Trends in Mortality From Ischemic Heart Disease in 50 Spanish Provinces

INTRODUCTION

Ischemic heart disease continues to be the leading cause of death in adults in developed countries. In Spain, it is the leading cause of death in men and the second leading cause of death in women.¹ In 1998, 22 352 men and 17 090 women died from ischemic heart disease in Spain, representing 37.09% and 23.42% of cardiovascular mortality for men and women respectively, and 11.86 and 10.08%, respectively, of all-cause mortality.

In Spain, several population registers are used to monitor acute myocardial infarct (AMI), including the MONICA database in Catalonia² and the IBERICA database.³ Using the IBERICA database, the incidence of AMI in the Spanish population aged 25 to 74 was estimated at 194 cases per 100 000 inhabitants/year for men and 38 cases per 100 000 inhabitants/year for women.⁴

These date indicate the importance of monitoring trends in ischemic heart disease and controlling its risk factors. Non-transmissible diseases, particularly cardiovascular disease, cancer and diabetes, are now considered a pandemic in the 21st century, and affect both developed and developing countries. The World Health Organization has promoted the monitoring of these diseases and their principal risk factors as part of an overall strategy for their prevention and control.

Monitoring trends over time and analyzing the geographic distribution of disease can be useful in detecting changes which suggest future risks for the population. Mortality data provide a good source of information as they are collected universally and their collection is highly standardized. For that reason, they allow comparisons to be made and are widely used to assess trends.

Spain has one of the lowest mortality rates from ischemic heart disease in the world, and the rate has been steadily decreasing over the past 25 years.¹ However, little is known about whether this general pattern also applies to different parts of Spain, nor whether there has been a trend towards a reduction in mortality from ischemic heart disease in individual provinces. The primary aim of the present study was to analyze mortality from ischemic heart disease in the population aged 35 to 64 years in different regions of Spain. A further aim was to identify regions in which significant deviations from the general trend might suggest a need for health care interventions.

PATIENTS AND METHODS

Mortality data were obtained from individual registers supplied by the National Institute of Statistics. The diagnostic codes selected were 410-414, which correspond to ischemic heart disease in the Ninth Revision of the International Classification of Diseases.

Statistical analysis

Mortality rates from ischemic heart disease in the population aged 35 to 64 were compared by province for two 4-year periods (1988-1991 and 1994-1997). The analysis took into account 3 age groups (35-44 years, 45-54 years and 55-64 years). For the two periods and the age groups mentioned, the denominators used were the populations in 1989 and 1995, respectively: these were calculated using interpolation methods based on the usual population statistics. Rates were compared using logarithmic linear regression models, in which it was assumed that the number of deaths in each age group and period showed a Poisson distribution.⁵ In formulating the model, the population was introduced as an offset or constant; the dependent variable were deaths, which were allowed to vary in terms of age and period. Age and period were incorporated into the model as explanatory variables. Independent models were adjusted for each province and sex, and the variables age and time-period were introduced as factors. The model was designed to provide the age-adjusted rate ratio and 95% confidence intervals (CI), and to compare the two time periods in each province.

Adjusted mortality rates were calculated for each province for 1998 using the direct method with the standard European population, and 18 age groups for each sex. Adjusted mortality rates per 100 000 person-years are presented for each Spanish province in 1998. The evolution of mortality rates in Spain as a whole were analyzed for the period 1980-1998. The statistical significance of trends was obtained by adjusting simple linear regression using the S-Plus program. The percentage change between 1980 and 1998 was calculated using the formula (T80 ­- T98)/T80x100, and the annual percent change was calculated using the formula recommended by López Abente et al:⁶ (exp [(ln (T80)-­ln (T98)]/10)-­1)x100.

Lastly, Spanish data were compared with European data provided by the World Health Organization,⁷ which include standardized mortality rates for the new standard world population for both sexes, all age groups, and per 100 000 person-years.

