Primary care gatekeeping during the COVID-19 pandemic: a survey of 1234 Norwegian regular GPs

Background In the Nordic healthcare systems, GPs regulate access to secondary health services as gatekeepers. Limited knowledge exists about the gatekeeper role of GPs during public health crises seen from the perspective of GPs. Aim To document GPs’ gatekeeper role and organisational changes during the initial COVID-19 lockdown in Norway. Design & setting A cross-sectional online survey was addressed to all regular Norwegian GPs (n = 4858) during pandemic lockdown in spring 2020. Method Each GP documented how patients with potential COVID-19 disease were triaged and handled during a full regular workday. The survey also covered workload, organisational changes, and views on advice given by the authorities. Results A total of 1234 (25.4%) of Norway’s GPs participated. Together, they documented nearly 18 000 consultations, of which 65% were performed digitally (video, text, and telephone). Suspected COVID-19 symptoms were reported in 11% of the consultations. Nearly all these patients were managed in primary care, either in regular GP offices (55.7%) or GP-run municipal respiratory clinics (40.7%), while 3.7% (n = 73) were admitted to hospitals. The GPs proactively contacted an average of 0.8 at-risk patients per day. While 84% were satisfied with the information provided by the medical authorities, only 20% were able to reorganise their practice in accordance with national recommendations. Conclusion During the early stage of the COVID-19 pandemic in Norway, the vast majority of patients with COVID-19-suspected symptoms were handled in primary care. This is likely to have protected secondary health services from potentially detrimental exposure to contagion and breakdown of capacity limits.


Introduction
In Nordic healthcare systems, GPs are assigned an essential role as gatekeepers in regulating access to secondary health services.Research on the gatekeeper function has mostly examined the role of GPs under stable societal conditions, affirming a relationship between a strong primary healthcare system and reduced morbidity, mortality, and even increased life expectancy. 1,2The Norwegian regular GP scheme is briefly described in Box 1.
Earlier outbreaks of infectious diseases sparked interest in how the primary healthcare service is reorganised to optimise triage and avoid hospital overload during infectious health crises.It is recommended to deliver targeted preventive advice, inform patients about current public health guidelines, and reach out to vulnerable patients regarding the potential risk for contagion.[10][11] Areas with weak primary health care experienced hospital overload and increased mortality. 8,12In Italy, where early European cases of COVID-19 arose, a lack of trustful communication between the GP service and the authorities was reported. 13While the healthcare system was dealing with a vast number of patients with COVID-19, other groups of patients chose to stay home.In England, fewer patients consulted GPs with symptoms that could potentially indicate cancer, giving rise to concerns about delayed diagnosis. 14A recent Canadian study found that family physicians all over the country were not well incorporated into the COVID-19 pandemic response. 15n Norway, several societal preconditions influenced the course of the COVID-19 pandemic (Box 2).A first societal lockdown occurred between 12 March and 15 July 2020.Both the Norwegian authorities and the Norwegian Medical Association immediately established information channels to reach GPs.Many primary care practices stopped offering routine physical appointments.This was economically feasible, as the authorities introduced reimbursements for digital triage and consultations using The Norwegian healthcare system is based on the principles of universal access and continuity of care. 35Since 2001, all Norwegian citizens may sign up with a GP (and change, if desired), and 99% have chosen to do so, although 250 000 people are currently on a waiting list.The system is financed by taxation, together with income-related employee and employer contributions and out-of-pocket payments (co-payments).Private medical insurance is limited.While national healthcare policy is controlled centrally, responsibility for the provision of primary health care is decentralised.GPs act as coordinators of municipal services and gatekeepers to specialised care.On average, a GP has a list of approximately 1050 patients and often provides other medical services in the municipality 1 day a week.In addition, GPs generally take part in 24-hour emergency care services and many voluntarily participated in municipal-run respiratory clinics during pandemic lockdown. 36x 1 The Norwegian GP scheme • An established welfare state and publicly organised health care: ensured rapid adaptation of healthcare reimbursement systems and implementation of digital care, paid sick-leave for registered workers in all sectors, support for financially threatened businesses, and so on.• High public trust and social cohesion: as a nation, Norway is characterised by high trust in the authorities, social equalisation, and a tendency to follow expert advice.telephone, video, and text-based e-consultations (that is, freely formulated questions and answers).7][18] Most patients who attended physical consultations had either undergone digital triage or resided in geographical areas with a negligible risk of infection.A Norwegian study has documented that many GP offices lacked personal protective equipment to meet the COVID-19 pandemic so that the staff experienced fear of not being able to diagnose and treat patients safely. 19n addition, GP-run respiratory clinics, also known as 'fever clinics', were rapidly established in most municipalities to offer safe clinical evaluations while minimising the spread of infection in general healthcare facilities.The respiratory clinics were mostly run by GPs pertaining to the regular GP scheme. 18ventually, most Norwegian GP practices adapted to the COVID-19 pandemic in terms of infection control measures and organisational changes. 16During the whole period, consultations related to pandemic-related issues (COVID-suspected symptoms, concerns and infection control information regarding COVID-19) were free of charge for patients. 20he early lockdown period in 2020, when digital consultations were used for an unprecedented, broad range of clinical problems, reflects a unique transition period in the delivery of primary health care. 15Knowledge gathered during this period may have high future relevance for pandemics or other abrupt societal changes.To the best of our knowledge, our study is unique, as it builds on real-time data collected during lockdown.
The aim of the present study was to document the gatekeeper role of and organisational changes by GPs during the initial COVID-19 lockdown in Norway.

