My understanding, confirmed by brief review of data on the ONS
website, is that the South Asian population in the UK has a younger age
distribution than the white population. In this case, would an indicator
based on those aged over 35 need further adjustment for the age
distribution beyond 35 in order to examine prescribing? Is it possible
that we are seeing an age effect in the negative correlation...
My understanding, confirmed by brief review of data on the ONS
website, is that the South Asian population in the UK has a younger age
distribution than the white population. In this case, would an indicator
based on those aged over 35 need further adjustment for the age
distribution beyond 35 in order to examine prescribing? Is it possible
that we are seeing an age effect in the negative correlation of
prescribing rates and ethnic minority population proportions?
Conflict of Interest
YHEC Ltd is a contract research company and
carries out projects for both the NHS/DH and for the pharmaceutical
industry. We are not conducting any current research on prescribing rates
for coronary heart disease nor do we see the paper or our comment as
impinging on any of our current work.
The Inverse Care Law, proposed by Julian Tudor Hart in 1971, states
that “the availability of good medical care tends to vary inversely with
the need for it in the population served.”[1]
A number of authors have
now claimed to have found instances of the Inverse Care Law operating in
practice.[2,3] Given the prominence that this ‘law’ has gained in the
health care literature over last thirty...
The Inverse Care Law, proposed by Julian Tudor Hart in 1971, states
that “the availability of good medical care tends to vary inversely with
the need for it in the population served.”[1]
A number of authors have
now claimed to have found instances of the Inverse Care Law operating in
practice.[2,3] Given the prominence that this ‘law’ has gained in the
health care literature over last thirty years, we were surprised to note
that Jordan et al. failed to make reference to it in their recent
article on the relationship between access to services and health.[4]
In this report, access to services was measured as both straight line
distances and car travel time to the nearest GP surgery and hospital as
well as the access domain of the Index of Multiple Deprivation 2000, which
combines measures of straight line distances to the nearest general
practitioner, primary school, food shop and post office. Amongst urban
wards, the authors report a consistent inverse association between
distance to services and both mortality and limiting long term illness
(LLTI) in individuals aged 0-64 years – although this association was
negligible in terms of the relationship between LLTI and distance to
hospitals.
Both premature mortality and LLTI are markers of need for health
services in themselves. In addition, they are both strongly associated
with deprivation in the UK,[5] and therefore a much broader marker of need
for health services. Jordan et al.’s results suggest that areas with
greater need for health services are nearer to and have greater access to,
or concentration of, both health and wider social services. This
conflicts with the Inverse Care Law which would predict that distance to
services should be greater, and therefore access poorer, in areas with
higher levels of need.
Are Jordan et al.’s results evidence that the Inverse Care Law is no
longer operating in the UK? Is it possible that over the last thirty
years, we have managed to redistribute primary care services, in
particular, so equitably that instead of deprivation, poor health and
greater need for services being associated with poor access to services,
they are now associated with greater access to services?
Alternatively,
it is possible that the Inverse Care Law has rarely operated in practice
in the UK in recent times and that ‘evidence’ for it has misinterpreted
the original formulation of the law and focused on use of services, rather
than provision of them.[2]
References
1. Tudor Hart J. The inverse care law. The Lancet 1971(7696):405-412.
2. Webb E. Children and the inverse care law. British Medical Journal
1998;316:1588-91.
3. O'Dea J, Kilham R. The inverse care law is alive and well in
general practice. Medical Journal of Australia 2002;177:78-79.
4. Jordan H, Roderick P, Martin D. The index of multiple deprivation
2000 and accessibility effects on health. Journal of Epidemiology &
Community Health 2004;58:250-257.
5. Acheson D. Report of the independent enquiry into inequalities in
health. London: Stationary Office, 1998.
Dr Leyland published an interesting and thought provoking article
about the variations in the risks of death by region in the United Kingdom.[1] Dr Leyland does not that migration might explain some of the
differences found in the study. Immigrants to a country tend to be
healthier than the base population.[2] In addition, workers tend to be
healthier than non-workers [3] which could influence the...
Dr Leyland published an interesting and thought provoking article
about the variations in the risks of death by region in the United Kingdom.[1] Dr Leyland does not that migration might explain some of the
differences found in the study. Immigrants to a country tend to be
healthier than the base population.[2] In addition, workers tend to be
healthier than non-workers [3] which could influence the composition of
people in high and low employment areas. High unemployment areas, which
tend to be less equal, might be less healthy because of healthy workers
migrating to jobs elsewhere in the UK.
Determining the cause of this differential mortality will be complex
at best and can only be done in a context where migration is controlled
for. Ecological studies, such as this one, are excellent at hypothesis
generation. However, it is important to realize that they cannot go
beyond the generation of a hypothesis to directly explain the association
that they present.[4] The increasing regional inequality in mortality
observed in this study is interesting but without a study that looks at
the individual level it is impossible to determine the correct social
policy response.
References
1. Leyland AH. Increasing inequalities in premature mortality in
Great Britain. J Epidemiol Community Health. 2004;58(4):296-302.
