Article Text
PDF
Abstract
Introduction The increasing demands made on emergency ambulance services contribute to inefficient, clinically inappropriate health care, and may delay the provision of emergency care to life-threatening cases. The hypothesis of this study was that the activity for the first year of operation of an emergency telephone consultation service contributed to a reduction in ambulance use in non-urgent cases and a decrease in the cost associated with despatching ambulances.
Methods The numbers of ambulance use and the emergency hospitalisation of ambulance cases were compared before and after the introduction of the Tokyo Emergency Telephone Consultation Centre (the #7119 centre). Public awareness of the #7119 centre in each region of Tokyo and the cost related to despatching ambulances were also investigated.
Results A total of 26 138 consultations was performed in the initial year. Compared with the previous year, the number of ambulance uses per 1 million people decreased (before 46 846, after 44689, p<0.0001). The emergency hospitalisation rate (EHR) of ambulance cases increased significantly because of the decreased proportion of non-urgent cases (before 36.5%, after 37.8%, p<0.0001). There was a statistical correlation between the awareness rate in each region and the change of after-hours EHR in adults (R=0.333, p=0.025). The total cost related to despatching ambulances was reduced by approximately ¥678 000 000 (£4 520 000) in the initial year.
Conclusion To date, the emergency telephone consultation service has contributed to the appropriate use of ambulances and a reduction of its cost in Tokyo.
- Despatch
- emergency ambulance systems
- prehospital care
- telephone consultation
- triage
Statistics from Altmetric.com
An increase in unnecessary emergency ambulance transfers contributes to inefficient, clinically inappropriate health care, and may delay the provision of emergency care to life-threatening cases.1 2 Telephone consultation services have been introduced in several countries and play an important role in the management of ambulance use and after-hours care. NHS Direct has been established in the UK since 1997 and was a cornerstone in the development of telephone consultation services.3 4 Telephone consultation was defined as to ‘assess and manage cases by giving advice or by referral to more appropriate services via telephone’.5 Similar systems have subsequently been established in Australia,6 Denmark,7 New Zealand,8 Sweden,9 Canada10 and the USA.11 The consultation also has basic concepts in common with the medical regulation system in the Services d'Aide Medicale Urgente (SAMU) of France and in other countries or regions.12 13
Similar problems have been experienced in Tokyo. Since the Tokyo Fire Department (Tokyo FD) introduced a campaign to dissuade citizens from calling for an ambulance in 2004, the number of ambulance uses decreased slightly; however, the decrease was less than expected. The Tokyo Metropolitan Government established a telephone consultation centre, which can cover all calls in Tokyo, on 1 June 2007, in order to reduce the use of ambulances.
The hypothesis of this study was that the introduction of the centre contributed to a reduction in ambulance use in non-urgent cases and a decrease in the cost associated with despatching ambulances.
Emergency medical services in Tokyo
Tokyo is the capital of Japan and the population was 12 898 939 in October 2008. The nationwide free emergency call number to despatch an ambulance is 119, which is managed by the Tokyo FD at the 119 centre. The main prehospital provider of ambulances is the emergency medical technician (EMT), who belongs to the Tokyo FD. They are classified as basic EMT, intermediate EMT and paramedics. According to the protocol established by the Tokyo Medical Control Council, the EMT and paramedics perform field triage and transport patients to an emergency hospital for definitive treatment.
Emergency hospitals are divided into three categories according to the proportion of the level of medical staff, skills and equipment, which are primary emergency clinics, secondary emergency hospitals and tertiary full-time emergency centres.
Excluding some recipients of public assistance, the entire Japanese population and foreigners who have residential status in Japan for 1 year or more can join the health insurance scheme. This scheme covers up to 70–80% of the medical costs of outpatient consultations and hospitalisation.
Overview of the Tokyo metropolitan emergency telephone consultation centre
Since 1 June 2007, the Tokyo Metropolitan Government has introduced the Emergency Telephone Consultation Centre (the #7119 centre), which is a 24-h and 7-day a week nurse-run telephone advice line, which aims either to refer callers to the most appropriate services, or to provide them with advice on how to care for their condition. The Tokyo centre was the first to be established as part of a nationwide project in Japan.
