Elder abuse in Lebanese primary care: prevalence and predictors from a cross-sectional study Authors: Lewaa Shoujaa, Marwan Azzam, Lina Abdouni, Reine Obeid, Danielle Mattar, Jana Ahmad, Wael Chami, Elie Toni Korbani, Jinan Usta, Umayya Musharrafieh, Georges Assaf Published: 2025 ABSTRACT AbstractBackgroundElder abuse is a growing public health concern with significant physical, psychological, and social consequences. In Lebanon, limited data on its prevalence and risk factors necessitate further investigation. This study estimated the prevalence of elder abuse among community-dwelling older adults in a primary healthcare setting in Lebanon and identified associated risk factors.MethodsWe conducted a cross-sectional study over 12 months at the outpatient Department of Family Medicine of the American University of Beirut Medical Center. We recruited participants aged 60 years and older and collected data on sociodemographic characteristics, frailty status, cognitive and psychological assessments, and elder abuse screening. We categorized abuse as physical, psychological, financial, and neglect-related mistreatment. We used descriptive statistics and multivariate logistic regression to identify risk factors.ResultsAmong 249 community-dwelling older adults, 19.7% (n = 49) reported at least one form of elder abuse. Psychological abuse was most common (16.1%), followed by neglect (8.8%), physical abuse (2.4%), and financial abuse (2.0%). Frailty independently increased the odds of abuse compared to robust individuals. Receiving financial support from children was independently associated with lower odds of abuse, while financial support from a spouse showed a borderline protective effect. Participants cared for by daughters or sons had significantly greater odds of reporting abuse compared to those cared for by a spouse. Each additional year of age corresponded to a 13% decrease in the odds of abuse.ConclusionsElder abuse is prevalent in Lebanese primary care, with psychological abuse the most reported type. Frailty, financial instability, and caregiving dynamics significantly influence abuse risk. Primary care systems should integrate routine abuse and frailty screening, strengthen caregiver support, and develop reporting pathways suitable for low-resource settings.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12877-025-06731-w.Keywords: Elder abuse, Frailty, Psychological abuse, Financial support, Primary healthcare, Lebanon, Risk factors Background According to the World Health Organization (WHO), elder abuse is defined as “a single or repeated act, or lack of appropriate action, occurring within any relationship where there is an expectation of trust, which causes harm or distress to an older person”. An older person is any individual aged 60 years and above [1]. Elder abuse includes physical, sexual, psychological, and emotional abuse; financial and material abuse; abandonment; neglect; and serious loss of dignity and respect [1]. A systematic review of 52 studies in 28 countries from diverse regions estimated that 1 in 6 people (15.7%) aged 60 years and older were subjected to some form of abuse. The pooled prevalence estimate was 11·6% (8·1–16·3) for psychological abuse, 6·8% (5·0–9·2) for financial abuse, 4·2% (2·1–8·1) for neglect, 2·6% (1·6–4·4) for physical abuse, and 0·9% (0·6–1·4) for sexual abuse [2]. New evidence indicates that the prevalence of elder abuse in both the community and in institutions has increased during the COVID-19 pandemic. A recent US study suggested that rates of elder abuse in the community may have increased by as much as 84% [3]. Worldwide, it is expected that the prevalence of elder abuse will increase exponentially as many countries have aging populations. This implies that the global number of victims of elderly abuse will increase even if the proportion of victims remains constant, growing to some 320 million victims by 2050 since the global population of people aged 60 years and older will increase to 2 billion by 2050 [1]. Lebanon’s older population faces a unique set of vulnerabilities stemming from the country’s ongoing socio-economic and political instability. Years of armed conflict, the 2019 financial collapse, the COVID-19 pandemic, and the Beirut port explosion have collectively exacerbated psychosocial and economic stressors [4]. Formal long-term-care capacity is minimal, and most facilities are private and prohibitively expensive [5]. Deep-rooted cultural norms further stigmatize institutional placement, compelling families to keep frail elders at home even when caregiving resources are inadequate [6]. This environment may inadvertently increase the risk of elderly abuse, particularly in multigenerational households or among