2. Muennig P and Fahs MC Health status and hospital utilization of
recent immigrants to New York City. Prev Med. 2002;35(3):225-31.
3. Li CY and Sung FC. A review of the healthy worker effect in
occupational epidemiology. Occup Med (Lond). 1999;49(4):225-9.
4. Grimes DA and Schulz KF. Descriptive studies: what they can and
cannot do. Lancet. 2002;359(9301):145-9.
Jousilahti and Salomaa appear unhappy with our response to their paper on the social patterning of serum inflammatory markers.
First they feel that we have misinterpreted their findings and
conclusions. We fail to see how. They found (as others have) that social
disadvantage was associated with increased inflammation – as indexed by
markers such as higher fibrinogen - and that this association...
Jousilahti and Salomaa appear unhappy with our response to their paper on the social patterning of serum inflammatory markers.
First they feel that we have misinterpreted their findings and
conclusions. We fail to see how. They found (as others have) that social
disadvantage was associated with increased inflammation – as indexed by
markers such as higher fibrinogen - and that this association was only
partly attenuated by adjustment for smoking, obesity and diagnosed
disease. They concluded that,
“Systemic inflammation is a potential mediator, especially among
young and middle aged men, for the association between socioeconomic
status and cardiovascular disease.”
For inflammation (as indexed by fibrinogen for example) to mediate an
association between social position and heart disease it would have to be
causally related to heart disease risk. Jousilahti and colleagues assume
such a causal relation and explicitly state that,
“Fibrinogen increases the risk of an acute coronary event through its
prothrombotic and rheological effects, and may also play a part in
atherosclerosis formation.”
The alternative explanation is that the association between
fibrinogen and heart disease is not causal. First, reverse causation
probably contributes. Atherosclerosis is an inflammatory condition
therefore established (including sub-clinical) disease would be associated
with increased inflammatory markers. But these markers do not cause
atherosclerosis any more than chest pain causes myocardial ischaemia.
Second, the association is likely to be confounded. Smoking, being obese
and being poor are all associated with increased fibrinogen and
independently with increased risk of heart disease. Within the non-causal
model, fibrinogen is essentially no more than a marker. A marker of
established disease, a marker of health damaging factors and, because
established disease and health damaging factors are socially patterned, a
marker of social position. If fibrinogen is merely a marker its social
patterning gives us no new insights into the social patterning of heart
disease.
Untangling competing theories of this nature is the bread and butter
of epidemiology and as epidemiologists we have an interest in any strategy
that might be helpful. Experimental studies are of course the mainstay
here. Random allocation of individuals to a condition of higher or lower
fibrinogen should ensure that any effects of fibrinogen subsequently seen
are not the result of confounding or reverse causation. Randomised trials
of manipulation of fibrinogen have not shown any important effects on
heart disease. [1] We think this is an important clue that the fibrinogen-
heart disease link is non-causal. Jousilahti and colleagues dismiss such
evidence on the basis that it is “scarce”. This is not the case: there are
a large number of trials of fibrates, which lower fibrinogen levels, and
the effects of these on coronary heart disease (CHD) have been
disappointing. There is some difficulty in interpretation, since the
fibrates also lower cholesterol levels and should produce reductions in
CHD risk through this effect. However, the fact that any observed
reduction in CHD risk is small, and certainly no greater than would be
anticipated by the cholesterol lowering effect, argues against a causal
role for fibrinogen.
Jousilahti and Salomaa appear to base their support for fibrinogen’s
causal candidacy on consistency of the observational evidence of
associations between higher serum fibrinogen and increased risk of heart
disease. Confounding and reverse causation of the type discussed above
would, of course, predict consistent associations in observational data
(because the associations that lead to the introduction of these biases
are often themselves consistent).
But if experimental evidence is difficult to interpret, and
conventional observational evidence consistently compromised by concerns
over reverse causation and confounding, how else can we progress our
understanding? One potential approach is Mendelian randomisation. [2] This
is not a panacea, but many epidemiologists have recognised and are excited
by its potential contribution. [3-10] Mendelian randomisation refers to
the random assortment of genes from parents to offspring that occurs
during gamete formation and conception. This leads to a distribution of
genotype in populations that is not, generally, confounded by lifestyle or
socioeconomic factors, unlike measured phenotypes such as fibrinogen
levels. The association between risk of a disease and a genetic variant
that mimics the biological link between a proposed exposure and disease is
not generally susceptible to the reverse causation or confounding that may
distort interpretations of conventional observational studies. Several
examples where the phenotypic effects of polymorphisms are well documented
provide encouraging evidence of the potential of Mendelian randomisation.
