The Annual Incidence of Vascular Dementia and Other Dementias: A Population-Based Study in Eastern Finland

Vascular dementia (VD) is widely recognized as the second most common cause of dementia after Alzheimer’s disease (AD), accounting for 15–20% of cases [1, 2]. AD, the leading cause, accounts for 60–70% [2‒4]. However, the literature suggests that a significant proportion of AD cases also have underlying vascular pathology, making mixed-type dementia (AD+VD) common [5, 6]. In a recent Swedish study, 23% of all dementia diagnoses accounted for mixed-type dementia [4]. Dementia with Lewy body dementia (LBD) and frontotemporal dementia (FTD) accounted for 3–11% and 1–8% of cases, respectively [4, 7‒9].

The incidence of dementia is expected to rise as the population ages [10]. Globally, populations are ageing, and the fastest growth is occurring in the age group of 65 and over. In Finland, the percentage of the population aged 65 and over is expected to grow to 27.6% by 2050 [11]. The number of people with dementia has been estimated to increase from 57.4 million cases globally in 2019 to 152.8 million cases in 2050 [12]. However, a stable or slightly declining trend in the age-specific incidence of dementia in high-income countries has been reported in the literature [12, 13]. Improved control of vascular risk factors and treatment of vascular diseases may partly explain this trend [14]. Additionally, dementias impose a massive economic burden, with global costs estimated to have reached 1.3 trillion dollars in 2019, encompassing direct medical and social care, and informal care costs [15]. Recent studies also indicate a positive correlation between total costs and disease stage [16].

Limited data on VD incidence in Finland have been available in recent decades, with most studies from the 1980s or 1990s, except for a recent Finnish national registry study [17‒20]. In Finland, the percentage of the population aged 65 and over has grown rapidly since the 1990s, from 13.4% in 1990 to 22.6% in 2020 [11]. These demographic changes, along with advancements in preventive medicine over past decades, emphasize the growing need for contemporary epidemiological data on VD and other dementias.

We conducted this study to obtain up-to-date epidemiological data on the incidence of VD and other dementias. Up-to-date data enable the examination of trends and the diagnostic sensitivity of dementia subtypes. Here, we report on our population-based study carried out in Finland.

Our objective was to determine the incidence of VD and other dementias in Northern Savo, Finland. The study data were created using the Kuopio University Hospital (KUH) memory disorder register and an anonymized KUH-based register of memory disorder diagnoses, created primarily for administrative purposes. The validity of the KUH registers was cross-checked by comparing the absolute incidence numbers from KUH with recorded numbers from two regional health centres in Northern Savo, ensuring consistency. Records from the regional health centres are locally maintained and were anonymized before being provided for this study. Demographic data for Northern Savo and its municipalities were provided by Statistics Finland, the Finnish authority for official statistics.

This population-based epidemiological study involved the Northern Savo area of Eastern Finland, the country’s sixth-largest region, with 248,190 inhabitants, representing 4.4% of the population [21]. The region comprises 18 municipalities, with Kuopio as the capital. Health services are delivered by KUH (central hospital), 3 regional hospitals, and 16 primary healthcare centres at the time of the study.

Primary health care is responsible for screening cognitive impairments and dementias, mostly carried out in outpatient care. Basic cognitive evaluations are conducted in primary health care and are specified in the Finnish National Current Care Guidelines for memory disorders [22]. Basic cognitive evaluations involve anamnesis and clinical examination, including a cognitive assessment, often CERAD. If a memory disorder risk is identified, patients are referred to a specialist. Additionally, a local clinical pathway clarifies responsibilities for cognitive impairment or dementia care among providers within the region.

Memory disorder diagnostics are performed by specialists through differential diagnosis, based on medical history, anamnesis, and clinical examination, including neuropsychological assessment if needed, structural brain imaging, laboratory tests, and additional testing like cerebrospinal fluid and functional brain imaging. While magnetic resonance imaging (MRI) is the golden standard, some municipalities may only have computed tomography access, which can lead to diagnostic differences.

In some municipalities, specialized geriatric memory clinics in primary healthcare centres are in charge of dementia diagnostics in their respective regions. However, neurology departments are responsible for the diagnostics of working-age and complex geriatric cases. All clinics follow current diagnostic criteria. Regardless of where the diagnosis is made, all new memory disorder cases in Northern Savo were intended to be registered in the KUH memory disorder register at the time of the study. Alternatively, anonymized information about the diagnoses was provided to KUH registers for administrative purposes, aiding in developmental purposes.

