Commentary: When bad news is worse than it sounds

Abstract

In their study of a quarter century’s worth of English deaths attributed to cardiovascular disease (CVD), Dr Asaria and colleagues have unearthed a trend that is worse than they report it to be. What is new in Asaria et al.’s article is the reported trend in geographical inequalities in suffering and dying from this disease. However, these authors succumb to a very common error at the start of their report, an explanation of which, I hope, serves to further strengthen their overall case and the importance of their work. In its substantive results, this article begins by stating that ‘When grouped by deprivation quintile, absolute inequality between most and least deprived wards narrowed over time in ages 30–64 years . . . ’. This is true, but the implication that such a trend implies a reduction in health inequalities is false. When overall mortality rates for a disease are falling then it is quite possible to record such narrowing, but for the narrowing not to imply a reduction in inequality. It is possible because of the biological impossibility for a widening to occur when that widening could require a mortality rate to fall below zero. In Table 1 of Asaria et al.’s article, the number 69 appears twice, and it is this rate and change of rate that helps explain why the bad news this article reveals is worse than it sounds. The first time 69 crops up is when it is revealed that, between 1982 and 1986, among all the least deprived quintile of wards, for every 100 000 women alive between the ages of 30 and 64, only 69 died each year from CVD. That is a very low rate of CVD deaths (twenty years on, even the best-off quintiles of men do not quite enjoy such a low rate). The second time 69 surfaces in Asaria et al.’s Table 1 is in the column recording the absolute reduction between 1982–86 and 2002–06 for each group by quintile of deprivation, sex and age. Between those dates, among all those wards in the most deprived quintile, for every 100 000 women aged between 30 and 64, some 69 fewer died each year of CVD in the later period than in the former. That is a greater absolute fall in CVD deaths for this age group of women than that enjoyed by any other quintile group (although even the best-off quintile of men enjoyed twice as fast a rate of improvement). Why are these two 69s of such importance? The reason is that when grouped by deprivation quintile, for absolute inequality (between most and least deprived wards) not to have narrowed over time in ages 30–64 years, but for the fall of 69 per 100 000 among the women living in the poorest areas to have still occurred, something impossible would have had to have taken place. A slightly greater absolute fall would have had to have taken place among the women resident in the least deprived fifth of wards. Standardized mortality rates there would have had to have fallen by at least 70 women not dying of CVD per 100 000 per year for the absolute gap to have widened. As they began at 69, the mortality rate would have had to have fallen to –1 deaths by 100 000 women per year. A rate of –1 implies reincarnation; in this case, the reincarnation of a woman living in the most affluent of areas would have had to have occurred or, to be a little more accurate, at least one such reincarnation for every 100 000 women aged between 35 and 64 living in such areas. In other words, when considering changes in inequality, the superficially positive first finding of this article is not at all positive; in the case of women, it is simply not possible for the absolute gap not to narrow, and in the case of men, it is highly implausible that any apparent narrowing should be seen as some kind of social inequality achievement. So, read again what most researchers who scan the abstract of the article will read, but now given what you know about how the narrowing was inevitable (sans reincarnation): ‘When grouped by deprivation quintile, absolute inequality between most and least deprived wards narrowed over time in ages 30–64 years, but increased in older adults’ (abstract), ‘ . . . relative inequalities worsened in all four age-sex groups, more so in young and middle-ages, pointing to persistent environmental, social and health system injustice’ (page 12). The kind of narrowing described here is akin to the narrowing in General Certificate of Secondary Published by Oxford University Press on behalf of the International Epidemiological Association