COVID-19 Continues to Burden General Practitioners: Impact on Workload, Provision of Care, and Intention to Leave

General practitioners (GPs), already in a profession with a high workload, have been at the frontline of providing COVID-19-related healthcare in addition to routine care. Our study examined the impact of pandemic-related consultations and changes in practice organization on GPs’ current workload and provision of healthcare in summer 2021 (May 2021–July 2021) and early 2022 (January 2022–February 2022). In total, 143 German GPs participated in an online survey in the summer of 2021. Of these, 51 GPs participated in the follow-up survey in 2022. Most GPs perceived an increase in consultation frequency, consultation times, and workload since the pandemic outbreak. Increased consultation times were related to the reduced provision of medical care to other patients with chronic diseases. More SARS-CoV-2 vaccination consultations were associated with reduced home visits, acute consultation times, and cancer screenings. A quarter of GPs considered leaving their job. Pandemic-related bureaucracy, restricted access to therapy and rehabilitation services specialized on COVID-19, unreliable vaccine deliveries, mandatory telematics-infrastructure implementation, and frequent changes in official regulations were the main reasons reported for dissatisfaction. Our results provide insights into how the pandemic continues to burden GPs’ work routines and how better working conditions in times of high demand could be achieved in future pandemics.


Introduction
The coronavirus-19 disease (COVID-19) pandemic developed into a persistent burden for inpatient and outpatient healthcare and limited the capacities of the healthcare system to pursue day-to-day services [1][2][3]. In the outpatient sector, general practitioners (GPs) have been at the frontline of the COVID-19 pandemic. Considering that general practice is already a profession with a high workload [4,5], COVID-19 placed a number of additional burdens on GPs [6]. The management and treatment of patients with acute COVID-19 and/or with persistent symptoms after infection (defined as "long COVID" [7]) is in most cases provided by GPs and might develop-due to high prevalence [8] and treatment uncertainty [9]-into a long-term burden for GPs [10][11][12]. Further, GPs in Europe widely joined the national SARS-CoV-2 vaccination campaign [13], placing greater demands on them. Besides complying with changing hygiene measures and government regulations [14,15], the situation is further exacerbated by recurrent staff shortages due to sickness or quarantine [14] and increased psychosocial stress on the practice teams [16,17].
At the beginning of the pandemic, GPs largely restructured their work routines [18][19][20]. In Germany, planned procedures, preventive consultations, or continued care for chronically ill patients had to be canceled or postponed [21][22][23]. This trend has also been observed in other countries [24]. Some patients with chronic diseases might also have avoided (primary) healthcare facilities in the first months of the COVID-19 outbreak, to not be a burden [25] or out of fear of being infected with SARS-CoV-2 [26]. In this early phase, GPs in Germany and in other countries were confronted with a reduction of consultations [23,27] and,

Recruitment Procedure
The data from the online surveys were collected in the Free State of Saxony, Germany, between May 2021 and July 2021 (defined as baseline 2021 in the following text). A followup survey was sent between January 2022 and February 2022 to all GPs who participated in the first assessment (defined as follow-up 2022 in the following text). The recruitment process of the first assessment has been described in more detail by Schrimpf et al. [9].
GPs were selected by the availability of an email address, on file with the Association of Statutory Health Insurance Physicians Saxony (Kassenärztliche Vereinigung Sachsen), and they were invited by email to voluntarily participate in this online survey. In May 2021, the first invitation email was sent to 1444 GPs. The email contained information about the purpose of the survey, data handling, and a link to the online survey. The access to the survey was generated by a TAN (transaction authentication number) to ensure that each GP participated only once. In June and July 2021, reminders containing the same information were sent to non-responders (following the recommendations of Edwards et al., [47]). In January 2022, an invitation to a follow-up survey was sent to all GPs who participated in the first survey, identified by their email addresses. Two reminders at the end of January and in mid-February 2022 were sent to non-responders (see recruitment process in Figure 1). et al. [9].
GPs were selected by the availability of an email address, on file with the Association of Statutory Health Insurance Physicians Saxony (Kassenärztliche Vereinigung Sachsen), and they were invited by email to voluntarily participate in this online survey. In May 2021, the first invitation email was sent to 1444 GPs. The email contained information about the purpose of the survey, data handling, and a link to the online survey. The access to the survey was generated by a TAN (transaction authentication number) to ensure that each GP participated only once. In June and July 2021, reminders containing the same information were sent to non-responders (following the recommendations of Edwards et al., [47]). In January 2022, an invitation to a follow-up survey was sent to all GPs who participated in the first survey, identified by their email addresses. Two reminders at the end of January and in mid-February 2022 were sent to non-responders (see recruitment process in Figure 1).

