Introduction

Primary care addresses the main health problems in communities, providing “promotive, preventive, curative, and rehabilitative services accordingly” [1]. In the Republic of Ireland, general practice teams are critical to primary care provision, providing an average of four GP consultations to each person in the state each year [2]. Almost all Irish general practitioners (GPs) have computerised their practices [3], with practices using one of four accredited electronic medical record (EMR) programmes to deliver care to patients [4]. The leading package is “Socrates”, used by more than 55% of GPs [5].

Preventive care provided routinely by Irish general practice teams includes the primary childhood immunisation programme (PCIP), which involves all vaccinations delivered in the first years of life [6, 7] and health screening via state-supported programmes such as the cervical screening programme “Cervical Check” [8].

The COVID-19 pandemic disrupted provision of routine healthcare due to a combination of demand-side issues, such as patient avoidance of healthcare settings due to fear of exposure and fear of burdening health systems; and supply-side issues, including service cancellations, staff redeployment, facility closures, and supply chain problems [9, 10]. As the virus spread and hospitals braced for the potential impact, opportunities for routine healthcare provision in primary care were disrupted [11] and concerns were expressed regarding knock-on effects of such missed opportunities [12, 13], including vaccination programmes and cancer screening.

This project aims to assess the effects of the pandemic on routine GP care activities by analysing summary electronic medical record data relating to patient record updates, appointments and medications prescribed.

Methods

This study used anonymized retrospective visit data from GP EMRs. Ethical approval was granted through the University of Limerick Hospital Group Research Ethics Committee (Reference 069/2021, approved 11th June 2021).

Setting

All thirty practices of the ULEARN-GP Research and Education network [14] in the Irish Midwest (counties Limerick, Clare and North Tipperary) using the Socrates EMR were invited to participate in this study by email. With a broad spread of rural and urban areas [15] and population of over 400 k (8% of national total), the Midwest is broadly representative of the national picture, though has a slightly older and more deprived population than that seen nationally [16]. Ten daytime general practice services (2 city, 2 large town, 4 small town and 2 rural practices) agreed to participate.

Data collection

Summary data pertaining to all entries to patient files, all patient appointments and all medicines prescribed during the 3 years 2019 to 2021 inclusive were collected. Standard reports available through the software were used to provide raw data for the study period. All data were anonymised irrevocably on site using custom software written by the lead author to process reports from the practice EMR. Anonymised data were sent back to the research team using the Health Service Executive secure email service “Healthmail”.

Analyses

Descriptive statistics using standard formulae and analyses were carried out using the open source statistical software package R [17] to assess trends in overall prescribing and overall practice activity over the study period, exploring in particular the provision of telephone consultations, childhood immunisations and cervical screening tests. Differences in proportions of consultation types from 2019 versus 2021 and in patients receiving telephone consultations versus all appointments by age group were assessed using the χ2 test.

Results

There were 1.18 million entries made to 31.9 k patient records in the ten practices over the study period, which translates as just over 1 entry per chart per month for every patient in the dataset. More than half of these entries were tasks or updates to records completed by the various general practice team members over the course of the normal working day. Of the total number of entries, 510 k were recorded as distinct appointments delivered by GPs, practice nurses and other healthcare professionals. Of these 510 k appointments, 1% were completed by GP interns (first year post graduation), 5% by GP registrars (GPs in training), 30% by practice nurses and 61% by GPs, with the remaining 2.5% of entries completed by other ancillary health professionals and students. Numbers of healthcare professionals working in the practices was similar across the study period. Across the ten practices, there were 376 distinct consultation types relating to appointments provided over the 3 years, which were classified into the eight Consultation type categories shown in Table 1.

Table 1 Overall appointment activity over the 3-year study period across 10 general practice clinics in the Midwest of Ireland, classified by consultation type. The right-most column examines for significant changes in proportions of the various consultation types from 2019 versus 2021
Full size table

There were 2.85 million medication items prescribed by the ten practices over the study period, which if taken across the study population of 31.9 k patients translates as just under 2.5 items per patient per month (or 7.5 items per quarter) over the study period. Figure 1 shows the breakdown of appointment and prescribing activity per patient per quarter.

Fig. 1
figure 1

Breakdown of appointment and prescribing activity per patient per quarter

Full size image

For average appointment count per patient by age group in the dataset for 2021, there were 3.7 appointments per patient aged 0–17 years, 5.4 appointments per patient aged 18–44 years, 6.4 appointments per patient aged 45–64 years and 10.8 appointments per patient aged 65 years-plus in 2021.

Finally, Table 2 demonstrates the breakdown of practice recording of main appointment types in 2021. For the differing age ranges in the left-most column of Table 2, the proportion of in each age grouping receiving care via teleconsultation was assessed using the χ2 test, yielding significant results (p < 0.01).