caregivers facing economic hardship. A qualitative study of elder abuse in Lebanon conducted in 2021 explored the different perceptions of elder abuse among different stakeholders such as home and institution-based older adults, their family members and caregivers and showed that behaviors considered abusive among Lebanese participants were comparable to those reported in international studies. These included physical abuse; sexual overtures; a failure to protect an elder from harm, psychological or emotional stressors; and neglect. The most commonly mentioned risk factors were a history of abuse in a seemingly dysfunctional family followed by caregivers’ lack of acceptance of the physical changes that accompany aging [6]. Elder abuse has serious physical and mental health, financial, and social consequences, including, for instance, physical injuries, premature mortality, depression and cognitive decline, financial devastation, and placement in nursing homes [1]. For example, a 2009 study showed that elders who have been abused have a 300% higher risk of death, compared to those who have not been mistreated [7]. Various risk factors contribute to elderly abuse. The risk factors can be individual, relational, and societal. Individual risk factors include the elder having a current diagnosis of mental illness, drug or alcohol abuse, physical health problems, high levels of stress, social isolation, an experience of traumatic or disruptive behaviors, poor training for their caregivers, inadequate coping skills, and exposure to abuse as a child [8]. The relationship between the elder and the abuser can have risk factors as well including high financial and emotional dependence of the elder on the abuser or a past family conflict [8]. Societal risk factors include placement in a nursing home or residential facility, residential staffing shortage, staff burnout, lack of qualified staff, or stressful working conditions [8]. Given the limited yet compelling high rates of elderly abuse seen in some local and regional studies, further research should be conducted to elucidate the prevalence of elderly abuse in Lebanon. Despite qualitative insights, there are limited primary-care–based prevalence estimates from Lebanon. Evidence on modifiable predictors such as frailty status, caregiving relationships, and financial dependency in this setting is particularly scarce. This study addresses these gaps by estimating prevalence in a primary-care clinic and modeling independent predictors to inform screening and policy. The study also aimed to alert health policy makers to this problem and establish policies such as strengthening legal protection, enhancing social services and improving response and reporting systems. Additionally, we aimed to raise public awareness about elder abuse while simultaneously empowering older adults. Methods Study design, participants and setting We conducted a cross-sectional study of community-dwelling older adults aged 60 years and older who attended the outpatient department (OPD) clinic of the Department of Family Medicine at the American University of Beirut Medical Center (AUBMC) over a 12-month period. Exclusion criteria included severe uncorrected hearing impairment or aphasia, as assessed by the nurse during the invitation process; impaired decisional capacity, determined through a decision-making capacity assessment during the verbal consent procedure; participants or caregivers who did not speak or understand Arabic or English; and participants not identified as caregivers. Data collection Eligible participants, who presented to the Family Medicine OPD at AUBMC to seek medical care between March 2023 and February 2024, were identified and approached by a trained nurse using an invitation script. Interested participants were escorted to a private room, where verbal consent was obtained. The use of verbal rather than written consent was approved by IRB and done to ensure patient safety, for fear of retaliation by the caregiver if abuse was present. To ensure participant safety, IRB approved incomplete disclosure was used during the consent process. This was done to minimize response bias and obtain accurate prevalence estimates. Participants were informed that the study examines “social experiences among older adults.” This description accurately reflected the scope of the study while avoiding priming effects that may occur if elder abuse was explicitly highlighted at the outset. This approach is consistent with The Council for International Organizations of Medical Sciences (CIOMS) and the World Health Organization (WHO) ethical guidelines permitting incomplete disclosure when the research involves no more than minimal risk, the omission is necessary