 7 2 6 3
  Some extent 8 10 6 8
  Small extent - 4 2 2
  Not at all 1  - - -
Concern for follow up 6 3 7 0.19 5 2 7 0.16
  Very worried 3 2 2 3
  Rather worried 5 9 4 6
  A little worried 5 5 4 2
  Not at all 3 - 4 3
Concern negative reaction 2 4 10 0.41 3 4 7 0.71
  Very worried - - - -
  Rather worried 3 1 1 2
  A little worried 6 8 8 5
  Not at all 7 7 5 7
Concern relationship 3 5 8 0.37 4 1 9 0.16
  Very worried 1 - - -
  Rather worried 2 - 1 2
  A little worried 6 10 4 6
  Not at all 7 6 9 6
Perceived organizational level barriers
  Lack of time 5 5 6 1.0 3 5 6 0.48
  Large extent 6 3 4 4
  Some extent 6 12 8 6
  Small extent 4 1 1 3
  Not at all - - 1 1
 Simmons et al. BMC Medical Education (2022) 22:597 Page 9 of 16

Table 2  (continued)
Intervention group (n = 16) Control group (n = 14)

Base-line 6 m Rank (n) p Base-line 6 m Rank p

Rank (n) Neg Pos Ties Neg Pos Ties

Lack of routines asking 2 4 10 0.32 1 7 5 0.03

  Large extent 5 3 4 1
  Some extent 8 9 5 7
  Small extent 3 4 3 5
  Not at all - - 1 1
Lack of routines managing 3 5 8 0.37 2 8 4 0.05
  Large extent 6 5 7 4
  Some extent 8 7 5 5
  Small extent 2 4 1 3
  Not at all - - 1 2

Changes between baseline and follow up regarding frequency of asking questions about elder abuse as well as changes in sense of responsibility and perceived
barriers for asking questions. A positive rank signifies a positive change, i.e., higher sense of responsibility and lower lever of perceived barrier. Significant changes
(p < 0.05) are written in bold and have been calculated using Wilcoxon’s signed rank test for paired samples

Table 3  Self-efficacy for asking older patients questions about abuse and managing the response
Intervention group Control group Between group comparisons
(n = 16) (n = 14)
Mean SD P-value Mean SD p-value Mean SD p-value

Self-efficacy asking questions

  Baseline 16.3 3.5 19.6 5.3 3.3 1.6 0.05
  6 months follow up 19.2 3.0 19.4 5.5 0.2 1.6 0.92
  Difference in mean 2.9 5.0 0.04 -0.3 5.2 0.84 3.2 1.9 0.10
Self-efficacy managing the response
  Baseline 24.3 7.3 24,9 8.2 0.7 2.8 0.81
  6 months follow up 28.1 7.6 27.8 11.4 -0.2 3.5 0.95
  Difference in mean 3.8 6.7 0.04 2.8 6.6 0.14 1.0 2.5 0.70
Significant changes (p < 0.05) between baseline and 6 months follow-up are written in bold and have been calculated using paired t-tests

discomfort and sadness, but also commitment and curi- Participants expressed that they had made associations
osity. Experiencing scenarios that were perceived as real- between what they were seeing and hearing during the
istic was an important factor in evoking the emotional educational day, and their own previous experiences. The
responses. Also, an increased awareness of elder abuse portrayal of various abuse-related scenarios, particularly
was articulated in all interviews. Participants reported in the videos and in the forum theatre, made the partici-
finding the subject more important because of the edu- pants reflect on their own experiences of similar situa-
cational day, as well as having more general knowledge tions. This prompted emotional reactions for some, e.g.
and a more comprehensive understanding of the issue. when realizing they hadn’t asked questions about abuse
Some participants stated that realizing how common and in situations where they now thought it would have been
under-diagnosed elder abuse is, as a result of listening to relevant. In addition, such realizations made them start
the theoretical lecture, had made a particular impact on to apply their acquired knowledge of elder abuse to real
them. life experiences.
“…I mean, I think the statistics were tough. Yes, that “When we saw the videos and other [participants]
was really the toughest part. […] statistics and this started to talk about what they’ve seen in wards
information about… […] how unusual it is for health and primary care offices, you realized ‘Oh my god,
care providers to ask [about elder abuse], and how so many things have been witnessed’. […] You started
many people go unnoticed”. (Participant 4). to think about what you’ve seen and experienced
Simmons et al. BMC Medical Education (2022) 22:597 Page 10 of 16

