Seguridad y Medio Ambiente FUNDACIÓN MAPFRE

Year 29 Nº 115 third quarter 2009

Accidents in the natural environmentSAFETY

This article presents the first known study of injuries caused by free-time activities in the natural environment and an estimation of the cost of taking the casualties to the health centres where they were treated. The injuries dealt with are only those that did not require mountain-rescue call-out and the area studied is the Aragón Pyrenees. This research work, which uses a trailblazing cost estimation methodology, could provide valuable insights for establishing good practices and improving prevention measures.

By : P. VELA1, G. BERNUÉS2, E. ANDRÉS*, A. CASTILLO2, C. EZQUERRA2, M. A. NERIN3, J. BADA4, y J. R. MORANDEIRA5 Primary Healthcare and Health Promotion Research Network. Aragón Health Sciences Institute (Instituto Aragonés de Ciencias de la Salud).

(1) Doctor in Medicine and Surgery. Specialist in Family and Community Medicine. Associate professor of Zaragoza Medical School. Head of Studies of the Family and Community Medicine Teaching Unit of the sectors Zaragoza III and Calatayud.
(2) Specialist in Family and Community Medicine.
(3) Doctor and assistant professor. Director of the master in Mountain Emergency Medicine, Zaragoza University.

(4) Graduate in Physical Education, associate professor. Sport and Health Sciences School. Zaragoza University.
(5) Tenured professor of Experimental Surgery. Coordinator of the University Courses of Specialisation in Emergency Mountain Medicine. Zaragoza University.
(*) Graduate in Statistical Techniques and Sciences, Hospital General de la Defensa. Zaragoza.

After a trawl of the bibliographical sources we found no epidemiological studies in Spain into the number and circumstances of healthcare services provided for wilderness accidents without mountain-rescue call-out. The only trustworthy statistics officially available are those kept on casualties that have been rescued by the mountain Guardia Civil in collaboration with the medical services of 061 Montaña, which has been dealing with mountain-rescue medical services since 1997. Two studies on this matter have been published, by Avellanas1 and Nerín2.

Even less is known about casualties that do not involve rescue call-out but use up local healthcare resources for examining and treating them.

Public healthcare in Aragón is broken down into geographical territories called areas. These areas are in turn divided into sectors and the sectors into health zones (figures 1 and 2).

Figure 1. Healthcare map of the Barbastro sector.

Figure 1.

Figure 2. Healthcare map of the Huesca sector.

Figure 2.

This study deals with area I, taking in the sectors of Barbastro and Huesca, and, therein, the health zones included in the Aragón Pyrenees, running from Echo in the west, bordering on Navarra, to Castejón de Sos in the east, bordering on the province of Lleida.

The Aragón Pyrenees records about half of the accidents of this type.

The aim of this study is to identify the risks of free-time activities in the natural environment throughout the Aragón Pyrenees, taking in the districts of Jacetania, Alto Gállego, Sobrarbe and Ribagorza, during the summer of 2008. The figures have been culled from questionnaires filled out by all people who received care services in the health centres operating in the abovementioned districts. Another aim is to calculate the costs generated, doing so by using the non-hospital-based methodology of ACGs (Adjusted Clinical Groups, formerly known as Ambulatory Care Groups or ACGs), taking into account the costs of tests and therapy and any conveyance costs by land or air.

The Aragón Pyrenees records about half the accidents of this type

Material and Methods

Descriptive design of prospective character.

The case definition includes individuals who were injured while taking part in a recreational activity in the study zone, from 1 May to 31 October 2008. This study defines recreational activity in the natural environment as any activity carried out for personal enjoyment and not involving a motor vehicle.

The cases were identified from a questionnaire filled out by the personnel attending patients in the health centre: doctor or nurse. If the healthcare personnel declined to cooperate, the figures were obtained retrospectively from patient healthcare reports filled out by the doctors on call.

All reports of the period under study were checked in the health centres of Castejón de Sos and Jaca, including their respective points of continuing care services in Benasque and Canfranc, respectively. From these the researcher chose those tallying with the «case» definition.

To crosscheck the figures obtained during the study period and ensure their quality, collaborators were also asked retrospectively for the emergency treatment sheets for the same period of 2007.