RESULTS

Table 1 compares provincial mortality rates from ischemic heart disease by sex in the Spanish population aged 35-64 using mortality rate ratios for the periods 1994-1997 and 1988-1991. In general, rates for both men and women were higher in 1988-1991 than in 1994-1997, as shown by the fact that the majority of the rate ratios were lower than 1. There was a statistically significant reduction in mortality among men in 27 provinces and among women in 12 provinces (Figure 1). In men, mortality rates increased between the two periods in only 5 provinces, though the increases were not statistically significant. The largest rate ratios were observed in Navarra, Orense, Guadalajara, and Toledo. During the second 4-year period (1994-1997), the population aged 35-64 in these provinces showed an excess risk of dying from ischemic heart disease of 17% in Navarra, 12% in Orense, and 10% in Guadalajara and Toledo compared with a similar population in the first 4-year period (1988-1991). The provinces of Burgos, Cantabria, and León showed the greatest reductions in mortality between the first and second periods, with a risk reduction of approximately 35% between the two periods. In women, the risk of dying from ischemic heart disease was higher in the second period in 10 provinces, although the increase in risk was only statistically significant in Ávila. Ávila is, however, a sparsely populated province and the absolute number of deaths is small, so this result should be treated with caution, due to the possible instability of the rates. An increase in the rate ratio was also observed in Segovia, Guadalajara, and Huelva, with an increased risk of dying from ischemic heart disease in these provinces of 55%, 31% and 25%, respectively.

Fig. 1. Provincial mortality from ischemic heart disease in Spain (1994-1997/1988-1991). Rate ratio in the population aged 35-64 years.

Table 2 shows the provincial mortality rates from ischemic heart disease in Spain in 1998, presented as rates adjusted by the standard European population and per 100 000 person-years for all age-groups and both sexes. The overall rate for Spain was 101.11 in men and 44.89 in women. The regions with the highest mortality rates from ischemic heart disease were Andalusia, Asturias, the Balearic Islands, the Canary Islands, Extremadura, Murcia, and the Community of Valencia. In men, the highest rates were observed in the canary Islands, followed by Seville, Alicante, and Cádiz. The lowest rates were in Soria, Segovia, Guadalajara and Cuenca. In women, the highest rates were also found in the Canary Islands and in Seville, whilst the lowest rates were found in Álava, Guadalajara, Zamora, Burgos, and Huesca. There was a notable difference in mortality rates between provinces; in men, the province with the highest mortality had 2.42 times more mortality than the province with the lowest mortality and in women the difference was 3.04. There were also notable differences between the sexes in terms of mortality attributable to ischemic heart disease.

Table 3 and Figure 2 show the mortality rates from ischemic heart disease in Spain from 1980 to 1998 for all age-groups and both sexes, adjusted by the standard European population, and per 100 000 person-years. In the period in question, mortality from ischemic heart disease steadily decreased. In 1980, mortality in men was 115.07 cases per 100 000 person-years and in women it was 50.54; 18 years later the corresponding rates were 101.11 in men and 44.89 in women. The percent change over the period was 12% in men and 11% in women, with an annual percent change in men of 1.30% and 1.19% in women. The decrease was slightly more marked in men, although mortality rates in men were double those of women. In men, the trend estimated by the linear regression model corresponded to a beta coefficient of ­-0.99 (P<.01), with a beta coefficient in women of -­0.41 (P<.01).

Fig. 2. Mortality from ischemic heart disease in Spain (1980-1998). Rates adjusted using the standard European population per 100 000 person-years. IHD indicates ischemic heart disease.

Table 4 shows the mortality from ischemic heart disease in other countries in the European Union, using adjusted rates for the standard world population. Rates are shown for 100 000 person-years and are provided for all age groups and both sexes. In both men and women, mortality rates from ischemic heart disease in Spain are amongst the lowest in Europe, with only France having a lower rate.

DISCUSSION

The results of this study show that mortality from ischemic heart disease in Spain has decreased significantly over the period studied. Overall, for all age groups and for the country as a whole, the tendency over the last 19 years has been towards a slight but significant reduction in mortality. Specifically, in the 35-64-year-old age group studied here, the results show a significant reduction in mortality from ischemic heart disease in 27 provinces in men and in 12 provinces in women. Only Ávila showed a statistically significant increase in mortality from ischemic heart disease, but this result should be treated with caution due to the possible instability of the rates there.

Mortality from ischemic heart disease in men has not, however, decreased significantly in 23 provinces, and in women, in the majority of provinces (n=38), mortality rates in 1994-1997 were not significantly lower than those observed 6 years earlier. In some cases, there is very little variation over the study period, a result which hints at a possible stabilization of mortality rates. In a small number of provinces, a non-statistically significant increase in mortality rates was observed, though this result needs to be confirmed over time. As a whole, our results indicate that the continuous reduction in mortality is not applicable to the whole of Spain, and that mortality is not decreasing in various regions.

