Causes of Death Abroad: Analysis of Data on Bodies Returned for Cremation to Scotland

Authors

  • Christopher A. Redman PhD,

    Corresponding author
    1. Health Protection Scotland, Clifton House, Glasgow, UK
    2. Brownlee Travel Clinic, Gartnavel General Hospital, Glasgow, UK
      Christopher A. Redman, PhD, Health Protection Scotland, Clifton House, 3 Clifton Place, Glasgow G3 7LN, UK. E-mail: christopher.redman@nhs.net
    Search for more papers by this author
    • Now retired.

  • Alice MacLennan RGN,

    1. Health Protection Scotland, Clifton House, Glasgow, UK
    Search for more papers by this author
  • Eric Walker MB, DSc (Hon), FRCP

    1. Brownlee Travel Clinic, Gartnavel General Hospital, Glasgow, UK
    Search for more papers by this author
    • Now retired.


Christopher A. Redman, PhD, Health Protection Scotland, Clifton House, 3 Clifton Place, Glasgow G3 7LN, UK. E-mail: christopher.redman@nhs.net

Abstract

Background. The majority of travelers from the UK and Scotland visit Europe, particularly Spain and France, as well as North America, yet surveillance that aids in travel medicine guidance continues to focus on infectious diseases relating to developing countries. Here, we report on causes of death among all bodies returned to Scotland for cremation.

Methods. Data collected by the Scottish Government on bodies being returned from abroad for cremation was collated for the period 2000 to 2004, and analyzed to identify the cause and location of death among travelers as well as to test the hypothesis that for death due to failure of the circulatory system among Scots there was a significant association between age at death and whether death occurred in Scotland or abroad.

Results. Of the 572 deaths reported between 2000 and 2004, 73% occurred in the European region and 10% in the Americas. With respect to the cause, trauma accounted for 20.4%, infectious diseases 1.5%, and other non-infectious causes accounted for 75.5% of deaths. Among the latter, the major cause of death was due to failure of the circulatory system (77.0%). A significant association was observed between death abroad due to failure of the circulatory system and younger age at death for all (χ2 = 26.9, df = 3, p < 0.001) and for males (χ2 = 20.7, df = 3, p < 0.001) but not for females (χ2 = 2.7, df = 1, p = 0.099).

Conclusions. The data indicates a low rate of death among Scots traveling abroad, with trauma and other non-infectious causes being the most common cause of death; failure of the circulatory system was the most common cause of death in the latter group. Europe and the Americas were the most common locations of death. Although travel health services should continue to advise travelers to developing countries on infectious disease risks, it is also important that travel health acts as venue for providing key advice and preventative means to all travelers, including those to developed countries. Those agencies, organizations, and companies who deal with travelers along their journey should also engage with travel health experts and practitioners to reduce the risk of adverse outcomes, including death, to travelers.

In travel medicine, a great emphasis is correctly placed on risk reduction of diseases with high incidence among travelers to developing countries,1–3 such as diarrhea4 and respiratory conditions,5,6 as well as those diseases which have substantial incidence in host countries and therefore pose a risk in terms of mortality or serious morbidity, eg, rabies7 or malaria.8,9 This emphasis is a consequence of travel medicine, as a specialty, arising out of the interaction between primary care health professionals and the increasing numbers of travelers who were traveling abroad and who consequently were seeking advice both before and after travel.10 Initially focusing on business travelers going to poorer countries,11 many of the risks that travel medicine dealt and continues to deal with were specific for people traveling from developed nations, to poorer, often-culturally different, and developing ones.

The fact remains, however, that the majority of travelers from the UK do not visit developing countries. In 2007, of the 69.5 million visits abroad by UK residents, 79% were to Europe12 and 7% to North America.12 Of the total visits over one third (36%) were to Spain and France. The proportions were similar for visits abroad by residents in Scotland12 with 78 and 10% of visits being to Europe and North America and 39% of visits being to either Spain or France.

There are difficulties in estimating adverse events among travelers with surveillance of travel-related incidents usually focused on infectious diseases.3 There is often no indicator of the proportion of events which were fatal, although exceptions do exist.3

Here, we report on analysis of causes of death among those returned to Scotland for cremations and test the hypothesis that there is a relation between death abroad from circulatory disease and age at death.

Methods

Data

In Scotland, permission to cremate remains requires rigorous checks concerning the cause of death under the 1935 Cremation (Scotland) Regulations, including a medical certificate of the cause of death signed by a doctor, as well as two cremation certificates signed by two additional doctors. The regulations were designed to introduce safeguards as it was considered that investigations into cremated remains would not allow further investigations concerning possible criminal matters afforded by investigations of an exhumed buried body. The regulations apply to all cremations in Scotland whether the death has occurred in Scotland or outwith Scotland. Upon return of a body from abroad for cremation, the cause of death is confirmed at the country of death by staff at the Scottish Executive Health Department (SEHD; now known as the Scottish Government Health Directorates) before permission being given to cremate the remains. If the cause of death cannot be ascertained to the satisfaction of SEHD, then permission to cremate the remains is refused.

