In this piece AL Bacharach praised a 1933 paper by Dr GCM M’Gonigle, the medical officer for health in Stockton-on-Tees. The paper analysed the effect of improved housing conditions on a population’s health, as measured by death rate. The piece is written in scientific terms, with little thought for the “control” group of the experiment – those who were left behind in the “Riverside area” of the slum. “The fate of the controls,” Bacharach assumed, was not “so unfortunate as to cause anyone of tender social conscience to resent their compulsory scientific service.” The result of the experiment might come as a surprise: between 1923 and 1927 and 1928 and 1932, the population that moved to the “hygienic Mount Pleasant estate” died 50 per cent faster than those in the Riverside area. These deaths were not due to infant mortality, but diseases such as measles, cancer, heart conditions, bronchitis and pneumonia. The deaths correlated with a diminished expenditure on food, which occurred because the rents of the families at Mount Pleasant had increased. Their income had not.
To be killed by improved housing conditions is a fate against which insurance premiums should be extremely low. Yet that, under certain conditions, it can happen and is happening has been proved, and proved by experiment.
In the social laboratory the successful consummation of a convincing experiment is a very rare event. First, the interference of the many variables always operating within human groups must be limited by the use of a large experimental group; sufficient results must be obtained to allow of statistical analysis. Second, the necessary technique to carry out that analysis must be available. Third, a “control” group, with which to compare the experimental group, must be at the disposal of the investigator. Fourth, perhaps most important of all, the investigator himself must be a man of the right type, knowing what information may be reasonably expected from the investigation, how to conduct the experiment so as to obtain that information, and exactly what conclusions may legitimately be drawn from it.
It is therefore not exaggeration to state that a recent paper by Dr GCM M’Gonigle, medical officer of health (MOH) for Stockton-on-Tees, must command the closest attention from everyone concerned with the relation of public health and economic conditions. Dr M’Gonigle’s paper was read in February before the Royal Society of Medicine, and has now been published in the April issue of that Society’s proceedings. It is earnestly to be hoped that it will also be published in some form more easily accessible to the non-medical reader.
The opportunity to carry out his experiments was seized by Dr M’Gonigle, with admirable foresight, in 1927. As a result of a survey carried out in 1919, it was decided to demolish a slum area, known as the “Housewife Lane area”, and to transfer its whole population to a new block of model dwellings, to be called the Mount Pleasant estate. The demolished area was part of a large slum, too large to be transferred as a whole. Its division into two parts was made along a purely arbitrary geographical line, and the inhabitants left behind, 1,298 people, in the “Riverside area”, formed an almost perfect control to the 710 individuals moved to Mount Pleasant. Nor, as we shall see, was the fate of the controls so unfortunate as to cause anyone of tender social conscience to resent their compulsory scientific service.
The main fact emerging among several others of great interest concerns the respective death-rates of the two populations. Figures for these are obtainable for each of the five years 1927-1932, and they are also given for each of the preceding five years. In all cases the crude and the “standardised” death-rates are both given, but only the latter will be considered here.
Stockton-on-Tees, with a population of just under 70,000, had, for the two periods in question, average death-rates of 12.32 (1923-1927) and 12.07 (1928-1931) per thousand; the corresponding figures for the whole of England and Wales were 10.48 and 10.30. In the Riverside area the average death-rate fell slightly from 26.10 (1923-1927) to 22.78 (1928-1932). But the transferred population, whose death-rate in the Housewife Lane area had been 22.91 (1923-1927), has been dying 50 per cent faster in the hygienic Mount Pleasant estate, where the average death-rate rose to 33.55 (1928-1932). All the other populations tabulated, in England and Wales, in Durham County, in the borough of Stockton-on-Tees, and in the Riverside area slum, show a decreased death-rate during the same period.
This increased death-rate is not due to an increase in infant mortality, in which Mount Pleasant shows a definite improvement over the Housewife Lane area. It is associated almost entirely with a percentage increase in the number of deaths from measles, cancer, heart conditions, bronchitis and pneumonia – three of them, it will be noted, definitely of infective origin, despite the immeasurably improved housing conditions.
Dr M’Gonigle has shown in his paper, almost beyond any shadow of doubt, that this increased death-rate is definitely correlated with a diminished expenditure on food, consequent on the increase in rents (about 4s. 6d. per week) payable by the families moved from the Housewife Lane area to Mount Pleasant, without any accompanying increase in the family income. To the dietitian it is of great interest to note that the economies have, in general, been made by the purchase of less, or poorer, food containing “first-class” protein, that is, protein of animal origin in meat and milk products. Among the employed sections of the Mount Pleasant population there has actually been a greater consumption of calories – as carbohydrate and low-grade fats and oils – than in the “controls” left in the Riverside area. There has also been an increase in the total protein, fat, and carbohydrate consumption among the employed in Mount Pleasant, but the unemployed show reductions in every item, again in both cases by comparison with the Riverside controls.
Dr M’Gonigle has also clearly demonstrated another fact of very great interest and, at present, completely baffling. Accompanying the increased death-rate in Mount Pleasant has been a rise in birth rate. The figures are Riverside area 36.51 (1923-1927) and 32.02 (1928-1932); Housewife Lane (1923-1927), before the transfer, 35.08. But the average figure for Mount Pleasant (1928-1932) is the extraordinarily high one of 44.25.
The investigation is not claimed to show a positive correlation between a low birth-rate and the consumption of high-grade protein, but that is what the facts seem to indicate. It would seem indeed strange if the kind of food best suited for keeping men alive is the kind most calculated to prevent their being born. Yet certain laboratory experiments have indicated some curious inter-relationships between successful gestation and the intake of protein and certain vitamins.
This work has opened up a large field to other investigators, and, it is to be hoped, for Dr M’Gonigle himself for many years, whether at Stockton or elsewhere. He deserves from all honest observers high praise and ungrudging thanks for a piece of work scientific both in its execution and in its conception. Whether it will be as far-reaching in its consequences as its conception and execution justify depends on forces and circumstances with which Dr M’Gonigle has nothing to do.