[2,11]
Youngman and colleagues, through their work on fibrinogen and heart
disease, recently exemplified this contribution. [12] In their
observational study, fibrinogen was associated with heart disease in the
usual way – a 0.12g/l increase conferring a relative risk of CHD of 1.20
(1.13-1.26). However serum fibrinogen is also influenced by a polymorphism
in the beta fibrinogen gene, presence of the T allele being associated
with a 0.12g/l increase in serum levels in this population. Presence or
absence of the T allele will not (apart from in very unusual
circumstances) be associated with any of the behavioural or environmental
correlates of fibrinogen that may confound associations with heart
disease, because genotype is randomly assigned at meiosis. Therefore
estimates of effects of this allele on CHD risks are in effect,
unconfounded, “intention to treat” estimates of the effect of the higher
fibrinogen levels associated with presence of the allele. This is the
principle of Mendelian randomisation. In their study, Youngman and
colleagues found that the relative risk of CHD associated with presence of
the T allele (i.e. the unconfounded effect of a 0.12g/l increase in
fibrinogen) was 1.03 (0.96-1.10). They interpreted this as strong evidence
increased fibrinogen did not cause increased risk of heart disease.
Jousilahti and Salomaa apparently disagree, both in relation to the
specific example of fibrinogen and with regard to the general usefulness
of the Mendelian randomisation principle. With respect, we found their
arguments difficult to follow and felt they suggested some
misunderstanding of the issues involved. For example they point out that a
number of genetic and environmental factors influence fibrinogen levels.
Well so they do, but this fact is irrelevant in the present context.
Presence of the beta fibrinogen T allele was probably not the most
important determinant of serum fibrinogen in the Youngman study
population. However it was an identifiable trait consistently associated
with higher fibrinogen levels. Consider another example. Within a
population the use of anti-hypertensive medicines (even if these are
widely and appropriately prescribed) will only make a small contribution
to variance in blood pressure. However this doesn’t mean that anti-
hypertensive drugs will not have a major influence on the sequelae of
elevated blood pressure in this population. The fact that there are many
other environmental and genetic factors contributing to blood pressure
other than anti-hypertensive drugs is irrelevant, as is the wide range of
factors contributing to variance in fibrinogen levels. Indeed, it is the
fact that fibrinogen levels are associated with so many environmental
factors that makes the association with coronary heart disease difficult
to study, due to the considerable potential confounding that can arise.
Jousilahti and Salomaa then appear to confuse arguments around the
shortcomings of association studies in identifying genetic determinants of
complex disease with arguments around the usefulness of Mendelian
randomisation. We have discussed the former (including the specific issues
of replicability and low population attributable risk of any single
factor) at length elsewhere. However these arguments do not apply in the
present context. [13] We [2,11] and others [5-7,14] have also discussed
the potential limitations of Mendelian randomisation in what we feel is a
coherent way. With regard to understanding the causal role of fibrinogen
the large sample sizes required is probably the major issue [15].
What is now termed Mendelian randomisation was first advanced as an
approach to clarifying issues of confounding and reverse causation 20
years ago. [16] As epidemiologists, predominantly working with
observational data we are well aware of these problems and are therefore
interested in the exploration of possible solutions. Others seem to share
this interest, and seem prepared to discuss the potential, and the
potential pitfalls, of this methodology coherently and constructively.
References
1. Meade T, Zuhrie R, Cook C, Cooper J. Bezafibrate in men with lower
extremity arterial disease: randomised controlled trial. BMJ 2002;325:1139
-41
2. Davey Smith G, Ebrahim S. “Mendelian randomisation”: can genetic
epidemiology contribute to understanding environmental determinants of
disease? Int J Epidemiol 2003;32:1-22.
3. Katan MB. Commentary: Mendelian randomization, 18 years on. Int J
Epidemiol 2004;33:10–11.
4. Wheatley K, Gray R. Commentary: Mendelian randomization—an update
on its use to evaluate allogenic stem cell transplantation in leukaemia.
Int J Epidemiol 2004;33:15–17.
5. Brennan P. Commentary: Mendelian randomization and
gene–environment interaction. Int J Epidemiol 2004;33:17–21.
6. Thomas DC, Conti DV. Commentary: The concept of ‘Mendelian
Randomization’. Int J Epidemiol 2004;33:21–25.
7. Tobin MD, Minelli C, Burton PR, Thompson JR. Commentary:
Development of Mendelian randomization: from hypothesis test to ‘Mendelian
deconfounding’. Int J Epidemiol 2004;33:26–29.
8. Keavney B. Commentary: Katan's remarkable foresight: genes and
causality 18 years on. Int J Epidemiol 2004;33:11–14.
9. Clayton D, McKeigue PM. Epidemiological methods for studying genes
and environmental factors in comples diseases. Lancet 2001;358:1356-60.
10. Keavney B. Genetic epidemiological studies of coronary heart
disease. Int J Epidemiol 2002;31:730-6
11. Davey Smith G, Ebrahim S. Mendelian randomisation: prospects,
potentials and limitations. Int J Epidemiol. 2004;33:30-42.
12. Youngman LD, Keavney BD, Palmer A et al. Plasma fibrinogen and
fibrinogen genotypes in 4685 cases of myocardial infarction and in 6002
controls: test of causality by “Mendelian randomisation”. Circulation
2000;102(suppl II):31-32.
