SummaryIn the latter part of a large hospital case-control study of the relationship of type of cigarette smoked to risk of various smoking-associated diseases, patients answered questions on the smoking habits of their first spouse and on the extent of passive smoke exposure at home, at work, during travel and during leisure. In an extension of this study an attempt was made to obtain smoking habit data directly from the spouses of all lifelong non-smoking lung cancer cases and of two lifelong non-smoking matched controls for each case. The attempt was made regardless of whether the patients had answered passive smoking questions in hospital or not.Amongst lifelong non-smokers, passive smoking was not associated with any significant increase in risk of lung cancer, chronic bronchitis, ischaemic heart disease or stroke in any analysis.Limitations of past studies on passive smoking are discussed and the need for further research underlined. From all the available evidence, it appears that any effect of passive smoke on risk of any of the major diseases that have been associated with active smoking is at most small, and may not exist at all.

A double-blind controlled trial of prophylactic factor VIII therapy has been carried out on nine severe haemophiliacs at the Lord Mayor Treloar College. Infusions were given once weekly and calculated to give a post-infusion plasma concentration of at least 0.25 I.U./ml of factor VIII. This regime reduced the overall bleeding frequency by 15%. The bleeding frequency in the first 3 days post-infusion was reduced by 66%. A moderate overall reduction in morbidity was also achieved. It is calculated that to reduce the incidence of bleeding in severe haemophiliacs by 15% would require a 73% increased usage of therapeutic materials. More than twice this amount of material is likely to be needed to reduce the bleeding frequency of the same group by 66%.

A mortality study of workers employed for at least one year between 1 January 1950 and 31 December 1975 at eight oil refineries in Britain has been carried out. Over 99% of the population were successfully traced to determine theirvital status at 31 December 1975. The mortalityobservedin the study population was compared with that which would be expected from the mortality rates for the all male population of England and Wales, and Scotland, with adjustment for regional variation in mortality for the English and Welsh refineries. The overall mortality observed was considerably lower than that expected on this basis, as was the mortality from heart disease, stroke, bronchitis, and pneumonia. The observed number of deaths from all neoplasms was also very much less than expected, a result almost entirely due to a large deficit of observed deaths from lung cancer. Raised mortality patterns were found in several refineries for cancers of the oesophagus, stomach, intestines, and rectum, although no location was consistently high for all these causes of death. Different yearof-entry cohorts and job groups were also affected. In general, mortality from these causes increased as length of service and interval from starting work increased. There were also significantly more observed deaths than expected from cancer of the nasal cavities and sinus, and melanoma. Further work is required to ascertain whether these are due to an occupational factor and, if so, to identify the physical or chemical nature of the risk.The aim of this study was to examine the patterns of mortality by cause for employees in oil refineries in Britain. In addition to providing background information on the health of the workers this could draw attention to specific problem areas that might require further investigation. A full report of the study has been completed.' The purpose of this article is to describe the method of data collection and analysis, to give the full results in detail, and to discuss some of the problems encountered in the approach selected and in the interpretation of results from this type of study.Previous studies on the hazards associated with oilThe carcinogenic properties of mineral oil have been well documented. Both experimental2 3 and occupational studies have been reported in the shale oil industry a,nd in the textile industry.4-7 Particular issues in the petroleum industry that haveRequests for reprints to L R.

SUMMARY.-A retrospective detailed study of 272 cases of breast carcinoma treated by radical mastectomy was published by Hamlin (1968). An (McWhirter, 1957), delay (Registrar General, 1967), clinical stage (Paterson, Tod and Russell, 1939), axillary node involvement (Myers, Axtell and Zelen, 1966), size of axillary nodes (Fisher, Slack and Bross, 1969), internal mammary node involvement (Handley and Thackray, 1954), malignancy grading of tumour (Bloom, 1950), host response (Hamlin, 1968), hormone balance (Hayward, Bulbrook and Greenwood, 1961), serum cholesterol levels (Juret, Aubert and de Kaou6l, 1967), treatment (Bloom, Richardson and Harries, 1962).It is clear from this list that many of these factors are interdependent. Age and menopausal status are related and both are probably related to hormone balance and perhaps to serum cholesterol levels. The size of the tumour, the delay in seeking advice, the clinical stage, and the node involvement are almost * Present address:

A mortality study of workers employed for at least one year between 1 January 1950 and 31 December 1975 at oil distribution centres from three oil companies in Britain has been carried out. Ninety nine per cent of the population were successfully traced to determine their vital status at 31 December 1975. The mortality observed in the study population was compared with that which would be expected from the mortality rates for all the male population of England and Wales. The overall mortality observed was considerably lower than expected on this basis as was the mortality from stroke, hypertensive disease, bronchitis, and pneumonia. The observed number of deaths from all neoplasms was also much less than expected as were the observed deaths from lung cancer. The observed deaths from ischaemic heart disease approximately equalled those expected overall and in each of the companies, however, and there was no evidence of a "healthy worker effect" for this disease group. The ratio of observed over expected deaths from ischaemic heart disease tended to decrease with increasing age at death, and for most of the job groups overall, the observed and expected deaths were about the same. Raised mortality patterns from ischaemic heart disease were found in several subgroups of the population of one company. Mortality from myelofibrosis and diseases of the lymphatic and haematopoietic tissue was slightly raised overall. Only myelofibrosis showed an overall excess but raised mortality was found in subgroups of the population defined by company, job, and length of service in several of the other neoplasms making up this disease group. The numbers of deaths from these causes were all small, making it difficult to exclude chance effects. Further work would be required to ascertain whether these results are due to an occupational factor and if so to identify the physical or chemical nature of the risk.