Travel time and distances to Norwegian out-of-hours casualty clinics

Material and method

Results

Table 1 provides an overview of the municipalities that were included and excluded. Of the 101 municipalities associated with OOH services in alternating locations, 23 had a mutual inter-municipal collaboration, 43 provided centralised night-time OOH services and 28 had other arrangements. Seven municipalities had multiple internal locations, including Oslo and Bergen with their city-district casualty clinics. The population in the excluded municipalities (OOH services by boat) accounted for 0.7 % of the total population in Norway as of 1 January 2011. Altogether 73 % of the municipalities, with 70 % of the total population, had permanent casualty clinics.

The distribution of the maximum average travel distances is shown in Figure 1. A maximum travel distance of 5 – 10 kilometres was the most usual (75 municipalities). The shortest estimated travel distance was 0.4 kilometres, while the longest amounted to 141 kilometres. The distribution of maximum average travel times is shown in Figure 2. The most frequent average travel time was 5 – 10 minutes (81 municipalities). The shortest travel time was one minute, the longest 131 minutes.

Figure 1  Histogram showing the distribution of maximum average travel distances for the municipalities included. Intervals of 5 km (N = 417)

Figure 2  Histogram showing the distribution of maximum average travel times for the municipalities included. Intervals of 5 minutes (N = 417)

Table 2 shows the median maximum travel times and distances with the 10th and 90th percentiles for all municipalities included. The municipalities belonging to Western Norway Regional Health Authority had the shortest maximum travel times and distances, while those in Central Norway Regional Health Authority had the longest travel times and distances to the OOH casualty clinic. The 10th percentile was lowest in municipalities under Northern Norway Regional Health Authority, for maximum time as well as distance, while the travel time for the 90th percentile was highest in those under the Northern and Western Regional Health Authorities.

Table 3 shows the median average distances and travel times to OOH services for municipalities that have a single permanent casualty clinic. The median average travel times and distances for municipalities with a permanent casualty clinic were shortest in the areas encompassed by Northern Norway and Central Norway Regional Health Authorities. In municipalities with a permanent casualty clinic (n = 316) the median average travel times and distances were lower than the maximum average travel times and distances in all municipalities seen as a whole (N = 417).

Table 4 shows that 238 (57 %) municipalities had more than 15 kilometres maximum average travel distance. These municipalities accounted for 21 % of Norway’s total population. Ninety-seven municipalities had more than 40 minutes of maximum travel time, and 40 municipalities had more than 60 minutes of maximum travel time. These municipalities accounted for 5 % and 2 % of the national population respectively. The largest proportion of the population that had more than one hour of maximum travel time to a OOH casualty clinic was found in areas under Northern Norway Regional Health Authority, whereas South-Eastern Norway Regional Health Authority had the lowest proportion. Table 5 shows that 33 085 inhabitants, equivalent to less than 1 % of the inhabitants in municipalities with a single permanent casualty clinic, had more than one hour of travel time to the nearest casualty clinic.

Table 6 shows travel times and distances in municipalities with varying types of alternating locations for their OOH services. The median maximum average travel times and distances were highest in municipalities providing centralised OOH services in the night-time and lowest in municipalities that have multiple, alternating locations for their OOH services internally.

Eight municipalities had population figures and travel times indicating that secondary on-call arrangements ought to be considered, according to the action plan (e-tab 7, e-tab 8). Two of these fulfilled both criteria for secondary on-call arrangements. One of these eight municipalities had a single permanent casualty clinic. Five municipalities had more than 40 minutes of travel time and more than 8 000 inhabitants, and five had more than 60 minutes of travel time and more than 4 000 inhabitants.

The population of the municipalities that were covered by the proposed recommendations for secondary on-call arrangements amounted to 70 422 persons, equivalent to 1.4 % of the national population. Eighty-nine municipalities with a total of 191 025 inhabitants had a maximum travel time of more than 40 minutes, but too few inhabitants to be affected by the action plan’s recommendations for secondary on-call arrangements. Of these, 50 municipalities with a total of 102 009 inhabitants (2.1 % of the national population) had a single permanent casualty clinic.

