An area-based description of closed books in general practices in Aotearoa New Zealand

Introduction

Primary Health Care (PHC) in Aotearoa New Zealand (NZ) is typically accessible through a general practice. These are community-based independent organisations delivering first level care by general practitioners (GPs), nurse practitioners, nurses, and, increasingly, a range of other PHC providers. People can enrol with a general practice, which gives them benefits such as lower consultation fees, prevention initiatives, and more coordinated care. General practices benefit from enrolment as they receive funding from the government based on the number and type of patients they have enrolled. Capitation funding is a large component of general practice income, as are out-of-pocket payments by the patients themselves.

General practices are able to restrict the enrolment of patients, either through selecting patients for enrolment or by not enrolling any new patients at all. A general practice that is not enrolling any new patients is said to have ‘closed books’.

Typically, a general practice closes its books when it has reached capacity, but this has flow-on effects for those patients who are not enrolled as they find it difficult or, more often, impossible, to get appointments at a closed books practice (Irurzun-Lopez,^A unpublished observations). If a patient has no open books practices available in their area, the choices for the patient are to (1) go to an accident and emergency clinic/after-hours clinic which are typically more expensive, (2) go to the emergency department at the local hospital, which is generally free to the patient but is discouraged, or (3) try to find an open books practice in another area. Patients and staff from multiple places around NZ report patients driving long distances to get to an open books practice.^1–3

Closed books have been a NZ issue for some time.^4,5 However, the problem has been worsened by the effects of the COVID-19 pandemic for two main reasons: (1) the extra health burden in PHC of dealing with COVID-19 prevention, protection, and illness; and (2) the border restrictions that the NZ government imposed that restricted foreign doctors and nurses from entering the country. NZ is highly dependent on overseas health personnel, with around 46% of GPs having been trained overseas.⁶ The extra work placed on GPs has led to lower work satisfaction with some leaving or reducing the hours they work, worsening the problem.⁷

The areas of interest in this study were NZ’s 20 District Health Board (DHB) districts. DHBs were administrative units through which government-funded health resources were managed, and while now amalgamated into a single entity (Te Whatu Ora or Health NZ), DHB districts are still used for reporting purposes.

The aim of this research was to look at which DHB districts were more likely to have closed books general practices, travel and cost impacts, and what characteristics of general practices and DHB districts were associated with open and closed books.

Methods

Results

Number and percentage of closed books general practices

Overall, there were 347 general practices with closed books. The Canterbury (45) and Southern DHB districts (32) had the greatest number, while West Coast (2), Tairāwhiti (3), and Lakes DHB districts (3) had the fewest (see Fig. 1a). Wairarapa DHB district had only one open books practice. Some places within DHB districts were particularly poorly served with both New Plymouth and Invercargill having no open books practices.

Fig. 1. 

The (a) number and (b) percentage of closed book general practices by DHB district and (c) the average distance travelled by road from a closed books general practice to an open books general practice by DHB district.

Altogether, 33% of general practices had closed books. Wairarapa (86%), Midcentral (81%), and Taranaki DHB districts (81%) had the greatest percentage of general practices having closed books while West Coast (14%) and Lakes DHB districts (16%) had the lowest (see Fig. 1b). The three Auckland DHB districts all had low percentages of closed books practices: Auckland (17%), Counties Manukau (17%), and Waitemata (18%).

In regional terms, the lower North Island has a clustering of DHB districts that had high rates of closed books practices.

Distance travelled and cost of substituting a closed books general practice with an open books general practice

The distance by road from a closed books practice to the nearest open books practice was calculated and averaged over the closed books practices in each DHB district (see Fig. 1c). The DHB districts where these distances were greatest were Wairarapa (26.9 km), Taranaki (18.8 km), and Southern (12.8 km), while it was less than 1 km for all three Auckland DHB districts. Across all DHB districts, the distance to an open books practice averaged 5.8 km.

Patients may choose to bypass the closest open books practice and go to the second closest practice. For the Wairarapa DHB district, the average distance to the second closest open books practices was 62.5 km, and for Taranaki DHB district it was 57.8 km. Fig. 2a displays the map of locations of open books practices (green) and closed books practices (red) in Taranaki. It clearly shows that there were large areas in the DHB district where it was difficult to access PHC enrolment.

Fig. 2. 

Open books (green) and closed books (red) general practices in (a) Taranaki DHB district and (b) the lower North Island.

Fig. 2b shows open books practices (green) and closed books practices (red) in the lower North Island. This map also shows large areas where it is difficult to enrol in a general practice.

