Barriers and unmet healthcare needs among patients in two Romanian hospitals

Introduction

Background

Romania has the lowest expenditures for outpatient services in Europe, while hospital-related costs account for a disproportionately large share of spending. This imbalance is driven by access barriers and unmet healthcare needs (UHNs). Assessing barriers to healthcare access is essential for strengthening primary healthcare (PHC) performance (1). These barriers include direct factors such as out-of-pocket costs, long waiting times, travel distances, poor transport, and underdeveloped infrastructure, as well as indirect opportunity costs like lost wages or caregiving responsibilities, which are often underestimated by policymakers (2). Understanding their mechanisms can help explain why care is often forgone.

The World Health Organization (WHO) uses the Tanahashi model to classify access barriers into availability, accessibility, acceptability, contact coverage, and effective coverage (3). These barriers often compound, particularly for marginalized or last-mile populations, intensifying disparities in care access (4). Studies show contact coverage barriers are most common, followed by those related to effective coverage. European statistics identify seven access barriers to healthcare: high costs, distance to medical facilities, lack of time, distrust in medical personnel, long waiting lists, fear (of doctors, hospitalization, or treatment), and the expectation that the health issue will resolve itself. In Romania, the predominant barrier is the high level of expenditure (5). Romania ranks second in Europe, after Greece, with an average of 12.3% of the low-income (first quintile) population aged 65 years and over reporting that they cannot afford the costs associated with healthcare. Countries experiencing a significant decline in the health status of the elderly also report the lowest levels of public health expenditure. While Romania and Bulgaria represent the lower extreme, Switzerland, Germany, and several Nordic countries demonstrate a substantially higher commitment to population health, with public spending levels more than triple those of Romania. Furthermore, Romania records the smallest share of spending dedicated to outpatient care (18%). More critical is the lack of investment in prevention, as Romania allocates the second-lowest amount to this end after Slovakia, and nine times less per capita than Germany. At the same time, inpatient expenditures account for 44%, the highest among European Union (EU) countries. According to 2023 European Union Statistics on Income and Living Conditions (EU-SILC) statistics, the proportion of the population experiencing UHNs in Romania increases significantly with age, a trend that is far more pronounced than the EU average. Indeed, Romanian residents over the age of 65 years are nearly three times more likely to experience healthcare deprivation than the EU average (6).

PHC typically serves as the entry point to care. One method to assess its effectiveness is through avoidable hospitalizations. Policy responses have included cost-sharing and PHC strengthening, though evidence on their effectiveness is mixed. While cost-sharing may reduce admissions, it can also harm population health and raise overall costs (7).

Some research finds that better PHC access lowers hospitalization rates, while others observe the opposite. Although visits to general practitioners (GPs) or specialists often correlate with fewer hospitalizations, findings vary, especially in regions like the U.S., where rural PHC is limited and often substituted by nurse practitioners. In contrast, countries like England, France, and Germany are expanding PHC to ease hospital burden, though evidence remains limited (8). In Romania, only 10.4% (58 of 558) of hospitals are located in rural areas. Furthermore, there are 1,645 patients per doctor in urban areas compared to only 144 in rural areas (9).

The concept of effective coverage, though less explored in the literature due to its complexity, can nonetheless be estimated through proxy indicators. Effective coverage refers to the proportion of the population that requires medical care and successfully accesses appropriate services of sufficient quality.

Barriers to healthcare access lead to UHNs, which both result from and perpetuate healthcare supply-demand imbalances. UHNs lack a universally accepted definition but are commonly described as either the gap between necessary and received services (10), or as the presence of healthcare needs for which individuals do not or cannot receive quality medical care (11). Supply-side factors include resource allocation (e.g., workforce, infrastructure, medications), while demand is shaped by population health, social norms, values, literacy, expectations, and previous experiences.

UHNs are part of the EU’s Social Scoreboard under the European Pillar of Social Rights, and the WHO has been urged to incorporate UHNs indicators in universal health coverage monitoring (12).

Rationale and knowledge gap

Studies utilizing the Tanahashi model indicate that contact coverage barriers are among the most prevalent in certain countries, followed by effective coverage barriers (3). The study’s rationale can be summarized as follows: the care process begins at the PHC level, and its deficiencies can be assessed by examining the probability of self-initiated admissions. Barriers to healthcare access generate UHNs. As such, these barriers are both a cause and a consequence of the prevalence and persistence of UHNs. Addressing UHNs requires better resource allocation, education, service quality, communication, and targeted research.

