Frailty and multimorbidity among adults aged 40–69 years in Montenegro: a cross-sectional pilot study Authors: Isidora Rubezic Lukic, Natasa Popovic, Ajša Rašljanin Kalač, Аndrea Radulović Published: 2025 ABSTRACT Abstract AbstractObjectivesThis pilot study aimed to estimate the prevalence of frailty, prefrailty and multimorbidity among adults aged 40–69 in Montenegro and to assess the feasibility of measuring these conditions in primary care to inform the design of a future large-scale study. We also aimed to examine differences in frailty scores and chronic disease burden across age subgroups, describe the profiles of prefrail/frail versus robust individuals and contextualise the findings within a broader public health framework.DesignCross-sectional pilot study.SettingPrimary healthcare centre, Podgorica, Montenegro.ParticipantsA total of 165 adults aged 40–69 were recruited during routine follow-up visits for chronic conditions or preventive check-ups with their primary care physicians. Patients with acute conditions or injuries were excluded.Primary and secondary outcomesPrimary outcomes were the prevalence of frailty and prefrailty (based on an adapted Fried frailty phenotype) and multimorbidity (defined as ≥2 coexisting chronic conditions). Secondary outcomes included comparisons of frailty scores and the number of chronic conditions across age subgroups (40–54, 55–64, 65–69), as well as sociodemographic and clinical differences between prefrail/frail and robust participants. Historic national mortality and healthcare utilisation trends since 2012 were also analysed.ResultsFrailty and prefrailty prevalence was 6.1% (95% CI 2.9% to 10.8%) and 46.7% (95% CI 38.8% to 54.5%), respectively, while multimorbidity affected 23% of participants (95% CI 16.8% to 30.2%). The number of chronic diseases and frailty score increased significantly across age groups (p<0.001 and p=0.003, respectively), with a marked increase in number of chronic diseases observed already after age 55. Participants categorised as prefrail or frail had higher body mass index (p=0.031) and more frequently presented with coronary artery disease (p=0.01) and multimorbidity (p=0.027) compared with robust individuals. Retrospective analyses showed a change in patterns of healthcare utilisation during the COVID-19 pandemic, accompanied by increased mortality among adults aged ≥55. This highlights the vulnerability of this age group and the fragility of the Montenegrin healthcare system during crises.ConclusionThis pilot study demonstrates that assessing frailty, prefrailty and multimorbidity in primary care settings in Montenegro is feasible and provides essential insights for designing a future large-scale study. Our preliminary evidence revealed a considerable proportion of prefrail individuals and a smaller proportion of frail individuals, as well as a notable burden of multimorbidity among adults aged 40–69 in Montenegro, particularly among those aged ≥55. This group could benefit from more timely preventive interventions at the primary care level, early detection of prefrailty, frailty and multimorbidity, which would ultimately enhance the resilience of the national healthcare system.Keywords: Frailty, Multimorbidity, Prevalence, Primary Health Care STRENGTHS AND LIMITATIONS OF THIS STUDY This cross-sectional pilot study was conducted in a real-world primary care setting, during routine daily practice. This cross-sectional pilot study was conducted in a real-world primary care setting, during routine daily practice. Frailty was assessed using a validated, adapted version of the Fried phenotype aligned with methods used in large population-based cohorts such as the UK Biobank. Frailty was assessed using a validated, adapted version of the Fried phenotype aligned with methods used in large population-based cohorts such as the UK Biobank. Complete data collection with no missing variables supports the feasibility of the research procedures. Complete data collection with no missing variables supports the feasibility of the research procedures. The small sample size and recruitment from a single geographical region may limit the generalisability of the findings. The small sample size and recruitment from a single geographical region may limit the generalisability of the findings. The cross-sectional design precludes assessment of temporal trends and causal relationships. The cross-sectional design precludes assessment of temporal trends and causal relationships. Introduction Frailty and multimorbidity are increasingly recognised as major challenges in ageing populations. Frailty, marked by diminished physiological reserves and increased vulnerability to stressors, and multimorbidity, defined as the coexistence of two or more chronic conditions, are particularly prevalent among adults aged 65 and older.1 2 These interrelated conditions carry significant implications for healthcare systems worldwide. Frailty is strongly associated with adverse outcomes such as increased mortality and hospitalisations, particularly when accompanied by multimorbidity.3,5 However, frailty, prefrailty and the rising prevalence of chronic