Table 1. Provincial Distribution of Protected Nature Sites *. Aragón and Provinces. 2006.
Aragón Huesca Teruel Zaragoza
No. Area No. Area No. Area No. Area
Total** 33 141.977 25 125.546 4 4.385 4 12.046
National Park 1 15.608,00 1 15.608,00 0 0,00 0 0,00
Nature Park 3 117.641,00 2 107.793,00 0 0,00 1 9.848,00
Nature Reserve 4 2.853,30 0 0,00 1 655,00 3 2.198,30
Natural Monument 24 3.072,48 22 2.698,00 2 374,48 0 0,00
Protected landscape 1 3.355,34 0 0,00 1 3.355,34 0 0,00

Unit of Area: hectares. Publication: © Instituto Aragonés de Estadística (IAEST), February 2007. Source: Instituto Aragonés de Estadística (IAEST), February 2007.
(*) The Protected Nature Sites considered are those declared as such under the Natural Spaces and Wild Flora and Fauna Conservation Act 4/89 (Ley de Conservación de los Espacios Naturales y de la Flora y Fauna Silvestres), and under the Aragón Protected Nature Sites Act 6/98 (Ley de Espacios Naturales Protegidos de Aragón).
(**) The Total Protected Area figure is not the sum of the rows since 553 hectares of Natural Monuments of the Pyrenean glaciers are included in the area of the Nature Park of Posets-Maladeta.

Table 2. Mountain Rescue 2008. Monthly Activity Summary.
No. of persons attended by 061 Montaña No. of primary services dealt with by 061 Montaña No. of secondary services dealt with by 061 Montaña No. of services cancelled
TOTAL 323 287 0 10
January 14 9 0 0
February 17 15 0 2
March 17 15 0 1
April 13 15 0 2
May 26 19 0 2
June 39 33 0 1
July 92 85 0 1
August 59 57 0 0
September 12 14 0 1
October 9 11 0 0
November 15 5 0 0
December 10 9 0 0

Cases involving people injured while not involved in a recreational activity were excluded, i.e., work-related accidents except for workers employed on open-air activities (monitors, guides, forest wardens).

In our working definition the cut-off age for adults and children is 18.

The figures collected include age, sex, date and time of the injury, where it happened, activity being carried out at the time, injury or illness type and occurrence, diagnosis and means of transport.

Cost estimation was done by means of the Adjusted Clinical Groups (ACG) patient distribution system

The specified hours of natural light run from 8:00 a.m. to 8:00 p.m. Multi trauma is defined as potentially fatal injuries in various parts of an individual’s body. The occurrence was defined as exacerbation when the origin of the call-out was clearly related to a pre-existing pathology (for example a heart attack in a person with a history of heart illness). The qualitative variables were described by means of distribution tables and bar graphs or sectors and were compared by means of the chi-square test and Fisher’s exact 2-tailed test for expected values of less than 5. If the chi-square test was significant, a standardised residuals analysis was carried out to pinpoint which category of variables the association occurred in. Quantitative variables were assayed by means of the parametric t-student test or ANOVA test and the non-parametric U-Mann-Whitney test and Kruskal-Wallis H test for comparing two or more groups of patients, as need be. The decision to use parametric or non-parametric tests was taken according to the normality or otherwise of the distribution of quantitative variables, cross checking this last hypothesis by means of the Kolmogorov-Smirnov test. A test p-value equal to or lower than 0.05 was considered to be statistically significant. Microsoft’s Access and Excel programmes and SPSS 15.0 for Windows were used for the statistical analysis.

Cost estimation was done by means of the Adjusted Clinical Groups (ACG) patient distribution system, a method developed by Johns Hopkins University. This is the population based case-mix system most widely used worldwide for the financing and management of patients in primary healthcare. Each ACG classifies people into single, mutually exclusive morbidity categories based on disease patterns and on the expected use of resources. The ACGs are obtained exclusively with diagnosis coding data, recorded in the medical reports or in the electronic medical records, plus the age and sex of the patients3.

An estimation was also made of the costs by means of Diagnosis Related Groups (DRGs). These constitute a hospitalisation classification system in which patients are grouped in clusters of similar clinical characteristics who would therefore be expected to use a similar amount of healthcare resources. The aim in doing so is to compare results even though the estimation is not used for studying non-hospital costs4.


Data Quality.

A month-by-month comparison was made of the total of patients attended in the various health centres. This was done by obtaining the patient variation rate in 2008 as compared to 2007.

An analysis of the number of cases dealt with, broken down by sex (table 3), shows a p-value associated with the test for equality of variance of 0.994, indicating that there was no significant difference between men and women attended in the health centres in both years.

