Step 1: Review and tuning of the PHAMEU framework and indicators

Following Donabedian, the PHAMEU framework distinguishes the broad domains structure, process and outcomes, which are broken down to dimensions, as depicted in Figure 2 (Donabedian, Reference Donabedian1988, Kringos et al., Reference Kringos, Boerma, Bourgueil, Cartier, Hasvold, Hutchinson, Lember, Oleszczyk, Pavlic, Svab, Tedeschi, Wilson, Windak, Dedeu and Wilm2010, EXPH, 2017).

Figure 2.

The PHAMEU framework of primary care.

We reviewed the framework in line with our requirements, resulting in the following decisions. The three dimensions of Domain 1 Structure of primary care (Governance; Economic conditions of primary care; Primary care Workforce Development) have been maintained.

Concerning Domain 2 (Process of primary care) we made a distinction between indicators that should be measured at system level and others that can more suitably be measured through surveys among providers and patients. Our main criterion was whether process indicators vary between care providers within countries or not. Only if they do not (substantially) vary, we have adopted them as system characteristics. The application of this criterion led to the removal of 24 indicators. Those we maintained were labelled ‘Systemic aspects of care delivery’. We use the term ‘systemic’ to emphasize that these indicators are measured at (national) health system level and not at the level of primary care practices.

Domain 3 (Outcomes of primary care) was excluded altogether as it does not measure the strength of primary care but the consequences of strong primary care. This led to the removal of 30 indicators on outcomes.

We identified a gap in the framework related to Domain 1 (Structure of primary care). We added indicators, covering national regulation that facilitate service delivery and local management. Such regulation can provide important incentives to organize the delivery of care professionally, e.g. by stimulating local practice management. Since the conceptualization of PHAMEU, regulatory aspects of primary care delivery have changed. We therefore included this new domain, and labelled it ‘Systemic aspects of facility management’.

In addition to the critical review of the framework, the PHAMEU indicators have been analyzed statistically to assess their contribution to the overall dimensions by computing item-total correlations. In previous studies, separate indicator items have rarely been used to characterize primary care systems. Usually, they have been combined into the dimensions through latent variable analysis. We have excluded 23 indicators with low correlations to the total score for each dimension. An example of an indicator, removed based on this analysis, is one about whether primary care has a separate department or unit within the Ministry of Health. This indicator had a low correlation to the overall dimension of Governance. The review of the PHAMEU framework (as described above) and the statistical analysis on the indicators’ contribution to the latent variables altogether resulted in exclusion of 77 from the original 122 indicators of PHAMEU.

Step 2: Merging the remaining indicators of PHAMEU with WHO PHC-IMPACT

WHO PHC-IMPACT encompasses a comprehensive set of indicators, developed by the World Health Organization Regional Office for Europe and aiming to generate performance intelligence to strengthen and monitor the potentials of primary care for the benefit of Universal Health Coverage (Barbazza et al., Reference Barbazza, Kringos, Kruse, Klazinga and Tello2019). For the practical use by countries, the indicators are provided in a so-called Indicator Passport, structured in domains, features and indicator questions. The main components of the WHO PHC-IMPACT framework are: Capacity, Performance and Impact of primary care (from right to left in Figure 3). From broad to specific, each of these components have been structured in six domains, 26 subdomains and 63 features, which have eventually been operationalized into 139 indicators. As the purpose of SiSPC is to characterize primary care specifically at the national level, the component of ‘Capacity’ and of ‘Performance’ (as far as process indicators at system level are concerned) are most relevant here.

Figure 3.

The framework underpinning PHC-IMPACT.

The evidence-base of PHC-IMPACT started from the PHAMEU systematic review which was updated by a literature review to identify frameworks and tools published between 2010 and the date of the search (2016–17) (Barbazza et al., Reference Barbazza, Kringos, Kruse, Klazinga and Tello2019).