13. Colhoun HM, McKeigue P M, Davey Smith G. Problems of reporting
genetic associations with complex outcomes. Lancet 2003;361:865-72.
14. Little J, Khoury MJ. Mendelian randomisation, a new spin or real
progress? Lancet 2003;362:930- 31.
15. Davey Smith G, Harbord R, Ebrahim S. Fibrinogen, C-reactive
protein and coronary heart disease: does Mendelian randomization suggest
the associations are non-causal? Q J Med 2004;97:163-166.
16. Katan MB. Apolipoprotein E isoforms, serum cholesterol, and
cancer. Lancet 1986;i:507–08. Reprinted Int J Epidemiol 2004;33:9.
We write to follow on from our eLetter published in 2003.[1]
As more information becomes available regarding the diagnosis and laboratory testing of SARS, the official number of laboratory confirmed SARS cases in Taiwan during the 2003 outbreak has been officially determined to be 346.[2]
The duration of the outbreak by onset date is February 25 to June 15. In order to take advantage of th...
We write to follow on from our eLetter published in 2003.[1]
As more information becomes available regarding the diagnosis and laboratory testing of SARS, the official number of laboratory confirmed SARS cases in Taiwan during the 2003 outbreak has been officially determined to be 346.[2]
The duration of the outbreak by onset date is February 25 to June 15. In order to take advantage of the newly-available data for modeling purpose, we use the up-dated cumulative case data of the 346 lab confirmed cases to fit the exponential curve with first-order autocorrelation in the error structure.[3] We divided the time duration of the outbreak into four time periods, the resulting estimated mean effective reproductive number of the observed time period R* of the curve-fitting and a chronology of the significant events related to the outbreak which occurred in Taiwan at the dividing point of each time period is given in the Table.
Table 1 Mean effective reproductive numbers R* for each of the four time periods with chronological events of relevance for the time periods.
Time Period
Mean
Std.
Err.
95 Lower
C.I.
95 Upper
C.I.
2/25-4/9
2.33401
0.20365
1.93487
2.73316
4/10-4/28
3.22814
0.34836
2.48578
3.97049
4/29-5/16
1.26861
0.03626
1.19029
1.34693
5/17-6/15
0.12190
0.00920
0.10301
0.14078
2/25 – Onset date of first confirmed case.
4/09 - Admission of first SARS patient to Ho Ping Hospital.
4/28 - Implementation of Level B quarantine and other interventions
measures.
5/16 – Change of leadership at Department of Health and CDC-Taiwan.
6/15 - Onset date of the last hospital infection.
The temporal fluctuation in the value of R* is further exhibited in figure 1 with similar trend as that of the mean effective reproduction number for various time intervals obtained in [1] using the probable cases.
Figure Histogram for mean effective basic reproduction number R* during the four time periods of SARS outbreak in Taiwan, 2003
The drastic decrease in the mean effective reproduction number after April 29 further confirms the fact that the turning point for the outbreak to subside had occurred around April 29 [4]. These result shows that the mathematical modeling methodology used here is inherently consistent, regardless of whether we use the cumulative probable case data as in [1], or the more restrictive but reliable cumulative laboratory confirmed case data.
References
1. Hsieh YH, Chen CWS. (2003) Re: Mathematical modeling of SARS: Cautious in all our movements [electonic response to the JECH Severe Acute Respiratory Syndrome Supplement] jech.com 2003http://jech.bmjjournals.com/cgi/eletters/57/6/DC1#66
The editorial by Gracia highlights the hidden nature of domestic
violence and he is right to outline the role of society in uncovering this
problem.[1] Definitional issues for epidemiologists are extremely complex
in this field, and are further complicated by differing definitions used
in criminal justice and voluntary organisations. There is little clear
consensus as to the definition of 'domestic' or '...
The editorial by Gracia highlights the hidden nature of domestic
violence and he is right to outline the role of society in uncovering this
problem.[1] Definitional issues for epidemiologists are extremely complex
in this field, and are further complicated by differing definitions used
in criminal justice and voluntary organisations. There is little clear
consensus as to the definition of 'domestic' or 'violence' and this has
hampered research, policy and practice.
Evidence of the benefit of screening in emergency departments is
considerably more limited than he suggests; there are no controlled trials
with meaningful outcomes.[2] Many studies have shown impressive 'before
and after' increases in detection rates, though whether these reduce
violence or morbidity is unproven.[3] The evidence needs to be
considerably stronger and consistent with National Guidelines before
screening is recommended.[4]
References
(1) Gracia E. Unreported cases of domestic violence against women:
towards an epidemiology of social silence, tolerance, and inhibition. J
Epidemiol Community Health 2004; 58(7):536-537.
(2) Jewkes R. Preventing domestic violence. BMJ 2002; 324(7332):253-
254.
(3) Morrison LJ, Allan R, Grunfeld A. Improving the Emergency
Department detection rate of domestic violence using direct questioning.
Journal of Emergency Medicine 2000; 19(2):117-124.
(4) Ramsay J, Richardson J, Carter YH, Davidson LL, Feder G. Should
health professionals screen women for domestic violence? Systematic
review. BMJ 2002; 325(7359):314.
