Organisational models of general practice and internal capacity for care coordination: a mixed-methods study from the Czech Republic

Abstract

Background

Primary care (PC) is increasingly expected to ensure coordination, continuity, and system efficiency in response to population ageing, multimorbidity, and workforce shortages. However, PC is delivered through diverse organisational models, and evidence on how these models relate to the internal capacity of practices to coordinate care remains limited. This study analyses how organisational models of general practice shape the internal capacity of PC and its ability to coordinate care, using the Czech Republic (CR) as a case.

Methods

The study used a sequential explanatory survey design with a subsequent interpretative expert consultation stage. A cross-sectional online survey of general practitioners (GPs) providing care to adult patients (n = 356) examined organisational models, workforce capacity, task delegation, and care coordination. Descriptive statistical methods were used. Written expert responses to a structured summary of preliminary survey findings were used to contextualise and interpret the observed patterns.

Results

Organisational models were associated with differences in workforce capacity, administrative support, patient load, and selected coordination-related conditions, while task distribution patterns were broadly similar across practice types.The predominance of small, professionally autonomous practices was associated with limited involvement of non-physician staff, constrained delegation, and high administrative workloads. Expert consultation suggested that these constraints were particularly pronounced in rural settings. Younger physicians were more likely to work in group-based or employment-oriented arrangements, indicating a mismatch between prevailing organisational models and workforce preferences. Limited diagnostic access and weakly institutionalised gatekeeping were further associated with lower coordination capacity.

Conclusions

Organisational models appear to play an important role in shaping the internal capacity of PC and its ability to coordinate care. The findings suggest that limitations in delegation and coordination reflect the interaction between practice-level organisational characteristics and broader institutional conditions, and are not fully explained by individual provider choices alone.

Trial registration

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12875-026-03406-2.

Background

The organisation of PC is central to shaping the effectiveness, coordination, and outcomes of healthcare systems [1, 2]. Strengthening PC has therefore become a key strategy for health system reform, reflecting its potential to improve population health while enhancing system efficiency and equity [3, 4].

Across Europe, PC systems are facing increasing pressure due to population ageing, the growing burden of chronic and multi-morbid conditions, rising patient expectations, workforce shortages, and the effects of recent crises, including the COVID-19 pandemic [5, 6]. These cumulative pressures have intensified efforts to identify how PC should be organised to function effectively under conditions of growing demand and complexity [7].

In many high-income countries, PC is predominantly delivered through small, professionally autonomous practices, which has important implications for how care is organised and coordinated and can complicate attempts to translate system-level ambitions into routine practice [8].

At the same time, PC is not organisationally uniform. Across countries, general practice is delivered through a wide range of organisational models that differ in their structural arrangements, including practice size, team composition, and degree of integration, from solo practices to larger, multidisciplinary and collaborative settings [9–11]. Emerging evidence shows that these organisational arrangements are not a neutral backdrop, but actively shape care coordination, workforce dynamics and service delivery, with collaborative and multidisciplinary models associated with more favourable outcomes in some contexts [12, 13].

Rather than focusing on specific team-based interventions or predefined care models, this study addresses the organisational conditions that shape what general practice is structurally able to deliver in everyday PC. In this context, internal capacity is conceptualised as an organisational property of general practice, referring to the structural ability of practices to allocate and redistribute tasks, integrate non-physician staff, and sustain coordination functions within existing institutional arrangements [9, 12].

Yet, despite growing expectations and increasing policy emphasis on integrated and coordinated PC, empirical evidence on how organisational models of general practice shape internal capacity—particularly with regard to care coordination—remains limited. Therefore, the aim of this study is to analyse how organisational models of general practice shape the internal capacity of PC and its ability to coordinate care. The analysis focuses on empirically observable differences between organisational models with regard to organisational structure, workforce capacity and task delegation, generational patterns among GPs, and the organisation and coordination of care.

Methods

The analysis presented in this study focuses on the PC system in the CR, characterised by a persistent dominance of small, professionally autonomous general practices alongside growing policy expectations regarding stronger care coordination, greater workforce capacity, and an expanded role for GPs in the healthcare system. PC is delivered by GPs for adults and by GPs for children and adolescents. This study focuses exclusively on GPs providing care to adult patients, who typically operate as self-employed providers contracted within the publicly funded health insurance system. While patient registration with a specific GP is not legally mandatory, access to routine PC is organised through a patient registration system [14].