yourself. That it [elder abuse] is much more common haps now…” (Participant 4)
than you think and something you need to open your
Participants also mentioned being equipped with more
eyes to…” (Participant 2).
tools to manage situations involving elder abuse, such as
The participants also described that during the day, having the pamphlets that were handed out during the
especially in the forum theatre and group exercises, dis- day that contained phone numbers and other contact
cussions would continuously arise between participants, information to supportive organizations and authorities.
which allowed them to share their own thoughts and They also cited being provided with ideas and inspiration
gain perspectives other than their own. Seeing and hear- concerning how and when to ask questions about abuse
ing their colleagues and the actors play out the scenes in and that they had learned strategies to manage situations
the forum theatre made the participants reflect on their where abuse could be suspected. However, some partici-
own choice of strategy, as well as providing them with the pants emphasized the importance of practising how to
opportunity to learn lessons from others. ask questions out loud about elder abuse, and that they
perceived that there had been too little time to do that
“And the thing about seeing…You learn from how
during the educational day.
your colleagues are managing it, what words they
are using. And when they got stuck, someone else
made a contribution. ‘Maybe you could do it [han- Area of responsibility
dle the situation] like this?’ Or ‘I would have done it All the participants expressed the importance of the
like that’. Then you also learn from your colleagues’ expectations on them as physicians and on their organi-
mistakes” (Participant 2). zations. Some clearly felt that it was their responsibil-
ity to investigate suspected ongoing elder abuse, and to
help patients who were suffering from abuse. However,
Motivational processes the informants also stated that the extent of their pro-
Most participants expressed that the educational day fessional responsibility needs to be more clearly defined
made them feel involved with and committed to care for and explained to them, that they needed to know where
victims of elder abuse. For instance, the forum theatre their responsibilities toward the patient ended, and other
evoked a feeling of involvement and some participants health care professionals’ and social welfare authori-
described how they felt compelled to take action to make ties’ responsibilities started. Staying within their area of
a difference for the better in the scene that was being responsibility was important for the informants and they
played out. An increased awareness of the issue, as well felt that stepping outside it would be difficult and could
as understanding that their actions could make a differ- put them in a difficult position. For instance, some felt
ence for the better, was mentioned as important in con- that they might be questioned for prioritizing the man-
veying motivation and interest. agement of abuse-related problems if it took time away
from their medically related tasks. All the participants
“… having time to talk about the subject and seeing also expressed how the patient had to take responsibility
situations… that I hadn’t for example paid attention as well, e.g. taking action to make a change and accepting
to before. And feeling that I could be of importance help that was offered to them.
[when encountering victims of elder abuse]. That I
could make a difference through my questions or “And I believe that if you talk about it more and
behaviour, that it could improve things for these peo- establish a norm that it’s important that we assist
ple… I felt like that created motivation” (Participant and aid in a certain way, then I think we employ-
3). ees will… work for that and take more responsi-
bility. The way things are now, I don’t feel like it is
When asked about how their way of working had that way […] It is not our responsibility […] we are
changed after the educational day, the participants par- supposed to manage other things. And if it is about
ticularly mentioned acting from the position of being being exposed [to abuse] in a relationship, it’s the
more aware of elder abuse, e.g. paying attention to warn- victim themself or other professionals who must take
ing signs and symptoms, asking direct questions and not responsibility.” (Participant 3).
being afraid to ask about abuse, more routinely screening
for elder abuse and considering types of abuse other than
physical. Feelings of insecurity and challenges in responding to elder
abuse:
“…But I have to say that in those situations when it’s
Some participants felt that the education did not pro-
more of an unpleasant atmosphere, I have never had
vide them with a solution to an important part of their
the courage to make a comment on that. But per-
 Simmons et al. BMC Medical Education (2022) 22:597 Page 11 of 16

insecurities, i.e. how to generally manage cases of elder care provider involved that the issue wouldn’t end with
abuse. Some respondents emphasized that they remained just their contribution, which could provide a sense
unsure of how to act in practice also after the educa- of security for them. Another approach to lessen their
tion. Not being able to present the patient with a definite insecurities was to seek support. During the interviews,
measure or solution left the participants with a feeling all the participants expressed a need for support to
of inadequacy. For instance, one participant shared that manage patients subjected to abuse. This support could
her insecurities about how to manage the situation could be e.g. practical routines, consulting more experienced
influence her to avoid the subject all together and refrain colleagues, social workers or for some a wish to hand
from asking questions about abuse. over responsibility for following up and managing the
situation to someone else.
“Often it’s like…What we hear is: ‘Don’t order those
lab tests, because it will result in incidental find- “…I often feel insufficient not knowing what resources
ings.’ And it is a bit the same situation here: ‘Don’t there are […] …Who can this person [the patient]
ask that question, because you will have a problem make contact with? Who do we usually send refer-
you can’t manage’. I think that is often the case. rals to? […] I wonder if there is something more I
And I feel like that’s a shame, but sometimes I feel could do, that I’m not aware of.” (Participant 3).
just like that: ‘What do I do with the answer?’.”
(Participant 2).
Integration of quantitative and qualitative results –
Participants required information about structured an overall picture of the educational effects
and uncomplicated ways to manage situations and An overview of both quantitative and qualitative results
sought the possibility to offer concrete measures to the and how the findings relate to each other can be found
patient, e.g. offering an appointment to someone in in Fig.  1. The results indicate that the model led to an
charge of follow-ups or making referrals to responsible increased propensity to ask older patients questions
departments or resources. This would assure the health about abuse and though uncertainties about the best

