Authors: Lamees Salameh, Amer Shtaya, Nehaya Abu Ya’qoub, Omar Younis, Suha Hamshari
Categories: Research, Medication adherence, Health services accessibility, Humanitarian crisis, Chronic disease, Conflict-affected populations, Palestine
Source: BMC Primary Care
Authors: Lamees Salameh, Amer Shtaya, Nehaya Abu Ya’qoub, Omar Younis, Suha Hamshari
Diabetes (15% prevalence), hypertension (24%), and dyslipidemia (66.4%) are major health challenges in Palestine, with poor control rates (≤ 16.7%) due to low medication adherence. Chronic disease management is further hindered by socioeconomic barriers, treatment complexity, and geopolitical instability, including restricted healthcare access. This study assesses adherence levels and identifies key barriers; such as cost, psychological stress, and war-related disruptions, to inform targeted interventions in conflict-affected region.
This cross-sectional study recruited 423 adult patients with diabetes, hypertension, and/or dyslipidemia from West Bank primary health care centers (March–June 2025) via convenient sampling. Using the MARS-10 questionnaire, adherence (primary outcome) and barriers (cost, conflict-related access, side effects) were assessed. A two-stage sampling approach selected 3 governorates (north/center/south) and 6 clinics (central/peripheral). Data collection involved structured interviews for illiterate participants.
Apparent medication adherence was low, largely reflecting access-constrained treatment interruption due to medication shortages and health-system disruption rather than behavioral non-adherenc: only 10.3% for diabetes, 10.0% for hypertension, and 7.8% for dyslipidemia. Key barriers included high medication costs (44.4%), multiple daily doses (38.5%), war-related unemployment (37.6%), and psychological distress (depression: 14.7%). Age, education, and income significantly influenced adherence (p < 0.05), while war-specific disruptions (checkpoints, clinic closures) they were not independently associated with Forced treatment interruptionin adjusted analyses; however, war-related unemployment emerged as a significant predictor of poor adherence. Findings highlight the urgent need for targeted interventions addressing financial, regimen complexity, and psychosocial challenges in conflict zones.
Chronic illness medication adherence in conflict zones is low due to systemic, financial, and emotional barriers. Effective solutions require psychosocial support, simplified regimens, and culturally sensitive approaches. Further research should explore long-term impacts of trauma, health beliefs, and adaptive behaviors to develop context-specific interventions.
Diabetes is a major public health concern with a global prevalence of 10.5% in those aged 20–79-year-old in 2021 [1]. Diabetes alone is responsible for over a million deaths annually, ranking it as the ninth most common cause of death worldwide [2]. The prevalence of diabetes among the Palestinian population in the West Bank, Gaza, and East Jerusalem is significantly high, estimated at around 15%. However, various reports suggest that the actual rate could be even higher, potentially ranging from 18% to 21% [3].
Together with diabetes, hypertension substantially contributes to the burden of chronic disease in Palestine, and both conditions frequently coexist and require long-term pharmacological management. Hypertension is a major risk factor for premature death worldwide [4]. Hypertension is a common 31.1% of the global population [5] but usually under detected health problem that affects hundreds of millions of people around the world, two thirds of which are in middle- and low-income countries, nearly half of them are unaware of having this “silent killer”, and when aware are inappropriately treated [4]. Worldwide, 21% of patients with Hypertension are uncontrolled [4], making them at high risk for heart attacks, heart hypertrophy, and ultimately heart failure. There are several factors that can exacerbate the negative health effects of hypertension, they include smoking, dyslipidemia, Forced treatment interruptionto treatment plan, stress, etc [6].
Dyslipidemia is defined as abnormal lipid metabolism. It is the most well-known risk factor for coronary artery disease which in turn is one of the leading causes of death on global scale [6], and the number one killer in Palestinian territories, comprising about 31.5% of all deaths in 2019 [7].
Diabetes, Hypertension and dyslipidemia are common in Palestine (9.1%, 24% and 66.4% respectively) [8–10]. They are poorly treated with only 9.5% of hypertensive patients achieving target blood pressure measurements (< 140/90) [9], and only 16.7% of diabetic patients achieving good glycemic control with HbA1c ≤ 7% [11]. To the best of our knowledge, there is a scarcity of published data on dyslipidemia control in Palestine, and existing studies have largely focused on diabetes and hypertension rather than lipid management.
Adherence to medication is defined as the degree to which patients take their prescribed medications as directed by their medical professionals. In general, 40–60% of patients with chronic conditions have problems with adherence to a prescribed regimen [12], which leads to substandard clinical outcomes. Poor medication adherence in HTN has negative effects on health status, increasing the risk for cardiovascular events, poor renal outcome, and all-cause mortality [13].
Adherence to medication can be influenced by several determinants such as patient-related factors, treatment-related factors, healthcare system factors, socioeconomic factors [14, 15], and geo-political factors that are unique to Palestinian occupied territories were people suffer a military occupation with consequent limitations to health services access and health equipment and medicinal shortages. This political and military situation severely impact the ability of patients to adhere to their prescribed treatments, exacerbating existing health problems and leading to poorer health outcomes [16, 17].
