BMJ No 7123 Volume 315
Paper Saturday 20/27 December Christmas 1997 issue
Death rates of characters in soap operas on British television: is a government health warning required?
Tim Crayford, Richard Hooper, Sarah EvansAbstract
Objective: To measure mortality among characters in British soap operas on television.
Design: Cohort analysis of deaths in EastEnders and Coronation Street, supplemented by an analysis of deaths in Brookside and Emmerdale.
Main outcome measures: Standardised mortality ratios and the proportional mortality ratio for deaths attributable to external causes (E code of ICD-9 (international classification of diseases, ninth revision).
Results: Staying alive in a television soap opera is not easy. Standardised mortality ratios for characters were among the highest for any occupation yet described (771 (95% confidence interval 415 to1127) for characters in EastEnders), and this was not just because all causes of death were overrepresented. Deaths in soap operas were almost three times more likely to be from violent causes than would be expected from a character's age and sex. A character in EastEnders was twice as likely as a similar character in Coronation Street to die during an episode.
Conclusions: The most dangerous job in the United Kingdom is not, as expected, bomb disposal expert, steeplejack, or Formula One racing driver but having a role in one of the United Kingdom's most well known soap operas. This is the first quantitative estimate of the size of the pinch of salt which should be taken when watching soap operas.
Introduction
Death is a fact of life in soap opera. It is only natural that producers should exaggerate the danger of real life to make series interesting, but by how much do they do this? What sorts of occupations would be as dangerous as being a character in a soap opera? Could Brookside Close be the most dangerous street in Britain? We set out to answer these questions in a hard hitting analysis of mortality in British television soap operas.
British soap opera has been a mirror to many contemporary social themes over the past 15 years, dealing with topics ranging from HIV and breast cancer to theft, mugging, and murder. This fact and the huge audiences that each programme draws each week has helped make soap opera one of the many ways by which people now normalise their own lives. In this respect, soap operas presumably contribute to people's knowledge of death as they have already done for subjects such as mental illness.(1)
Could it be, however, that these mirrors of our daily lives do not quite reflect reality when dealing with death? Or are deaths in soap operas just convenient ways to raise audience figures while distorting the perception and possibly fear of violence in society?
Methods
We studied mortality in four British soap operas on television - Coronation Street, EastEnders, Brookside, and Emmerdale - from 1985, when the newest of the four, EastEnders, was first broadcast, to the middle of 1997. A literature review using Medline found little other literature specifically on mortality in soap opera.(2) We next sought data from the producers on the dates of deaths or losses to follow up of all named characters, together with their ages and the dates of their first appearances, but we were successful only in the case of Coronation Street. Information obtained from a variety of internet sites was of sufficient quality to enable a suitable cohort dataset to be compiled for EastEnders. For Brookside and Emmerdale we were able to produce only a list of deaths. Causes of death were divided into external (E code) and other causes, using codes of ICD-9 (international classification of diseases, ninth revision). External causes included car accidents, murders, drug overdoses, suicides, and an aeroplane crash. For comparative data we used the Office for National Statistic's CD ROM of 20th century mortality, which lists age and cause specific death rates for the United Kingdom from 1900 to 1995.(3) We assumed that these rates were the same in 1996 and 1997 as in 1995.
| Analysis of data Survival in Coronation Street and EastEnders was analysed using Kaplan-Meier survival curves and Cox regression, with age as a time dependent covariate. We calculated the proportional mortality fraction for deaths from external causes for all four soap operas. This is defined as the proportion of all deaths that are attributable to a particular cause. It was compared with the proportion of deaths from external causes which would be expected in the population of the United Kingdom at precisely the ages and dates observed, which gives the proportional mortality ratio.(4) The likelihood of a given death having an external cause was compared across soap operas using logistic regression to adjust for age at death and for sex. |  | |
| EastEnders: Arthur Fowler is mourned by his wife and sons |
Results
| Survival analysis There was some indication of a difference in survival for characters in EastEnders and Coronation Street (P<0.10). Over a short interval of time, such as the duration of an episode, a character in EastEnders was nearly twice as likely to die as was a similar character in Coronation Street (hazard ratio 1.9 (0.9 to 3.9)). The figure shows the survival curves for the two soap operas, with subjects divided into three groups according to the age at which they were introduced. Table 1 shows the five year survival rates obtained from these curves. People suffering from many forms of cancer and other serious diseases have better five year survival rates than these characters seem to have. There was no difference in survival at five years between the soap operas in characters under 29 years old. The figure also shows that people moving to Coronation Street between the ages of 30 and 44 went on to lead charmed lives, while their peers living in Albert Square dropped like flies. The survival analysis assumes that people who are written out of the serials without having to die - because their characters have become dull or dreary to the audience - are not at an increased risk of death relative to people who stay. In fact, many disappearances from the square or the street may have been precipitated by an increased risk (witness the recent departure of the Jackson family from EastEnders because of threats to their lives), and this would tend to mean that our estimates of survival were conservative. Standardised mortality ratios |  | |
| Kaplan-Meier survival curves for characters in Coronation Street and EastEnders in three age groups. Lines for general population show survival of cohort from population of United Kingdom who were followed up from 1985, when they were aged 20, 35, and 50 respectively(3). |
| Table 1 - Five year survival rates in percentages (with SEs) of characters in two British soap operas according to age at first introduction and in comparison with general population | |||
| Age (years) | Coronation Street | EastEnders | General population* |
|---|---|---|---|
| 0-29 | 90 (5.4) |
91 (4.0) |
99.7 |
| 30-44 | 100 | 79 (8.6) |
99.5 |
| 45 and over | 86 (4.6) |
73 (12.1) |
97.3 |
| *Survival in 1990 of people aged 20, 35, and 50 in 1985.(3) | |||
These figures are higher even than the highest stable estimates of occupational mortality from the Office for National Statistics for 1980.(5) The most dangerous occupations in the United Kingdom at that time had standardised mortality ratios of only 308 (foremen in the printing in industry) or 413 (foremen in metal furnace industries). Actors had a standardised mortality ratio of only 99.