A steering council, which consists of the Tokyo Medical Association, the Bureau of Health and Welfare of the Tokyo Metropolitan Government and the Tokyo FD direct the #7119 centre. The #7119 centre is located next to the 119 centre in the Tokyo FD. It involves call handlers and nurses who are employed by the Tokyo FD, and physicians who are registered as emergency telephone consultation doctors of the Tokyo Medical Association. Telephone consultations for citizens in Tokyo are responded to by an emergency telephone consultation team, which includes one physician, three to four nurses and four to eight call handlers in two shifts of 12 h. The computer-programmed medical protocols were developed by a working group on protocols of the Emergency Telephone Consultation Centre of the Committee of Emergency Medicine in the Tokyo Medical Association. There are 98 protocols including 18 for paediatric cases. This is similar to other systems, such as NHS Direct in the UK. The priority determination for each consultation was provided and was classified into four categories based on perceived severity, as follows: red, immediate action by sending ambulance by 119; orange, immediately seek help within approximately 1 h; yellow, urgently seek help within approximately 6 h; and green, non-urgent case to seek help within the next 24 h.
Methods
Study design
This is a pre and post-interventional study with the intervention being the establishment of a call centre and the dependent variable being the #7119 centre of ambulance use. The hospital admission percentage is a nested cohort within each primary cohort. We retrospectively reviewed the previous system data of the Bureau of Health and Welfare of the Tokyo Metropolitan Government and the Tokyo FD. The investigated period for before the introduction of the #7119 centre was from 1 June 2004 to 31 May 2007, and that for after was from 1 June 2007 to 31 May 2008 (A). The period before the introduction was divided into three annual groups ((B1): 1 year before (June 2006 to May 2007); (B2): 2 years before (June 2005 to May 2006) and (B3): 3 years before (June 2004 to May 2005)). In addition, to show the long trend of annual ambulance use and emergency hospitalisation, we collected data from 1976 to 2008, and its duration of each annual data was from January to December, respectively.
First, we reviewed activity data for the initial year of the #7119 centre. The data included the number of calls, average number of calls per hour, characteristics of callers or patients (address, age and gender), the number of despatched ambulances, outcomes by the #7119 centre and the number of emergency hospitalisation cases transported by ambulance.
Second, we compared the following annual rates between the periods (A) and (B1), (B1) and (B2) and (B2) and (B3): ambulance use per 1 million people (excluding secondary transport cases), after-hours ambulance use per 1 million people, emergency hospitalisation rate (EHR) of ambulance cases (excluding secondary transport cases) and after-hours EHR. We also determined the annual change in the rates for each factor. These were calculated as following; (100×(data of next year minus data of current year)/(data of current year)) (%). Furthermore, subgroup analyses were performed for adults and paediatric cases (classification for paediatric patients; ie, those aged less than 15 years).
Third, the public awareness of the #7119 centre was evaluated using a two-stage stratified random sample postal survey, which was performed twice (July 2007 and July 2008) after the introduction of the #7119 centre. The rate of awareness was defined as the number of citizens within each region of Tokyo who stated that they were aware of the #7119 centre divided by the population of each region. We then examined the relationship between awareness of the #7119 centre in each region and the factors described above. Subgroup analysis was also performed for adult and paediatric cases.
Finally, the cost-effectiveness of this project was evaluated. The cost associated with the establishment of the centre during the initial year including personnel expenses and depreciation was calculated. Then, the difference of the expenditure of ambulance use between period (B1) and period (A) was counted. The expenditure on despatching one ambulance was estimated as approximately ¥45 000 (£300, if £1=¥150) based on the previous report of the Tokyo Metropolitan Government in 2002.
Statistical analysis
Chi-square tests were used to evaluate the change in ambulance utilisation before and after the introduction of the #7119 centre. Spearman's rank correlation analyses were performed to assess the correlations between the ratio of the #7119 awareness among the local residents, the public awareness rate and emergency hospitalisation of ambulance cases. These analyses were performed using SPSS, version 16.0.
To clarify the characteristics for each region, geographical information system (GIS) analysis was applied. The GIS is used to integrate, store, edit, analyse and display data that refer to location. The GIS application allows users to analyse spatial information, edit data, edit maps and present the results of all these operation. In our study, the ‘map overlay’ function was selected as a tool of MapInfo Professional, version 9.5 (Pitney Bowes Inc, New York, New York, USA). The function can classify two indices on a map at the same time. We chose the average awareness rate and the after-hours EHR in adults as the two indices.
Results
Activity data for the first year operation of the #7119 centre
The total number of calls to the #7119 centre was 268 094 (732.5/day). For approximately 90% (238 388) of all calls, the callers requested information regarding available hospitals or clinics. For these callers, the call handlers gave information about the hospitals or clinics and obtained an address and telephone number; a telephone consultation with a nurse or physician was not necessary for these calls.
The total number of telephone consultations was 26 138 (71.4/day), and were more frequent over weekends or during national holidays. The number of consultations increased during the after-hours period of hospitals or clinics, and the daily peak was at approximately 20:00 h. The percentage of patients who were male, female, or not recorded was 52.8%, 46.1% and 1.0%, respectively. The mean (SD) age of the patients was 28±27 years (0–103; n=26 012).