The current organisational configuration of Czech PC is shaped by a strong path-dependent development linked to the post-socialist transformation of the healthcare system. Prior to 1989, healthcare provision was highly centralised and organised through territorially defined public institutions. The healthcare reforms of the 1990s led to rapid decentralisation and the dissolution of integrated district-based structures, alongside the expansion of independently operating ambulatory providers and a strengthened role of outpatient specialist care [15]. These reforms were accompanied by regulatory arrangements, including restrictions on prescribing and referral practices for GPs, which contributed to a fragmented organisation of care and limited formal integration between levels of the healthcare system. Key structural features established during this transition have persisted into the present system, despite more recent policy efforts aimed at redefining and strengthening the role of GPs within Czech healthcare [16].

National workforce data indicate pronounced demographic ageing among GPs, with a growing share of physicians aged 60 years and older, raising concerns about the sustainability and internal capacity of PC provision [17].

At the same time, international comparative evidence from the OECD Patient-Reported Indicator Surveys (PaRIS) indicates that Czech PC performs relatively strongly on patient-reported measures of continuity and perceived care coordination, while levels of trust in the healthcare system remain below the average of participating EU countries [18]. These system characteristics form the contextual background for the empirical analysis.

Study design

This study employed asequential explanatory survey design with a subsequent interpretative expert consultation stage. In line with the logic of sequential explanatory design [19], this approach was appropriate for the research objective because the survey identified observable organisational patterns across practices, while the subsequent expert consultation helped to contextualise and interpret these patterns within broader structural and policy conditions.

First, a cross-sectional online survey was conducted. Following the preliminary analysis of the survey data, a structured expert consultation was carried out. The aim of this phase was not to generate new empirical categories, but to capture expert perspectives on the observed patterns and to support the interpretation of the survey findings. Finally, findings from the survey and the expert consultation were brought together at the interpretation stage.

Survey and sampling

The study drew on an online survey of GPs providing care to adult patients. A total of 356 respondents completed the questionnaire (see Additional File 1), representing approximately 6.6% of all GPs in the CR. As this was a non‑probabilistic, self‑selected sample, the findings reflect the characteristics of respondents rather than those of the full GP population.

Participation was voluntary and anonymous. Data collection took place between January and March 2024.

Survey invitations were distributed through two channels. First, GPs listed in the National Register of Healthcare Providers maintained by the Institute of Health Information and Statistics of the CR were contacted via the email addresses recorded in the register [17]. Second, the invitation was shared in a closed Facebook group for GPs with approximately 2,700 members. The questionnaire was hosted on the Google Forms platform.

The questionnaire was developed specifically for this study to capture how different organisational models of general practice operate from the perspective of physicians working within them and to examine empirically observable differences between these models in terms of organisational structure, workforce capacity, task delegation, generational patterns among GPs, and the organisation and coordination of care. The initial draft was prepared by the first author and subsequently discussed with the co-authors and practising general practitioners, whose comments informed revisions to the wording, structure, and relevance of individual items. The instrument was developed exclusively for the purposes of this study and was not intended as a validated scale measuring a single construct.

The questionnaire included closed-ended (single- and multiple-choice) and open-ended items, primarily requesting numerical responses. It was organised into thematic blocks covering organisational models, staffing composition, staff competencies, practice infrastructure, number of registered patients, care organisation, waiting times for acute and non-acute consultations, and selected aspects of care coordination. Internal capacity was not measured as a single validated construct, but was examined indirectly through observable organisational characteristics related to workforce capacity, task delegation, administrative support, appointment organisation, diagnostic access, and selected coordination indicators. The questionnaire was designed to capture the structural and organisational characteristics of practices rather than the individual motivations or causal mechanisms underlying the reported practices.

For analysis, organisational models were grouped into four categories reflecting differences in organisational structure and employment arrangements: solo, group, network-based, and employed practices. This four–category classification was used to enhance analytical clarity and to present the Czech organisational context in terms that are accessible to an international readership (see Supplementary Figure S1).

Solo practices included single-physician practices, regardless of whether they operated independently or within a polyclinic. Group practices comprised two or more GPs, including both formally integrated teams and practices sharing premises without full organisational integration. Network-based practices operated as part of a provider network or chain with shared ownership or management. Employed GPs worked under an employment contract within hospital outpatient departments.

Expert consultation

The expert consultation consisted exclusively of written responses. All experts received the same structured comment template summarising five key findings from the preliminary survey analysis (see Additional File 2). The purpose of this stage was to obtain contextual, structural, and policy-oriented reflections on the identified patterns. The consultation was thus structured around predefined survey findings and designed to elicit expert reflections on these patterns in a consistent written format. Experts were selected based on their professional expertise and standing in PC. Invited experts included representatives of professional associations of GPs and nurses, as well as health policy stakeholders.

Seven experts (four women and three men) were invited, of whom five provided written responses. Expert input was used to contextualise the quantitative findings and to inform interpretation of structural and workforce-related implications. All participating experts consented to having their professional roles disclosed. An overview of expert characteristics is provided in Table 1.

