936 uals treated with
Hypothesis
drugs which
D. F. HORROBIN Clinical Research Institute and University de Montreal, 110 Pine Avenue West, Montreal, Quebec H2W 1R7, Canada
Summary
Evidence that
schizophrenia may be a prostaglandin deficiency disease comes from three main sources: (1) all effective antischizophrenic drugs stimulate prolactin secretion and prolactin is a potent stimulator of prostaglandin synthesis; (2) schizophrenics are resistant to pain and inflammation and are free of rheumatoid arthritis and there is increas-
ing evidence that prostaglandins play important roles in pain, inflammation, and rheumatoid arthritis; (3) high doses of drugs recently shown to be prostaglandin antagonists cause schizophrenia-like syndromes. The hypothesis is not necessarily inconsistent with current transmitter theories of schizophrenia since prostaglandins modify transmitter secretion and action. It does indicate radically new approaches to investigation, treatment, and drug design not suggested by the transmitter concepts.
play a key role in arthritis, in other inflammatory reactions, and in pain. 16 17 If a P.G. deficiency were involved in schizophrenia there should be evidence of reduced inflammatory responses and susceptibility to inflammatory conditions in schizophrenics. Nissen and Spencer" first drew
ALL effective antischizophrenic drugs stimulate prolactin secretion possibly because of a dopamine-blocking effect. Stimulation of such secretion is probably a better indicator of effective antipsychotic action than other more conventional tests of dopamine-blocking activity.1 I have suggested that stimulation of prolactin secretion may be an essential part of the action of current anti-
schizophrenic drugs.2-4 Prolactin seems to act in both the mammary gland and in vascular walls by stimulating the secretion of prostaglandins (P.G.S).S-7 There is indirect evidence that its renal effects may also be mediated by p.G.S.8 It is therefore possible that stimulation of P.G. synthesis may be a general mechanism of prolactin action. p.G.s can modify both the presynaptic release of transmitter substances and their post-synaptic actions.9-12 Very low concentrations of P.G.S seem to consistently potentiate catecholamine effects whereas higher concentrations may be either excitatory or inhibitory.’3 Some P.G.S in certain systems have "bell-shaped" dose response curves and the effects of both an excess and a deficiency of these P.G.s are therefore similar.14 Injection of P.G.S into animals has produced catatonic states1S raising the possibility that schizophrenia may be a disease in which P.G. levels are excessive. However, because of the bell-shaped response curves the consequences of a P.G. excess may be similar to those of P.G. deficiency. I suggest that schizophrenia is a P.G. deficiency disease. This proposal is based on the prolactin and therefore P.G.-stimulating properties of antischizophrenic drugs, on the resistance to pain, inflammation, and inflammatory diseases shown by schizophrenics and on the occurrence of schizophrenia-like psychoses in individ-
P.G.S
the fact that rheumatoidarthritis and schizophrenia seem to be mutually exclusive conditions. An Australian study has confirmed this; in 301 women with schizophrenia not one with clinical or radiological evidence of rheumatoid arthritis could be found." Schizophrenics seem to be resistant to the inflammatory consequences of histamine injections2O and to systemic effects of injections of typhoid vaccine.21They have an increased threshold to painful stimuli22 and diminished blood-pressure and pupillary responses to pain.2324 They may be very resistant to cardiovascular shock resulting from surgery or burns.25 The vasodepressor actions of P.G.s may be a cause of shock and there is evidence that schizophrenics have increased vasoconstrictor reactivity26 which could be caused by low P.G. levels. There is therefore a good deal of indirect evidence to suggest that schizophrenics may not be able to attention
to
synthesiseP.G.s normally. DRUGS, HORMONES
INTRODUCTION
antagonsist
AND SCHIZOPHRENIA
There is good evidence that
DEFICIENCY DISEASE
P.G.