It turned out that 21 indicators that we included from the PHAMEU list were (more or less) the same in PHC-IMPACT. Indicators that were in PHC-IMPACT, but not in the remaining PHAMEU indicator set, and satisfied our requirements, related to:

• - A licensing system for family physicians (FPs) or primary care nurses

• - Support for family carers

• - Primary care benefit package

• - Tasks of FPs formally defined

• - Age balance in active FPs

• - Time FP trainees spend in primary care

• - Predominant practice settings of FPs

• - Quality assurance mechanisms

• - Patient experiences measured and published

• - Quality of care processes implemented in primary care practices

• - Role for representatives of citizens/patients

• - Screening programmes operating via primary care

• - Prescriptions by FP without specialist recommendation

These topics have been taken into consideration to be elaborated in indicators for SiSPC.

Step 3: Searching for other frameworks and extracting relevant elements

From our scan of the literature, we evaluated nine frameworks. They revealed new domains, new dimensions and related new indicators.

Within the domain of the Structure of primary care, we added the dimension of ‘Information structures’, to explicitly cover issues like access to and use of shared medical records and a learning primary care system.

Within the (new) domain of ‘Systemic aspects of facility management’, professionalization of practice management resulted as a potential field of interest. Furthermore, ‘Composition of the primary care team’ was mentioned as potentially relevant in Systemic aspects of service delivery.

Finally, we found a few indicator fields that, in this stage, did not fit well in the proposed domains (see Table 1). We parked them – provisionally – in the domain ‘Not fitting in above Domains’. In later stages of the project, as domains further got their shape, it turned out that these indicator fields could be absorbed by one of the other domains.

Table 1.

Overview of domains and indicator fields from other primary care-relevant frameworks

The results of our third step, screening of other frameworks, are summarized in Table 1.

In screening the other frameworks, we found a category of indicators which was missing both in PHAMEU and PHC-IMPACT: indicators for the context in which primary care operates. We included ‘Context’ in our framework, be it with a special status, as these indicators do not inform about the strength of primary care in a specific country. Nevertheless, they are important in understanding why primary care is stronger in some countries than in others. Context aspects concern the broader health system; the political, social and cultural influences on the primary care system (Sidel and Sidel, Reference Sidel and Sidel1977, Kringos et al., Reference Kringos, Boerma, Van Der Zee and Groenewegen2013b), and the non-health care determinants of health (Whitehead, Reference Whitehead1990, Whitehead and Dahlgren, Reference Whitehead and Dahlgren2006), as shown in Table 2.

Table 2.

Overview of indicator fields on context

Step 4: Integrating topical issues

The fourth step resulted from our aim to have an up-to-date system of indicators, reflecting current (and future) challenges and needs in primary care systems that are not already incorporated in previously considered frameworks. Topical issues in primary care have been identified through a review of authoritative reports of professional and international organizations that identify insights, visions and innovations that should make primary care fit for the future.

We have identified 14 challenges and issues that play (or will play) a role in primary care. These are listed below with our decision whether each of them were covered or not by the current system of indicators. If not, one or more new indicators were added.

• Health system resilience, dealing with pandemics (Kruk et al., Reference Kruk, Myers, Varpilah and Dahn2015, Thomas et al., Reference Thomas, Sagan, Larkin, Cylus, Figueras and Karanikolos2020, EXPH, 2020, OECD, 2021). Resilience of primary care is covered by a number of separate indicators in the current set.

• Environmental footprint of health care (Lenzen et al., Reference Lenzen, Malik, Li, Fry, Weisz, Pichler, Suveges Moreira Chaves, Capon and Pencheon2020, Gonzalez-Holguera et al., Reference Gonzalez-Holguera, Gaille, Del Rio Carral, Steinberger, Marti, Bühler, Kaufmann, Chiapperino, Vicedo Cabrera, Schwarz, Depoux, Panese, Chèvre and Senn2022, Klemenc Ketis and Rochfort, Reference Klemenc Ketis and Rochfort2022). A new indicator was added to Primary care context.

• Primary care in depopulating regions (Bosmans et al., Reference Bosmans, Boerma and Groenewegen2021) and shortage of primary care staff (WHO, 2016a, OECD, 2020). This is covered by indicators on incentives for working in remote areas.

