In a 10-year follow-up study of 543 men and 180 women seen in a community survey in 1958 mortality was examined in relation to the 1958 haematological data, and 91% of the survivors were re-examined. Analyses based on the 1958 haemoglobin and packed cell volume estimations suggested that those with values near the mean may subsequently have lower death rates. Higher death rates occurred in those having low serum iron levels. Comparisons of haemoglobin concentrations and of packed cell volumes in 1958 and 1968 show correlation coefficients (r) between 0.30 and 0.60 in various subgroups.

In epidemiology it is essential that a definite physiological or pathological condition should be studied in a defined community. These requirements are sometimes difficult to meet because, although the population to be studied should be representa tive of a large community, the measurements involved usually require elaborate apparatus and clinical examination rooms which are not easily mobile. Hitherto it has been the policy of the Epidemio logical Research Unit (ERU) to study intensively a specific disease in selected but limited communities (Miall and Cochrane, 1956) as in the Rhondda Fach, Vale of Glamorgan, Staveley, Annandale, or Wensleydale, with the belief that although these communities are certain to experience unique genetic, cultural, economic, and climatic conditions yet a pattern will emerge of meaningful associations between pathological conditions and the environ ment. This policy has been a success in the past, but it has always been open to the criticism that the associations that were found were also unique to the selected population and could not be safely applied to the general population. One way of confirming such epidemiological results is to repeat the investigations in a different community with a different environment, as was done for the surveys of blood pressure (Miall, 1960), bronchitis (Higgins, 1960), rheumatoid arthritis (Miall, Ball, and Kellgren, 1958), and anaemia (Kilpatrick, 1961), in which the studies in the mining valley of the Rhondda Fach were repeated in the agricultural community of the Vale of Glamorgan or elsewhere. A second method is to investigate a condition in large communities of over 100,000 persons where a wider range of modern conditions of class, of industry and commerce, and of age and sex can be seen than in a small defined community; this method is being used in the Rhondda Borough and in Cardiff City.

Journal Article Iron Deficiency Anaemia in Adults: Prevalence and Prevention Get access A Jacobs, MD, MC Path, A Jacobs, MD, MC Path Senior Lecturer in Hœmatology Welsh National School of Medicine, Cardiff Search for other works by this author on: Oxford Academic Google Scholar G S Kilpatrick, MD, MRCP (Ed), G S Kilpatrick, MD, MRCP (Ed) Senior Lecturer in Medicine Welsh National School of Medicine, Cardiff Search for other works by this author on: Oxford Academic Google Scholar J L Withey, MB, BS, MC Path J L Withey, MB, BS, MC Path Lecturer in Hœmatology Welsh National School of Medicine, Cardiff Search for other works by this author on: Oxford Academic Google Scholar Postgraduate Medical Journal, Volume 41, Issue 477, July 1965, Pages 418–424, https://doi.org/10.1136/pgmj.41.477.418 Published: 01 July 1965

The serum vitamin B12 concentration was estimated in a random sample of two communities; an industrial and an agricultural one. A total of 890 people were studied. The frequency distribution of serum vitamin B12 concentration shows that the range of values is wide and that the serum vitamin B12 level falls with increasing age. For general purposes the results of this study may be summarised by standardising at age 50 years, using the regression coefficient — 1.254 μμg per ml per year of age, which summarises the gradient with age over the whole age range studied. Thus the adjusted means and standard errors, 255.2 ± 5.45 μμg per ml for the industrial men and 262 ± 10.06 μμg per ml for the women, compare closely with the agricultural community where the figures for the men are 260 ± 9.52 and for the women 272 ± 8.45 μμg per ml.

It is generally accepted that differences in the mean haemoglobin level in the two sexes become apparent after puberty, and this is thought to be due mainly to a gradual rise in the mean level in boys, which continues until about the age of 16 or 17 years (Sunderman, MacFate, MacFayden, Stevenson, and Copeland, 1953).However, in addition to age and (after puberty) to sex, many other factors are likely to be associated with haemoglobin level.These probably include family size, though this may be merely a reflection of differences in diet.It also seems likely that body mass may be a further asso- ciated factor and, if this is true, some, if not all, of the difference in mean haemoglobin between the two sexes may simply reflect differences in body mass.This possibility does not appear to have been examined.The report which follows presents the results of a survey of haemoglobin levels in school children aged 14 years in eleven Cardiff schools, undertaken to determine the distribution of haemoglobin levels and to examine the associations between those levels and sex, family size, height and weight, and (in girls) menstruation.METHOD An attempt was made to see all children, born in

was well known. But these preparations failed for the unexpected reason that the patients developed resistance to the intrinsic factor, and actual antibodies to it were found in their blood.' 2 In an attempt to overcome this difficulty Drs. J. G. Heathcote and F. S. Mooney3' prepared a material of peptide nature from a Streptomyces mutant which they considered had intrinsic-factor activity, and they gave a preparation combining crystalline vitamin B12 and this peptide to patients with pernicious anaemia. They reported good results, but it seemed possible these were due only to the vitamin-BI2 content of the preparation. W. B. Castle,3 for instance, pointed out that to establish the claim that the peptide had intrinsic-factor activity Heathcote and Mooney ought to have shown that it would produce a response in a patient with untreated pernicious anaemia when a

When it was suggested that pernicious anaemia could be successfully treated by an oral preparation of vitamin B52 as a peptide complex (Heathcote and Mooney, 1958) it was decided to treat a group of patients suffering from pernicious anaemia with this preparation. We decided to regard a response as satisfactory if the treatment was able to maintain not only the haemoglobin level but also the serum-vitamin-B12 concentration at a satisfactory level.

The rheumatoid pneumoconiosis syndrome was originally described as a characteristic chest radio- graphic appearance in coal-workers associated with clinical rheumatoid arthritis (Caplan, 1953).This association was later confirmed as real and not fortuitous by Miall, Caplan, Cochrane, Kilpatrick, and Oldham (1953) using an unselected population.The original communication defined the radio- graphic picture as one showing multiple, well- defined round opacities 0.5 to 5 cm. in diameter distributed throughout the lung field.We will call this the classical appearance.It was noted that cavitation and calcification of the opacities was frequent.There was a tendency for the opacities to appear suddenly before, coincident with, or after the onset of arthritis.Reference was also made to the fact that only a few round opacities may be present, and in many instances the lesions become confluent and may be indistinguishable radio- graphically from progressive massive fibrosis.It was suggested that more cases would be recognized if it were realized that there is often a mixed radio- graphic picture of round opacities and opacities indistinguishable from progressive massive fibrosis and tuberculosis.Subsequent observations on more than 550 cases of the syndrome have confirmed in the main the original radiographic description and in particular emphasized the high incidence of cavitation as a characteristic feature.Additional experience among coal-workers has suggested an association of clinical rheumatoid arthritis with discrete nodular opacities 0.3 to 1 cm. in diameter, i.e., somewhat smaller than was originally found, and classifiable under the International