Studies of changing inequalities in receipt of procedures like that
carried out with respect to revascularization by Hetemaa et al.[1] need to
be undertaken with an appreciation of the statistical tendency whereby the
rarer an outcome the greater the relative difference in rates of
experiencing it and the smaller the relative difference in rates of
avoiding it.[2-6]
Studies of changing inequalities in receipt of procedures like that
carried out with respect to revascularization by Hetemaa et al.[1] need to
be undertaken with an appreciation of the statistical tendency whereby the
rarer an outcome the greater the relative difference in rates of
experiencing it and the smaller the relative difference in rates of
avoiding it.[2-6]
Most research into inequality in the health arena examines morbidity
and mortality. Typically these are examined in terms of relative
differences in experiencing adverse outcomes. In recent decades most,
though not all, adverse health outcomes have been declining and relative
differences in experiencing them have been increasing. Generally these
increases have been regarded as reflecting meaningful worsening of the
relative situation of disadvantaged groups, but without recognition of the
extent to which such increases may be solely the consequences declining
prevalence of the outcomes or recognition that relative differences in
experiencing the opposite outcome may be declining. Whether the observed
patterns of changing relative differences are more than or less than those
that would be expected to flow solely from declines in the prevalence of
the outcome has gone unexamined, though it is not clear that there are
effective tools to answer such questions.[2,6]
Research into inequalities in the receipt of beneficial procedures,
on the other hand, has generally examined rates of experiencing the
favorable outcome (i.e., receipt, rather than denial, of the procedure).
Because rates of receiving these procedures usually have been increasing,
relative differences in rates of receiving them have been declining,
though relative rates of failing to receive them have been increasing. A
greater increase in rates of receiving the procedure experienced by groups
with lower baseline rates of receiving the procedures (relative to proxy
for need), such as that found in the study by Hetemaa et al., is a
corollary to this pattern, as is a smaller decrease in rates of failing to
receive the procedure. Table 1 in the study provides many illustrations
of the pattern. For example, the overall female rate of receiving
revascularization procedures increased 71 percent more than the male rate
(a 317% increase for women compared with a 186% increase for men), but the
female rate of failing to receive the procedure declined by 22 percent
less than the male rate (a 20% decline for women compare with 26% decline
for men). Correspondingly the relative difference in rates of receiving
the procedure decreased while the relative difference in rates of failing
to receive the procedure increased.
The authors note an expectation of declining inequality based on what
has been observed in other situations where there occurred an increased
supply of coronary revascularization procedures. Yet, the pattern of
declining relative differences in receipt of procedures, not only for
revascularization but for all procedures that are increasing, is generally
to expected to occur solely as a result of the increase in supply, as is
an increase in the relative difference in failing to receive the outcome.
Whether either change reflects a meaningful change in inequalities – i.e.,
one that is not solely a consequence of the increasing availability of the
procedure – requires a closer examination. Again, however, it is not
clear that there exist effective tools for doing so.
The above-described tendency is pertinent not only to comparisons of
changes over time, but to all comparisons of relative differences in
settings with differing overall frequencies of an outcome. The authors
note smaller relative differences in procedures in districts where the
procedures are more common and larger relative differences among persons
over 70 (where procedure are rarer). This pattern is to be expected
simply because of the differing frequencies of the procedures in the
different settings. And one would likely find the reverse pattern if one
examined rates of failing to receive the procedure.
That is not to say that these patterns will be observed with respect
to every comparison of the size of relative differences in varying
temporal, demographic, or geographic settings. For factors other than the
referenced statistical tendency are at work as well. Nevertheless, one
cannot evaluate those factors without appreciation of the purely
statistical aspects of the observed patterns.
In the United States, citing the receipt or non-receipt of
mammography as an example, the National Center for Health Statistics
(NCHS) recently recognized that the size and the patterns of change in
health inequalities may turn on whether one examines the favorable or the
adverse outcome.[7] It recommended that all relative differences between
groups be measured in terms of adverse outcomes. If the recommendation is
followed, in many situations where relative differences were perceived to
be declining – as, for example, in the case of male-female
revascularization rates in Finland – the differences would instead be
perceived to be increasing. But NCHS has yet to acknowledge that relative
differences in rates of experiencing favorable and adverse outcomes tend
to change systematically in opposite directions as the prevalence of each
outcome changes or to suggest a means of identifying changes in inequality
that are not solely the consequence of changes in prevalence.