• e-Health care (OECD, 2020). Indicators related to e-health access to primary care practices and the availability of reliable online patient information were added.

• Community involvement (DeCamp et al., Reference Decamp, Dukhanin, Hebert, Himmelrich, Feeser and Berkowitz2019, Eder et al., Reference Eder, Carter-Edwards, Hurd, Rumala and Wallerstein2013, Modigh et al., Reference Modigh, Sampaio, Moberg and Fredriksson2021, Sharma and Grumbach, Reference Sharma and Grumbach2017). This is covered by a number of indicators in the current set.

• Dealing with multimorbidity (Adan et al., Reference Adan, Gillies, Tyrer and Khunti2020, WHO, 2016b, Rijken et al., Reference Rijken, Hujala, Van Ginneken, Melchiorre, Groenewegen and Schellevis2018). How multimorbidity is dealt with in primary care largely varies between providers and, therefore, it can better be measured at provider or practice level. No addition is needed.

• Mental health and primary care (Smit et al., Reference Smit, Hill, Walton, Kendall and De Lepeleire2020). This field is covered by an indicator on coverage in national policy documents and by an indicator on the role of primary care in access to specialized mental health care. No addition is needed.

• Primary care and social care collaboration (RCGP, 2019, EuroHealthNet, 2022). This field is covered by an indicator on coverage in national policy documents. No addition is needed.

• Continuity of care out-of-hours (RCGP, 2016, Steeman et al., Reference Steeman, Uijen, Plat, Huibers, Smits and Giesen2020). This is covered by an indicator on family physician involvement in out-of-hours care. No addition is needed.

• Care team well-being, Quadruple Aim (Bodenheimer and Sinsky, Reference Bodenheimer and Sinsky2014). Indicators on the work-private balance were added.

• Social prescribing (NHS, 2019, Buck and Ewbank, Reference Buck and Ewbank2020, WHO, 2022). We have added social prescribing under ‘Systemic aspects of service provision’.

• Extension of family physician training duration (RCGP, 2019, Burgmann et al., Reference Burgmann, Paier-Abuzahra, Sprenger, Avian and Siebenhofer2023). This is represented by an indicator on the length of family medicine training. No addition is needed.

• Population health approach, outreaching (RCGP, 2019, De Maeseneer, Reference De Maeseneer2017, Jiao et al., Reference Jiao, Slemon, Guta and Bungay2022, NASEM, 2023, RVS, 2023). Important conditions for such approaches are registration of patients with a primary care provider (list system) and availability and exchange of relevant patient/population data. As these are covered no addition is needed.

• Learning health system based on primary care data (IOM, 2007, Friedman et al., Reference Friedman, Rubin, Brown, Buntin, Corn, Etheredge, Gunter, Musen, Platt, Stead, Sullivan and Van Houweling2015). This is sufficiently covered as a result of the previous steps, in particular in the dimension information structures within the domain of structure of primary care. No addition is needed.

Step 5: Processing feedback on the development process and provisional indicator system

The fifth and final step in the process of developing SiSPC consisted of consulting experts and processing their feedback. The consortium partners who implemented the PaRIS project, provided input online throughout the project and at physical meetings in September 2023 and 2024. In an early phase of the development of SiSPC, we organized a workshop at the yearly conference of EFPC, attended by some twenty participants from various countries (including e.g. Austria, Azerbaijan, Belgium, and France) and with backgrounds in general practice, policy making and research. Some of the challenges to strengthening primary care, mentioned in the workshop, were:

• - Fragmentation of payment systems for primary care providers, leading to less efficient and less satisfactory organization of primary care;

• - Unclear roles and responsibilities of stakeholders involved in the governance of primary care;

• - Availability and implementation of self-monitoring devices for patients with chronic conditions;

• - Use of primary care-generated information for surveillance;

• - Organization of out-of-hours services regarding to the role of FPs and accessibility to patients.