Survey
The questionnaires were self-developed in the Department of General Practice of Leipzig University by an interdisciplinary research team (medical scientists, biochemist,

Survey
The questionnaires were self-developed in the Department of General Practice of Leipzig University by an interdisciplinary research team (medical scientists, biochemist, and GPs) in a multi-stage revision process. The development process was complemented by an extensive literature search aimed at identifying relevant factors for barriers in ambulatory care related to patients with COVID-19 or long COVID. The final versions of both the initial and the follow-up survey can be found in Supplementary Tables S1 and S2. For the webbased evaluation, the software LimeSurvey (http://www.limesurvey.org/) was used, hosted on a secure server of the Leipzig University Computer Center. The completion of the first online survey took approximately 15 min and the follow-up survey took approximately 10 min.
Participating GPs were asked in both surveys to click the "I agree" button on the online informed consent form. Then, the survey started and comprised of the following topics: (1) demographics (e.g., age, sex, practice information), (2) patients with long COVID in the practice (e.g., frequency of consultations, difficulties during treatment, need for information or support), and (3) impact of the COVID-19 pandemic on practice and patient care. The response formats were multiple choice answers, rating scales (1-10), and free text entries.
Prior to implementation, the questionnaires underwent a think-aloud pre-testing [48] aimed at identifying problems or misunderstandings related to each item. The provisional questionnaires were filled out by five GPs, who were instructed to think aloud while answering each item and report every spontaneous thought. After completing the questionnaires, the GPs were briefly interviewed about general issues with the questionnaires (e.g., length, structure, and general comprehensibility). After pre-testing, the provisional questionnaires were adjusted and further developed.

Coding of Free Text Entries
Participating GPs were asked to indicate in free text fields if needed, additional problems, and needs related to the treatment of COVID patients as well as reasons for wanting to quit their jobs. Additional general comments and wishes could be entered at the end of the survey. Free text entries were independently coded by two authors of this study (AS, AB). Categories were derived inductively during the coding, either indicating a major category or a subcategory. The assignments were compared and differences in coding were discussed until an inter-coder agreement was reached for each discrepancy.

Statistical Analyses
All statistical analyses were carried out using IBM SPSS Statistics 27 (Armonk, NY, USA) with a two-sided α-level of 0.05. For descriptive statistics, missing values in single variables were considered by presenting frequencies as % (n/n valid ). Continuous variables were presented as mean (M) ± standard deviation (SD).
Univariate analyses of variance (ANOVAs) were used for analyzing differences in continuous variables between groups. Repeated measures ANOVAs were applied with the within-subject factor "time" (baseline 2021, follow-up 2022) for the metric variables "numbers of patients with long COVID", "Being able to address medical and/or psychological needs of patients with COVID-19", and "Perceived limited provision of satisfactory medical care to other patients with chronic diseases" to measure differences between the two time points. Estimated effect sizes were reported using partial eta squared (η p 2 ). For all univariate and repeated measures ANOVAs indicating a significant main effect, least significant differences tests were utilized to determine the origin and direction of the effect, in which case we report M ± SD.
Further, a two-sided bivariate correlation was calculated to analyze the association of two continuous variables (number of patients with long COVID treated in the practice and the satisfactory provision of medical care to other patients with chronic diseases).

Sample Characteristics
Of the 186 GPs who participated in this study (13% total response rate), 45 GPs left the survey incomplete after the first page. In total, 143 GPs were included in the analyses at baseline. Of these participants, 51 completed the follow-up survey in January or February 2022. We further differentiated practices stated to have or not to have a special focus Healthcare 2023, 11, 320 5 of 21 on COVID-19, indicating that especially practices with more health insurance approved physicians specialized in COVID-19 management (Table 1). Practices with a focus on COVID-19 treated on average more patients with long COVID (symptoms lasting between 4 to 12 weeks: M = 20.1, SD = 14.3; symptoms lasting more than 12 weeks: M = 9.8, SD = 8.1) than standard practices (symptoms lasting between 4 to 12 weeks: M = 11.2, SD = 10.9, F(1, 129) = 5.451, p = 0.021, η p 2 = 0.041; symptoms lasting more than 12 weeks: M = 4.5, SD = 5.9, F(1, 128) = 6.345, p = 0.013, η p 2 = 0.047). Additional percentages, means, and standard deviations for GP sample characteristics and practice information can be found in Table 1. A direct comparison of GP sample characteristics between baseline 2021 and follow-up 2022 can be found in Supplementary Table S3. Data are presented as mean, standard deviations, and percentage (n/n valid ). *, multiple responses possible.