One might think that the NCHS focus on adverse outcomes would be
especially inappropriate for something like revascularization, since, even
among those hospitalized for cardiac heart disease, revascularization is
not appropriate for everyone. The point, however, is that the value of
health inequality studies lies in identifying changes that are not solely
the result of changes in prevalence. Neither changes in relative
differences in receipt of procedures nor changes in relative differences
in denial of procedures – nor changes in absolute differences (which here
favored men)[2,6] – seem to serve that purpose.
James P. Scanlan
References
1. Hetemaa T, Keskimäki I, Manderbacka, et al. How did the recent
increase in the supply or coronary operations in Finland affect
socioeconomic and gender equity in their use? J Epidemiol Community
Health 2003;57:178-185.
2. Scanlan JP. Can we actually measure health disparities? Chance
2006:19(2):47-51:
http://www.jpscanlan.com/images/Can_We_Actually_Measure_Health_Disparities.pdf.
3. Scanlan JP. Measuring health disparities. J Public Health Manag
Pract 2006;12(3):294 [Lttr]:
http://www.nursingcenter.com/library/JournalArticle.asp?Article_ID=641470.
4. Scanlan JP. Race and Mortality. Society. 2000;37(2):19-35:
http://www.jpscanlan.com/images/Race_and_Mortality.pdf.
6. Scanlan JP. The misinterpretation of health inequalities in the
United Kingdom. Paper presented at: British Society for Population Studies
Annual Conference 2006, Southampton, England, Sept. 18-20, 2006:
http://www.jpscanlan.com/images/BSPS_2006_Complete_Paper.pdf.
7. Keppel K., Pamuk E., Lynch J., et al. Methodological issues in
measuring health disparities. Vital Health Stat 2005;2 (141):
http://www.cdc.gov/nchs/data/series/sr_02/sr02_141.pdf.
In their prospective study, Baibas et al (JECH 2005;59(4):274-
8)showed that, in a Greek mountain village at 950 metres, total mortality
and not merely coronary mortality was lower than in two lowland villages.
What follows assumes that the cancer figures included within "other
causes" follow this pattern.
In 'Geographic Cancer Risk and Intracellular Potassium/Sodium
ratios'.Cancer Detection...
In their prospective study, Baibas et al (JECH 2005;59(4):274-
8)showed that, in a Greek mountain village at 950 metres, total mortality
and not merely coronary mortality was lower than in two lowland villages.
What follows assumes that the cancer figures included within "other
causes" follow this pattern.
In 'Geographic Cancer Risk and Intracellular Potassium/Sodium
ratios'.Cancer Detection and Prevention 1986;9:171-94, B. Jansson reported
a high intracellular K+/Na+ ratio correlating with low cancer mortality.
Apart from diet, the favourable ratio of ions is promoted, inter alia at
higher altitudes by an effect involving the sodium pump. Among ten studies
supportive of altitude-low cancer effects cited by Jansson, several show
an altitude effect mainly at sites exposed to air, viz. mouth,
esophagus,larynx and lung. Even 250 metres can make a difference. Jansson
had ideas about the mechanism possibly involved in this cancer
prophylaxis.
Recent animal studies, if applicable to humans, permit further
speculations about a mechanism, viz. relative inhibition of Na+K+ATPase
(causing low intracellular K+/Na+) at low altitudes, leading to cell
detachment from one another and from substrate (Contreras et al.Journal
of Cell Science 1999;112:4223-32).Such detached cells would normally
die from the form of apoptosis named anoikis, but if the cells express the
activated neurotrophic receptor trkB, via a pathway through PI(3)K and
Akt, non-malignant cells can be converted into highly tumorigenic cells
(Douma et al.Nature 2004;430:1034-9).One assumes other factors must be
present.
To isolate the contribution to longevity of physical exertion under
hypoxia, as proposed by Baibas et al, I suggest studying mortality in
towns built on a high plateau, compared with those at similar mean
elevations, which are spread up and down a mountainside.
We read with interest the recent
editorial setting an agenda for future research considering neighbourhood
influences on health.[1] Whilst agreeing with many of the points raised in the editorial we take issue with the proposed use
of multiple membership multilevel models to explore contextual effects at
different points in time.