Patients with diabetes, hypertension or dyslipidemia often exhibit varying adherence levels due to multiple barriers, such as forgetfulness, side effects and poor communication with health care providers [18, 19]. Treatment complexity; including sophisticated dosing schedules, polypharmacy, or specific administration instructions (e.g., taking with food), can further reduce compliance. Additionally, challenges arise when integrating therapy into daily life, especially with frequent dosing, conflicting instructions, or special storage needs. Cognitive impairments, low health literacy, and inadequate support systems may also inhibit adherence. The ongoing Gaza conflict contributes also to high levels of stress, anxiety, and depression which are considered strong psychological barriers to medication adherence [20]. In conflict-affected settings such as the West Bank, geopolitical instability creates a wide range of structural and psychosocial barriers that directly influence patients’ ability to adhere to long-term treatment. Mobility restrictions imposed by checkpoints and road closures can delay or prevent access to clinics and pharmacies, leading to missed appointments, medication shortages, or interruptions in prescription refills. Additionally, periodic clinic closures and disruptions in supply chains can reduce the availability of essential medications. The economic consequences of conflict—including unemployment, loss of income, and increased financial strain—further limit patients’ capacity to prioritize healthcare expenses. These structural challenges are compounded by psychological stress, anxiety, and uncertainty, all of which can reduce motivation, impair self-management, and undermine adherence to chronic disease treatment. Making these mechanisms explicit provides a clearer foundation for understanding the multifaceted ways in which geopolitical factors shape health behaviors in conflict settings.
In Palestine, there are comparatively few studies on medication adherence [21]. Filling this evidence gap is critical because the absence of context-specific adherence data limits the ability of policymakers and clinicians to design interventions that directly address the structural, socioeconomic, and conflict-related barriers affecting patients in the West Bank. By generating empirical data tailored to this unique setting, the study provides a necessary foundation for developing targeted, evidence-based strategies aimed at improving long-term disease management.The aim of our study is to estimate the prevalence of adherence to Hypertension, diabetes and Dyslipidemia medications, and to identify the barriers and associated factors influencing medication adherence.
A cross-sectional study among 423 patients recruited from Primary Health Care (PHC) facilities in the West Bank, Palestine, between March and June 2025. While clinic selection followed a stratified random process to enhance geographic representation, participant recruitment within clinics was based on convenience sampling due to feasibility constraints in a conflict-affected setting.
All adult Palestinians (≥ 18-year-old) who have Diabetes, HTN and or Dyslipidemia attending governmental PHC centers in the West Bank were selected.
The sample size for our study was estimated using the OpenEpi© Version 3, the OpenEpi Sample Size Calculator manufactured by the WHO [22]. It was computed using the following \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ n = [DEFFNp(1-p)]/ [(d^{2}/Z^{2}_{1-\alpha/2}(N-1) + p*(1-p)]
The calculation was based on an assumed prevalence of medication Forced treatment interruptionof 45.8%, as reported in a previous Palestinian study [21], with a precision level of 0.05 and a 95% confidence interval. Using these parameters, the minimum required sample size was estimated to be 382 participants. To account for potential non-response, a total of 423 participants were recruited. A two-stage stratified sampling approach was used to select study sites. In the first stage, the West Bank was stratified into three regions (north, center, and south), and one governorate from each region was randomly selected. In the second stage, two Ministry of Health PHC centers were selected from each chosen governorate (one central and one peripheral clinic), resulting in a total of six clinics. Within each selected clinic, convenience sampling was used to recruit participants. Eligible patients attending the clinics during the data collection period were approached in the waiting areas and invited to participate until the target sample size was reached.The number of governorates in the West Bank is 11, a list was made, and then one governorate from each region was randomly selected using a smartphone Random Generator Spin Wheel app. According to the Health Annual Report–Palestine 2021, the Ministry of Health operates 491 PHC centers, 439 of which are in the West Bank (North: 169 centers, Center: 113 centers, South: 157 centers) [23]. For each selected governorate, a list of governmental PHC centers was prepared. From this list, two PHC centers per governorate were the central PHC center and one peripheral PHC center selected randomly. Participants were recruited using a waiting-room convenience sampling approach from six Ministry of Health (MoH) primary healthcare clinics. Recruitment was conducted on regular clinic working days (Sunday to Thursday) during morning hours (8:00 am–2:00 pm), when caregivers commonly accompanied patients to scheduled appointments. During the recruitment period. This participant flow has been documented to improve transparency regarding recruitment and participation. In this conflict-affected setting, medication