Table 1.

List of experts

Expert	Expert name	Expert position
1	Kateřina Javorská	General Practitioner; Member of the SVL ČLS JEP Committee
2	Bohumil Seifert	Vice President and Scientific Secretary, the SVL ČLS JEP committee
3	Dana Jurásková	Former Minister of Health of the Czech Republic
4	Ludmila Bezdíčková	Member of the Committee of the Association of General Practitioners of the Czech Republic and the SVL ČLS JEP; Head of the Department of General Practice, IPVZ
5	Jan Přáda	Vice President of the Czech Medical Chamber

Abbreviations: SVL ČLS JEP Society of General Practice, Czech Medical Society of Jan Evangelista Purkyně, IPVZ  Institute for Postgraduate Medical Education

Written responses were read in full and interpreted in relation to the preliminary survey findings, with attention to areas of convergence, divergence, and contextual elaboration. Interpretation of expert responses was discussed among the authors prior to integration with the survey findings. Because experts responded to a structured summary of findings prepared by the authors, their input should be understood as interpretative consultation linked to the authors’ analytical framing of the identified patterns, not as independent validation of raw survey data.

Data analysis

Survey data were exported from Google Forms and analysed in Microsoft Excel using descriptive statistical methods. Statistical significance was not assessed, as the analysis was designed to provide a descriptive characterisation of the data. Given the variability and frequently asymmetric distribution of several parameters, results were presented using descriptive measures, including means, medians, and proportions, without the application of inferential statistical tests. Results were stratified by organisational model, age of the GP, and region.

Written expert responses were analysed using a qualitative interpretative approach. Each author first read the written responses independently and identified comments considered most relevant for interpreting the survey findings. These interpretations were then compared and discussed collectively, and the final use of expert comments in the manuscript was agreed among the authors. The analysis focused on how expert responses contextualised, nuanced, or expanded the authors’ preliminary interpretation of organisational, workforce-related, and coordination-related patterns identified in the survey. Particular attention was paid to areas of convergence, divergence, and added contextual insight across expert perspectives. Interpretation was informed by the multidisciplinary composition of the research team, which included clinical, economic, and comparative health policy perspectives.

Ethics

The study protocol was reviewed and approved by the Ethics Committee of the Faculty of Biomedical Engineering, Czech Technical University in Prague (Approval No. C63/2025) (see Additional File 3).

Results

Respondent characteristics

The survey included 356 GPs for adults. The sample was predominantly female (66.6%). The age distribution was concentrated among middle-aged and older physicians, with a median age of 51 years. The youngest age group (25–30 years) comprised 2.2% of respondents. Participants were recruited from all regions of the CR, with the highest proportions in Prague and the South Moravian Region (see Table 2). The gender distribution closely reflects national workforce data reported by the Ministry of Health [20].

Table 2.

Respondent characteristics

	n	%
Sex
Women	237	66.6
Men	118	33.1
Not reported	1	0.3
Age (years)
Median	51.1
25–30	8	2.2
31–40	66	18.5
41–50	103	28.9
51–60	100	28.1
≥ 61	79	22.2
Region
Capital City of Prague	50	14.0
South Moravian	45	12.6
Central Bohemian	41	11.5
Moravian-Silesian	28	7.9
Ústí nad Labem	26	7.3
South Bohemian	23	6.5
Vysočina	23	6.5
Olomouc	23	6.5
Plzeň	21	5.9
Hradec Králové	20	5.6
Liberec	20	5.6
Zlín	17	4.8
Pardubice	13	3.7
Karlovy Vary	6	1.7

The following sections present findings on organisational models, workforce capacity, task delegation, and care coordination, integrating quantitative results with expert commentary.

Organisational models and workforce capacity

The distribution of organisational models indicates a predominance of solo practices within Czech PC (Table 3). Solo practices represented the predominant model, accounting for approximately 65% of respondents. Of these, 29% were administratively affiliated with a polyclinic or another healthcare facility, while continuing to operate as independent offices. Group and team-based practices accounted for 29.5% of respondents. 3.4% reported employment in network-based practices, and 2% practised as employed GPs.

Table 3.

Distribution of respondents by organisational model

	Solo	Group	Network-based	Employed GPs
n	232	105	12	7
%	65.1	29.5	3.4	2.0

Multidisciplinary organisational arrangements were uncommon in Czech PC. The involvement of physicians from other specialties within the same workplace was reported by 9% of respondents, primarily within traditional polyclinic settings.

Organisational models showed variation in terms of practice size and staffing capacity. Variation in practice scale and internal workforce resources by organisational model is summarised in Table 4. The mean number of full-time equivalent (FTE) GPs per site was highest in network-based practices (3.5 FTEs) and lowest in solo practices (1.0 FTE).