Fig. 1  Overview of results. Quantitative data collected at baseline and 6-month follow-up in both the intervention and control groups (blue), and
qualitative data collected post-intervention (green). Both the propensity to ask older patients questions about abuse and providers’ perceived ability
to manage the response (yellow arrows) were improved in the intervention group at follow-up. The change was possibly mediated by an increased
self-efficacy for asking questions and a higher awareness about the issue. Also, the results indicated that the perceived ability to manage the
response affected the propensity for asking questions
Simmons et al. BMC Medical Education (2022) 22:597 Page 12 of 16

possible response to elder abuse remained, participants [40]. One explanation for the higher self-efficacy for ask-
also reported higher self-efficacy for managing cases of ing questions as well as managing the response found at
elder abuse post intervention. The results were possibly follow-up is likely using forum theatre as a method. The
mediated by an increased self-efficacy for asking ques- scenes played out functioned as a form of skills train-
tions and a higher awareness about the issue post inter- ing in which participants and drama teachers together
vention, both of which were reflected in the quantitative explored different ways of acting. Informants articu-
as well as qualitative results. Also, the results indicated lated in the interviews that the forum theatre and group
that the perceived ability to manage the response, includ- discussions stimulated reflections on their own way of
ing availability of guidelines affected the propensity for acting and provided them with new insights and perspec-
asking questions. As illustrated in Fig. 1 the quantitative tives. Similarly, using interactive teaching techniques in
(blue) and qualitative (green) results were in agreement education about elder abuse has previously been found to
and complement each other, which is further elaborated be more effective than non-interactive teaching interven-
on in the discussion. tions [30] and forum theatre has been found useful when
practising communication skills in health care education
Discussion [29].
The main aim of this mixed method pilot study was to The mean self-efficacy for asking questions was higher
evaluate an educational model to increase health care in the control group compared to the intervention group
providers’ preparedness to care for older adults subjected at baseline (borderline significant p = 0.05, Table 3). This
to abuse. Our findings indicated a positive effect, moti- finding may have several reasons, e.g., 64% of participants
vating a full-scale test of the intervention. in the control group compared to 31% in the intervention
group had experiences of asking older patients questions
Propensity to ask older patients questions about abuse about abuse at baseline and a larger proportion of par-
Participants in the intervention group reported a signifi- ticipants in the control group compared to intervention
cant increase in frequency of asking questions after the group reported previous education about both violence
intervention, which was not found in the control group. in close relationships and elder abuse (Table  1). These
Likewise, in the qualitative interviews, participants differences between the groups were not statistically
expressed that they had changed their practice after the significant – potentially because of the low number of
education by, for example, being more attentive to signs participants—but might partially explain the difference
of abuse and not hesitating to ask questions. Altogether, in self-efficacy between the two groups at baseline. The
this indicates that the education led to an increased pro- differences also indicate that potential confounding fac-
pensity to ask older patients questions about abuse. The tors may be unevenly distributed between the two groups
change was likely in part mediated by a higher commit- and therefor the within group differences pre- and post-
ment to care for victims and an increased awareness of intervention may be a more reliable measure of the effect
the issue, both of which were clearly articulated in the of the education than the in-between groups analysis.
interviews. Also, the quantitative analysis showed that Our within group analysis revealed that the self-efficacy
fewer respondents regarded their own lack of awareness increased significantly at follow up in the intervention
as a barrier to asking questions about abuse at follow-up. group, but not in the control group, which indicates an
Higher awareness after education about elder abuse is effect of the education. However, this needs to be veri-
unsurprising and has also previously been reported [30, fied in a full-scale test of the model in which multivariate
39]. Some mediators of the increased awareness were analyses including potential confounding factors can be
articulated in the interviews, e.g. providing knowledge included and more reliable in-between group compari-
about different aspects of elder abuse during the lec- sons can be made.
tures and case descriptions in the short films and forum
theatres. Perceived ability to manage the response
Another possible mediator for the increased propen- We found a significant increase in self-efficacy for man-
sity to ask questions about elder abuse was the improved aging the response in the intervention group at follow-
self-efficacy found in the intervention group at follow-up. up, which was not found in the control group. However,
High self-efficacy indicates feeling more secure about in the qualitative interviews, the informants expressed
how to perform a certain task, in this case asking ques- that they remained insecure about how to manage
tions about abuse and managing the response. Increased cases of elder abuse, and that this instilled in them a
self-efficacy for handling intimate partner violence has feeling of inadequacy. One potential explanation for
previously been associated with a higher likelihood of this discrepancy between the quantitative and quali-
screening for intimate partner violence in health care tative results is that the education might have opened
 Simmons et al. BMC Medical Education (2022) 22:597 Page 13 of 16