The authors indicate their intention to clear up the definition of
EBPH defining it "as a public health endeavor in which there is an
informed, explicit, and judicious use of evidence that has been derived
from any of a variety of science and social science research and
evaluation methods."[1] It was very reassuring to see the authors clearly
include ‘lay-knowledge’ as a ‘type of evidence in public h...
The authors indicate their intention to clear up the definition of
EBPH defining it "as a public health endeavor in which there is an
informed, explicit, and judicious use of evidence that has been derived
from any of a variety of science and social science research and
evaluation methods."[1] It was very reassuring to see the authors clearly
include ‘lay-knowledge’ as a ‘type of evidence in public health decisions’
that may help shape Evidence Based Public Health (EBPH).
The literature does not even provide a clear definition of Evidence
Based Health Care (EBHC) and there is confusion between authors as to
whether the concepts of EBHC and Evidence Based Medicine (EBM) are
comparable and the terminology interchangeable. The most frequently used
definition of EBHC is indeed synonymous with EBM and defined as "the
conscious explicit and judicious use of current best evidence in making
decisions about the care of individual patients," (Sackett 1996). The
inadequacy of this definition, by apparently excluding the patient from
the decision process, motivated its revision by Sackett to its ‘new
improved’ version, "the integration of best research evidence with
clinical expertise and patients values," (Sackett 2000) which highlights
the contemporary view of the patient as a partner in the concept.
How ironic that the majority of writers continue to refer to the
original ‘time expired’ definition. Other authors, (Jadad 2000) have
continued to use the 1996 EBM terminology in defining EBHC. Muir Gray
attempted to differentiate EBM from EBHC, "EBM is the provision of
appropriate healthcare for individual patients." Whereas in the UK, the
responsibility for the healthcare of populations is clearly allocated to
health authorities and to management, where the concept of evidence-based
healthcare was invented," and "the development of EBHC was stimulated and
facilitated by the NHS R&D program."
Now all that remains is to clarify the differences between EBHC and EBPH
or do we need evidence based public health care (EBPHC) to bridge the gap?
Let Goodman sound a final warning note, "At its core, evidence based
practice rests on a supposition which, while probably true, itself has
unclear evidentiary support"
Reference
1. Lucie Rychetnik, Penelope Hawe, Elizabeth Waters, Alexandra Barratt, and Michael Frommer. A glossary for evidence based public health. J Epidemiol Community Health 2004; 58: 538-545.
Dear Editor – Battersby et al present a method of performing equity
audit where data on incidence, deprivation and surgical resection rates of
non-small cell lung cancer are compared.[1] Deprivation was measured
using the Index of Multiple Deprivation (IMD) 2000[2] and all analyses
were performed at the primary care trust (PCT) level. Battersby et al
report no statistically significant associations between the...
Dear Editor – Battersby et al present a method of performing equity
audit where data on incidence, deprivation and surgical resection rates of
non-small cell lung cancer are compared.[1] Deprivation was measured
using the Index of Multiple Deprivation (IMD) 2000[2] and all analyses
were performed at the primary care trust (PCT) level. Battersby et al
report no statistically significant associations between their measure of
deprivation and age and sex standardised incidence of non-small cell lung
cancer. This is highly unusual and contrary to findings from a large
number of different populations.[3-9] Without clear evidence that there
is something exceptional about the population studied by Battersby et al,
the lack of association between deprivation and incidence of lung cancer
is likely to be an artefact.
Two possible explanations of Battersby et al's failure to find an
association between deprivation and incidence of lung cancer are possible.
Firstly, calculating deprivation at the PCT level may be highly
inaccurate. The IMD 2000 is a ward level variable and the authors of this
study use a weighted average to determine IMD 2000 scores for PCTs.
However, with an average of almost 22 wards per PCT in England and an
average PCT population of over 128 000 in the current study, it is not
clear that “the use of a weighted average IMD score is an appropriate way
of including deprivation in the analysis”.[1] Secondly, with only 17 data
points, it is likely that Battersby et al's study lacks sufficient power
to detect any but the strongest of associations.
Although the focus of Battersby et al's article is primarily on a
methodological technique, rather than on the particulars of the example
data used, the failure to detect an association between deprivation and
incidence of lung cancer suggests some problem with this technique.
Furthermore, presentation of these results to non-specialised audiences,
as is intended, may give the impression that deprivation is not an
important determinant of incidence of lung cancer – which the majority of
other evidence suggests it is.
Although public health practice requires methods that are quick, easy
and cheap to perform, this should not be at the expense of accuracy to
such a degree that important, strong and known associations are
overlooked.
References
(1) Battersby J, Flowers J, Harvey I. An alternative approach to
quantifying and addressing inequity in healthcare provision: access to
surgery for lung cancer in the east of England. Journal of Epidemiology
& Community Health 2004;58:623-625.
(2) Department of the Environment Transport and the Regions. Indices
of Deprivation 2000. Regeneration Research Summaries 2000;31.