to maintain data integrity, and participants are not misled about procedures that could affect their willingness to participate. At the end of their participation, all individuals received a full debriefing explaining the specific focus on elder abuse, the rationale for the initial wording, and the importance of their contribution. They were also given the opportunity to withdraw their data. This strategy balanced the need for valid data with the ethical obligation to respect autonomy of the participants and maintain transparency. Sensitive topics such as abuse or assault were addressed with confidentiality emphasized. Data was collected anonymously, and any accompanying person was asked to remain in the waiting area during the interview. If the accompanying person was the primary caregiver, they were invited to participate in a separate interview after obtaining the older participant’s consent. If the caregiver declined, only the patient was included in the study. Assessment instruments Elder abuse was assessed using an interviewer-administered questionnaire adapted from the Elder Abuse Suspicion Index (EASI) [9], covering four WHO-defined domains: physical, psychological/emotional, neglect, and financial abuse. Cognitive function was measured using the Animal Naming Test (semantic verbal fluency), in which participants name as many animals as possible within 60 s—a widely validated and culture-fair measure of language and executive function [10]. Frailty was assessed using the FRAIL Scale [11], which includes five components: fatigue, resistance, ambulation, illnesses (≥ 5 chronic conditions or medications), and weight loss. Scores range from 0 to 5, with higher scores indicating greater frailty. Psychological well-being was evaluated using the WHO-5 Well-Being Index, a validated, five-item scale assessing subjective mental well-being [12]. For the caregiver component, selected questions were adapted from the Caregiver Abuse Screen (CASE) [13] to evaluate caregiver strain and the potential for mistreatment. All tools underwent forward–backward translation into Arabic and were reviewed by bilingual geriatric experts for semantic and cultural equivalence (Appendix I). Sample size We calculated sample size using a single population proportion formula, assuming a 95% confidence level. Since the prevalence of elder abuse in Lebanon was not known, we used a proportion of 43.7% as a proxy estimate [14], with a margin of error set at 5%. This estimate was derived from a rural Egyptian study, given the absence of Lebanese prevalence data at the time of design. Egypt was considered regionally and socio-culturally comparable, which justified its use for sample size estimation in our context [14]. Therefore, the sample size needed was calculated to be 379. Although our a-priori calculation indicated a required sample of 379 participants, enrollment was limited to 249 due to time and resource constraints, reducing study power and increasing the likelihood of Type II error, particularly for modest associations. Statistical analysis Descriptive analysis was performed by calculating frequency and percentages for categorical variables and mean and standard deviation for continuous variables. Multivariate logistic regression, and correlation analyses were performed using Pearson’s chi-square test or Cramér’s V for categorical variables, and t-tests/ANOVA for continuous predictors. Spearman’s correlation was used for continuous non-normally distributed variables. These correlation analyses were performed to identify risk factors for elder abuse, with P < 0.05 for statistical significance. Multicollinearity was assessed using variance inflation factors (VIF), and no variable exceeded the conventional threshold (VIF < 5), indicating acceptable levels. Missing data were handled via listwise deletion; no imputation was performed given the relatively small number of missing cases, except for ‘earnings’ where missingness was too high for meaningful modeling and thus excluded. Ethics approval and consent to participate This study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Ethical approval was obtained from the Institutional Review Board of the American University of Beirut (Approval ID: SBS-2022-0151). IRB approved verbal informed consent was obtained from all participants prior to the interview. Safety and ethical measures for abused older adults Older adult participants were invited to complete the questionnaire in a private room without the presence of any accompanying individuals, including caregivers. Older adult participants were assured that their identities would be kept confidential and there would be no adverse consequences if they chose to withdraw from the study. All participants