conditions like cardiovascular disease, diabetes and cancer can manifest earlier, during midlife, a transitional period marked by the onset of accelerated functional decline.5,8 Notably, over half of individuals with multimorbidity are under the age of 65.9 Despite this, research on frailty and multimorbidity in middle-aged individuals remains limited, leading to their under-recognition in this demographic.410,17 This is particularly relevant in regions with lower socioeconomic development, where frailty and multimorbidity often emerge a decade earlier than in high-income settings.9 18 Montenegro, a middle-income country in Southeast Europe, is undergoing a demographic transition marked by population ageing and a significant burden of chronic non-communicable diseases (NCDs).19 20 According to national statistics, adults aged 40–64 represent approximately one-third of the total population (33%).19 Many of them are facing health challenges such as hypertension (HTN), diabetes, cardiovascular disease and cancer, alongside behavioural risk factors such as smoking, physical inactivity and overweight.20 Although Montenegro shows comparable prevalence of NCDs and risk factors to those of high-income countries, differences exist in socioeconomic conditions and healthcare system capacity to resist disruptions. For comparative context, in table 1, Montenegro’s healthcare and sociodemographic profile is placed alongside the UK’s, a well-resourced, high-income system more resilient to disruptions. Lower average income and educational attainment contribute to increased health vulnerability.19 The healthcare system in Montenegro provides near-universal coverage through a three-tiered structure, but the country has fewer physicians and nurses per capita than European Union benchmarks.20 Despite population ageing and growing health needs, the number of hospital beds has slightly declined over the past two decades, indicating stagnation in infrastructure development. Preventive services and screening programmes remain underused, contributing to the continued high burden of NCDs.20 Table 1Sociodemographic and health context of middle-aged population in Montenegro and UK(1) PopulationMontenegroUKTotal population in 2023, absolute number, (%)623 633 (100)1968 265 209 (100)56Total population in 1991615 035 (100)1957 438 700 (100)57Aged 40–64 in 2023, absolute number, (%)208 630 (33.4)1921 819 280 (32)56Aged ≥65 in 2023, absolute number, (%)105 008 (16.8)1912 924 166 (18.9)56Life expectancy in 2023 (years)77.65881.458(2) Socioeconomic indicators in 2023–2024Education attainment, %Ages 15–64Ages 25–64Primary school and less20.85917.860Secondary52.75929.560Higher (Bachelor, Master, Doctoral level)26.25952.760Unemployment rate, %12.2584.258GDP per capita, $12 9455853 19558World Bank income groupUpper-middle61High61(3) Healthcare system and access to primary care in 2024Universal healthcare coverageYes20Yes62Out of pocket spending, %3820Low62People unable to afford medical care, %0.920Low62Number of physicians per 100 000 of population (WHO European Region=389)27520<30062Number of hospital beds per 100 000 of population (WHO European Region=470)3812023062Childhood vaccination rates, % for measles containing vaccine in 202233209062Doctors—general practice/family medicine, %18.42020.462Urban/rural distribution*69% urban6385% urban63(4) Burden of chronic diseases and risk factorsPremature mortality due to major chronic diseases, %20641164Overweight/obesity, %82.76491.564Elevated blood pressure, %29.16415.264Tobacco use contributing to mortality, % of all deaths13.42011.162Major non-communicable diseases=cardiovascular disease, cancers, diabetes and chronic obstructive pulmonary disease; overweight=BMI ≥25 m2/kg; obesity=BMI ≥30 m2/kg, elevated blood pressure=systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg.*Relevant to access to care and risk factor distribution.BMI, body mass index; GDP, gross domestic product. Sociodemographic and health context of middle-aged population in Montenegro and UK Major non-communicable diseases=cardiovascular disease, cancers, diabetes and chronic obstructive pulmonary disease; overweight=BMI ≥25 m2/kg; obesity=BMI ≥30 m2/kg, elevated blood pressure=systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg. Relevant to access to care and risk factor distribution. BMI, body mass index; GDP, gross domestic product. Given these demographic, socioeconomic and healthcare system challenges, addressing the needs of middle-aged adults becomes an urgent priority. This population, often overlooked in frailty and multimorbidity research, may represent a critical window for early detection and intervention, offering an opportunity to delay or prevent disease progression and reduce long-term healthcare system strain. The aim of this pilot study was to estimate the prevalence of frailty, prefrailty and multimorbidity among adults aged 40–69 in Montenegro and to evaluate the feasibility of assessing these conditions in primary care as a basis for a larger study. Additional objectives were to examine differences in frailty scores and chronic disease burden across age subgroups