(3) Singh G, Miller B, Hankey B. Changing area socioeconomic patterns
in US cancer mortality, 1950-1998: Part II - lung and colorectal cancers.
Journal of the National Cancer Institute 2002;94(12):916-925.
(4) Kreiger M, Quesenberry C, Peng T et al. Social class,
race/ethnicity, and incidence of breast, cervix, colon, lung, and prostate
cancer among Asian, black, Hispanic, and white residents of the San
Francisco Bay Area, 1988-92 (United States). Cancer Causes and Control
1999;10:525-537.
(5) Brown J, Harding S, Bethune A et al. Incidence of health of the
nation cancers by social class. Population Trends 1997;90:40-77.
(6) Pollock A, Vickers N. Breast, lung and colorectal cancer incidence
and survival in South Thames Region, 1987-1992: the effect of social
deprivation. Journal of Public Health Medicine 1997;19(3):288-294.
(7) van Loon A, Brug J, Goldbohm Ret al. Differences in cancer
incidence and mortality among socio-economic groups. Scandinavian Journal
of Social Medicine 1995;23(2):110-120.
(8) van Loon A, Goldbohm R, van den Brandt P. Lung cancer: is there an
association with socioeconomic status in The Netherlands? Journal of
Epidemiology & Community Health 1995;49(1):65-69.
(9) Hein H, Suadicani P, Gyntelberg F. Lung cancer risk and social
class. The Copenhagen Male Study - 17 year follow up. Danish Medical
Bulletin 1992;39(2):173-176.
The letter by Boyle points out that there is not sufficient evidence to
recommend screening for domestic violence, which is not to say that there
is sufficient evidence to recommend against screening.[1] Recommendations for
screening or routine inquiry for domestic violence have been made on other
grounds.
For example, the US Preventive Services Task Force concluded that
“there is insuffici...
The letter by Boyle points out that there is not sufficient evidence to
recommend screening for domestic violence, which is not to say that there
is sufficient evidence to recommend against screening.[1] Recommendations for
screening or routine inquiry for domestic violence have been made on other
grounds.
For example, the US Preventive Services Task Force concluded that
“there is insufficient evidence to recommend for or against the use of
specific screening instruments to detect family violence, but
recommendations about physical abuse patients may be made on other
grounds” [2]. David Atkins, coordinator of the US Preventive Services Task
Force, said that “asking patients about abuse as part of routine history-taking may be indicated on the basis of the substantial prevalence of
undetected abuse among women, the potential value of this information in
the care of the patient, and the low risk of harm in asking” [3].
Also the Canadian Task Force on Preventative Health Care adopted
the position that there is insufficient evidence to recommend for or
against routine screening for violence although, despite the lack of
evidence, concluded that the prevalence of and significant impairment
associated with violence against women make it important to maintain a
high index of suspicion when assessing patients [4-6]. Other medical
organizations and accreditation bodies do recommend screening women or
routine enquiry for domestic violence in health settings [7-10].
In Australia different initiatives are seeking to incorporate routine
screening programs for domestic violence in health services and their
evaluations support screening [11-14].
The editorial by Gracia pointed to the hidden nature of domestic
violence, and one of the advantages of routine enquiring is that it uncovers
hidden cases of domestic violence helping to break the social silence,
inhibition and climate of tolerance surrounding the victims [15, 16]. For
Bradley et al screening should be thought of as a way of uncovering and
reframing a hidden stigma.[17] The importance of inquiring about domestic
violence for increasing identification rates is also recognized in a
influential review which concluded that there was insufficient evidence to
recommend screening programmes. Ramsay et al acknowledge that “we know
that introducing a programme is likely to increase the number of women
experiencing domestic violence who are identified by health professionals,
but not that subsequent interventions are effective” [18]. It is true that
lack of evidence about effectiveness may discourage health services to
implement them, but again, as Jewkes argues, not only the question of what
is effectiveness in this context has not been resolved, but also equating
lack of evidence to ineffectiveness is premature [19]. For example in a
workshop hosted by the Centers for Disease Control and Prevention in USA,
participants agreed that screening for domestic violence should be
evaluated scientifically but there was also consensus that physicians
should not wait for the results of randomized clinical trials to begin
screening in their own practices [3].
Among the benefits of routine screening, not always easy to evaluate
in clinical trials (but also worth evaluating), are the increase of
awareness about domestic violence both among health professionals and the
wider community, greater pressure for increasing high quality training
resources for health professionals, improvement of knowledge, skills and
sensitive attitudes among professionals, greater pressure to increase
resources for victims after disclosure, benefits as a result of early
intervention for victims both physical and psychological, increase of the
social visibility of this problem, changes on social tolerance of violence
in relationships, and also greater awareness and perceived relevance among
the research community and funding agencies. And all these potential
benefits may contribute to a coordinated response aiming to uncover
hidden cases of domestic violence, reduce the social silence, tolerance
and inhibition surrounding the victims, and progressively melt the iceberg
of domestic violence against women.