Data on all bodies returned including age and sex of deceased and cause and country of death were kept in handwritten form. This data was collated by Health Protection Scotland (HPS) in a Microsoft Access database. The cause of death was categorized by a Consultant Epidemiologist (EW) and Nurse (AM) as to whether the cause of death was due to traumatic, infectious, or other non-traumatic, non-infectious causes. Those other non-traumatic, non-infectious causes of death were then also matched to International Classification of Diseases (ICD)-10 codes and categorized accordingly: eg I00 to I99; diseases of the circulatory system constituted one category. Where there was more than one cause of death which could be mapped to an ICD-10 code, the underlying cause was used for categorization.

Incidence data on death by cause and age for the year 2002 (mid-point of 2000–2004) were obtained from the General Register Office for Scotland (GROS2002). Data on age distribution for UK travelers abroad in 2002 were obtained from published data from the International Passenger Survey13 (IPS2002).

Analysis

Analysis was carried out on the most recent 5-year period available being data pertaining to bodies returned between 2000 and 2004, inclusive. Descriptive statistics were calculated using Microsoft Excel and Minitab.

Analysis to test the hypothesis that there was a significant association between age at death from circulatory diseases and whether death occurred abroad or in Scotland was carried out in two ways. In method A, which allowed the association to be tested for males and females, the age distribution of death from circulatory diseases from GROS2002 was used to calculate the number of expected deaths (E) among the age groups from the cremation database. χ2 analysis was used to estimate whether there was an association between E and O (the observed number of deaths observed in the cremation database). For method B, the age distribution of death by age group from circulatory diseases from GROS2002 was applied to the population of UK travelers going abroad in 2002 (IPS2002) to calculate the numbers of expected deaths among UK travelers. This age distribution was then applied to the cremation data to estimate the numbers of expected deaths (E). A χ2-test was used to determine if there was a significant association between the age distribution E and O, the observed number of deaths.

Limitations

As outlined in the “Introduction” section there are always difficulties in estimating the range of causes of both morbidity and mortality among travelers abroad. Where the death of a British National occurs abroad, it (1) must be registered according to the law of that country and (2) should be reported to the British Consul who may be able to arrange for the death to be registered in the UK as well. With respect to the data for analysis there are severe limitations to allow analysis of UK citizens dying abroad. In the case of consular data, there is no obligation on relatives of the deceased to notify the consulate, the data itself is not centrally collated, and where it exists it depends on the information supplied by a relative of the deceased who may not be in a position to provide the cause of death. In the case of burials in Scotland, on return to Scotland the Registrar of Births, Deaths, and Marriages for the district where the funeral is to take place must be informed in order for burial to take place. However, no data are collected or retained on where the death occurred for further analysis. In the case of cremation in Scotland, it is only because additional permission of the SEHD is required for remains to be cremated that data on cause and location of death is collated.

While approximately 60% of deaths in Scotland result in cremation,14 there is no similar estimate of what proportion of deaths among Scottish travelers abroad result in cremation in Scotland as opposed to burial in Scotland or disposal abroad. For the above reasons, it is not possible to state how representative the sample used in this analysis is of the population of Scottish travelers dying.

Although cause, date, and location of death were available for the analysis, additional data on traveler type, time the deceased spent abroad before death, and data on risk factor/underlying conditions would have aided in discrimination of possible effectors on death. With respect to the cause of death bias may also have been introduced due to differences in recording the cause of death between different countries including Scotland or even inaccuracy in the cause of death communicated to the SEHD.

The data also did not allow the distinction to be made between Scots living abroad (eg, expatriates) and Scots traveling abroad (eg, on holiday). This may have introduced bias into any comparisons with the reference Scottish population, as factors related to long-term residence abroad may have affected the cause and age at death.

In addition, the lack of age-categorized denominator data for Scottish travelers necessitated the assumption that age distribution of UK travelers abroad was representative of Scottish travelers abroad to analyze the relationship between age at death due to circulatory disease and whether death occurred abroad or not.

Finally, there are significant limitations related to the comparability of traveling and non-traveling Scots, where, for example, the Scottish population will include those who for health reasons are unable to travel. In comparing across the age range 25 to 64, it was hoped to eliminate some of this bias associated with underlying conditions and ability to travel associated with older age.

Results

A total of 587 bodies were returned to Scotland for cremation between 2000 and 2004. Of these, 177 (30.2%) were females and 408 (69.5%) were males; 2 (0.3%) were not recorded for sex. The mean age at death was 57.8 years (range 0–93 years; median 61 years).

