Primary care: the four attributes that define it

A middle-aged woman goes to the service closest to her home because of fatigue that has lasted for weeks. She is seen by someone who has never met her, leaves with a request for tests and the advice to see a specialist. She comes back three months later, for another appointment with another person, and nobody knows where the tests ended up. She had access to a health service. She did not have primary care in the sense the word carries in the literature.

Primary care is not a synonym for a cheap service, a small clinic or a quick appointment. It is a level of care defined by functions that can be described and measured. This text presents those functions, shows how researchers described them as assessable dimensions and discusses what the evidence associates with strong primary care, along with the limits of that association.

Key points

• Primary care is defined by four core attributes: first contact, longitudinality, comprehensiveness and coordination.
• The attributes work together; a service can be accessible and still fragmented.
• First contact is not just opening the door: it depends on the service and on the person's ability to reach and use care.
• Longitudinality is the relationship over time, not the number of appointments.
• Comprehensiveness is handling most needs at that level itself, and recognizing what does not belong there.
• Coordination is answering for the patient's path when they need other services.
• Systems with strong primary care are associated with better health indicators, but most of this evidence is ecological and subject to confounding.

Summary

• From Alma-Ata to Astana: where the idea comes from
• The four Starfield attributes
• First contact: the entry point and access
• Longitudinality: the relationship over time
• Comprehensiveness: handling most needs in the same place
• Coordination: when care needs to leave primary care
• How to measure the strength of primary care
• What the evidence associates with strong primary care and its limits
• Frequently asked questions
• Final checklist

From Alma-Ata to Astana: where the idea comes from

The most cited definition was born at an international conference. The Alma-Ata Declaration described primary care as the first level of contact of individuals, families and communities with the health system, bringing care close to where people live and work, and as the first element of a continuous process of care (WHO, 1978). Two ideas that return throughout this text are already there: entry point and continuity.

Forty years later, the Astana Declaration renewed the commitment. The signatories stated that strengthening primary care is the most inclusive, effective and efficient path to improving physical and mental health and social well-being, and that it is the cornerstone of a sustainable system for universal health coverage (WHO and UNICEF, 2018). The document also acknowledges that health needs remain unanswered in all parts of the world and that fragmented, unsafe or low-quality care can no longer be tolerated (WHO and UNICEF, 2018).

The declarations set the political direction, but they do not say how to recognize, in a concrete facility, whether primary care works. For that, research needed operational attributes. The most influential list comes from the work of Barbara Starfield and colleagues, revisited in a review that became a reference (Starfield, 2005).

The contrast with other settings helps show what is at stake. In humanitarian crises, the priority is to measure and reduce excess deaths, as discussed in the text on why the mortality measure matters to the clinician. Primary care is about continuous, basic care that exists before and after crises and that defines how much the system can handle day to day.

The four Starfield attributes

Starfield, Shi and Macinko summarized primary care in four main features: first-contact access for each new need; long-term care centered on the person, not the disease; complete care for most needs; and coordinated care when the person needs to be seen elsewhere (Starfield, 2005). Primary care is considered good to the extent that it fulfills these four functions (Starfield, 2005).

For some purposes, the authors add guidance for the family and the community (Starfield, 2005). A later review recalls that the global goal proposed by the WHO in 2001 included six domains: first contact, longitudinality, comprehensiveness, coordination, person- or family-centred care and community orientation (Shi, 2012). The first four form the core; the other two widen the focus from the individual to the group.

The central point is that the attributes reinforce one another. First contact without longitudinality produces episodic care. Longitudinality without comprehensiveness produces a relationship in which almost everything is referred. Comprehensiveness without coordination leaves the patient alone when they need to leave the service. The table summarises how each attribute appears in practice and how to recognise when it is missing.

Attribute	What it means	Sign that it is present	Sign that it is missing
First contact	The service is the entry point for each new need	The person knows where to go first and manages to be seen in time	New problems go straight to the emergency department or the specialist
Longitudinality	A continuous relationship with a professional or team over time	The team knows the patient's history and baseline	Each appointment starts from scratch, with someone else
Comprehensiveness	The service handles most needs, from prevention to palliative care	Promotion, prevention, treatment and rehabilitation happen in the same place	Almost every complaint becomes a referral
Coordination	The service answers for the patient's path when other levels come in	There is a referral letter, information sent back and a revised plan	Tests and opinions get lost between services

A consensus study with experts reinforces this reading. Among 25 operationalised primary care attributes, only five were considered specific to this level: first-contact accessibility, relational continuity, family-centred care, population orientation and intersectoral work (Haggerty, 2007). The others, such as technical quality and communication, matter at any level of care.