Table 3. Distribution of the individuals attended for injuries suffered while participating in recreational activities
2007 2008
Frequency Percentage Frequency Percentage
Men 789 0,54 613 0,58
Women 670 0,46 438 0,42

As regards the age at the moment of the accident, there was no significant difference. In 2007 the average age was 31.74 (SD: 18.80) and in 2008 it was 32.42 (SD: 42.86). The p-value of the test to check the difference between both averages was 0.999.

In conclusion, the data quality is good and ipso facto the results obtained therefrom.

Of the total casualties dealt with from 1 May to 31 October 2008, 58.3% were male and 41.7% female. The patient’s average age was 32.42 (CI 95%: 29.77-35.06), being slightly higher in males at 43.28 (31.19-55.36) as against 33.29 (31.88-34.71). Nonetheless, the difference was not significant (p-value of the non-parametric U-Mann-Whitney test: 0.988).

Hiking was the commonest activity being carried out at the moment of the accident, accounting for nearly 62% of cases

The commonest accident time-band was each side of midday, from 10:00 to 14:59. The accident frequency breakdown by time-bands is shown in graph 1. In general the time-band running from 10 to 20 hours pooled 83.96% of the accidents recorded.

Graph 1. Accident frequency breakdown by timebands. (%)

Graph 1. Accident frequency breakdown by timebands. (%)

There were no statistically significant differences in the accident time-band breakdown between men and women (p-value associated with the chi-square test 0.381), nor by age brackets (p-value 0.173). The accident breakdown by time-bands was also similar in all months studied (p-value 0.850).

The most frequent activity at the moment of the accident was hiking, accounting for almost 62% of the accidents (graph 2).

Graph 2. Activity being carried out at the moment of the accident. (%)

Graph 2. Activity being carried out at the moment of the accident. (%)

The gender difference of the activities (p-value of the chi-square test < 0.001) was most pronounced in the activity of mountain biking, where the standardised residuals show that men have far more accidents in this sport than women. In the other activities there were no significant differences.

As for the causes of the accident, table 4 shows the frequency and percentage of the various causes analysed.

There are gender differences in the accident causes (p-value associated with the chi-square test = 0.034). An analysis of standardised residuals shows that only women were involved in the collisions between people. There was also a significant difference in falls to the same level, where the residual in the case of the woman is significant, showing that, in our study at least, women are more prone to falls (graph 3).

Graph 3. Causes of the Accidents. (%)

Graph 3. Causes of the Accidents. (%)

Table 4. Frequencies of the Accident Causes
Code Accident Causes Frequency Percentage
1 Common illness 9 1,20
2 Landslide 0 0,00
3 Fall to a different level 70 9,32
4 Fall to the same level 289 38,48
5 Collision between people 3 0,40
6 Falling rocks 12 1,60
7 Activity-related illness 32 4,26
8 Improper equipment 8 1,07
9 Going astray 1 0,13
10 Anchor failure 1 0,13
11 Impact from objects 54 7,19
12 Lack of skill for the activity 43 5,73
13 Weather 10 1,33
14 Other 219 29,16

Most of the patients visiting the health centre had sprains, bruises or grazes and wounds, accounting for 73.58 % of the total casualties. Table 5, analysing the p-value of the chi-square test to check for gender differences in any of the consequences analysed, shows that all are higher than 0.05 except for grazes and wounds, where the p-value is 0.001. In other words, there are statistically significant differences only in this group, where an analysis of the standardised residuals shows that women suffered far fewer grazes and wounds as a result of the accident than men.

Table 5. ADG Distributio
ADG Frequency Percentage
Acute: minor 117 10,98
Acute: minor-primary infections 4 0,38
Acute: major 7 0,66
Allergies 26 2,44
Likely to recur: discontinuous-primary diseases 4 0,38
Chronic illness dealt with by primary healthcare: stable 2 0,19
Chronic illness dealt with by primary healthcare: unstable 1 0,09
Chronic illness dealt with by primary healthcare: unstable-ophthalmology 1 0,09
Dermatology 1 0,09
Traumas/adverse effects: minor 385 36,12
Traumas/adverse effects: major 468 43,90
Psychosocial: acute. minor 1 0,09
Psychosocial: Recurrent or persistent, unstable 1 0,09
Signs/symptoms: minor 12 1,13
Signs/symptoms: uncertain 1 0,09
Signs/symptoms: major 15 1,41
Discretional group 19 1,78
Dentistry 15 0,09

As regards the time spent carrying out the activity before the accident occurred, most of the individuals had been engaged for 1 to 4 hours (44.09%). Nonetheless, mention should also be made of the 5% who had been carrying out the activity for more than 1 day (trekking). A gender breakdown shows no significant differences in the time spent on the activity before the accident. The p-value of the chi-square test to analyse the relation between qualitative variables was 0.584. (graph 4).