While research on UHNs is concentrated in the U.S. and Canada (13), interest is growing in Europe as limited data obstructs public health policy reform (14). Romania remains underrepresented, with only three identified studies (15). Despite growing recognition of the need for inclusive approaches involving not only researchers and policymakers, but also patients and the public, such efforts remain rare across Europe (16). Most UHNs data are derived from self-reported measures, yet challenges remain in understanding factors such as demand behaviors, supply characteristics, and care quality. Since most studies rely on national-level statistical data, there is a notable lack of disaggregated approaches that consider factors such as residential environment, types of health conditions, sociodemographic characteristics, health-related beliefs and attitudes, and institutional discrimination.

Measurement methods may be subjective (patient-given assessment) or objective (clinical data) (17). Common metrics like primary care consultation rates and resource availability fail to reflect unutilized care needs, which are better captured by proxy indicators derived from surveys (13). Research gaps remain in the identification of measurable proxies that closely align with specific UHNs categories (18). Self-reported UHNs refers to whether an individual needed examination or treatment for a specific type of healthcare but did not receive or seek such care (19). Effective healthcare coverage is quantified by hospitalization needs, care quality, and medical professionals’ expertise, which is then compared to patient-reported UHNs (20). To date, no studies have been conducted in Romania on UHNs among hospitalized patients.

Objective

The main objective of this article is to identify PHC access barriers and resulting UHNs in the patient group interviewed, focusing on care-seeking behaviors. To achieve the objective, two dependent variables are considered: (I) the number of visits to a GP in the past 12 months (contact coverage); and (II) the mode of hospital admission (effective coverage). Additionally, the purpose of the GP visit is analyzed as an explanatory variable in the initiation stage of the care process.

The specific objectives are to: (I) classify barriers to PHC access using the Tanahashi model; (II) investigate the mechanisms underlying these barriers; and (III) assess the extent to which PHC underutilization contributes to increased hospital utilization.

The prespecified hypotheses are:

• Out-of-pocket expenditure constitutes the main barrier to PHC access;
• Individuals who regularly consult GPs are less likely to require emergency hospitalization;
• Healthcare-seeking behavior influences both the probability and the mode of hospitalization.

The UHNs are analyzed in relation to how patients seek hospital care. Alongside standard admission types (emergency and referred), self-initiated hospitalizations are examined as a sign of PHC underuse. We present this article in accordance with the STROBE reporting checklist (available at https://jphe.amegroups.com/article/view/10.21037/jphe-25-51/rc).

Methods

This is an analytical, cross-sectional, multi-center, hospital-based study. Its scope is to identify barriers and UHNs among hospitalized patients. The study design is depicted in Figure 1. Data collection took place between April and June 2024 in two distinct Romanian hospitals: a clinical facility in Bucharest and a county emergency unit in Ploiesti. The convenience selection of these two healthcare facilities was based on the institutional accessibility and the authors’ ability to ensure the data collection process oversight. Even though the selection includes the two urban hospitals, it still tackles the profound urban-rural disparities from the following perspectives: (I) systemic concentration of care: while the hospitals are geographically located in urban centers, they serve as the main care facilities for surrounding rural regions. (II) Regional specificity (Prahova County): 14 out of 15 hospitals are located in urban centers. Consequently, rural residents are often compelled to seek emergency and continuous care in urban hospitals, as rural facilities are virtually non-existent. (III) Primary care disparity: the lack of rural infrastructure is compounded by a shortage of primary care. This disparity often drives rural patients to bypass local primary care and initiate hospital admission. (IV) Sample composition: the data should reflect this migratory pattern of care.

Figure 1 Study design and conceptual framework. Overview of the cross-sectional hospital-based study, including setting, sampling strategy, questionnaire development, theoretical frameworks Andersen’s behavioral model and Tanahashi’s model, and outcome structure assessing healthcare barriers and unmet healthcare needs. GP, General practitioner.

To ensure a demographically balanced sample, three different wards were selected: (I) Urology (clinical hospital), with a predominantly older, male population; (II) Pneumology (emergency hospital), for its wider representation; (III) Obstetrics-Gynecology (emergency hospital), to represent a predominantly younger, female population. Also, the selection of these three medical specialties was driven by their importance in hospital admissions. Even after the coronavirus disease 2019 (COVID-19) crisis, respiratory conditions ranked as the leading factors of hospital morbidity. At the same time, urological ailments represented significant burdens, particularly renal insufficiencies, accounting for almost a quarter of hospitalizations. Additionally, Romania systematically holds the highest hospitalization rates in Europe for conditions originating in the perinatal period.

The main objective of the sampling strategy was to achieve a demographically balanced sample across age groups and genders, rather than to perform a comparative analysis between clinical wards. The three different wards served as strata to capture a heterogeneous patient population that reflects the reality of hospital admissions. A stratified proportionate sampling strategy was conducted to ensure that each ward (Urology, Pneumology, and Obstetrics-Gynecology) is represented according to its clinical volume. Within these strata, a consecutive convenience sampling approach was applied, where all eligible patients admitted on designated data-collection days were invited to participate. The target sample size was determined based on previous monthly admission data for each ward, aiming for a ~10% representation of the 3-month admissions. The final analysis was subsequently calibrated against the actual admission figures recorded between April and June 2024.