Causes of Death

The cause of death was recorded in 572 (97.4%) patients (Table 1). Of these, only 9 (1.5%) were due to infectious causes; one of these was due to cerebral malaria, one due to a viral hemorrhagic fever, and the remainder due to septic shock. Trauma accounted for 120 deaths (20.4%), while other non-infectious causes accounted for 443 (75.5%) deaths.

Table 1.  Causes of death by WHO region among those returned to Scotland for cremation in 2000 to 2004
ClassificationAfrica n (%)Europe n (%)MediterraneanAmericas n (%)South East Asia n (%)Western Pacific n (%)Cruise n (%)Grand total n (%)
Infectious2(22)5(56)2(22)0(0)0(0)0(0)0 (0)9(100)
Traumatic4(3)81(68)10(8)12(10)6(5)6(5)1 (1)120(100)
Other16(4)335(76)25(6)47(11)6(1)13(3)1 (0)443(100)
 Circulatory14(4)254(74)21(6)38(11)3(1)10(3)1 (0)341(100)
 Respiratory2(5)27(66)2(5)6(15)3(7)1(2)0 (0)41(100)
 Gastrointestinal0(0)18(90)1(5)1(5)0(0)0(0)0 (0)20(100)
 Neoplasm0(0)16(89)0(0)2(11)0(0)0(0)0 (0)18(100)
 Renal0(0)5(100)0(0)0(0)0(0)0(0)0 (0)5(100)
 Liver0(0)4(100)0(0)0(0)0(0)0(0)0 (0)4(100)
 Other0(0)11(79)1(7)0(0)0(0)2(14)0 (0)14(100)
Unknown2(13)10(67)2(13)1(7)0(0)0(0)0 (0)15(100)
Grand total24(4)431(73)39(7)60(10)12(2)19(3)2 (0)587(100)

The causes of many of the 120 traumatic deaths were often difficult to accurately ascertain. In most cases (N = 95, 79.2%) they were broadly described as accidental deaths. The remainder consisted of those who died by suicide (17, 14.2%) and conflict (3, 2.5%); the cause was unrecorded in 5 (4.2%).

Among those deaths which were neither caused by trauma nor infection (Table 2), the major cause of death was failure of the circulatory system (341, 77.0%) which contributed to 52.0% of all deaths. This was followed by failure of the respiratory (41, 9.3%) and gastrointestinal (20, 4.5%) systems with neoplasm accounting for 18 deaths (4.1%).

Table 2.  Numbers of deaths by system affected for non-infectious and non-traumatic causes
SystemTotal% Non-infectious% All
Circulatory34177.052.0
Respiratory419.36.3
Gastrointestinal204.53.0
Neoplasm184.12.7
Renal51.10.8
Liver40.90.6
Other143.22.1
Grand total443100.067.5

Geography of Death

In all categories, the majority of deaths occurred in European (EU) destinations (Table 1). For traumatic deaths, Europe contributed to 68% (81) of deaths followed by the Americas (12, 10%), and the Mediterranean region (10, 8%). Similarly, of the 341 deaths due to failure of the circulatory system, 74% (254) occurred in Europe, followed by the Americas (38, 11%), and the Mediterranean region (21, 6%).

The five countries where most deaths occurred were all EU: being Spain (195, 33%), France (34, 6%), Greece (28, 5%), Portugal (28, 5%), and Netherlands (25, 4%). The most common non-EU countries where deaths occurred were the Americas (21, 5%), United Arab Emirates (15, 3%), Canada (13, 2%), Australia (9, 2%), and Iraq (7, 1%).

Age at Death From Diseases of the Circulatory System and Travel

Comparison of the age distribution of death from failure of the circulatory system between the deaths abroad (Figure 1A and B) and the Scottish population (Figure 1C and D) suggested that a higher proportion of deaths were occurring in lower age groups among those who died abroad. It was decided to test for any association between age at death and location of death (abroad/not abroad) across the age range 25 to 64.

Figure 1.

Age distribution among those dying from circulatory disease. (A) Males returned to Scotland for cremation in 2000 to 2004. (B) Females returned to Scotland for cremation in 2000 to 2004. (C) Males dying from circulatory disease among the Scottish population (GROS2002). (D) Females dying from circulatory disease among the Scottish population (GROS2002).

Using Method A, a significant association was found between death abroad and age at death for all (χ2 = 26.9, df = 3, p < 0.001) and for males (χ2 = 20.7, df = 3, p < 0.001), but not for females (χ2 = 2.7, df = 1, p = 0.099); numbers of females were too low for analysis across four age groups. For Method B, which sought to estimate an expected age distribution of death among travelers by using data from the International Passenger Survey (IPS2002), a significant association was found between death abroad and age at death for all (χ2 = 21.3, df = 3, p < 0.001).

