|
 |
|
|||||||
|
|||||||||
Electronic Letters to:
|
|
eLetters: Electronic letters published:
-
![[Read eLetter]](/icons/misc/sectionDOWN.gif)
Interpreting increasing health inequalities in Spain
- James Scanlan (24 April 2006)
-
absolute and relative health inequalities
- Valerie A Kay (1 March 2006)
-
Money cannot change testosterone levels...
- James M. Howard (26 January 2006)
|
|
|||
|
James Scanlan, Attorney None
Send letter to journal:
jps{at}jpscanlan.com James Scanlan
|
The findings by Regidor et al.1 of increasing health inequalities in Spain during periods of decreasing socioeconomic inequality must be evaluated in light of certain statistical tendencies generally found when the prevalence of adverse health outcomes is changing. As can be observed in virtually any data set showing the proportions of two groups falling below and above various points on a continuum, the rarer and outcome, the greater the relative difference in experiencing it and the smaller the relative difference in avoiding it. For example, when a test cutoff is lowered, relative differences between the failure rates of two demographic groups tend to increase while relative differences in pass rates tend to decline; when poverty declines, relative differences in poverty rates of two demographic groups tend to increase while relative differences in rates of avoiding poverty tend to decline.2,3,4,5 Similarly, when adverse health outcomes decline, as occurred in Spain during the period studied by Regidor, one would expect an increase in the relative difference between the rates at which advantaged and disadvantaged groups experience the outcomes, and such increase may be sufficient to more than offset any reduction in health-related difference between the two groups that might have arisen from a reduction in socioeconomic inequality. Thus, one cannot interpret a simple increase in the rates of experiencing an adverse outcome to mean any true adverse change in the relative health of the disadvantaged group – “true change” meaning one that is not solely a consequence of the decline in the prevalence of the adverse outcome These tendencies make the interpretation of changes in relative rates of experiencing (or avoiding) an outcome problematic. The interpretation of changes in absolute differences is also problematic. We will observe in the types of data referenced above that when almost the entire population experiences the adverse outcome the absolute difference is small; as the outcome declines toward a point where a majority of the population no longer experiences it, the absolute difference grows larger; and as the outcome continues to decline the absolute differ grows small again.2 So certain changes in absolute differences will flow solely from changes in the prevalence of the outcome. Usually mortality in published studies tends to be reported in terms of rates that are low enough that overall declines might be expected to lead to reductions in absolute differences. In fact one observes this quite often. Thus, in circumstances where the reported rates are relatively low, we might be inclined to regard an increase in absolute differences (as observed in Regidor's study) as reflecting a true worsening of the relative health of the disadvantaged group. Yet seemingly low mortality rates are often a function of convention, as where low yearly mortality rates are reported in circumstances where the ten year rates are quite high ¬– that is, more in the range where overall declines tend to lead to increases in absolute differences. Further, low rates reported in studies are composites of varied rates among subpopulations (e.g., younger and older age groups), some of which may have quite high rates. Declines in overall prevalence among such subpopulations can lead to increasing absolute differences between advantaged and disadvantaged groups within the subpopulations in ways that can cause overall absolute difference also to increase notwithstanding what seems to be a low overall prevalence rate. Such factor may also lead to an increase in the overall relative difference in avoiding the outcome, even though, consistent with expectation in time of a decline in the prevalence of the outcome, the relative difference in avoiding the outcome is declining within each subpopulation. Regidor et al.state that “[t]he greatest reductions in health inequalities will be achieved when interventions focus on the health problems that occur most frequently even if problems that occur less frequently show a pronounced gradient in relative terms.” There certainly is something to be said for focusing attention on health disparities with large absolute differences rather than large relative differences, given that more common problems tend to show large absolute differences and small relative differences while rarer problems tend to show small absolute difference and large relative differences. And it has been noted that problems with large relative differences tend not to be the main contributors to the overall health inequalities.6 But researchers need to recognize that when common problems are growing less prevalent, relative differences in experiencing the outcomes are likely to increase, and absolute differences may increase as well, solely as a function of the change in the prevalence of the outcome. Interpreting the meaning of observed changes in either relative or absolute differences – with respect either to the particular problems or the overall mortality rates of which the particular mortality rates comprise a part – is a complex undertaking. 