Abstract
INTRODUCTION
Mild cognitive impairment (MCI) is a common mental disorder affecting around 16% of elderly people without dementia [1]. The incidence rate of the condition has been estimated at approximately 63.6 per 1,000 person-years [2]. MCI is considered an intermediate state between normal cognition and dementia [3]. Studies of patients in German primary care practices have shown that one in four patients develops dementia within the three years following initial MCI diagnosis, underlining the importance of intensive management and treatment of this predementia condition [3].
The risk factors of MCI have been analyzed by several authors in recent years [4–12]. A 2001 study showed that elevated midlife serum cholesterol levels significantly increased the risk of developing MCI, where the regression model was adjusted for age and body mass index [4]. Two years later, it was discovered that ethnicity, educational level, low modified Mini-Mental State Examination and Digit Symbol Test scores, cortical atrophy, infarctions identified by MRI, and measurements of depression were positively associated with MCI diagnosis [5]. In Germany, Luck and colleagues found in 2010 that older age, vascular diseases, the ApoE ɛ4 allele, and subjective memory complaints increased the odds of developing MCI in the future [6]. More recently, a study including 1,278 individuals suggested that traumatic brain injury may decrease the age of onset of MCI [10].
Based on the findings of these previous works, the goal of the present study was to reanalyze the risk factors for MCI development in German primary care practices.
METHODS
Database
The Disease Analyzer database (IMS Health) compiles drug prescriptions, diagnoses, and basic medical and demographic data obtained directly from the computer systems of a representative sample of resident physicians throughout Germany. The data are generated directly from the computers in the physicians’ practices via standardized interfaces and provide daily routine information on patients’ diseases and therapies. Each practice transmits patient data stored in the physician’s computer to IMS on a monthly basis. Before transmission, the data are encrypted for data protection purposes and contain, in similar scope and detail, the information in the patient files in the doctor’s practice.
The Disease Analyzer database provides a complete listing of all relevant patient details for each practice. The data obtained directly from the practice computers are checked for plausibility, linked to relevant additional information such as the price of medicinal products, ATC- (The Anatomical Therapeutic Chemical Classification) and ICD- (International Classification of Diseases) coded, saved, and updated on a monthly basis. In compliance with the applicable data protection legislation, all information contained in the database is anonymized.
The validity of the reported data is monitored by IMS based on a number of quality assurance criteria (e.g., completeness of documentation, linkage of diagnoses and prescriptions) [13]. Finally, this database has already been used in several studies focusing on dementia and other psychiatric disorders [14–16]. Because the present study entailed the retrospective analysis of anonymized data from primary care practices all over Germany, no specific ethical consent was obtained.
Study population
A total of 3,604 patients with initial diagnosis of MCI (ICD 10: F06.7) were included in the study between January 2010 and December 2015. In addition, 3,604 controls without MCI were also included and matched with cases on the basis of age, sex, type of health insurance, and physician. Practice visit records were used for both cases and controls to verify that there had been 3 years of continuous follow-up prior to the index date. Thus, a total of 7,208 subjects were observed.
Study outcome
The main outcome of the study was the risk of MCI depending on predefined risk factors. Several disorders potentially associated with MCI were determined based on primary care documentation using ICD-10 codes: diabetes (E10-E14), hypertension (I10), obesity (E66), hyperlipidemia (E78), stroke including transient ischemic attack (I63, I64, G45), Parkinson’s disease (G20, G21), intracranial injury (S06), coronary heart disease (I20–I25), depression (F32-33), anxiety disorder (F41), sleep disorder (G47), and mental and behavioral disorders due to alcohol use (F10). Psychiatric diagnoses including MCI documented by primary care physicians were initially made and secured by neurologists/psychiatrists.
Statistical analyses
Descriptive analyses were obtained for all demographic variables and mean±SDs were calculated for normally distributed variables. Multivariate logistic regression models were fitted with MCI as a dependent variable and other disorders as potential predictors. p-values <0.05 were considered statistically significant. Analyses were carried out using SAS version 9.3.
RESULTS
Patient and control characteristics
Socio-demographic data pertaining to the individuals included in this study are shown in Table 1. The mean age was 75.2 years (SD: 9.1 years) and 45.3% of patients were men. Several disorders occurred significantly more frequently in cases than in controls: hypertension (70.5% versus 54.6%), hyperlipidemia (43.8% versus 27.4%), diabetes (32.3% versus 24.1%), coronary heart disease (28.6% versus 19.5%), depression (26.6% versus 13.0%), history of stroke (23.5% versus 12.3%), sleep disorder (18.5% versus 10.6%), obesity (9.9% versus 5.1%), anxiety (9.5% versus 3.6%), Parkinson’s disease (3.5% versus 2.1%), mental and behavioral disorders due to alcohol use (2.0% versus 0.9%), and intracranial injury (1.3% versus 0.4%).
Associations with MCI
The results of the multivariate logistic regression model for the development of MCI in German primary care patients are displayed in Table 2. MCI development was found to be associated with 12 disorders: intracranial injury, anxiety disorder, depression, mental and behavioral disorders due to alcohol use, stroke, hyperlipidemia, obesity, hypertension, Parkinson’s disease, sleep disorder, coronary heart disease and diabetes, with odds ratios ranging from 1.13 (diabetes) to 2.27 (intracranial injury).