actions. INFLAMMATION, PAIN,
SCHIZOPHRENIA AS A PROSTAGLANDIN
have
AND SCHIZOPHRENIA
We have shown that the antimalarials chloroquine and quinine are P.G. antagonists.2’ 2a We have found that quinacrine has a similar effect (unpublished work). If schizophrenia is a P.G.-deficiency disease, treatment with antimalarials might cause a psychotic syndrome in normal individuals or precipitate relapse in those with schizophrenic tendencies. The schizophrenic-like state which can be induced by quinacrine used to be well known29 and chloroquine has also been reported to produce schizophrenic features .30 31 Suppression of prolactin secretion should also cause a reduction in P.G. synthesis and a precipitation of schizophrenia in susceptible individuals. Levodopa, which suppresses prolactin secretion, has been reported to do this.32 There is good evidence that cortisol can block the effect of prolactin and other substances on P.G. secretion.33-36 Steroid-induced schizophrenia-like syndromes could be. related to this. Thyroid deficiency may also perhaps reduce P.G. secretion36 and this could be a contributing factor to myxoedema madness. THE DEFECT IN SCHIZOPHRENIA
The genetic evidence suggests that schizophrenia could be an inborn error of metabolism. If schizophrenia is caused by a deficiency of a P.G. or a substance such as an endoperoxide or a thromboxane related to the P.G.S then several different enzyme defects could produce a similar end result. The P.G.s are manufactured from the essential fatty acids stored as complexes with the membrane phospholipids. The precursors are split from the membrane phospholipids by phospholipase A which is probably the enzyme regulated by prolactin and cortisol. 37 Defective enzyme systems for converting the essential fatty acids
937 to arachidonic and dihomo-y-linolenic immediate acid, the precursors of the 2 and 1 series of or defective P.G.S, phospholipase could lead to P.G. defiAn abnormality of the cyclo-oxygenase which ciency. the converts precursors to the P.G. endoperoxides would a general deficiency of P.G.s and p.G.-like lead to also in the enzymes after the endoperoxDefects materials. to abnormal balances between endoperides would lead oxides, P.G.s, and thromboxanes.38-4o There is thus a clear possibility that a number of different enzyme abnormalities could lead to the same final result. There is also a likelihood, because of abnormalities of essential fattyacids intake, or of vitamin cofactor requirements by the various enzymes, that dietary changes could alleviate or aggravate schizophrenia. Three precise predictions follow from the hypothesis which should allow it to be tested :1. Arachidonic acid, possibly the most important P.G. precursor, should alleviate schizophrenia. It is essential to note that it must not be administered with other polyunsaturated fatty acids which can block its conversion to P.G.s.4l 42 2. A factor limiting the effectiveness of many of the prolactin-secreting anti schizophrenic drugs may be their frequent possession of membrane-stabilising actions. Membrane-stabilising actions may be indicators of P.G. antagonist effects.28 The drugs are therefore stimulating P.G. synthesis via prolactin but antagonising the actions of the P.G.S produced. If the hypothesis is correct a good antischizophrenic drug should be a powerful stimulator of prolactin secretion but, should not have membrane-
found in the diet
stabilising properties. 3. Drugs which inhibit P.G. synthesis or antagonise P.G. action should, if given in sufficiently high doses, produce schizophrenia-like syndromes. Since the P.G.s appear to modulate synaptic function this hypothesis is not necessarily inconsistent with other current hypotheses, for example, the one relating to dopamine. It does, however, point to radically new experimental and therapeutic approaches not suggested by the transmitter theories. I thank Dr T. J. Crow for telling me about the relationship between and rheumatoid arthritis. This work was supported by the Canadian and Quebec Medical Research Councils, the National Cancer Institute, the Muscular Dystrophy Association of Canada, and the Cancer Research Society of Montreal.
