Article Text

Original research
Trends in prevalence and incidence of registered dementia and trends in multimorbidity among patients with dementia in general practice in Flanders, Belgium, 2000–2021: a registry-based, retrospective, longitudinal cohort study
  1. Simon Gabriël Beerten1,
  2. Antje Helsen1,
  3. Jan De Lepeleire1,
  4. Frans Boch Waldorff2,
  5. Bert Vaes1
  1. 1Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium
  2. 2Department of Public Health, Research Unit for General Practice and Section of General Practice, University of Copenhagen, Copenhagen, Denmark
  1. Correspondence to Dr Simon Gabriël Beerten; simon.beerten{at}kuleuven.be

Abstract

Objectives With the ageing of our population, it seems plausible that the prevalence of both dementia and multimorbidity will increase in the following decades. The aim of this study is to examine the trends in prevalence and incidence of registered dementia and trends in multimorbidity in patients with dementia in general practice in Flanders.

Design Retrospective, longitudinal cohort study.

Setting Primary care practices across Flanders, Belgium.

Participants Patients included in the Intego database.

Methods Data were collected from the Intego database, a Belgian general practice registration network, from 1 January 2000 to 31 December 2021. Joinpoint regression, the Cochran-Armitage test and Jonckheere-Terpstra test were used for the trend analysis.

Results Data from 149 492 unique patients aged 65 years and older were available. From 2000 to 2021, 3835 incident cases of dementia were found. The age-adjusted prevalence of registered dementia significantly increased during this study period, from 1.19% to 2.43% (average annual percentage change (AAPC) 3.3; 95% CI 2.7 to 4.0). Incidence increased from 3.68 to 5.86 per 1000 patient years overall (AAPC 1.8, 95% CI −2.0 to 5.7), but declined in recent years (annual percentage change −8.1, 95% CI −14.8 to −0.8). Almost three-quarters of the patients with dementia (74.8%) suffered from multimorbidity (three or more comorbidities) and this increased significantly during the study period (p=0.0031). By 2021, 86.7% and 74.8% of the patients with dementia suffered from two or more or three or more chronic conditions, respectively. Hypertension (47.9%), osteoarthritis (29.7%) and lipid metabolism disorders (25.7%) were the most prevalent conditions.

Conclusions The prevalence of registered dementia doubled over a 22-year time period, mirroring the increasing health burden by this disease globally. Furthermore, three-quarters of the patients with dementia suffered from multimorbidity, underlining the urgent need to implement comorbidity management and patient-centred care in dementia.

  • Dementia
  • EPIDEMIOLOGY
  • PRIMARY CARE

Data availability statement

Data are available on reasonable request. The datasets generated and analysed during the current study are not publicly available due to inclusion of protected health information but can be made available subsequent to further deidentification on reasonable request to the corresponding author (SGB).

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • The main strength of this study is the inclusion of primary care data instead of hospital data.

  • We used two operational definitions of multimorbidity in our analyses.

  • Our study sample is representative for the Flemish population.

  • We did not have information on how the physicians diagnosed individuals with dementia, or on the severity of the studied comorbidities.

Introduction

Dementia is an age-related clinical syndrome that comprises multiple underlying pathophysiological changes in the brain, which eventually lead to an impairment of mental functioning, activities of daily living and behaviour. The WHO states that there are 10 million new diagnoses of dementia annually and that 50 million people are currently living with dementia worldwide.1 Furthermore, the total number of people living with dementia is expected to nearly double every 20 years.1–3 In Flanders, the estimated prevalence of dementia increases from 10% at the age of 65 to 20% at the age of 80 and up to 40% at an age of 90 years and older.4 5 Dementia is more actual than ever, since the WHO has recognised the disease as a public health priority.1 The management of health and social care for people with dementia should comprise early diagnosis, effective treatment, education and support for carers and advanced care planning. For many of these aspects, primary care practitioners are best placed to coordinate care and to deliver person-centred care.2

Another age-related phenomenon is multimorbidity, defined as the coexistence of two or more chronic diseases in the same person.6 7 This matter too has gained interest over the last years, because studies suggests that multimorbidity is associated with an increased risk of mortality, poor functional status, reduced quality of life and greater use of healthcare services.8 Multimorbidity implies a shift from the single disease paradigm to patient-centred and integrated care. This should also be one of the core competencies of a primary care practitioner.6 9–11 With the ageing of our population it seems plausible that the prevalence of both dementia and multimorbidity will increase in the following decades.4 9 This is an important evolution because growing evidence suggests that additional comorbidity can accelerate the cognitive decline in patients with dementia. In addition, dementia can complicate diagnosis and clinical care of other medical conditions and may undermine the patients’ self-management.12 Nevertheless, only limited research has been performed on the changes in dementia prevalence and incidence and in related multicomorbidity and comorbidity. This is reflected in the lack of accurate data on dementia prevalence and in the lack of applicable guidelines on multicomorbidity and comorbidity in patients with dementia.13