First contact: the entry point and access

The first attribute says that primary care should be the place sought first for each new problem. The expert consensus defined first-contact accessibility as the ease with which a person obtains the needed care, including guidance and support, with the professional of their choice, within a time frame suited to the urgency of the problem (Haggerty, 2007). The definition has three parts: ease, choice and time.

The same group separated this idea from another, accommodation: the way the service organises hours, scheduling and walk-in care to receive people with different capacities for contact (Haggerty, 2007). A service can be nearby and formally free and still fail at first contact because it only schedules at a time when the person is working.

A widely used conceptual model shows that access arises from the interaction between the service and the person. On the service side, the authors describe five dimensions: being recognisable, being acceptable, being available and accommodating, being financially possible and being adequate (Levesque, 2013). On the population side, they describe five corresponding abilities: perceiving the need, seeking care, reaching the service, paying and engaging in care (Levesque, 2013).

This model helps the clinician see barriers that do not show up in the schedule. It is worth asking whether the person knows they can come to this service for this type of problem, whether the hours, the transport and the way of booking an appointment fit their routine and whether there is a direct or indirect cost, such as missing work, that delays seeking care. It also matters to know whether the service is culturally acceptable and speaks a language the person understands and whether, after arriving, they can take part in decisions and follow the plan.

The language barrier is a concrete example of acceptability. The text on the impact of professional interpreters on outcomes shows how communication changes what happens after first contact.

First contact also has a filtering function. A review article describes, among the roles of primary care, coordinating all health problems, including acting as a filter for referral to more specialized services (van Weel, 2018). When the front door works, the specialist receives cases that are already organized; when it fails, the specialist receives whatever arrived by chance.

Longitudinality: the relationship over time

Longitudinality is person-centered care over time. A multidisciplinary review identified three types of continuity present in all the disciplines studied (Haggerty, 2003):

• informational continuity: the use of information about past events and personal circumstances to make current care appropriate to each person;
• management continuity: a consistent and coherent approach to a health problem, one that responds to changes in the patient's needs;
• relational continuity: an ongoing therapeutic relationship between the patient and one or more professionals.

The authors point out that time distinguishes continuity from other attributes, such as the quality of communication in a single consultation (Haggerty, 2003). An isolated consultation can be excellent and still have nothing longitudinal about it. They also warn that many measures count chronological patterns of visits without measuring whether the patient experiences connected and coherent care (Haggerty, 2003).

The Starfield review lists benefits associated with prior knowledge of the patient. Knowing the patient increased the chance that the physician would recognize psychosocial problems that affect health, and continuity and first contact were associated with more efficient consultations, fewer laboratory tests and lower costs (Starfield, 2005). For clinical reasoning, this means a baseline: the professional knows how that person usually is and notices the change.

The ongoing relationship also changes the conversation. Listening for the cues the patient offers, discussed in the text about the cues the patient offers, becomes easier when there is a shared history. Difficult conversations, such as those about end-of-life care planning, depend on trust built earlier, as the text about how to conduct the planning conversation shows.

There is also a diagnostic advantage that is rarely remembered. When the team is going to review the same patient, it can plan the reassessment instead of resolving everything in a single consultation, a strategy described in the text about planned reassessment. Longitudinality is what makes that plan viable.

Comprehensiveness: solving most of it in the same place

Comprehensiveness is the capacity to respond to most of a population's health needs at the level of care itself. The expert consensus defined comprehensiveness of services as the offer, direct or indirect, of a full range of services: health promotion, prevention, diagnosis and treatment of common conditions, referral, management of chronic diseases, rehabilitation and palliative care (Haggerty, 2007).

The same work separated comprehensiveness of services from care of the whole person. The latter was defined as how much the professional investigates and considers the physical, emotional and social aspects of health and the context of the community (Haggerty, 2007). A service can offer many procedures and still treat the person as a sum of organs.

In international comparisons, comprehensiveness was one of the traits that separated countries with strong primary care from the rest. In the Starfield review, the characteristics present in high-scoring countries and absent in low-scoring ones were the degree of comprehensiveness, that is, how much professionals offered a broad range of services instead of referring, and family orientation (Starfield, 2005).

Comprehensiveness has a direct clinical consequence. The more the service resolves, the less the patient circulates among specialists, and the less they are exposed to tests and interventions that do not change management. Starfield includes among the mechanisms of benefit of primary care the role of reducing unnecessary and potentially harmful specialized care (Starfield, 2005). The text on how the harm of overdiagnosis reaches the patient shows the other side: excess investigation also hurts.

