High-contact paternal occupations, infection and childhood leukaemia: five studies of unusual population-mixing of adults

Abstract

Summary The hypothesis has been tested that, among excesses of childhood leukaemia associated with extreme population-mixing, the incidence is higher for the children of men in occupations involving contact with many individuals (particularly children), as noted in certain childhood infections. Data on childhood leukaemia were examined from five previous studies of the author in which significant excesses had been found associated with population-mixing involving adults. Occupational titles were categorized according to the estimated level of work contacts as medium, high, very high or indeterminate. Occupations involving frequent contact with children were categorized as having a very high contact level given the high frequency of exposure to the infection postulated as underlying childhood leukaemia. There was a significant positive trend (P < 0.001) in childhood leukaemia risk at ages 0-14 years across the occupational contact categories from the reference group (comprising the medium and low plus indeterminate categories) through high to very high (i.e. high-child) contact categories in the combined data from the author's five studies of adult population-mixing; this significant trend also applied at ages 0-4 (P < 0.001) and 5-14 (P < 0.01) years. The excess in the high category was mainly because of paternal occupations connected with the construction industry and transport, suggesting a broader definition of the 'very high' contact category. No sign of these excesses was found in a limited examination of the question outside areas of population-mixing using mortality data for childhood leukaemia in the general population of England and Wales. The findings represent the first individual-based support for infection underlying childhood leukaemia that is promoted by population-mixing, as well as further support for the role of adults in transmission of the infection. Keywords: childhood leukaemia; infection; paternal occupationThe hypothesis of an infective basis in childhood leukaemia is strongly supported by the findings in studies of areas (mainly rural) affected by unusual population-mixing. Such situations would be conducive to mini-epidemics of an underlying infection to which leukaemia is an uncommon response and, in a large series of examples studied, significant excesses of childhood leukaemia have been found (Kinlen, 1988;Kinlen et al, 1990Kinlen et al, , 1993aKinlen et al, , 1995Langford, 1991;Kinlen and John, 1994;Petridou et al, 1996;Stiller and Boyle, 1996;Alexander et al, 1997). These community-based findings make a persuasive case for an infective basis for childhood leukaemia. In the absence of the demonstration of evidence of different exposures to the causative agent (here impossible), other case-control differences would be valuable as additional support, although it may be noted that such evidence is often lacking in illnesses that are known to be uncommon responses to infections. The design of the population-mixing studies necessarily means that data amenable to case-control comparisons ...

A case–control study of childhood leukaemia and paternal occupational contact level in rural Sweden

“…In Sweden, children start primary school later than in many countries, at age 7 years, but earlier attendance in the smaller mixing units of kindergartens (from age 3 or 4 years) is common, though probably less so in rural areas. Against this background and the positive findings of two previous studies of rural Britain (Kinlen, 1997;Kinlen and Bramald, 2001), the results of the present study in Sweden are noteworthy: specifically, the significant relative excess of leukaemia in rural areas among the children aged 0 -4 years with fathers in high contact occupations, as well as the significant positive trend in these areas across categories of increasing contact level.…”

“…These were allocated as far as was possible to the contact categories used in our previous studies (Kinlen, 1997;Kinlen and Bramald, 2001) in which a much larger number of different occupational titles were available for allocation to the categories by six independent advisors. Two separate (but partly overlapping) classifications were applied, as in the earlier studies.…”

“…'High contact' paternal occupations have already been associated with an increased incidence of childhood cytomegalovirus infection and paralytic poliomyelitis. An examination of this question within the excesses of childhood leukaemia associated with extreme rural population mixing found a greater risk among the children of men in high contact jobs (Kinlen, 1997). Also, a recent case -control study found a significant positive trend in risk among pre-school children across three levels of increasing paternal occupational contacts in rural Scotland (Kinlen and Bramald, 2001) with each of the two classifications of contact level developed in the 1997 study.…”

Paternal occupational contact level and childhood leukaemia in rural Scotland: a case–control study

“…'High contact' occupations have already been associated with an increased incidence of childhood cytomegalovirus infection and paralytic poliomyelitis. Furthermore, in situations of extreme rural population mixing which had produced excesses of childhood leukaemia, risk was greater among the young children of those whose work involved many community contacts (Kinlen, 1997). No such increases have been found in mortality studies of the question in the (largely urban) general population of England and Wales (Kinlen, 1997;Fear et al, 1999).…”

“…Furthermore, in situations of extreme rural population mixing which had produced excesses of childhood leukaemia, risk was greater among the young children of those whose work involved many community contacts (Kinlen, 1997). No such increases have been found in mortality studies of the question in the (largely urban) general population of England and Wales (Kinlen, 1997;Fear et al, 1999). As an urban population would be expected to have the raised levels of immunity to infective agents often associated with high population density, we have considered the question still open as to whether high levels of paternal occupational contacts have effects on the incidence of childhood leukaemia in young children in a very rural area, where the prevalence of susceptible individuals would be unusually high.…”