schizophrenia
REFERENCES
J. A., Smalstig, E. B., Sawyer, B. D. Psychopharmacologia, 1974, 40, 123. 2. Horrobin, D. F. Prolactin: Physiology and Clinical Significance; p. 158. Lancaster, 1973. 3 Horrobin, D. F., Karmali, R. A., Mtabaji, J. P., Manku, M. S., Nassar, B. A. Postgrad. med. J. 1975, 452, suppl. 2, p. 80. 4 Horrobin, D. F. Prolactin 1976; p. 158. Montreal, 1976. 5. Rillema, J. A. Nature, 1975, 253, 466. 6. Horrobin, D. F., Manku, M. S., Karmali, R. A., Nassar, B. A., Greaves, M. V. Lancet, 1974, ii, 1154. 7. Mtabaji, J. P., Manku, M. S., Horrobin, D. F. Can. J. Physiol. Pharmac. 1976, 54, 357. 8. Mtabaji, J. P., Robinson, C. J., Manku, M. S., Cronin, D., Horrobin, D. F. J. Endocr. (in the press). 9. Hedqvist, P. in Prostaglandin Synthetase Inhibitors (edited by H. J. Robinson and J. R. Vane); p. 303. New York, 1975.
1. Clemens,
10. Hedqvist, P. in Advances in Prostaglandin and Thromboxane Research (edited by B. Samuelsson and R. Paoletti); p. 357. New York, 1976. 11. Horrobin, D. F., Manku, M. S., Karmali, R. A., Nassar, B. A., Davies, P. A. Nature, 1974, 250, 425. 12. Manku, M. S., Horrobin, D. F. Prostaglandins, 1976, 12, 369. 13. Manku, M. S., Mtabaji, J. P., Horrobin, D. F. ibid. (in the press). 14. Horrobin, D. F., Manku, M. S. Med. Hypotheses, (in the press).
15.
Coceani, F., Pace-Asciak, C. R. in Prostaglandins: Physiological, Pharmacological and Pathological Aspects (edited by S. M. M. Karim); p. 1. Lancaster, 1976. M. W. in
Prostaglandins: Physiological, Pharmacological and Pathological Aspects (edited by S. M. M. Karim); p. 293. Lancaster,
16. Greaves,
1976. 17. Robinson, D. R., McGuire, M. B., Levine, L. Ann. N.Y. Acad. Sci. 1975,
256, 318. 18. Nissen, H. A., Spencer, K. A. New Engl. J. Med. 1936, 214, 576. 19. Mellsop, G. W., Koadlow, L., Syme, J., Whittingham, S. Aust. N.Z. J. Med. 1974, 4, 247. 20. Matthysse, S. A. Rev. Med. 1975, 26, 551. 21. Loumos, S. A.M.A. Archs Neurol. Psychiat. 1952, 68, 69. 22. Malmo, R. B. J. Personality, 1951, 19, 359. 23. Earle, A., Earle. B. V. J. nerv. ment. Dis. 1955, 121, 132. 24. May, P. R. A. J. ment. Sci. 1948, 94, 89. 25. Huxley, J., Mayr, E., Osmond, H., Hoffer, A. Nature, 1964, 204, 220. 26. Shattock, M. J. ment. Sci. 1950, 96, 32. 27. Manku, M. S., Horrobin, D. F. Prostaglandins, 1976, 12, 789. 28. Manku, M. S., Horrobin, D. F. Lancet, 1976, ii, 1115. 29. Greiber, M. F. Am. J. Psychiat. 1947-8, 104, 306. 30. Burrell, Z. L., Marinez, A. C. New Engl. J. Med. 1958, 258, 798. 31. Bomb, B. S., Bedi, H. K., Bhatnagar, L. K. Trans. R. Soc. trop. Med. Hyg.
1976, 69, 523. Yaryura-Tobias, J. A., Diamond, B., Merlis, S. Curr. ther. Res. clin. Exp. 1970, 12, 528. 33. Horrobin, D. F., Mtabaji, J. P., Manku, M. S. Endocrinology, 1976, 92,
32.
406. 34.
Horrobin,
D.
F., Mtabaji, J. P., Manku, M. S. Med. Hypotheses, 1976, 2,
219.
Kantrowitz, F., Robinson, D. R., McGuire, M. B., Levine, L. Nature, 1975, 258, 737. 36. Tashjian, A. H., Voelkel, E. F., McDonough, J., Levine, L. ibid. 739. 37. Rillema, J. A., Anderson, L. D. Biochim. biophys. Acta, 1976, 428, 819. 38. Malmsten, C., Hamberg, M., Svensson, J., Samuelsson, B. Proc. Natn. Acad. Sci. U.S.A 1975, 72, 1446. 39. Needleman, P., Kulkarni, P. S., Raz, A. Science, 1977, 195, 409. 40. Moncada, S., Needleman, P., Bunting, S., Vane, J. R. Prostaglandins, 1976, 12, 323. 41. Pace-Asciak, C., Wolfe, L. S. Biochim. biophys. Acta, 1968, 152, 784.
