Prevalence and relation of dementia to various factors in Parkinson's disease


Abdul Qayyum Rana, MD, FRCPC, Parkinson's Clinic of Eastern Toronto and Movement Disorders Center, 404-2863 Ellesmere Road, Scarborough, ON M1E 5E9. Canada. Email:


Aims:  Parkinson's disease is a chronic neurodegenerative disorder characterized by bradykinesia, rigidity, and resting tremor. Dementia, among its non-motor symptoms, is a debilitating complication affecting intellectual functioning. The aim of the present study was to determine the prevalence of dementia in Parkinson's disease and its relation to age, gender and stage of the disease.

Methods:  A retrospective chart analysis was performed on Parkinson's disease patients seen in a community-based Parkinson's disease and movement disorder clinic between 2005 and 2010.

Results:  A total of 310 patients were included in this survey, among whom 61 patients (19.7%) with Parkinson's disease met the criteria for dementia. Age was found to be a significant factor in developing dementia, with 90% of patients with dementia aged ≥70. Gender, however, was not correlated with dementia in Parkinson's disease. On analysis of stage at which dementia developed, progression of the disease was positively correlated with prevalence of dementia.

Conclusions:  As age increases, the chances of developing dementia increase. Dementia, contrarily, is not selective between genders. The likelihood of developing dementia increases as the stage of disease advances. Further research is required in order to understand underlying mechanisms of dementia in Parkinson's disease.

PARKINSON'S DISEASE (PD) is a neurodegenerative disease primarily characterized by bradykinesia, rigidity and resting tremor. Its neuropathology manifests in various forms, including both motor and non-motor symptoms. Dementia, a non-motor symptom of PD, is a cognitive disorder characterized by a general loss of intellectual abilities involving impairment of memory, judgment and abstract thinking as well as changes in personality. A prevalence study of the global population predicted that 81.1m people will have dementia by 2040. This translates into seven new cases every second.1 Various other studies have investigated the association and incidence of dementia in PD. According to Anderson, dementia may develop in 20–40% of patients diagnosed with PD.2 Most patients with PD may initially have mild cognitive deficits as a result of neuropathological changes caused by PD itself. Some of the symptoms of Parkinson's disease dementia (PD-D) include impaired planning, lack of organization of goal-directed activity, difficulty with set shifting, visuospatial deficits, bradyphrenia, impaired problem solving, and hindrance in learning and memory, in general. Most patients also exhibit signs of fluctuation in attention and cognition, mood and personality problems, hallucinations, and psychosis. Language and praxis, however, remain intact until the late stage of the disease. The risk for developing dementia is sixfold higher in PD patients compared to age-matched controls.3–6 Approximately 40% of patients may meet criteria of dementia at some stage in Parkinson's disease. Once patients develop evident dementia, their condition may deteriorate further with increasing disability and the need for long-term placement and care. The pathological changes in PD-D consist of Lewy body accumulation, particularly when they affect the neocortex.7,8

Initially, PD-D patients experience slower thinking than normal and may require additional time to respond to questions. Also, patients with memory and cognitive deficits (without PD) are likely to develop PD later in the course of that illness. Most patients with Parkinson's diseases have only mild memory problems at the beginning but as time and disease progress, the condition worsens. Sometimes these cognitive and behavioral problems can become even more disabling than actual motor disability. Depression, levodopa-induced hallucinations, tobacco consumption, advanced age at the onset of Parkinson's disease, and disease severity are important predictors of the progression of cognitive impairment in Parkinson's patients.2,9,10

Previous studies have indicated that improper functioning of the dopaminergic, cholinergic, and adrenergic systems results in cognitive defects.2 Thus, cholinesterase inhibitors, like the ones used in Alzheimer's disease treatment, may provide some relief to PD-D patients, further suggesting a similar mechanism at the neuropathological level.2 Effective treatments include proper diet and lifestyle, pharmacologic treatments, and exercise and physical therapy. Although exercise is not directly linked to dementia, it has been shown to improve mental state, in general, allowing one to cope better with dementia.2

The aim of the present was study was to describe the prevalence of PD-D and its relation to stage of PD, age and gender. Dementia on its own is a disabling condition and when it is coupled with PD, it is a frequent cause of long-term care placement. Hence, patients should be thoroughly screened over the course of the PD for dementia.