Graph 4. Breakdown of the time spent on the activity before the accident. (%)

Graph 4. Breakdown of the time spent on the activity before the accident. (%)

As for age, 73.03% of the casualties were adults. There is, however, a high percentage of children attended in our clinics for various reasons. If we break down the age at which the accident occurred by gender we find a p-value associated with the T-Student test of 0.099. The fact that this is higher than the significance level of 5% tells us that there are no gender-based differences in the age of the casualty.

Most of the patients visiting health centres had sprains, bruises or grazes and wounds, accounting for 73.58% of the total casualties

If we divide the age into ten-year brackets we find that the most frequent age bracket is 0-10, representing 20.13%. Nonetheless, the youngest, up to 20, account for 52.2% of the total patients dealt with (graph 5).

Graph 5. Age breakdown of casualties. (%)

Graph 5. Age breakdown of casualties. (%)

If we analyse the possible existence of a relation with the gender variable, we find a 0.518 p-value associated with the chi-square test, showing that age is evenly distributed by gender. Neither were there any differences in the age breakdown according to the hours spent on the activity before the accident (0.178 p-value associated with chi-square test). There is, however, a relation with the month-of-accident variable (p-value of the chi-square test < 0.001). A study of the standardised residuals here shows that youngsters aged 10 to 20 are those who have the highest number of accidents in July and August.

The months in which most accidents were recorded were July and August. These two months together accounted for 59.83% of the visits. An analysis by gender shows that the number of accidents is similar for men and women; the p-value of the chi-square test was 0.204 (graph 6).

Graph 6. Accident breakdown by months. (%)

Graph 6. Accident breakdown by months. (%)

The accident incidence rate in 2008 was 3.30%, and in 2007 it was 5.22% (No. of cases/No. of visitors). If we calculate the incidence rate per person we find a rate of 593% for 2008 and 816% for 2007 (No. of cases/No. of exposure days).

Economic Analysi

To make an estimation of the healthcare cost of the accidents occurring during recreational activities in the study period we drew from the national benchmark figures of the Diagnosis Related Groups (DRGs) obtained from the Ministry of Health and Consumer Affairs and also the weight and cost of the national healthcare service for each one of the DRGs5.

Two intermediate indicators are needed for calculating the cost of attending patients injured by recreational activities: firstly, the case-mix index of attending the casualty is calculated by means of a weighted average of the DRGs, obtaining a value of 0.66767; secondly the Case Complexity Classes (CCCs) are calculated by multiplying the total of DRGs (or patients attended) by their weighted average; in our case, 1068 x 0.66767 = 713.068032.

The total cost according to the cost report of the national health system for the DRGs corresponding to the patients attended in our consulting rooms was €90,092.71 a year. Considering that the study period was six months, this is tantamount to €45,046.36 in the period running from May to October. If we divide this cost by the CCC, we obtain an average cost per DRG of €63.17; in our case this would give a total cost of €67,468.33. This is the outcome if we establish the cost by means of hospital case-mix groups.

The Adjusted Clinical Groups, (formerly Ambulatory Care Groups or ACGs) are a diagnosis grouping system that classifies people by their morbidity burden over a period of time (generally six months or one year).

On the basis of this classification system the patient population diagnoses are pooled into fairly homogenous groups in terms of resource use. It was therefore used at the start to explain use patterns. The ACG classification is based on age, sex and reasons for visiting the health centre or diagnosis codes according to the International Classification of Diseases (ICD-10-CM) of each patient.

We used a four-stage process to obtain the ACGs of an ICD. The first stage groups the ICD-10-CM diagnosis codes into 34 ambulatory diagnostic groups (ADGs) (a patient may have one or more ADGs); the second into 12 collapsed ambulatory diagnostic groups (CADG); the third into 25 major ambulatory categories (MAC) and the fourth into one single case-mix group. (table 5).