adherence as measured by the MARS-10 reflects both medication-taking behavior and the ability to access medicines. Because patients cannot take medications they do not possess, barriers related to drug availability, cost, and access were analyzed separately as structural determinants of adherence rather than In this conflict-affected setting, medication adherence as measured by the MARS-10 scale reflects both medication-taking behavior and the ability to access medicines. Because patients cannot take medications they do not possess, barriers related to medication availability, cost, clinic closure, and access constraints were considered structural determinants of adherence rather than behavioral non-adherence. This distinction is particularly important in humanitarian and conflict settings, where treatment interruption may be externally imposed. A strict cut-off (score = 10) was used to define optimal adherence. We acknowledge that this definition is conservative and may underestimate adherence in real-world settings, particularly in conflict-affected contexts.Convenience sampling was used to recruit participants within each selected center. Eligible adult patients who attended the clinics during the data collection period were approached, informed about the study, and invited to participate. Recruitment continued until the required sample size was reached.Measurement tools and data collection. Data were collected using a self-reported questionnaire that was distributed to participants in the waiting room of each clinic. For illiterate patients and those with reading difficulties, a trained data collector made the interview and wrote down their answers after gaining their informed consent. The questionnaire underwent a rigorous cultural adaptation process, including forward-backward translation, expert panel review by two senior family medicine specialists, and piloting among 20 patients. The development of the questionnaire was informed by the methodologies employed in the studies of [24, 25] with particular consideration given to the shared cultural context of the Arab region. The questionnaire covered four #### Part 1 Socio-Demographic characteristics of These Age, Sex, Marital status, Residential place (city {urban}, village or camp {rural}), occupation, education, monthly income, smoking status, Alcohol and drug abuse, and weight and height. #### Part 2 Associated factors affecting adherence to medication. These duration of the disease, complexity of medications, number of daily doses, coexisting conditions, frequency of clinic visits, distance to the clinic, medication side effects, and the availability of medications at the governmental pharmacy. #### Part 3 Assessment of adherence to medications using MARS-10; a 10-item, self-reported measure of adherence which also reflects the degree of compliance. This scale has proven its validity and reliability especially with treatment of long-term conditions including dyslipidemia, diabetes and hypertension [26] Items Q1–Q6, Q9, and Q10 are negatively worded and were scored as No = 1 and Yes = 0, reflecting non-adherent behaviors. In contrast, items Q7 and Q8 are positively worded items that assess favorable beliefs about medication; these were therefore scored as Yes = 1 and No = 0. Total scores range from 0 to 10, with a score of 10 indicating full adherence and scores below 10 indicating Forced treatment interruption. For linguistic validation, the questions were translated into Arabic and then backwards into the original language by an independent translator. #### Part 4 Barriers to adherence to medications were measured using the 14-item CULIG’s scale, a valid and reliable measure [27] that helps to estimate prevalence of different causes to Structural and access-related barriers to adherence. These include ( " I was not at home”, ” The drug was not available due to short supply”, " I take a number of drugs several times a day”, " The drug was too expensive”, " I was afraid of developing drug dependence”, “I felt well”, " I wanted to avoid side effects”, " I did not want other people seeing me taking drug”, “I consumed all of it”, “I had problems with medication timing”, " I felt depressed or broken”, " I was sleepy at medication time”, " My doctor frequently changes my therapy”, " I felt the drug to be toxic – harmful”). Another 5 structured questions were added to cover possible unique barriers to medication adherence during the ongoing conflict in Gaza and The West Bank; (“ I felt no need to take medications while I’m stressed by war and might die at any time”, " I’m not insured, so, I can’t buy my medications”, " I recently became unemployed because of war, so I can’t afford my medications”, “The clinic is closed most of the time, so I can’t refill my medications”, “I can’t reach out the PHC facility due to check points or blockades”). For linguistic validation, the original scale questions were translated to Arabic then backwards to English with the help of an independent translator. The Cronbach’s Alpha was calculated at 0.71, indicating acceptable internal consistency and plausibility of the questionnaire. ### Data analysis Data entry and all analyses were carried out using the Statistical Packages for Social Sciences (SPSS) for windows, version 27.0. The descriptive statistics described the study sample by showing the means, standard deviations and ranges for continuous variables while presenting proportions, frequencies and percentages for categorical variables. The 1^ry^ outcome is categorical, so comparison with other categorical variables was done using the Chi Square test, while comparison with continuous variables was done