References
(1) Boyle A A. Domestic violence screening, evidence is lacking [electronic response to Gracia E. Unreported cases of domestic violence against women: towards an epidemiology of social silence, tolerance, and inhibition.] jech.com 2004
http://jech.bmjjournals.com/cgi/eletters/58/7/536
(2) US Preventive Services Task Force. Guide to Clinical Preventive
Services (2nd ed.) Baltimore, MD: Wilkins and Wilkins, 1996.
(3) Cole, T. Is domestic violence screening helpful? JAMA
2000;284:551-3
(4) MacMillan HL, Wathen CN with the Canadian Task Force on
Preventive Health Care. Prevention and treatment of violence against
women: Systematic review & recommendations. CTFPHC Technical Report
#01-4. London, ON: Canadian Task Force. September, 2001
(5) Wathen CN, MacMillan HL, with the Canadian Task Force on
Preventive Health Care. Prevention of violence against women.
Recommendation statement form the Canadian Task Force on Preventive Health
Care. CMAJ 2003;169:582-4.
(6) MacMillan HL, Wathen CN. Violence against women: integrating the
evidence into clinical practice. CMAJ 2003;169:570-1.
(7) American Medical Association, Council on Scientific Affairs.
Violence against women: Relevance for medical practitioners. JAMA
1992;267:3184-9.
(8) Waalen J, Goodwin MM, Spitz AM, et al. Screening for intimate
partner violence by health care providers: barriers and interventions. Am
J Prev Med 2000;19:230-7.
(9) Department of Health. Domestic violence: a resource manual for
health care professionals. London: Stationery Office, 2000.
(10) Rhodes KV, Levinson W. Interventions for intimate partner
violence against women: clinical applications. JAMA 2003;289:601-5.
(11) New South West Health. Policy and procedures for identifying and
responding to domestic violence. NSW Health Department, Sydney, 2003.
(13) Women’s Health Strategy Unit. Northern Territory Department of
Health and Community Services. Domestic violence screening pilot
evaluation report. Department of Health and Community Services. Northern
Territory Government, 2003.
(14) Laing L, with Australian Domestic & Family Violence
Clearinghouse. Routine screening for domestic violence in health services.
Australian Domestic & Family Violence Clearinghouse, 2001.
(15) Gracia E. Unreported cases of domestic violence against women:
towards an epidemiology of social silence, tolerance, and inhibition. J
Epidemiol Community Health 2004;58:536-537
(16) Taket A, Nurse J, Smith K, et al. Routinely asking women about
domestic violence in health settings. BMJ 2003;327:673-6.
(17) Bradley F., Smith M., Long J, et al. Reported frequency of
domestic violence: cross sectional survey of women attending general
practice. BMJ 2002;324:1-6
(18) Ramsay, J, Richardson J, Carter YH, et al. Should health
professionals screen for domestic violence? Systematic review. BMJ
2002;325:314.
(19) Jewkes R. Preventing domestic violence. BMJ 2002;324:253-4.
Gracia is right about the "softer" benefits of routine inquiry for
domestic violence and the role of societal attitudes is crucial in
overcoming this problem.[1] Certainly the role of public awareness campaigns,
though difficult to evaluate, has probably reduced social tolerance
towards violence against women.
Evidence and the lack of evidence is a tricky issue for domestic violence
screening. This seems t...
Gracia is right about the "softer" benefits of routine inquiry for
domestic violence and the role of societal attitudes is crucial in
overcoming this problem.[1] Certainly the role of public awareness campaigns,
though difficult to evaluate, has probably reduced social tolerance
towards violence against women.
Evidence and the lack of evidence is a tricky issue for domestic violence
screening. This seems to be an area where all the normal "rules" about
screening seem to be ignored.[2] Screening differs very much from routine
inquiry that is more flexible and intuitive, but this distinction is often
forgotten.[3] Atkins is wrong to suggest that there is a low risk of harm
in asking.[4]
A proportion of women find inquiry unacceptable and this may damage
the therapeutic relationship with an individual provider.[5,6] The most
dangerous time for a woman in an abusive relationship is when she decides
to leave, the risk of violence escalates. Healthcare providers need to be
trained properly to deal with the resultant problems that disclosure can
cause.
The risks are further compounded where there are young children living
with the victim. The association between child protection and domestic
violence is unclear, though female victims of assault are much more likely
to be known to child protection services.[7] Activation of child
protection procedures is not without considerable risk and potential for
distress. Interestingly "acceptability" is much greater in the peri-natal
population than in primary care.[8] Time is a vital resource and it is
not clear whether there is enough time for this among busy clinicians.
Clinicians may be prepared to make more time if they could be persuaded of
the benefits of their efforts.
Randomised controlled evidence about the benefits of interventions is
conflicting.[9,10] This is partly because interventions may be
culturally specific, the interventions are complex and follow-up in these
women can be difficult. However, routine inquiry should only proceed when
there is enough evidence to demonstrate that the risks are outweighed by
the benefits and this should require RCT level evidence. Proceeding
beforehand may cause more problems than it would solve.