the eyes of some participants to the deficiency in clear operational tasks of their clinic. This seems to be an
guidelines and organizational preparedness to care for important barrier to getting involved in cases of elder
victims of elder abuse. In the interviews, the informants abuse and hence needs to be more clearly addressed in
expressed a need for support, as well as structured and a future full-scale test of the education. Only medical
concrete measures to offer the patient, as a means to interns, i.e. early-career physicians, were included in this
lessen their insecurities and to provide the best possi- study, which might have contributed to uncertainties
ble care for victims. To date, however, there are no evi- about what the professional role entails. However, simi-
dence-based programmes that can be generally applied lar to our findings, previous research has suggested that
to victims of elder abuse [41, 42]. This is in part due to health care providers do rely on other occupations, e.g.
a paucity of studies on the subject, but also a reflection social workers, to manage elder abuse issues [14].
on the complexity of the issue, where all victims have
individual needs [43, 44]. Hence, the sought-after clear- Limited ability to manage the response as a barrier
cut way of managing the response is perhaps not pos- to asking questions about abuse
sible to achieve. However, though still feeling insecure, The perceived ability to manage cases of elder abuse was
informants reported in the interviews that they felt also found to be related to the propensity to ask ques-
better equipped with tools and strategies to detect and tions, e.g. one of the participants interviewed mentioned
manage elder abuse after the education than before and that insecurities about how to manage cases might dis-
as previously mentioned, self-efficacy for managing the courage her from enquiring about elder abuse altogether.
response increased at follow up. Altogether, this indi- Likewise, concerns about not being able to provide a
cates that the education might provide health care pro- proper follow-up and lack of routines for managing cases
viders with skills to administer the best possible care were reported as barriers to asking questions about abuse
within the available societal support system. by many participants in both the intervention and con-
One theme recurring in the qualitative interviews was trol group. These barriers were unfortunately not affected
the sense of responsibility. The participants expressed in by the intervention.
the interviews that they perceived health care organiza- In the control group we found an unexpected signifi-
tions and themselves as health care providers as respon- cant decrease in considering lack of routines for asking
sible for detecting and managing elder abuse. This is questions and lack of routines for managing the response
concurrent with the quantitative results, in which most as barriers to asking questions. At approximately the
respondents reported a high sense of responsibility both same time as the intervention was carried out, new
at baseline and follow-up. There were no significant guidelines on how to manage violence in close relation-
changes concerning estimation of own responsibility ships, were introduced in the region where the study
for identifying victims at baseline compared to follow- was carried out. The section about elder abuse was con-
up. This could possibly be attributed to a ceiling effect. siderably extended in the new version of the guidelines
Already at baseline 10 out of 16 participants rated both and more practical advice on how to manage cases were
health care system responsibility and own responsibil- introduced. There is no obvious explanation for why
ity as the highest level (large extent). In a future study, this should have affected the control group more than
response categories need to be changed to avoid a ceiling the intervention group, but it is possible that—for some
effect. This is important, considering that a strong sense unknown reason—more participants in the control group
of professional responsibility has previously been associ- compared to the intervention group were aware of the
ated with having experiences of talking to older patients new guidelines and hence were less inclined to consider
about abuse [9]. lack of routines as a barrier for asking questions and
Though feeling responsible, the participants elabo- managing the response.
rated in the interviews on uncertainties about their role
in managing cases of elder abuse, e.g. what they were Implications for future education about elder abuse
expected to do and what they should leave for others The benefits of using different pedagogical strategies to
to do. The Swedish National Board of Health and Wel- evoke interest and commitment towards caring for older
fare states in its directive that whenever a patient shows adults subjected to abuse was a recurrent subject in the
signs or symptoms that indicate abuse, health care pro- interviews, e.g. one participant stated that the statistics
viders have a responsibility to ask questions and provide were the most striking part, while another underlined
contact with relevant societal resources [45]. Even so, that the films with patient cases elicited the most emo-
participants expressed concerns in the interview that it tions and interest. Building on previous experiences and
might not be appropriate to prioritize managing abuse- learning from each other in group discussions, and espe-
related issues, because it was not considered part of the cially forum theatre, were also repeatedly mentioned as
Simmons et al. BMC Medical Education (2022) 22:597 Page 14 of 16