The percentage of calls by patients themselves, family members, others, unknown, and not recorded was 31.4%, 62.6%, 5.5%, 0.2% and 0.2%, respectively.
In terms of the outcomes of telephone consultations, the proportion of cases who were transferred to 119 and who were sent an ambulance was 12.8%. The proportion of cases who were provided information on emergency hospitals/clinics and who were advised to seek an urgent appointment at the hospitals/clinics or who were advised to refer to the callers' family doctors was 66.9%. For 1.9% of calls, the case was connected with other community services, for 0.5%, the call was terminated by the caller and 18.3% of calls had other outcomes. Emergency hospitalisation was recorded for 30.8% (1000 cases) of 3252 who were transferred to a hospital by an ambulance (figure 1).
Outcomes of calls to the #7119 centre.
The long trend of annual ambulance use per 1 million people excluding secondary transport and EHR of ambulance cases
As shown in figure 2, the annual ambulance use per 1 million people increased year by year and there was a peak in the number in 2005. On the other hand, the EHR decreased year by year with a plateau during recent approximately 10 years.
Long trend of annual ambulance use per 1 million people excluding secondary transport and emergency hospitalisation rate (EHR) of ambulance cases during from 1976 to 2008. Triangle dots represent the annual numbers of ambulance use per 1 million people excluding secondary transport (×1000), and round shaped dots represent those of EHR of ambulance cases (%).
Comparison of ambulance use and after-hours ambulance use before and after the introduction of the #7119 centre
Based on the data from the Tokyo FD, the number of ambulances used over the 1-year period after the introduction of the #7119 centre decreased by 22 215 ((B1) vs (A): 592 814 vs 570 599) and the number of ambulances used per 1 million was statistically reduced compared with that of the previous year ((B1)) ((B1) vs (A): 46 846 vs 44 689, p<0.0001). Although the number of ambulances used decreased significantly each year during the study period, the annual change in the rate of ambulance decrease from (B1) to (A) was greater than that for the other years. In addition, the after-hours ambulance use per 1 million people was also significantly reduced during (A) compared with (B1) ((B1) vs (A): 31 965 vs 30 370, p<0.0001) (table 1, figure 3).
Comparison between before and after introduction of the #7119 centre
Annual ambulance use, change in use and emergency hospitalisation rate. The white coloured column shows the values related to ‘the after introduction of the #7119 centre’, and the grey column shows ‘the before’. (A) Ambulance use per 1 million people in each period. (B) Annual change in ambulance use per 1 million people in each period (%). (C) Emergency hospitalisation rate in each period (%).
EHR and after-hours EHR of ambulance cases
The EHR decreased annually before the introduction of the #7119 centre. However, the rate after its introduction was statistically higher than that of (B1) ((B1) vs (A): 36.5% vs 37.8%, p<0.0001). The rates in both paediatric and adult cases were also higher after the introduction than that of (B1) (paediatric cases: 10.4% vs 11.4%, p<0.0001; adult cases: 38.7% vs 40.0%, p<0.0001).
The after-hours EHR of ambulance cases for all cases and for adults was also higher after the introduction of the #7119 centre (A) than those of (B1) (all cases: 29.4% vs 29.9%, p<0.0001; adult cases: 28.5% vs 30.7%, p<0.0001). There was no difference in after-hours EHR for paediatric case between the two periods (B1; 15.3% vs A; 14.9%, p=0.2528) (table 1, figure 3).
Public awareness of the #7119 centre
Responses were obtained from 53.4% (1603/3000) and 50.8% (1524/3000) of the samples mailed in July 2007 and July 2008, respectively. There was no significant relationship between the public awareness rate in each area and the difference in the rate of ambulance use or after-hours ambulance use after the introduction of the #7119 centre. There was also no statistical difference between the awareness rate in each area and the difference of EHR, after-hours EHR or after-hours EHR in paediatric cases. With the increase of awareness of the #7119 centre in each area, the difference of after-hours EHR in adults slightly increased after its introduction (R=0.333, p=0.025) (table 2). The colour coding scheme by GIS analysis demonstrated that there were few areas of deep red and deep green (figure 4).
Relationship between public awareness of the #7119 centre and the change in ambulance use or EHR
Relationship between the public awareness of the #7119 centre and differences (data of ‘A’ minus ‘B1’) in the after-hours emergency hospitalisation rate in adults in each region of Tokyo (geographical information system analysis). (A): After the introduction of the #7119 centre. (B1): 1 Year before the introduction. Colour coding scheme: the red zones show areas with an increase in the emergency hospitalisation rate (EHR) and a low public awareness rate, the green zones show areas with a small difference in EHR and a high public awareness rate, and the yellow zones show areas with good correlation between differences in EHR and the public awareness rate.