The present study was carried out using official statistics. The MONICA study suggested that official statistics tend to exaggerate the trend towards reduced mortality in comparison with data obtained from clinical records of patients in the population. While official statistics suggested that mortality in the MONICA population was decreasing at an annual rate of 4%, data for the same population from the MONICA study suggested a rate of reduction of 2.7% in men and 2.1% in women.² If the same bias existed in the present study, the result would be an even more notable stabilization of rates than that observed.

With respect to the geographic distribution of mortality, the results of the present study indicate that north-south differences still exist, with higher mortality rates in the south. Other important regional differences persist, and mortality rates are still higher in men than in women. Mortality, however, remains low in comparison to other developed countries. In general, the results of the present study are consistent with those from earlier studies8, including those using aggregated data from different regions⁹ or municipalities,10 as well as those which focused on specific regions such as Catalonia,¹¹ Andalusia,¹² the Community of Valencia,¹³ Murcia¹⁴ or the Canary Islands.¹⁵

Trends in mortality and the geographic distribution of cardiovascular disease can be explained by various factors including geographical differences, changes in the prevalence of cardiovascular risk factors,^16-18 differences in socioeconomic level, and differences in the quality of and access to health care. Factors such as the environment or tourism are have also been shown to be associated with differences in cardiovascular mortality. With regard to diet, the low consumption of fish and wine in Spain could explain the apparent contradiction between high levels of cardiovascular mortality and a low consumption of saturated fats in Mediterranean regions.¹⁹ Results from the Four Provinces Study²⁰ show that children living in cities where mortality from ischemic heart disease is high have higher body mass indices, a higher caloric intake, and a higher consumption of cholesterol and sodium. Diabetes also shows a north-south pattern;¹⁰ the highest mortality rates are observed in Extremadura, Andalusia, the Levant, and the Canary Islands. The IBERICA study³ also showed a greater than 25% prevalence of diabetes in the Mediterranean regions studied, particularly Valencia and Murcia. Both of these regions have high vascular mortality rates.

The results of the MONICA study²¹ also show, however, that the classic risk factors for cardiovascular mortality only explain 15% of mortality in women and 40% in men. The Eight Provinces Study²² suggested that socioeconomic aspects and social inequalities might also affect mortality from ischemic heart disease, and the small areas Mortality Atlas provides evidence of an association between the distribution of cardiovascular mortality and patterns of illiteracy, unemployment and overcrowding.¹⁰ It has also been shown that socioeconomic level is the variable which best explains geographical variations in the prevalence of overweight and obesity in Spain.²³

These factors are not directly susceptible to modification by the health care system, although they can be taken into account when assigning health care resources. The impact of existing inequalities of access to quality health care has been demonstrated by the MONICA study,² which showed that decreases in cardiovascular mortality are strongly associated with improvements in treatment for heart disease. The IBERICA study³ showed that that the hospital management of coronary patients varied across Spain, that there differences in the use of diagnostic and therapeutic resources, and differences in mortality. Studies which have investigated the hospital-based care of AMI patients, such as the PRIAMHO,²⁴ ARIAM,²⁵ and PREVESE II²⁶ studies, all showed that there are differences in access to health care and that aspects such as secondary prevention, or delays between the onset of AMI symptoms and health care, could be improved.

In terms of limitations, the present study suffers from the problems inherent in obtaining data from death certificates, although in the case of ischemic heart disease, death certificates in Spain have been shown to be of good quality.²⁷ As mentioned, biases revealed by the MONICA study are small and not significant when referring to a given period. They tend to magnify reductions in mortality when trends are analyzed. Some of the geographical differences found may be due to errors in classifying place of residence, an error which is particularly relevant in tourist areas, where mortality rates may be overestimated by t he inclusion of the deaths of non-residents. Such an effect has been observed in tourist areas in the Canary Islands.¹⁵ These limitations should be taken into account when interpreting the results of the study.

In conclusion, this study has shown that the reduction in mortality from ischemic heart disease observed in Spain as a whole is not homogenous across the country. The high prevalence of cardiovascular risk factors, together with geographical variations in the quality of health care resources and the access to these resources, provide ample opportunity for action and improvement. The results of the present study should be useful in prioritizing between geographical areas when deciding on where health care interventions should be focused.

ACKNOWLEDGEMENTS

The authors would like to thank Dr. Gonzalo López-Abente for his advice during the present study.

See editorial

---

Correspondence: Dra. R. Boix Martínez.
Centro Nacional de Epidemiología. Instituto de Salud Carlos III.
Sinesio Delgado, 6. 28029 Madrid. Spain.
E-mail: [e-mail]