important elements of the education. In a future full- research and might have affected the results. It is possi-
scale test of the education, it might be beneficial to inte- ble that the interviewer’s own experience and thoughts
grate theory and group discussions even more, to further about the education affected the questions asked dur-
stimulate collaborative learning and exchange of ideas. ing the interviews, as well as interpretation of the data.
More emphasis during the education needs to be These risks were handled by using an interview guide to
directed at what to do when an older adult reports expe- steer the interviews and by letting another researcher
riences of abuse. As mentioned before, there is no single (KS) with no relation to the participants take part in the
solution for how cases of elder abuse should be handled. analysis. Also, continuous discussions of coding and ten-
There are, however, some general principles that could tative subcategories and categories within the research
be further outlined during the intervention, e.g. what group stimulated reflexivity in the analysis, reducing the
trauma-informed care response entails for this patient risk of reproducing preconceived ideas. It has previously
group [46]. Also, more emphasis should be directed at been suggested that when insider and outsider research-
explaining the societal resources that are available for ers collaborate, this provides a possibility to gain a more
victims. To further make visible what can be done in profound understanding of the phenomenon studied [48,
individual cases, a brief remark after each forum theatre 49].
could also be made about potential ways of helping in
each specific case. In that way participants are provided Conclusion
with examples of how to manage a few role-model cases, In this mixed method pilot study, we tested an educa-
which they can then return to and contemplate when tional model aiming at improving health care provid-
faced with cases in clinical practice. ers’ preparedness to care for older adults subjected
to abuse. Results indicate that the education led to an
Limitations increased propensity to ask questions about abuse, possi-
In this pilot study, the number of participants was low, bly through raising awareness and commitment to iden-
and a rather large proportion were not retained at follow- tifying and helping in cases of elder abuse, as well as by
up. Also, only female participants agreed to participate increasing participants’ self-efficacy for asking questions.
in the qualitative interviews. Hence, results should be The results concerning managing the response to elder
interpreted with caution and all results need to be repli- abuse were more ambiguous: self-efficacy for manag-
cated in a full-scale study. However, the quantitative and ing cases of elder abuse increased at follow-up, but both
qualitative results point in the same direction, towards a the quantitative and qualitative results indicated that
positive effect, which increases the validity of the results. uncertainties about how to manage cases of elder abuse
Those who were retained in the intervention group were remained. Hence, one important lesson learned was that
more likely than those lost to follow-up to report previ- in future tests of the model, more focus needs to be put
ous education concerning violence in close relationships on how to manage cases of elder abuse. A full-scale test
at baseline, possibly indicating a higher interest in the of the model is currently being conducted. A study proto-
issue. It is possible that the intervention was less effective col has been published [47] and the study is registered at
among participants lost to follow-up considering they clinicaltrials.gov (register no NCT05065281).
might have been less committed to the issue to begin
with. Also, the group of participants was homogeneous
Abbreviation
in that they were young, early-career physicians. It is REAGERA-P: Responding to Elder Abuse in GERiAtric care – Provider
probable that the results would have been different if the questionnaire.
group had been more heterogeneous regarding e.g. pro-
fession or clinical experiences. This will be explored in Supplementary Information
a future full-scale test of the model that is on-going and The online version contains supplementary material available at https://​doi.​
org/​10.​1186/​s12909-​022-​03653-8.
will provide the opportunity to test for differences e.g.
between different professional categories [47].
Additional  file 1. REAGERA-P.
The qualitative interviews were in some cases con-
Additional  file 2. Interview guide.
ducted several months after the intervention took place,
which means there is a risk of recall bias. However, the
Acknowledgements
qualitative data was rich in content, which indicates
The authors wish to acknowledge Dr Stina Garvin and colleagues at the
that limitation of recall was not of great importance residency programme for pedagogical leadership at Region Östergötland for
for the results. The author who conducted the qualita- encouragement and important input during the early phase of the project.
We also acknowledge Karolin Olstam for her important help in creating
tive interviews (AM) was a participatory observer dur-
the short films used for group discussions, as well as help in organizing the
ing the education. This can be labelled a form of insider
 Simmons et al. BMC Medical Education (2022) 22:597 Page 15 of 16