Method Setting, study design, and data collection
A cross-sectional online survey was addressed to all Norwegian GPs (n = 4858) between 14 April and 3 May 2020 during the first national COVID-19 pandemic lockdown.The questionnaire contained 170 items.First, each responder answered generic questions covering workload and change in practice organisation.Workload was measured by counting the number of consultations (physical, video, telephone, and text-based freely formulated questions and answers).Organisational changes were measured through questions on GPs' number of days in practice and conditions to reorganise working practices in accordance with the national recommendations.To investigate GPs' situations during lockdown, we included questions regarding perceived information and recommendations from the health authorities, GPs' access to personal protective equipment, and personal concerns regarding COVID-19-infection risk.
Then, the GPs were asked to document all clinical activities on one typical full workday at the GP office.To investigate the gatekeeper function, the number of patients with suspected COVID-19 symptoms was registered, together with information on whether the patient was handled locally by the GP, referred to a GP-run municipal respiratory clinic, or admitted to a hospital.Furthermore, the GPs reported whether they proactively contacted vulnerable patients owing to risk of a serious course of eventual COVID-19, the number of patients or next of kin with questions and concerns about COVID-19, and the number of patients who received tailored advice related to COVID-19 in case of symptoms, concerns, or other needs for infection control information.The GPs also registered whether they were concerned about possible delayed somatic diagnosis or treatment (that is, related to cancer) and about patients' lockdown-associated psychosocial difficulties.We consulted the Checklist for Reporting Results of Internet E-Surveys to develop the survey and report its results. 21The survey was pilot tested by a panel of experienced GPs.It took approximately 90 minutes to complete the full survey.Two articles about GPs' reported use of video consultations have been published from the same material. 17,22o reach responders, we collaborated with Norwegian Health Informatics (NHI), a web-based portal that hosts an online clinical decision support product (NEL), to which approximately 98% of all Norwegian GPs subscribe. 23An invitation was sent to all NEL subscribers including a unique email link to the survey, ensuring both the authenticity and anonymity of the responders.In addition, we used large social media groups of GPs to stimulate participation and invite GPs who did not receive a personal email.Several reminders were sent by email and social media.Corrected for this, the response rate is 26% of the GPs who de facto received the invitation.The survey was conducted through Netigate, an application for internet surveys.

Statistics
The data were analysed using SPSS (version 26.0).We excluded answers with extreme outliers that were most likely wrong (for example, 141 consultations on a working day; n = 3).Questions eliciting GPs' viewpoints were scored on a 5-point Likert scale and then combined into three categories.More participants (n = 1234) completed the generic questions, including workload estimation, than registration of the full working day (n = 910).Therefore, we calculated the total number of consultations by multiplying the average consultations per day (19.7 based on all 1234) with the number of responses to the triage questions (901-910).We performed a multivariable regression analysis to investigate the effect of GPs' gatekeeping in relation to experience, recourses to adaptation of practices according to the authorities' recommendations, and how information from the authorities was received.

Results
In total, 1234 of the 4858 invited GPs (25.4%) partici pated and answered the generic questions.Of those invited, 910 (18.7%) completed registration of the working day, and we received answers from between 904 and 913 GPs to the questions concerning gatekeeping.The 1234 GPs reported an average of 19.7 (95% confidence interval [CI] = 19.6 to 19.9) consultations per day.Our material of 901-910 responses therefore covers approximately 18 000 consultations (17 767-17 945).The characteristics of the responders are presented in Table 1.