35.
Reviews of Books Hypertension
.
A Policy Perspective. MILTON C. WEINSTEIN and WILLIAM B. STASON. Cambridge, Massachusetts and London: Harvard University Press. 1977. Pp. 243.$15, 11.25.
SUBSPECIALISATIONS in medicine and society quickly reach points of diminishing returns because the attendant proliferation of technical languages creates barriers to the widespread understanding and application of new findings. To counter this problem "translators" have arisen--clinical pharmacologists, clinical epidemiologists, clinical psychologists, and the
like. Medicine has
not been so fortunate with administrators, however; perhaps the languages are too disparate or perhaps the purposes are truly antithetical. This book is a substantial
step towards reducing the communication barrier between clinicians and health-policy planners. It is one of the first products of the Harvard Center for the Analysis of Health Practices, a multidisciplinary "think tank" of economists, physicians, epidemiologists, ethicists, statisticians and others. The authors, a policy analyst and a cardiologist of epidemiological orientation, have mastered each other’s language and have broken new ground to develop a mathematical model for decision-making in the allocation of health-care resources for the detection and treatment of hypertension. Beginning with the principle that society’s interests will be served best by achieving maximum benefit for minimum expenditure, they concisely but clearly describe the tools of cost-effectiveness analysis and then consider the relevant management issues of community screening, expected benefits of treatment, and erosion of benefits by side-effects and patient non-compliance. The data on these issues are put into the model and conclusions about resource allocation emerge. For example, if one must work within budgetary restraints, it is possible to improve the proportion of controlled hypertensives in the community be spend-
ing relatively more on follow-up (including compliance manoeuvres) than on screening. The model is robust: fed accu-
Hypothesis
drugs which
D. F. HORROBIN Clinical Research Institute and University de Montreal, 110 Pine Avenue West, Montreal, Quebec H2W 1R7, Canada
Summary
Evidence that
schizophrenia may be a prostaglandin deficiency disease comes from three main sources: (1) all effective antischizophrenic drugs stimulate prolactin secretion and prolactin is a potent stimulator of prostaglandin synthesis; (2) schizophrenics are resistant to pain and inflammation and are free of rheumatoid arthritis and there is increas-
ing evidence that prostaglandins play important roles in pain, inflammation, and rheumatoid arthritis; (3) high doses of drugs recently shown to be prostaglandin antagonists cause schizophrenia-like syndromes. The hypothesis is not necessarily inconsistent with current transmitter theories of schizophrenia since prostaglandins modify transmitter secretion and action. It does indicate radically new approaches to investigation, treatment, and drug design not suggested by the transmitter concepts.
