Stroke is a leading cause of death and acquired adult disability and it is estimated that globally each year 5.8 million people die from stroke [1]. In high income countries it has been stated that approximately 1 in 20 adults is affected by stroke and the incidence of acute cerebrovascular events (strokes and transient ischemic attack combined) exceeds the incidence of acute coronary heart disease [2]. In the recent INTERSTROKE study five risk factors including hypertension, current smoking, abdominal obesity, diet and physical activity, were associated with 80% of the global risk of stroke and interventions to reduce these factors are therefore likely to reduce the stroke incidence [3]. The observation period in this study covers a time period where smoking prevalence among Danes in general declined and treatment of hypertension and hypercholesterolemia gradually improved [4, 5]. During the same period life expectancy and mean age of the population increased which could be followed by a higher absolute number of stroke events. Increasing life expectancy and mean age of the population would therefore be likely to increase the absolute number of stroke events. Different trends in stroke incidence have been reported worldwide, and it is currently not known if the stroke incidence has changed in Denmark [6–9]. Temporal changes in admission and diagnostic practices may tend to increase the overall hospitalization rates, which is a burden to health facilities even if it does not reflect an actual increase in stroke incidence rates. Therefore the aim of this study was to examine temporal trends in stroke admissions in Denmark in the period 1997–2009.

This nationwide study showed that in the Danish population there was a constant rate of hospitalized stroke events of 3.2/1000 PY from 1997–2009. A decrease in the stroke hospitalization rate was seen when adjusted for age. IRs for hospitalized ischaemic stroke increased during the study period whereas the IRs for hospitalized unspecified stroke decreased. Compared to the control group the stroke population was generally older and suffered from more comorbid conditions.

Different trends in stroke incidence have been reported previously in different countries. In a systematic review of 56 population-based studies from 2009, based on multiple sources of information, the authors reported a 42% decrease in stroke incidence in high-income countries, and a more than 100% increase in low to middle income countries [7]. The World Health Organization (WHO) Monitoring Trends And Determinants In Cardiovascular Disease (MONICA) stroke project did not report a definite change in stroke incidence over time [9]. Multinational comparison is possible with appropriate data quality although caution is required, as stated by others [11]. The Copenhagen City Heart Study, a prospective study based on discharge diagnosis as well as supplementary patient files, found that stroke incidence from 1976–1993 remained stable among men aged 45–64 years and women aged 45–84 years, whereas a statistically significant decrease was found in men aged 65–84 years [12]. Similarly, another Danish prospective study based on clinical stroke diagnoses evaluated by a trained neurologist, in the community of Frederiksberg, Copenhagen, found an increase of 42% in stroke incidence for men aged 65–84 years from 1972–1974 to 1989–1990 [13]. There were no changes in IRs for women. In Lund, Sweden, in a study involving retrospective screening of medical journals, increasing IRs were found, mainly confined to people < 75 years of age [14]. In a recent cohort study from Rotterdam, the Netherlands, based on stroke diagnosis from general practitioners as well as hospital files reviewed by a stroke neurologist, the authors found that IRs from 1990–2008 increased in men, but not in women [15].

The observation period in this study covered a time period where smoking prevalence among Danes in general declined and treatment of hypertension and hypercholesterolemia gradually improved [4, 5]. On the other hand the incidence of atrial fibrillation (AF) is likely to have increased, partly due to increasing population size as well as a larger proportion of elderly [16, 17]. AF is an important risk factor for stroke and is estimated to be associated with approximately 1/5 of ischaemic stroke events, which could increase in the absolute number of stroke events due to the rise in AF incidence [18, 19]. Furthermore, it has been suggested that an increase in stroke occurrence might be due to increased survival after ischaemic heart disease [20]. Increased focus on and awareness of stroke symptoms in both the general population and in the health care system may lead to improved detection of less-severe strokes and which could increase the number of patients discharged with a diagnosis of stroke [21]. Introduction of thrombolytic treatment given to carefully selected patients may have further increased the focus on early recognition and diagnosis of stroke events, as the chances of successful outcome decrease by the minute after onset of stroke symptoms [22, 23]. This may also have resulted in a tendency in the current era to perform cerebral CT scanning of patients with milder cases of strokes, which would not have been diagnosed before. As such, the fact that the number of patients diagnosed with ischaemic stroke increased over time in our study may be explained, in part, by increased use of cerebral imaging procedures. During the later years more hospital units have gained access to CT and MRI scanners that make it more likely that the stroke diagnosis has become more specific. Therefore, the increasing IRs of ischaemic strokes observed in our study is more likely to reflect the implementation of scanners in the Danish health care system than an actual increase in actual ischemic stroke rates. This notion may also explain why the IRs of patients admitted with unspecified stroke decreased in the observed time period. The decreasing rates may again reflect that more and more patients are scanned and may therefore receive a more exact stroke diagnosis.

Recent WHO estimates based on reviews of studies including more than 20 million subjects, project that the number of strokes in European countries is likely to increase from 1,1 million per year in 2000 to 1,5 million per year in 2025 [24]. Moreover, the median age in these populations is envisioned to rise from 37.7 years in 2000 to 47.7 years in 2050 along with a decrease in the total number of Europeans. Overall, these projections suggest that the absolute number of strokes will increase in future decades, whereas the number of younger adults will decrease. This development is likely to put further burden on the welfare society and health care systems in European countries [25, 26].

This study reflects the number of patients discharged from hospital and every patient admitted for a stroke calls for hospital facilities as well as urgent attention from physicians and nurses. The large group of patients in need of assistance and rehabilitation is also a significant expense for the health care system as half of all stroke patients suffer from permanent ill-effects and almost one-out-of-four acquire disabilities to such an extent that help to everyday activities is necessary [27]. Furthermore there is an additional cost due to the decreased employability and early retirement of stroke patients, as well as the fact that one third of all stroke patients suffer from depression [28]. The socioeconomic importance of the disease cannot be underestimated.

This study found a constant rate of stroke hospitalization from 1997–2009. The overall rate of stroke hospitalized stroke events adjusted for age was reduced during the study period. This study was the first to investigate trends in patients discharged with a first-time stroke diagnosis in Denmark. Effective preventive and treatment strategies are very much needed, as the burden of stroke will increase as the elderly segment of the population increases.