Given the fact that Belgium struggles with an ageing population, and that the dementia burden is thus likely to become bigger in the future, we will attempt to fill this gap in literature. We aim to provide a frame of reference regarding the evolution and the extent of the problem of dementia and its comorbidities, to inform and guide care policy. To that end, this study analysed the trends in prevalence and incidence of registered dementia and the trends in dementia-related multicomorbidity and comorbidity from 2000 to 2021 in general practice.

Methods

Data collection

Data were obtained from the Intego database, a Belgian general practice-based morbidity registration network at the Department of General Practice of the KU Leuven.14 In 2021, the Intego database included data from over 700 000 different patients. This practice population was calculated to represent 4.76% of the Flemish population by age and gender. As of 2022, Flanders accounts for approximately 58% of the total Belgian population.

The Intego database contains pseudonymised information about diagnoses, drug prescriptions, laboratory results and biomedical parameters. All data are obtained from the electronic health records of over 100 practices evenly spread throughout Flanders.14 15 These data are all registered using computer-generated keywords, internally linked to codes. New diagnoses are thus classified according to a thesaurus automatically linked to the International Statistical Classification of Diseases of Primary Care (ICPC-2) and the International Statistical Classification of Diseases and related Health Problems-10.14 16 These classification systems are accepted by the WHO for coding in primary care.17 Currently, new data are collected on a weekly basis, subsequently entered into a central database and then historically accumulated for each patient. The quality of the data is guaranteed first by selecting only those general practices with a high registration performance (>80% coded diagnoses). Second, the database is externally validated by means of national and international comparisons and internally validated using tracers as a measure of correctness.14 16

From this database, we selected all incident cases older than 65 years of age for whom the ICPC-2 code P70 (dementia) was registered for the first time, between 1 January 2000 and 31 December 2021. Prevalent cases were defined as having had a diagnosis of P70 ever before. No further differentiation was made among the subtypes of dementia. The yearly contact group (YCG) was used as the denominator. The YCG for a given year is defined as all patients who attend a certain practice at least once in that year. Research demonstrated that in countries without capitation the YCG approximated 80% of the practice population, and this measure has been validated before.18 Furthermore, in the elderly, the YCG nearly approached the entire practice population, since most elderly tended to visit their GP at least once annually.18

Multimorbidity and comorbidity

Multimorbidity is generally defined as the coexistence of two or more chronic diseases in the same person, whereas comorbidity is defined as any disease or risk factor that may interact with one index disease.6 In our study, multimorbidity was analysed as either the coexistence of two or more chronic diseases or as the coexistence of three or more chronic conditions in the same patient, in order to identify those patients with higher care needs.9 Dementia itself was accounted for as one chronic condition. This implied that multimorbidity was present from the moment that one or two other chronic conditions were diagnosed, depending on the operational definition mentioned above.

We analysed comorbidities in incident dementia cases. A comorbidity was considered present when it was diagnosed at any point before the dementia diagnosis.

A list of chronic conditions, as implemented earlier19 and adapted by Smeets et al16 for similar research, was used to study multimorbidity. This list is available in attachment. Parkinson’s disease (N87) and HIV/AIDS (B90) were excluded from the index list because of possible interaction with the aetiology of dementia. Psychiatric comorbidities were excluded from the list as well, because of diagnostic difficulty between psychiatric and dementia symptoms and the risk of potential misclassification, especially with early symptoms.20 We did perform a separate analysis afterwards. The same index list of chronic diseases was used to examine the most prevalent chronic comorbidities in patients with dementia.

For comparison, we added a control group which was matched on age, gender, practice and YCG, with a ratio of 3 controls to one case. The same analysis as explained below was then carried out for this ‘dementia-free’ control group.