Table 2 Breakdown of activity (general, teleconsultation, other) by practice for 2021
Full size table

Discussion

Summary of main findings

This study demonstrates that general practice activity increased from 2019 to 2021, with particularly large increases in teleconsulting and vaccination activity, likely driven directly by COVID-19. In addition, activity levels in cancer screening (when permitted again in 2021) and childhood vaccination programmes were not adversely affected by the pandemic. General practice continued to deliver these essential preventative activities. There was a notable shift towards more granular recording of activity and while lack of standardisation across practices introduced some analysis challenges, grouping of activity facilitated comparisons between practices and over time. Over the course of the study period, delivery of GP care for 31.9 k patients involved provision of 510 k consultations recorded in the EMR but involved more than twice this number of record entries in patient charts by GP teams. This speaks to the hidden workload of general practice that is not readily assessed by standard metrics, which tend to focus on consultation numbers [18], which will miss a large portion of work required to deliver general practice-based care [19, 20].

Comparison with existing literature

General practice care delivery in Ireland is a team effort, with nurses being responsible for almost a third of appointment activity in the study, with GPs and GPs in training being responsible for approximately two-thirds, which is in keeping with national estimates [5]. As may be expected, patients at extremes of age have increased care needs [2]. While COVID-19-related illness, long-COVID and COVID-19 vaccinations have all been drivers of increased GP workload in Ireland in recent years [21, 22], demographic trends leading to a higher proportion of patients in older age groups also continue to exert demand side pressures on general practice [23].

Overall, consulting activity reduced for a period in early-mid 2020, before rebounding in 2021 (see Fig. 1). Prescribing activity did not dip to the same degree, and indeed is in keeping with the generally upwards trend seen nationally [22]. Growth of teleconsulting, on top of the pre-pandemic consultation workload, means that GP teams are likely now busier than ever before.

Adapting to workload challenges and pandemic pressures through use of teleconsulting seems apparent in our study, although the pre-pandemic picture seems to be incomplete. Just 2% of activity in 2019 was recorded as teleconsultations, which is much lower than the 10.5% reported elsewhere [24]. It is felt that better recording of this workstream occurred as GP teams adapted appointment booking systems and their workflow to cope with additional demand for teleconsultations. This adaptation led to teleconsultation appointments being recorded more faithfully as GPs diverted some of their time to dealing with this substantial shift in workload. National data suggests that teleconsulting activity is falling, with 60% of GP consults in October 2020–March 2021 (a period of very high COVID-19 activity) being delivered face-to-face, compared to 84% in early 2022 [25, 26]. This may indicate patient and/or GP preference for face-to-face consults [27, 28].

The higher likelihood of teleconsultations being used for younger age groups is perhaps expected given their increased levels of comfort with technology and less complex overall care needs.

In any case, consequences of the dramatic increase in recorded teleconsultation activity for overall workload and unintended consequences require further study. While potential to increase inequality of care provision to already disadvantaged groups [29], medicolegal uncertainty for care providers [30] and uncertainty around impact on antimicrobial stewardship [31] mean we should tread carefully, it is perhaps the potential to undermine the “establishment and maintenance of relational trust, with a negative impact on the quality of care and patient safety” [32] that requires most consideration. Interestingly, patients seem to value teleconsultation with the primary care physician they know well [33], which highlights the existing relationship we have with patients, often formed over years, as the foundation for high-quality and effective care [34].

Regarding provision of routine care, childhood vaccinations increased over the course of the pandemic, which is in keeping with national trends [35], and may be important given recent measles outbreaks across Europe [36]. Cervical screening tests dipped in 2020 before rebounding in 2021, which is in keeping with a national closure of the programme for 3 months in 2020 [37].

Strengths and limitations

This analysis of this large body of real-world data offers an overview of general practice activity in the Midwest of Ireland over the study period. Analysis of all EMR entries, in addition to prescribing and consulting activity, highlights the “hidden workload” relating to administrative and follow-up tasks that is critical to well-functioning primary care.

Recruitment via email to a University-associated research network may introduce selection bias, and generalisability may be adversely affected by the selection of one specific EMR and the Irish Midwest as our study setting. Incomplete or inaccurately recorded EMR data could also affect validity. Finally, consequences stemming from teleconsultations (e.g. did a teleconsultation lead to a prescription or a future appointment?) were not recorded in this study.

Conclusions

As seen internationally [38], general practice in Ireland was challenged and changed by the pandemic, adapting swiftly despite resource constraints and maintaining care provision where permitted. Activity is now being recorded more faithfully by practices, particularly for teleconsultations, and while evidence around teleconsulting’s long-term benefit is lacking, this element of GP workflow seems to be here to stay [2].

Study of GP EMRs can offer useful insights into workflow trends and while standardisation of recording activity is desirable, we must avoid burdensome EMR activities, such as excessive data entry or administrative work that is not valuable to the clinician or the patient [39, 40], as this may interfere with direct patient care and contribute to physician burnout [41].

GP EMR data can validate data from other sources [35], while providing more granular information relating to patient and healthcare professional characteristics, to aid timely assessment of immediate challenges such as pandemics and long-term planning in the face of changing population and demographic trends.