Workload since the Outbreak of the Pandemic
Information on changes in workload, economic situation of the practices, frequencies of consultations, and time requirements for consultations for both time points can be found in Table 2. In general, most GPs perceived an increase in the frequency of patient visits, consultation times, and workload since the outbreak of the pandemic. Between baseline 2021 and follow-up 2022, the frequency of patient visits continued to increase. The economic situation of the practices only worsened in a minority of practices.

Consultations Related to COVID-19 in GP Practices
Comparison between the two surveys: The number of patients with long COVID remained stable from baseline 2021 (M = 11.9 patients with long COVID symptoms lasting between 4 to 12 weeks, M = 5.9 patients with long COVID symptoms lasting more than 12 weeks) to follow-up 2022 (M = 10 patients with long COVID symptoms lasting between 4 to 12 weeks, M = 6.8 patients with long COVID symptoms lasting more than 12 weeks) in GP practices (Table 3). Repeated measures ANOVAs revealed no significant differences between the two time points (patients with long COVID symptoms lasting between 4 to 12 weeks: F(1, 48) = 0.081, p = 0.777, η p 2 = 0.002; patients with long COVID symptoms lasting more than 12 weeks: F(1, 49) = 1.687, p = 0.200, η p 2 = 0.033).  Data are presented as mean, standard deviations, and percentage (n/n valid ). n.a., not assessed.

Baseline 2021:
The majority of GPs (51.9%) reported treating patients with acute COVID-19 every week in their practice. Similarly, most GPs (44.3%) stated they treated patients with long COVID symptoms lasting between 4 to 12 weeks at least once a week in their practice. In contrast, patients with long COVID symptoms lasting more than 12 weeks were largely treated at least once a month (44.6%) in GP practices (Table 3).
Follow-up 2022: GPs were asked to estimate how many out of 100 patients in their practice are currently visiting for pre-defined counseling needs. Importantly, GPs reported that, in total, 29.4 out of 100 patients are currently visiting the practice because of COVID-19/long COVID or SARS-CoV-2 vaccination issues. In addition, GPs were asked to indicate which medical services can currently only be offered at a reduced capacity. Self-payer services, preventive health check-ups, and preventive cancer screenings have been offered less by 66%, 58%, and 44% of respondents, respectively. Further details for the variables assessed can be found in Table 3.

Provision of Care for Patients with Long COVID: Identified Problems and Need for Support
Comparison between the two surveys: GPs were asked at both time points whether they were able to address the medical and/or psychological needs of patients with acute COVID-19 or long COVID during consultations on a scale from 1 = "not able" to 10 = "fully able". The ratings did not change between baseline 2021 and follow-up 2022 (F(1, 48) = 0.096, p = 0.758, η p 2 = 0.002; Table 4). Fewer appointments can be currently offered in the following services *: n.a.
Acute consultations 20% Care of chronically ill patients 40% Preventive cancer screenings 44% Preventive health check-ups 58% Self-payer services 66% Home visits 24% Nursing home visits 22% Being able to address medical and/or psychological needs of COVID patients during consultations 1 = "not able", 10 = "fully able" 6.7 ± 2 6.7 ± 2 Perceived limited provision of satisfactory medical care to other patients with chronic diseases 1 = "fully limited", 10 = "not limited" Data are presented as mean, standard deviations, and percentage (n/n valid ). *, multiple responses possible, n.a., not assessed.

Baseline 2021:
GPs were asked at baseline to identify current problems related to the treatment of patients with long COVID and the GPs' need for information. The long course of the disease (78.9%) was rated as the main problem during the treatment of patients with long COVID. The introduction of national guidelines on long COVID (63.9%) was identified as the main need for support of GPs. All results can be found in Table 4. GPs were additionally able to make individual comments on both questions. The lack of and need for specialists and facilities to treat patients with long COVID were mentioned by most respondents. The results of these free-text answers can be found in Supplementary  Table S4.