This is a retrospective cross-sectional study of 310 PD patients assessed and followed in a community-based Parkinson's disease and movement disorder clinic between 2005 and 2010. The stage of PD was diagnosed according to the modified Hoehn and Yahr (HY) classification system, which encompassed stages from 1 to 5, increasing from least to most severe.

In the patient group, 61 (19.7%) met the criteria for dementia while the other 249 did not (Table 1). The mean age of the entire group was 74.8 ± 10.28 years. On average, the men were 74.8 ± 10.29 years old and the women were 74.7 ± 10.30 years old. The study included 175 male patients (56.5%), of whom 40 (12.9%) developed dementia. In contrast, among 135 female patients (43.5%) only 21 (6.77%) showed signs of dementia. Approximately 39.4% of the patients were diagnosed with HY stage 2, 17.4% with stage 2.5, 20.3% with stage 3, and 15.2% with stage 4. The other stages, stage 1, stage 1.5, and stage 5, were excluded during the data analysis due to lack of patients in these stages. This would interfere with statistical analysis and thus, were omitted.

Table 1.  Patient characteristics vs presence of PD-D
 PD-D%PD-nonD%Total%Age (years) mean ± SD
  1. PD-D, Parkinson's disease with dementia. PD-nonD, Parkinson's disease without dementia.

Male4012.9013543.5517556.4574.8 ± 10.29
Female216.7711436.7713543.5574.7 ± 10.30
Stage 100.0020.6520.6569.5 ± 16.26
Stage 1.500.0020.6520.6551.0 ± 8.49
Stage 2206.4512239.3514245.8170.4 ± 10.56
Stage 2.592.904514.525417.4275.1 ± 10.34
Stage 3175.484614.846320.3276.8 ± 6.85
Stage 4154.843210.324715.1680.6 ± 7.77
Total6119.6824980.32310100.0074.8 ± 10.28

Further, the ethnicity of each patient was also recorded. The patients were distributed across seven different groups, namely: Afro-American (0.97%, n = 3), Asian (6.13%, n = 19), Caucasian (57.10%, n = 177), Indian (26.77%, n = 83), Mediterranean (2.58%, n = 8), Middle Eastern (1.61%, n = 5), and South American (4.84%, n = 15).

All of the present patients had established diagnosis of PD and were on various combinations of PD medications including levodopa (300–2000 mg/day) with/without entacapone (600–1600 mg/day), pramixole (0.75–4.5 mg/day), ropinirole (0.75–24 mg/day), emantadine (100–300 mg/day), selegiline (5–10 mg/day), and rasagiline (0.5–1.0 mg/day). The combination of these drugs and dosage varied with stage of disease and acquired resistance to a particular drug.


These patients were further studied and tests were carried out in accordance with DSM-IV11 to deduce the presence of dementia. At baseline, the patients did not show signs of dementia. Those who had dementia only due to PD were included in statistical analysis. Further demographic data regarding age and gender were also gathered. Any atypical patient was excluded from the study because they could potentially confound the findings. Age, gender, and stage of PD were studied in relation to dementia in PD patients.

The PD-D patients were followed from their PD diagnosis to the development of dementia. Parkinson's disease patients took a mean of 3.6 ± 3.2 years to develop dementia in this group.

The method of the present study conforms to the provisions of the Declaration of Helsinki in 1995 (as revised in Tokyo 2004) and was approved by the ethics committee of Centenary Hospital, part of Rouge Valley Health System.

Statistical analysis

For age analysis, a two-tailed t-test assuming unequal variances (P < 0.05) was used, while two-tailed chi-squared tests (P < 0.05) were used to determine the significance of gender and stage in PD-D. An inter-stage analysis was performed comparing each stage to one another using two-tailed chi-squared tests (P < 0.05).


Although the present patients were seen from 2005 to 2010, data were collected in January 2011. Among those patients who did develop dementia, the HY stage and the age at which they did so was recorded and compared to that of those who did not develop dementia over the span of 5 years (2005–2010).

The results indicate that age is a significant factor in developing dementia (t(115) = 1.98, P < 0.05). The youngest patient with dementia was 53 years old. The mean age of the PD-D patients was 79.7 ± 8.01 years while the mean age of the non-PD-D patients was 73.6 ± 10.4 years. This indicates that as age increases, the likelihood of developing dementia increases.