Data analysis

Five-year age groups were formed starting from 65 years of age to 90 years and older. First, we calculated the prevalence (/100 patients) and incidence (/1000 patient years) of dementia by gender and by age. Next, the trends in dementia prevalence and incidence rates for every age group and gender were calculated from 2000 to 2021 using a joinpoint regression analysis. A joinpoint is a point in the trend curve where a statistically significant change in the trend over time is observed. The annual percentage change (APC) and the average APC (AAPC) can be calculated from this joinpoint regression model.16 21 For this joinpoint regression analysis the SEER*Stat software (Joinpoint Trend Analysis software from the Surveillance Research Programme of the US National Cancer Institute (http://surveillance.cancer.gov/jointpoint)) was used. To calculate the prevalence of multimorbidity (/100 patients) and its time trend in incident dementia cases, the 22-year time period was divided into five intervals (2000–2003, 2004–2007, 2008–2011, 2012–2016 and 2017–2021) and the Cochran-Armitage test was used. This test for trend is used to associate a binary response to an ordinary variable with k categories.16 22 This analysis was performed for the two operational definitions of multimorbidity. Finally, the trend in comorbidity profiles in incident dementia cases was explored for the most prevalent comorbidities, using the Jonckheere-Terpstra test for these same time intervals. The Jonckheere-Terpstra trend test is used to associate a continuous response with an ordinal variable with k categories.16 22 Both analyses were performed using the R Software V.4.0.3 (Free Software Foundation, Boston, Massachusetts, USA) (DescTools and clinfun packages).16 For each analysis, the level of significance was set on p<0.05.

Patient and public involvement

None.

Results

Study population

Between January 2000 and December 2021, we found a total of 3835 incident cases with registered dementia aged 65 years and older. The mean age of the study population was around 82 years old. Over 60% of the study population were women, and their share remained stable during the study period.

Trends in age-standardised prevalence and incidence of registered dementia (2000–2021)

The age-standardised prevalence of registered dementia showed a significant upward trend for the entire study population from 1.19% in 2000 to 2.43% in 2021 (AAPC 3.3; 95% CI 2.7 to 4.0). A similar, significant upward trend was seen in males (AAPC 4.2; 95% CI 3.4 to 5.0), with a significant APC of 5.7% (95% CI 5.0% to 6.4%) from 2000 to 2017, afterwards remaining stable. The overall trend for women was significantly rising as well, consisting of a steep increase in prevalence from 2002 to 2005 (AAPC 11.4; 95% CI 2.4 to 21.2) and less steep, positive trends from 2005 to 2012 (AAPC 1.3; 95 % CI 0.1 to 2.5) and 2012 to 2016 (AAPC 5.7; 95% CI 2.1 to 9.5). Furthermore, these rising trends showed a more than doubling in prevalence for the total, female and male populations (figure 1, table 1, online supplemental appendix figure A1).

Table 1

Trends in dementia prevalence (/100) for every 5 year age group and gender from 2000 to 2021

Figure 1

Age-standardised prevalence (/100) and incidence (/1000) of dementia in the Intego database for the entire population, females and males, 2000–2021.

The age-standardised rate for incidence showed a non-significant rising trend for the total and female population (AAPC 1.8; 95% CI −2.0 to 5.7 and AAPC 1.5; 95% CI −2.6 to 5.8, respectively), with a significant decrease from 2018 to 2021 (APC −8.1; 95% CI −14.8 to −0.8 and APC −8.7; 95% CI −15.9 to −0.9, respectively). Yet in men, a significant rise in incidence was noted from 3.40 per 1000 patient years in 2000 to 5.39 per 1000 patient years in 2021 (AAPC 3.3; 95% CI 2.2 to 4.5) (figure 1, table 2).

Table 2

Trends in dementia incidence rates (/1000) for every 5-year age group and gender from 2000 to 2021

By the end of the study period in 2021, the prevalence of registered dementia ranged from 0.64% at the age of 65–69 to 2.3% at the age of 75–79 and up to 10.32% at the age of 90+ years old. (table 1) The prevalence per age group showed a significantly rising trend for the age groups from 65 to 74 years old, 80–84 and 90+ years old, as demonstrated by their respective AAPC’s in table 2 . For the age group 75–79 and 85–89 years in men, an overall significant upward trend was seen (table 1, online supplemental appendix figure A1).

Table 3

Prevalence (/100) of multimorbidity as >2 or >3 chronic conditions and comorbidity profiles in incident dementia cases and matched controls using the Jonckheere-Terpstra and the Cochrane-Armitage tests for trend from 2000 to 2021

The AAPCs of the age-specific and gender-specific incidence rates showed mostly positive and significant increases in the 75+ age groups (table 2). Again, a significant decline in the incidence rate was seen for age groups 75–79 (only for women) and 90+ (both total and in women).