DISCUSSION
The present retrospective study including 3,604 MCI cases and 3,604 controls unaffected by MCI showed that this cognitive condition was associated with 12 psychiatric and medical disorders to a significant extent.
The presence of intracranial injury led to a 2.3-fold increase in the risk of developing MCI. This association has been the center of controversial research in recent years [10, 17–20]. In 2008, Rapoport et al. found no significant relationship between MCI and intracranial injury as rates of MCI or dementia were no higher in patients with traumatic brain injury than in controls [17]. Dams-O’Connor et al. corroborated these findings five years later, estimating that individuals aged 65 years and over who had reported a history of intracranial injury were not at an elevated risk of being diagnosed with dementia or Alzheimer’s disease [18]. By contrast, a retrospective cohort study performed between 2005 and 2011 in the United States found a 1.46-fold increase in the odds of developing dementia after brain trauma [19]. Interestingly, moderate to severe trauma had a significant impact across all ages, whereas mild injury was a significant risk factor in older patients only [19]. Three hypotheses may explain the close relationship between cognitive impairment and traumatic brain injury: (i) the triggering of a neurodegenerative process, (ii) the acceleration of a neurodegenerative cascade that already existed before the trauma, and (iii) the reduction of cognitive reserve [19]. In line with the work of Gardner and colleagues, Li et al. discovered that traumatic brain injury is a risk factor for cognitive decline in older adults, and is associated with an earlier age of onset of MCI and Alzheimer’s disease [10].
Another important result of the present study is the significant relationship it reveals between MCI and both anxiety and depression. Again, several authors have obtained contrasting findings in this regard. For example, in their 2009 study, Becker and colleagues discovered no strong evidence to support the hypothesis that mood disturbance is linked with future diagnosis of dementia [21]. Conversely, another work published the same year and including 1,487 Chinese individuals showed a positive association between depression and cognitive decline in men [22]. In 2010, using data from the Framingham Heart Study, Saczynski et al. found that depressed patients have a more than 50% higher risk of developing dementia and Alzheimer’s disease [23]. The role of depression in the etiology of cognitive decline may be explained by its negative impact on sleep and behavior, which could indirectly favor neurodegeneration. The association between MCI and both depression and anxiety may also involve diabetes. There is evidence for a bidirectional relationship between depression and diabetes [24]. Since the latter condition is a known risk factor for cognitive impairment and dementia [7, 25], it could be behind the positive association between anxiety/depression and MCI. In line with these two hypotheses, our analysis showed that the presence of sleep disorders or diabetes increased the odds of being diagnosed with MCI. Interestingly, obesity, hyperlipidemia, hypertension, and coronary heart disease were additional risk factors for the diagnosis of cognitive impairment. This result may be explained by the fact that people with these chronic conditions are often affected by type 2 diabetes, emphasizing the close relationship between all these disorders.
MCI was also associated with mental and behavioral disorders due to alcohol use. More than a decade ago, it was found that there was a U-shaped relationship between alcohol consumption and the risk of MCI in patients aged between 65 and 79 years [26]. In 2007, a study including 121 MCI patients discovered that consuming no more than one unit of alcohol per day may decrease the rate of progression of MCI to dementia [27]. Two years later, Xu et al. corroborated these findings, estimating that consuming small amounts of alcohol may reduce the rate of cognitive decline in people affected by MCI, whereas alcohol misuse may be a risk factor for its progression to dementia [28]. Since the association between alcohol and MCI is not linear, it has been suggested that the alcohol consumption does not have a direct impact on cognitive functions but rather is a more general indicator of a person’s lifestyle, either healthy or unhealthy [26].
Finally, a history of stroke or Parkinson’s disease was associated with a higher risk of being diagnosed with MCI. These results are in line with the literature, as a previous observational analyses indicated that almost one in three individuals develops MCI in the three years following the first stroke [29]. In 2009, Knopman and colleagues discovered that stroke was associated with both nonamnestic and amnestic MCI to a significant extent [30]. Stroke had an impact on all cognitive domains other than memory, and particularly on attention and executive functions [30]. Numerous studies have examined the relationship between Parkinson’s disease and MCI. A Danish work showed that the mean annual decline on the Mini-Mental State Examination (MMSE) is one point in patients affected by this neurological disorder, highlighting the importance of intensive management and treatment for individuals with Parkinson’sdisease [31].
The present study is subject to several limitations. First, the assessment of MCI diagnoses and co-morbidities was based solely on ICD codes entered by general practitioners. No data from neuropsychiatric practices were analyzed, as the documentation of co-morbidities such as diabetes, obesity or coronary heart disease by neuropsychiatrists is often incomplete. Another limitation of this study was the unavailability of information concerning patient compliance with regard to the treatments prescribed. In addition, data on socioeconomic status and physical activity were lacking. The main strengths of the study are the number of patients and the number of primary diagnoses included in the statistical analysis.
Overall, the present study found that twelve disorders were associated with MCI to a significant extent. The three diagnoses with the strongest association were intracranial injury, anxiety and depression. Further analyses are needed to gain a better understanding of the potential impact of the treatment of these disorders on the risk of developing MCI. Moreover, the research in this area is very important as new medications were developed which could potentially delay the onset of dementia [32].