While the study sample consists of hospitalized patients, this cohort should not be assumed to have fully met healthcare needs. Instead, they represent a unique group that experiences significant system-internal barriers. The unmet needs within this so-called privileged sample are most evident in two areas: contact coverage and effective coverage. The heavy reliance on GPs predominantly for bureaucratic functions (referrals), coupled with a high rate of hospital self-admitted patients suggest bypass behaviors indicating existing UHNs.

The data collection instrument is a 22-item structured questionnaire interviewer-administered, adapted from established, validated frameworks (Appendix 1). Regarding the face and content validity, the questionnaire underwent a first assessment performed by the nurses involved in the study, followed by a formal validity assessment by a panel of five experts, including medical and public health specialists.

Data were collected by trained nurses through a consecutive sampling approach. All initial participants were informed about the study’s scope and objectives, the voluntary nature of their participation, and the data confidentiality. Written informed consent was obtained prior to data collection. Consenting patients aged over 18 years who were cognitively capable of completing the survey were included. Patients in critical condition or unable to communicate efficiently were excluded.

The sociodemographic section was structured using Andersen’s Behavioral Model of Health Service Use (21). The covariates (sociodemographic variables), classified according to Andersen’s model, include three categories: predisposing factors (gender, age, education); enabling factors (monthly income, residence); and need factors (presence of at least one chronic disease, medical specialty—Urology, Pneumology, Obstetrics-Gynecology).

The barriers to healthcare were adapted from Tanahashi’s model (3). The explanatory variables categorized according to this model are: (I) accessibility barriers, including financial and logistical obstacles (excessive costs, lack of time or inconvenient medical facility hours, long travel distance or time, lack of transportation, physical incapacity to travel); (II) acceptability barriers, including preferences for alternative care options (self-treatment, consulting a GP, consulting a specialist, seeking advice from a pharmacist, relying on family or friends, using online resources, consulting a nurse, or calling emergency services), emotional factors (fear of receiving bad news, fear of pain, past negative experiences, embarrassment), and organizational concerns (avoiding contact with other patients, fear of healthcare-associated infections); (III) initiation of the care-seeking process (contact coverage), assessed through the purpose of the GP visit (obtaining a referral, routine check-up, presence of symptoms).

These models were selected to provide a comprehensive view of barriers and UHNs. To this end, two proxy-dependent variables were used: (I) number of GP visits over the preceding twelve months representing contact coverage, and (II) mode of hospital admission (self-referred, GP-referred, specialist-referred) reflecting effective coverage. Patient-initiated admissions encompass unscheduled arrivals via emergency medical services (ambulance) as well as presentations by private transport (e.g., family, friends, or self-driving).

The number of GP visits was assessed using survey questions aligned with those from major European health studies. In the EU-SILC survey, the corresponding question (coded PH008) is formulated as follows: “During the past 12 months, about how many times have you consulted a GP (including home visits by the doctor)?” Similarly, in the ESS, the corresponding variable (coded tmcnsdc) is phrased as: “Consulted GP, how many times in the last 12 months?”

The mode of hospital admission, although not systematically recorded in European health statistics, is of particular relevance for identifying UHNs. High rates of self-referred and emergency hospital admissions may indicate healthcare avoidance or delays in seeking medical attention, both of which are associated with UHNs.

All procedures performed in this study were in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of the Clinical Hospital “Prof. Dr. Th. Burghele”, Bucharest (No. 2649/14.03.2024), and the Ethics Committee of the County Emergency Hospital “Constantin Andreoiu” Ploiesti, Romania (No. 21999/10.05.2024), and informed consent was obtained from all individual participants.

Statistical analysis

All statistical analyses were conducted using the cloud-based version of Jamovi (latest release). A Poisson regression model was initially fitted for the count outcome (number of GP visits); however, because overdispersion was observed (variance exceeding the mean), a negative binomial regression model was applied. Regression coefficients are reported as incidence rate ratios (IRRs) with 95% confidence intervals (CIs). Statistical significance was evaluated at P<0.05. Model fit and overdispersion are assessed using residual deviance and the chi-squared/degrees-of-freedom ratio. For the second dependent variable, the mode of hospital admissions, univariate logistic models are employed to assess the associations with predictors. Model specification was theory-driven, with all variables identified a priori from Andersen’s and Tanahashi’s frameworks entered simultaneously into the regression models (enter method), without stepwise selection.