using the independent t-test. A p-value of less than 0.05 was considered statistically significant. ### Ethical consideration Participants were informed about the purpose of the study, their right to decline participation or withdraw at any time without consequence, and the assurance of confidentiality and anonymity of their responses. Ethical approval was obtained from the Institutional Review Board (IRB) at An-Najah National University (Ref: Med. Oct. 2024/23). The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki and the Belmont Report. For data confidentiality, no personal identifiers were collected, and all questionnaires were assigned anonymized numerical codes accessible only to the research team. Data were stored securely and used solely for research purposes. ## Results ### Sociodemographic and clinical characteristics of the study population A total of 423 adult patients diagnosed with diabetes, hypertension, and/or dyslipidemia participated in the study. As detailed in Table 1, the majority of participants were female (61.7%) compared to male participants (38.3%), with most aged between 55 and 64 years (39.5%), followed by those aged 65 years or older (31.4%). Most participants were married (78.7%), resided in villages (69.7%), and had either a basic education (32.9%) or secondary education (31.2%). More than half of the participants (54.8%) were classified as obese, while 34.9% were overweight. Only 10.1% had normal body weight, and 0.2% were underweight. A significant portion of the study population were unemployed (63.8%), and nearly all participants reported having health insurance (96.6%), the vast majority of whom were covered by governmental insurance. Almost half of the participants (49.2%) reported a monthly income of less than 1500 NIS, and 28.1% earned between 1500 and 2999 NIS per month. The prevalence of smoking among participants was relatively low (16.8%), and alcohol or drug use was negligible, with nearly all respondents denying alcohol consumption or illicit drug use. Table 1Sociodemographic characteristics of the study population (*N* = 423)VariableFrequency (%)Sex Male162 (38.3) Female261 (61.7)Age (years) 18–3917 (4.0) 40–54 106 (25.1) 55–64167 (39.5) ≥ 65133 (31.4)Marital status Married333 (78.7) Single/Widowed/Divorced90 (21.3)Place of residence Village295 (69.7) City125 (29.6) Camp/Other3 (0.7)Education level Illiterate52 (12.3) Basic139 (32.9) Secondary132 (31.2) Bachelor's or higher100 (23.6)Occupation Public Sector48 (11.3) Private Sector58 (13.7) Military4 (0.9) Retired43 (10.2) Unemployed270 (63.8)Health Insurance Insured (mainly governmental) 402 (96.6) Denominator varies due to missing responses Uninsured14 (3.4)Monthly Income (NIS) < 1500208 (49.2) 1500–2999119 (28.1) 3000–500066 (15.6) > 500030 (7.1)BMI Category Underweight1 (0.2) Normal42 (10.1) Overweight145 (34.9) Obese228 (54.8) Denominator varies due to missing responsesSmoking status Current or Former Smoker71 (16.8) Non-smoker352 (83.2)Alcohol use Yes4 (0.9) No419 (99.1)Drug use No423 (100) As shown in Table 2, nearly half of the participants (46.6%) reported having three chronic conditions, while 34.0% had two and 19.4% had only one. The majority had been diagnosed with hypertension (88.4%), diabetes (89.1%), and dyslipidemia (86.9%) for more than one year. Among patients with hypertension, 40.7% were taking two or more antihypertensive medications. Similarly, 68.3% of patients with diabetes were prescribed two or more antihyperglycemic agents, and 79.9% reported taking two or more daily doses. Medication side effects were least common among patients taking antihypertensive medications (8.0%), compared with antihyperglycemic (19.2%) and lipid-lowering medications (14.2%). Regarding medication availability, antihypertensive drugs were reported as unavailable by 40.9% of participants, while 47.2% of patients with diabetes and 45.3% of those with dyslipidemia reported lack of medication availability at clinics. Table 2Clinical characteristics related to diabetes, hypertension, and dyslipidemia#ItemFrequency (%)1How many chronic diseases do you have (e.g. Diabetes, HTN, Dyslipidemia, IHD)?Only one82 (19.4)Two144 (34)Three197 (46.6)Hypertension n = 3272When were you diagnosed with HTN?<1 year38 (11.6)≥ 1 y - <10 years176 (53.8)≥10 years113 (34.6)3How many antihypertensive drugs do you take?Only one medication194 (59.3)Two or more medications133 (40.7)4How many doses of your Anti-hypertensive medication do you take daily?Only once223 (68.2)Two or more104 (31.8)5Do you complain of anti HTN medication side effects?Yes26 (8)No301 (92)6Are your anti HTN medication(s) available at clinic?Yes72 (22.1)No133 (40.9)Sometimes120 (36.9)Diabetes n = 3127When were you diagnosed with T2DM?<1 year34 (10.9)≥ 1 y - <10 years165 (53.1)≥10 years112 (36)8How many antihyperglycemic drugs do you take?Only one medication98 (31.7)Two or more medications211 (68.3)9How many doses of your Anti-hyperglycemic medication do you take daily?Only once62 (20.1)Two or more246 (79.9)10Do you complain of anti-Diabetic medication side effects?Yes59 (19.2)No249 (80.8)11Are your anti Diabetic medication(s) available at clinic?Yes70 (22.7)No146 (47.2)Sometimes93 (30.1)Dyslipidemia n = 30912When were you diagnosed with Dyslipidemia?< 1 year40 (13.1)≥ 1 y - <10 years198 (64.7)≥10 years68 (22.2)13How many lipid lowering agents do you take?Only one medication278 (90.3)Two or more medications30 (9.7)14Do you complain of anti-Dyslipidemia medication side effects?Yes44 (14.2)No265 (85.8)15Is your lipid lowering agent(s) available at clinic?Yes74 (23.9)No140 (45.3)Sometimes95 (30.7)16How many times did you visit the clinic throughout the last year (since Gaza conflict)?Once every month147 (34.8)Once every 3 months91 (21.6)Once every 6 months49 (11.6)Once every year135 (32)17What is the distance (by car) for the clinic you follow up in?