were provided verbal informed consent during the interview; written informed consent was not obtained to protect all the participants’ privacy, given the sensitivity of the study topic. Moreover, participants were identified with codes to ensure participant anonymity. Older adult participants who screened positive for any form of abuse were offered a referral to the Social Services Department at AUBMC for further assessment and management. They were provided with the contact information for the department should they have any questions or need assistance. The contact number for social services at AUBMC was printed anonymously on the consent form given to the older adults participants, and it was explained to them how to safely contact the number in case the need arises. Moreover, contact information of participants who screened positive for any type of abuse and expressed strong emotional distress (example: crying, anger, panic attack) was shared with social services at AUBMC for active and close follow up. Participants who screened positive for poor psychological wellbeing, via the World Health Organization (WHO) Well-Being Index, were scheduled a follow-up appointment with their primary care provider, if consented, for further assessment and management. The Lebanese Ministry of Public Health (MOPH) does not currently have a reporting mechanism for cases of elder abuse. This was confirmed via direct contact with the Minister of Public Health. Furthermore, no non-governmental organizations (NGOs) were found to offer support to this at-risk population. In addition, healthcare professionals in Lebanon are not legally mandated to report cases of elder abuse to authorities. Results A total of 250 older adults were surveyed, one of whom tested positive for cognitive impairment and was subsequently excluded from the analysis. Of the remaining 249 participants, 49 (19.68%) were classified as having experienced at least one form of abuse. Psychological abuse was the most commonly reported type (n = 40), followed by neglect (n = 22), physical abuse (n = 6), and exploitation (n = 5). Participants were allowed to select more than one type of abuse. The mean age of the overall sample was 69.25 ± 7.28 years, while the mean age of those reporting abuse was 69.27 ± 7.32 years, showing no statistically significant difference (p = 0.96). Similarly, no significant differences were found in cognitive assessment scores (16.75 ± 4.76 vs. 16.89 ± 5.02; p = 0.35) or IADL scores (6.12 ± 1.69 vs. 6.14 ± 1.63; p = 0.71). A significantly higher proportion of females reported experiencing abuse compared to males (24.44% vs. 14.04%; p = 0.04). Differences in marital status, education level, monthly earnings, and employment status were observed but did not reach statistical significance (p = 0.17, p = 0.73, p = 0.77, and p = 0.06, respectively). The presence of financial support did not differ significantly between the abused and non-abused groups (p = 0.10), while the source of financial support was associated with reported abuse (p = 0.05). Participants receiving support from pensions or siblings were more likely to report abuse. Housing situation was also associated with reported abuse (p = 0.05), with individuals living in a family member’s or friend’s home (40%) reporting higher rates of abuse compared to those living in their own apartments (18.59%) or in rental units (19.68%). Factors such as number of cohabitants, smoking status, and alcohol intake were not significantly associated with elder abuse (p > 0.05 for all). As shown in Table 1, a significant association (p = 0.01) was found between caregiver identity and elder abuse. Older adults whose primary caregivers were paid helpers (50%) or daughters (41.67%) were more likely to report abuse compared to those cared for by a spouse (9.38%) or living alone (22.43%). Frailty status was also significantly associated with elder abuse (p = 0.003), with frail (28.95%) and pre-frail (24.8%) participants reporting higher rates of abuse than robust participants (8.14%). Similarly, participants with poor psychological well-being (34.29%) were more likely to report abuse compared to those with good psychological well-being (17.29%; p = 0.02). Table 1Associations between participant characteristics and elder abuse (Total n = 249; abused n = 49; not abused n = 200)VariableTotal (n = 249)Abused (n = 49)Not Abused (n = 200)p-valueAge (mean ± SD)69.25 ± 7.2869.27 ± 7.3269.24 ± 7.290.96Cognitive Assessment Score (mean ± SD)16.75 ± 4.7616.89 ± 5.0216.72 ± 4.700.35Cognitive Performance0.003* Low (< 13)7212 (17.0%)60 (83.0%) Moderate (13–18)8723 (26.0%)64 (74.0%) High (> 18)8914 (16.0%)75 (84.0%) IADL Score (mean ± SD)6.12 ± 1.696.14 ± 1.636.11 ± 1.710.71Sex0.04* Male114 (45.78%)16 (14.04%)98 (85.96%) Female135 (54.22%)33 (24.44%)102 (75.56%)Marital Status0.17 Single29 (11.69%)8 (27.59%)21 (72.41%) Married176 (70.97%)29 (16.48%)147 (83.52%) Widowed33 (13.31%)7 (21.21%)26 (78.79%) Divorced10 (4.03%)4 (40%)6 (60.00%)Education Level0.73 Illiterate16 (6.45%)2 (12.5%)14 (87.50%) Intermediate & below47 (18.95%)12 (25.53%)35 (74.47%) Secondary level88 (35.48%)17 (19.32%)71 (80.68%) University96 (38.71%)17 (17.71%)79 (82.29%)Currently Working0.06 Yes73 (29.32%)9 (12.33%)64 (87.67%) No176 (70.68%)40 (22.73%)136 (77.27%)Housing Situation0.05 Own apartment156 (62.65%)29 (18.59%)127 (81.41%) Family/friend’s apartment20 (8.03%)8 (40%)12 (60.00%) Rental apartment73 (29.32%)12 (19.68%)61 (80.82%)Frailty Status0.003* Frail38 (15.26%)11 (28.95%)27 (71.05%) Pre-frail125 (50.2%)31 (24.8%)94 (75.20%) Robust86 (34.54%)7 (8.14%)79 (91.86%)Psychological Well-Being0.02* Poor35 (14.06%)12 (34.29%)23 (65.71%) Good214 (85.94%)37 (17.29%)177 (82.71%)Primary Caregiver0.01* Spouse64 (26.78%)6 (9.38%)58 (90.62) Daughter24 (10.04%)10 (41.67%)14 (58.33) Son13 (5.44%)2 (15.38%)11 (84.62) Paid Helper8 (3.35%)4 (50%)4 (50.00) Alone107 (44.77%)24 (22.43%)83 (77.57)Categorical variables collapsed as shown in the methods (for example financial support: yes vs. no). Total N per group is reported in the table header* Statistical significance is set at p < 0.05 Associations between participant characteristics and elder abuse (Total n = 249; abused n = 49; not abused n = 200) Categorical variables collapsed as shown in the methods (for example financial support: yes vs. no). Total N per group is reported in the table header * Statistical significance is set at p < 0.05 Multivariate analysis Table 2and Fig. 1 present unadjusted and adjusted odds ratios for factors associated with elder abuse. The results indicate that age was not significant in the crude model (cOR = 1.00, p = 0.96) but emerged as a potential protective factor after adjustment (aOR = 0.87, p = 0.01), meaning that with each additional year of age, the odds of reporting abuse decreased by approximately 13%. Table 2Adjusted odds ratios for predictors of elder abuse (Total n = 249; abused n = 49; not abused n = 200)VariableAdjusted ORp-valueAge0.870.01*Sex (Female vs. Male)1.600.50Financial Support from Spouse (Yes vs. No)0.100.05Financial Support from Children (Yes vs. No)0.110.01*Housing Situation (Living in Family/Friend’s Home vs. Own Apartment)1.240.83Frailty Status (Robust vs. Frail)0.130.04*Psychological Well-Being (Good vs. Poor)0.480.32Primary Caregiver (Daughter vs. Spouse)17.480.01*Primary Caregiver (Son vs. Spouse)18.110.03** Statistical significance is set at p < 0.05 Adjusted odds ratios for predictors of elder abuse (Total n = 249; abused n = 49; not abused n = 200) * Statistical significance is set at p < 0.05 Forest plot of adjusted odds ratios (ORs) with 95% confidence intervals for predictors of elder abuse. The red dashed line at OR = 1 indicates no effect While male sex was significantly associated with elder abuse in the crude model (cOR = 1.98, p = 0.04), this relationship was no longer significant after adjustment (aOR = 1.60, p = 0.50), suggesting that gender alone may not independently influence the likelihood of reported abuse once other variables are considered. Housing situation also demonstrated an association with elder abuse risk. Older adults living in a family member’s or friend’s home had higher odds of reporting abuse compared to those living in their own apartment (cOR = 2.92, p = 0.03; aOR = 1.24, p = 0.83). Financial support from spouses (aOR = 0.10, p = 0.05) and children (aOR = 0.11, p = 0.01) appeared to be protective factors, reinforcing the role of financial security in mitigating vulnerability. Frailty status remained significantly associated with elder abuse: robust older adults were less likely to report abuse compared to frail individuals (aOR = 0.13, p = 0.04). Although better psychological well-being was associated with lower odds of abuse in the crude model (cOR = 0.40, p = 0.02), this relationship was no longer significant after adjustment (aOR = 0.48, p = 0.32), indicating that psychological distress may be correlated with but not necessarily causal for elder abuse. Caregiving relationships also showed a strong association: older adults cared for by a daughter (aOR = 17.48, p = 0.01) or a son (aOR = 18.11, p = 0.03) had significantly higher odds of reporting abuse compared to those cared for by a spouse. Discussion This study shows a substantial prevalence of elder abuse in Lebanon’s primary care setting, with nearly one in five older adults reporting at least one form of mistreatment. Psychological abuse emerged as the most common, aligning with global data from Yon