The obituary columns of the BMJ have become more informative since editorial
policy tried to encompass all medical deaths early in 1995. Our impression was that the new
columns showed that doctors born in the Indian subcontinent died earlier than those born in
the United Kingdom. We tested this impression in a systematic analysis.
Methods and
results
We reviewed 572 obituaries in the BMJ from April 1995 to December 1995 and
in the issue of 6 April 1996 for the doctor's age, country of birth, and specialty. Age at
death was recorded in whole years, and the significance of the differences in means assessed
using Student's t test. No correction for continuity was applied to take account of
the month of death. To increase the size of groups for comparison, countries of origin and
specialties were grouped as in table 1, comparisons being made with the largest group,
United Kingdom and primary care respectively. Three junior doctors who had not selected a
specialty were excluded from the analysis of specialty.
| Table 1 - Longevity of doctors related to country of origin and specialty |
|
No of doctors |
Mean (SD) age (years) |
Difference in means* |
Student's t |
P value |
| Country of origin |
| United Kingdom |
469 |
75.2 (13.4) |
|
|
|
| Republic of Ireland |
12 |
80.1 (7.9) |
4.9 |
1.25 |
0.21 |
| English speaking world (Australia, New Zealand, United States, Canada, South Africa, and the Caribbean) |
27 |
81.5 (9.8) |
6.3 |
2.40 |
0.017 |
| Rest of Europe |
22 |
80.2 (14.9) |
5.0 |
1.68 |
0.093 |
| Indian subcontinent (India, Pakistan, Bangladesh, and Burma) |
21 |
61.8 (14.8) |
13.5 |
4.47 |
< 0.001 |
| Rest of world (Africa and rest of Asia) |
21 |
65.0 (18.7) |
10.2 |
3.33 |
0.001 |
| Specialty** |
| Radiology |
24 |
78.1 (12.8) |
3.9 |
1.26 |
0.21 |
| Hospital medicine |
137 |
77.5 (11.4) |
3.3 |
2.21 |
0.028 |
| Laboratory medicine |
46 |
77.2 (12.0) |
3.0 |
1.33 |
0.18 |
| Surgery |
91 |
75.7 (13.8) |
1.2 |
0.84 |
0.40 |
| Primary care |
192 |
74.2 (14.3) |
|
|
|
| Obstetrics and gynaecology |
22 |
73.1 (11.4) |
1.1 |
1.33 |
0.18 |
| Public health |
7 |
73.0 (12.9) |
1.2 |
0.22 |
0.83 |
| Psychiatry |
24 |
70.8 (13.6) |
3.4 |
1.09 |
0.28 |
| Anaesthetics |
26 |
66.4 (19.9) |
7.8 |
2.48 |
0.014 |
| *Compared with United Kingdom for country of origin and with primary care for specialty. **Unclassified for three junior doctors. |
Doctors
from the Indian subcontinent (P< 0.001) and from the rest of the world (P = 0.001) died
before those from the United Kingdom, while those from the rest of the English speaking
world lived longer (P = 0.017) (table 1). The mean age at death of those in primary care was
74.2 years. Anaesthetists died younger than those in primary care (P = 0.014), while primary
care practitioners died before physicians in hospital medicine (P = 0.028). The oldest
survivor was a general practitioner, who died at 101. Differences between the United Kingdom
and the Indian subcontinent and the United Kingdom and the rest of the world were maintained
when anaesthetists were excluded ( t = 4.21, P< 0.001; t = 3.00, P< 0.005
respectively) but, when we considered only those born in the United Kingdom the difference
between primary care practitioners and anaesthetists was no longer significant.
Comment
Our data set is incomplete because we could not to consider all the
factors that we would like in such a selective review. Firstly, we did not record the sex of
the doctor, but in the cohort of doctors (40 000) surveyed by Doll et al in 1951 less
than 15% were women.(1) Secondly, not all deaths are recorded in an obituary. For
example, given that the number of doctors registered in 1950 was 50 000 and that the average
age at death is 75 years,(2) 1250 doctors would be expected to die each year, or 959
in our study period. We had 572 obituaries, or about 55% of expected deaths. Finally, annual
death rates in doctors should be related to those in comparable social groups. It would of
course also have been the counsel of perfection to compare doctors living in India with
those in the United Kingdom. Nevertheless, our results are a stimulus for further
investigation.
The earlier death of those born in India rather than the United Kingdom
may reflect the tendency of Indians to coronary heart disease.(3) However, this would
not explain why those born in the rest of the world also died earlier.(3) Could it be
that the NHS does not deal kindly with these foreign entrants to the service?
The earlier
deaths of anaesthetists, although not significant among doctors born in the United Kingdom,
may warrant investigation in a larger population. Hospital physicians seem to live longer
than general practitioners. Perhaps they show a higher degree of professionalism in dealing
with their intimations of mortality than their less specialised colleagues. Hospital
practitioners usually retire at 65, whereas many general practitioners used to work longer,
but varying retirement age does not seem to alter life expectancy.(4)
These results
could be validated by using more com-plete data on doctors in whom an accurate cause ofdeath
has been determined. Such investigations using doctors as guinea pigs would make any
findings more likely to be immediately translated into public health measures.
We
thank Miss Miranda Wright for abstracting the data to a computer record.
Funding:
None.
Conflict of interest: None.
Charing Cross and Westminster Medical School,
London W6 8RF
D J M Wright,
reader in medical microbiology
A P Roberts,
senior
lecturer in medical microbiology
Correspondence to:Dr Wright.
References:
1 Doll R, Peto R, Wheatley K, Gray R,
Sutherland I. Mortality in relation to smoking: 40 years of observations on male British
doctors. BMJ 1994;309 :901-11.
2 Office of Population and Census.
Mortality series. London: HMSO, 1955:38 (table 15). (DHI, No 27.)
3 Balarajan R.
Ethnicity and variations in the nation's health. Health Trends
1995/6;27 (4):114-9.
4 Niemi T. Retirement and mortality. Scand J Soc Med
1980;8 :39-41.
