First of all we might as well launch our stat attack. This graph gives you the numbers of children, of all ages, in the Home for those years (including 1924 when it was in Glenamaddy) and note that this is an estimate for many years:
This is the number of deaths of all children in the Home as derived from the death certificates (there were no deaths in 1961):
The national Irish Infant Mortality Rate (IMR), i.e. (number of deaths of <1 aged infants in a given year/live births in that year) X 1000, for those years is given here:
Although we aren't using the exact same type of statistic, of necessity because we don't have the full data to work on as described below under 'sources', this graph compares the national IMR with that in the Home:
So for those years, 1925-60 excluding 1961 when there were no deaths in the Home but also when it was being wound down so it would be very difficult to estimate numbers, our best statistic is that on average you have a national IMR of 60 and a Home IMR of 128, but because of the unsatisfactory way we have to calculate the Home figure we have to estimate that the gap is in fact wider than this.
Hence the six million dollar question is why is the Home rate seemingly more than double the national rate? I think there are five basic biological, statistical and understandable reasons why this was so:
1. One thing that has exercised the minds of statisticians and historians for those years with regard to the Irish Infant Mortality Rate is that in urban areas, particularly Dublin, we find that the rate is much higher, in fact its much closer to the Home rate. So for example when the national rate of 66 was calculated for 1938, in Dublin Borough it was 102 and Cork 103.(1) Also this urban/rural gap closes about the early to mid 50s at the same time that the Home rate falls to below the National one. So it seems to this observer that the special problems that caused the high IMR in Irish urban areas of that time, and which were solved by the mid 50s, could shed a light on the problems that the Home faced during the same time. It seems there are two basic reasons why the Dublin/urban Infant Mortality Rate was so much higher, and these were:
a) Sanitary issues in the overcrowded Dublin slums. In some cases families in apartment blocks, and even whole streets, were using the same toilet facilities and also the water supply was in some cases also primitive which in turn had a bad effect in terms of deadly infant infections. So if you read up on the literature of this you will find that the state put great store in more modern sewage and water facilities and felt that this was a key reason why the Dublin IMR rate eventually fell.
Of course we hope that the Home facilities were quite good and not at all in that primitive state but that is not to say that this didn't effect the Home IMR. The point is that on occasion government inspectors would go into homes in the area and if they found that some children were living in apalling sanitary conditions they sometimes took the children away and placed them in the Home. So in a sense children from the very worst sanitary conditions are actually in the Home and this it was felt did impact on the cause of death there, as the Inspector mentioned in 1947:
b) It seems clear that the second big reason why the rate was much higher in Dublin etc was because in the overcrowded conditions in the cities the very contagious and then deadly infant diseases of measles, influenza, whooping cough and diphtheria were able to spread in a type of pandemic fashion across the infant population whereas in the more spread out rural areas, especially if children were held back from school, these then incurable diseases could be more easily contained. Again the rate fell as you get into the 50s because modern medicine developed better drugs and vaccines to deal with these diseases."First there is the constant risk of infection brought in from outside by the admission of whole families of itinerants, destitutes, evicted persons etc." (2)
This I think is clearly the story in the Home as well. They seemed to have great problems with these terrible diseases for the simple reason that they had a lot of children housed together in the same building. This was hardly their fault, and it was hardly the fault of anybody because the state could surely not have afforded at the time to have say 10 different Children's Homes in a given county so of necessity they had to place these children together but this was bad news when it came to the deadly wave of measles attacks etc that clearly affected the Home.
This is a very major issue as a half glance at the death certificates will show. So for example if we take the worst years in the Home we find that in 1926 64% of the deaths were attributed to measles, in 1936 it was 48% and in 1947 35%. Also in 1933 37.5% of the deaths were due to Whooping cough while in 1943 24% of the child deaths were due to the same disease while 20% were due to influenza in that year. John Cunningham who stayed in the Home also refers to it:
But are we to blame the staff for not being able to cure infants of these then incurable diseases? There was talk of isolation chambers, and in fact the staff frequently pressed the state for funds to create facilities like this but improvements were slow in coming and anyway those diseases are exceptionally contagious.(4) Its very unlikely they could have contained them to only a part of the Home without elaborate, and then very expensive and rare, isolation chambers."Some of the women, like my surrogate mother (Mary), stayed on for life - possibly because they were briliant nurses for newborn babies that were there in lines in cots. They had to be brilliant nurses for the infants were then potential victims for diseases which we now regard as just a nuisance, like measles, whooping cough, scarlet fever, the mumps." (3).
Hence it seems to this observer that there are good reasons why the Home rate mimicked the much greater Dublin/urban IMR of that time, rather than the lower national rate, and for reasons that were not the fault of the staff there.