Provision of Care for Other Patients since the Outbreak of the Pandemic
Comparison between the two surveys: GPs were asked whether they felt that the pandemic limited the provision of satisfactory medical care to other patients with chronic diseases on a scale from 1 = "fully limited" to 10 = "not limited". The ratings changed between baseline 2021 and follow-up 2022 (F(1, 48) = 11.287, p = 0.002, ηp 2 = 0.190; Table 4 Figure 3A).

Follow-up 2022:
GPs were asked to indicate which healthcare services were currently less likely to be offered. The results showed that especially preventive cancer screenings, preventive health check-ups, and self-payer services were currently reduced ( Table 4). In addition, GPs who reported currently offering fewer acute consultation times than usual administered more SARS-CoV-  Figure 3B). Further, the percentage of patients currently visiting for SARS-CoV-2 vaccinations was related to GPs' ability to pro-vide preventive cancer screenings, showing that GPs who currently reduced their cancer screening services had a higher proportion of SARS-CoV-2 vaccination consultations in their practice (M = 18.8, SD = 11.5) compared to GPs with no changes in cancer screening services (M = 12.4, SD = 7.9; F(1, 46) = 5.178, p = 0.028, η p 2 = 0.101). No relationship between SARS-CoV-2 vaccination consultations and other services has been found.

Intention to Leave
At follow-up, GPs were able to indicate if they considered quitting their job in the last 12 months, which 26.5% affirmed ( Table 2). Age did not differ between GPs who affirmed (M = 47.7, SD = 9.2) and those who did not affirm having considered quitting (M = 49.5, SD = 9.2). We further found that GPs who considered quitting their job also currently treated more patients with long COVID (patients with symptoms lasting between  Figure 2B). GPs were able to indicate in free text fields reasons for considering leaving their job. We identified the following main reasons: increase in workload and administrative tasks, demanding patients, as well as the handling of the pandemic by politicians, health authorities, and media. All results of these free text answers can be found in Supplementary  Table S5.

Additional Comments and Wishes
Baseline 2021: At the end of the baseline survey, participants were asked to provide additional comments regarding long COVID, and 33 GPs filled in the free text field. A summary of statements can be found in Supplementary Table S6. The majority of statements were related to the current treatment of long COVID and GPs' individual observations regarding the long course of the disease. Many GPs wished for better therapy and rehabilitation options for their patients-also with respect to psychotherapy-and described the current possibilities as insufficient: "It is difficult to get a rehabilitation place (which is also time-consuming and help is needed). The same applies to initiating psychological co-treatment. The health insurance companies do not support me as a doctor and my patients (e.g., I went through depressing written disputes about quarantine/AU [certificate of incapacity for work])." (female GP, 49 years old) "Direct and timely access to rehabilitation and specialist care must be organized! The best way is via a central coordination office. It is essential to set up a quota for psychotherapy for these patients!" (female GP, 42 years old) In addition, GPs reported the observation of strong psychological comorbidity in their patients with long COVID. Whereas some GPs see an increase in psychosomatic symptoms after infection with SARS-CoV-2, others attribute these symptoms to a preexisting psychological condition: "In my patients I see predominantly psychological impairments, especially an increase in anxiety/neurotic symptoms accompanied by physical and cognitive stress insufficiency. It is difficult to differentiate whether the physical limitations are a consequence of the psychological impairments." (female GP, 45 years old) "I am concerned that this disease is drifting more into the psychosomatic domain. Apart from a long feeling of illness, there is no tangible value and no recovery criterion except for the patient's subjective statements." (male GP, 59 years old) "More than genuine "post/long COVID symptoms", we observe an aggravation of already psychologically pre-altered patients in connection with COVID-19 without objectifiable pathological organic findings." (male GP, 39 years old) Only a few GPs mentioned perceived issues with media coverage, research, and politics and their influence on the practice and patients: "The extensive "nocebo education" provided by the media and the constant change of information are counterproductive for physical and psychological convalescence." (male GP, 39 years old) "The state has failed and these polls are far too late. Last year's discussions [2020] were a disgrace to the academy. Germany is stuck in the Middle Ages when it comes to communication between the university and the front." (male GP, 35 years old) Follow-up 2022: At the end of the follow-up survey, participants were asked about their wishes for improvements, and 38 GPs filled in the free text field. A summary of statements can be found in Table 5. The majority of statements were related to politics and regulations, indicating that GPs wished for a substantial reduction in bureaucracy and administrative work: "At the moment, I am only 50% GP and 50% practices organizer. Bureaucracy is not diminishing, since Corona, it massively increased (through constant change of billing codes, diagnosis codes and combinations, official orders)." (male GP, 54 years old) Further, many GPs wished for more reliable policy announcements and a reduction of political short-notice decisions during the pandemic. Along the same line, many GPs wished for reliable vaccine dose orders to organize their practice and for SARS-CoV-2 vaccine offers outside the practice: "Planning security. A reasonable, comprehensible, and not constantly changing strategy in pandemic control and vaccine supply. Relief through sufficient vaccination services outside the practice. We can well and safely secure the infection event and the outpatient care of patients suffering from COVID-19 if we are not responsible for the quarantine regulations and we also receive the ordered vaccine. Compulsory vaccination of staff in our facilities will lead to staff shortages and I worry that, then at the latest, we will only be able to provide minimal patient care and the quality of care can no longer be guaranteed due to overwork of the remaining staff." (female GP, 36 years old) It was further mentioned that the mandatory implementation of telematics infrastructure (enabling an electronic patient file, electronic prescriptions, and electronic certificate of incapacity for work) during an already pandemic-related high workload additionally burdened GP practices: "The IT innovations are justified, plausible, and at some point perhaps also facilitating/helpful. Currently, however, these things represent an additional burden! It would therefore be helpful to postpone them or to implement simplified processes!" (male GP, 43 years old) "It would help to be released from the burdensome and largely pointless expansion of digitization applications. I am not an opponent of digitization, but the currently planned measures such as e-prescription, e-AU [certificate of incapacity for work], and e-PA [patient file] predominantly cost time, money, non-existent mental reserves without visible practicability for the general public." (female GP, 46 years old) n * = statements in this category, ** % = percentage of GPs who stated an insight from this category.