education. Also, drama teachers Patrik Almgren, Jenny Hälleråd and Lena 4. Schofield MJ, Powers JR, Loxton D. Mortality and disability outcomes
Carlsson played an invaluable role when conducting the forum theatre. of self-reported elder abuse: a 12-year prospective investigation. J Am
Geriatr Soc. 2013;61(5):679–85.
Authors’ contributions 5. Dong XQ. Elder Abuse: Systematic Review and Implications for Practice. J
All the authors contributed to the design of the study. JS was primary Am Geriatr Soc. 2015;63(6):1214–38.
investigator and had the overall responsibility for planning and conducting 6. Fraga Dominguez S, Storey JE, Glorney E. Help-Seeking Behavior in
the study. JS and ML together planned the education and the quantitative Victims of Elder Abuse: A Systematic Review. Trauma Violence Abuse.
data collection. JS performed the quantitative data analysis. AM, JS and KS 2021;22(3):466–80.
together planned the qualitative data collection and AM recruited informants. 7. World Health Organization. Ageing and Life Course Unit & Université de
AM conducted the interviews, and coded the transcripts, supervised by KS. Genève. Centre interfacultaire de gérontologie. (‎2008)‎. A global response
Tentative categories were discussed first by AM and KS and later also with JS. to elder abuse and neglect : building primary health care capacity to deal
AM wrote the first draft of the method and results concerning the qualitative with the problem world-wide : main report. World Health Organization.
data. JS incorporated AM’s text into a first version of the whole manuscript https://​apps.​who.​int/​iris/​handle/​10665/​43869
and wrote the subsequent drafts. All authors contributed to and approved the 8. Touza GC. Influence of health personnel’s attitudes and knowledge in the
final manuscript. detection and reporting of elder abuse: An exploratory systematic review.
Psychosoc Interv. 2017;26(2):73–91.
Funding 9. Motamedi A, Ludvigsson M, Simmons J. Factors associated with health
Open access funding provided by Linköping University. This work was sup- care providers speaking with older patients about being subjected to
ported by the Swedish Crime Victim Fund, under grant no. 03384/2019. JS was abuse. J Elder Abuse Negl. 2022;34(1):20–37.
also supported by Region Östergötland under grant no. RÖ-937398, as well 10. Simmons J, Wenemark M, Ludvigsson M. Development and validation
as funding through the residency programme for pedagogical leadership in of REAGERA-P, a new questionnaire to evaluate health care provider
Region Östergötland (no specific grant). preparedness to identify and manage elder abuse. BMC Health Serv Res.
2021;21(1):473.
Availability of data and materials 11. Erlingsson CL, Carlson SL, Saveman BI. Perceptions of elder abuse: voices
The datasets generated and analysed during the current study are not publicly of professionals and volunteers in Sweden–an exploratory study. Scand J
available and are not available in its entity from the corresponding author on Caring Sci. 2006;20(2):151–9.
request. This is due to the small sample size, which would make it impossible 12. Alt KL, Nguyen AL, Meurer LN. The Effectiveness of Educational Programs
to maintain participant privacy and confidentiality if made publicly available. to Improve Recognition and Reporting of Elder Abuse and Neglect: A
Data not including background characteristics or other information that could Systematic Review of the Literature. J Elder Abuse Negl. 2011;23(3):213–33.
potentially be used to identify participants is available from the corresponding 13. Corbi G, Grattagliano I, Sabbà C, Fiore G, Spina S, Ferrara N, et al. Elder
author upon reasonable request. abuse: perception and knowledge of the phenomenon by healthcare
workers from two Italian hospitals. Intern Emerg Med. 2019;14(4):549–55.
14. Schmeidel AN, Daly JM, Rosenbaum ME, Schmuch GA, Jogerst GJ. Health-
Declarations care Professionals’ Perspectives on Barriers to Elder Abuse Detection and
Reporting in Primary Care Settings. J Elder Abuse Negl. 2012;24(1):17–36.