and to compare sociodemographic and clinical characteristics of prefrail/frail versus robust participants, as well as to contextualise the findings within national mortality and healthcare-utilisation patterns. Methods Study design and recruitment of participants We recruited patients aged 40–69 who were attending routine follow-up appointments for chronic conditions, or preventive check-ups with their primary care physicians at the primary healthcare centre in Podgorica. Patients with acute conditions or injuries were excluded from the study. Recruitment was conducted by two physicians concurrently engaged in the pilot study at any given time over a 12-month period, from June 2024 to June 2025. Although two physicians were engaged at a time, they were not always the same individuals; in total, five different physicians participated in the recruitment process over the course of the study. The recruitment was performed in consecutive order—participants who qualified according to recruitment criteria were invited to participate, and those who accepted were recruited. The sample size was calculated using the standard formula for estimating proportions, n=Z2×P×(1−P)/d2, where Z is the Z-score for the desired confidence level (1.96 for 95%), P is the expected prevalence and d is the acceptable margin of error. No national data are available on the prevalence of frailty, prefrailty or multimorbidity among middle-aged adults in Montenegro, while studies from other populations report considerable variation in prevalence estimates for all three outcomes.59 21,23 To ensure adequate precision, we conservatively assumed a prevalence of 50%, which yields the maximum required sample size for a given margin of error. Using a 95% confidence level and a 10% margin of error, the minimum required sample size was calculated to be 96 participants. A 10% margin of error is generally considered acceptable for pilot studies aiming to estimate proportions.24 25 To improve the reliability of estimates, enable subgroup comparisons and account for potential exclusions, we aimed to recruit at least 96 participants. The age range of 40–69 years was selected to ensure compatibility with the UK Biobank data set and Hanlon’s study on frailty in middle age, which was based on participants from the UK Biobank cohort.5 The UK Biobank specifically recruited individuals aged 40–69 years at baseline, as stated on the official UK Biobank website.26 We aligned our inclusion criteria with those of the UK Biobank cohort to enable meaningful comparisons and enhance the global interpretability of our pilot cohort data from Montenegro. Frailty status was assessed for each participant, and data on relevant chronic diseases was documented. Additionally, demographic and lifestyle information, including sex, age and smoking status, was collected, along with anthropometric measurements such as height, weight and body mass index (BMI). Patient and public involvement Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research. Frailty assessment Frailty was assessed using an adapted Fried frailty phenotype following the method described by Hanlon et al,5 who adapted the original phenotype proposed by Fried et al.3 The original Fried phenotype includes five criteria: self-reported unintentional weight loss in the last year (>10 pounds), self-reported exhaustion, weakness (measured grip strength, adjusted for sex and BMI), slowness (gait speed measured over 15 feet) and low physical activity (quantified via Minnesota Leisure Time Physical Activity Questionnaire).3 The adaptation enables alignment with previous large-scale population-based studies (eg, UK Biobank) while maintaining feasibility in primary care settings. The adapted method also uses five criteria: Exhaustion – Self-reported: ‘Over the past two weeks, how often have you felt tired or had little energy?’ (response: more than half the days or nearly every day=1, other=0). Exhaustion – Self-reported: ‘Over the past two weeks, how often have you felt tired or had little energy?’ (response: more than half the days or nearly every day=1, other=0). Unintentional weight loss – Self-reported: ‘Compared with one year ago, has your weight changed?’ (response: yes, lost weight=1, other=0). Unintentional weight loss – Self-reported: ‘Compared with one year ago, has your weight changed?’ (response: yes, lost weight=1, other=0). Low physical activity: Self-reported physical activity level in the past 4 weeks (response: none or light activity with a frequency of once per week or less=1, medium or heavy activity or light activity more than once per week=0). Low physical activity: Self-reported physical activity level in the past 4 weeks (response: none or light activity with a frequency of once per week or less=1, medium or heavy activity or light activity more than once per week=0). Slowness – Self-reported: ‘How would you describe your usual walking pace?’ (response: slow=1, other=0). Slowness – Self-reported: ‘How would you describe your usual walking pace?’ (response: slow=1, other=0). Weakness: Grip strength measured by digital Camry dynamometer EH101, graded by sex-specific and BMI-specific cut-offs (1 point). Grip strength was measured for both hands, and the higher value was used for analysis. A study by Guerra et al27 confirmed reliability and concurrent validity of the Camry dynamometer compared with the Jamar dynamometer (the gold standard), indicating that the Camry device is suitable for frailty assessments involving grip strength. Weakness: Grip strength measured by digital Camry dynamometer EH101, graded by sex-specific and BMI-specific cut-offs (1 point). Grip strength was measured for both hands, and the higher value was used for analysis. A study by Guerra et al27 confirmed reliability and concurrent validity of the Camry dynamometer compared with the Jamar dynamometer (the gold standard), indicating that the Camry device is suitable for frailty assessments involving grip strength. Participants were subsequently categorised as non-frail (robust) (0 points), prefrail (1–2 points) or frail (≥3 points). Assessment of the presence of chronic diseases and multimorbidity The presence of common NCDs, which significantly contribute to global mortality,28 was recorded using a questionnaire completed jointly by the patient and their physician. Physicians also accessed the patient’s electronic medical records to ensure a comprehensive overview of chronic conditions. The following NCDs were documented: HTN, diabetes mellitus type 1 and 2, coronary artery disease (CAD), chronic obstructive pulmonary disease, a history of stroke, dementia and information on the three most prevalent cancers: prostate, lung and colorectal cancer in men, and breast, lung and colorectal cancer in women. Multimorbidity was assessed using two approaches. First, the total number of chronic conditions present in each patient was recorded as a quantitative measure. Second, a qualitative approach was used to determine the presence or absence of multimorbidity, with patients classified as having multimorbidity if they had two or more coexisting chronic conditions.2 Retrospective descriptive analysis of national mortality rates and utilisation of healthcare To analyse mortality trends, we used publicly available data from the Statistical Yearbook of the Montenegro Statistical Office (MONSTAT).19 From this source, we extracted annual mortality rates for age groups categorised in 5-year intervals for the period from 2012 to 2023. MONSTAT defines the death rate as the number of deaths in relation to the estimated midyear population, expressed per 1000 inhabitants. Healthcare utilisation for the period 2012–2021 was analysed using publicly available data from the Statistical Yearbooks on Population Health and Healthcare in Montenegro, published every 5 years by the Institute of Public Health of Montenegro.29 These yearbooks provided information on the number of initial and follow-up visits to primary healthcare facilities for specific disease groups, including infectious diseases, tumours, endocrine, nutritional and metabolic disorders, cardiovascular diseases and neurological conditions. Additionally, data on hospitalisations in secondary and tertiary levels of healthcare for these disease groups were included. Since the statistical publications did not include information on associated comorbidities, only the primary diagnosis for each hospitalisation was considered. The number of primary healthcare visits and hospitalisations was expressed per 100 000 of Montenegro’s total population, based on the 2011 Census data. To facilitate interpretation of patterns of healthcare utilisation, the percentage of hospital visits out of total healthcare visits (primary care and hospital together) was calculated for each year. Statistics Descriptive statistics was used to summarise cross-sectional data: categorical variables were expressed as percentages, while quantitative variables were presented as medians and SD. Prevalence estimates for frailty, prefrailty and multimorbidity were accompanied by 95% CIs calculated using the Clopper-Pearson method. Data on mortality and healthcare utilisation was also analysed descriptively. The age distributions of the entire Montenegrin population and our study population were compared using the two-tailed t-test. The normality of numerical variables was assessed using the Shapiro-Wilk test. The cross-sectional sample was divided into three age groups (40–54, 55–64, 65–69), and group differences in the number of chronic conditions and frailty score were assessed using the Kruskal-Wallis non-parametric test, followed by pairwise post hoc comparisons (Dunn’s test with Bonferroni adjustment). To compare the prefrail/frail group with the robust participants, the χ² test was applied for categorical variables, and the Mann-Whitney U test was used for numerical variables. Where expected cell counts were low, Fisher’s exact test was used as an alternative to the χ² test. Only conditions with five or more cases were included in group comparisons using χ² or Fisher’s exact tests. P values <0.05 were considered statistically significant. Statistical analyses of cross-sectional data were conducted using SPSS software (V.30.0.0). Line graphs for mortality and healthcare