play a key role in arthritis, in other inflammatory reactions, and in pain. 16 17 If a P.G. deficiency were involved in schizophrenia there should be evidence of reduced inflammatory responses and susceptibility to inflammatory conditions in schizophrenics. Nissen and Spencer" first drew
ALL effective antischizophrenic drugs stimulate prolactin secretion possibly because of a dopamine-blocking effect. Stimulation of such secretion is probably a better indicator of effective antipsychotic action than other more conventional tests of dopamine-blocking activity.1 I have suggested that stimulation of prolactin secretion may be an essential part of the action of current anti-
schizophrenic drugs.2-4 Prolactin seems to act in both the mammary gland and in vascular walls by stimulating the secretion of prostaglandins (P.G.S).S-7 There is indirect evidence that its renal effects may also be mediated by p.G.S.8 It is therefore possible that stimulation of P.G. synthesis may be a general mechanism of prolactin action. p.G.s can modify both the presynaptic release of transmitter substances and their post-synaptic actions.9-12 Very low concentrations of P.G.S seem to consistently potentiate catecholamine effects whereas higher concentrations may be either excitatory or inhibitory.’3 Some P.G.S in certain systems have "bell-shaped" dose response curves and the effects of both an excess and a deficiency of these P.G.s are therefore similar.14 Injection of P.G.S into animals has produced catatonic states1S raising the possibility that schizophrenia may be a disease in which P.G. levels are excessive. However, because of the bell-shaped response curves the consequences of a P.G. excess may be similar to those of P.G. deficiency. I suggest that schizophrenia is a P.G. deficiency disease. This proposal is based on the prolactin and therefore P.G.-stimulating properties of antischizophrenic drugs, on the resistance to pain, inflammation, and inflammatory diseases shown by schizophrenics and on the occurrence of schizophrenia-like psychoses in individ-
P.G.S
the fact that rheumatoidarthritis and schizophrenia seem to be mutually exclusive conditions. An Australian study has confirmed this; in 301 women with schizophrenia not one with clinical or radiological evidence of rheumatoid arthritis could be found." Schizophrenics seem to be resistant to the inflammatory consequences of histamine injections2O and to systemic effects of injections of typhoid vaccine.21They have an increased threshold to painful stimuli22 and diminished blood-pressure and pupillary responses to pain.2324 They may be very resistant to cardiovascular shock resulting from surgery or burns.25 The vasodepressor actions of P.G.s may be a cause of shock and there is evidence that schizophrenics have increased vasoconstrictor reactivity26 which could be caused by low P.G. levels. There is therefore a good deal of indirect evidence to suggest that schizophrenics may not be able to attention
to
synthesiseP.G.s normally. DRUGS, HORMONES
INTRODUCTION
antagonsist
AND SCHIZOPHRENIA
There is good evidence that
DEFICIENCY DISEASE
P.G.
actions. INFLAMMATION, PAIN,
SCHIZOPHRENIA AS A PROSTAGLANDIN
have
AND SCHIZOPHRENIA
We have shown that the antimalarials chloroquine and quinine are P.G. antagonists.2’ 2a We have found that quinacrine has a similar effect (unpublished work). If schizophrenia is a P.G.-deficiency disease, treatment with antimalarials might cause a psychotic syndrome in normal individuals or precipitate relapse in those with schizophrenic tendencies. The schizophrenic-like state which can be induced by quinacrine used to be well known29 and chloroquine has also been reported to produce schizophrenic features .30 31 Suppression of prolactin secretion should also cause a reduction in P.G. synthesis and a precipitation of schizophrenia in susceptible individuals. Levodopa, which suppresses prolactin secretion, has been reported to do this.32 There is good evidence that cortisol can block the effect of prolactin and other substances on P.G. secretion.33-36 Steroid-induced schizophrenia-like syndromes could be. related to this. Thyroid deficiency may also perhaps reduce P.G. secretion36 and this could be a contributing factor to myxoedema madness. THE DEFECT IN SCHIZOPHRENIA
