Genetical Causes of Schizophrenic Symptoms
Monthly Review of Psychiatry and Neurology, Basel, Vol. 112 No. 1 (1947)
In a work that came out early in the war, M. Zehnder (Monatschr. Psychiat. 1940, 103, 230) gave an account of 92 schizophrenic brothers and sisters. The account is of more than common interest, because the writer gives us details of the disease in every one of these men and women. With the help of these details, and of the science of numbers of today, one is able to go farther than Zehnder in getting out facts of interest. She did no more than make a division of the 92 persons into two groups of families, one in which the picture of the disease in the separate brothers and sisters was a like one, and one in which it was not like. In this way Zehnder came to certain opinions. It is not my purpose to give, myself, any opinion on Zehnders' views, but in place of that to make a wider use of her observations.
Zehnder's material is made up of 5 families of 4 brothers and/or sisters, 6 families of 3, and 24 families of 2. There are in addition 3 families of 2, two to a birth. Of these last, in one family the two persons were two‑egg; of the other two families we have no knowledge. So, if we keep out these last four persons, we have 36 families, which make up 73 groups of two; for each family of three makes 3 separate groups of two, and each family of four makes 6 separate groups of two. In what comes after, keeping out the example of sex‑division, we shall have to do with these 73 groups of two.
Division by Sex.
Taking out those brothers and sisters whose birth was on the same day, we have 72 groups of two: ‑ 13 MM, 29 MW and 30 WW, for there are 89 women to 55 men. That is to say there are 43 groups of like sex to 29 groups of opposite sex. The amount by which the the like‑sexed are greater in number than the opposite‑sexed is probably no more than we are able to put to the accounter of chance (X2= 2,01, P < 0,05).[1] But the tendency is the same as in the observations of L. S. Penrose (J. Ment. Sei. 1942, 88, 308). The rule is general in diseases of the mind that two relations of any degree are more frequently of the same than of opposite sex.
Time of Starting of the Disease.
In this quality we have something to be measured, which is far from frequent in psychiatry. In schizophrenia one is frequently not able to say within wide limits of error when a man gets ill for the first time; but against that, one is able to give without error the time when he first came into hospital. Zehnder has made a note of this in every one of her examples. That number of years old at which a man first goes into hospital may, with two brothers, be the same, or like, or widely different; and we can put a measure to the tendency for it to be much or little or like different. If we put into relation with one another, two by two in every one of the 73 groups, the number of years old the men and women were at the time of coming into hospital, this measure will be given by the correlation coefficient r.
where x and y = the number of years old, N the number of persons, and the sense of S is that addition is to be made of the values. In this material r = +0,55 ± 0,08. It is of great interest that r is as high as + 0,5, for this is about the value of r in the comparison of brothers and sisters in such things as weight, and how high they are; and it is the value to be looked for in the comparison of brothers or of sisters in qualities dependent on a great number of genes. For two brothers will normally have half their genes the same and half different. A naturally to be formed opinion, then, would be that the time of getting ill in schizophrenia is dependent for the greater part on a number of genes working together. To put this theory to the test, the working out of the values of r with others sorts of relation, such as sons and daughters of brothers and sisters, would be a help.
Kraepelinian Picture
In most of the examples Zehnder has given a diagnosis, and we may say:
Catatonia is seen in 3, 4, 5, 6, 12, 14, 15, 16, 17, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 42, 43, 44, 47, 49, 50, 51, 57, 58, 59, 61, 63. 64, 65, 67, 69, 70, 71. 72, 73, 77, 78, 83, 84, 85, 89, 90, 91.
Hebephrenia in 1, 2 '7, 8, 13, 19, 20, 37, 38, 39, 40, 45, 46, 48, 56, 75, 80, 81.
Paranoid Schizophrenia in 9, 10, 11, 41, 52, 53, 54, 55, 60, 62, 66, 74, 76, 79, 82, 86, 88, 92.
