Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Has equity been achieved?
- Dr Arash Rashidian (11 September 2002)
-
Corrections to paper
- Sonia Saxena, Professor Azeem Majeed (25 June 2003)
-
The risk of redundant publication when using archived data
- James Y Nazroo, Harriet Becher, Yvonne Kelly, Anne McMunn (16 July 2003)
|
|
|||
|
Dr Arash Rashidian, Health Services Research Scholar Department of Health Sciences, University of York, York, YO10 5DD
Send response to journal:
|
Dear Sir or Madam, Although Saxena et al (1) paper is a highly readable paper, there is a fundamental concern. They have surveyed self (or parents') reported health status of children. Therefore, their conclusion of "children's health status ... did not vary significantly by social class, which implies that equity in this area has been partially achieved" is misleading. Clearly, what they have found is 'equity' in reported health status and no more conclusions can be drawn out of this data. I would like to attract authors attention to Sen's editorial (2). Sen has clearly shown that in Indian states with far worse state of health, literacy and wealth (and therefore lower socioeconomic status) the reported morbidity is significantly lower than states with higher levels of literacy, wealth and life expectancy. He writes: "the lowness of reported morbidity runs almost fully in the opposite direction of life expectancy, in interstate comparisons". Another point that can be added to their discussion is around the lower use of outpatient clinics by Asian children. I would like to add that many factors other than general practitioners' referral behaviour could have influenced this finding (3). Ethnic minority families may live in over-stretched areas with limited outpatient capacities. It is also interesting to study the ethnic background of GPs' who work in those areas to see whether it can help explaining this finding. It is also unclear how they have derived conclusions about 'Irish' group, while respondents had chosen their ethnic background from a variety of ethnic categories including 'white' and 'other' as well as non-white categories. It is likely that many Irish families have introduced their ethnic background as 'white', making it difficult to re-group them, as it has been mentioned by authors. References: 1. Saxena S, Eliahoo J and Majeed A. Socioeconomic and ethnic group differences in self reported health status and use of health services by children and young people in England: cross sectional study. 2002; 325: 520-523 2. Sen A. Health: perception versus observation. 2002; 324: 860-861. 3. Ethnic differences in invasive management of coronary disease: prospective cohort study of patients undergoing angiography. 2002; 324: 511-516. |
|||
|
|
|||
|
Sonia Saxena, lecturer in primary care Research and development directorate, 112 Hampstead Road London NW1 2LT, Professor Azeem Majeed
Send response to journal:
|
Dear editor, Saxena S, Eliahoo J, Majeed, A. Socio-economic and ethnic group differences in health status and use of health services by children and young people in England: cross-sectional study BMJ vol 325 7.9.2002 2002 Because of editing changes at the proof stages in this paper, the Methods section gives the erroneous impression that the authors were responsible for the data collection. Unfortunately, this error was not corrected when the proofs were returned, The Health Survey for England was actually carried out by the Joint Health Surveys Unit at University College London and the National Centre for Social Research. We used data supplied by the Essex Data Archive for our analysis. Due to an authors error, the first part of reference 11 is also incorrect and should read: Office for National Statistics. Health Survey for England: the health of minority ethnic groups '99. London 1999: HMSO, 1999. Dr Sonia Saxena
Competing interests: None declared |
|||
|
|
|||
|
James Y Nazroo, Reader in Sociology Department of Epidemiology and Public Health, UCL, 1-19 Torrington Place, London, WC1E 6BT, Harriet Becher, Yvonne Kelly, Anne McMunn
Send response to journal:
|