< 10 minutes139 (32.9)10-29 minutes213 (50.5)30- 59 minutes47 (11.1)>= 1 hour23 (5.5)18Do you -usually- stop your medications when you feel sick?Yes53 (12.6)No367 (87.4)Percentages are calculated based on the number of respondents for each question. Missing responses account for differences in total N ### Medication adherence rates Regarding medication adherence as assessed using the MARS-10 scale notably, approximately 85% of participants reported partial or complete inability to obtain their prescribed medications from governmental health centers. Under these conditions, low adherence scores likely reflect forced treatment interruption rather than intentional medication-taking behavio. Only 10.3% of patients with diabetes (32 out of 312) were classified as adherent to their anti-diabetic medications. Among patients with hypertension, 10.0% (32 out of 320) were adherent to their prescribed antihypertensive medications. In contrast, adherence among patients with dyslipidemia was even lower, with only 7.8% (24 out of 307) adhering to their lipid-lowering medications. To provide a fuller depiction of adherence behavior, we also examined the distribution of MARS-10 scores. The median (IQR) MARS-10 score was 8 (7–9) in diabetes, 8 (7–9) in hypertension, and 8 (7–9) in dyslipidemia. Additionally, the proportion scoring ≥ 9/10 was 37.5%, 36.5%, and 32.6%, respectively. Table 3 shows the responses to the MARS-10 questions. Table 3Responses to medication adherence rating scale questions (MARS-10)QuestionsDiabetesHypertensionDyslipidemiaFrequency (%)YesFrequency (%)NoFrequency (%)YesFrequency (%)NoFrequency (%)YesFrequency (%)NoDo you ever forget to take your medication?130 (41.7)182 (58.3)153 (46.8)174 (53.2)149 (48.1)161 (51.9)Are you careless at times about taking your medication?47 (15.1)265 (84.9)59 (18.1)267 (81.9)70 (22.5)241 (77.5)When you feel better, do you sometimes stop taking your medication?29 (9.3)283 (90.7)32 (9.8)294 (90.2)32 (10.3)279 (89.7)Sometimes if you feel worse when you take the medication, do you stop taking it?25 (8)288 (92)30 (9.2)296 (90.8)27 (8.7)283 (91.3)I take my medication only when I am sick21 (6.7)291 (93.3)27 (8.3)299 (91.7)24 (7.7)286 (92.3)It is unnatural for my mind and body to be controlled by medication52 (16.7)260 (83.3)47 (14.5)278 (85.5)44 (14.1)267 (85.9)My thoughts are clearer on medication141 (45.2)171 (54.8)133 (40.9)192 (59.1)124 (39.9)187 (60.1)By staying on medication, I can prevent getting sick205 (65.7)107 (34.3)211 (64.7)115 (35.3)203 (65.3)108 (34.7)I feel weird on medication36 (11.5)276 (88.5)33 (10.1)293 (89.9)37 (11.9)273 (88.1)Medication makes me feel tired and sluggish58 (18.5)255 (81.5)51 (15.7)273 (84.3)57 (18.3)254 (81.7) ### Factors associated with medication adherence Analysis of factors associated with adherence revealed statistically significant associations with several sociodemographic variables, as shown in Table 4. Age was significantly associated with adherence to antihypertensive medications (*p* = 0.017). Educational status was also significantly associ. ated with adherence to hypertension medications (*p* = 0.034). Income level was associated with adherence to diabetes (*p* = 0.048) and dyslipidemia medications (*p* = 0.040). No statistically significant associations were found between adherence and sex, residence, health insurance, or smoking status for any of the three conditions. Table 4Sociodemographic factors associated with medication adherenceDiabetesHypertensionDyslipidemiaAdherent FrequencyNot Adherent FrequencyP ValueAdherent FrequencyNot Adherent FrequencyP ValueAdherent FrequencyNot Adherent FrequencyP ValueAge0.9830.0170.92918-39190100440-5476566666355-6412111712010115≥ 65129519938101Education level0.5050.0340.597Illiterate536534138Basic Education810081058102Secondary Education158915781085University447360547Postgrad.08112011Monthly Income0.0480.1300.040< 1500 NIS2013422141171421500-2999 NIS6874893843000-5000 NIS246441143>5000 NIS413218314 As shown in Table 5, statistically significant associations were observed between disease duration and adherence for both hypertension (*p* = 0.011) and diabetes (*p* = 0.043), i.e. those who have hypertension or diabetes for 1 to 10 years were more likely to have adherence challenges. For dyslipidemia, the complaing of anti-Dyslipidemia medication side effects was significantly associated with adherence (*p* = 0.020). Other clinical factors, including the number of medications taken, presence of side effects, and medication availability at clinics, did not show statistically significant associations with adherence. In addition, no significant associations were found between adherence and the number of clinic visits in the past year, the distance to the clinic, or whether participants usually stopped their medications when feeling sick. Table 5Associations between clinical variables and medication adherenceQuestionsHypertensionAdherent FrequencyNot Adherent FrequencyP valueWhen were you diagnosed with HTN?0.011 < 1 year432 ≥ 1 y - <10 years13159 ≥10 years1495How many antihypertensive drugs do you take?0.427 Only one medication18171 Two or more medications14116How many doses of your Anti-hypertensive medication do you take daily?0.098 Only Once18199 Twice or more1488Do you complain of anti HTN medication side effects?0.257 Yes124 No31263Are your anti HTN medication(s) available at clinic?0.461 Yes763 No16113 Sometimes9109DiabetesAdherent FrequencyNot Adherent FrequencyP valueWhen were you diagnosed with