View report page

There have been a number of surveys of old people, but none that we knew
of had attempted full medical examinations; so that accurate information about
the health of people over 60 was lacking. Dr. Smith chose the villages of
Tunstall and Silksworth in his area in County Durham as typical of a middleand
working-class urban district. The population was 7,500, the density of
houses and persons per acre slightly better than average, and in other respects
such as the infectious disease rates and death rates the area was about average.
The workers are employed mostly in the coal-mining and shipping industries
and small shops.
Arranging the Medical Examinations.
The names and addresses of all people over 60 were obtained by using every
source of information possible such as the voting lists, all organisations and
clubs, the churches, welfare officers and the medical practitioners. The
complete list contained 550 names or 7.3 per cent, of the population, which is
slightly less than in comparable areas of England and Wales.
By courtesy of the consulting physician Dr. o. Olbrich, the facilities for
medical examination in the department of geriatrics at Sunderland General
Hospital were placed at Dr. Smith's disposal. He arranged for about 60 people
from the list to be invited each week to attend the hospital. About half of them
accepted; but some, perhaps feeling quite well and not in need of medical
attention, some perhaps fearing something being found out, some saying they
did not wish to tempt providence, some recently attended by their own doctors
or already attending hospital, and some ill in bed—and for other reasons were
reluctant or refused to attend. Dr. Smith visited most of them and persuaded
a number to come. Some private car owners including the Red Cross volunteered
to provide transport. The number finally persuaded to attend was 238
out of the original 550.
The medical examinations were very thorough indeed, covering every
system of the body and including X-rays and laboratory tests. The findings
were entered on the hospital clinical records.
Factor Analysis.
We decided to do a factor analysis of the findings to see if they revealed
any factors. This procedure is used to test if there are any common influences
which can account for the differences amongst the people examined; it also
serves as a means of classifying them. We selected certain information from
the clinical records such as sex, age, haemoglobin level, blood pressure, and the
diseases found. Some persons had two diseases, some three, and a few four.
We divided these details into 42 sub-classes. We then made up a table showing
how many people fell into each sub-class. This table constitutes the first step
in the analysis of the factors.
The whole analysis is a complicated mathematical procedure. It took
many many hours of work at odd times. One part required the use of Hollerith
punch-card and sorting machinery and the use of the electronic calculator.
15