Discussion
The present study investigated the impact of the COVID-19 pandemic on GPs' workload, quality of patient care provision, intention to leave, and working conditions in 143 German general practices in 2021 and followed up with 51 of those in 2022. Most GPs perceived an increase in the frequency of patient visits, consultation times, and workload since the outbreak of the pandemic. At baseline 2021, increases in consultation times were related to perceived limitations in the satisfactory provision of medical care to other patients with chronic diseases. At follow-up in 2022, an increase in the number of SARS-CoV-2 vaccination consultations conducted was associated with reduced care services in the practices, such as home visits, acute consultation times, or cancer screenings. Better access to therapy and rehabilitation for patients with long COVID, especially psychotherapy, was identified as the main need of GPs. We further found that a quarter of GPs considered leaving their job, which was related to the current number of patients with long COVID in the respective practices. Increased administrative tasks, unreliable vaccine dose deliveries, simultaneous introduction of telematics-infrastructure implementation, as well as the handling of the pandemic by politicians, health authorities, and media were identified as reasons for dissatisfaction. Our results provide insights into how the pandemic continued to burden GPs and their work routines between 2021 and 2022.

Pandemic's Influence on Workload
Most GPs in our study perceived an increase in workload, consultation times, and frequency of consultations since the outbreak of the pandemic, the latter further amplified between 2021 and 2022. Our results are in line with both qualitative research [12] and survey results [49] from Germany and other countries [45,50]. In addition, a recent study using medical record data reported that the average number of consultations in GP practices in the summer of 2021 increased by 18% as compared to a comparable period in 2019 [51]. Longer consultation times might have resulted from increased patient requests [37] and from the double burden of providing regular care and COVID-19-related care [12]. Especially communicating COVID-19-related information, e.g., about vaccines, potential symptoms, or testing, has been found to be common in GP practices [27,38,40].

Consultations Related to COVID-19 and Other Services
Compared to the baseline survey [9], we found that the number of patients with long COVID in GP practices remained stable from 2021 to 2022. At follow-up in 2022, we further found that almost one-third of all consultation issues were related to COVID-19, including patients with acute/long COVID-19 and SARS-CoV-2 vaccinations, indicating that these healthcare services came at the expense of other essential services. Our results showed that especially care of chronically ill patients, preventive cancer screenings and health checkups, as well as self-payer services, were reduced by many GPs. In line with our findings, previous research reported a substantial decrease in new cancer diagnoses, especially in GP practices, since the outbreak of the pandemic [22,52] as well as a disruption of chronic disease management [53]. The pandemic's negative impact on primary care for non-COVID patients due to shifted resources has also been discussed in other studies [12,54]. Although medical record data showed a general increase in GP consultations since the pandemic, diagnoses of new diseases dropped by 6% between these two periods [51]. An additional physician questionnaire revealed substantial reductions in home visits and opening hours as well as suspended check-ups and delayed consultations for high-risk patients by physicians in Germany during the pandemic, further indicating a shift in healthcare services [51].