Comprehensiveness does not depend on doctors alone. In many systems, part of the care is done by other trained professionals, and this expands what the unit can resolve. The text on the conditions for task shifting to work discusses what is needed so that this division does not reduce quality.

Coordination: when care needs to leave primary care

Every service has limits. Coordination is the attribute that ensures that, when the patient needs another level, someone is accountable for the path. Starfield describes it as coordinated care when it must be sought elsewhere (Starfield, 2005). In practice, coordination brings together information and management: knowing what happened outside and adjusting the plan based on that.

Coordination connects with two types of continuity. Informational continuity ensures that the specialist's result comes back and is used; management continuity ensures that the plans of different professionals do not contradict each other (Haggerty, 2003). When one of these threads breaks, the patient becomes the only bearer of their own history.

The signs of fragile coordination tend to appear in routine: referrals without an explicit clinical question, test results ordered by another service that no one reviews, different medication lists in each service, the specialist's reply that does not reach the original team, and the patient who has to repeat the whole history at every visit.

Teamwork within the unit is part of this. The text on what gets in the way of team reasoning shows how communication failures between professionals become failures of care. In extreme situations, such as disasters, coordination of chronic disease care becomes even more visible, as described in the text on what to do after the first week.

How to measure the strength of primary care

Assessing primary care requires describing each attribute through indicators. A systematic review analyzed 85 studies published between 2003 and 2008 and identified ten central dimensions of a primary care system (Kringos, 2010). Structure brings together three dimensions: governance, economic conditions and workforce development. Process brings together four: access, continuity, coordination and comprehensiveness. Outcomes bring together three: quality, efficiency and equity (Kringos, 2010).

The model makes clear that Starfield's four attributes are the process, what happens between professional and patient, and that they depend on a structure that sustains them. Without funding, without professionals and without clear access rules, the attributes do not hold up.

The same researchers applied a standardized instrument in 31 European countries, with 77 structure and process indicators (Kringos, 2013). One finding stands out: there was no correlation between access, continuity, coordination and comprehensiveness across countries (Kringos, 2013). That is, being good in one attribute does not guarantee being good in the others. A system can have easy access and poor coordination, and vice versa.

In low- and middle-income countries, an international initiative launched in 2015 by multilateral organizations and a philanthropic foundation proposed a conceptual model to measure and improve primary care (Bitton, 2017). The authors, who are part of this initiative, argue that the central functions are universal: services that are accessible at first contact, continuous, coordinated, complete and person-centered (Bitton, 2017). The fact that they are part of the project they describe is a point to consider when reading.

The source of the information also matters. In expert consensus, some attributes are better measured by patient report, such as first contact and relational continuity, and others by administrative data or the medical record (Haggerty, 2007). Asking the patient whether they have a usual place and a usual person remains one of the most direct ways to assess primary care.

What the evidence links to strong primary care, and its limits

The Starfield review gathered studies on the supply of primary care physicians, on people with a regular source of care and on comparisons between countries (Starfield, 2005). In a comparison of 11 industrialized countries, those with stronger primary care generally had healthier populations, and certain policy aspects proved important for sustaining that strength (Starfield, 2005).

A study of 18 high-income countries, from 1970 to 1998, totaled 504 observations and used fixed-effects regression (Macinko, 2003). The strength of the primary care system was associated with lower overall mortality, lower premature mortality and lower premature mortality from some causes, such as cardiovascular disease and respiratory diseases, even after adjustment for national income, total number of physicians and proportion of older adults (Macinko, 2003).

Starfield proposes six mechanisms that would explain the benefit, alone or in combination (Starfield, 2005):

1. greater access to needed services;
2. better quality of care;
3. more focus on prevention;
4. early management of health problems;
5. cumulative effect of the main features of primary care;
6. reduction of unnecessary and potentially harmful specialized care.

A review article summarizes the same direction: primary care improves health system performance by reducing total spending while improving population health and access (van Weel, 2018).

The limits matter, and the authors themselves point them out. Macinko and colleagues recall the ecological nature of the analysis and acknowledge that they did not directly address the possibility of endogenous variables, for lack of adequate instruments (Macinko, 2003). Starfield admits that the results may be overinterpreted: places with stronger primary care may also be those with stronger social policies in other areas, and the mere presence of physicians does not guarantee that all groups can use the service (Starfield, 2005). A broad review also describes much of the evidence as ecological studies, with geographic units as the unit of analysis (Shi, 2012).

These limits do not cancel the evidence, but they change how to use it. An ecological study shows that places with more of one thing have more of another; it does not prove that a change in an individual patient will produce the same effect. The text on study designs and their weaknesses helps separate association from causality. It is the consistency across different designs, the plausibility of the mechanisms and the stable direction of the findings that support the conclusion that investing in strong primary care is worthwhile.

Frequently asked questions

What are the four attributes of primary care?

First contact, longitudinality, comprehensiveness and coordination. Starfield describes them as first-contact access for each need, long-term person-focused care, complete care for most needs and coordination when care must be sought elsewhere (Starfield, 2005).

Is primary care the same as a health post?

No. The name of the service does not guarantee its functions. A facility can be the first place people go and still lack continuity or coordination. The assessment looks at the attributes, not the sign on the door.

What is the difference between longitudinality and continuity?

In practice the terms overlap. The Haggerty review describes three types of continuity, informational, management and relational; Starfield's longitudinality corresponds mainly to relational continuity, the relationship with a professional or team over time (Haggerty, 2003).

Does being good at one attribute guarantee being good at the others?

No. In a study of 31 countries, access, continuity, coordination and comprehensiveness did not correlate with one another (Kringos, 2013). Each attribute must be assessed separately.

Does strong primary care reduce mortality?

It is associated with lower mortality in studies from countries and regions, even after adjustments, but most of these studies are ecological and may be confounded by other social policies (Macinko, 2003). The evidence is consistent, not definitive.

Does this text guide individual management?

No. It presents a model for understanding and evaluating primary care services and does not replace the clinical assessment of each patient.

Final checklist

• Ask the patient whether they have a place and a reference professional for new problems.
• Check whether opening hours, transport, cost and language allow first contact in time.
• Use the accumulated history and the patient's baseline to interpret new complaints.
• Resolve within the service itself what belongs there, from prevention to palliative care, before referring.
• When referring, write down the clinical question and agree on how the answer will come back.
• Review results and reports coming from other services and adjust the plan.
• Assess each attribute separately; good access does not guarantee good coordination.
• Read the evidence on primary care bearing in mind that much of it is ecological.

References

1. Bitton A, Ratcliffe HL, Veillard JH, et al. Primary health care as a foundation for strengthening health systems in low- and middle-income countries. J Gen Intern Med. 2017. PMID 27943038. DOI 10.1007/s11606-016-3898-5.
2. Haggerty J, Burge F, Lévesque JF, et al. Operational definitions of attributes of primary health care: consensus among Canadian experts. Ann Fam Med. 2007. PMID 17664500. DOI 10.1370/afm.682.
3. Haggerty JL, Reid RJ, Freeman GK, et al. Continuity of care: a multidisciplinary review. BMJ. 2003. PMID 14630762. DOI 10.1136/bmj.327.7425.1219.
4. Kringos DS, Boerma WG, Hutchinson A, van der Zee J, Groenewegen PP. The breadth of primary care: a systematic literature review of its core dimensions. BMC Health Serv Res. 2010. PMID 20226084. DOI 10.1186/1472-6963-10-65.
5. Kringos D, Boerma W, Bourgueil Y, et al. The strength of primary care in Europe: an international comparative study. Br J Gen Pract. 2013. PMID 24267857. DOI 10.3399/bjgp13X674422.
6. Levesque JF, Harris MF, Russell G. Patient-centred access to health care: conceptualising access at the interface of health systems and populations. Int J Equity Health. 2013. PMID 23496984. DOI 10.1186/1475-9276-12-18.
7. Macinko J, Starfield B, Shi L. The contribution of primary care systems to health outcomes within Organization for Economic Cooperation and Development (OECD) countries, 1970-1998. Health Serv Res. 2003. PMID 12822915. DOI 10.1111/1475-6773.00149.
8. World Health Organization. Declaration of Alma-Ata. International Conference on Primary Health Care, Alma-Ata, 6-12 September 1978. Official WHO document.
9. World Health Organization and United Nations Children's Fund. Declaration of Astana. Global Conference on Primary Health Care. 2018. WHO/HIS/SDS/2018.61. Official WHO document.
10. Shi L. The impact of primary care: a focused review. Scientifica (Cairo). 2012. PMID 24278694. DOI 10.6064/2012/432892.
11. Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. Milbank Q. 2005. PMID 16202000. DOI 10.1111/j.1468-0009.2005.00409.x.
12. van Weel C, Kidd MR. Why strengthening primary health care is essential to achieving universal health coverage. CMAJ. 2018. PMID 29661815. DOI 10.1503/cmaj.170784.