Trends in multimorbidity and comorbidity profiles (2000–2021)

This analysis revealed a high prevalence of registered multimorbidity. When defined as the coexistence of two or more chronic conditions, almost 90% of the patients with dementia suffered from multimorbidity. This proportion did not increase significantly (p=0.5749) over the past 22 years. When defined as the presence of three or more chronic conditions, the prevalence of multimorbidity was lower. Nonetheless, the prevalence remained high and demonstrated a significant increase from 64.5% to 74.8% (p=0.0031) during the study period as well (table 3).

Hypertension (47.9%), osteoarthritis (29.7 %) and lipid metabolism disorders (25.7%) had the highest prevalence and remained the most prevalent comorbidities throughout the entire study period. A significant increase in prevalence was seen for atrial fibrillation, hypertension, atherosclerosis, ischaemic heart disease, diabetes, lipid metabolism disorders, urinary incontinence and back pain. Ischaemic heart disease (p=0.0163) was the only comorbidity to show a slight but significant decrease in prevalence by 2021. Chronic kidney disease showed a slight non-significant decline. The prevalence of the other conditions remained stable. The most common comorbidities at the start and at the end of the study period are listed in table 4.

Table 4

Most frequent comorbidities at the start and at the end of the study period

Discussion

The most interesting findings of our analyses were the overall significantly rising trends and doubling of the prevalence of registered dementia from 2000 to 2021. The incidence rates increased as well, but with a decline in recent years. Regarding multimorbidity, our results revealed that the vast majority of the patients with dementia suffered from two or more chronic conditions. Furthermore, the prevalence of registered multimorbidity (3+ comorbidities) in dementia patients showed a significantly increasing trend over the past 22 years.

Comparison with other studies

Trends in dementia prevalence and incidence

The proportion of women in our analysis corresponded to the proportion mentioned in similar studies. Bauer et al,23 Clague et al24 and Chen et al25 reported a female proportion of 70.4%, 70.6% and 63.6%, respectively.

The prevalence of dementia in this analysis was lower than previously reported in literature. According to the latest report of the Alzheimer Cooperative Valuation in Europe (Alcove) from 2013, the estimated prevalence of dementia in patients aged 65 years and older in the whole of Europe was 7.23%.26 The Belgian Qualidem project stated that prevalence numbers for dementia in Belgium varied from 6.3% to 9.3%.27 Both organisations thus reported a higher prevalence than in our results. This difference could be explained by the study design and data collection method. Real-world data were used in which a dementia diagnosis was based on routine clinical recording instead of an assessment through a set-out protocol. This may have led to an underestimation because of multiple reasons. A first explanation is underdiagnosis. This could have been due to difficulties with early dementia diagnosis,28 although over the past years diagnostic criteria and neurological examination and imaging have improved. Also, therapeutic nihilism, the belief that diagnosis is less useful because of the limited treatment options, could have played a part in underdiagnosis.23 24 Second, under-registration could also explain the low prevalence numbers. Reasons for under-registration could have been the perception of the dementia diagnosis as a loaded term, the registration as a symptom diagnosis ‘memory disturbance’ (ICPC-2 code P20) instead of dementia or diagnosis in the hospital or at a nursing home. The lower prevalence in our analysis, however, was in line with the results of other studies with similar study designs. In a retrospective, longitudinal cohort study on data of the Clinical Practice Research Datalink in the UK, Donegan et al reported a rising prevalence of dementia in an all-age population of 0.41% in 2005 to 0.82% in 2015.29 Also, a cross-sectional analysis performed by Clague et al24 on data of patients aged 65 or older from 314 general practices, provided by the Primary Care Clinical Informatics Unit in the UK, demonstrated a dementia prevalence of 3.6%.24

The prevalence of dementia increased from 2000 to 2021. The trend analyses for dementia prevalence showed significantly upward trends for nearly all age groups, except 75–79 and 85–89 years old. In addition to a clear increase in prevalence, a more than doubling of the total age-adjusted prevalence from 1.19% to 2.43% was noted during the study period. For the age-specific and gender-specific prevalence too, a doubling was observed. These findings correspond to the prediction of the WHO in 2012 that the number of people living with dementia worldwide would double in the following 20 years.1 For high-income countries, however, a decreasing trend in incidence has been described, with a later onset of dementia, known as the ‘compression of cognitive morbidity’.30–32 We found a significantly decreasing incidence of registered dementia in the last years of our analysis. This corresponds to a finding already reported by Satizabal et al, in an analysis of Framingham Heart Study participants.33 This was hypothetically linked to improvements in cardiovascular health.

