Healthcare users’ evaluations of general practice: a survey among Danish men aged 45–70 years

Summary

In this national survey of general practice care in Danish men aged 40–75 years, we found participants generally evaluated health care as good; however, variation persisted in some of the elements of health care experienced by different groups of healthcare users.

Strengths and limitations

A major strength of this study is the large sample size, and that it is possible to control for many sociodemographic factors.

This study does, however, have many limitations. It was a secondary analysis of data from a larger web-based survey with hypothetical vignettes that focused on patient involvement in medical decision-making about testing for cancer of the prostate, rather than patients’ evaluations of general practice specifically. Furthermore, items were purpose-designed, modified from the PSQ-18, rather than directly taken from, for example, the EUROPEP. Questions such as ‘The medical treatment that I have received has been almost perfect’, like its original wording in the PSQ-18, may have led to bias towards positive feedback;²⁰ rephrasing it as an open-ended question — for example: ‘How would you rate the overall quality of medical treatment received?’ — might have decreased bias. Similarly, the statement ‘Doctors are good at explaining the cause of medical examinations’, like its counterpart in the PSQ-18 (and item 12 in the Danish Patients Evaluate General Practice instrument), may not have adequately elicited patients’ perceptions of the health care they received. For example, it would be better to explore whether doctors explain the diagnosis, and address the patient’s ideas, concerns, and expectations. These last aspects provide a better understanding of an effective patient–doctor consultation, whereas explaining the cause of a medical examination is only a small part of the consultation and does not completely reflect its effectiveness. In addition, the inversion of the ‘showing consideration’ item (‘Sometimes doctors ignore what I’m telling them’) may have introduced bias or made comparisons with non-inverted scores more difficult.

The study explicitly focused on general practice health care in Denmark, and we investigated only those evaluations of GPs that were made by men aged 40–75 years. It remains unknown to what extent results may be generalised to other groups of healthcare users. In Heje et al’s previous study, there was no general significance of gender with regard to the assessment of GP care.⁶

The 30% response rate, although not uncommon in research using a web-based survey design,²¹ was not optimal. In addition, it raises the question of whether survey participation might have been higher among men who could identify some issues around symptoms of prostate disease and, therefore, whether selection bias was introduced; this cannot be ruled out, even though our sample had been found to be representative of adult men when compared with Danish and international data.

The survey item relating to chronic illness was far from being an accurate representation of the disease burden in Denmark. According to estimations from the Danish National Institute of Public Health, approximately 20% of the Danish population, or 700000–800000 adult Danes, will experience mental health problems during the course of 1 year;²² as such, it is clear that mental ill health forms an important part of the disease burden and including perceptions of that population would have provided a more accurate representation of experiences of health care across a wider spectrum of the population.

It should be remembered that Likert ratings reflect evaluations of general practice health care, rather than actual healthcare quality. Previous studies have suggested that correlations between clinical quality and interpersonal aspects of care are low, and that patient experience and clinical outcomes likely represent distinct aspects of healthcare quality.^23,24 Hence, as suggested by Prang et al, patient experience is a valuable marker of patient-centered care and good customer service, rather than a proxy for clinical outcomes.²⁴

We used municipality-level information as a proxy for responders’ income, together with self-reported working status. The nested design of this part of the study, including some predictors on a municipality level (although most were on an individual level), could possibly have masked some associations, due to individuals who are uncommon for their place of living (for example, a high-income person living in a low-income municipality) not contributing to possible associations.

Finally, it should be noted that differences in the evaluation of GP care were relatively small and, in many instances, less than 0.2 points on the five-point Likert scale, giving rise to the question of their clinical interpretation and significance. However, given the desire to create higher quality general practice health care through research-based information,⁵ the present findings, regardless of effect sizes, can be considered as one source of data among others to inform the prioritisation of further research and improvement initiatives.

Comparison with existing literature

A major finding of the present study is that a large majority of healthcare users agreed or strongly agreed that their GP treatment had been “almost perfect”. Even if one could argue that there is still room for improvement, finding about 9 in 10 healthcare users who perceive GP healthcare as almost perfect would seem pleasing to many. Our findings align with results from a cross-sectional study dating from 2008, in which 50774 Danish adults completed the EUROPEP questionnaire and 82.1% responded that their GP’s ‘thoroughness’ was at least ‘good’.⁵ Our findings are also consistent with those of a 2018 survey on patient satisfaction with out-of-hours primary care, in which 82.5% of participants expressed satisfaction with care.²⁵ In a later national cluster randomised case–control study among 3609 Danish citizens aged 18 years or older, 66.7% of the responders reported receiving ‘exceptional’ or ‘good’ general practice care during the previous 12 months.²⁶ By way of comparison with international findings, in Aljohani et al’s survey from 2022, approximately 85% of participants expressed satisfaction with general practice.²⁷