et al. [2], which reported psychological abuse as the most prevalent globally at 11.6%. However, our estimate of 19.68% is higher than global pooled estimates, possibly due to the interplay of socioeconomic, cultural, and systemic factors unique to Lebanon. Age and reported abuse In adjusted models, each additional year of age was associated with lower odds of reported elder abuse (aOR = 0.87). Population-based studies have shown higher mistreatment prevalence among the “younger-old” (60–74 yrs) than the “older-old” (≥ 75 yrs). For example, the U.S. National Elder Mistreatment Study reported that “younger old respondents experienced significantly more verbal and financial abuse than their older old counterparts” [15], while the National Social Life, Health and Aging Project found a similar pattern for any family-perpetrated abuse [16]. A scoping review by Pillemer et al. likewise concluded that shared-household abuse “tends to peak in the early old-age cohorts” [14]. We interpret this association cautiously for two reasons. First, our cross-sectional design supports association, not causation. Second, the observed inverse age gradient is susceptible to selection and measurement biases, notably survivor bias. Older cohorts in a clinic-attending sample may be healthier or better supported and those at highest risk (e.g., severe neglect or violence) may be institutionalized, deceased, or not attending clinic, thereby under-represented. The oldest-old may be less willing or able to disclose mistreatment because of fear of repercussions, dependency on caregivers, cognitive/sensory limitations, cultural norms, or interviewer presence (reporting bias) [2, 17, 18]. Frailty and dependence as structural vulnerabilities A particular finding was the association between frailty and elder abuse. Abuse prevalence climbed from 8.1% in robust elders to 28.9% in those classified as frail, underscoring frailty as a structural vulnerability. Frailty increased the odds of abuse substantially, a result consistent with studies from Mexico City and China [19, 20]. Frailty denotes not just physical weakness, but a reduction in the physiological capacity to respond to stressors, often leading to functional dependence. From a theoretical standpoint, this places older adults in a structurally dependent position, intensifying power asymmetries in caregiving relationships. In the caregiver stress model, greater dependency by care recipients correlates with higher caregiver burden, which can escalate into neglect or abuse if not supported by formal services [21]. This mechanism is exacerbated in Lebanon where home caregiving is the norm due to cultural repulsion to institutional care and limited availability of elder facilities [6]. As the caregiving role intensifies, especially without formal training, unmet needs and chronic stress may manifest as psychological mistreatment or neglect. Psychological well-being and abuse: a bidirectional risk Poor psychological well-being was associated with higher abuse prevalence. This may result from a bidirectional pathway. Abuse can lead to depression, anxiety, and withdrawal [22], while poor mental health may impair older adults’ assertiveness or ability to resist or report mistreatment. Additionally, caregivers may interpret depressive or anxious symptoms as burdensome, thereby reinforcing negative interactions [15]. Salminen-Tuomaala et al. [23] found that mental illness among abused older adults reduced social participation, which can further perpetuate isolation and vulnerability. The compounded effects of financial insecurity, chronic illness, and reduced autonomy likely play into this cycle. Caregiver identity: intergenerational caregiving dynamics, stress, and abuse Caregivers’ identity played a substantial role in abuse risk. Adults cared for by sons or daughters had significantly higher odds of experiencing abuse than those cared for by spouses. This points to deeper structural and emotional tensions in intergenerational caregiving. The “intergenerational caregiving dynamics” between parent and child may generate resentment, especially in resource-scarce households. Daughters in particular, often expected to provide emotional and physical care, may experience “caregiver role strain” if also managing employment and childcare [24]. Paid caregivers also exhibited high abuse rates, possibly due to poor regulation and oversight. Unlike formal institutional settings, in-home caregiving often lacks monitoring, increasing the risk of mistreatment, whether due to neglect, emotional detachment, or exploitation [25]. Economic dependence and abuse risk Our findings reaffirm the well-established relationship between financial dependence and abuse [26]. Participants supported