Pandemic's Influence on the Provision of Patient Care
Shifted resources to pandemic-related care might come at the expense of chronically ill patients. Our results support this assumption: at baseline in 2021, GPs rated the satisfactory provision of medical care to patients with chronic diseases as more limited, the more patients with long COVID were currently treated in their practices and the more they perceived an increase in time for patient consultations since the outbreak of the pandemic. Importantly, the perceived limited provision of satisfactory medical care to patients with chronic diseases further amplified between 2021 and 2022, indicating an ongoing burden of the pandemic on patient care. Although patient consultations dropped during the first wave of the COVID-19 pandemic [23,27], during the following waves primary care was overwhelmed by the double burden of managing pandemic-related care and routine care with the same pre-pandemic resources [12,54,55], potentially leaving patients with chronic conditions underserved [22,43,44].
In addition, in 2021, GPs whose practices had a worsening economic situation since the outbreak of the pandemic rated their ability to address COVID patients' needs as lower compared to GPs with an unchanged or improved economic situation. This finding indicates that adjustments to the billing system in the context of changes in patients' needs and consultation reasons since the pandemic might be necessary and at the same time beneficial for patient care.
At follow-up in 2022, especially the administration of SARS-CoV-2 vaccinations was associated with reduced offers of medical services, such as acute consultation times, home visits, or preventive cancer screenings. Since April 2021, GP practices were allowed to join the national vaccination campaign in Germany [13,56]. The proportion of SARS-CoV-2 vaccines administered in GP practices in January and February 2022 was significantly higher than at the beginning of the vaccine campaign in April 2021 [57], potentially influenced by both the COVID-19 winter surge and the dismantling of high-capacity but expensive mass vaccination centers since September 2021 [46]. Our results suggest that the reduction of mass vaccination sites, such as vaccine centers, increased the burden on GP practices and came at the expense of other essential primary care services. We argue that external vaccination offers, especially in the upcoming COVID-19 waves, might be beneficial to relieve the burden on GPs and ensure the delivery of routine primary care services.

Intention to Leave
We asked GPs at follow-up in 2022 whether they were considering leaving their job in the last 12 months, which one-quarter of the respondents affirmed. Our results resonate with a previous study from the US, showing that 24% of participating physicians indicated a moderate to high likelihood to leave their current practice within the next two years [58]. Especially increases in workload and COVID-19-related stress have been found to be associated with the intention to leave [58], which was also next to patients' attitudes as stated in free text fields by GPs in our study. In addition, the age of the respondents was not a determinant for considering leaving practice, which was also found in a previous study with German GPs prior to the pandemic [59]. We further found that a greater number of patients with long COVID treated in GP practices was related to GPs considering quitting and might contribute to an increased burden. In line, previous studies found that frequent contact with patients with COVID-19 was related to higher scores in burnout and lower job satisfaction in health professionals [34][35][36]. In addition, qualitative data showed that the double burden of maintaining regular healthcare and COVID-19-related healthcare was perceived to be exhausting [12,37,42]. Generally, studies showed that the pandemic placed high psychological burdens on healthcare workers [17,60], which potentially accelerated during the course of the pandemic [49]. Our results are of concern, as a previous study showed that GPs' intention to leave patient care was a predictor of actually leaving their job [61]. The additional pandemic-related workload might therefore be a catalyst for GPs' intention to leave.