We found that GPs were given higher scores by older healthcare users. International studies have previously suggested a relationship between healthcare user age and a more positive evaluation of healthcare.^28–30 Likewise, in Aljohani et al’s survey, being aged >65 years was significantly associated with satisfaction.²⁷ As suggested in those previous studies, the association between satisfaction and older age may be rooted in: a patient’s longstanding relationship with their GP; the patient having a more realistic view of health and health care, expectations, and doctors’ skills due to their own life experience; and, perhaps, older people having a more forgiving way of judging — being a ‘grateful generation’ — with regard to health care. In a nationwide Danish patient evaluation survey that included 28260 patient evaluations, a positive GP assessment was found to be strongly associated with increasing patient age and increasing frequency of attendance;⁶ furthermore, patients reporting a chronic condition were more likely to be positive. Our study adds nuances to those previous findings by suggesting that evaluations of general practice care may vary among patients with different chronic illnesses. Contrary to our findings, in that same study, no association was found between a positive GP assessment and patients’ education level.⁶

Finally, we found that healthcare users in areas with higher tax incomes assessed GP care more critically. These findings echo those of previous international research that found lower levels of satisfaction among patients who were affluent.^9,31 As previously suggested, there may be socioeconomic differences in health service expectations, as people with more socioeconomic resources may expect their doctor to communicate particularly well and provide clear information.^9,32

In a Norwegian survey from 2017, responders found, to a large extent, that GPs were showing interest in their situation, providing sufficient information, and including them as much as they would like in decisions.³³ In our study, more than three-quarters of healthcare users reported that their GPs’ skills in explaining medical procedures were good, while slightly less than two-thirds of responders thought their GPs were good at listening and showing consideration for them. Although the reasons for this finding are unknown, a tentative explanation could be that high workload and signs of burnout are common among GPs,³⁴ and stress and burnout may affect listening skills, as well as the competency to show consideration for patients, be attentive, and listen to others.^35,36

In contrast with previous findings from Denmark⁵, the evaluation of GPs’ skills in terms of showing consideration for their patients in the present study was least positive. Furthermore, in our study, we found that responders in urban areas tended to evaluate the explanation of medical procedures as better. By way of comparison, in Iqbal et al’s 2021-study from Scotland, individuals residing in remote and rural areas tended to have the highest satisfaction with their general practice in terms of patient-centered care.⁹

Implications for research and practice

To continuously achieve the best possible quality in general practice, knowledge is needed about where, in particular, to prioritise improvement efforts.⁵ The findings suggested persisting inequalities in aspects of GP care that may need to be addressed, as well as the importance of doctors’ skills in showing consideration for their patients, informing them, and providing patient-centered care. Studies are needed to both deepen our understanding of healthcare users’ views on GP care and validate the findings of the present analyses. As an example, our finding of an age gradient in healthcare users’ evaluation of GP care should be followed up to clarify to what extent this mirrors the evolvement of a more demanding population of healthcare users that necessitates change at GP level (for example, relating to further education or training) or at higher organisational levels that requires intensified information to accommodate the expectations of younger healthcare users.

Our findings that suggested that participants with cerebrovascular disease have a slightly less positive evaluation of GP care also merits further investigation to verify whether this reflects a common experience in this group of citizens and, if so, to better meet the needs of individuals with cerebrovascular disease in general practice. In the same way, studies are needed to clarify healthcare users’ evaluations of mental health care in general practice.

In future studies, more accurate financial status should also be collected; this could be done directly from responders or through linkage with tax record registries. Last, but not least, in many countries, women face unique challenges in accessing health care, and their satisfaction levels may differ significantly from those of men. As such, including women in future studies would help to identify any gender-specific issues and provide a more comprehensive understanding of healthcare perceptions across the entire Danish population, thereby ensuring that healthcare services accommodate the needs of all patients. Research findings may then inform the design of means to improve the delivery of general practice services. Improvement initiatives should also include measures of clinical outcomes and patient satisfaction to obtain a comprehensive evaluation of care.

The findings of this study suggested that, in spite of GP care being evaluated highly overall, evaluations vary among healthcare users with different sociodemographic backgrounds, personality styles, and health problems; this underscores the need for GPs to be attentive to patients' different needs, views, and concerns.