Discussion

This study shows that the median maximum travel time to OOH services in Norwegian municipalities in 2011 was 22 minutes, and the median maximum travel distance was 19 kilometres. Twenty-three per cent of the municipalities, with five per cent of the national population, had more than 40 minutes of median maximum average travel time, and 10 % of the municipalities, with 2 % of the national population, had more than 60 minutes of median maximum average travel time. Among the municipalities that participate in OOH arrangements with a single permanent casualty clinic, the median average travel time was 16 minutes and the median average travel distance was 13 kilometres. Altogether 47 % of these municipalities, with 17 % of the inhabitants, had travel distances exceeding 15 kilometres, i.e. travel distances that according to previous studies (1, 2) are associated with a reduced use of OOH services.

Among the municipalities that have a permanent casualty clinic it is interesting to note that the median travel times and distances are considerably lower in Northern Norway than in other parts of the country. Most likely, this reflects the fact that the population in the North is more typically concentrated around the centre of the municipalities, where the casualty clinic also tends to be located.

One strength of this study is that is has been possible to estimate maximum average travel times and distances for as many as 417 out of 430 municipalities. These figures are especially interesting in terms of preparedness and illuminate the need for extra measures in addition to standard OOH services, even though for some municipalities this will be relevant only during parts of the day. For municipalities in which the provision of OOH services is spread over several municipalities, the longest travel distances will most often be relevant only during those times of the day when the activity is at its lowest, since local emergency services are provided within short travel distances at times when demand is at its highest.

Inhabitants in municipalities with a single casualty clinic face the same travel distances at all times and will hence be able to serve as a better basis for prognoses of the rate of use of OOH services in the population seen in relation to distance. In municipalities that provide OOH services in alternating locations across several municipalities, the inhabitants will in most cases need to travel only a short distance to reach these services, even though the maximum distance (for example during the night) may be long. The consultation rate may therefore be far higher than the maximum average travel time would indicate. Figures from municipalities with a single permanent casualty clinic are somewhat less representative, since they comprise only 73 % of the municipalities with 70 % of the total national population.

A weakness of the study is that the National Emergency Primary Health Care Registry has proven to contain a number of incorrect data, despite the fact that the information has been provided by the local emergency services themselves. Although the information has been checked in cases of doubt, we cannot exclude the possibility that incorrect information on the organisation of OOH services may have been included for some municipalities.

The method that was used to estimate distances has proven robust (2), but since the distances apply to the municipal level, it is possible that analyses using postcodes as units could have yielded a more representative description of actual average travel times. We chose not to undertake analyses at the postcode level, however, since no statistics on the use of OOH services are available for each postcode. It is also possible to investigate distances to OOH services from each individual address point, but we did not have the resources to retrieve the necessary information from Statistics Norway.

The larger the units into which we divide the material, the more dominant the central areas will be. Some municipalities have only a single postcode from which to estimate distances. This may cause the travel distance to be underestimated, since the post office and the casualty clinic are often located near each other.

The distribution of travel times corresponds to travel distances, with some right-side displacement of the curve. This is partly because some municipalities rely on ferries, meaning that the journey takes longer than the travel distance in kilometres would indicate, and the fact that average driving speeds are lower over short distances. We consider the chosen limit of 15 kilometres to be relevant, but it is essentially arbitrary. A limit of 12.5 kilometres would most likely have been more correct, since this is the shortest distance at which we observed a statistically significant reduction of the contact rate when looking at the confidence intervals of the regression analysis, as described in a previous article (1).

No equivalent investigations of the distance to OOH services have been undertaken in Norway. Internationally there are few examples of OOH arrangements that correspond to the Norwegian one, and we are not aware of any comparable studies from other countries. The South Denmark Regional Authority has defined a service goal saying that the distance to the nearest place for a consultation should be less than 30 kilometres, although a little longer in certain areas and during the night (8). In Norway, 137 municipalities with a total of 400 000 inhabitants would fail to satisfy a requirement for a maximum average travel distance of 30 kilometres.