T2DM?0.043 < 1 year231 ≥ 1 y - <10 years12151 ≥10 years1893How many antihyperglycemic drugs do you take?0.250 Only one medication890 Two or more medications24185How many doses of your Anti-hyperglycemic medication do you take daily?0.076 Only Once359 Twice or more29215Do you complain of anti-Diabetic medication side effects?0.809 Yes554 No27220Are your anti Diabetic medication(s) available at clinic?0.783 Yes861 No16129 Sometimes885DyslipidemiaAdherent FrequencyNot Adherent FrequencyP valueWhen were you diagnosed with Dyslipidemia?0.954 < 1 year337 ≥ 1 y - <10 years15179 ≥10 years662How many lipid lowering agents do you take?0.912 Only one medication22253 Two or more medications225Do you complain of anti-Dyslipidemia medication side effects?0.020 Yes044 No24236Is your lipid lowering agent(s) available at clinic?0.504 Yes666 No13125 Sometimes590 ### Structural and behavioral barriers to medication adherence Reported barriers were analytically classified into structural access-related barriers and behavioral adherence-related factors As shown in Table 6, among the 19 assessed barriers to medication adherence, the most frequently reported were high medication cost (44.4%), taking multiple medications daily (38.5%), being away from home at medication time (37.6%), and recent unemployment due to war (37.6%). Additional barriers included unavailability of medications (32.0%), intentional avoidance due to side effects (22.0%), and psychological factors such as feeling depressed or broken (14.7%) and being sleepy at medication time (17.7%). Table 6Frequency of Reported Barriers to Medication AdherenceI was not at homeFrequency (%)YesFrequency (%)No159 (37.6)264 (62.4)The drug was not available due to short supply135 (32)287 (68)I take a number of drugs several times a day163 (38.5)260 (61.5)The drug was too expensive188 (44.4)235 (55.6)I was afraid of developing drug dependence58 (13.7)365 (86.3)I felt well112 (26.5)309 (73.4)I wanted to avoid side effects93 (22)330 (78)I did not want other people seeing me taking drug31 (7.3)392 (92.7)I consumed all of it120 (28.4)303 (71.6)I had problems with medication timing91 (21.6)331 (78.4)I felt depressed or broken62 (14.7)361 (85.3)I was sleepy at medication time75 (17.7)348 (82.3)My doctor frequently changes my therapy36 (8.5)387 (91.5)I felt the drug to be toxic – harmful40 (9.5)383 (90.5)I felt no need to take medications while I’m stressed by war and might die at any time70 (16.5)353 (83.5)I’m not insured, so, I can’t buy my medications47 (11.1)376 (88.9)I recently became unemployed because of war, so I can’t afford my medications159 (37.6)264 (62.4)The clinic is closed most of the time, so I can’t refill my medications63 (14.9)360 (85.1)I can’t reach out the PHC facility due to check points or blockades79 (18.7)344 (81.3) Several barriers showed statistically significant associations with Forced treatment interruption (Table 7). For antihypertensive medications, significant associations were found with taking multiple medications daily (*p* = 0.012), feeling depressed or broken (*p* = 0.040), being sleepy at medication time (*p* = 0.041), and being away from home at medication time (*p* = 0.050). For diabetes medications, Forced treatment interruptionwas significantly associated with feeling depressed or broken (*p* = 0.040) and being sleepy at medication time (*p* = 0.015). In contrast, no individual barrier showed a statistically significant association with Forced treatment interruptionto dyslipidemia medications. War-related barriers were among the most common challenges reported by participants (Table 6). These recent unemployment due to war (37.6%), inability to reach primary healthcare centers because of checkpoints or blockades (18.7%), clinic closures (14.9%), feeling no need to take medications while under war-related stress and fear of imminent death (16.5%), and lack of insurance coverage (11.1%). None of the war-related factors showed a statistically significant association with medication adherence, despite being frequently reported (Table 7). Table 7Associations between barriers and medication adherence for diabetes, hypertension, and dyslipidemiaDiabetesHypertensionDyslipidemiaAdherent FrequencyNot Adherent FrequencyP ValueAdherent FrequencyNot Adherent FrequencyP ValueAdherent FrequencyNot Adherent FrequencyP ValueI was not at home0.1760.0500.123 Yes810861075107 No241722618219176The drug was not available due to short supply0.6930.8450.834 Yes991101018100 No231882218716182I take a number of drugs several times a day0.8500.0120.200 Yes1211561237124 No201652616617159The drug was too expensive0.3500.1380.673 Yes121341013110134 No201462215814149I was afraid of developing drug dependence 0.2790.0630.936 Yes241144337 No30239312450.83121246I felt well Yes8800.8367715760.634No241992521619206I wanted to avoid side effects0.8230.740.433 Yes6643693 60 No262162922021223I did not want other people seeing me taking drug0.4190.2460.384 Yes124021020 No312563226824263I consumed all of it0.4140.5380.064 Yes785780387 No251952520921195I had problems with medication timing0.2570.3550.081 Yes464462269 No282162822622213I felt depressed or broken0.0400.553 Yes1470.040150247 No312333123922238I was sleepy at medication time0.0410.273 Yes1580.015152253 No312223123722230My doctor frequently changes my therapy0.4970.686 Yes2270.752126220 No302533126322263I felt the drug to be toxic – harmful0.7530.805 Yes2260.752229228 No302543026022255I felt no need to take medications while I'm stressed by war and might die at any time0.441510.6230.777 Yes3444347 No292362823821236I'm not insured, so, I can't buy my medications0.5090.5800.295 Yes424536428 No282562725320255I recently became