We aimed to identify all dementia diagnoses made in Northern Savo in 2020 by using the International Classification Codes 10th Revision (ICD-10) diagnosis codes to identify memory disorder diagnoses from the registers. The VD diagnoses were identified with the header code of F01 (VD). To identify the cases with mixed pathology (AD+VD), we included codes of F00.2* (dementia in AD, atypical or mixed type) and G30 (AD). The diagnoses of AD, LBD, and FTD were identified using codes F00 and G30, G31.0, and G31.83, respectively. For 36.6% of cases, we also had data on both the sex and age of the patients at the time of diagnosis. To observe incidence at the municipality-specific level, we linked each diagnosis to the patient’s municipality of residence. The observation period was from January 1 to December 31, 2020. Data collection was conducted in 2020 and finalized in 2021. To calculate the incidence rates (IRs), we collected demographic data for Northern Savo and its municipalities. The size, age, and gender distribution of the region and municipalities at the start of the observation period (1/2020) were provided by Statistics Finland.

The data were analysed using SPSS Statistics Version 25.0 (IBM, Armonk, NY). Demographic data were expressed as frequencies and percentages or means, standard deviations, and ranges. Analyses were used for comparison between incidences and municipalities. IRs with 95% confidence intervals (CIs) (per 100,000) were computed using Poisson regression model. The IRs were calculated by dividing the number of new cases by the population considered and expressed per 100,000 of the population per year. In the age-specific incidence analyses, cases under 60 years of age were excluded due to the presence of only one incident case in that group. The data were analysed at both regional and municipal levels.

A total of 202 new diagnoses of dementia with vascular component or vascular alone were collected from registers during the 2020 observation period. Among these, 182 mixed-type dementias and 20 cases of pure VD were diagnosed in 2020 in Northern Savo. Table 1 provides demographic information on the diagnosed cases. Of the diagnosed patients, 60.8% were women. The median (SD) age was 80.0 (7.8) years, ranging from 59 to 95 years. From the regional perspective, the annual incidence of dementias with vascular component in Northern Savo in 2020 was 81.4 (95% CI: 70.9–93.4) per 100,000. In the age group of 65 and over, the incidence was 310.8 (95% CI: 280.6–357.0) per 100,000.

Of the incident cases, a total of 55 cases were residents of the city of Kuopio. Among those aged 65 and over in Kuopio, the annual incidence of dementias with a vascular component in 2020 was 210.7 (95% CI: 161.8–274.5) per 100,000. The annual incidence varied strongly between municipalities, from 17.4 (95% CI: 2.46–123.8) to an incidence of 1,160.6 (95% CI: 816.2–1650.3) per 100,000 in the age group of 65 and over in the municipality with the highest incidence.

Table 2 provides information by age and sex on the incident cases of dementia associated with vascular disease. Row and column percentages were calculated. Among both sexes, most diagnoses occurred in the 80–84 age group, with a strong decline in incidence in age groups over 90 years. There were no incident male cases in the 90 and over age group.

Furthermore, Table 3 presents IRs and proportions of major dementia subtypes in Northern Savo in 2020. The annual incidence of AD was 204.6 (95% CI: 187.6–223.2) per 100,000. For LBD and FTD, the corresponding numbers were 9.3 (95% CI: 6.2–13.9) and 4.4 (95% CI: 2.5–8.0), respectively. In our study population, of all incident cases of dementia, the proportion of dementia associated with vascular disease (pure VD and AD+VD) was 23.8%, and the proportion of pure VD was 2.4%. AD, LBD, and FTD numbers were 60.0%, 2.7%, and 1.3%, respectively.

To our knowledge, this is the first population-based study to provide updated incidence numbers of VD (pure and mixed type) in Finland. Our primary goal was to obtain up-to-date epidemiological data on VD in Finland. In our study, dementia associated with vascular pathology accounted for 23.8% and AD for 60.0% of all incident dementia cases. There were only 20 cases of pure VD, with the rest representing mixed-type dementia (AD+VD). For LBD and FTD, the corresponding percentages were 2.7% and 1.3%, respectively. In previous studies, VD is generally recognized as the second most common cause of dementia after AD, accounting for about 15–20% of cases [1‒3]. In a Swedish study in 2021, 23% of dementia cases accounted for mixed-type (AD+VD) dementia and 17% for VD [4]. A recent Finnish national registry study reported VD accounting for 15.4% of cases in 2021 [20]. Our findings are somewhat in line with the previous ones. However, in our study, the incidence of pure VD was considerably low. In a Finnish population-based study in the 1990s, dementia was associated with vascular pathology in 36.1% of cases [17]. Compared to the proportion of dementia cases associated with vascular diseases, there is a decline of 12.3 percentage points to the current numbers.

It has been suggested that advances in the treatment of vascular risk factors may have contributed to a decrease in the incidence of VD [14]. Total brain volumes remain larger today, with less cerebral small vessel disease compared to the 1990s [23]. Furthermore, studies suggest a decline in the age-specific incidence of dementia in Western countries since the late 20th century [13, 14, 23].

In our study population, the percentage of the population aged 65 and over was 25.9%, compared to 22.6% nationally in Finland in 2020 [11]. The United Nations reported that the corresponding percentage for Europe was 12.7% in 1990 and 19.1% in 2020, and is projected to reach 28.1% by 2050 [11]. The proportion of elderly individuals in our study population is already approaching the projected European level for 2050, emphasizing the need for up-to-date data.