Results

Across the Urology, Pneumology, and Obstetrics-Gynecology wards, 232 patients were interviewed, with 217 (93.5%) meeting the criteria for final analysis. This validated sample accounts for 12.5% of the 1,731 total admissions across the three wards. Specifically, the sample captured 13.4% of Urology admissions (135/1,009), 14.7% of Pneumology admissions (40/273), and 9.4% of Obstetrics-Gynecology admissions (42/449). The aggregate margin of error was ±6.22% (95% CI), with ward-specific margins ranging from ±7.85% to ±14.41%. Fifteen questionnaires (nine for Urology, three for Pneumology, and three for Obstetrics-Gynecology) were excluded due to incomplete sociodemographic data, specifically missing age or income information due to privacy concerns. The attrition rate was consistent across all three wards (averaging 6.5%). Since these exclusions represented a small fraction of the total sample and were distributed proportionally across wards, the risk of selection bias is considered minimal. The baseline demographic and clinical characteristics of the cohort are presented in Table 1.

The study cohort is characterized by a slight male predominance and a significant concentration of older adults, while the geographical distribution shows a primary urban concentration. Socioeconomic and demographic profiles reveal a financially vulnerable patient base.

Contact coverage

Due to the count nature of the outcome and the presence of overdispersion (χ²/df=2.94 in a Poisson model), a negative binomial regression was employed to examine the association between sociodemographic characteristics (Andersen’s model), the encountered barriers (Tanahashi’s model) and the annual number of GP visits. The statistically significant predictors for the GP visit frequency are presented in Table 2.

Sociodemographic characteristics

Significant associations were found between GP visits and both gender (predisposing) and presence of chronic diseases (need). Specifically, men had a 32% lower rate of GP visits compared to women. Patients with chronic diseases exhibited a 51% higher rate of GP visits compared to those without chronic conditions. The model demonstrated good fit, with a residual deviance of 242.83 and no evidence of significant overdispersion (χ²/df=1.09). These findings indicate that female gender and presence of chronic disease are important predictors of increased GP visits in this population.

Encountered barriers

Significant predictors of fewer GP visits included the lack of time barrier (accessibility) and two healthcare-seeking behavior characteristics (acceptability). Specifically, the lack of time reduced the frequency of GP visits by half. Also, individuals who rely on pharmacists had a 39% lower rate of GP visits. Patients who seek advice from family or friends had a 44% lower frequency of visits to a GP. The model demonstrated good fit, with a residual deviance of 205 and no evidence of significant overdispersion (χ²/df=1.02). These findings confirm that lack of time and avoidant healthcare-seeking behavior are important predictors of decreased GP visits.

To verify if the regression analysis regarding gender is not confounded by the specific wards selected, a sensitivity analysis excluding the Obstetrics-Gynecology was conducted. Results confirmed that gender remains a significant predictor of GP visit frequency. Specifically, men had a lower expected visit rate compared with the reference category (IRR =0.695, 95% CI: 0.527–0.915, P=0.01).

Effective coverage

The statistically significant predictors for the mode of hospital admission are presented in Table 3. When focusing specifically on patient-initiated admissions, among all predictors examined, financial constraints (accessibility) were significantly associated with patient-initiated admissions. Higher reported cost burden induced lower odds of patient-initiated admissions. Regarding the GP-referred admissions, visiting the GP for a referral significantly decreased the odds of GP-referred hospital admissions. A significant association was also observed for patients with chronic conditions who are more than twice as susceptible to being hospitalized through a GP referral.

Discussion

Key findings

Contact coverage

The frequency of GP visits is significantly influenced by gender, chronic illness, time constraints, support from family or friends, and pharmacist advice. Gender significantly influences the frequency of GP visits. As a predisposing factor, it underscores a persistent underutilization of primary care by men. This is not just an issue of time constraints, but more likely a habit outcome of specific psychological barriers. The findings of this study align with European statistics, showing that in Romania, women were more likely to visit a GP in the past 12 months (55.1%) compared to men (42.7%) (22). The lower frequency of visits to a GP among men suggests a significant gender gap, which highlights the need for further qualitative, in-depth cultural and psychological inquiries. Future research should examine how masculine norms deter seeking primary care. Policies should develop gender-sensitive interventions seeking countermeasures for these avoidance behaviors.

The higher GP visit rate among chronic patients (need factor) validates that the GP serves as the primary hub for long-term management, which may be essential for achieving effective coverage. The finding that lack of time (accessibility) reduces visits by half is critical. This indicates that the opportunity cost of seeking care is prohibitive, and it leads to missed opportunities for early intervention. This finding is consistent with data from Greece and Italy, where time constraints, alongside high costs, are the most frequently cited reasons for foregoing medical care, often due to family or professional obligations.