Discussion

There is a great deal of literature in travel medicine on deaths among travelers relating to travel to remote areas,15 deaths during the journey,16,17 and deaths due to specific causes, eg, infectious diseases,18 accidents,19–21,22 cardiovascular disease,19,20 and envenomation.23 This analysis was carried out to estimate the causes of death among travelers from Scotland abroad and to test whether travel altered the risk of dying from circulatory disease among Scots abroad. The data highlighted the low proportion of infection-related deaths and the high proportion of deaths due to failures in the circulatory system and to accidents.

For the 5-year period 2000 to 2004, there were 572 reports on the cause of death compared to 952 deaths reported in a similar study published in 199124 for the 15-year period 1973 to 1988. This observed increase in average number of cremations among travelers per year (114.4 per year in this study compared with 63.5 previously24) may reflect either increased numbers of deaths abroad as observed elsewhere22 and/or an increase in preference for cremation observed in the UK population.14 If the former then this may merely reflect the increase in travel observed among the UK population.12 That being said the UK Office of National Statistics estimated 8.3 million visits abroad by Scots, of which 40% were to EU countries. Assuming that bodies returned for cremation represented 57.4%,14 then this suggests a low death rate in the order of 12 deaths per 100,000 visits.

A large proportion of deaths (20%) were caused by trauma, of which the majority was accidental. Accidental deaths among travelers have been observed to be increasing in US citizens and it has been argued that pretravel advice tends to focus on infectious disease risk as opposed to risks that cause injury.22 Personal preparedness and planning is important in increasing safety and decreasing the risk of accidents among travelers who due to unfamiliarity with local conditions or changed personal behavior are at increased risk of death due to drowning21,25 and car accidents22,26,27; children may be particularly vulnerable.25

In terms of Scottish travelers, it is interesting to note the high proportion of deaths due to circulatory causes (52%), although the proportion is less in this study than that observed by Paixao and colleagues24 at 69%. In that study it was proposed that, among the elderly, deaths abroad may have occurred in their home country had they not traveled. However, our observation that for death due to failure of the circulatory system among those aged 25 to 64, the age at death among those whose bodies were returned for cremation was younger compared to that of the reference Scottish population, raises the possibility that this difference is linked to travel abroad. A number of factors related to travel abroad may detrimentally affect those with preexisting circulatory conditions including warm climate,28 the journey,29 and lifestyle changes,30 such as increased exertion or changes in diet and/or environmental factors.31 The relationship between age at death from cardiovascular disease has been observed among US citizens abroad,32 where 49% of deaths were due to this cause, with the highest proportion of deaths occurring in Western Europe. Cardiovascular death rates among US citizens abroad were found to be higher than among those at home aged 35 to 44.

Considering that many of the travelers died in Southern Europe where the incidence of cardiovascular mortality is much lower than that of Scotland,33 it would be interesting to study at which stage of the journey deaths due to failure in the circulatory system occur. Couch29 noted in an analysis of sudden death due to coronary arteriosclerosis that incidence among visitors was four times that of the local population and suggested that stress due to changing time zones or travel may have contributed. In another study of ischemic heart disease among residents of New York City,34 it was observed that increased deaths due to ischemic heart disease were observed among visitors to that city, while residents away from the city were observed to have lower numbers than expected. This effect was again tentatively linked to stress associated with living in New York for both residents and visitors alike.

In summary, while travel consultations in the UK generally focus on those traveling to high-risk destinations for infectious diseases, the largest proportion of travelers are to Europe. Here, we observe that the largest numbers of deaths among Scots travelers occurred in Europe and, to a lesser degree, the Americas, in the main due to natural causes. As to the observation concerning age at death from circulatory system failure and travel abroad, additional research is required on which, if any, aspects of travel exacerbate existing conditions.29 Considering the relatively low death rate, prospective studies would be resource intensive and require large numbers to produce statistically meaningful data. Nonetheless, a body of evidence exists which highlights natural causes, such as coronary heart disease,19,24,32 and injury22,24–26,32 as major causes of death among travelers. Certainly, travel health services should move beyond advising travelers to developing countries on infectious disease risks, to becoming venues for providing key advice and preventative means to all travelers, including those to developed countries. In addition, those agencies, organizations, and companies who deal with travelers along their journey should also engage with travel health experts and practitioners to reduce the risk of adverse outcomes, including death, to travelers.

Acknowledgment

We acknowledge the advice and assistance of Prof. Chris Robertson of the University of Strathclyde with respect to the analysis of circulatory disease deaths with respect to age.

Declaration of Interests

The authors state they have no conflicts of interest to declare.

Ancillary