Transfer of care and inbox management in primary care: a survey on medico-legal responsibility awareness and administrative burden in Aotearoa New Zealand

“I spent 10–15 minutes a day just dealing with copied inpatient radiology reports from the hospital; that’s at least 60 minutes a week and 48 hours a year.” – One general practitioner

Transfer of care is defined as a series of steps intended to maintain coordinated and continuous healthcare as patients move between different physical locations or levels of care.^1,2 Current mechanisms of transfer of care in Aotearoa New Zealand are known to increase primary care’s administrative burden and uncertainty as to who is clinically responsible for ongoing patient care. Significant contributors include routinely copied radiology, laboratory and histology reports for inpatient and outpatient encounters by hospital and specialist services (HSS) as well as unidirectional non-closed loop transfer of care from electronic discharge summaries (EDS). Consequently, multiple adverse patient events have resulted, leading to investigation by the Health and Disability Commissioner.^3,4 Copying results to another clinician is initiated either manually on the paper request form or a tick box on the electronic order form. Sometimes, there are even “built-in” processes wherein the HSS requester has no control over whether a copy is sent to the patient’s enrolled primary care clinician. Medico-legally, there is an expectation for primary care clinicians to act on significant abnormal results if they are a recipient of this copied information, even though they did not order the test and often are not aware of the clinical context.^5,6

Administrative burdens, especially relating to the electronic health record (EHR), are well-documented contributors to job dissatisfaction, reduced patient-facing time, cognitive overload and increased burnout among primary care clinicians and represent a significant challenge for workforce retention.^7,8,9,10

Despite the introduction of TestSafe (an online web portal that allows Northern region primary care clinicians to access all test results ordered by HSS) in 2010, the widespread culture of sending copied results to primary care has persisted and continues to grow, especially as healthcare documentation is seeing an increase in digital adoption.

In April 2024, Te Whatu Ora – Health New Zealand published a guidance document titled “Transfer of Care and Test Results Responsibility” which outlined principles to ensure test results are followed up safely in both inpatient and outpatient settings.¹¹ It affirms that the clinician who ordered the test is responsible for reviewing and actioning the results regardless of subsequent transfer of care, unless explicitly agreed to and documented. If action is required when copying results to another clinician, a documented handover, with agreement from the receiving clinician to accept the responsibility, involving a closed loop communication, is expected. Finally, as part of safety netting, any clinician copied into a result which is significantly abnormal needs to ensure appropriate action has been taken. Unfortunately, the document does not offer further recommendations on how these principles can be operationalised as they are not currently common practice.

Additionally, access to general practice in Aotearoa New Zealand continues to decline in response to factors such as increased patient complexity and clinical demand, a capitation funding formula that is slow to adapt and a workforce shortage from both an intake and exit perspective.¹² In the 2024 Ministry of Health – Manatū Hauora survey,¹³ 75.6% of adults and 66.6% of children reported visiting general practice at least once in the past 12 months (compared with 78.9% and 74.5% respectively in 2019). “Time taken to get an appointment was too long” was the most significant barrier to general practice access (25.7% of adults and 18.5% of children in 2024, 21.2% and 14.8% in 2023 and 11.6% and 8% in 2022), and the rates of patients who visited the Emergency Department at least once in the past 12 months are also on the rise (17.8% of adults and 21.3% of children in 2024 versus 14.2% and 12.7% in 2020). This is a worrying trend as international evidence has consistently highlighted the importance of investing in primary care as a critical solution to addressing rising demand on HSS and achieving downstream savings in the healthcare system.^14,15,16,17

This study aimed to assess primary care clinicians’ preferences regarding transfer of care in the Northern region. We hypothesised that most clinicians would opt out of receiving copied results when variables such as medico-legal responsibility and sustainability of their profession were considered. Additionally, we sought to evaluate the proportion of non–patient-facing clinical time spent on administrative tasks as well as clinician wellbeing and long-term career plans.

Methods

This was an anonymous cross-sectional survey designed for clinically active primary care clinicians (general practitioners [GPs], urgent care physicians, nurse practitioners [NPs]) whose main place of work is in the Northern region. The Northern region encompasses four districts (previously known as district health boards), namely Te Tai Tokerau, Waitematā, Te Toka Tumai Auckland and Counties Manukau; it also comprises 37.7% of Aotearoa New Zealand’s total population. These geographical locations were selected due to TestSafe access and coverage for primary care.