Perceived Burdens and Need for Support
In free text fields, GPs expressed at baseline in 2021 the urgent need for specialists, outpatient and rehabilitation clinics, and psychotherapies for patients with acute COVID-19 and long COVID. The expansion of care services related to COVID-19 as well as structured concepts was also requested by German patients [62]. Our results emphasize the benefits of the previously described interdisciplinary, multi-sectoral, and interprofessional approach to the management of patients with acute COVID-19 and long COVID [62][63][64], it being able to meet the varying needs of affected patients.
At follow-up in 2022, especially bureaucracy and administrative work, which increased during the pandemic, were perceived as burdens and were also reported elsewhere [27,65]. Inflated bureaucracy and over-regulation have been discussed as shortcomings of the governments' pandemic regulation attempts [66] and should be further evaluated and revisited to reduce the workload of healthcare providers. Reliable policy announcements were additionally mentioned by GPs in our study as being needed for better planning security. In line, constant changes in official information or announcements have been found to also generate stress, confusion, and workload in healthcare providers of other countries [27,37,39,41,67]. Reduction of information in terms of frequency and quantity might therefore be beneficial in future pandemics to increase GPs' and patients' adherence to these announcements and regulations. Some GPs also wished for constant vaccination opportunities outside the practice and for reliable vaccine dose deliveries. In Germany, SARS-CoV-2 vaccinations have been found to be associated with high efforts and administrative work and to be insufficiently remunerated in GP practices [68]. These time expenditures increased the GPs' workload and may have resulted in reduced capacities for other healthcare services. In addition, GPs could order vaccines only in limited quantities and vaccine doses were delivered depending on availability, further increasing GPs' planning insecurity and workload. Lastly, GPs perceived the implementation of telematics infrastructure at the time of the pandemic as an additional burden. Coincidentally, the obligation to establish certain telematics infrastructure functions in German medical practices was introduced during the COVID-19 pandemic [69], which came with initial technical difficulties, such as malfunction or compatibility issues. As has been shown elsewhere [70,71], GPs in our study stated a general affinity for digitalization and acknowledged the benefits. However, due to initial problems, implementation was time-consuming, in parallel with an already increased workload caused by the pandemic. In particular, the perception of time savings through digitalization was mentioned by physicians in other studies as a facilitator for adopting digitalization [70,71]. In sum, our results indicate that increases in workload for GPs during the pandemic had multiple drivers, including bureaucracy related to pandemic regulation, frequent changes in official information and legislation, being responsible for the main coverage of the national SARS-CoV-2 vaccination strategy, and the simultaneous requirement to implement new, not fully developed, telematic infrastructures.

Limitations
Our study has limitations. First, given the nature of the study and the sample size, a selection bias might have occurred. All answers were self-reports and might be imprecise due to subjective perceptions. Some of the reported results might be influenced by other factors than pandemic-related work changes. In addition, we did not conduct a power analysis, wherefore the data are to be considered exploratory and cannot be generalized. Second, our questionnaire is not a valid scale as we did not develop and assess several items measuring a construct related to workload, provision of patient care, or job satisfaction, but rather investigated single-item responses. Single-item responses were chosen over scales to reduce the length and monotony of the questionnaire and, hence, increase willingness to participate in a population with time constraints. However, studies showed that single-item responses might be as reliable as multiple-item scales, especially for less complex constructs (e.g., [72][73][74]). Third, due to a cross-sectional study design, we do not have data on GPs' status before the pandemic and can only depict subjective perceptions of changes since the pandemic. Lastly, the study was conducted in one federal state in Germany. Differences (e.g., in pandemic regulations, vaccine supply, or case incidences of COVID-19) between federal states in Germany as well as between European countries limit the generalizability or comparability of our findings.

Implications
Our research contributes to a better understanding of the ongoing impact of the COVID-19 pandemic on the provision of primary healthcare and GPs' satisfaction with their working conditions. As GPs' workload was already high in pre-pandemic periods [4], the current conditions have been described as unsustainable [6]. Our results might therefore have some general implications. Considering the average age of the GPs, demographic changes, and the expected decrease in treatment capacities in the future, the COVID-19 pandemic might act like a magnifier of the upcoming problems and distribution battles, such as prevention and screening vs. acute treatment as well as the lack of referral of patients to specialists or clinics. The GP as the gatekeeper would become the universal treatment provider. Further, the profession of general practice is already suffering from a lack of attractiveness for young doctors [75,76] and is perceived to have a disproportionately high amount of administrative tasks and comparatively low income [77]. The extraordinarily high workload and increased administrative tasks experienced by GPs during the pandemic might further deter young graduates to pursue a career in general practice.
To reduce further dissatisfaction, burnout, or even withdrawals from practices, our results highlight the following potential areas for improvement (Box 1).