utilisation trends were created in LibreOffice V.7.6.7.2 and online application Drawio.30 Results Patients’ sociodemographic and clinical characteristics Out of 170 patients initially screened, 5 were excluded due to not meeting the age criteria (2 were younger than 40 and 3 were older than 69), resulting from a recruitment error. The final study sample included 165 participants. No participants had missing data. Among recruited participants, 30.3% were male and 69.7% female, with a median age of 56 years (table 2). Table 2 further summarises the sociodemographic and clinical characteristics of the cross-sectional study population. Regarding frailty status, most participants were categorised as either frail (6.1%; 95% CI 2.9% to 10.8%) or prefrail (46.7%; 95% CI 38.8% to 54.5%). Overall, 23.0% of participants (95% CI 16.8% to 30.2%) were identified as having multimorbidity, defined as the presence of two or more chronic conditions. Table 2Sociodemographic and clinical characteristics of participantsVariable, unit of measurementAge, median±SD56±7.9Sex, absolute value, (%)* Male50 (30.3) Female115 (69.7)Smoking, absolute value, (%) People who currently smoke64 (38.8) People who have quit smoking40 (24.2) People who have never smoked61 (37)BMI, absolute value, (%) BMI ≤18.52 (1.2) BMI >18.5–2572 (43.6) BMI >25–3067 (40.6) BMI >3024 (14.5)Frailty, absolute value, (%) Non-frail (score 0)†78 (47.3) Prefrail (score 1–2)77 (46.7) Frail (score 3–5)10 (6.1)Number of frailty indicators present in a patient, absolute value, (%) 0 indicators78 (47.3) 1 indicator48 (29.1) 2 indicators29 (17.6) 3 indicators8 (4.8) 4 indicators2 (1.2) 5 indicators0 (0)Frailty indicators, absolute value, (%) Fatigue58 (35.2) Weight loss28 (17) Low physical activity15 (9.1) Slowness27 (16.4) Weakness10 (6.1)History of chronic diseases, absolute value, (%) HTN95 (57.6) DM28 (17) CAD14 (8.5) Stroke4 (2.4) Cancer9 (5.5) Dementia1 (0.6) COPD2 (1.2)Number of coexisting chronic diseases, absolute value, (%) 0 conditions62 (37.6) 1 condition65 (39.4) 2 conditions29 (17.6) 3 conditions7 (4.2) 4 conditions2 (1.2)Multimorbidity,‡ absolute value, (%) Yes38 (23) No127 (77)*Per entire group of patients, (n=165).†Per adapted version of the Fried frailty phenotype.‡Multimorbidity is present if the patient is concurrently diagnosed ≥2 chronic conditions.BMI, body mass index; CAD, coronary artery disease; COPD, chronic obstructive pulmonary diseases; DM, diabetes mellitus type 1 and 2; HTN, hypertension. Sociodemographic and clinical characteristics of participants Per entire group of patients, (n=165). Per adapted version of the Fried frailty phenotype. Multimorbidity is present if the patient is concurrently diagnosed ≥2 chronic conditions. BMI, body mass index; CAD, coronary artery disease; COPD, chronic obstructive pulmonary diseases; DM, diabetes mellitus type 1 and 2; HTN, hypertension. Analysis of the age distribution of the people in the cross-sectional study and corresponding population from retrospective analysis shows that there is no significant difference between these two populations (figure 1). Comparison of the age distribution in the cross-sectional pilot study with historical data on general population in Montenegro based on 2023 census. The values are expressed as a percentage of total number of people aged 40–69 for the respective group, t-test, p=1.00. Number of chronic diseases and frailty scores by age group A significant increase in the median number of chronic diseases occurs already at age ≥55 (40–54 vs 55–64: p<0.001; 55–64 vs 65–69: p=0.004; overall p<0.001) (figure 2a). Additionally, the median frailty score also increased with age, with a statistically significant difference observed between the youngest and the oldest age group (40–54 vs 65–69: p=0.005; overall p=0.003) (figure 2b). Number of chronic diseases and frailty score by age groups. (a) Median number of chronic diseases increases significantly already at age 55 (age groups 40–54 vs 55–64 vs 65–69: median±SD=0±0.63 vs 1±0.78 vs 2±1.07 with overall p<0.001). (b) Median frailty score in the youngest group is significantly lower than the oldest group of participants (age groups 40–54 vs 55–64 vs 65–69: median±SD=0±0.69 vs 1±0.99 vs 2±1.26 with overall p=0.003). Asterisk denotes statistically significant pairwise comparisons with p<0.05. Differences in sociodemographic and clinical characteristics between prefrail/frail and robust participants Sociodemographic and clinical characteristics of prefrail/frail versus robust participants are presented in table 3. The prefrail/frail group had a significantly higher BMI (27±4.2 vs 25±4.2, p=0.031). The prevalence of CAD (7.3% vs 1.2%, p=0.010) and multimorbidity (15.8% vs 7.3%, p=0.027) was also significantly higher in this group (table 3). Table 3Sociodemographic and clinical characteristics across different frailty statusesVariable, unit of measurementPrefrail/frailn=87Robustn=78P value*Sex, absolute value, (% male)28 (17)22 (13.3)0.579Age, median±SD58±8.254±50.085Smoking ever, absolute value, (% yes)57 (34.5)47 (28.5)0.485Smoking current, absolute value, (% yes)33 (20)31 (18.8)0.811BMI, median±SD27±4.225±4.2 0.031 HTN, absolute value, (%)54 (32.7)41 (24.8)0.217DM, absolute value, (%)18 (10.9)10 (6.1)0.179History of CAD, absolute value, (%)12 (7.3)2 (1.2) 0.010 Cancer, absolute value, (%)3 (1.8)6 (3.6)0.231Number of chronic conditions, median±SD1±1.031±0.70.117Multimorbidity† present, absolute value, (% yes)26 (15.8)12 (7.3) 0.027 *The χ² test was applied for categorical variables and Mann-Whitney U test was used for numerical variables. Where expected cell counts were low, Fisher’s exact test was used as an alternative to the χ² test. Per cents were calculated per entire group of patients (n=165). Bold font signifies statistically significant results.