So for catatonia we have the groups C‑C 36, C‑not C 18, not C‑not C 19 (X2 = 16,9; P < 0,001), and for hebephrenia H‑H 11, H‑not 117, not H‑not 11 55 (X2 = 28,5; P < 0,001). So for all three sorts of picture there is a strong tendency for like to go with like, a strong positive correlation. The material of other workers may be taken in comparison. So B. Schulz (Zeitschr. f. d. ges. Neur. Psychiat., 1932, 143, 175) says of his material of schizophrenic brothers and sisters, given in his Table 20:
"Wenn wir in dieser Tabelle nur die 3 reinen Gruppen betrachten, so finden wir allerdings eine deutliche Korrelation. Aber diese Korrelation finden wir nicht mehr bei Einbeziehung auch der gemischten Probandengruppen und gemischten Sekundärfälle." His material is smaller than Zehnder's, and certainly such positive correlation as there is might be caused by chance. F. J. Kallmann (The Genetics of Schizophrenia, New York, 1938) gives figures that are of interest here:
Kallmann says about this table:
”In this way we are able to show that only about half the schizophrenias in the children and grandchildren of our probands correspond to the disease form of the related probands (53,2 percent and 58,3 percent respectively)… We may conclude that the individual form of schizophrenia must be determined not solely by the special nature of the hereditary predisposition, but by a series of other factors."
But Kallmann has overlooked that, if there were no positive correlation, the coining together of pictures of a like sort would be even less frequent. From his table it may be worked out that in place of the numbers 49, 31 and 13, the numbers probable by chance are 30,0, 16,6 and 8,6. So that in place of 93/160 = 58 %, chance would give us 55,2/160 = 34 %. The difference is a great one (X2 = 39,6; P < 0,001). Though this is not Kailmann's opinion, his numbers make it very probable that the disease picture is partly at least dependent on the genes, which is in agreement with the suggestion given by Zehnder's material.
Outcome of the Disease.
In every one of her examples, Zehnder gives her opinion on the outcome ‑ dementia, chronicity or remission:
Dementia in 1, 2, 3, 4, 8, 13, 16, 17, 18, 19, 20, 21, 25, 26, 27, 28, 29, 30, 31, 32, 33, 35, 36, 37, 38, 39, 42, 43, 44, 47, 51, 56, 57, 58, 59, 61, 62, 64, 67, 68, 69, 70, 71, 73, 74, 76, 78, 85, 88, 91, 92. Chronicity in 7, 9, 10, 11, 15, 23, 41, 52, 60, 63, 77, 80, 82, 89. Remissions in 5, 6, 12, 14, 22, 24, 34, 40, 45, 46, 48, 49, 50, 53, 54, 55, 65, 66, 72, 75, 79, 81, 83, 84, 86, 87, 90.
These numbers may be grouped D‑D33, D‑not D 25, not D‑not D 14 (X2 = 4,2; P <0,05). So there is a more than chance positive correlation between the brothers and sisters in this respect.
Body Development.
In only 51 persons has Zehnder given details of body development, pyknic, athletic or asthenic. Testing the numbers gives no sign of a more than chance similarity between the brothers and sisters. Certainly no measures are given.
The material for the above discussion has been taken from what Zehnder herself has said. But there are other details in her accounts of which we are able to make use.
Hallucinations
From Zehnder's accounts we may take it that hallucinations were present in:
Auditory: 2, 6, 9, 10, 11, 14, 15, 17, 18, 19, 21, 22, 25, 26, 27, 31, 32, 33, 34, 41, 42, 49, 50, 51, 52, 59, 60, 61, 63, 65, 67, 68, 72, 73, 74, 76, 78, 82, 83, 87, 89, 91, 92.
Visual: 5, 6, 9, 17, 24, 38, 61, 67, 70, 72, 76, 83.
Other senses: 1, 6, 23, 36, 38, 41, 43, 44, 50, 51, 53, 54, 56, 59, 63, 65, 72, 76, 80, 82.
There is no sign of any correlation between the brothers and sisters in this respect.
Destruction of thought processes.
This is to be seen in 1, 2, 3, 4, 10, 13, 16, 18, 20, 21, 22, 28, 30, 39, 42, 44, 46, 47, 48, 50, 57, 59, 61, 64, 65, 70, 73, 77, 79, 82, 85, 91, 92.
There is no correlation between brothers and sisters.
Damage to Feeling.
Feeling (affect) was in many poor and feeble: 13, 18, 19, 20., 25, 26, 27, 30, 32, 33, 35, 36, 37, 38, 39, 40, 42, 43, 44, 47, 48, 51, 58, 59, 61, 62, 64, 67, 69, 70, 71, 73, 74, 75, 88, 89, 91, 92.
There is here a positive correlation (X2 = 4,4; P < 0,05).
Catatonia.