Among the acceptability variables, support from family or friends and advice from pharmacists show statistically significant negative correlations with GP visit frequency across the sample. The lower visit rates among those relying on pharmacists and family or friends (acceptability) suggest a “substitution effect” that undermines contact coverage. This pattern suggests low acceptability of formal primary care services, leading to UHNs. Furthermore, this behavior bypasses the formal diagnostic integrated in the GP’s record, inducing an important barrier to the continuity of primary care. Additionally, the data shows a high reliance on GPs primarily for bureaucratic purposes, specifically for obtaining referrals, rather than for clinical check-ups or symptom management. This suggests that even for those connected to the system, primary care often functions as a hurdle rather than a point of effective intervention.

Effective coverage

The substantial percentage of self-admitted patients to the hospital without a referral from a GP or specialist indicates a breakdown in the standard healthcare pathway. This bypass behavior highlights that even when services are available, the formal navigation process is sufficiently fragmented or inefficient that patients resort to direct hospital entry to ensure they receive care.

While financial constraints significantly reduce the odds of patient-initiated hospital admissions, they do not affect the frequency of GP visits. This suggests that primary care remains an accessible healthcare entry point. Instead, the hospital admission is perceived as a financial burden stemming from hospital-specific costs, including informal payments, and leading to UHNs. Consequently, healthcare policies should address more thoroughly the out-of-pocket burden associated with inpatient stays.

The outcome that patients with chronic ailments are twice as likely to be admitted via GP referral underscores the critical role of primary care. It can also highlight that the primary care system lacks the resources to manage these patients in the community. Thus, the dual observation of increased GP utilization and higher referral-based hospitalization among chronic patients should not be viewed as a contradiction.

The finding that GP visits for referrals decrease the likelihood of admission underscores the value of the primary care sector. By providing an alternative to hospitalization, GPs act as critical “gatekeepers” in the healthcare system. It should be added that GP referrals are primarily intended to direct patients towards specialists and clinical tests, rather than for hospitalization.

Strengths and limitations

This study supports the notion that systematic analysis of hospital admissions can serve as a metric for evaluating PHC quality (7) and contributes to the limited existing research on the correlation between primary care efficiency and hospitalization rates (8).

Data were collected from two urban hospital centers in Bucharest and Ploiesti, which may limit the generalizability of the findings. However, this hospital selection mirrors the structural reality of the Romanian healthcare system, where most hospitals are concentrated in urban areas. Future research should include smaller community hospitals to further explore the nuances of local-level barriers and UHNs.

The purposive selection of specific wards (Urology, Pneumology, and Obstetrics-Gynecology) allowed for a balanced and diverse representation of both age groups and gender. Finally, by focusing on hospitalized patients, we uncover “hidden” unmet needs that remain unresolved even after the hurdle of admission has been cleared.

Several limitations of this study must be acknowledged. Given the modest size of the sample, we acknowledge that statistical power is limited and that effect estimates may be unstable. We also emphasize that the findings should not be interpreted as definitive predictors, but rather as preliminary signals warranting validation in larger and more diverse cohorts. While the 3-month timeframe (April–June 2024) provided a strong snapshot of patient experience, it may not account for seasonal variations in hospital admission patterns.

Another limitation stems from the involvement of healthcare professionals in administering questionnaires, which may have led to social desirability bias, particularly regarding out-of-pocket expenditures, organizational barriers, and emotional obstacles encountered. Future research should aim to provide a clearer understanding of the structure and nature of out-of-pocket expenditures. Also, this study is subject to Berkson’s and “survivorship” biases because the cohort consists exclusively of hospitalized patients. By definition, these individuals have successfully navigated initial systemic barriers to achieve admission, excluding the most vulnerable populations who may have forgone care entirely due to financial constraints. However, we argue that this “privileged” cohort provides a unique lens into the internal failures of the healthcare system. Even among those who secured admission, significant unmet needs were observed. For example, the high reliance on GPs primarily for bureaucratic referrals rather than clinical management suggests a breakdown in primary care contact coverage. Additionally, the high rate of patient self-admission without prior referrals indicates a bypass of the standard healthcare pathway, a form of effective coverage failure.

Comparison with similar research and explanations of findings

EU-SILC data indicate that elderly European women perceive greater healthcare access barriers than men. Yet, European statistics record 55.1% of Romanian women visiting a GP in the past year compared to 42.7% of men (22), likely reflecting higher chronic illness prevalence (23), or psychological factors.

As mentioned in the literature, chronic illness correlates strongly with healthcare barriers (16,24). However, European healthcare data reveal that the association between chronic disease prevalence and GP visit frequency varies across countries: Italy reports the most frequent yearly medical consultations despite having the lowest chronic illness rates, while Nordic countries show the reverse. Hungary encounters both a substantial prevalence of chronic diseases and a high medical consultation rate of 9.5 visits per capita. Romania averages 5.5 visits annually, which places it in the intermediate range compared to other countries (22). Notably, Romanian adults aged 35–44 years report the lowest rates of chronic illnesses in the EU, yet escalate sharply after age 65 years (23). This trend is also observed in other lower-income countries such as Bulgaria, Slovakia, Greece, and Hungary. In contrast, developed nations, particularly Nordic countries, experience less pronounced declines in perceived health with aging (25). The abrupt transition for the elderly population suggests cumulative effects of weak long-term preventive care, poor living conditions, and limited resources for the primary care system (24,26-29).