Survey content development was designed with input from Medical Protection Society advisors, the primary care advisor from the Health and Disability Commission, clinical directors from the primary health organisations (PHOs) in the Northern region and local GPs. The survey contained 47 questions in total and was largely multi-choice (see Appendix A for full survey questions).

The survey was hosted on Research Electronic Data Capture (REDCap) and distributed electronically to primary care clinicians with support from the PHOs, the Royal New Zealand College of General Practitioners (RNZCGP), the Royal New Zealand College of Urgent Care (RNZCUC), Nurse Practitioner New Zealand (NPNZ) and the Goodfellow Unit (an educational organisation for primary care operating out of the Department of General Practice and Primary Health Care at the University of Auckland). Additional distribution was through a private Facebook group consisting of GPs in Aotearoa New Zealand with verified identity. The survey link was active over a 2-month period from August 2023 to October 2023.

The co-primary endpoints were: the proportion of primary care clinicians who were aware of their medico-legal responsibility upon being copied into tests they did not initiate; and the proportion of primary care clinicians who preferred to receive these copied results in their electronic inbox. Secondary outcomes evaluated primary care professionals’ preference on transfer of care, education on discharge summary composition, number of patient-facing versus non–patient-facing clinical hours, career trajectory and current job satisfaction. Respondents were asked to exclude any off-site teaching (e.g., lecturing at a university), continuing medical education and any business-related administration when calculating the total hours spent on clinical work. Hours spent beyond their usual worktime were to be included, e.g., working through breaks or remotely checking their electronic inbox after hours and during weekends.

Power calculation

Based on Te Kaunihera Rata o Aotearoa | Medical Council of New Zealand (MCNZ) data, 1533 doctors work as GPs and 122 doctors work in urgent care in the Northern region.¹⁸ Based on 95% confidence, a 5% margin of error and a conservative estimated response distribution of 50% and 80% power, the number of respondents required was calculated at 308 for GPs and 93 for urgent care physicians. Given there were no specific data available for the number of NPs working in primary care, we estimated there were 167 NPs in the Northern region based on a recent New Zealand Nursing Council report, which approximated that 60% of 702 NPs work in primary care, and further extrapolating that to 39.7% working in the Northern region (using the GP distribution ratios in Aotearoa New Zealand).¹⁹ Thus, 117 NP respondents were required to reach the same power calculation for GPs and urgent care physicians.

Statistical analysis plan

Descriptive statistics were used to summarise participant characteristics and survey responses, with frequencies and percentages reported for categorical variables and medians with interquartile ranges for continuous variables. The Wald method was used to calculate 95% confidence intervals. Comparisons between the GP sub-group and the Te Kaunihera Rata o Aotearoa | Medical Council of New Zealand 2022 workforce survey cohort were performed using Pearson’s Chi-squared test. Fisher’s exact test was used to assess differences in test result responsibilities and opinions on care transfer across the four professional groups surveyed.

The distribution of total, patient-facing and non–patient-facing weekly working hours by profession was analysed using one-way analysis of variance (ANOVA) or Kruskal–Wallis one-way analysis of variance on ranks, depending on data distribution. Improbable responses, such as total work hours exceeding 80 hours per week or patient-facing hours exceeding total reported hours, were excluded to maintain data integrity. Sub-group analyses for GPs explored associations between the proportion of non–patient-facing hours and various participant characteristics, as well as survey responses relating to medico-legal awareness and preferences for routine inbox items, using the Kruskal–Wallis one-way analysis of variance on ranks. Post hoc pairwise comparisons were performed using Tukey’s Honest Significant Difference test or Dunn’s Test for Multiple Comparisons, as appropriate. Statistical significance was set at p<0.05, and p<0.025 for pairwise post hoc testing.

Additionally, multivariate negative binomial regression was used to analyse the proportion of non–patient-facing hours worked per week, examining the influence of age, gender, ethnicity, vocational scope qualification, years of primary care experience, and practice type (Cornerstone certified, Very Low Cost Access [VLCA], set number of enrolled patients). Each factor was assessed individually (univariate) and collectively (multivariate) to identify which remained significant after accounting for potential confounding.

Ethics approval

Ethics approval for the study was granted by the Auckland Health Research Ethics Committee in August 2023 (reference: AH26309).

Results

Over the 2-month survey period, we received a total of 506 responses. Of these, 36 responses were excluded as their main work site was not located within the four districts of the Northern region. Out of the remaining 470 responses, 428 identified their primary role as GPs, 15 as urgent care physicians, 9 working in other primary care (such as aged residential care, hospice care, skin cancer and student health) and 18 as NPs. While over 99% of survey responses were complete, data analysis shown in the tables exclude any missing responses.