The genetic evidence suggests that schizophrenia could be an inborn error of metabolism. If schizophrenia is caused by a deficiency of a P.G. or a substance such as an endoperoxide or a thromboxane related to the P.G.S then several different enzyme defects could produce a similar end result. The P.G.s are manufactured from the essential fatty acids stored as complexes with the membrane phospholipids. The precursors are split from the membrane phospholipids by phospholipase A which is probably the enzyme regulated by prolactin and cortisol. 37 Defective enzyme systems for converting the essential fatty acids
937 to arachidonic and dihomo-y-linolenic immediate acid, the precursors of the 2 and 1 series of or defective P.G.S, phospholipase could lead to P.G. defiAn abnormality of the cyclo-oxygenase which ciency. the converts precursors to the P.G. endoperoxides would a general deficiency of P.G.s and p.G.-like lead to also in the enzymes after the endoperoxDefects materials. to abnormal balances between endoperides would lead oxides, P.G.s, and thromboxanes.38-4o There is thus a clear possibility that a number of different enzyme abnormalities could lead to the same final result. There is also a likelihood, because of abnormalities of essential fattyacids intake, or of vitamin cofactor requirements by the various enzymes, that dietary changes could alleviate or aggravate schizophrenia. Three precise predictions follow from the hypothesis which should allow it to be tested :1. Arachidonic acid, possibly the most important P.G. precursor, should alleviate schizophrenia. It is essential to note that it must not be administered with other polyunsaturated fatty acids which can block its conversion to P.G.s.4l 42 2. A factor limiting the effectiveness of many of the prolactin-secreting anti schizophrenic drugs may be their frequent possession of membrane-stabilising actions. Membrane-stabilising actions may be indicators of P.G. antagonist effects.28 The drugs are therefore stimulating P.G. synthesis via prolactin but antagonising the actions of the P.G.S produced. If the hypothesis is correct a good antischizophrenic drug should be a powerful stimulator of prolactin secretion but, should not have membrane-
found in the diet
stabilising properties. 3. Drugs which inhibit P.G. synthesis or antagonise P.G. action should, if given in sufficiently high doses, produce schizophrenia-like syndromes. Since the P.G.s appear to modulate synaptic function this hypothesis is not necessarily inconsistent with other current hypotheses, for example, the one relating to dopamine. It does, however, point to radically new experimental and therapeutic approaches not suggested by the transmitter theories. I thank Dr T. J. Crow for telling me about the relationship between and rheumatoid arthritis. This work was supported by the Canadian and Quebec Medical Research Councils, the National Cancer Institute, the Muscular Dystrophy Association of Canada, and the Cancer Research Society of Montreal.
schizophrenia
REFERENCES
J. A., Smalstig, E. B., Sawyer, B. D. Psychopharmacologia, 1974, 40, 123. 2. Horrobin, D. F. Prolactin: Physiology and Clinical Significance; p. 158. Lancaster, 1973. 3 Horrobin, D. F., Karmali, R. A., Mtabaji, J. P., Manku, M. S., Nassar, B. A. Postgrad. med. J. 1975, 452, suppl. 2, p. 80. 4 Horrobin, D. F. Prolactin 1976; p. 158. Montreal, 1976. 5. Rillema, J. A. Nature, 1975, 253, 466. 6. Horrobin, D. F., Manku, M. S., Karmali, R. A., Nassar, B. A., Greaves, M. V. Lancet, 1974, ii, 1154. 7. Mtabaji, J. P., Manku, M. S., Horrobin, D. F. Can. J. Physiol. Pharmac. 1976, 54, 357. 8. Mtabaji, J. P., Robinson, C. J., Manku, M. S., Cronin, D., Horrobin, D. F. J. Endocr. (in the press). 9. Hedqvist, P. in Prostaglandin Synthetase Inhibitors (edited by H. J. Robinson and J. R. Vane); p. 303. New York, 1975.
1. Clemens,
10. Hedqvist, P. in Advances in Prostaglandin and Thromboxane Research (edited by B. Samuelsson and R. Paoletti); p. 357. New York, 1976. 11. Horrobin, D. F., Manku, M. S., Karmali, R. A., Nassar, B. A., Davies, P. A. Nature, 1974, 250, 425. 12. Manku, M. S., Horrobin, D. F. Prostaglandins, 1976, 12, 369. 13. Manku, M. S., Mtabaji, J. P., Horrobin, D. F. ibid. (in the press). 14. Horrobin, D. F., Manku, M. S. Med. Hypotheses, (in the press).
15.