Despite low GP visit costs, out-of-pocket expenses remain a major barrier: 12.3% of low-income Romanians aged 65+ years could not afford care in 2023, second only to Greece. The trend is emerging among higher-income young adults as well (30). In the U.S., financial barriers cause 20% of chronically ill patients to skip checkups, worsening outcomes (31). Significantly, in this study, cost was cited by nearly a third of respondents as the main access barrier. Other studies signal a weak link between perceived limitations and actual costs (25,32-35). In 2021, two-thirds of healthcare spending in Romania went to medications, with co-payments ranging from 10–80% (36).

Aligning with data from Greece and Italy, where time and cost are common barriers (33,37), in Romania, individuals aged 45–64 years are most affected. Despite mid-level consultation rates, Romania has the EU’s lowest share of people (49.1%) accessing GPs (22). This may be due to high self-rated health; three-quarters of Romanians report good or very good health, though perceptions drop sharply post-65 years (38). Another factor is undervaluing preventive care: only a third of Romanians seek professional help when ill; many prefer self-treatment or expect recovery without intervention (30).

In countries with weak healthcare systems, families often bear responsibility for elder care, accepting their decline in health and reinforcing beliefs that medical help for the elderly is unnecessary except in acute cases (26). Though this is less pronounced in Europe, Romanians aged 65+ years are almost three times more likely than the EU average to report UHNs (30). Barriers include medication management, lack of institutional support, communication issues, mobility limitations, increased risk of falls, treatment costs, lack of specialized healthcare facilities, and distrust or fear of medical professionals (11,28).

Emergency admissions account for nearly half of all hospitalizations in Romania, followed by GP (28.8%) and specialist referrals (12.7%) (39). Romania spends the lowest share of its healthcare budget on outpatient services in the EU. This supports using emergency admission rates as a metric for primary care quality (7,40).

As mentioned earlier, cost concerns are protective against emergency admissions. Despite formal entitlement to hospital care, 18% of Romanians in 2022 reported informal payments or offerings of expensive gifts to doctors and nurses, more than quadruple the EU average (4%) (36). These findings align with studies linking cost-sharing to reduced hospital use (7).

Romania’s hospital-centric spending stems partly from widespread self-initiated admissions, bypassing GPs. By contrast, most EU countries emphasize outpatient services. Romania also ranks second-lowest in prevention spending, investing nine times less than Germany (6).

Implications and actions needed

Understanding UHNs can help identify critical gaps in the healthcare delivery process that require targeted improvements. Proposed solutions for reducing UHNs include optimizing the allocation of available resources (with an emphasis on health literacy), improving service quality, enhancing communication, and conducting further research into the mechanisms of primary medical care avoidance. The most important action is to profoundly reconsider the role of the PHC sector. Policies should transform the primary care model to ensure effective coverage and reduce avoidable hospital admissions.

Conclusions

This study identifies annual GP visits among hospitalized patients as a proxy for inadequate resource allocation and underuse of primary care in Romania, where outpatient services receive the lowest budget share in all of Europe. Most consultations were for referrals, highlighting inefficient use of primary services.

High rates of self-initiated hospital admissions further indicate unmet needs and delayed care. However, regular GP use was linked to reduced emergency hospitalizations, underscoring primary care’s protective role. Even GP consultations intended primarily for bureaucratic referrals effectively divert outpatients from becoming inpatients. Contrary to the first initial hypothesis, out-of-pocket expenditure did not emerge as the main barrier to PHC access (contact coverage), where time constraints and acceptability factors were more significant. Instead, financial burdens acted as a critical barrier to hospital-level care (effective coverage), where a cost constraint significantly reduced the odds of patient-initiated admissions.

The second hypothesis that regular GP consultations reduce hospitalization was only partially supported and highly context-dependent. The data suggest that referral-seeking behavior (visiting the GP specifically for a referral) was the primary driver of lower GP-referred admission odds, suggesting that the GP often functions as an administrative gatekeeper rather than a longitudinal clinical manager who prevents the need for admission. Thus, the third hypothesis that healthcare-seeking behavior influences the mode of hospitalization was strongly confirmed. Also, acceptability barriers, such as relying on pharmacists or social networks, were associated with a reduction in GP visits, which subsequently altered the pathway to the hospital by bypassing primary care.