The survey response rate is estimated at 27.9% for GPs, 12.3% for urgent care physicians and 10.8% for NPs in the Northern region.^18,19

Respondent characteristics

Respondent demographics and practice characteristics are displayed in Table 1. NPs are reported together due to the small numbers of respondents (16 in general practice and two in other primary care). Ethnicity reporting follows the same single prioritised ethnicity method as used in the 2022 MCNZ workforce survey.¹⁸

The largest group of respondents are from the 55–64 years age group, followed by the 35–44 years age group. Of the respondents, 60.9% are female and 46% identify as NZ European/Pākehā for their ethnicity. Over 80% of respondents hold a specific vocational scope qualification and have over 6 years of experience in primary care (minimum 3 years training for GPs and 4 years each for urgent care physicians and NPs). Some general practice and urgent care respondents do not hold their respective vocational scopes as this is not legally required in Aotearoa New Zealand. Nearly 80% respondents were from Cornerstone accredited practices that have voluntarily obtained the RNZCGP certification for equitable and high-quality care. Nearly 40% respondents work in VLCA practices, which receive extra funding to reduce patient co-payments for serving high-needs populations.

General practice operates largely in two models: a set number of patients enrolled directly under a primary care clinician’s name, where all associated results and notifications go to a specific clinician’s inbox; or administrative work from all enrolled patients are shared amongst all doctors within one practice. More GPs (63.1%) identified as being in the former model of care.

There were 428 GP respondents (exceeding the 308 required to power the study) where a meaningful statistical comparison was possible with MCNZ data of doctors reporting working predominantly in general practice (Table 2). Our GP respondents’ age and gender distributions were statistically different to MCNZ data, with higher representation for those in the under 35 years, 35–44 years and 45–54 years age groups, and those identifying as female. Ethnicity and their main work site district in the Northern region were not statistically different. However, it should be noted that MCNZ identifies main work location based on a doctor’s annual practicing certificate renewal address.

View Table 1–6.

Medico-legal awareness, Transfer of Care and Test Results Responsibility (Table 3)

Two-thirds (66%) of respondents were aware of their medico-legal responsibility when being copied into results of tests they did not initiate. However, 68.4% of these respondents were only made recently aware in the past 3 months.

With the exception of NPs, most respondents had a preference to not receive copied radiology, histology, microbiology, haematology and biochemistry test results. Preference for endoscopy and cardiology test results were split evenly. Doctors largely preferred not to be sent notifications related to outpatient clinic referrals initiated by HSS, but opinions were split among NPs. Hospital admission notifications were generally preferred by participants (61.9%) in comparison to hospital discharge notifications (37.7%), which would be a duplicate to the EDS (if finalised at time of patient discharge).

There was strong support from almost all respondents to use the National Enrolment Service (NES) database (patient enrolment information held by the Ministry of Health – Manatū Hauora which determines General Practice capitation funding) as source of truth for where HSS direct all patient-related communications to, including copied reports and notifications.

From a transfer of care perspective, four out of five respondents agreed with the statement that “any results copied to primary care (including the associated responsibilities) need to be discussed with and accepted by a relevant clinician in primary care (e.g. GP, NP or a practice nurse)”. An electronic system that would allow the recipient to accept or decline the transfer of care requests was the most preferred communication option (48.7%), followed by a discussion over the phone (33.4%).

Most respondents agreed that there was a need for standardised education on discharge summary composition both at medical student and junior doctor levels.

Work-life balance and career trajectory (Table 4)

Of the respondents, 41.5% did not feel they were able to maintain a good work-life balance. Despite similar burnout rates in GPs and NPs, urgent care physicians still reported positively regarding their work satisfaction (86.6% were “moderately satisfied” or “very satisfied”). Regardless of the reported wellbeing indicators, all primary care professions in the survey have reported an overall reduction in their patient-facing hours in the past 5 years, with increased administrative burden as one of the main contributors.

Of the respondents, 26.4% expressed an intention to leave primary care in the next 3 years, but only GPs identified that increased administrative burden was the main reason to alter their career trajectory. Overall, GP respondents were the least inclined (35.7%) to recommend their profession as a career in Aotearoa New Zealand.

Non–patient-facing clinical time

Nine improbable responses were excluded from data analysis (Table 5) where the reported patient-facing hours exceeded total clinical hours reported or where the total hours exceeded 80 hours per week.