References
(1) Gracia E. Re: author's reply [electronic response to Boyle AA. Domestic violence screening, evidence is lacking] jech.com 2004 URL direct link to eLetter
(2) The National Screening Committee. 2004.
http://www.nsc.nhs.uk/uk_nsc/uk_nsc_ind.htm
(3) Taket A, Nurse J, Smith K, Watson J, Shakespeare J, Lavis V et al. Routinely asking women about domestic violence in health settings.
BMJ 2003; 327:673-676.
(4) US Preventive Services Task Force. Guide to
Clinical Preventive Services, 2nd ed. Baltimore: Wilkins & Wilkins. 1996.
(5) Gielen AC, O'Campo PJ, Campbell JC, Schollenberger J, Woods AB,
Jones AS et al. Women's opinions about domestic violence screening and
mandatory reporting. American Journal of Public Health 2003; 19(4):279-
285.
(6) Partovi SN, Nelson BK, Bryan ED, Walsh MJ. Faculty triage
shortens emergency department length of stay. Academic Emergency Medicine
2001; 8(10):990-995.
(7) Ward L, Shepherd J, Edmond AM. Relationship between adult
victims of assault and children at risk of abuse. BMJ 1993; 306:1101-1102.
(8) Bindman AB, Grumbach K, Keane D, Rauch L, Luce JM. Consequences
of queuing for care at a public hospital emergency department. JAMA 1991;
266(8):1091-1096.
(9) Sullivan CM, Bybee D, I. Reducing violence using community-based
advocacy for women with abusive partners. J Consult Clin Psychol 1999;
67(1):43-53.
(10) Dunford FW, Huizinga D, Elliott DS. The role of arrest in
domestic assault: the omaha police experiment. Criminology 1990; 28(2):183
-206.
Dear Editor
My understanding, confirmed by brief review of data on the ONS website, is that the South Asian population in the UK has a younger age distribution than the white population. In this case, would an indicator based on those aged over 35 need further adjustment for the age distribution beyond 35 in order to examine prescribing? Is it possible that we are seeing an age effect in the negative correlation...
Dear Editor
The Inverse Care Law, proposed by Julian Tudor Hart in 1971, states that “the availability of good medical care tends to vary inversely with the need for it in the population served.”[1]
A number of authors have now claimed to have found instances of the Inverse Care Law operating in practice.[2,3] Given the prominence that this ‘law’ has gained in the health care literature over last thirty...
Dear Editor
Dr Leyland published an interesting and thought provoking article about the variations in the risks of death by region in the United Kingdom.[1] Dr Leyland does not that migration might explain some of the differences found in the study. Immigrants to a country tend to be healthier than the base population.[2] In addition, workers tend to be healthier than non-workers [3] which could influence the...
Dear Editor
Jousilahti and Salomaa appear unhappy with our response to their paper on the social patterning of serum inflammatory markers.
First they feel that we have misinterpreted their findings and conclusions. We fail to see how. They found (as others have) that social disadvantage was associated with increased inflammation – as indexed by markers such as higher fibrinogen - and that this association...
Dear Editor
We write to follow on from our eLetter published in 2003.[1]
As more information becomes available regarding the diagnosis and laboratory testing of SARS, the official number of laboratory confirmed SARS cases in Taiwan during the 2003 outbreak has been officially determined to be 346.[2]
The duration of the outbreak by onset date is February 25 to June 15. In order to take advantage of th...
Dear Editor
The editorial by Gracia highlights the hidden nature of domestic violence and he is right to outline the role of society in uncovering this problem.[1] Definitional issues for epidemiologists are extremely complex in this field, and are further complicated by differing definitions used in criminal justice and voluntary organisations. There is little clear consensus as to the definition of 'domestic' or '...
Dear Editor
The authors indicate their intention to clear up the definition of EBPH defining it "as a public health endeavor in which there is an informed, explicit, and judicious use of evidence that has been derived from any of a variety of science and social science research and evaluation methods."[1] It was very reassuring to see the authors clearly include ‘lay-knowledge’ as a ‘type of evidence in public h...
Dear Editor – Battersby et al present a method of performing equity audit where data on incidence, deprivation and surgical resection rates of non-small cell lung cancer are compared.[1] Deprivation was measured using the Index of Multiple Deprivation (IMD) 2000[2] and all analyses were performed at the primary care trust (PCT) level. Battersby et al report no statistically significant associations between the...
Dear Editor
The letter by Boyle points out that there is not sufficient evidence to recommend screening for domestic violence, which is not to say that there is sufficient evidence to recommend against screening.[1] Recommendations for screening or routine inquiry for domestic violence have been made on other grounds.
For example, the US Preventive Services Task Force concluded that “there is insuffici...
Gracia is right about the "softer" benefits of routine inquiry for domestic violence and the role of societal attitudes is crucial in overcoming this problem.[1] Certainly the role of public awareness campaigns, though difficult to evaluate, has probably reduced social tolerance towards violence against women. Evidence and the lack of evidence is a tricky issue for domestic violence screening. This seems t...
Pages