2. We can assume I think that the vast majority of the illegitimate children in the Home, which again would be about three quarters of the total children in the institution (in 1928 the Home had 118 children of whom 96 were illegitimate (5) but the latter proportion might have been somewhat less for the War period), would be first births to young unmarried mothers (who obviously, in almost all cases, went without asistance from fathers or a wider family). Its interesting then to read about one wide ranging 2011 study of infant mortality data in 55 low and middle income countries which shows that:
While its true that this study is also talking about social issues, which we hope do not then impact our analysis of the Home in 1925-61, nonetheless they found that there are both "biological and social mechanisms" at work in giving this high infant mortality rate in children of young first time mothers. These biological reasons may therefore be another cause why the infant mortality rate was higher in the Home than nationally and through no fault of the staff there."The first-born children of adolescent mothers are the most vulnerable to infant mortality and poor child health outcomes. Additionally, first time mothers up to the age of 27 have a higher risk of having a child who has stunting, diarrhoea and moderate or severe anaemia." (6)
I should add here, for the sake of completeness, two other potential biological issues that could impact on the IMR in the Home. The first point is a rather dark and, hopefully, not important one to make which is that in three cases 'consanguinity' is mentioned in the death certificates. This seems a very strange reason to give as a cause of death unless the writer knows that the parents of the dead child are 'very' closely related, e.g. brother and sister or father and daughter? Obviously this would then cause serious biological reasons for a higher IMR and again through no fault of the staff although this is only true of three certificates anyway.
The second point is that the Inspector in 1947 raised the prospect of venereal disease being a factor in the health of the children, complaining that there is insufficient testing for it:
Presumably he had reason to believe that this was strongly merited, maybe because e.g. a significant proportion of the mothers were working as prostitutes? Without knowing why he felt so strongly about VD being a factor in the infants health we cannot judge now the significance of these diseases in the IMR in the Home but at any rate it could be another biological reason for the high IMR and again through no fault of the staff."Secondly, there is no routine examination and testing for venereal diseases." (7)
3. While the Home tried to insist that mothers stay for at least a year after the birth with their child nonetheless there were in fact also a proportion of totally abandoned children who had to be brought up in the Home without any mother and this must have been difficult and also understandably impact on the mortality rates. You hear stories of the Gardai placing an infant in the Home who was found abandoned (8) and of mothers being chased for maintenance who completely deserted their children,(9) of children sent to the Home because their mother died when they were very young (10) - or even perhaps during childbirth? -, and even of infants being left behind on trains who were sent to the Home,(11) etc. At least for the early period this seems to have been quite a factor in the mortality rate as reported here from 1924 while the Home was still in Glenamaddy but about to go to Tuam:
But surely we can always accept that motherless infants - no matter how much care they get - are bound to have a higher mortality rate than those with mothers present - which would obviously be the case for the vast majority of the infants in the national mortality figure - especially for the very early weeks or months and again this is surely not the fault of the Home."...but the death rate amongst the infants has been higher than it ought to have been because of the difficulties of rearing motherless babies." (12)
4. Although the facts are somewhat slight in this case nonetheless it seems fair enough to speculate that there may be a selection factor causing bias in our statistics whereby some of the legitimate children in the Home may have been sent there because their parents were unable to rear them at home due to some chronic type of illness or disability. This I admit is speculative but surely its reasonable to guess that since the Bon Secour nuns in the Home ran an acute hospital in Tuam during this period that maybe in cases where an ill child couldn't be cared for at home, and had a chronic illness/disability not urgent enough for an acute hospital, that the child would have been admitted to the Home as a last resort? Again we have this from John Cunningham:
Hence if a significant proportion of the infants in the Home were admitted there at least partly because they were too 'sickly' to be reared in their own homes then we have a clear reason why the IMR would be higher in the Home than the general population."My mother died shortly after I was born and because of the close proximity of the Children's Home to my family home, and because rearing a sickly infant would be well-nigh impossible for a widower who already had four young children, I was reared in the Children's Home, though regularly spending some time at home as well." (13).
Meanwhile reading the 1947 report its interesting to speculate that they seem to have a disproportion of disabled infants living there:
Possibly whats happening is that some parents, maybe on the advice of medical professionals, faced with rearing disabled children are in fact placing them in the Home?"...a baby with mis-shapen head and wizened limbs...deaf and dumb who is awaiting a vacancy in the Institute for Deaf and Dumb...there is an albino child...5 years, atrophied areas - hands growing near shoulders..." (14)
Also there are very consistent reports of a large proportion of infants described as 'mentally defective', or even 'idiots', being in the Home, for example in the above 1947 report, in the death certificates, and in some media reports of the time including this one from the closing of the Home in 1961 when they found:
Of course this has generated much outrage in the popular press but I think there is no point is getting excited about the accepted medical terms of the day, surely they are describing here Down Syndrome children? After all how would they know that a child of age about 1 was mentally defective anyway, unless in fact they are basing their diagnosis on the well known physiological characteristics of Down Syndrome children? The 'unduly high' number mentioned is then probably a reflection of the fact that a Down Syndrome child was frequently sent into the Home during this period, when our understanding of mental handicap in children was slighter than it is today."...an unduly high number of mentally defective children in the Home." (15)
But Down Syndrome children frequently have unique medical problems caused by their special genetic condition, as often do other disabled and, obviously, chronically ill children, which unfortunately makes the rate of mortality in those children higher than the normal infant population, much higher probably at this period when these medical problems were less understood than now. So if we find that Down Syndrome was a common reason why legitimate children were sent into the Home, and the frequent reports of 'mentally defective' children make this very likely and a significant statistical bias, then we find again a good reason why the infant mortality rate is higher in the Home than outside, caused not by bad care but only by genetics!