We should thus further investigate whether this increasing trend reflects a true increase in prevalence. A true increase in prevalence could be caused by a growing elderly population, increased incidence and a prolonged survival of patients with dementia. However, given our study method and given the results of a declining incidence, it seems probable that the rising prevalence numbers could at least partly be explained by an increased awareness of dementia and a better registration in electronic medical records.29

Trends in multimorbidity and comorbidity profiles

The data on multimorbidity prevalence showed significantly rising trends from 2000 to 2021, if defined as the presence of three or more chronic conditions. More than half of the patients with dementia suffered from two or more additional comorbidities and therefore from an even greater care need than those with only one additional chronic condition.9 These results corresponded to the findings of a similar study performed by Doraiswamy et al, who reported a prevalence of 61% of 3 or more comorbidities in Alzheimer’s disease patients.34 Again, whether this rising trend reflects a true increase in prevalence or is merely due to an improved registration should be investigated further. As expected, more than 70% of the patients with dementia are suffering from multimorbidity.

The findings on the most common comorbidities in patients with dementia were in line with our expectations and with the results of similar studies (online supplemental appendix table A1). Bauer et al and Clague et al reported hypertension, lipid metabolism disorder, type 2 diabetes, ischaemic heart disease, atherosclerosis, heart failure, cerebrovascular accident and osteoarthritis among the most prevalent comorbidities.23 24 Hypertension, lipid metabolism disorder and atherosclerosis, however, are very common in subjects without dementia as well (as evidenced by our matched control group) and do not necessarily bring about a large increase in care needs. In contrast, other frequent chronic conditions such as malignancy, ischaemic heart disease and diabetes do increase care needs. Also, the significant rise in prevalence of urinary incontinence is noteworthy, because this is known to be a risk factor for permanent stay in a residential setting.23

Implications for policy and practice

In this study, we found that the prevalence of dementia unsurprisingly kept increasing, highlighting the problem of our ageing population. Furthermore, the high and increasing burden of comorbidity in dementia complicates the problem of the disease itself and makes the need for ‘comorbidity-centred’ care in this population dramatically urgent. Future research should thus focus on making comorbidity care a necessary pillar of patient-centred care in the older population.

Strengths and limitations

The strengths of this study included the use of data from an existing primary care database instead of in-hospital data,9 23 the use of two operational definitions of multimorbidity,9 the exclusion of acute conditions to avoid unnecessary inflation of multimorbidity prevalence35 and a minimisation of the risk of loss to follow-up by calculating multimorbidity in incident dementia cases. Lastly, our study population was a representative sample of the Flemish population, although representativeness for other populations might be limited.

A limitation of this study was the fact that diagnoses were based on physician documentation and could not be validated by clinical examinations, neither for dementia, nor for comorbidity. Also, the database did not provide any information on whether the dementia diagnosis was made by a general practitioner or through a diagnostic evaluation in a memory clinic. Although useful in its own right, a study in general practice might miss nuances and trends unique to the hospital situation. Ideally there would be multiple sources of data, thus limiting confounding and increasing external validity.36 Third, we may have missed some patients with early stage dementia, because we did not include patients with memory disturbance (ICPC-2 code P20). Another limitation was the lack of information on the severity of comorbidities, which made it impossible to assign a weight to different chronic conditions and to classify them by their respective care needs.10

Conclusion

This study demonstrated that the incidence of registered dementia rose significantly over the past 22 years. The prevalence of registered dementia increased significantly in nearly all age groups and even more than doubled for the entire population. The incidence rose as well, but with a significant decline in recent years.

Around three-quarters of the patients with dementia appeared to be suffering from multimorbidity by means of two additional chronic health conditions and this proportion showed a rising trend as well. Further research on how to implement the management of comorbidity and patient-centred care in the care for patients with dementia is needed in order to set up applicable guidelines for this specific population of elderly.

Data availability statement

Data are available on reasonable request. The datasets generated and analysed during the current study are not publicly available due to inclusion of protected health information but can be made available subsequent to further deidentification on reasonable request to the corresponding author (SGB).

Ethics statements

Patient consent for publication

Ethics approval

The procedures were approved by the Belgian Privacy Commission (no. SCSZG/13/079) and the ethical review board of the Faculty of Medicine at the KU Leuven (no. ML1723). Intego was waived the need for individual informed consent, but operates under an opt-out procedure for patients who do not wish their data to be included. This was approved by the aforementioned ethical review board.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Contributors AH and SGB wrote the manuscript. BV, JDL and FBW provided input on the study design and cowrote the paper. BV acts as the study's guarantor. All authors approved the final manuscript.

  • Funding This work was supported by a research project from the Research Foundation Flanders (FWO).

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.