Table 4.

Personnel capacity of practices by organisational model

	Solo	Group	Network-based	Employed GPs
Mean number of GPs (FTE)/practice (SD)	1.0 (0.1)	2.3 (1.2)	3.5 (3.6)	3.3 (1.9)
Median number of GPs (FTE)/practice (IQR)	1.0 (0.0)	2.0 (1.1)	1.0 (3.8)	3.0 (2.8)
Mean number of NPHCP (FTE) per GP (SD)	1.1 (0.6)	1.3 (0.9)	0.9 (1.0)	2.6 (2.3)
Median number of NPHCP (FTE) per GP (IQR)	1.0 (0.0)	1.0 (1.4)	0.8 (0.7)	1.9 (2.5)
Single-GP–single-NPHCP practices (%)	75.9	59.0	75.0	28.6
Practices with > 1 NPHCP per GP (%)	17.7	40.0	8.3	28.6
Practices with administrative staff (%)	14.2	20.0	33.3	57.1
Mean number of administrative staff (FTE) per practice (SD)*	0.8 (0.4)	1.0 (1.4)	1.8 (1.4)	2.3 (1.3)
Median number of administrative staff (FTE) per practice (IQR)*	1.0 (0.8)	0.7 (0.5)	2.0 (2.2)	2.0 (0.8)

Abbreviation: NPHCP Non-physician healthcare personnel

* Values refer only to practices employing administrative staff. Practices without such personnel were not included in the calculation

This variation was also visible in the staffing structure of practices. Overall, the availability of non-physician healthcare personnel was low. The mean ratio of non-physician healthcare personnel to GP FTEs was 1.16. Nearly 70% of practices followed a one-physician–one-non-physician staffing model (most frequently in solo practices and least frequently among employed GPs). Fewer than 25% of practices employed more than 1 non-physician healthcare personnel (FTE)/GP.

Approximately 18% of practices employed administrative personnel, with the highest proportion in practices of employed GPs and the lowest in solo practices.

The delineation of responsibilities for non-physician personnel was predominantly informal. Competencies were most defined verbally (39.9%) or specified in employment contracts or their appendices (27.8%), whereas 9.5% of respondents reported that staff roles were not defined at all.

Expert consultation also helped to contextualise these findings by highlighting concerns about the sustainability of solo practices, particularly in rural areas and the limited institutional support for team-based models.

Expert No. 1 offers the following comment on this finding: “The predominance of solo practices, combined with the unfavorable age structure of physicians, indicates that this model is unsustainable in the long term, particularly in rural areas. In my opinion, the very low proportion of group and team practices and the near absence of multidisciplinary centers significantly reduce the system’s ability to adapt to the growing need for healthcare services”.

In this context, he also described the limited support for group and team practices in the CR as an important structural constraint on the development of primary care. Similarly, Expert No. 5 emphasised that the dominance of individual VPL practices, often with limited personnel and technical resources, may constrain capacity development, care integration, and the adaptability of the system.

Generational patterns in organisational models of practice

Organisational models varied by physician age. A larger proportion of physicians aged 51–60 years and ≥ 61 years worked in solo practices.

Among younger physicians, organisational models were more evenly distributed and characterised by a greater prevalence of group practices and network-based models of care. The proportion of physicians working within group and network-based organisational models was lower in older age groups. The employment model within hospital settings remained marginal across all age groups (see Table 5).

Table 5.

Organisational models by physician age group

Age group (years)	Solo	Group	Network-based	Employed GPs
25–30	37.5%	37.5%	25.0%	0%
31–40	39.3%	39.2%	14.3%	7.1%
41–50	61.5%	34.6%	1.9%	1.9%
51–60	76.0%	23.0%	0%	1.0%
≥ 61	83.3%	16.6%	0%	0%

Percentages are calculated within age groups

Expert consultation also helped to contextualise these age-related differences by pointing to several possible factors associated with younger physicians’ organisational preferences. Younger physicians were described as more likely to prefer team-based models that allow shared responsibilities, more predictable working hours, and a stronger focus on clinical work. Experts also noted that younger physicians were aware of the administrative burden associated with establishing and managing independent practices. Expert No. 5 comments on this: “Younger physicians prefer an employee or shared practice model, which allows for the distribution of responsibilities, predictable working hours, and a better work-life balance. At the same time, they can devote most of their time to actual medical work and do not have to deal with all the administrative tasks associated with running a practice—neither during nor after working hours. However, the current system only reflects these preferences to a limited extent, creating a latent staffing crisis: the system relies on an organizational model that holds little appeal for the next generation”. Experts also suggested that these preferences may be shaped not only by work–life balance considerations, but also by broader concerns regarding the predictability of system support and confidence in ongoing reform efforts. Relatedly, Expert No. 4 pointed to the potential relevance of training focused on the management of general practice and the development of training centres to support future practice leadership.