Swedish authorities have investigated distances to the healthcare centres («vårdcentraler»), which to a certain extent are comparable to the Norwegian daytime casualty clinics. The study showed that 6 739 persons, equivalent to 0.073 % of the population, had more than 40 minutes of travel time to the nearest centre in 2012. The proportion was largest in Norrbotten county, where this applied to 0.9 %. Eighty-one per cent of the population had less than five minutes of extra driving time to the second closest healthcare centre (9).

Fortunately, few people need to travel a very long distance (> 60 minutes of travel time) to an OOH casualty clinic in Norway, but this nevertheless applies to more people than is indicated by the tables in this study. At the periphery of municipalities that cover a large geographical area there are numerous inhabitants who need to travel for more than 40 or 60 minutes to reach emergency medical services, even though the estimated average for the municipality in question is far lower.

In future reforms of the municipal structure and the organisation of emergency medical services – with increasingly large units – there is an increasing risk of overlooking the inhabitants at the periphery, since they will contribute less to the estimated average, even though the absolute number of inhabitants remains unchanged. Care should therefore be taken in making decisions that apply to the inhabitants on the periphery based on aggregated data for the entire municipality or OOH district.

These findings show that if the population criteria for secondary on-call arrangements described in the proposal from the National Centre for Emergency Primary Health Care for an action plan at the municipal level (4) are applied, very few inhabitants will be included. When applied at the municipal level, the recommendations appear not very suitable or appropriate.

Nearly 200 000 persons live in municipalities in which the maximum travel times are so long that they would be covered by the distance criteria in the proposed recommendations for secondary on-call arrangements, but where the population is too small to be affected by the population criteria. Since we also know that many inhabitants have long travel distances, even though the average travel distance in the municipality as a whole is shorter than the stipulated limits, we must assume that several hundred thousands of inhabitants face travel times of more than 40 minutes. These will risk being left without acceptable primary emergency services should the municipalities choose to adhere strictly to the proposed recommendations for secondary on-call arrangements in the action plan.

One weakness of this study is that we have assessed how large a proportion of the population is covered by the recommendations on the basis of the municipal structure in 2011, and not on the basis of real population concentrations independently of the municipal boundaries, as suggested by the recommendations in the proposed action plan (4). For example, two or more neighbouring municipalities that are located far away from the casualty clinic may have a total population that exceeds the limits in the recommendations, and it is conceivable that population concentrations within a single municipality could be affected by the recommendations without these being applicable to the municipality as a whole.

We nevertheless claim that the results presented here may indicate that the proposed recommendations apply to an insufficient proportion of the population, and that it could be appropriate to remove or adjust the limits in the population criteria. The National Centre for Emergency Primary Health Care has submitted a proposal to the governmental committee for emergency medicine for a service requirement saying that at least 90 % of the inhabitants in an OOH district should have less than 40 minutes of travel time to the nearest casualty clinic (10).

A new municipal structure consisting of fewer but much larger municipalities will present further challenges if municipalities are used as the unit of analysis, and even more so if a number of such enlarged municipalities cooperate in forming inter-municipal OOH districts. A sustainable plan for what an acceptable travel time should be and the recommended threshold for when secondary on-call arrangements should be organised should also in the future be based on actual population concentrations and distances to the permanent casualty clinic, independently of present and future municipal boundaries. For example, secondary on-call arrangements could be organised through contracts with GPs’ offices in the peripheral areas of OOH districts.

Conclusion

In more than half of all municipalities, the inhabitants face travel distances that are so long that the activity of the OOH services might be lower than expected. Moreover, many municipalities have travel distances that are so long that secondary on-call arrangements should be considered, even where the population is small. The proposed recommendations for on-call arrangements should be adjusted to ensure provision of adequate primary medical services to as many people as possible.