Table 6 shows the ACG grouping of the diagnoses. The average costs per visit were obtained from the work of Sicras 6, throwing up a primary healthcare cost of €34.71. This is tantamount to a total cost of €48,884.41.

Table 6. AGC Diagnosis Grouping
ACG Frequency Percentage Cost per visit (€) Total cost per ACG (€)
Total 1.068 100,00 34,71 48.884,41
300 130 12,17 5,84 759,30
600 26 2,43 30,28 787,15
1800 4 0,37 26,41 105,65
1000 2 0,19 33,52 67,05
1100 1 0,09 27,38 27,38
2100 1 0,09 26,67 26,67
1800 854 79,96 52,82 45111,13
2300 1 0,09 58,84 58,84
2700 14 1,31 53,84 753,70
3200 15 1,40 30,22 453,37
2800 19 1,78 36,02 684,29
3600 1 0,09 49,89 49,89

The result shows a lower cost than the APDRG method. This is due basically to the fact that the ACGs group the ICD10s in homogenous resource-consumption groups in relation to primary healthcare. In our case this undervalues the cost, since we have to deal with many traumas and fractures that call for specialised medical resources, such as X-ray diagnosis or stitches.

The figures obtained from the Civil Protection and Security Service of the Interior Directorate General of the Government of Aragón give an hourly helicopter cost of €46007,8. A helicopter covers 20 kilometres in about 5 minutes; adding on another 5 minutes for taking off and landing, this gives a cost of €766.66 for 20 kilometres. The cost of the flight from Zaragoza, where the helicopter is based, to the accident site also has to be factored in. In the case in hand the helicopter flew from Benasque to Boltaña, since the rescue was carried out by the Guardia Civil helicopter, whose costs we have not been able to determine. We also know that the ambulance cost amounts to €20 per kilometre when it is simply a case of transporting patients and €30 per kilometre for a mobile ICU9.

Two sorts of patient-transporting arrangements were taken into account to ascertain the total number of trips made: firstly that the patient was taken from the accident site to the health centre making the first check and, secondly, that the patient was taken from the health centre to the hospital (some cases involved a combination of both and this was factored in for calculating the total ambulance journey).

The breakdown of patient-transporting arrangements is shown in Table 7 and in Graph 7.

Graph  7. Patient-transporting means used

Graph 7. Patient-transporting means used

Table 7. Breakdown of patient-transporting means
Journeys Frequency Percentage
Ambulance 35 92,11
By helicopter 1 2,63
ICU 2 5,26

Google Maps TM were used to calculate the distance in kilometres from the accident site to the health centre. In the five cases with no record of the means of transport used from the accident site, the missing data was averaged in. When transport was laid on from the health centre to the hospital, the kilometres were obtained in the same way as that outlined above. When there was no mention of the hospital in question, the distance to the nearest hospital was taken.

Once the distance in kilometres had been obtained for each patient, the transport costs were calculated. The average distance in kilometres from the accident site to the health centre was 20.39 kilometres (CI 95%: 14.13-26.65), and from the health centre to hospital it was 66.79 kilometres (CI 95%: 58.29-75.28).

The transport cost to the health centre and to the hospital is shown in table 8, where the final column gives the total cost. The average transport cost to the health centre from the accident site was €374.22 (€414.31) (standard deviation: €276.84 (€414.22) and from the health centre to hospital was €1036 (standard deviation: €748.34); this gives a total transport cost, regardless of the means of transport used, of €1410.86 (standard deviation: €828.46). The total cost amounted to €55,023.67.

Table 8. Casualty Transporting Costs
Cost to health centre Cost to hospital (€) Total cost (€)
Total 14.594,67 40.429,00 55.023,67
Average 374,22 1.036,64 1.410,86

A calculation was also made of the healthcare costs of emergency treatment in the hospital the casualty was sent on to10. We hence found that each X-ray had an average cost of about €38.08 and that a fracture had an additional emergency-service cost of €106.78. There were hence 24 patients taken to hospital to find out the severity of the injury by X-ray, amounting to a total healthcare cost of €913.92 in X-ray plates. Likewise, 24 patients taken to hospital were found to have a fracture, increasing the initial cost by €2562.72.

The transport cost plus emergency hospital treatment amounted to €60,063.54.


Despite the considerable number of people who like to undertake open-air activities, there are few studies dealing specifically with the injuries occurring during activities in the natural environment and the epidemiology thereof.