Practice capacity and task distribution

Non-physician healthcare personnel were primarily involved in basic clinical and administrative tasks, including specimen collection, administration of injections and vaccinations, and anthropometric measurements. Their involvement in patient education and care was reported by 46.1% of respondents. The transfer of prescribing and documentation-related responsibilities to non-physician staff remained relatively uncommon, reported by fewer than 20% of practices.

Approximately 80% of practices did not employ administrative staff, and administrative tasks were typically handled by GPs or non-physician healthcare personnel. Where administrative personnel were present, they performed mainly organisational and financial tasks, patient communication, and documentation processing, with less frequent involvement in clinical activities. Across practices, physicians performed the majority of both clinical and administrative tasks.

These overall patterns were also reflected when examining task distribution by organisational model. Across solo, group, network-based and employed GPs practices, GPs consistently performed the largest share of clinical procedures and prescribing/documentation tasks. Non-physician healthcare personnel contributed substantially to basic clinical activities, administrative tasks and patient education in all practice types, with similar proportions across models. Administrative staff, where present, were primarily engaged in administrative and organisational tasks, with higher involvement observed in network-based and employed practices. Tasks reported as “not performed” remained infrequent across models, although slightly more common in solo practices. Overall, the distribution of responsibilities across staff categories showed comparable patterns regardless of organisational model, with the most notable variation relating to the availability and involvement of administrative personnel (see Table 6).

Table 6.

Distribution of selected tasks by staff category and organisational model

Staff category	Organisational model	Clinical procedures	Administrative/organizational tasks	Patient education and care	Prescribing/documentation tasks
GP	Solo	33.2%	45.8%	50.6%	81.0%
	Group	31.6%	39.9%	54.3%	78.4%
	Network-based	32.3%	30.0%	56.0%	80.0%
	Employed GPs	28.6%	33.9%	63.0%	88.2%
Non-physician healthcare personnel	Solo	57.5%	49.1%	46.0%	18.6%
	Group	60.1%	49.9%	44.9%	19.6%
	Network-based	62.1%	56.2%	40.0%	20.0%
	Employed GPs	61.2%	46.4%	37.0%	11.8%
Administrative staff	Solo	0.2%	1.8%	0.3%	0.4%
	Group	0.7%	6.4%	0.0%	1.5%
	Network-based	0.8%	12.3%	0.0%	0.0%
	Employed GPs	0.0%	14.3%	0.0%	0.0%
Not performed	Solo	9.1%	3.3%	3.1%	0.0%
	Group	7.6%	3.8%	0.8%	0.5%
	Network-based	4.8%	1.5%	4.0%	0.0%
	Employed GPs	10.2%	5.4%	0.0%	0.0%

Expert comments on these findings highlighted the administrative workload as an important constraint on practice functioning and task delegation. They also noted that the absence of administrative support and the limited delegation of responsibilities often result in physicians and non-physician healthcare personnel performing tasks that could be transferred to other support roles. Expert comments further suggested that limited financial capacity may be an important barrier to expanding non-physician and administrative roles, alongside broader organisational and institutional constraints. As Expert No. 5 noted, “The minimal use of non-medical staff reduces the effectiveness of care. The absence of administrative staff and the low level of delegation of responsibilities result in highly qualified physicians routinely performing tasks that could safely be delegated to other professions. But just as physicians should not be performing nursing tasks, nurses should not be overwhelmed with tasks that could be handled by receptionists, administrative staff, and others.”

Access to care, referral patterns, and diagnostic capacity

Indicators of patient load, time allocation, care organisation, and diagnostic resources varied across practice types. The indicator ‘registered patients per GP FTE’ captures the number of registered patients in relation to available GP full-time equivalent capacity; it does not directly measure the overall size of the practice or the total volume of care provided. The mean number of registered patients per GP FTE was 1,726 (median 1,700), ranging from 1,874 in solo practices to 1,028 among employed GPs. The mean number of consultation hours per GP FTE was 27 h per week (median 30 h). The greatest time burden was reported by physicians in network-based practices (37 h weekly), whereas employed GPs and those working in solo practices devoted an average of 27 h per week to direct patient care. Administrative tasks accounted for a mean of 10.3 h per week, with slightly higher values in solo and group practices.