Primary care clinicians reported working a median of 37 hours in total per week (Interquartile Range [IQR]: 28–45), of which 30% (IQR: 20–43%) was non–patient-facing (Table 5). While total hours did not significantly differ between the four professions, urgent care physicians were found to spend proportionally less time on non–patient-facing tasks compared to GPs, NPs and other primary care physicians.

Further subgroup analysis, conducted specifically for GPs, examined the proportion of non–patient-facing clinical time in relation to respondent characteristics and their preferences for receiving specific copied test results (Table 6). Post hoc analysis showed significantly lower proportions of non–patient-facing hours by GPs aged <35 years (versus GPs aged 45 years and older) or 35–44 years (compared with those over the age of 65 years), GPs of Other non-European ethnicity (compared with Chinese, Indian, Other European and New Zealand European GPs), and GP trainees/registrars who were 2nd year and above (compared with those with a vocational scope in general practice; Dunn’s Test for Multiple Comparisons, p<0.025). Significantly higher proportions of non–patient-facing hours were reported by female GPs, GPs with over 10 years of experience (compared with GPs with less than 5 years’ experience), GPs with a set number of patients enrolled under their care and GPs who work in a non-VLCA practice (Dunn’s Test for Multiple Comparisons, all p<0.025). Awareness of secondary medico-legal responsibility and preference to receive copied reports were not shown to affect the participants’ proportion of non–patient-facing clinical time.

Discussion

This anonymous survey set out to clarify where front-line primary care clinicians stand on the topic of transfer of care and how it affects their proportion of non–patient-facing time, wellbeing and long-term career plan.

We found a substantial number of primary care clinicians that were not aware of their medico-legal responsibility when copied into results (over one in three for GPs, over one in four for urgent care and over one in five for other primary care and NPs). In fact, of the participants who were aware, only one in four became aware in the past 3 months, indicating the same survey would have yielded a result where less than half of the participants were aware of their medico-legal responsibility had the survey been disseminated 3 months prior. This follow-up question was added due to recent renewed interest in general practice on how to manage the continued surge of non–patient-facing clinical workload and webinars on this topic led by the author for education and survey recruitment purposes. These findings are in-line with other studies showing poor awareness in clinicians, including those in the HSS, in the respective medico-legal topics surveyed (e.g., documentation, confidentiality, consent, child protection, driving clearance), highlighting this field as a knowledge gap that commonly exists across the workforce.^20,21,22,23

It was also confirmed with the medico-legal advisers that misdirected test results also confer the same secondary responsibility to the incorrectly copied clinician. At present, misdirection of results frequently occurs when HSS administrative staff do not check (or have access to) primary care enrolment details on the NES database during routine patient encounters. The default response is to update the primary care provider details on the HSS system based on what the patient reports. Misdirection of clinical information not only results in additional administrative and clinical burden to primary care to notify HSS and correct but confers a clinical risk for patients due to the time it takes for these reports to be redirected to the correct recipient. It is not uncommon that EDS would handover to primary care to follow-up pending or planned diagnostic tests for patients on discharge. Furthermore, there are times when the correct recipient for these electronic documents and results is “no recipient”, as the patient’s primary care enrolment has expired and not been renewed.

In patient safety literature, communication failure is widely recognised as one of the leading causes of adverse patient outcomes and, as a result, various versions of handover templates (e.g., SBAR, ISBAR, ISOBAR) and checklists (e.g., peri-operative surgical checklists) have been used to reduce communication errors.^{24,25,26,27,28} However, when it comes to transferring of patient care back to the community, there is a stark contrast where there is no standardised education or expectation to utilise closed loop communication for the clinician completing the EDS document (often the most junior member on the team).

Furthermore, most doctors preferred not to receive radiology, histology, microbiology, haematology and biochemistry results that were not ordered by themselves as this creates confusion as to identifying the main responsible clinician, as well as it being unnecessary, as all Northern region primary care providers can access TestSafe. Although arguably this may result in an overall safer outcome for patients by having primary care act as an additional safety net for the requesting clinician, this contradicts Health New Zealand – Te Whatu Ora’s “Transfer of Care and Test Results Responsibility” and doesn’t include a closed loop communication that four out of five respondents in this survey have expressed support for.¹¹

Additional qualitative work will need to be done to understand why there is a split preference for endoscopy and cardiology reports. However, one of the possible contributing factors is that endoscopy reports are often also used as a transfer of care document, e.g., asking primary care to follow up on biopsy results to initiate Helicobacter pylori treatment or re-refer for repeat procedure in a few years’ time. It is uncommon for primary care to receive a separate transfer of care communication via a separate document. Inclination to receive cardiology investigations could be due to perceived usefulness of having these reports conveniently stored on the primary care patient management system (PMS), rather than needing to separately log onto TestSafe to retrieve the results.