The strengths of this study are as follows. The study population was geographically defined, and the catchment area was well defined. The study population represents 4.5% of the Finnish population and is representative of the national population with some exclusions. In 2019, Northern Savo was the unhealthiest region in Finland, in terms of both the age-standardized and unadjusted morbidity indices, according to the Finnish Institute for Health and Welfare’s Morbidity Index, which evaluates multiple disease groups and compares regional morbidity to the national level [24]. Notably, in Northern Savo, a rural area in Finland, cholesterol levels have been observed to be higher compared to urban areas like Turku and Helsinki [25]. Additionally, the educational level of the areas differs. In 2020, in Northern Savo, 45.1% of the population aged 15 and over had completed upper secondary education, and 28.9% had completed tertiary education, compared to 40.6% and 32.6%, respectively, nationally [26]. Furthermore, health care in Finland is largely centralized, and screening and diagnostics for cognitive problems are conducted in the public sector. In Finland, National Current Care Guidelines contribute to consistent diagnostics, while local clinical pathways further standardize practices [22]. As a result, dementia diagnostics are quite standardized. Additionally, in our study, all patients underwent brain imaging by MRI or computed tomography before diagnostics, depending on the local imaging resources. However, MRI is the preferred and suggested imaging modality for its superior sensitivity and specificity in detecting mild vascular pathologies [27]. Furthermore, the local registers used for our case ascertainment are considered comprehensive and are expected to include all diagnosed cases. As a result of these factors, the study is considered to have a high level of accuracy.

There are limitations to our study as follows. Given its regional focus and the heterogeneity between populations in Finland, a national study with a methodology designed to ensure comparable accuracy would be needed for broader representativeness. Moreover, the reported IRs are likely minimum estimates of the true IRs, as some incident VD cases are likely to remain undiagnosed despite efforts to enhance diagnostic accuracy. Cases are more likely to remain unrecognized by health care if there is a more apparent aetiology for cognitive impairment, such as alcohol abuse or traumatic brain injury. Additionally, for stroke patients with severe motoric symptoms, cognitive impairment might receive less attention and, thus, is more likely to remain undiagnosed. Comorbidities affect the validity of diagnosis. Pathological autopsy studies would be the most accurate way of approaching the case. In past autopsy studies, cerebral infarcts have been found in about half of the cohorts of the oldest-old [28, 29]. Additionally, both sex and age data were available for only 36.6% of the incident cases, which may impact the reliability of age- and sex-specific analyses. The cases with both age and sex data do not appear to be a systematically selected subset of the population, suggesting that selection bias is unlikely. Our findings on the age-specific incidence of dementia associated with vascular disease appear to be in line with a recent Finnish national registry study, which reported the highest incidence of new dementia cases in the 75–84 age group [20]. However, future studies with more complete datasets are needed.

Moreover, despite centralized health care in Finland, not all patients seek care for symptoms. Previous studies suggest that older people, especially males, are more often undiagnosed, and those over 90 years are less likely to be diagnosed [30]. Our results were consistent with these findings, as the incidence declined strongly in age groups of 90 and over. According to a Finnish study, dementia is also under-documented in primary health care [31]. Additionally, Danish studies suggest that the validity of registers is greater for AD compared to other ICD-10 subtypes, including VD [32]. Corresponding data are unavailable for Finnish registers.

Our findings indicate significant municipal variation in VD incidence in Northern Savo, independent of age distribution. The IRs in the age group of 65 and older ranged from 17.4 to 1,160.6 per 100,000. Healthcare resource shortages are likely to impact the extensity of diagnostics. For instance, the availability of neuroimaging, particularly MRI, and personnel resources differ municipally and in respect of time.

To our knowledge, the registers used in our study are inclusive of all VD cases to the highest possible extent. However, as a register-based study, diagnosis uncertainty exists. A prospective study with individually confirmed diagnoses would enhance reliability. However, this study provides updated VD incidence numbers in Finland until further research is conducted.

We gratefully thank J. Salpakari for assistance with data collection and T. Selander for statistical guidance.

The Research Ethics Committee of the Northern Savo Hospital District provided ethical permission for the study (ID: 1321/2018). The research followed good scientific practice, and all registered data were anonymized and unique identity codes were blinded. The research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki (and followed the provisions of the Finnish Medical Research Act). Written informed consent was not required.

The authors have no conflicts of interest to declare.

A Tuominen was funded by Finnish Cultural Foundation (No. 65211984).

A. Tuominen was mainly responsible for drafting the manuscript. V. Kärkkäinen made substantial administrative contribution to the study. V. Jetsonen and O. Lappalainen contributed to the acquisition of data for the work. A. Koivisto and P. Jäkälä contributed to the conceptualizing and designing of the study. All authors participated in reviewing the work critically.

Research data are not publicly available due to ethical and privacy reasons.