With respect to care organisation, 68.2% of respondents reported using designated time slots for different types of visits, whereas 29.9% did not employ such scheduling structures. Designated appointment slots were most common in group practices (71.9%), followed by solo (67.5%) and network-based practices (66.7%), whereas the proportion was substantially lower among employed GPs (28.6%). While 96.6% of practices reported providing acute care within 24 h, only 18% were able to offer timely non-acute appointments. Waiting times exceeding 48 h for non-urgent visits were most frequently reported among employed GPs (71.4%) and network-based practices (58.3%) (see Table 7).

Table 7.

Practice workload and access indicators by organisational model

	Solo	Group	Network-based	Employed GPs
Registered patients per GP (FTE), mean (SD)	1874 (592.5)	1410 (558.2)	1682 (1063.3)	1028 (470.7)
Registered patients per GP (FTE), median (IQR)	1800 (550.0)	1300 (708.3)	1300 (910.0)	931 (297.0)
Consultation hours per GP (FTE) per week, mean (SD)	27 (4.5)	33 (5.5)	37 (10.9)	27 (7.4)
Administrative hours per GP (FTE) per week, mean (SD)	10.6 (6.1)	10.3 (5.2)	5.6 (4.0)	9.0 (3.6)
Practices with designated visit-specific time slots (%)	67.5	71.9	66.7	28.6
Practices providing acute care within 24 h (%)	97.8	94.3	91.7	100
Practices with waiting time > 48 h for non-urgent visits (%)	51.9	55.3	58.3	71.4

The role of the GP as the formal first point of contact is not legislatively established in the Czech context. Nevertheless, 80% of respondents indicated that specialist services required a referral from a GP in more than half of cases.

In terms of diagnostic capacity, practices were typically equipped with basic instruments, such as point-of-care testing (POCT) analysers, electrocardiography, otoscopy, and pulse oximetry. More advanced diagnostic modalities were available in a minority of practices. Ultrasound examinations were available on-site in 74 practices, while an additional 25 provided these services on a limited basis. Five respondents reported fully independent provision of ultrasound diagnostics. Although 96% of respondents reported access to radiography within seven days, prolonged waiting times for imaging results, extending to several weeks, were reported sporadically in some regions. The mean waiting time for non-acute computed tomography (CT) was 34.7 days and for magnetic resonance imaging (MRI) 54.3 days (see Supplementary Figure S2).

Expert consultation also provided contextual insight into the observed variation in diagnostic access and referral-related practices. Differences in reported waiting times for imaging services, particularly radiography, were attributed to heterogeneous regional conditions. Experts also commented on the limited formalisation of gatekeeping in the Czech system, while noting that many patients nevertheless access specialist care through GPs in practice. Some expert comments further suggested that regional variation in diagnostic access may reflect differences in practice setting, organisational conditions, and the broader coordination of care.

Discussion

This study examined how organisational models of general practice are related to the internal capacity of PC and its ability to coordinate care in the CR. The findings suggest that differences in workforce capacity, care coordination, and selected workload indicators are associated with practice-level organisational characteristics and broader institutional conditions, and cannot be understood solely as the result of individual provider choices. They also point to a growing institutional mismatch between the continued predominance of small, professionally autonomous practices and the preferences of younger physicians for more collaborative or employment-based arrangements. In this study, internal capacity is understood as an organisational property examined indirectly through observable characteristics related to workforce capacity, task delegation, administrative support, diagnostic access, and selected coordination indicators.

Differences in staffing, delegation, and coordination appear to reflect a historically embedded organisational settlement dominated by small, professionally autonomous practices. In line with previous research on the limits of system steerability in settings characterised by strong professional autonomy and small practice size [8, 9], this configuration is associated with a fragmented organisational landscape with limited opportunities for systematic task delegation, team-based care, and stable coordination arrangements.

This organisational configuration has implications for workforce capacity and task distribution within practices. The concentration of clinical, organisational, and administrative responsibilities within a single professional role limits the systematic involvement of non-physician health professionals and administrative staff, increasing workload pressures on GPs [21].

As observed in other PC systems dominated by small autonomous practices, limited task delegation may reflect organisational environments that provide limited infrastructure for expanded team-based models [9, 12]. Expert consultation further suggested that these constraints are particularly pronounced in rural settings, where limited opportunities for collaboration and substitution intensify pressures on workforce capacity, consistent with evidence from rural PC systems elsewhere [22].

Taken together, organisational fragmentation appears to be an important structural factor shaping both internal capacity and the ability of PC to fulfil its coordinating role within the healthcare system. Rather than indicating a clear advantage of any single organisational model, these findings align with evidence highlighting persistent trade-offs: small autonomous practices combine local adaptability with structurally limited internal capacity and heightened vulnerability to workforce shortages, particularly in rural contexts [23, 24]. The analysis does not imply a normative preference for larger or more integrated models, but shows how differences in practice size, staffing, administrative support, and organisational integration are associated with distinct structural capacities and constraints.