Successes of primary care in participants’ countries were mentioned as well, in particular:

• - Availability and use of guidelines;

• - Availability of (shared) data from electronic medical files.

Concerning community participation in primary care, the following issues were suggested:

• - Community participation is relevant both at the level of primary care practices and in health care institutions at the national level (e.g. insurance bodies);

• - The importance of non-disease-specific, national patient platforms;

• - The existence of legislation or national support for patient councils at primary care practice level.

In the December 2024, the draft of SiSPC was sent to experts in 34 countries. By the end of February 2025 and after only one reminder, experts from 25 countries had responded (see the acknowledgement at the end of this article). They all were positive about the initiative to develop a new system of indicators for the strength of primary care. More general remarks and suggestions that were made concern:

• - Balanced focus

A general problem identified is whether there is enough focus on chronic disease or too much. The PaRIS project focussed on primary care for people living with chronic conditions. However, primary care also has an important role in prevention and care for acute episodes. SiSPC aims to provide a balance between attention to chronic care and other roles of primary care. Many of the aspects of strong primary care are especially relevant for people with a chronic disease (e.g., coordination and comprehensiveness of primary care). At the same time, we have included indicator items relating to prevention and acute care.

• - Focus on FPs

The indicators that constitute SiSPC mainly refer to FPs. Ideally, all other professions active in primary care teams should also be considered. We have concluded that this is unfeasible for several reasons. First, FPs are still the backbone of primary care; secondly, equal attention to all other primary care disciplines would multiply the number of indicators, while there is little information on other professions. However, we have devoted attention to practice nurses, whose profession is increasingly important for strong primary care, especially in the care for people with a chronic disease.

• - The time dimension

The SiSPC indicator system does not contain items on changes over time. The measurement of the SiSPC indicator system will represent the situation in the early 2020’s. A first indication of changes may come from a comparison with the PHAMEU indicator system. However, more importantly, we hope that SiSPC will be updated in line with future rounds of the PaRIS project. This will show future changes in the strength of primary care. Basically, SiSPC will provide repeated cross-sections.

• - Relevance of answering categories

In some countries, the answering categories for indicator items do not cover the specific situation in that country. Reviewers suggested that there could be options for ‘Not applicable’, ‘Partial’ or ‘no available data’ for some indicators. Also, we should provide the option ‘Other’ with the possibility of free-text input more often. We have reviewed all indicator items to see whether additional answering options are needed. Related to this is how to deal with federal countries that have differences in their primary care system in federal states or autonomous provinces. Here, the focus remains on the communalities at country level.

• - A core set of indicators

It was suggested to define a core set of indicators for the strength of primary care. We are not in favour of this. The indicator items refer to dimensions of strong primary care and after data collection will be combined through a statistical model. A core set of indicators may be easier for data collection but would miss out important areas. The statistical analysis will provide information on which indicator items are less relevant for an overarching dimension and that can be skipped in future data collections.

Specific additions resulting from the experts’ remarks are listed in Table 3. We left out smaller changes in wording of indicator items and in the answering options.

Table 3.

Feedback resulting in an removal, addition or change by domain and dimension

The resulting full system of SiSPC indicators

As a result of the five steps, SiSPC consists of the following four domains (and dimensions within these domains):

• - Context of Primary Care (Population, Economy, Social- and cultural values, Welfare benefits and social protection, Education (-related) resources, Life style, Health system overall);

• - Structure of Primary Care (Governance, Economic & Financial Conditions, Workforce Development, Information Structures);

• - Systemic Aspects of Facility Management (Scale of Primary Care Delivery, Systems /structures for Quality Assurance and Safety, Practice Management Incentives, Community Involvement);

• - Systemic Aspects of Care Delivery (Accessibility, Comprehensiveness, Continuity, Coordination).

The full system of indicators is provided in the appendix; a selection of one item for each dimension is given in Table 4.

Table 4.

Overview of the domains and dimensions (with the number of indicators) and one exemplary SiSPC indicator for each dimension and their response categories