1. Regidor E, Ronda E, Pascual C, Martinez D, Calle ME, Dominguez V. Decreasing socioeconomic inequalities and increasing health inequalities in Spain: A case study. Am J Pub Health. 2006;96:102-108. 2. Scanlan JP. Can we actually measure health disparities? Chance. 2006:19(2): ____. In press. 3. Scanlan JP. Measuring health disparities. J Public Health Manag Pract. 2006;12(3):294. 4. Scanlan JP. Race and Mortality. Society. 2000;37(2):19-35. 5. Scanlan JP. Divining difference. Chance. 1994;7(4):38-9, 48. 6. Wong MD, Shapiro MF, Boscardin WJ, Ettner SL. Contributions of major diseases to disparities in mortality. N Engl J Med. 2002; 347:1585 -92. |
|||
|
|
|||
|
Valerie A Kay, graduate student & policy adviser
Send letter to journal:
vakay{at}students.latrobe.edu.au Valerie A Kay
|
The article by Regidor et al states "Our results show that redistribution of income might achieve greater social justice but probably does not lead to reduced health inequalities ...". However I question whether this can be justified on the evidence provided. As I understand it, Spain's level of income (compared to other countries) has risen over the period 1985-2000. Therefore it is possible that the absolute income differential between the lowest and highest quartile has increased although the ratio has decreased. For example using the ratios provided (2.88 in 1985 and 2.27 in 2000) and holding income constant in dollar terms, it would be possible that the 1985 incomes were say, $100 per week in the lowest quartile and $288 in the highest, while the incomes in 2000 were $300 per week in the lowest quartile and $681 in the highest. Thus the absolute difference in 1985 would be $188 while the absolute difference in 2000 would be $381. If there is a reasonably linear relationship between income and health differentials, then we would expect greater health inequalities in 2000 on that basis. Moreover, while the authors have touched on some of the other factors that might influence this result, they have not looked at them in detail. For example in some countries higher income women were more likely than lower income women to be smokers in earlier decades but have also been more likely to give up smoking more recently. Patterns like this would influence the differentials in cancer rates. I agree with the authors' comments that improvements in health and public health messages in recent decades may have been more likely to benefit, or influence, higher income groups, but I don't think that their suggestion that reducing income inequality won't affect health inequalities can be justified on their evidence. No historical analysis can every say what might have happened, therefore they also can't rule out the possibility that health inequalities might now have been worse in Spain if the ratio of income inequality had also increased. |
|||
|
|
|||
|
James M. Howard, Biologist independent
Send letter to journal:
jmhoward{at}anthropogeny.com James M. Howard
|
It is my hypothesis that testosterone is directly causative of human evolution (Androgens in Human Evolution, Rivista di Biologia / Biology Forum 2001; 94: 345-362). Therein, I suggest that periodically excessive testosterone accumulates within populations as increases in the percentage of individuals of higher testosterone. This excessive testosterone produces adverse effects on these individuals. It is known that testosterone increases adverse health. "The suppressive effects of male sex hormones on immune functions have been observed in a wide variety of disease processes and appear to be testosterone-mediated." (Aging Male. 2005 Sep-Dec;8(3):166-74) Excessive testosterone increases learning problems. This will manifest itself as low socioeconomic achievement in populations which increasingly depend upon advanced learning for employment. I suggest the percentage of individuals in low socioeconomic levels is increased. An input of monies into these groups will not change the higher percentage of individuals of higher testosterone and the consequences therefrom. In fact increased monies may actually increase the percentage of individuals of higher testosterone. This may explain the findings of Regidor, et al. |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |