Patients values regarding primary health care: a systematic review of qualitative and quantitative evidence

Studies were eligible for inclusion if they:

Studies were excluded if they:

We sought evidence by systematically searching for original research articles in PubMed and EMBASE (Ovid) (Supplementary file 1) and additional searches in Google Scholar. Articles from January 2009 to May 2020 were included. We identified values and preferences important to patients related to PHC services. Considering values are a complex concept, the inclusion criteria were extended by identifying articles that might implicitly research and explain patient values. The search terms were selected from key terms often used in literature for describing patients' values such as satisfaction, important aspects or factors, expectations, priorities, preferences, and experiences [12‐14, 16]. We used MeSH terms, keywords, and synonyms to search the articles. The main search terms were "Primary health care", "Patient", "value", and "access" (Supplementary file 1). Search terms were kept broad to capture as many relevant articles within our study objectives.

ABP and JM performed the title, abstract, and full-text screening independently. Through online meetings with all team members, we discussed and resolved disagreements. Due to the variability in methods and data collected in the included studies, meta-analysis was not performed. The data extraction process was guided by the Joanna Briggs Institute (JBI) mixed methods data extraction form [20]. Extracted data were analyzed thematically and synthesized narratively by ABP, RSP, and DW, with input from JM. Firstly, we performed inductive coding to allow concepts and themes to emerge from the data. Coding and concepts were discussed, refined, and finalized with all authors. Secondly, a deductive phase was conducted to categorize the findings. The concepts developed by Bastemeijer et al. [16] were used as an initial guide to classify our findings. This systematic review was presented according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [21].

The quantitative and qualitative studies' quality was assessed using the JBI Critical Appraisal Checklist for Analytical Cross-sectional Studies [17]. This checklist was developed collaboratively, approved by the JBI International Scientific committee, and used in previously published systematic reviews [18‐20]. In addition, qualitative studies were assessed using Consolidated Criteria for Reporting Qualitative Studies (COREQ) [22, 23] by ABP and JM. DW and RSP reviewed and gave input to the appraisal results.

After removing duplicates, a total of 1,819 articles were identified and screened for titles and abstracts, which resulted in 70 articles eligible for full-text screening (Fig. 1). Among them, six full texts were not available. The authors then performed full-text screening for the rest of the articles, resulting in 16 papers and two additional papers from reference searching that met the inclusion criteria. Articles were excluded because of the following reasons: twenty-two articles did not directly reveal patient values; eight papers were about specific aspects and values that had been pre-determined by investigators; six articles focused on other types of care such as emergency; five articles were about scoring and ratings; three studies did not meet the method inclusion criteria (such as using discrete choice experiment), and two studies were not clearly from the patients' point of view. Characteristics of the included studies are described in Tables 1 and 2.

The selected qualitative studies met between 9 to 10 from 10 JBI criteria, and 18 to 27 from the 32 COREQ criteria (Supplementary file 2). Half of the qualitative studies explicitly explained the interviewers' characteristics, but only one study presented a discussion concerning non-participation. One study had a coding tree description, and the other studies may have put the coding tree directly into themes or subheadings in the main text. Most of the quantitative studies met the critical appraisal criteria checklist. However, the criteria on appropriate exposure measurement, confounding identification, and mitigation strategies did not apply for some of the studies (Supplementary file 2). All authors (ABP, RSP, JM, DW) judged the included studies to be of sufficient quality.

Out of eighteen included studies, none came from low-income settings, thirteen originated from high-income countries, and the rest were from middle-income countries (country category according to The World bank) [24]. Nine studies were from European countries. Most respondents aged 18 and above. Vulnerable populations identified within the included studies were elderly, lesbian women, people with mental illness and substance abuse, and ethnic minority groups. No studies were about parent's or children's values.

To provide a clearer view at which level the values occur within the PHC system, we categorized the patient values into four groups, as illustrated in Figs. 2 and 3:

Values and related quotations from the included qualitative studies are shown in Table 3.

From our analysis, there are two values closely associated with the patient: autonomy and privacy.

We found that patient values were predominantly associated with the values of the GP. Patients hold high expectations towards their GPs even though the system and regulation also contribute to patients’ experience. Some values related to the GP were especially crucial, to such an extent that it would be intolerable if violated by the physician. Some studies found that patients would rather change providers if their personal values are breached. However, in practice the GPs might not be fully aware of this situation.

Although the essential values of primary care arising from the included studies were mainly similar, different social determinants such as age, gender, ethnicity, disease, and rurality might influence patients' values regarding PHC. Findings from the studies varied, particularly appearing from the analysis in some quantitative studies. According to two studies, the most important values do not differ between gender, although women assigned higher values [12, 32]. Specifically, for vulnerable groups, the value of non-discriminatory services should be seen as a top priority, because it could potentially influence their decision to visit a particular PHC [27, 29, 30, 33, 36, 37]. In a survey in Switzerland, participants with chronic diseases valued the aspect of being seen as a person and continuity of care higher than people without chronic illness with OR 2.21 and 1.92, respectively [12]. In contrast, the study in Australia found that having a chronic disease and rurality do not differ in all their modelling in essential aspects of GP [32]. However, age and GP visit frequency did. According to Sebo et al., elderly and patients at large practices are more stringent [39].

Values related to the general practitioner were paramount and apparent in most studies from high- and middle-income countries. This may reflect that even in a good PHC system, patients place high expectations on their doctor's interpersonal skills. Patients value primary care doctors who have not only medical-related competence but also virtuous characteristics and can address their patients' as a person, as in findings from previous studies [41‐43]. A study from the UK found the most critical component of general practitioner consultation was having their problems addressed seriously, which increases trust and confidence in GP. Findings from other countries also emphasized this value. The picture of an ideal physician may not always be realistic. However, it reflects the need for primary care providers to pay attention to tailored care to address patients' specific needs.

In healthcare and daily life, vulnerable groups experience stigma and discrimination daily outside healthcare, from stigma related to the socio-economy, and criminalization, to unsettled housing, which can become notable barriers to accessing PHC. Our study highlights that vulnerable subgroups may prioritize a particular value regarding PHC differently than general adult patients. For example, the elderly and indigenous accentuated their need for autonomy. In contrast, lesbians and patients with mental disorders prioritized inclusivity and understanding their psycho-social context. In Europe, patients don't feel discriminated against in PHC [27]. However, this is not the case in other parts of the world, including studies from Brazil and Canada. Hence, it is crucial to operationalize an inclusive care provision at PHC with patients' views.

Seven studies from different countries valued the continuity of care, which is unique for PHC. For patients needing multiple visits or those with chronic illness, continuity of care is regarded as relatively more critical, similar to findings from previous studies [44, 45]. Continuity of care impacts health outcomes and may lead to decreased mortality in primary and secondary care [45]. Through this continuity of care, patients experience the benefit in that the doctor remembers patients’ information, so they do not have to repeat the same story. This can save the uncomfortable process for the patient of explaining their medical condition and establish trust and a good relationship [12, 27, 32]. PHC may operationalize continuity of care differently, either at the healthcare facility or doctor level. In the findings, patients value the latter.

Furthermore, referral provision is also unique to PHC. Our findings highlight the importance of the PHC doctor's decision to provide a referral. However, some patients perceived reluctance and experienced difficulties related to the referral. Referral provision-related policies may restrict doctors from providing referrals [4, 8]. Due to limited understanding of this gatekeeping system, patients may demand that GPs give referrals to a higher level of care [31]. Some patients perceive GP's gatekeeping role as a barrier to reaching specialist care [35], creating a dilemma for the doctor.

We noted that although privacy protection increased patient satisfaction [34], it is rarely explicitly framed. Privacy protection is closely related to patients' trust and care-seeking should be possible without any privacy infringement [46‐49]. Attention to patient privacy should be reflected at least in two aspects: during the visit and on patients' information. Research on privacy in healthcare recently leaned towards discussing electronic patient records and online data protection [49‐53]. Despite being equally important, there was little discussion of the other dimensions of privacy, such as how the patient felt and experienced their privacy. Privacy during PHC visits might not be an issue in some countries, but this gap suggests there is scarce evidence about how and which aspects of privacy patients value in primary care.

Autonomy in primary care can be particularly challenging for vulnerable groups. Respecting a patient's autonomy can be, at a certain point, challenging when in conflict with the value of evidence-based medicine [54], for example, for doctors to find the balance between their medical knowledge and patients' wishes. Research on patients with chronic illness found that autonomy is recognized as a value underlying patients' demand for quality services, while paternalism and ‘knowing better’ can reflect a lack of recognition of patients' autonomy [55]. Autonomy is also closely related to the second theme; doctor-patient interaction relates to the values of shared-decision making and empowerment.

According to the IOM and Picker institute, values should guide clinical decisions [56, 57]. Patient values identified in our study can be used as a basis for PHC aiming at patient-centered care. This kind of patient-centered approach could be implemented, for example, into a guideline that incorporates patients' values [58, 59].

To our knowledge, this is the first systematic review on patient values regarding primary care. We used rigorous methods through the search steps and assessment of the articles. We used the most relevant studies, suggesting that the values reflected are related to the current situation. The studies included different populations, and countries, thus capturing various settings and circumstances.

A limitation of this study is that it only included studies published in English. Related to the included studies, elaboration and description of particular values were often unavailable. As a result, we could not compare possible different meanings concerning a specific value in different studies.