Ethics approval and consent to participate 15. Wagenaar DB, Rosenbaum R, Page C, Herman S. Primary care physicians
The study was approved by the Swedish Ethical Review Authority (ID nos. and elder abuse: current attitudes and practices. J Am Osteopath Assoc.
2020–02548 and 2020–04862) and was conducted in accordance with the 2010;110(12):703–11.
principles of the Declaration of Helsinki. All participants gave their written 16. Mohd Mydin FH, Othman S. Elder Abuse and Neglect Intervention in the
informed consent at the time of participation. Clinical Setting: Perceptions and Barriers Faced by Primary Care Physi-
cians in Malaysia. J Interpers Violence. 2020;35(23–24):6041–66.
Consent for publication 17. Kennedy RD. Elder abuse and neglect: the experience, knowledge, and
Not applicable. attitudes of primary care physicians. Fam Med. 2005;37(7):481–5.
18. Mohd Mydin FH, Othman S, Choo WY, Hairi NNM, Hairi FM, Syed Karim
Competing interests SN, et al. Clinical recognition of elder maltreatment and intention to
The authors declare that they have no competing interests. report among primary care doctors. J Elder Abuse Negl. 2020;32(1):72–83.
19. Cooper C, Selwood A, Livingston G. Knowledge, detection, and reporting
Author details of abuse by health and social care professionals: a systematic review. The
1
 Department of Acute Internal Medicine and Geriatrics in Linköping, American journal of geriatric psychiatry : official journal of the American
and Department of Health, Medicine and Caring Sciences, Linköping Univer- Association for Geriatric Psychiatry. 2009;17(10):826–38.
sity, Linköping, Sweden. 2 Department of Emergency Medicine in Norrköping, 20. Lachs MS, Pillemer KA. Elder Abuse. N Engl J Med. 2015;373(20):1947–56.
and Department of Health, Medicine and Caring Sciences, Linköping Univer- 21. Saveman B-I, Norberg A, Hallberg IR. The Problems of Dealing with Abuse
sity, Linköping, Sweden. 3 Department of Psychiatry in Linköping, and  Depart- and Neglect of the Elderly: Interviews with District Nurses. Qual Health
ment of Biomedical and Clinical Sciences, Linköping University, Linköping, Res. 1992;2(3):302–17.
Sweden. 4 Department of Health and Caring Sciences, Faculty of Health 22. Brijnath B, et al. "Build Rapport, Otherwise No Screening Tools in the
and Life Sciences, Linnaeus University, Kalmar, Sweden. World Are Going to Help”: Frontline service providers’ views on current
screening tools for elder abuse." The Gerontologist. 2020;60(3):472-82.
Received: 31 March 2022 Accepted: 22 July 2022 23. Pickering CEZ, Ridenour K, Salaysay Z, Reyes-Gastelum D, Pierce SJ. EATI
Island – A virtual-reality-based elder abuse and neglect educational
intervention. Gerontol Geriatr Educ. 2018;39(4):445–63.
24. Fisher JM, Rudd MP, Walker RW, Stewart J. Training Tomorrow’s Doctors to
Safeguard the Patients of Today: Using Medical Student Simulation Train-
References ing to Explore Barriers to Recognition of Elder Abuse. J Am Geriatr Soc.
1. Yon Y, Mikton CR, Gassoumis ZD, Wilber KH. Elder abuse prevalence in 2016;64(1):168–73.
community settings: a systematic review and meta-analysis. Lancet Glob 25. Boal A. Theatre of the oppressed. London: Pluto Press; 2000.
Health. 2017;5(2):e147–56. 26. Bruggemann AJ, Persson A. Using forum play to prevent abuse in health
2. Dong X, Chen R, Chang ES, Simon M. Elder abuse and psychological well- care organizations: A qualitative study exploring potentials and limita-
being: a systematic review and implications for research and policy–a tions for learning. Educ Health (Abingdon). 2016;29(3):217–22.
mini review. Gerontology. 2013;59(2):132–42. 27. Zbikowski A, Brüggemann AJ, Wijma B, Swahnberg K. Counteracting
3. Fisher BS, Zink T, Regan SL. Abuses against older women: prevalence and Abuse in Health Care: Evaluating a One-Year Drama Intervention with
health effects. J Interpers Violence. 2011;26(2):254–68. Staff in Sweden. Int J Environ Res Public Health. 2020;17(16):5931.
Simmons et al. BMC Medical Education (2022) 22:597 Page 16 of 16