A recent paper used multiple
membership models to analyse the effect of area of residence observed over 9
years on individuals’ health (measured just at the final time point).[2] An accompanying editorial pointed out the problems with
the use of such models.[3] Since we agree that understanding the longitudinal
influence of area of residence over time will form a key part of the
neighbourhood research agenda we expand on the reasons that make multiple
membership models unsuitable for important research questions pertinent to the
field and propose an alternative model.
If person P1 lived in area A1
at time T1 and in area A2 at time T2 then from a life course
perspective we might expect there to be a contribution of both areas to that
person’s measured or reported health at time T2. Similarly, if person P2 moved in the opposite direction then we would expect
a contribution from both areas. The multiple membership model,
however, makes two simplifying and unrealistic assumptions. Firstly, the effect
of each area is assumed to be the same at both times, having the same effect on
person P1 at time T1 as on person P2 at time T2. This means assuming that
the area effect is not dependent on the period; the regeneration or decay of
areas is ignored. Secondly, the multiple membership model
requires explicit weights to be attached to the contribution of each area to an
individual’s health. Suggesting that this should be done on the basis of the
time spent in each area[1] is to assume that the effect of an area is constant
irrespective of the stage of the life course – in other words, such an analysis
assumes that the risk associated with neighbourhoods accumulates at a steady
rate throughout the life course. So if persons P1 and P2
spent an equal amount of time in each area, the direction of movement (from A1 to A2 or from A2
to A1) is ignored.
Alternative models – such as a critical period model – cannot be investigated
under such a framework.[4] Yet the influence of the social environment along the
life course has been shown to differ for specific diseases.[5]
The authors propose extending the
above model to one where health, as well of area of residence, is measured at
both time points.[1] In addition to the assumptions noted above, the
suggestion that the area of residence at time T2 affects health measured at time T1 clearly departs from the usual model of causation.
A more suitable model for such data
is the cross-classified multilevel model. Figure 1 illustrates the
classification diagram for individuals (at level 1) whose health is measured at
time T2 and who live in a
cross-classification of areas at times T1
and T2. Every individual
lives in an area at each time point, whether these are different areas (P1 and P2) or the same area (P3 and P4).
When health is measured at T1
individuals are nested within areas in a strict hierarchy; when health is
measured at T3 there is a
three-way cross-classification of areas at T1,
T2 and T3. The cross-classified
model does not make the assumption that area effects will be the same at the
two time points; in fact, the assumption that the effects for one area at
different times are uncorrelated is likely to lead to conservative estimates of
the variance components. However, the free estimation of the variances
associated with each time point means that a variety of life course models can
be examined without the need for prior assumptions.
Level 1:
Person
Level 2(1):
Neighbourhood at time T1
Level 2(2):
Neighbourhood at time T2
Figure 1 Multilevel structure
of individuals nested within a cross-classification of areas at two time points
(T1 and T2). Diagram shows person P1 moving from area A1
to area A2 and person P2 moving from area A2 to area A1.
Alastair H Leyland
Senior Research Scientist
MRC Social and Public Health
Sciences Unit, Glasgow, Scotland
Øyvind Næss
Senior Scientist
Epidemiological Division, National
Public Health Institute, Oslo, Norway
Correspondence to:
Alastair H Leyland
MRC Social and Public Health
Sciences Unit
4 Lilybank Gardens
Glasgow G12 8RZ
Scotland
References
1.Kawachi I, Subramanian SV. Neighbourhood influences on
health. Journal of Epidemiology and
Community Health 2007; 61:3-4.
2.Chandola
T, Clarke P, Wiggins RD, et al. Who you live with and where you live: setting
the context for health using multiple membership multilevel models. Journal of Epidemiology and Community Health
2005; 59:170-5.
3.Leyland AH. Assessing the impact of mobility on health: implications for lifecourse epidemiology.Journal of Epidemiology and Community Health 2005; 59:90-1.
4.Kuh
D, Ben-Schlomo Y, Lynch J, et al. Life course
epidemiology. Journal of Epidemiology and
Community Health 2003; 57:778-83.
5.Næss
Ø, Strand BH, Davey Smith G. Childhood and adulthood
socioeconomic position across 20 causes of death. A
prospective cohort study of 800 000 Norwegian men and women.Journal of Epidemiology and
Community Health (in press).