Cost-effectiveness of the #7119 centre
The cost associated with the establishment of the centre during the initial year including personnel expenses and depreciation was approximately ¥321 380 000 (£2 142 533, £1=¥150). The estimated decrease of cost associated with ambulance despatching was ¥999 675 000 (£6 664 500) calculated by a decrease in the number of despatched ambulances (22 215) and the expenditure on despatching one ambulance (¥45 000). Overall cost-effectiveness was estimated to a level of ¥678 295 000 (£4 521 967).
Discussion
Although the ambulance usage rate was slightly lower, 22 215, which was the actual value of decrease of ambulance use had a great impact on the cost-effectiveness. The resource corresponding to the decrease in the cost can be allocated in the other activities of emergency medical services. From a viewpoint of a human resource, the decrease in ambulance use can also be considered beneficial not only for personnel directly involved in ambulances, but also for healthcare professionals, especially those involved at emergency hospitals.
The #7119 centre had an impact on changes in the severity of symptoms of ambulance users because the rate of after-hours EHR by ambulance after the introduction was slightly but statistically higher than that before. It seemed likely that the activity of the #7119 centre contributed to the change in the behaviour of the #7119 callers and decreased unnecessary ambulance transport. Previous reports similarly described that after-hours house calls by general practitioners14 15 and telephone consultations to ED16 had been shown to decrease because of the change in the behaviour of those who consult the telephone consultation service. However, it was not possible to compare these results indiscriminately because the emergency care system was different in each country. In our study, the difference of after-hours EHR in adults after the introduction of the #7119 centre was greater in regions with greater awareness of the #7119 centre. Although it was not a strong correlation and the awareness rate was still low, the activities of the #7119 centre including the campaign effect influenced the decrease of unnecessary ambulance transports.
Elaborating on the impact from a user's perspective, the #7119 centre was useful for citizens who did not fully comprehend the severity or emergency of the sudden injury or illness, because the centre was able to offer several choices, including an ambulance request according to the patient's condition, and this point is consistent with previous reports. Moreover, the #7119 centre was beneficial for mothers with children. Because 30% of the calls forwarded by the #7119 centre to 119 were immediately hospitalised, the #7119 centre offers an alternative for people who may be hesitant about calling 119. It is important to be able to screen serious patients from those who hesitate to use an ambulance without fully comprehending their severity.
The first limitation of this study is the definition of the appropriate use of an ambulance. Cases requiring emergency treatment without hospitalisation should also be considered as appropriate cases requiring ambulance transport. We do not currently have a method to identify patients who need emergency treatment without hospitalisation. The second is the bias of morbidity during the study period. To our knowledge, although there were no epidemic episodes that may have affected morbidity before or after the introduction of the #7119 centre, this was not evaluated in detail. The third limitation is a problem concerning the population. The population was defined as the number of people for which the residents' cards were reported to the office of each city, town and village. Based on this definition, there may be many emergency patients but a small population in the downtown region of Tokyo, which is mainly composed of office space, whereas suburbs and commuter towns may have relatively few patients but a large population. The other limitation is that secondary transport cases were included in the analysis of the after-hours ambulance utilisation and emergency hospitalisation. In general, after-hours secondary transport cases include patients who were transferred from a primary or secondary emergency hospital to a tertiary emergency hospital to provide more appropriate care based on the severity of the patient's condition, and who needed emergency hospitalisation. The number of these transport cases could strongly affect the distribution of each category of emergency hospital.
Finally, the Ministry of Home Affairs showed a decreasing tendency of ambulance use on a nationwide scale in a report in September 2008, which was derived from data of a slightly different period to that in our investigation. A nationwide ambulance-use control campaign was also considered to have had an effect on the decrease.
Conclusion
In its first year of operation, the emergency telephone consultation service through the #7119 centre contributed to the appropriate use of ambulances in Tokyo. The activities of the centre produced a statistically significant drop in ambulance use and a decrease in unnecessary ambulance transport. These results brought a reduction in its cost of approximately ¥678 000 000 (£4 520 000). We can allocate the resource corresponding to the decrease in the cost in other emergency medical service activities instead of the inappropriate use of ambulances. Further research, especially analysis of the burden of emergency departments including walk-in patients, is necessary to determine the clinical impact of this service.
References
Footnotes
Members and investigators who participated in the Steering Council of Tokyo Emergency Telephone Consultation Centre are listed in the appendix.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.