Triage and treatment of potentially infected patients
On their documented working day, 11.1% (n = 1997) of the consultations were with patients with suspected COVID-19 symptoms.In Table 2, we show how these patients were handled.Nearly all these patients were managed in primary care, either in regular GP offices (55.7%) or GP-run municipal respiratory clinics (40.7%), while 3.7% (73/1997) were admitted to hospital, corresponding to 0.4% (73/17904) of the total number of consultations.
We performed a multivariable regression analysis to study whether admission rates to hospitals were affected by the GP's work experience (±5 years), financial and practical resources for adapting practice in accordance with the authorities' recommendations, and how information from the authorities was received (see Supplementary Information S1).We found no differences in referral rates, as these background variables explained only 0.3% of the variation in referral rates (R 2 = 0.003), and neither was statistically significant.

Medical help and preventive care for vulnerable patients
During lockdown, the GPs reported selected aspects regarding the composition of patients.
In Table 3

Organisational changes
During the pandemic lockdown, the GPs reported a 12% increase in time devoted to direct patient contact (in contrast to municipal and administrative tasks, and so on) to a total of 4.6 weekdays: 3.5 days (95% CI = 3.4 to 3.6) at the GP office and 1.1 days (95% CI = 1.0 to 1.2) working from home.
In Figure 1, we show that pandemic-related information provided by the Norwegian health authorities and the Norwegian Medical Association combined was regarded as satisfactory by 84% of the GPs.However, 56% disagreed that they had sufficient personal protective equipment, and only 20% were able to reorganise their practice in full accordance with the national recommendations at this early stage of the pandemic.
Figure 2, shows GPs' personal concerns regarding COVID-19 infection in the early phase of the pandemic.While 76% were worried that they could spread the virus to their patients and 75% were concerned they would spread it to their family, 55% were worried about becoming infected themselves.discussed, our participants appeared relatively representative of regular GPs in Norway. 17Female and doctors aged <30 years were slightly overrepresented, as were GPs from Trøndelag County, where the research group is based.Collaboration with NHI gave all subscribers an email with a unique link to our survey. 23This method made it slightly less accessible for the approximately 2% of GPs who do not subscribe to NHI.The use of Netigate prevented multiple answers from the same source, which is considered a strength.Alternative recruitment methods, such as through the Norwegian Medical Association, would likely encounter similar limitations.It can be assumed that their mailing lists do not encompass all Norwegian regular GPs.
National Norwegian data show that the total number of admissions per day with suspected COVID-19 peaked at 325 per day during early lockdown and then gradually decreased. 24Our findings of 73 admissions from one-fifth of Norway's GPs of patients with suspected COVID-19 during early lockdown fits well with this, indicating reasonable representativeness of the entire population.
A few additional admissions directly from the respiratory clinics in these early weeks of lockdown, are not measured in this study, but can be considered negligible in number.

Comparison with existing literature
Our study has added to the evolving literature about the role of primary care during the COVID-19 pandemic. 15,16,18,22,25By international comparison, Norway had very low and even subnormal mortality rates during the pandemic. 26This outcome has complex explanations, 26 as outlined in Box 2 and a Norwegian report on the pandemic from 2022. 27It is beyond the scope of this paper to deliver exact estimates of the contribution of primary care.It is, however, relevant to note how the regular Norwegian GP scheme, based on continuity of care (Box 1), provided favourable premises for triage of COVID-suspect cases and preventive care for vulnerable patients in general, as recommended during infection outbreaks. 3,5Such emphasis was rapidly facilitated by Norwegian authorities through changes in the tariff system. 20Our data indicate that one in five patients who consulted their GP received tailored advice.This equalled twice the number of patients with suspected COVID-19 symptoms, indicative of efforts among GPs to provide preventive medicine to their list population.In Norway, patient selection was carried out using the GP's electronic journals and various digital tools. 20hile there was a general drop in hospital admissions, data from Statistics Norway show that the number of consultations in primary care was maintained during the pandemic. 27,28This validates our findings that the average GP conducted around 20 consultations on a typical workday, similar to pre-pandemic levels. 17,28The high level of digital consultations among the responders was similar to national numbers during this phase of the pandemic. 28The increase in reported clinical working hours despite an unchanged number of consultations is most likely explained by time-consuming information updates from the authorities and implementation of infection control measures. 29ur findings related to GPs' worries about overlooking or undertreating potentially serious illnesses can be readily explained by restrictions during lockdown.Many of the digital consultations would, under normal circumstances, have been physical with possibilities for ordinary clinical examinations.The high focus on COVID-19 might have discouraged patients from seeking help for other types of problems. 14rust in the medical authorities is central for a well-functioning healthcare system. 27Associations between patient trust in governmental information and high compliance have previously been found. 30,31Our finding that Norwegian GPs generally trusted information from the medical authorities and their medical association can be seen to reflect a well-functioning system for crisis-related communication in the healthcare system.
Our findings of a lack of personal protective equipment and difficulties reorganising GP practices in the early phase of the pandemic have also been shown by others. 16,18Lack of equipment might explain why so many GPs were concerned that they might get infected and spread COVID-19 to their patients and families.With sufficient equipment in the early phase of the pandemic, GPs could have handled more patients in their own offices, reducing the need for GP-run respiratory clinics.
Previous studies have indicated that experienced GPs tend to admit fewer patients to hospitals than inexperienced colleagues. 32,33Our findings, however, indicate that for COVID-related admissions during the pandemic, this might not have been the case.One explanation could be that COVID-19 was a previously unknown disease and admissions were to a high extent guided by algorithmic recommendations with limited room for clinical deliberations. 34