Healthcare barriers encompass both accessibility and acceptability factors. While primary care (contact coverage) is hindered by time constraints, financial burdens are obstacles to hospital self-admission (effective coverage). Men, chronic patients, and those who bypass formal consultations seeking advice from pharmacists, family or friends are particularly susceptible to UHNs.

A key finding of this study is the tier-specific nature of cost barriers. The research confirms that financial constraints primarily impact effective coverage rather than contact coverage. Conversely, the presence of chronic conditions exacerbates gaps in both contact and effective coverage, highlighting a significant intersection of medical complexity and systemic vulnerability.

By combining contact and effective coverage, this study provides a framework to monitor primary care performance. Findings can inform policies aimed at improving access, promoting prevention, and reducing avoidable hospital use.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jphe.amegroups.com/article/view/10.21037/jphe-25-51/rc

Data Sharing Statement: Available at https://jphe.amegroups.com/article/view/10.21037/jphe-25-51/dss

Peer Review File: Available at https://jphe.amegroups.com/article/view/10.21037/jphe-25-51/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jphe.amegroups.com/article/view/10.21037/jphe-25-51/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in this study were in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of the Clinical Hospital “Prof. Dr. Th. Burghele”, Bucharest (No. 2649/14.03.2024), and the Ethics Committee of the County Emergency Hospital “Constantin Andreoiu” Ploiesti, Romania (No. 21999/10.05.2024), and informed consent was obtained from all individual participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Handbook for conducting assessments of barriers to effective coverage with health services: in support of equity-oriented reforms towards universal health coverage. Geneva: World Health Organization; 2024.
2. Allin S, Masseria C. Unmet need as an indicator of health care access. Eurohealth 2009;157-10.
3. Tanahashi T. Health service coverage and its evaluation. Bull World Health Organ 1978;56:295-303.
4. Tønnessen-Krokan M, Bringedal Houge A. Complex emergencies: overcoming barriers to health care. Scand J Public Health 2022;50:312-7. [Crossref] [PubMed]
5. Analyzing and Overcoming Access Barriers to Strengthen Primary Health Care. Pan American Health Organization; 2023 [cited 2026 Feb 26]. Available online: https://iris.paho.org/handle/10665.2/58876
6. Eurostat. Health care expenditure by function [Statistical database]. 2024. Data extracted on 03/10/2024. Available online: https://ec.europa.eu/eurostat/web/main/data/database
7. Van den Heede K, Van de Voorde C. Interventions to reduce emergency department utilisation: A review of reviews. Health Policy 2016;120:1337-49. [Crossref] [PubMed]
8. Oh NL, Potter AJ, Sabik LM, et al. The association between primary care use and potentially-preventable hospitalization among dual eligibles age 65 and over. BMC Health Serv Res 2022;22:927. [Crossref] [PubMed]
9. National Institute of Statistics (INS). The Activity of the Sanitary and Healthcare Network in 2023 [Statistical report]. Bucharest, Romania: National Institute of Statistics; 2024.
10. Carr W, Wolfe S. Unmet needs as sociomedical indicators. Int J Health Serv 1976;6:417-30. [Crossref] [PubMed]
11. Rosenberg M, Kowal P, Rahman MM, et al. Better data on unmet healthcare need can strengthen global monitoring of universal health coverage. BMJ 2023;382:e075476. [Crossref] [PubMed]
12. Seventy-Sixth World Health Assembly Resolutions and Decisions Annexes. WHA76/2023/REC/1. World Health Organization; 2023. Available online: https://apps.who.int/gb/ebwha/pdf_files/WHA76-REC1/A76_REC1_Interactive_en.pdf
13. Kocot E. Unmet Health Care Needs of the Older Population in European Countries Based on Indicators Available in the Eurostat Database. Healthcare (Basel) 2023;11:2692. [Crossref] [PubMed]
14. Smith S, Connolly S. Re-thinking unmet need for health care: introducing a dynamic perspective. Health Econ Policy Law 2020;15:440-57. [Crossref] [PubMed]
15. Maslyankov I. Unmet healthcare needs in Southeastern Europe: a systematic review. J Med Access 2024;8:27550834241255838.
16. European Commission. Directorate General for Health and Food Safety. Defining value in ‘value-based healthcare’: report of the Expert Panel on Effective Ways of Investing in Health (EXPH). [Internet]. LU: Publications Office; 2019 [cited 2024 Nov 5]. Available online: https://data.europa.eu/doi/10.2875/35471