Differences in registered patients per GP FTE should not be interpreted as differences in overall practice size. The indicator reflects patient load relative to available GP capacity, not the total registered population of a practice. Lower patient numbers per GP FTE in group or employed practices therefore do not necessarily imply that these practices manage fewer patients overall. Conversely, higher patient numbers per GP FTE in solo practices may signal a greater concentration of workload on individual physicians and lower organisational redundancy. Such practices may carry larger patient lists per physician without necessarily having greater ability to absorb additional demand, physician absence, administrative workload, or coordination needs. Expert consultation also highlighted that rural practices may operate under different diagnostic, referral, and reporting conditions than urban practices, suggesting that part of the observed variation in patient load and accessibility may reflect geographic context rather than organisational model alone. Because GPs in the Czech Republic do not operate as formal gatekeepers, the number of registered patients per GP FTE cannot be interpreted in the same way as in systems where primary care controls access to specialist services. As one expert noted, gatekeeping is not strongly institutionalised in the Czech insurance-based system, yet most patients still enter the system through their GP, indicating that patient flow patterns do not necessarily reflect formal gatekeeping structures.

Beyond organisational fragmentation, limited internal capacity appears to be a closely related factor shaping PC functioning. In this context, internal capacity refers not to physician numbers but to the organisational ability to allocate tasks, delegate responsibilities, and integrate non-physician staff into everyday care processes. At the organisational level, internal capacity helps connect system-level expectations for coordination and the practical realities of everyday care delivery. This interpretation aligns with comparative evidence demonstrating the central role of team composition and practice-level organisational characteristics in structuring PC work [12]. In systems dominated by small, autonomous practices, such capacity may remain structurally constrained.

Limited delegation may reflect organisational environments lacking administrative support, stable team arrangements, and clearly institutionalised roles for non-physician personnel [25]. Cross-national evidence further shows that delegation is associated primarily with practice size, team composition, and organisational infrastructure more consistently than with individual attitudes [26]. Attempts to expand delegation in the absence of such infrastructure may increase coordination costs and workload rather than reduce them, as supervisory and accountability responsibilities remain concentrated at the physician level [27]. This interpretation is further supported by mixed-methods research indicating that cumulative administrative workload and organisational change contribute to decisions to leave general practice or reduce clinical commitment [28].

The broadly similar distribution of tasks across organisational models is therefore an important finding in itself. Although group and network-based practices may provide somewhat greater access to administrative staff, the results did not indicate substantially broader redistribution of clinical, prescribing, or documentation-related tasks. Expert consultation suggested that financial constraints and the absence of dedicated funding for non-clinical and most non-physician roles restrict the development of broader team structures. Current reimbursement arrangements primarily support physician work and only partially remunerate nursing activities, while offering no systematic support for other health or administrative professions. This limits the capacity of practices to employ additional staff and constrains opportunities for meaningful task redistribution. These organisational and institutional conditions may help explain why group and network-based practices did not exhibit substantially higher levels of delegation, despite intuitive expectations to the contrary.

From this perspective, internal capacity can be understood as an organisational property rather than an individual capability, dependent on stable team structures, administrative support, and embedded non-physician roles that enable sustainable delegation in everyday practice. A similar structural pattern has been observed in nurse-led PC models operating in fragmented healthcare systems [13].

Expert consultation identified administrative workload as a central feature of GP practices and suggested that it may shape the scope for task delegation. One expert described a “competency cascade”, whereby responsibilities are transferred from specialists to GPs and subsequently to non-physician staff.

The limited presence of roles commonly established in other health systems was also described as a sign of institutional rigidity that may limit adaptation to increasing patient complexity.

A generational dimension also becomes visible in interaction with the organisational constraints identified above. Younger physicians are more frequently oriented towards collaborative, group-based, or employment-based arrangements, whereas older physicians remain predominantly in solo practice models. This pattern is consistent with international evidence documenting changing career preferences among younger GPs, including stronger preferences for group-based practices, salaried or hybrid employment arrangements, and reduced administrative burden [22, 29–31].

These generational differences point to a growing institutional mismatch between the historically dominant model of independent practice ownership and contemporary career trajectories. This shift may reflect changing preferences in relation to organisational models that concentrate clinical, administrative, and entrepreneurial responsibilities within a single role. Generational change thus highlights the limited adaptability of practice models anchored in individual ownership and raises questions about the long-term sustainability of PC systems that fail to align organisational capacity with workforce expectations [32].

With respect to care coordination, our findings point to structural limitations in PC’s ability to fulfil its coordinating role within the healthcare system. GPs are widely involved in coordinating patient pathways, yet this role is not consistently supported by clearly defined authority, adequate access to diagnostic resources, or protected time for coordination activities.

