Statistics from Altmetric.com
Sir,—We were most interested to read the report by Reitsma et al on heart failure hospitalisations in the Netherlands.1 While these authors found a trend in hospitalisations similar to our report from Scotland they point out that their absolute hospitalisation rates were lower.2There are at least three likely explanations for this.
First, Reitsma et al excluded a number of ICD codes for heart failure that we included in our report: ICD9 codes 425.4 (primary cardiomyopathy), 425.5 (alcoholic cardiomyopathy), and 425.9 (secondary cardiomyopathy, unspecified). These codes accounted for 4.4% of our total cases in 1990.
Second, and much more importantly, the prevalence of coronary artery disease, the major cause of heart failure, is much higher in Scotland than in the Netherlands. For example, the age adjusted mortality rate per 100 000 for men aged 35–74 in 1992 was 535 in Scotland and 248 in the Netherlands (ICD codes 410–414, standardised to England and Wales population 1972). The respective rates for women were 218 and 84.
Third, a considerably higher proportion of the Scottish population are elderly—the age groups with the highest incidence and prevalence of heart failure. The proportion of the population aged 75–79, 80–84, and 85+ in the Netherlands in 1993 was 1.96, 1.16, and 0.68%, respectively; in the UK these proportions were 2.37, 1.55, and 0.85%. For women in the Netherlands the proportions were 3.07, 2.29, and 1.86; in the UK they were 3.53, 2.86, and 2.49%.
In summary, when these differences are considered the findings of Reitsma et al are consistent with ours in the Scottish population.
This letter was shown to the authors, who replied as follows:
We highly appreciate the letter by McMurray and Morrison in which they mention several causes for the higher number of hospitalisations for heart failure in Scotland compared to the Netherlands. We endorse their viewpoints on these factors, especially their remark on the difference in prevalence of coronary artery disease between the two countries.
In addition to the explanations brought forward by McMurray and Morrison, other factors such as hospital admission policies and coding practice may influence the number of first listed discharge diagnosis, especially in a complex and chronic condition like heart failure. Differences in these factors are difficult to assess and need more attention in future research. For the sake of comparison the use of age specific discharge rates with smaller age intervals (5 or 10 years) or standardisation to a widely available standard (the European population standard) should be encouraged.1-1