28. Sevrain-Goideau M, Gohier B, Bellanger W, Annweiler C, Campone M,

Coutant R. Forum theater staging of difficult encounters with patients to
increase empathy in students: evaluation of efficacy at The University of
Angers Medical School. BMC Med Educ. 2020;20(1):58.
29. Middlewick Y, Kettle TJ, Wilson JJ. Curtains up! Using forum theatre to
rehearse the art of communication in healthcare education. Nurse Educ
Pract. 2012;12(3):139–42.
30. Mydin FHM, Yuen CW, Othman S. The Effectiveness of Educational Inter-
vention in Improving Primary Health-Care Service Providers’ Knowledge,
Identification, and Management of Elder Abuse and Neglect: A System-
atic Review. Trauma Violence Abuse. 2021;22(4):944–60.
31. Creswell JW, Plano Clark VL. Choosing a mixed methods design. In
Creswell JW, Plano Clark VL. Designing and Conducting Mixed Methods
Research. 2nd ed. California: SAGE Publications Inc; 2011. p. 53–106.
32. Bazeley P. Integrative Analysis Strategies for Mixed Data Sources. Am
Behav Sci. 2012;56(6):814–28.
33. Simmons J, Wiklund N, Ludvigsson M, Nägga K, Swahnberg K. Validation
of REAGERA-S: a new self-administered instrument to identify elder abuse
and lifetime experiences of abuse in hospitalized older adults. J Elder
Abuse Negl. 2020;32(2):173–95.
34. Miller WR, Rollnick S. Motivational interviewing: Helping people change,
3rd edition. New York, NY, US: Guilford Press; 2013.
35. Bandura A. Guide for constructing Self-efficacy scales. In: Pajares F, Urdan
T, editors. Self-Efficacy Beliefs of Adolescents. Greenwich, Conneticut:
Information Age Publishing; 2006. p. 307–37.
36. Graneheim UH, Lundman B. Qualitative content analysis in nursing
research: concepts, procedures and measures to achieve trustworthiness.
Nurse Educ Today. 2004;24(2):105–12.
37. Hsieh H-F, Shannon S. Three Approaches to Qualitative Content Analysis.
Qual Health Res. 2005;15:1277–88.
38. Malterud K. Qualitative research: standards, challenges, and guidelines.
The lancet. 2001;358(9280):483–8.
39. Khiya Marshall, Jeffrey Herbst, Candace Girod & Francis Annor. Do
interventions to prevent or stop abuse and neglect among older adults
work? A systematic review of reviews, Journal of Elder Abuse & Neglect.
2020;32(5):409-33. https://​doi.​org/​10.​1080/​08946​566.​2020.​18199​26
40. Lawoko S, Sanz S, Helstrom L, Castren M. Screening for Intimate Partner
Violence against Women in Healthcare Sweden: Prevalence and Determi-
nants. ISRN nursing. 2011;2011: 510692.
41. Fearing G, Sheppard CL, McDonald L, Beaulieu M, Hitzig SL. A systematic
review on community-based interventions for elder abuse and neglect. J
Elder Abuse Negl. 2017;29(2–3):102–33.
42. Baker PRA, Francis DP, Hairi NN, Othman S, Choo WY. Interventions
for preventing abuse in the elderly. Cochrane Database of Systematic
Reviews. 2016;Issue 8. Art. No.: CD010321. https://​doi.​org/​10.​1002/​14651​
858.​CD010​321.​pub2.
43. Ludvigsson M, Wiklund N, Swahnberg K, Simmons J. Experiences of
elder abuse: a qualitative study among victims in Sweden. BMC Geriatr.
2022;22(1):256.
44. Simmons J, Wiklund N, Ludvigsson M. Managing abusive experiences: a
qualitative study among older adults in Sweden. BMC Geriatr. 2022;22(1):456.
45. Swedish National Board of Health and Welfare. Våld i Nära relationer [Violence
in close relationships], Regulation number SOSFS 2014:4. Stockholm; 2014.
https://​www.​socia​lstyr​elsen.​se/​globa​lasse​ts/​share​point-​dokum​ent/​artik​elkat​
alog/​fores​krift​er-​och-​allma​nna-​rad/​2014-5-​7.​pdf. Accessed 1 Aug 2022.
46. Ernst JS, Maschi T. Trauma-informed care and elder abuse: a synergistic
alliance. J Elder Abuse Negl. 2018;30(5):354–67.
Ready to submit your research ? Choose BMC and benefit from:
47. Ludvigsson M, Motamedi A, Westerlind B, Swahnberg K, Simmons J.
Responding to Elder Abuse in GERiAtric care (REAGERA) educational
• fast, convenient online submission
intervention for healthcare providers: a non-randomised stepped wedge
trial. BMJ Open. 2022;12(5): e060314. • thorough peer review by experienced researchers in your field
48. Brannick T, Coghlan D. In Defense of Being “Native”: The Case for Insider • rapid publication on acceptance
Academic Research. Organ Res Methods. 2007;10(1):59–74.
• support for research data, including large and complex data types
49. Hanson J. Educational developers as researchers: the contribution of
insider research to enhancing understanding of role, identity and prac- • gold Open Access which fosters wider collaboration and increased citations
tice. Innov Educ Teach Int. 2013;50(4):388–98. • maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in pub- Learn more biomedcentral.com/submissions
lished maps and institutional affiliations.