unemployed because of war, so I can't afford my medications0.550.0880.666 Yes711281198111 No251682417016172The clinic is closed most of the time, so I can't refill my medications0.6000.3140.549 Yes641744242 No262392524522241I can't reach out the PHC facility due to check points or blockades0.8170.9690.772 Yes756655454 No252242623420229 ### Multivariable analysis of factors associated with medication adherence In multivariable logistic regression analyses **(**Table 8), war-related unemployment was independently associated with lower medication adherence among patients with diabetes mellitus (aOR = 0.28, 95% CI: 0.10–0.79; *p* = 0.016) and hypertension (aOR = 0.25, 95% CI: 0.08–0.79; *p* = 0.018). Monthly income was also significantly associated with adherence in the hypertension and dyslipidemia groups; participants earning 1500–2999 NIS per month had lower odds of adherence compared with those earning < 1500 NIS (hypertension: aOR = 0.24, 95% CI: 0.07–0.80; *p* = 0.020; aOR = 0.20, 95% CI: 0.05–0.77; *p* = 0.020). No significant associations were observed for age, sex, educational status, disease duration, number of medications, dosing frequency, medication side effects, clinic availability, or other war-related barriers across the three conditions. Table 8Significant factors associated with medication adherence in multivariable logistic regressionVariableCategoryDiabetes Adjusted OR (95% CI)*p*-valueHypertension Adjusted OR (95% CI)*p*-valueDyslipidemia Adjusted OR (95% CI)*p*-value **Monthly Income (NIS)** ≤ 1500 (reference)1.00—1.00—1.00—1500–29990.399 (0.142–1.125)0.0820.236 (0.07–0.796) **0.020** 0.203 (0.053–0.774) **0.020** 3000–50000.231 (0.044–1.218)0.0840.706 (0.173–2.877)0.6280.134 (0.016–1.151)0.067> 50002.367 (0.537–10.425)0.2551.262 (0.207–7.694)0.8011.404 (0.279–7.061)0.681 **War-related unemployment** No (reference)1.00—1.00—1.00—Yes0.282 (0.101–0.787) **0.016** 0.246 (0.077–0.785) **0.018** 0.651 (0.207–2.052)0.464 Only variables with statistically significant associations (*p* < 0.05) in multivariable logistic regression models are presented. All models were adjusted for sociodemographic characteristics, disease-related factors, medication-related variables, and war-related barriers. ## Discussion This study found exceptionally low levels of medication adherence among patients with chronic illnesses living in conflict-affected areas. Only about 10% of participants were adherent to hypertension and diabetes medications, and adherence to dyslipidemia treatment was even lower, at under 8%. These figures are far below those typically reported in more stable environments, where adherence among chronic disease patients usually ranges between 50% and 60% [28]. They are also lower than recent figures from Gaza, where adherence to medications dropped from 85% to below 45% following the escalation of warfare since October 7th [29]. Pre-war studies in Palestine reported higher medication adherence and disease control than observed in the present study, with adherence estimates of approximately 40–60% and better glycaemic and blood pressure control during relatively stable periods. Although adherence was suboptimal even before the conflict, these data provide contextual benchmarks suggesting a further decline during the recent escalation Stirratt et al. (2018) and Al-Qasem & Smith (2011) [12, 28]. A critical methodological consideration in interpreting our findings is the distinction between behavioral non-adherence and forced treatment interruption due to health-system access failure. Classical adherence frameworks conceptualize non-adherence as patients’ medication-taking behavior among those who possess medicines. In our study, however, approximately 85% of participants reported incomplete access to prescribed medications due to shortages, financial constraints, unemployment, or clinic closure. Under these circumstances, the low adherence rates observed primarily reflect health-system collapse and access deprivation rather than intentional or behavioral non-adherence. Accordingly, our findings should be interpreted as access-constrained adherence within a conflict-affected health system. Beyond these immediate barriers, the conflict appears to have created both structural and psychological barriers that significantly affect patients’ ability to follow their treatment plans. Commonly cited challenges included the high cost of medications (reported by 44.4% of participants), the complexity of taking multiple medications daily (38.5%), and being away from home or asleep during scheduled dosing times (37.6% and 17.7%, respectively). These issues are consistent with findings from other regions and populations [30], yet the presence of ongoing conflict seems to magnify their impact. War-related factors such as job loss (37.6%), healthcare facility closures (14.9%), and emotional distress (14.7%) were frequently mentioned. While these specific barriers were not significantly associated with adherence in the regression analysis, this may be because such hardships affected nearly all participants, regardless of their adherence behavior an observation also noted in studies from other conflict zones like Syria and Yemen [31]. Psychosocial challenges also played a prominent role, especially among individuals managing hypertension and diabetes. Symptoms of depression, physical exhaustion, and disruptions to daily routines were closely linked with missed doses. These findings align with earlier research showing that psychological distress is a strong predictor of Forced treatment interruptionin chronic disease populations [30, 32]. Moreover, the burden of managing multiple medications (observed in over two-thirds