I wish to thank Dr Hanna and her
colleagues for this excellent study (1). I would like to share below a few
comments.We can read, under the “Social
Acceptability” heading:
“Within
the Bangladeshi community smoking was not acceptable as Islam forbids addiction
to any substance. However, it was agreed that smoking was a habit for some
Muslims, although much less acceptable in women than in men. Smoking using a
hookah was uncommon in Scotland owing to the absence of strong sunlight for
drying the tobacco. It was more acceptable to chew paan, which was common
among women and men. It was thought that truthful answers to questions on
smoking might be more likely if the questions were put by a doctor or by an
independent researcher.”
First
off, Islam does not “forbid” many things. It is an extremely tolerant religion;
so tolerant that even Western tobacco control activists are often amazed to see
how anti-tobacco campaigns are difficult to implement in the corresponding
countries (2):
« Lâ
′ikrâha fî-d-dîn » (Let There Be No Compulsion in Religion)
(Qur’ân: II, 255)
Concerning
hookah (shisha, narghile) smoking, I think the questionnaire could have been
enhanced at this point for two main reasons:
1-because
of the tremendous recent development of hookah smoking in the world, already
called an epidemic by some researchers;
2-the
interviewees were probably thinking of the traditional raw tobacco usually
prepared in their remote country. However, more and more people, in the United
Kingdom and other countries of the world, now smoke a hookah with a
ready-to-use tobacco or non-tobacco molasses based mixture called tobamel or
“mu‘assel” (i.e. honeyed in Arabic)(3). In these conditions, their
answer was expected and, I would add, naïve: no sun so no sun-cured tobacco…
This
adapted questionnaire by Hanna and colleagues is, I insist, original and
excellent and I have no doubt that “the methods and lessons are applicable
internationally”. It is not biased as it actually happened with another one
in Lebanon where the interviewees did not know that some questions related to
the supposed established detrimental health effects of hookah smoking were, in
fact, referring to a study based on a “waterpipe” smoking machine in a
laboratory and powered by a type of charcoal (quick self-lighting) thatis not used in their country (4).
(1) Hanna L, Hunt S, Bhopal RS. Cross-cultural adaptation of
a tobacco questionnaire for Punjabi,Cantonese, Urdu and Sylheti speakers:
qualitative research for better clinical practice, cessation services and
research . Journal of Epidemiology and Community Health
2006;60:1034-1039.
(2) Chaouachi K:
Le narguilé : analyse socio-anthropologique. Culture,
convivialité, histoire et tabacologie d’un mode d’usage populaire du tabac.
Doctoral thesis, Université Paris X (France). [Eng.:
Narghile (hookah): a Socio-Anthropological Analysis. Culture,
Conviviality, History and Tobaccologyof a Popular Tobacco Use Mode]. Published by ANRT (Lille), 420
pages.
(3) Chaouachi K. A Critique of the WHO's TobReg "Advisory
Note" entitled: "Waterpipe Tobacco Smoking: Health Effects, Research
Needs and Recommended Actions by Regulators. Journal of Negative Results in
Biomedicine2006 (17 Nov); 5:17.
(4) Chaaya M., Roueiheb
Z.E., Chemaitelly H., Azar G., Nasr J. and Al-Sahab B. Argileh smoking
among university students: A new tobacco epidemic. Nicotine &
Tobacco Research. 2004
Jun; 6 (3):457-63.
The results of the study by Fairley and Leyland [1] of changing
social class inequalities in perinatal outcomes in Scotland must be
interpreted in light of the statistical tendency whereby the rarer an
outcome the greater the relative differences in experiencing the outcome
and the smaller the relative difference in avoiding the outcome.[2-6] In
times of declines in adverse outcome (the more common...
The results of the study by Fairley and Leyland [1] of changing
social class inequalities in perinatal outcomes in Scotland must be
interpreted in light of the statistical tendency whereby the rarer an
outcome the greater the relative differences in experiencing the outcome
and the smaller the relative difference in avoiding the outcome.[2-6] In
times of declines in adverse outcome (the more common situation in recent
decades), the relative difference in experiencing the outcome will tend to
increase solely as a consequence of the decline in prevalence. But such
increases in relative difference in experiencing the outcome – which
usually are attended by declines in the relative difference in rates of
avoiding the outcome as well as declines in the absolute difference
between rates of experiencing (or avoiding) – ought not to be regarded as
a meaningful worsening of the relative situation of disadvantaged groups.