Box 1. Potential improvements of GPs' working condition.
Healthcare 2023, 11, x FOR PEER REVIEW 17 of 22

Implications
Our research contributes to a better understanding of the ongoing impact of the COVID-19 pandemic on the provision of primary healthcare and GPs' satisfaction with their working conditions. As GPs' workload was already high in pre-pandemic periods [4], the current conditions have been described as unsustainable [6]. Our results might therefore have some general implications. Considering the average age of the GPs, demographic changes, and the expected decrease in treatment capacities in the future, the COVID-19 pandemic might act like a magnifier of the upcoming problems and distribution battles, such as prevention and screening vs. acute treatment as well as the lack of referral of patients to specialists or clinics. The GP as the gatekeeper would become the universal treatment provider. Further, the profession of general practice is already suffering from a lack of attractiveness for young doctors [75,76] and is perceived to have a disproportionately high amount of administrative tasks and comparatively low income [77]. The extraordinarily high workload and increased administrative tasks experienced by GPs during the pandemic might further deter young graduates to pursue a career in general practice.
To reduce further dissatisfaction, burnout, or even withdrawals from practices, our results highlight the following potential areas for improvement (Box 1). Box 1. Potential improvements of GPs' working condition.
(1) The primary care sector carries the main quantitative burden of care for patients with acute COVID-19 and long COVID as well as SARS-CoV-2 vaccinations. GPs would therefore potentially benefit from extended treatment options, including additional external vaccination offers and referral options for patients with acute COVID-19 or long COVID to specialists or rehabilitation to increase time for non-COVID-19-related healthcare. Information and updates on these additional local specialized care offers should be easily available for GPs. In addition, multidisciplinary teams and the possibility to allocate tasks to practice nurses might further reduce GPs' burden. (2) In the course of this pandemic or in future pandemics, GPs can be supported by revisiting, suspending, or outsourcing pandemic-related documentational and administrative work. (3) Changes in official regulations and legislation should be reduced to a minimum to increase compliance. (4) Practices should be able to postpone the implementation of obligatory changes in practices' structures during times of extraordinarily high workload, such as telematics infrastructures. These implementations should also be fully developed and should not cause additional workload. (5) Finally, feeling valued for the daily responsibilities as well as financial incentives could further increase satisfaction for staff working in general practices and might compensate for the higher workload, as has also been reported previously [58].

Conclusions
GPs are at the forefront of providing COVID-19-related healthcare in addition to routine care. We confirm that the pandemic continues to aggravate GPs' working conditions and affect other essential healthcare services. We found evidence that, without political mitigation measures, the pandemic might accelerate GPs' intention to leave the practice. Especially reductions in bureaucracy, the provision of additional vaccination sites or referral options for patients with acute COVID-19 or long COVID, and the option to postpone telematic infrastructure implementations in times of increased workload might contribute to the alleviation of GPs' current and future working conditions. Our findings provide insights into how future pandemics could be handled to achieve better primary care working conditions in times of high demand.

Conclusions
GPs are at the forefront of providing COVID-19-related healthcare in addition to routine care. We confirm that the pandemic continues to aggravate GPs' working conditions and affect other essential healthcare services. We found evidence that, without political mitigation measures, the pandemic might accelerate GPs' intention to leave the practice. Especially reductions in bureaucracy, the provision of additional vaccination sites or referral options for patients with acute COVID-19 or long COVID, and the option to postpone telematic infrastructure implementations in times of increased workload might contribute to the alleviation of GPs' current and future working conditions. Our findings provide insights into how future pandemics could be handled to achieve better primary care working conditions in times of high demand.

Supplementary Materials:
The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/healthcare11030320/s1, Table S1: GP questionnaire translated into English (baseline 2021); Table S2: GP follow-up questionnaire translated into English (follow-up 2022); Table S3: Characteristics of participating GPs at baseline and follow up (GPs in the follow-up survey also participated in the baseline survey); Table S4: Other problems and needs related to the treatment of patients with COVID-19 or long COVID: content analysis of free text answers at baseline 2021; Table S5: GPs' reasons for wanting to leave their job (n = 13): content analysis of free text answers; Table S6: GPs' general comments on long COVID at baseline 2021 (n = 33): content analysis of free text answers.

Institutional Review Board Statement:
The study was carried out in accordance with the Declaration of Helsinki and the study protocol was approved by the research ethics committee of Leipzig University (reference number 157/21-ek).

Informed Consent Statement:
Online informed consent was obtained from all participants. They did not receive an incentive for their participation. No personal data besides age, sex, and education level were assessed.

Data Availability Statement:
The data that support the findings of this study are available on request from the corresponding author.