With the exception of urgent care physicians, primary care professions with work more closely related to patient long-term care are noted to spend a median of 31% and average 33.5% of total hours on non–patient-facing clinical time. This was not unexpected due to a continued flow of discharge summaries, specialist clinic letters and diagnostic tests that are not always related to direct patient consultation. The RNZCGP conducted a study in 2024 and reported 30% of total work time spent on non-patient contact activities (different to this survey in that it included continued medical education and business-related administration as part total work time).²⁹ Similar figures were noted in other studies; 23.7% from 142 family physicians by reviewing their EHR event logs over a 3-year period and 25% from 1,343 members of the Ontario College of Family Physicians in a cross sectional survey.^30,31 While the widespread introduction of EHR has been shown to improve efficiency in certain workflows and patient care, the subsequent increased time spent interacting with EHRs have also resulted in a wealth of literature noting increased associated physician burnout, not limited to primary care providers.^32,33,34,35

Proficiency in information technology systems (e.g., the ability to touch type, utilisation of artificial intelligence to take and summarise consultation notes or navigate the PMS with quick actions/smart phrases) were not evaluated in this survey due to the survey aim and length consideration, even though they have been identified to affect a clinician’s HER efficiency.^36,37 Interestingly, while awareness of medico-legal responsibility or previous experience in medico-legal matters have been shown to affect clinician behaviour,^38,39,40 we found no significant difference in the proportion of non–patient-facing clinical time spent for our GP respondents.

Our survey revealed significant issues with GPs’ work-life balance and career intentions in Aotearoa New Zealand. In fact, our GP respondents reported a stronger intention to retire earlier compared with a 2024 RNZCGP workforce survey (which only included GPs with specialist vocational scope qualification); 30% versus 21% in the next 3 years, 21% versus 23% in 3 to 5 years and 49% versus 56% in over 5 years.⁴¹ Despite a recent commitment by Health New Zealand – Te Whatu Ora to achieve 300 general practice registrar intakes per year by 2026 and increased funding for training NPs, our survey has highlighted a fundamental problem of the perceived unattractiveness of general practice as a viable career option.⁴² Efforts to increase intake will be significantly undermined if workforce output continues to decline from the hidden cost of administrative burden (i.e., reduction of patient-facing clinical hours) and failure to retain our training and emerging workforce.

Limitations and strengths

One of the limitations of this study was the low numbers of non-GP respondents. While it is true they can be recipients of EDSs and copied results, the low response was accepted at the end of the survey closing period. For urgent care this was due to recognition of the acute and episodic nature of their work. For NPs and other primary care doctors, this was due to the difficulty in ascertaining their numbers in the Northern region, and it also became apparent that there were limited outreach modalities for these two professions.

Even though we recognise the limitations of extrapolating findings, the transfer of care mechanism and circumstances described in the study is not unique to the Northern region. All regions in Aotearoa New Zealand utilise EDS and clinic letters as their main method for transfer of patient care. Despite TestSafe being unique to the Northern region, South Island | Te Waipounamu region has HealthOne, and most primary care providers in the Midland and Central regions have some form of limited access to view hospital diagnostic results.

A strength of this paper lies in the comprehensive nature of transfer of care related topics covered in the survey. Medico-legal awareness was also ascertained early in the survey to ensure this is not a confounding factor with analysis of all subsequent responses both in preference to receive copied results and the impact on their hours.

Suggested future research directions include repeating the survey for primary care providers in different regions, further qualitative work on identifying the reasoning for a clinician’s preference to receive copied results and modifying the survey to ascertain the viewpoints of HSS clinicians regarding their awareness of medico-legal responsibility of copied results and preference for transfer of care.

Conclusion

Our survey aimed to evaluate primary care professionals’ understanding and preferences in various aspects of transfer of care including their long-term career trajectory. It affirms that there is currently a medico-legal knowledge gap present in our primary care workforce. While it does not seem to affect clinician time spent on non–patient-facing clinical work, most would prefer not to be routinely copied into results of tests they did not order unless a closed loop communication with the requester has been undertaken. To improve primary care as a viable profession and improve productivity, a collaborative effort between primary care and Health New Zealand – Te Whatu Ora will be required to address the current less-than-ideal method of transfer of care and realise the principles of “Transfer of Care and Test Results Responsibility”.

View Appendix.