Coceani, F., Pace-Asciak, C. R. in Prostaglandins: Physiological, Pharmacological and Pathological Aspects (edited by S. M. M. Karim); p. 1. Lancaster, 1976. M. W. in
Prostaglandins: Physiological, Pharmacological and Pathological Aspects (edited by S. M. M. Karim); p. 293. Lancaster,
16. Greaves,
1976. 17. Robinson, D. R., McGuire, M. B., Levine, L. Ann. N.Y. Acad. Sci. 1975,
256, 318. 18. Nissen, H. A., Spencer, K. A. New Engl. J. Med. 1936, 214, 576. 19. Mellsop, G. W., Koadlow, L., Syme, J., Whittingham, S. Aust. N.Z. J. Med. 1974, 4, 247. 20. Matthysse, S. A. Rev. Med. 1975, 26, 551. 21. Loumos, S. A.M.A. Archs Neurol. Psychiat. 1952, 68, 69. 22. Malmo, R. B. J. Personality, 1951, 19, 359. 23. Earle, A., Earle. B. V. J. nerv. ment. Dis. 1955, 121, 132. 24. May, P. R. A. J. ment. Sci. 1948, 94, 89. 25. Huxley, J., Mayr, E., Osmond, H., Hoffer, A. Nature, 1964, 204, 220. 26. Shattock, M. J. ment. Sci. 1950, 96, 32. 27. Manku, M. S., Horrobin, D. F. Prostaglandins, 1976, 12, 789. 28. Manku, M. S., Horrobin, D. F. Lancet, 1976, ii, 1115. 29. Greiber, M. F. Am. J. Psychiat. 1947-8, 104, 306. 30. Burrell, Z. L., Marinez, A. C. New Engl. J. Med. 1958, 258, 798. 31. Bomb, B. S., Bedi, H. K., Bhatnagar, L. K. Trans. R. Soc. trop. Med. Hyg.
1976, 69, 523. Yaryura-Tobias, J. A., Diamond, B., Merlis, S. Curr. ther. Res. clin. Exp. 1970, 12, 528. 33. Horrobin, D. F., Mtabaji, J. P., Manku, M. S. Endocrinology, 1976, 92,
32.
406. 34.
Horrobin,
D.
F., Mtabaji, J. P., Manku, M. S. Med. Hypotheses, 1976, 2,
219.
Kantrowitz, F., Robinson, D. R., McGuire, M. B., Levine, L. Nature, 1975, 258, 737. 36. Tashjian, A. H., Voelkel, E. F., McDonough, J., Levine, L. ibid. 739. 37. Rillema, J. A., Anderson, L. D. Biochim. biophys. Acta, 1976, 428, 819. 38. Malmsten, C., Hamberg, M., Svensson, J., Samuelsson, B. Proc. Natn. Acad. Sci. U.S.A 1975, 72, 1446. 39. Needleman, P., Kulkarni, P. S., Raz, A. Science, 1977, 195, 409. 40. Moncada, S., Needleman, P., Bunting, S., Vane, J. R. Prostaglandins, 1976, 12, 323. 41. Pace-Asciak, C., Wolfe, L. S. Biochim. biophys. Acta, 1968, 152, 784.
35.
Reviews of Books Hypertension
.
A Policy Perspective. MILTON C. WEINSTEIN and WILLIAM B. STASON. Cambridge, Massachusetts and London: Harvard University Press. 1977. Pp. 243.$15, 11.25.
SUBSPECIALISATIONS in medicine and society quickly reach points of diminishing returns because the attendant proliferation of technical languages creates barriers to the widespread understanding and application of new findings. To counter this problem "translators" have arisen--clinical pharmacologists, clinical epidemiologists, clinical psychologists, and the
like. Medicine has
not been so fortunate with administrators, however; perhaps the languages are too disparate or perhaps the purposes are truly antithetical. This book is a substantial
step towards reducing the communication barrier between clinicians and health-policy planners. It is one of the first products of the Harvard Center for the Analysis of Health Practices, a multidisciplinary "think tank" of economists, physicians, epidemiologists, ethicists, statisticians and others. The authors, a policy analyst and a cardiologist of epidemiological orientation, have mastered each other’s language and have broken new ground to develop a mathematical model for decision-making in the allocation of health-care resources for the detection and treatment of hypertension. Beginning with the principle that society’s interests will be served best by achieving maximum benefit for minimum expenditure, they concisely but clearly describe the tools of cost-effectiveness analysis and then consider the relevant management issues of community screening, expected benefits of treatment, and erosion of benefits by side-effects and patient non-compliance. The data on these issues are put into the model and conclusions about resource allocation emerge. For example, if one must work within budgetary restraints, it is possible to improve the proportion of controlled hypertensives in the community be spend-
ing relatively more on follow-up (including compliance manoeuvres) than on screening. The model is robust: fed accu-