The DHB districts where the percentage of closed books were high typically had small populations. Although 15% of the DHB districts had more than 80% of their general practices with closed books, these only covered 7% of the national population. On the other hand, 30% of DHB districts had less than 20% of their general practices closed to enrolments but 46% of the national population lived in those DHB districts.

Overall, the change in cost of a consultation when substituting a closed books practice with the nearest open books practice was an average gain of $2. The DHB districts with the greatest average gains were Capital and Coast ($15) and Auckland ($12), and the DHB districts with the greatest average losses were West Coast (suppressed due to small numbers) and Northland ($15).

Modelling of closed books general practices

No association was found between the proportion of closed books practices in a DHB district and (1) the number of GPs per head of population (P = 0.5747), (2) the number of consults per enrolees per year (P = 0.8304), (3) the change in population between June 2019 and June 2022 (P = 0.2065) or (4) the proportion of people enrolled in PHC in a DHB district (P = 0.9082).

Logistic regression was used to model whether a general practice had open or closed books in relations to fees, in $10 bands, for a consultation for a person aged 18–64 years without a CSC. Fig. 3 shows this relationship between fees and having closed books (P < 0.0001). When fees are low or high, the general practice is less likely to have closed books. When the fees are towards the middle of the fee schedule, the general practice is more likely to have closed books.

Fig. 3. 

The graph of odds ratios for the association between a general practice having closed books and the fees charged in general practice.

The final model looked at the association between open or closed books and practice type (VLCA or not VLCA). There was a strong association with practice type (P < 0.0001) where general practices who were VLCA were less likely to have closed books (odds ratio 0.49, 95% CI 0.36–0.65, P < 0.0001).

Discussion

General practices with closed books can add barriers to accessing PHC. When closed books general practices are concentrated in one area, the effects of these barriers can be exacerbated. This study has shown that there were a large number and proportion of general practices with closed books within DHB districts across Aotearoa New Zealand, as reported on 1 June 2022. For some locations, this had the potential to be onerous for patients, with large travel distances, and hence travel times and costs, to get to the nearest open books practice (measured from a closed books practice). In this analysis, we have considered the nearest open books practice as a substitute for a closed books practice; however, some patients might not wish to, or be able to, substitute the nearest open books practice eg due to the nearest practice charging higher fees, people not being comfortable at the practice, or people having a history of unpaid bills. On the other hand, patients may feel forced to remain at an unsuitable practice as they have no reasonable alternative. Substitution with another practice therefore depends on time, logistics, levels of comfort, and financial ability to travel. The impact of closed books is to limit people’s ability to choose their preferred/most suitable provider.

The NZ Geographic Classification for Health (GCH) classifies areas into five categories of rurality: two types of urban areas and three types of rural areas.¹⁴ In terms of the GCH, those DHB districts with over 80% of general practices with closed books (Taranaki, Midcentral, and Wairarapa) have similar characteristics. They have one urban area, classified at level 2 (U2), surrounded by rural areas at all three levels (R1–R3). While these characteristics are not unique to these DHB districts, the top five of six DHB districts with the highest percentage of closed books share these characteristics. The previously mentioned paper found that the highest crude mortality rates were found in U2, followed near equally in R1 and R2, suggesting the need for health services is higher relative to other areas.^14,15

There are known difficulties with accessing PHC in rural areas. In NZ, previous problems with data quality have limited the ability to do true comparisons between rural and urban populations, but evidence suggests there is disparity in disease incidence, access to services, and outcomes, especially for rural Māori and disabled people.^16–20

We know from the NZ Health Survey that there are populations who experience barriers in accessing PHC, with cost being the biggest barrier.²¹ In pooled data from that survey (2017/18 to 2020/21), about 14% of the adult population did not have a GP consultation in the year prior, when in need, because of cost.²¹ Moreover, Jatrana and Crampton examined the extent to which financial barriers to accessing PHC had an effect on health outcomes, through a panel study.²² They found that for all health outcomes studied (physical, mental, and self-rated health), missing visits with a GP were associated with deterioration of health outcomes.²² Furthermore, not being enrolled in PHC at all is associated with poorer health outcomes. An analysis of all amenable deaths in NZ between 2008 and 2017 found that those who died were more likely to not be enrolled, after adjusting for age, sex, ethnicity, and deprivation.²³

The inability to enrol nearby, together with the extra time and cost associated with travelling to get to an open books practice, hampers multiple dimensions of access to health as conceptualised by Levesque;²⁴ from the ability to reach (transport, mobility, social support), ability to seek (personal and social values, culture, autonomy), ability to pay (income, social capital), and ability to engage (information, caregiver support). Thus, not just the high prevalence of closed books but also their distribution, as shown by this analysis, intensifies the negative impact on access, compounding effects of barriers to care, having to consider extra time, economic costs and perhaps risking cultural acceptability when accessing services in a different setting. This cultural impact has been noted, for example, for rural–urban differences, which, in NZ, also interplay with ethnicity, with Māori having greater representation in rural areas, R2 and R3, than in other areas.^14,24 Higher travel distance also makes personal mobility, availability of transportation, and occupational flexibility more relevant.