by children or spouses had lower odds of abuse, whereas those dependent on retirement income or distant relatives faced elevated risk. In Lebanon, the financial collapse has strained intergenerational households, with older adults sometimes perceived as economic burdens. These pressures may breed conflict, particularly when financial dependency is not matched with emotional mutuality or respect. Mosafer et al. [27] identified low financial literacy and lack of legal safeguards as key predictors of financial abuse. In this study, economic hardship combined with a lack of reporting infrastructure may mask the true prevalence of such abuse. Regional comparison: elder abuse in the Arab context A regional lens reveals both commonalities and distinct patterns. For example, abuse prevalence in Saudi Arabia has been estimated at 25–40% [24], and in Egypt, at 43.7% [28]. These rates, higher than Lebanon’s, may reflect differing methods, thresholds for reporting, or structural factors. However, the findings highlight the need for region-specific policy responses. Policy and practice implications Our findings highlight the urgent need for a national framework that addresses elder abuse within the health system. Several areas warrant attention, including systematic abuse screening in primary care, caregiver support, integration of mental health services, and stronger legal and financial protection for older adults. However, based on the predictors identified in this study, particularly frailty, financial dependency, and caregiver relationship, three immediate priorities emerge: Integrate frailty and abuse screening into routine primary care visits.Embedding brief, validated tools for frailty and elder abuse detection within existing primary care workflows would enable early identification of at-risk older adults. This can be implemented through the Family Medicine and community clinics with minimal resource expansion. Integrate frailty and abuse screening into routine primary care visits. Embedding brief, validated tools for frailty and elder abuse detection within existing primary care workflows would enable early identification of at-risk older adults. This can be implemented through the Family Medicine and community clinics with minimal resource expansion. Enhance caregiver training and support.Given the strong association between caregiver relationship and reported abuse, structured programs that address caregiver stress through respite services, mental health counseling, and subsidized training, could reduce the risk of mistreatment among dependent older adults. Enhance caregiver training and support. Given the strong association between caregiver relationship and reported abuse, structured programs that address caregiver stress through respite services, mental health counseling, and subsidized training, could reduce the risk of mistreatment among dependent older adults. Implement financial protection mechanisms for vulnerable elders.Because financial dependency emerged as a significant predictor, establishing safeguards such as mandatory co-signature for asset transfers, standardized banking protocols for older adults, and social support for low-income elders, may prevent economic exploitation and improve autonomy. Implement financial protection mechanisms for vulnerable elders. Because financial dependency emerged as a significant predictor, establishing safeguards such as mandatory co-signature for asset transfers, standardized banking protocols for older adults, and social support for low-income elders, may prevent economic exploitation and improve autonomy. These steps represent practical, evidence-informed entry points for improving elder protection in Lebanon. Broader reforms, including legal safeguards, remain essential but require longer-term policy development. Strengths and limitations A major strength of our study is its focus on primary care settings, which provides a unique perspective on elder abuse within a healthcare context. Additionally, the use of standardized tools for assessing frailty, cognitive status, and psychological well-being enhances the robustness of our findings. This study, while offering valuable insights into elder abuse in Lebanon, is subject to several limitations. First, its cross-sectional design restricts our ability to draw causal inferences. Recruiting 249 instead of the planned 379 may have limited our ability to detect smaller effect sizes. Although associations between frailty, caregiver identity, and abuse were statistically significant, the temporal direction of these relationships cannot be established. For example, while frailty may increase vulnerability to abuse, abuse itself could exacerbate frailty through neglect or psychological