The reference bibliography checked in databases (EMBASE and PUBMED) showed no studies like the one in hand now. All referred to emergency attention in rural areas11 or mountain-rescue call-outs to deal with mountain-climbing or skiing incidents12,13,14.

US, Canadian and Australian studies refer to interventions by the rescue services and forest wardens to deal with visitors to national parks. In these studies11,14,15, the methodological limitations are similar and they are based on a partial and insufficient collection of data from the healthcare forms and are also constrained by the difficulty of establishing the population at risk, to be able to calculate the incident rate. In our case, moreover, reference is made only to healthcare services provided in the summer months.

The results in terms of the age of the participants, sex differences, number of hours spent on the activity before the accident happened, etc. do not differ from one study to another. The differences they show all stem from the different data sources.

Neither have any reference works been found on the cost of dealing with these injuries by means of the case-mix methodology on diagnosis related groups and, more specifically, for ambulatory attention by the Adjusted Clinical Groups (ACGs). The only similar studies carried out in Spain refer to attention given in some health centres16.

Awareness raising campaigns like «Montañas Seguras MIDE» could help to reduce the false sense of security that «nothing will happen to me»

The results probably understate the cost of the healthcare services on the assumption that not all episodes were culled and recorded. Nonetheless, this research opens up the possibility of using this methodology for calculating and forecasting the cost of dealing with injuries of this type and reflecting the importance and social and economic benefit of prevention campaigns.

In this study 58.3% were male and 41.7% were women. These injury rates do not imply a greater participation by men in open-air activities17,18. Many studies on recreational activities in the natural environment bear out these results19,20; once again, however, they are not based on the real number of participants. These results, therefore, do not represent injury risk based on exposure to given activities.

The study of Gentile et al21. uses person-days (number of casualties/study period days) to measure exposure. In our case the calculation gives high figures because the study period was very short (180 days). The gender differences detected in this research no not support a claim that gender is a greater risk factor when participating in open-air recreational activities. It is nonetheless worthwhile pointing out in our study that the age bracket with the highest number of casualties is the 0-10 group, probably due to the high number of people dealt with who were taking part in youth camps during the study period.

The average age of patients was 32.42 (CI 95%: 29.77-35.06), being slightly higher in males at 43.28 (31.19-55.36) as against 33.29 (31.88-34.71). Nonetheless, the difference was not significant (p-value of the non-parametric U-Mann-Whitney test: 0.988).

The average age of the casualties differs in the various studies on recreational activity injuries, according to the activity in question or the study site20, 21, 22. These results show that young males aged 10 to 35 are most prone to suffer injuries when taking part in recreational activities; this probably reflects the fact that this age bracket is most given to participate in activities of this sort.

Most of the patients attended in health centres had sprains, bruises or grazes and wounds, accounting for 73.58% of the total persons with some type of injury. Injuries of this type, frequent but undramatic, tend to pass under the radar of the press and the public at large. This may lull them into a false sense of security and underestimation of the risk involved23. Awareness-raising campaigns like «Montañas Seguras MIDE»24, could help to lessen this false sense of security that «nothing will happen to me», leading people to overestimate their own strength and underestimate the risk level25.

The results of this study serve as a starting point for new research into the epidemiology of open-air or nature-related injuries. They bear out many of the assumptions on the cause of these injuries, stressing their potential severity and long term consequences.

It is crucial to follow the three key principles for preventing wilderness injuries: planning, preparation and forecasting of the possible problems to bring down the number and severity of the injuries 26.

Primary prevention is based on basic precautions. The people involved need to be properly trained for activities that call for an extra effort in a hostile environment; they need to be sure of their own limits and skills and keep within them and they also need to use the right equipment in good conditions of maintenance.

Education and social pressure from key groups or persons could encourage participants to abide by these good practices

Secondary prevention involves taking along a well-provisioned first aid kit suited to the activity involved and an ad hoc means of communication such as a mobile phone or radio phone. Families and security forces should be informed of the activity to be carried out, to prevent a non-life-threatening injury from worsening into pathologies of greater severity27.

In general there is no regulation covering open-air recreational activities, so education and social pressure from key groups or persons could play a vital role in encouraging participants to abide by good practices28, the aim of which is simply to reduce the frequency and gravity of any accidents that may crop up and lay on the necessary aid resources when they do occur.29

Future research work should include studies to assess the total number of participants per activity. Common denominators would enable the risks of the various activities to be compared30, 31.