A departure from the pattern in the 1980s, however, might suggest an
improvement in that situation. [2,6]
The same tendency influences the differing patterns observed among
groups categorized by characteristics that are related to overall risk,
such as are shown in Tables 3-5 of the study. The authors note that
inequalities among lone mothers are smaller than among married mothers
despite the fact that lone mothers suffer greater socioeconomic
disadvantage and ill health than married mothers. But one should expect
to find greater relative social class differences in adverse perinatal
outcome rates among married mothers simply because adverse outcome rates
are lower among married mothers. The relative difference in avoiding
these outcomes, however, generally would be smaller.[2,4,6]
For the same reason, inequalities measured in relative rates of
experiencing adverse perinatal outcomes will tend to be greater among the
(lower risk) age 20-34 group than the (higher risk) under 20 group. That
the relative difference is greater among the 35 or above group than the 20
-34 group, assuming that the 35 or above group is at higher risk of
adverse outcomes than the 20-34 group, does not mean the statistical
tendency is not present. Rather, it merely suggests that certain factors,
possibly including the implications of smoking noted by the authors,
outweigh the statistical tendency.
The above observations apply as well to differences in the relative
index of inequality, which measure is largely a function of relative
differences in rates of experiencing an outcome.[2,6] The odds ratios
shown in the tables raise somewhat different issues. In the case of the
outcome frequencies at issue here, odds ratios tend to approximate
relative risks of experiencing the outcome, and the observations regarding
the sizes of such relative risks are unlikely to be less pertinent because
the figures shown in the tables are odds ratio. It is true that relative
difference patterns measured in odds ratios will not vary depending on
whether one examines the adverse or the favorable outcome. But that does
not mean that differences between odds ratios are in some manner
reflecting differences between the sizes of inequalities that are not
solely the function of differences in prevalence in the various settings,
whether defined temporally or demographically, that are being compared.
Like other measures of differences between the rates at which two groups
experience or avoid such outcomes, odds ratios tend to change solely as a
result of changes in prevalence, though less predictably than changes in
relative risks.[2,6]
James P. Scanlan
References
1. Fairley L, Leyland AH. Social class inequalities in perinatal
outcomes: Scotland 1980-2000. J Epidemiol Community Health 2006;601:31-
36.
2. Scanlan JP. Can we actually measure health disparities? Chance
2006:19(2):47-51:
http://www.jpscanlan.com/images/Can_We_Actually_Measure_Health_Disparities.pdf.
3. Scanlan JP. Measuring health disparities. J Public Health Manag
Pract 2006;12(3):294 [Lttr]:
http://www.nursingcenter.com/library/JournalArticle.asp?Article_ID=641470.
4. Scanlan JP. Race and Mortality. Society. 2000;37(2):19-35:
http://www.jpscanlan.com/images/Race_and_Mortality.pdf.
6. Scanlan JP. The misinterpretation of health inequalities in the
United Kingdom. Paper presented at: British Society for Population Studies
Annual Conference 2006, Southampton, England, Sept. 18-20, 2006:
http://www.jpscanlan.com/images/BSPS_2006_Complete_Paper.pdf.
Dear Editor
Studies of changing inequalities in receipt of procedures like that carried out with respect to revascularization by Hetemaa et al.[1] need to be undertaken with an appreciation of the statistical tendency whereby the rarer an outcome the greater the relative difference in rates of experiencing it and the smaller the relative difference in rates of avoiding it.[2-6]
Most research into inequa...
Dear Editor
In their prospective study, Baibas et al (JECH 2005;59(4):274- 8)showed that, in a Greek mountain village at 950 metres, total mortality and not merely coronary mortality was lower than in two lowland villages. What follows assumes that the cancer figures included within "other causes" follow this pattern.
In 'Geographic Cancer Risk and Intracellular Potassium/Sodium ratios'.Cancer Detection...
Dear Editor
...
Dear Editor,
I wish to thank Dr Hanna and her colleagues for this excellent study (1). I would like to share below a few comments....
Dear Editor
The results of the study by Fairley and Leyland [1] of changing social class inequalities in perinatal outcomes in Scotland must be interpreted in light of the statistical tendency whereby the rarer an outcome the greater the relative differences in experiencing the outcome and the smaller the relative difference in avoiding the outcome.[2-6] In times of declines in adverse outcome (the more common...
Pages