harm. Second, data were collected through self-reported interviews, which are inherently vulnerable to recall bias and social desirability bias given the sensitive nature of the questions. Cultural taboos surrounding family dynamics and abuse may have led participants to underreport mistreatment, particularly in a collectivist society where maintaining family honor is paramount. Furthermore, since the same trained team conducted all interviews, interviewer bias may have occurred through subtle differences in tone, rapport, or question delivery. These biases would most likely lead to underestimation of true abuse prevalence. Third, the sample was drawn from a single academic medical center in Beirut. This may limit generalizability to rural populations or individuals not engaged in the healthcare system. Older adults who attend outpatient services are likely more mobile, socially supported, and cognitively intact, potentially underrepresenting more vulnerable subgroups such as the homebound or those with severe cognitive impairment. Fourth, the study did not include detailed qualitative data that could have explained the contextual distinctions of abuse, including emotional dynamics, cultural pressures, and individual caregiving experiences. Lastly, although several factors were explored, including frailty, psychological health, and financial dependency, other relevant variables such as caregiver burden, elder personality traits, past trauma history, and social support networks were not systematically assessed. Future studies should adopt a more holistic, mixed-methods approach to capture these multidimensional influences. Conclusions This study provides critical insights into the prevalence and risk factors of elder abuse in Lebanon. Key findings highlight the role of frailty, psychological well-being, financial dependency, and caregiver identity in shaping abuse risks. Given these findings, targeted interventions should focus on strengthening caregiver support, enhancing financial security for older adults, and incorporating frailty assessment into routine geriatric care. Our findings emphasize the need for interventions from policy makers, such as the Lebanese MOPH, to set guidelines for mandatory screening of elder abuse in healthcare centers and develop a reporting mechanism for healthcare providers to report applicable cases. There is also a need for developing a support mechanism for this at-risk population as there are currently no organizations in Lebanon that offer that kind of service. Further longitudinal studies are needed to better understand causality and to develop effective prevention strategies for elder abuse. Supplementary Information Supplementary Material 1. Acknowledgements The authors express their sincere gratitude to all participants who contributed to this study. We also extend our appreciation to the nurses of the OPD clinic of the Family Medicine Department for their invaluable assistance in subject recruitment. The authors also recognize the assistance of ChatGPT, an AI language model developed by OpenAI, in revising and refining the manuscript. The model’s contributions were instrumental in enhancing the clarity, coherence, and overall quality of the paper. Abbreviations World Health Organization Outpatient department American University of Beirut Medical Center The Council for International Organizations of Medical Sciences Elder Abuse Suspicion Index Caregiver Abuse Screen The Lebanese Ministry of Public Health Non-governmental organizations Instrumental activities of daily living Adjusted odds ratio Crude odds ratio Authors’ contributions LS, MA, JU, UM and GA conceived the idea of the study. DM, JA, WC and ETC collected the data. RO conducted the data analysis and contributed to the results write-up. LS, MA, LA, RO and GA drafted the manuscript. LA and GA were responsible for the review and editing of the paper, in addition to supervising the overall project. All authors read and approved the final manuscript. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Data availability The datasets generated and analyzed during the current study are not publicly available to protect patients’ confidentiality but are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This study was approved by the Institutional Review Board at the American University of Beirut (Approval ID: SBS-2022-0151). Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Lewaa Shoujaa and Marwan Azzam contributed equally to this work. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The datasets generated and analyzed during the current study are not publicly available to protect patients’ confidentiality but are available from the corresponding author on reasonable request.