This study is the first known one of its type to come up with estimates of injuries occurring during activities in the natural environment, in potentially hostile wilderness areas but not requiring the call-out of mountain rescue services.

It could provide valuable insights for establishing good practices.

It is also a trailblazer in the use of the case-mix methodology for estimating derived costs, to ascertain potential savings that may stem from systematic prevention campaigns.

Despite its drawbacks, the culling of information on emergency healthcare in health centres is an important monitoring measure to this end.

Using IT resources for data collection purposes will improve recording quality. This information will help in carrying out prevention projects. It could also be used for drawing up training programmes not only for healthcare personnel but also for forest wardens, civil protection staff, open air activity monitors, all of whom may have to deal with such injuries in rural areas. This will improve the efficiency of prevention activities.


The study has been approved by Aragón’s Clinical Research Ethical Committee and has been conducted on a 2007 research grant from FUNDACIÓN MAPFRE. There is no conflict of interest in relation hereto


Mª Antonia Nerín Rotger y Juan de Dios Bada Jaime, Castejón de Sos. Clara Cortés Martín y Mª Pilar Gistau Torres, Aínsa. Guillermo Bernués Sanz, Lafortunada. Jesús Sánchez Sanz, Jaca. José Mª Borrel Martínez, Ayerbe. Gisela Jordán Lanaspa, Berdún. Charo Casado Ortiz, Hecho. Azucena Soria, Broto. Ana Delia Castillo and Cristina Ezquerra, Biescas. Isabel Cuenca Peña. Tarazona.


RESCUE. Recovery and evacuation of casualties. DRG. . Diagnosis Related Group (usually shortened in medical and costs-related works to DRG) is a system for classifying hospital cases into about 500 groups thought likely to have a similar hospital resource use. Cases are classified in terms of the international codes of the International Classification of Diseases (ICD-10), the type of surgical procedure carried out, the age, sex and the presence of complications or comorbidities. The purpose of this classification is to group diseases for assigning a monetary value to each one and thereby improve the management of hospital costs. CASE MIX. The case mix complexity term has been used to refer to an interrelated but distinct set of patient attributes, including the severity of the illness, its prognosis, difficulty of treatment, need of medical action and resource use intensity. Each one of these attributes has a very precise meaning describing a particular aspect of a hospital’s case mix.
ADJUSTED CLINICAL GROUPS© (ACG©).. These are the case mix system of ambulatory activity, designed to radically improve knowledge of ambulatory healthcare, just as DRGs did for hospital healthcare 10 years ago. ACGs© are the property of the Johns Hopkins University, which created them and has run many pilot schemes in several countries, such as Canada, Sweden and Spain, with similar health systems. On the basis of this ongoing experience ACG©s have become today’s most promising instrument for running ambulatory healthcare, especially primary healthcare and for capitation-based financing.