Comparative research shows that the institutionalisation of gatekeeping varies substantially across health system types and is less consistently formalised in insurance-based systems, where coordination responsibilities are often distributed across multiple actors rather than clearly anchored in PC [33, 34]. In such contexts, gatekeeping tends to operate through partial, voluntary, or incentive-based arrangements rather than as a clearly institutionalised mechanism, resulting in heterogeneous and often informal coordination practices [35].

These coordination gaps may reflect a mismatch between the coordination responsibilities assigned to PC and the institutional resources, diagnostic access, and protected time available to perform them effectively. At the same time, OECD PaRIS data indicate that patients in CR report comparatively high satisfaction with care coordination in international comparison [18]. This contrast suggests that patient-perceived coordination may reflect relational continuity and professional commitment rather than systematically institutionalised coordination mechanisms, and therefore does not contradict the structural constraints identified in this study.

Taken together, these coordination problems appear to reflect the combined effects of organisational fragmentation, limited internal capacity, and constrained task delegation across organisational models of general practice. At the system level, the findings point to a policy paradox in the organisation of PC [36]. Although integrated care, interdisciplinary collaboration, and strengthened coordination are widely articulated as policy objectives [3, 4], these expectations are layered onto practice structures with limited staffing, administrative support, and coordination capacity. As a result, organisational fragmentation continues to shape how care is delivered in practice, with implications for equity, continuity, and the predictability of patient pathways.

This study suggests that limitations in delegation, workforce use, and coordination are shaped by the interaction between practice-level organisational characteristics and broader institutional conditions. The findings also indicate that internal capacity and coordination can be understood as organisational properties shaped by practice size, staffing, administrative support, patient load, geographic context, and broader institutional conditions, which together influence what PC is able to deliver in practice.

Expert consultation also reflected a more cautious assessment, suggesting that without targeted interventions in organisational and institutional structures, improvements in accessibility, effectiveness, and equity may remain limited.

Several limitations should be considered when interpreting these findings. The analysis relies on self-reported data and may therefore be subject to reporting bias. Another significant limitation is the low number of physicians (6.6%) who completed the questionnaire and thus participated in the study. In addition, the study was based on a non-probabilistic, self-selected sample, and the findings should therefore be interpreted as reflecting the characteristics of respondents rather than those of the full GP population. The sample included a somewhat lower proportion of physicians aged 60 years and older compared with national workforce data, which may affect the interpretation of age-related patterns. The distribution of respondents across organisational models was uneven, resulting in smaller quantitative representation of some practice types. The study focuses on a single national context, limiting direct generalisability to other health systems. In addition, the study design does not permit causal inference regarding relationships between organisational models and care outcomes. However, the aim of the study is analytical rather than causal, andthe findings remain relevant for understanding how practice-level organisational characteristics and broader institutional conditions may shape internal capacity and care coordination in this context.

Conclusions

This study suggests that organisational models of general practice are important in shaping the internal capacity of PC and its ability to coordinate care. Drawing on evidence from the CR, the findings indicate that variation in workforce capacity, administrative support, patient load, and coordination-related conditions is associated with practice-level organisational characteristics and broader institutional conditions. At the same time, task distribution patterns remained broadly similar across organisational models, suggesting that delegation is constrained across practice types and cannot be understood solely as the result of individual professional choices.

Conceptualising internal capacity as an organisational property highlights how limitations in delegation and coordination may reflect fragmented practice structures and limited institutional support. These findings suggest that efforts to strengthen care coordination and integration may remain limited unless the organisational foundations of general practice are addressed. The findings also point to a growing institutional mismatch between traditional ownership-based practice models and the preferences of younger physicians for more collaborative or employment-based arrangements. Although focused on a single national context, the organisational dynamics highlighted in this study may also be relevant to other health systems facing similar constraints.

Abbreviations

CR

Czech Republic

GP

General Practitioner

PC

Primary Care

FTE

Full-time equivalent

Authors’ contributions

KH: Investigation; Formal analysis; Data curation; Visualization; Writing – review & editing. MC: Investigation; Writing – review & editing. ZK: Conceptualization; Methodology; Investigation; Writing – original draft; Writing – review & editing; Supervision. All authors contributed to the interpretation of the results and approved the final manuscript.

Funding

This research received no external funding.

Data availability

The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study protocol was reviewed and approved by the Ethics Committee of the Faculty of Biomedical Engineering, Czech Technical University in Prague (Approval No. C63/2025). The study was conducted in accordance with the Declaration of Helsinki.

Participants were informed about the purpose of the study and participation was voluntary and anonymous. Completion of the questionnaire was considered as provision of informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.