17. Ke XT, Wang CL, Salmon JW, et al. Unmet needs as indicator of improving chronic care delivery system in China. Chronic Dis Transl Med 2021;7:1-13. [Crossref] [PubMed]
18. Aragon Aragon MJ, Chalkley M, Goddard M. Defining and measuring unmet need to guide healthcare funding: identifying and filling the gaps. Work Pap 141cherp Cent Health Econ Univ York. 2017. Available online: https://ideas.repec.org/p/chy/respap/141cherp.html
19. Moran V, Suhrcke M, Ruiz-Castell M, et al. Investigating unmet need for healthcare using the European Health Interview Survey: a cross-sectional survey study of Luxembourg. BMJ Open 2021;11:e048860. [Crossref] [PubMed]
20. Ranjan A, Thiagarajan S, Garg S. Measurement of unmet healthcare needs to assess progress on universal health coverage - exploring a novel approach based on household surveys. BMC Health Serv Res 2023;23:525. [Crossref] [PubMed]
21. Andersen RM, Davidson PL. Improving access to care in America: Individual and contextual indicators. In: Andersen RM, Rice TH, Kominski GF. editors. Changing the US Health Care System: Key Issues in Health Services Policy and Management. 3rd ed. San Francisco: Jossey-Bass; 2007:3-31.
22. Eurostat. Consultation of a medical doctor per inhabitant [hlth_hc_phys2__custom_13124872]. Data extracted on 03/10/2024.
23. Eurostat. People having a long-standing illness or health problem, by sex, age and degree of urbanisation [hlth_silc_19__custom_12993301]. Data extracted on 24/09/2024.
24. Yoon YS, Jung B, Kim D, et al. Factors Underlying Unmet Medical Needs: A Cross-Sectional Study. Int J Environ Res Public Health 2019;16:2391. [Crossref] [PubMed]
25. OECD, European Union. Health at a Glance: Europe 2022: State of Health in the EU Cycle. OECD; 2022 [cited 2024 Nov 9]. (Health at a Glance: Europe). Available online: https://www.oecd-ilibrary.org/social-issues-migration-health/health-at-a-glance-europe-2022_507433b0-en doi:10.1787/507433b0-en
26. Gao Q, Prina M, Wu YT, et al. Unmet healthcare needs among middle-aged and older adults in China. Age Ageing 2022;51:afab235. [Crossref] [PubMed]
27. Kowal P, Corso B, Anindya K, et al. Prevalence of unmet health care need in older adults in 83 countries: measuring progressing towards universal health coverage in the context of global population ageing. Popul Health Metr 2023;21:15. [Crossref] [PubMed]
28. Yi K, Kim S. Patient Perspectives of Chronic Disease Management and Unmet Care Needs in South Korea: A Qualitative Study. J Patient Exp 2023;10:23743735231213766. [Crossref] [PubMed]
29. Rahman MM, Rosenberg M, Flores G, et al. A systematic review and meta-analysis of unmet needs for healthcare and long-term care among older people. Health Econ Rev 2022;12:60. [Crossref] [PubMed]
30. Eurostat. Self-reported unmet needs for medical examination by sex, age, main reason declared and income quintile [hlth_silc_08__custom_13143896]. Data extracted on 03/10/2024.
31. Hawks L, Himmelstein DU, Woolhandler S, et al. Trends in Unmet Need for Physician and Preventive Services in the United States, 1998-2017. JAMA Intern Med 2020;180:439-48. [Crossref] [PubMed]
32. Scîntee G, Mosca I, Vlădescu C. Can people afford to pay for health care? New evidence on financial protection in Romania. Copenhagen: WHO Regional Office for Europe; 2022.
33. Cavalieri M. Geographical variation of unmet medical needs in Italy: a multivariate logistic regression analysis. Int J Health Geogr 2013;12:27. [Crossref] [PubMed]
34. Can people afford to pay for health care? Evidence on financial protection in 40 countries in Europe. Copenhagen: WHO Regional Office for Europe; 2023.
35. Cylus J, Papanicolas I. An analysis of perceived access to health care in Europe: How universal is universal coverage? Health Policy 2015;119:1133-44. [Crossref] [PubMed]
36. OECD/European Observatory on Health Systems and Policies. România: Profilul de țară din 2023 în ceea ce privește sănătatea. State of Health in the EU. Paris: OECD Publishing; Brussels: European Observatory on Health Systems and Policies; 2023.
37. Pappa E, Kontodimopoulos N, Papadopoulos A, et al. Investigating unmet health needs in primary health care services in a representative sample of the Greek population. Int J Environ Res Public Health 2013;10:2017-27. [Crossref] [PubMed]
38. Eurostat. Self-perceived health by sex, age and degree of urbanisation [hlth_silc_18__custom_13155104]. Data extracted on 06/10/2024.
39. INMSS. Available online: http://www.drg.ro/index.php?p=indicatori&s=2024_an#form, data extracted on 27/09/2024.
40. Bigby J, Dunn J, Goldman L, et al. Assessing the preventability of emergency hospital admissions. Am J Med 1987;83:1031-6. [Crossref] [PubMed]