To finish off the point about statistical bias we should point out that we are not talking about a static population here. At least after 1 year, and possibly before that in some cases anyway, (16) mothers are taking their children out of the home and also there are children being fostered and adopted out all the time. But the point is that presumably the ones going out of the home in those cases are the healthy children, doubtless a mother will not take a sickly child out nor will it be fostered or adopted, while the unhealthy ones remain so it provides a good reason why in the unique conditions of the Home you would have a higher IMR than in the normal population. But this is complicated by the fact that most of the children who leave would be over the age of 1, which is an age usually above that recorded in the IMR anyway and hence it shouldn't be too big a factor in creating bias in our statistics.
5. But there is one other very important statistical issue here that I think anyway could have a very big impact in artificially increasing the IMR in the Home as opposed to the national rate. In both cases we are basing our figures on deaths registered, ultimately, in the national register of deaths in Dublin which had been started in Victorian times. While these are incredibly valuable records for genealogy and all kinds of statistics nonetheless for this period it has to be pointed out that the degree of registration of births and deaths was far more patchy across Ireland at that time than people realise. For example note this quote from the Oireachtas in 1926:
Where Mr Farren is talking there about poor people not needing a certificate he is referring particulary to the fact that somewhat wealthier people with life insurance or property would need a certificate but the poor people he mentions would not. Obviously babies and infants are not likely to fall into either category so they are probably less likely to be registered."Sir Edward Bigger: Fully 70 per cent. of the deaths that occur in many districts are uncertified by any medical attendant. There is nothing to indicate the cause of death, and the whole of our statistics about deaths are worth nothing and are absolutely useless.
Mr. Farren: The people that you have in mind here are very poor people and these people do not really require a certificate, because the number of uninsured amongst those people would be relatively very small. They are nearly all insured in the Prudential or the Royal Liver.
Sir Edward Bigger: Speaking of the urban districts that is true.
Mr. Farren: Yes, and in the rural areas also it is largely true." (17)
If you read the Dail and Seanad reports this was a constant complaint about Irish death statistics,(18) even for example the CSO estimated that as late as 1949 some 11% of deaths were uncertified.(19) (By the way, to clarify, there is a difference between the two, some deaths were registered without being certified - by a doctor - for various reasons but the two problems were linked, as pointed out in the Dail in 1973:
As late as a study conducted in a number of parishes in the West of Ireland in 1966-69 which double checked the death registers with funeral entries in parish registers and other sources, we find, by some calculations, that about 10% of deaths were not registered:"The incidence of uncertified deaths is falling steadily and in 1970 represented 1.5 per cent of all deaths registered. It is hoped that this progress will be maintained. An associated problem is the failure to register deaths." (20))
Now the point here is: could this cause some bias in our death certificates that could be a factor here? Well you see in the institution there was a dedicated Medical Officer, doctor, paid for by the state so naturally it was a simple matter to get a doctor to register the deaths, i.e. that Medical Officer."In 1966-69 out of a total of 806 deaths 90.2% were registered and 85% were certified whereas in 1974-77 out of a total of 885 deaths 95.2% were registered and 92.7% were certified, an improvement of 4.8% and 5.3% respectively." (21)
But picture the scene in a rural cottage in Galway in 1926 or whatever. Obviously nearly all births that time - and for all the period 1925-60 - were at home and delivered with the assistance of a midwife not a doctor. So for example in the Home certificates we have a few babies who unfortunately died after only a few hours and I would say the chances are very high that if that happened in our hypothetical cottage they just aren't going to go to the trouble of registering that death, or birth indeed. And since we know that the main bias in favour of registering people was if they had property or life insurance, both conditions highly unlikely to pertain to an infant, we can say that infant mortality figures are very likely to be heavily under reported in the national death statistics.
Actually this is a recognised phenomenon seen around the world, where in some cases you can get artifically high IMR rates not because the actual rate is high but because the reporting is better in a hospital setting:
Then we have a clear reason why the Home infant death rate would be much higher than the national one, because many people across the country probably didn't bother registering these type of infant deaths frequently at that time but with a resident doctor they did register them promptly in the Home. And if it was as high as 70% in some areas where deaths are unreported then we could be onto something here as the real cause of the bias in these figures.(23)"Another seemingly paradoxical finding, is that when countries with poor medical services introduce new medical centers and services, instead of declining, the reported Infant Mortality Rates often increase for a time. This is mainly because improvement in access to medical care is often accompanied by improvement in the registration of births and deaths. Deaths that might have occurred in a remote or rural area, and not been reported to the government, might now be reported by the new medical personnel or facilities. Thus, even if the new health services reduce the actual IMR, the reported IMR may increase." (22)
For these five reasons then, statistical, biological and understandable, I think the difference between the IMR rate nationally and locally in the Home can be satisfactorily explained without casting aspersions at the staff in the Home at that time.