Implications for research and practice
Our study has highlighted the value of strong primary health care during public health crises.It has documented how regular GPs in Norway, based on defined patient list responsibility, could work effectively as gatekeepers to secondary care during a pandemic lockdown, while also remaining attentive to other patients' important but less acute needs.Our study has further emphasised the need for useful information, trust in advice from the authorities, enough protective equipment, and resources to reorganise local GP practices.The findings have relevance for future health service planning with respect to public health crises.

Funding
After the survey was conducted, BLN and TMJ had salaried part-time positions at the Norwegian Centre for E-health Research in Tromsø, Norway.The Netigate licence was funded by the General Practice Research Unit at NTNU.

Ethical approval
Participation was voluntary and anonymous.We did not elicit sensitive information or demographic characteristics that could reveal the identity of the GPs or patients, such as patient age, sex, specific diagnoses, or other sensitive information.Distribution of the survey to GPs' email addresses was handled by an independent party (Norwegian Health Informatics).No linkage key was established, and the participants' IP numbers were not accessible.Further approvals were thereby not required, according to Norwegian health research legislation and verified by the Norwegian Centre for Research Data (NSD).

Provenance
Freely submitted; externally peer reviewed.
38Clear communication to the general public: from the start of the pandemic the Norwegian Ministry of Health established an authoritative and clear information strategy, providing advice on how to behave in the lockdown situation and tackle eventual symptoms.•Societallockdown:closure of schools, working from home if possible, regulations regarding physical distancing, social gatherings and travel, and so on.•Establishment of effective communication channels to the healthcare sector: both the Norwegian Ministry of Health and the Norwegian Medical Association rapidly developed systems for communicating updates and recommendations regarding handling of the COVID-19 pandemic.These systems were in place but not fully developed when data collection for the present study took place.•Demographiccharacteristics: in a European perspective, Norway is an affluent country with a relatively dispersed population of approximately.5.4 million inhabitants, with 83% living in urban regions.The age distribution and proportion of immigrants is comparable with many European countries.38Box 2 Contextual factors affecting the course of the COVID-19 pandemic in Norway when our study was conducted

Table 1
Characteristics of participating GPs (n = 1234) Total responders: Sex n = 1224, Years of experience as a GP, n = 1233, Citizens in the municipality of the practice n = 1233, Employment n = 1231, Number of patients on GP's list n = 1230.Research Norberg BL et al.BJGP Open 2024; DOI: 10.3399/BJGPO.2023.0095 , we present COVID-related issues and concerns during a full-working day.In 9.0% (1613/17866) of the consultations, the GPs reported concerns about potentially delayed diagnoses of serious non-COVID-related diseases such as cancer.Twice as many people received 'tailored advice' concerning COVID-19 than the number who presented with suspected COVID symptoms (4.5 versus 2.2 per day), and the GPs proactively contacted on average one (0.8) at-risk patient per day.Concerns for lockdown-associated psychosocial difficulties were reported at 2.3 (95% CI = 2.1 to 2.4) a day.

Table 2
GPs' handling of the 1997 registered patients with suspected COVID-19 symptoms Based on an average of 19.7 consultations per day.b Municipal respiratory clinics in primary care were run by GPs at suitable locations. a

Table 3
COVID-related issues and concerns during a typical, full-working day during the first societal lockdown in Norway a Based on an average of 19.7 consultations per day Research Norberg BL et al.BJGP Open 2024; DOI: 10.3399/BJGPO.2023.0095