Sings of catatonia, stupor, negativism, etc., were seen in. 10,
15, 25, 26, 28, 32, 35, 36, 38, 39, 49, 50, 51, 62, 64, 69, 70, 71, 81, 84, 85.
The small positive correlation might be caused by chance.
Violent Conditions.
These conditions (Erregungszustände) are noted in: 1, 3, 4, 5, 6, 7, 8, 10, 11, 12, 16, 17, 18, 19, 23, 24, 25, 28, 29, 30, 32, 34, 35, 36, 39, 42, 46, 49, 50, 51, 52, 53, 56, 57, 58, 61, 73, 78, 79, 85, 87, 88, 89, 90, 91.
The positive correlation is quite strong (X2 = 6,2; P < 0,02).
Depressive Colouring.
This is noted in: 1, 5, 6, 7, 8, 14, 15, 16, 18, 21, 23, 45, 46, 50, 54, 55, 56, 57, 58, 64, 65, 67, 70, 71, 72, 73, 75, 78, 83, 84, 86, 89, 90. The positive correlation is a strong one (X2 = 29,8; P < 0,001).
Paranoid Behaviour.
A paranoid way of looking at the environment, false beliefs of ilidoing by others, etc., are specially noted in:
1, 2, 3, 7, 8, 9, 10, 11, 16, 18, 19, 20, 21, 25, 27, 28, 29, 30, 31, 32, 34, 38, 41, 42, 43, 44, 45, 49, 51, 52, 54, 55, 56, 58, 60, 62, 63, 66, 67, 68, 71, 72, 74, 75, 80, 82, 87, 88, 91, 92.
The positive correlation is quite strong (X2 = 5,7; P < 0,02).
From our observations we may say, then, that in qualities of the disease such as time of falling ill, outcome, Kraepelinian picture, damage to powers of feeling, catatonia, violent conditions, depressive colouring and paranoid behaviour, brothers and sisters are more like one another than may be accounted for by chance. The opinion is a natural one that the cause of this is to be looked for in the genes, not the gene of schizophrenia itself, but the great number of genes of small effect, half of which are common to any two brothers or sisters.
Correlations between Disease Qualities.
It is probable that the observations we have looked at are dependent to some degree either on one another or on some common tendency. For example, how old a man is may have an effect on more than one of the qualities of the disease. In the table below are noted the correlations between separate observations, that is how strong the tendency is for two observations to be made on one and the same person. We see that catatonia is more frequent in the young and in the asthenic, and has no tendency to paranoid development, in place of that to conditions of violence. Hebephrenia is seen in young and old, has no special correlations with other observations, but is far from frequently the cause of auditory hallucinations. Paranoid schizophrenia is more frequent in the old and those of pyknic body development; it has a strong tendency to chronicity but not to dementia or depressive colouring. If the outcome is towards dementia, damage of thought processes and of powers of feeling will be seen, but not depressive colouring or violent conditions. Chronicity of the disease process makes auditory hallucinations probable and damage of powers of feeling not so. A tendency to remissions is seen frequently with depressive colouring, far from frequently with damage of feeling powers and paranoid tendency. Persons of the pyknic sort have strong tendencies to paranoid pictures, less so to hallucinations of hearing, and frequently are without damage of thought processes. Asthenics have a tendency to catatonia, to violent conditions and to depressive colouring but very little to paranoid schizophrenia. In younger persons catatonia and violent conditions are frequent, in the older paranoid schizophrenia and chronicity.
Summary.
The data provided in an earlier work of M. Zehnder have been analysed statistically. The material consisted of 92 schizophrenic brothers and sisters. Analysis shows that there are family resemblances too great to be accounted for by chance in age of onset, Kraepelinian picture, course and outcome of the disease, degree of affective dilapidation, presence of catatonic signs, occurrence of states of exictement, of a depressive colouring, and of paranoid development. It is suggested that individual differences of these kinds, though not determined by the presence or absence of the specific schizophrenic gene, are genetically caused and are to be attributed to the action of modifying genes. In addition, the material suggests that various features of the illness are correlated among themselves. Thus particular symptoms are found to be associated with an early and a late onset, with pyknic and with asthenic habitus, with a psychosis that runs a remitting, a chronic or a dementing course.
[1] For the system by which these numbers are worked out, see II. A. Fisher, "Statistical Methods for Research Workers", 9th edition, London, 1944. P = measure of how probable it is that the tendency seen in the numbers is to be accounted for as a chance effect.