of diabetic patients in this study) likely contributes to treatment fatigue, making it harder for patients to stay on track with their medications [30]. Sociodemographic factors also influenced adherence patterns. Older age and higher levels of education were linked to better adherence among hypertension patients, while higher income levels were associated with better adherence across all three conditions studied. These results are in line with earlier studies highlighting the role of financial stability and health literacy in promoting effective disease management [33, 34]. On the other hand, individuals with lower income or educational attainment often face added obstacles such as cost concerns and limited understanding of their treatment, both of which can reduce adherence, especially in unstable or resource-limited settings. Another key finding was the link between longer disease duration and lower adherence, particularly among patients with hypertension and diabetes. This suggests that while familiarity with treatment regimens may increase over time, the long-term demands of chronic disease management (especially under stressful conditions like war) can lead to treatment fatigue and disengagement [35]. These findings reflect broader behavioral theories. For example, the Health Belief Model proposes that an individual’s perception of illness severity, perceived barriers, and confidence in managing their condition all influence adherence. Similarly, the Burden of Treatment Theory suggests that the cumulative demands of managing chronic illness; including taking medications, attending appointments, and coping with side effects, can overwhelm patients, particularly in environments marked by instability and insecurity [36]. The very access-constrained adherenceobserved in this study can be explained using established behavioral frameworks. Within the Health Belief Model, high medication costs, regimen complexity, and psychological distress function as dominant perceived barriers that outweigh perceived benefits of long-term treatment in a conflict setting. Consistent with the Burden of Treatment Theory, multimorbidity and multiple daily doses increase treatment workload beyond patients’ coping capacity, leading to treatment fatigue and disengagement. The insurance–cost paradox, whereby most participants were insured yet nearly half reported cost barriers, reflects medication stock-outs, forced private purchasing, and war-related income loss that undermine the protective effect of insurance. Interestingly, although most participants reported having some form of health insurance, nearly half still identified medication costs as a major obstacle. This indicates that insurance alone is not sufficient to shield patients from out-of-pocket expenses; especially in areas where healthcare systems are under strain and local economies have been disrupted by conflict, as is the case in Palestine where blockades affected medicinal supply [37]. Taken together, these findings have clear implications for practice and policy in conflict-affected settings. Given that medication cost was the most commonly reported barrier, strengthening public-sector drug supply chains and reducing out-of-pocket expenditures should be policy priorities, as insurance alone does not ensure affordability or access. Clinically, the strong association between Forced treatment interruptionand regimen complexity highlights the need for treatment simplification strategies, including once-daily dosing, fixed-dose combinations, and regular medication reviews, particularly for patients with multimorbidity. In addition, the high prevalence of war-related unemployment and psychological distress underscores the importance of integrating social protection and mental health support into primary care to address the economic and emotional drivers of Reasons for non-adherence. ### Limitations This cross-sectional study design prevents establishing causal relationships. The results are specific to a population under active conflict, and therefore may not be generalizable to patients in more stable settings. participants were recruited using a waiting-room convenience sampling strategy from MoH clinics, which may have led to the under-representation of other pateints. particiapnts who were unable to travel due to illness severity, financial constraints, or movement restrictions, as well as those receiving care in private or UNRWA healthcare facilities, were less likely to be included. Consequently, the experiences of the most severely affected particiapnts may not be fully captured. Additionally, the data in this study were self-reported, which may be subject to recall bias and social desirability bias, potentially affecting the accuracy of the responses. Although validated adherence scales were used, their application in a conflict setting may conflate behavioral non-adherence with forced treatment interruption due to access failure. ## Conclusion Medication adherence among individuals with chronic illnesses in conflict-affected regions is notably low, influenced by a combination of systemic challenges, financial instability, emotional strain, and complex treatment regimens. Addressing this issue requires more than just ensuring medication availability; it calls for integrated approaches that include psychosocial support, consideration of patient beliefs, simplified treatment plans, and context-sensitive delivery mechanisms. Further longitudinal and qualitative studies are essential to explore how conflict-related trauma, health perceptions, and coping behaviors evolve over time, and to guide the development of effective, locally relevant interventions.