†Multimorbidity is present if the patient is concurrently diagnosed ≥2 chronic conditions.BMI, body mass index; CAD, coronary artery disease; DM, diabetes mellitus; HTN, hypertension. Sociodemographic and clinical characteristics across different frailty statuses The χ² test was applied for categorical variables and Mann-Whitney U test was used for numerical variables. Where expected cell counts were low, Fisher’s exact test was used as an alternative to the χ² test. Per cents were calculated per entire group of patients (n=165). Bold font signifies statistically significant results. Multimorbidity is present if the patient is concurrently diagnosed ≥2 chronic conditions. BMI, body mass index; CAD, coronary artery disease; DM, diabetes mellitus; HTN, hypertension. Retrospective descriptive analysis: mortality rates and healthcare utilisation in Montenegro The retrospective analysis revealed a marked increase in mortality rates during 2021, coinciding with the COVID-19 pandemic. This rise was particularly pronounced in adults aged 55–59, 60–64 and 65–69, with mortality rates exceeding their highest recorded levels since 2012 by 16%, 18% and 29%, respectively (figure 3). Mortality rates in individuals aged 40–69 in Montenegro from 2012 to 2022. The mortality rate represents the number of deaths in relation to the estimated midyear population, expressed per 1000 inhabitants. In the same period, a decline was observed in both primary healthcare visits and hospital admissions for cardiovascular diseases, endocrine, nutritional and metabolic disorders, neurological and malignant diseases. While primary healthcare visits for infectious diseases decreased, hospital admissions for these conditions increased substantially (figure 4a,b). As a result, the proportion of hospital visits relative to total healthcare visits (hospital+primary care) for chronic conditions remained stable in 2021 compared with the period from 2012. However, for infectious diseases, this ratio surged to 2.5 times the previously highest recorded level (figure 4c), primarily due to increased hospitalisations for infectious diseases (figure 4b). Healthcare utilisation by diagnostic groups from 2012 to 2021. (a) Number of primary healthcare visits for each year by groups of diagnoses expressed per 100 000 people. (b) Number of hospital visits for each year by groups of diagnoses expressed per 100 000 people. (c) Percentage of hospital visits relative to all healthcare visits by groups of diagnoses. Discussion This pilot study provides the first estimates of frailty, prefrailty and multimorbidity prevalence among adults aged 40–69 in Montenegro, revealing concerning patterns with significant implications for healthcare planning. We found that slightly more than half of adults aged 40–69 already meet the criteria for either prefrailty or frailty (46.7% prefrail and 6.1% frail) and nearly one in four adults aged 40–69 lives with multimorbidity (23%), (table 2). Notably, both the median number of chronic diseases and frailty score increase significantly among individuals aged ≥55 (figure 2). The estimated prevalence of frailty and prefrailty in our study (table 2) is higher than that reported by Hanlon et al, who used the same methodology (adapted Fried’s phenotype): 6.1% vs 3% for frailty, and 46.7% vs 38% for prefrailty, while the prevalence of multimorbidity (23%) is lower than reported by Hanlon in the UK (33%).5 While multimorbidity reflects the mere number of chronic conditions a person has, frailty is more sensitive to the severity of those conditions, which suggests that the observed differences between the two populations may reflect variations in healthcare system performance. In Montenegro, less efficient preventive and screening programmes—compounded by socioeconomic factors such as poverty and lower educational attainment (table 1)—likely result in many chronic conditions remaining undiagnosed until they reach advanced stages, when complications and functional decline have already set in. Moreover, we applied a relatively strict definition of multimorbidity, including only major chronic diseases that contribute to mortality, while Hanlon et al and many other studies report higher prevalence in middle age by including broader diagnostic categories.5 16 17 The observed differences in frailty prevalence between our study and Hanlon’s are in line with previous research showing that low socioeconomic status is associated with earlier onset, higher prevalence and