  1. Avellanas Chavala, M.L., «Los accidentes de montaña en España: Análisis de la situación actual, sobre un estudio epidemiológico de los últimos 25 Años (1969-1993)» [doctoral thesis]. Zaragoza: Universidad de Zaragoza; 1995.
  2. Nerín M, Estado actual de la prevención de los accidentes de montaña en Aragón. Cultura Ciencia y Deporte. 2005;(1):75- 86.
  3. ACG - Adjusted Clinical Groups (Agrupador). Available at: acg-adjusted-clinical-groupsagrupador. Visited on 20 March 2009.
  4. Proyecto de estimación de pesos y costes de los procesos de hospitalización en el Sistema Nacional de Salud – reseña metodológica estudio 2006. Disponible en: estadisticas/docs/Notas_metodologicas_ GRD_2006.pdf. Visited on 7 January 2009.
  5. Op. Cit. (4).
  6. Sicras-Mainar A; Serrat-Tarres J;Navarro-Artieda R; et al. Posibilidades de los grupos clínicos ajustados (Adjusted Clinical Groups- ACG´s) en el ajuste de riesgos de pago capitativo. Rev. Esp. Salud Publica vol.2006. (80) 1: 55-65.
  7. Comunicación personal. Servicio de Seguridad y Protección Civil de la Dirección General de Interior del Gobierno de Aragón. 24 March 2009.
  8. Pliego de condiciones técnicas para la contratación de un «servicio de transporte sanitario y de emergencias mediante helicóptero», disponible en http://www.boa.aragon. es/cgi-bin/BoaAA/BRSCGI? CMD=VEROBJ&MLKOB= 223448953944. Visitado el 17 de marzo de 2009.
  9. Comunicación personal. 061 Aragón. Servicio Aragonés de Salud. 23 March 2009.
  10. Comunicación personal. Dirección de Administración y Servicios Generales. Hospital de Barbastro (Huesca). Aragonés de Salud. 24 March 2009.
  11. Leemon D, Schimelpfenig T. Wilderness injury, illness, and evacuation: National Outdoor Leadership School's incident profiles, 1999-2002. Wilderness Environ. Med. 2003;14(3):174-82.
  12. Chavala M.L.A. The mountineering accidents in Spain. ARCH. MED. DEPORTE 1997 14:60 (321 - 325) Olivera, J., «Las actividades físicas de aventura en la naturaleza: análisis sociocultural». Apunts Educ Fis Esports. 1995; 41: 5-8.
  13. Finlay J. Wild, MBChB. Epidemiology of Mountain Search and Rescue Operations in Banff, Yoho, and Kootenay National Parks, 2003–06. Wilderness and Environmental Medicine. 2008. 19, 245 251.
  14. Ela GK. Epidemiology of wilderness search and rescue in New Hampshire, 1999-2001. Wilderness Environ. Med. 2004 15(1):11-7.
  15. Flores AH, Haileyesus T, Greenspan AI. National estimates of outdoor recreational injuries treated in emergency departments, United States, 2004-2005. Wilderness Environ. Med. 2008;19(2):91-8.
  16. Op. cit. (6)
  17. Outdoor Industry Foundation. Active Outdoor Recreation Participation Study, 2005. Full Report. Accessible at: http://www.outdoorfoundation. org/research.participat ion.2005.html. Consultado el 15 de marzo de 2009.
  18. Bowker JM,Murphy HK,Cordell DBK, et al. Wilderness and primitive area recreation, participation and consumption; an examination of demographic and spatial factors. J Agric Appl Econ Assoc. 2006;38: 317-326.
  19. Gretchen K. Ela, MD. Epidemiology of Wilderness Search and Rescue in New Hampshire, 1999–2001. Wilderness and Environmental Medicine., 2004; 15:11-17.
  20. Bradford D Stephens; Douglas S Diekema; Eileen J Klein. Recreational Injuries in Washington State National Parks. Wilderness & Environmental Medicine. 2005; 16: 192-197.
  21. Gentile DA, Morris JA, Schimelpfening T, et al. Wilderness injuries and illness. Ann Emerg Med. 1992; 21: 853-861.
  22. Montalvo R, Wingard DL, Bracker M, et al. Morbidity and mortality in the wilderness. Western Journal of Medicine. 1998 ;168(4): 248-254.
  23. Powell C. The Perception of Risk and Risk Taking Behavior: Implications for Incident Prevention Strategies. Wilderness & Environmental Medicine. 2007 Apr 1;18(1): 10-5.
  24. Montañas Seguras. Federación Aragonesa de Montañismo. Accessible at: http://www.euromide. info/index.php?option=com_co ntent&view=article&id=44&Itemid= 55. Visited on 15 March 2009.
  25. Op. Cit. (24)
  26. Auerbach PS. Ed. Wilderness Medicine. 4th ed. St Louis MO; Mosby Inc; 2001.
  27. Sholl JM. Curcio EP 3rd. An introduction to wilderness medicine. Emergency Medicine Clinics of North America. 2004; 22(2):265-79.
  28. US Department of Agricultural Forest Service. National Survey on Recreational and Environment. American’s participation in outdoor recreation: results fron NSRE. Accessible at: /Nsre/nsre2.html . Consultado el 15 de marzo de 2009.
  29. Hargarten S. Injury control research and wilderness medicine: a babe dangling in de woods. Wilderness Environ. Med. 1999; 10:2.
  30. Bauer R, Körmer K, Sector M. Scope and patterns of tourist injuries in the European Union. Taylor & Francis. 2005.
  31. Bilan des interventions effectuées par les services de secours en montagne au cours des mois de juin - juillet - août et septembre 2006. Systeme National d’Observation de la Securite En Montagne. Accessible at: http://www.securiteconso. org/IMG/pdf/Fiche_accident_ montagne_Ete_2006.pdf Visited on 10 February 2009.