A review was done on models of patient enrolment.²⁵ It noted that when designing a patient enrolment system, one of the considerations that needs to be made is ‘making provision for alternate care and informational continuity if the nominated GP is unavailable or no appointment is available within a reasonable time’, with one of the weaknesses of a patient enrolment system being that it ‘does not meet patient needs when traveling, commuting, seeking second opinion’ and that ‘GP shortages limit competition between providers’. The analysis here confirms that such weaknesses are being experienced, and even accentuated, with the perception that the enrolment system has become gridlocked in places.⁷ It is necessary at this point to look for alternatives to ensure appropriate levels of care, given the prevalence and distribution of practices not accepting new patients, and this detrimental effect on access. This is particularly important in a country like NZ, where general practices act as gatekeepers to the rest of the healthcare system, eg for prescription medicines, referrals for diagnostic services, and referrals for specialist hospital care.

This study looked at a range of variables to see if they were associated with closed books. Consultation fees were found to be significant, with the relationship being non-linear. General practices with fees towards the middle of the range were more likely to have closed books. It could be that general practices with low consultation fees could be staying open to enrolment as they are highly dependent on capitation funding, but some practices could also see it as their mission to serve all who need PHC. Previous work has documented the financial difficulties of general practices with a concentration of high-needs patients, arising from insufficient compensation from capitation funding together with fees charged to patients being capped at a low level; this financial difficulty may lead these practices to make extra efforts to remain open to enrolment despite capacity limitations.^26–28

General practices with the highest fees could remain open to enrolment as the demand for their services is lower because of the cost to the patient. It appears that it is those practices with fees in the mid-range that can afford to close their books. A respondent in a survey in general practice said ‘From [a] business point of view there is a disincentive to close your book but at some point, patient care has (to) take priority over financial benefit’.⁷ That VLCA practices were more likely to have open books fits with the results that lower fees practices were more likely to have open books.

While we have considered a range of variables that could be associated with closed books, some of these could only be considered at the DHB level where there are only a small number of observations. It could be that there was not enough power to detect a real association between closed and open books and the variables we examined. However, scatter plots were examined between the variables and the percentage of closed books and there seemed to be no evidence of any association. While the Wairarapa DHB district was in the top three for general practise utilisation (suggesting high needs), population change (high growth) and access to PHC enrolment (high enrolment rates), the next two DHB districts with high percentages of closed books were both ranked at 9th (out of 20) or much lower on those variables. We found no significant association between closed books and access to PHC at the DHB level, whereas other studies have shown otherwise. For example, a longitudinal cohort study in Quebec, Canada found that attachment to a family physician improved access to and continuity of PHC.²⁹

This study adds to existing literature in showing the effects of closed books on restricting access to care (unequal distribution, travel distance, cost). These factors add barriers to the population’s ability to access basic health services when in need. NZ is currently in the middle of significant health system reforms, signalled in a Prime Minister and Cabinet white paper released in April 2021, with new agencies created on 1 July 2022.³⁰ Changes in funding and contracting began to take place in May 2023, and are likely to continue for some time to come.³¹ One of the drivers for the changes is the problem of the ‘postcode lottery’, where people receive different levels of health care depending on where they live. We hope these area-based findings, and the findings from our wider study on closed books,⁷ including recommendations for improvements (Mohan,^B unpublished observations) will contribute to the design of a more accessible and equitable PHC system.

Data availability

The data are publicly available (citations provided) or available on request from the MoH.

Conflicts of interest

The authors declare that they have no competing interests.

Declaration of funding

Funded by the Lottery Health Research Funding (Aotearoa New Zealand) LHR-2022-186638. The funders did not have any role in the design, analysis, interpretation of data or writing of the manuscript.

Author contributions

Conceptualisation and funding acquisition: MTIL, MP, MJ, JC; Methodology, Software, Data curation, Formal analysis: MP; Writing – original draft: MP, MTIL. Writing – Review and editing: MP, MTIL, NM, MJ, JC; Project Administration: MP, MTIL, NM. All authors have read and agreed to the final version of the manuscript.