Dear Editor The risk of redundant publication when using archived data A recent publication in the BMJ,[1] using data that had previously been analysed by ourselves,[2] flags up a potentially very serious problem regarding the use of archived data for novel research. The paper published findings from secondary analysis of the 1999 Health Survey for England, with a focus on the health of children and young adults from seven ethnic groups (Afro-Caribbean, Indian, Pakistani, Bangladeshi, Chinese, Irish and ‘General population’) and their use of health services. The tables in both the short and longer (available as electronic copy only) versions of the paper showed apparently novel findings on the relationship between ethnicity and: socioeconomic characteristics (class, income and benefit receipt), acute illness, chronic illness, treated asthma, treated infections, major and minor ‘incidents’ (i.e. accidents), visits to GP, hospital outpatients visits and treatment as a hospital inpatient/day patient. At first glance these findings are very important. There have been no data covering this range of ethnic groups or health outcomes in a nationally representative sample apart from those arising from the 1999 HSE. Indeed there are few published studies of the health of ethnic minority children in the UK. However, a reading of the chapter on children’s health[2] in the 1999 HSE report[3] shows that all, but one, of the tables simply duplicate a sub-set of the previously published findings. While the approach to analysis varies a little between the paper and the report, the pattern of findings and the conclusions drawn are no different. Indeed, rather than providing a more detailed level of analysis, the work in the BMJ paper is done at a cruder level than that in the report. For example, the age range covered is 2 to 20 rather than 2 to 16 and data are presented for boys/girls/men/women as one combined group, rather than separately. And, the table that is not a duplication of our earlier work can only be found in the electronic version of the paper (table 6 showing ethnic differences in use of health services adjusted for age, gender, class and reported health status). The publication of such papers does point to the need to take care when using archived data. As members of teams that typically archive the data sets we generate (such as the HSE series, the Fourth National Survey of Ethnic Minorities,[4] the English Longitudinal Study of Ageing[5] and the Millennium cohort[6]), we are great supporters of both the archiving of data and their use by other researchers for secondary analysis. This is often a cost effective way of conducting research, both in terms of financial cost and in terms of burden on those giving up their time to participate in studies. There are well-recognised difficulties with using archive data sets for analyses that were not considered when the originating study was designed. To these we add the risk of redundant publication, which is a risk that can be guarded against. We recommend that analysts using archived data carefully read the outputs of the originating study and that these outputs are fully referenced – we note that the paper referred to here did not adequately reference the 1999 HSE report and that although the authors have recently written to amend this ommission,[7] they still do not reference the chapter that specifically reported the data they used. And we also recommend that both reviewers for journals and editors of journals carefully check the outputs from the originating study before they accept a paper as an original publication. If editors do this, it will also allow them to identify potentially useful reviewers from within the team that generated the data, who will be able to comment on the suitability of the data for the purposes of the secondary analysis. Yours James Nazroo
1. Saxena S, Eliahoo J, Majeed A. Socioeconomic and ethnic group differences in self reported health status and use of health services by children and young people in England: cross sectional study. BMJ 2002;325:520-523. 2. Nazroo J, Becher H, Kelly Y, McMunn A. Children’s Health. In Health Survey for England: The Health of Minority Ethnic Groups, B Erens, P Primatesta, G Prior (eds.), London: The Stationery Office, 2001:345-395. http://www.official-documents.co.uk/document/doh/survey99/hse99-13.htm 3. Erens B, Primatesta P, Prior G. (eds.) Health Survey for England: The Health of Minority Ethnic Groups London: The Stationery Office, 2001. http://www.official-documents.co.uk/document/doh/survey99/hse99.htm. 4. Modood T, Berthoud R, Lakey J, Nazroo J, Smith P, Virdee S, Beishon S. Ethnic Minorities in Britain: Diversity and Disadvantage. London: Policy Studies Institute, 1997. 5. Nazroo J. The English Longitudinal Study of Ageing (ELSA): A new data resource on health, economic position and quality of life for older people. Generations Review 2001;11:14. 6 Kelly, YJ, Marmot, M. The Millennium Cohort Study: Children of the New Century. The Journal of Circumpolar Health Proceedings from the 2nd Conference on Epidemiological Studies in Europe, Oulu, Finland 2002. 7. Saxena S. Corrections to paper. BMJ 25th June 2003. Competing interests: None declared |
|||