faster progression of frailty, even in midlife.31 32 Other studies that included middle-aged populations reported similar or slightly lower frailty prevalence,10 14 21 except in cases where a higher proportion of older adults was included, as frailty prevalence increases with age.12 15 In the age-stratified analysis, both the number of chronic diseases and the frailty score increased with age (figure 2). A significant rise in the number of chronic diseases was already observed from the age of 55 onwards (figure 2a). Similar findings were reported by Van Oostrom et al, who demonstrated that the prevalence of multimorbidity increases substantially in the 55–64 age group.33 These data support the notion that age 55 may represent a critical turning point for preventive strategies. Identifying and supporting this subgroup through targeted interventions, improved screening and strengthened primary care systems could help mitigate the growing burden of multimorbidity and frailty. The significance of our findings is reflected through the national mortality data during the COVID-19 pandemic: increased mortality disproportionately affected those aged 55 and older, particularly in 2021 when the healthcare system faced major disruptions (figure 3). That year, Montenegro recorded excess death rates nearly twice the average of the WHO European Region, with a sharp increase in mortality observed after age 55.20 These trends likely reflect a combination of underlying population health vulnerabilities (such as frailty and multimorbidity) and weaknesses in the healthcare system. Frailty and multimorbidity are both well-established independent predictors of adverse COVID-19 outcomes.34,36 A meta-analysis by Zhang et al, which included over 23 000 COVID-19 patients, identified frailty as a strong independent predictor of mortality.35 Similarly, Iaccarino et al reported that age and multimorbidity were key determinants of COVID-19-related mortality.36 Excess mortality was also driven by disruptions within the healthcare system, including disrupted service delivery, limited infrastructure and inadequate protection for healthcare workers. National healthcare utilisation data supports this: in 2021, there was a marked decline in primary care visits and hospitalisations for chronic conditions (figure 4a–c), while hospitalisations for infectious diseases increased. Univariate analyses revealed that frail and prefrail individuals had higher BMI and a higher prevalence of both multimorbidity and CAD than robust individuals (table 3). These trends are consistent with findings reported in the literature. Obesity has been identified as a risk factor for the development of both prefrailty and frailty.37,39 Obesity is accompanied by metabolic and inflammatory alterations that may contribute to a parallel reduction in muscle mass—sarcopenia.40 41 Importantly, sarcopenia is not only a recognised risk factor for frailty but is also embedded within the very criteria used to define frailty, as reduced grip strength represents one of the components of Fried’s frailty phenotype. The coexistence of both conditions, known as sarcopenic obesity, has been increasingly recognised as an important mechanism linking obesity to functional decline and accelerated frailty progression.42 Similarly, CAD not only contributes to the onset and progression of frailty but also worsens clinical outcomes in already frail individuals.43,46 For example, frail patients with CAD have been shown to experience poorer outcomes, including higher failure rates following percutaneous coronary interventions.46 Regarding the association with multimorbidity, Vetrano et al reported that most frail individuals are multimorbid, while a subset of multimorbid individuals are frail (72% and 16%, respectively).47 Recognising this interplay is important, as the combined effects of these conditions on health outcomes are often multiplicative.4 48 From a methodological standpoint, this study confirmed that the research procedure itself is both feasible and straightforward, as evidenced by the absence of missing data. It also provided prevalence estimates that will allow for precise sample size calculations for future studies using narrower CIs and smaller margins of error. Although this pilot study provided valuable insights and laid the groundwork for the design of a larger-scale study, several limitations must be acknowledged. First, the small sample size and recruitment from the central region may limit the generalisability of the findings. The small number of participants recruited by two physicians reflects the constraints of real-world primary care settings in Montenegro, where limited time and high patient loads often restrict physicians’ ability to engage in research activities. This highlights the need to involve a greater number of physicians in future and to enhance their motivation. Although nearly half of Montenegro’s population resides in the central region19 where this study was conducted, future larger studies should include participants from the coastal and north regions of Montenegro to ensure representativeness. Second, the cross-sectional