For more detailed data, we obtained figures on excess mortality from the insurance industry (Munich Reinsurance Company's occupational rating guide, 1996). We estimated the standardised mortality ratio that would arise if, instead of becoming characters in one of the soap operas, the people studied had chosen some other hazardous profession. Table 2 shows the results in decreasing order of foolhardiness. Being an EastEnders character headed the list, beating Formula One motor racing to the top spot. Life on Coronation Street was still more risky than being a bomb disposal expert or a steeplejack.
| Table 2 - Standardised mortality ratios (SMRs) for various high risk groups in comparison with general population | |
|---|---|
| SMR | |
| EastEnders characters | 771 |
| Formula One drivers | 581 |
| Coronation Street characters | 353 |
| Oil rig divers | 235 |
| Bomb disposal experts | 196 |
| Steeplejacks | 148 |
| General population | 100 |
In total, there were 17 deaths in EastEnders compared with 26 in Brookside and 28 in Emmerdale. Although we did not have data on the denominator population for Brookside and Emmerdale, these data suggest that these soap operas would have had even higher standardised mortality ratios than EastEnders or Coronation Street. Brookside Close and Emmerdale could well be the most dangerous streets in the United Kingdom.
Proportional mortality ratios
Table 3 shows the proportional mortality ratio for deaths from
external causes in each of the four soap operas. Logistic regression
analysis found no differences between them in the number of deaths from
external causes after adjustment for age at death and for sex. Because
of the comparatively young ages at which subjects tended to die, the
expected proportions of deaths with external causes were high.
Nevertheless, in each soap opera there were at least 2.5 times the
expected numbers of deaths from external causes, and the proportional
mortality ratio for all soap operas combined was 2.7 (2.3 to 3.1)
(table 3). Note that in the United Kingdom deaths coded E account for
only 3.2% of all deaths because most deaths occur among elderly people
and external causes of death are rare in this group of
people.(3) But overall in the soap operas they accounted for
64% of all deaths, partly because of the younger age of the characters
and partly because the sets are more violent places to
live.
| Table 3 - Total numbers of deaths and numbers of deaths from external causes in four British soap operas | |||||
| Soap opera | Total No of deaths | No of deaths from external causes | PMF (%) | Expected PMF in age matched population(%) | PMR (95% CI) |
|---|---|---|---|---|---|
| Coronation Street | 14 | 6 | 43 | 17 | 2.5 (1.4 to 4.6) |
| EastEnders | 17 | 11 | 65 | 22 | 3.0 (2.1 to 4.1) |
| Brookside | 26 | 20 | 77 | 28 | 2.7 (2.2 to 3.3) |
| Emmerdale | 28 | 17 | 61 | 24 | 2.5 (1.9 to 3.4) |
| All | 85 | 54 | 64 | 24 | 2.7 (2.3 to 3.1) |
| PMF=proportional mortality fraction. PMR=proportional mortality ratio. |
|||||
Discussion
| This paper has proved what has been long suspected to be the case:
Brookside Close, Coronation Street, Albert Square, and Emmerdale are
highly dangerous places to live. Characters tend to die young and from
a variety of obscure and often violent causes, ranging from the mystery
virus in Brookside, which killed three, to a plane crash
in Emmerdale, which killed four.
Of course soap opera has to be melodramatic to be interesting, but should not the portrayal of death be a little more reflective of real life? It seems sad that for soap operas to hold our interest they have to be about as dangerous as Formula One racing. We hope this paper will stimulate further investigation and debate into the two soap operas for which we were unable to produce a comprehensive cast list. In the meantime, however, characters in these serials would be advised to wear good protective clothing (designed to withstand sharp implements, sudden impacts, and fire) and to receive regular counselling for the psychological impact of living in an environment akin to a war zone. We apologise in advance to the estate agents covering these areas, because for the rest of us the advice is clear: don't buy your next house in Albert Square, Brookside Close, Emmerdale, or Coronation Street. |  | |
| Brookside: Gladys meets her controversial end with the help of her family |
| Key Messages | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
We thank the following for their help: Daran Little,
Coronation Street archivist; John Peake of Inside
Soap magazine; Simon Harris from Scottish Legal Life; the
rec.arts.tv.uk.eastenders newsgroup; and Julia Bunting
from the Office for National Statistics.
Funding: None.
Conflict of interest: None.
Department of Public Health and
Epidemiology,
King's College Hospital,
London SE5 9RS
Tim
Crayford, senior clinical fellow
Richard Hooper,
statistician
Sarah Evans, research
associate
Correspondence to: Dr Crayford
email: t.crayford@kcl.ac.uk
References 1 Gerbner G, Gross L, Morgan M, Signorielli N. Portrayals of
mental illness in daytime television serials. Journalism
Quarterly 1985;62:384-7, 449.
2 Gerbner G, Gross L, Morgan M, Signorelli N. Health and
medicine on television. N Engl J Med 1981;305:901-4.
3 Office for National Statistics. Twentieth century
mortality on CD ROM (1990-1995). London: ONS, 1996.
4 Breslow N E, Day N E. Statistical methods in cancer
research. II. The design and analysis of cohort studies. World
Health Organisation-International Agency for Research on Cancer, 1987.
5 Office of Population Censuses and Surveys.
Classification of occupations 1980. London: OPCS, 1980.
Home | Current issue | Past issues | Classified ads | Career Focus | Feedback
Collections | About this site | About the BMJ | BMA | Medline