View report page

The two smogs, both of which covered five days, were somewhat different in that in
December, 1957, smoke was 84 per cent. and sulphur dioxide was 120 per cent. of the
corresponding volumes in January, 1956.
Figure 4 (page 14) shows the seven-day moving average of deaths occurring in
London (excluding the 90 deaths in the Lewisham rail disaster on the 4th December)
over the period 20 November to 18 December, i.e., the smog days and about two weeks
cither side of them, for all deaths and for deaths affecting those aged under 70 years
and 70 years and over. Because of the use of seven-day moving averages the figure for
any one day is influenced by the number of deaths on that day and three days either
side of it, hence the thickened lines of the curves include all the figures affected partly
and wholly by deaths over the foggy days. The estimation of the number of excess
deaths is determined, somewhat arbitrarily, by Unking, by the dotted line, the daily
trend of deaths before and after the subsidence of the peak figure of 189 on the 6th and
7th December. On this basis the number of excess deaths represented by the bulge in
the mortality curve works out at some 300 deaths. The rail disaster deaths, though an
indirect consequence of fog, are excluded because the purpose of this study is to arrive
at an estimation of the number of deaths directly attributable to the physical action of
smog upon human beings.
Before accepting this estimate it is necessary to study the general background of the
weather and mortality at this time. It was cold during the period of the smog with night
temperatures below freezing point. There was no cold spell immediately following
this smog with which to contrast the effect of cold and smog, as was done with the
smog of January, 1956 (see page 206 of my 1956 report), but the even colder weather
in the week of the 19th—25th January, 1958, produced no such similar disturbance in
the level of mortality. A further complication was the presence of the influenza epidemic
in the autumn of 1957, but the peak mortality from this cause and associated respiratory
diseases occurred in late October, after which the number of deaths fell to a little above
the normal seasonal level, rising again in mid-November and continuing to rise after
the smog episode to the highest figure of the winter in the week after Christmas.
Viewed in perspective, the bulge in the curve due to smog occurred on a rising gradient
of mortality, in a period when deaths were at a level higher than usual following an
excess of deaths due to the influenza episode. Of the causes of death during the smog
and the week following, the greatest proportionate rise was in diseases of the respiratory
system, especially in bronchitis; there was also the customary rise in the number of
'mentions' of respiratory disease on death certificates. After consideration of all these
aspects there is no reason to doubt the validity of the estimate of 300 excess deaths
directly attributable to the smog of December, 1957, or 400 if the deaths resulting from
the Lewisham rail disaster be included.
An interesting feature of this smog, in contrast to the smog of January, 1956, was
that it affected both those under 70 and those over 70 years of age. In January, 1956,
it was the old people who died; in this smog when there was relatively less smoke and
more sulphur dioxide the under seventies also died, but on the basis of a detailed inquiry
of about one-third of London the excess deaths in this group were mainly among
persons aged between 50 and 70 years of age. It would be foolish to make deductions
from the experience of these two episodes about the relative importance of smoke and
sulphur dioxide on mortality and this research must be carried on over a series of smogs
before any firm conclusions can be drawn.
1 would like to add here my thanks to the medical officers of health of the metropolitan
boroughs who are co-operating in this research by supplying the mortality
data on which these findings are based.
15
B