design prevents assessment of frailty trajectories and causal relationships. A longitudinal design will be necessary in future studies to better understand progression patterns and identify individual risk trajectories over time. Third, women accounted for 70% of participants despite making up 50.8% of the Montenegrin population.19 Nevertheless, sex distribution did not differ significantly between the prefrail/frail and robust groups (table 3). The higher proportion of female participants likely reflects the fact that women are more likely to attend regular check-ups and preventive examinations.49 Additionally, four out of the five recruiting physicians were women, which may have influenced female participation, as some women may prefer to consult female doctors.50 Therefore, future studies should aim to include a more gender-balanced group of physicians in the recruitment process. Fourth, the recruitment strategy may have introduced some degree of selection bias, as participants attending follow-up appointments for chronic conditions may have been more likely to be prefrail or frail. However, those attending preventive check-ups may have been more health-conscious and generally healthier than the broader middle-aged population, which could partly offset this effect. Another limitation of our study is that we assessed multimorbidity by focusing only on major NCDs that substantially contribute to mortality, rather than using a comprehensive list of conditions for multimorbidity assessment. This pragmatic approach was chosen to ensure feasibility and diagnostic accuracy in routine primary care, but it may underestimate the true burden of multimorbidity. Finally, we used a modified version of Fried’s frailty phenotype that incorporates self-reported measures, and we acknowledge their limitations, including potential biases related to recall errors, subjective interpretation and cultural influences. Nonetheless, this approach has been validated in large epidemiological studies and is widely accepted in the literature.551,54 Importantly, self-reported measures offer practical advantages in primary care and resource-limited settings, as they are quick to administer, non-invasive and minimally disruptive to clinical workflows. Future studies should, where feasible, consider including additional objective performance-based measures to strengthen assessment robustness. Besides its methodological implications, this pilot study demonstrated the potential of the primary healthcare setting to enable early detection of prefrailty, frailty and multimorbidity. Primary healthcare serves as the first point of contact for most individuals, often before serious complications arise. This gives primary care physicians an advantage in identifying prefrailty, frailty and multimorbidity early, enabling interventions that are more effective than those initiated later in life when deficits become irreversible.55 Our findings also emphasise the importance of establishing health registers for chronic diseases for detection of multimorbidity. Conclusion This pilot study provides the first data on frailty and multimorbidity among adults aged 40–69 in Montenegro, identifying adults 55 and older as a particularly vulnerable group. Despite its exploratory nature, the findings already show that simple frailty assessments are both feasible and valuable in primary care, enabling early detection of at-risk individuals, when interventions are more effective. Integrating routine frailty screening and strengthening chronic disease surveillance through screening and registries could help delay disease progression and reduce future healthcare burdens. The study sets the foundation for larger, longitudinal research to validate these results and guide targeted, system-level interventions to improve population health and resilience of Montenegro’s healthcare system. Footnotes Funding: This research is funded through a grant from Ministry of Education, Science and Innovation of Montenegro (Grant Agreement No. 04-082/23-2528/1). Prepub: Prepublication history for this paper is available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-107788). Data availability free text: The pilot data set associated with this paper is deposited on Zenodo under DOI https://doi.org/10.5281/zenodo.15530367 65 and will be made publicly available upon publication of the manuscript. Data on mortality and healthcare utilisation are publicly accessible on the official websites of MONSTAT and the Institute of Public Health of Montenegro. Patient consent for publication: Not applicable. Ethics approval: The Ethics Committee of the Primary Healthcare Centre in Podgorica (No. 5/17-4714 and No. 5/17-9353, the latter reflecting an extension of approval period). Participants gave informed consent to participate in the study before taking part. Provenance and peer review: Not commissioned; externally peer reviewed. Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research. Data availability statement Data are available in a public, open access repository. References Data Availability Statement Data are available in a public, open access repository.