Furthermore, gender, according to the results, is not correlated with the presence of dementia in PD (χ2(1,n = 310) = 2.57; P = 0.1089). Approximately 22.9% (n = 40) of all the male patients (n = 175) had PD-D while only 15.6% (n = 21) of all the female patients (n = 135) had dementia.

The HY stage did correlate with development of dementia when considering all stages at once (χ2(3, n = 306) = 9.59, P = 0.0224). Out of 306 patients considered for chi-squared analysis, 61 (19.9%) had dementia. Among them, 20 patients were in HY stage 2 and 15 patients were in stage 4.


We found that 19.9% of the present patients with PD had dementia. There was no difference in men and women. Progression of PD stage and age correlated with prevalence of dementia in PD patients.

According to the present results, age is a determining factor in developing PD-D as is reported in many other studies.2,6,12–16 As one ages, the likelihood of developing PD-D increases. By age 85, patients in the present sample had a 67.2% incidence rate of PD-D. A clinic-based sample, as reported by Levy et al., also produced similar results, having an incidence rate of 65% by age 85.14 Approximately 90% of the present patient sample with PD-D was age ≥70, further strengthening the claim that the evolution of PD-D occurs around age 70.14

As age increases, biological and psychological changes begin to occur. Dopamine levels decrease with age, allowing older individuals to be more susceptible to PD.17 Moreover, Alzheimer's disease, the most common form of dementia, is correlated with high amounts of oxidative stress and an increase in amyloid-β production in the brain.18 Therefore, the combination of lower dopamine levels and high levels of oxidative stress leads to dementia in PD.17

Stage was, also, significantly correlated with PD-D according to the present results. In agreement with this, previous studies have found a positive correlation of PD-D with the progression of PD.16,17,19 Hughes et al. reported that incidence of PD-D was related to greater severity of neurological and motor symptoms, which is a result of disease progression.15 High severity of extrapyramidal signs with older age poses the highest risk of developing PD-D.14 HY stage greater than 2 has been proposed to be a predictor of PD-D.6

The results suggest that gender is independent of the development of dementia. Forty male patients out of 175 men with PD (23%) had dementia, while 21 women (16%) showed symptoms of dementia out of a total of 135. The frequency of PD-D in men was slightly higher, but insignificant. Whether gender is a determining factor for PD-D is still controversial. The basis of argument lies in the hormonal differences between men and women. Estrogen replacement therapy (ERT) in post-menopausal women is shown to decrease the likelihood of developing dementia in female PD patients.20 Pre-menopausal women produce high levels of estrogen while post-menopausal women produce much less estrogen, similar to the levels of men, who compensate by using androgens as antagonists. Post-menopausal women are highly likely to develop dementia compared to pre-menopausal women.21 Some other studies have suggested a significant correlation between male gender and dementia due to this reason of different hormones.12,15,22 In contrast, other studies have found no such association with gender.4,14 All of the present female patients were post-menopausal and, perhaps due to this, the present results support the latter view in which development of dementia does not show gender specificity. This was important in order to control for hormone variations.

The present sample size was relatively large and the patient group diverse. Nevertheless, the present study had certain limitations with regard to patient demographics. A significant number of patients were elderly immigrants with limited command of English and a poor educational background. Their children had arrived in Canada and later sponsored them. Some of these patients were not even able to read and write in their own native languages. Drawing and copying skills were also limited because some of them had never used a pen. Thus, it was not possible to standardize Mini-Mental State Examination21 score as well as years of education across the group.

There is evidence that PD dementia is associated with a cholinergic deficit, supporting the therapeutic role of cholinesterase inhibitors, which are already the first-line management in other types of dementia, such as in treatment of Alzheimer's disease.23 Another study suggested that biomarkers are helpful in identifying patients at risk for cognitive decline. For example, reduction of the biomarker CSF amyloid β (1-42) and low levels of plasma epidermal growth factor are important predictors of cognitive deficit in patients with PD. They may provide beneficial prognostic information, particularly once other risk factors for cognitive decline are present, and they may help in monitoring the responses of patients to new management therapies.24–26 Cognitive decline is not uncommon in PD in that it can occur in a significant proportion of patients, even at the onset of the disease process.

Perhaps further research into such biomarkers and an increased understanding of the underlying mechanism of dementia in Parkinson's disease will lead to better treatments and the possibility of preventing or delaying the onset of PD-D.


The authors would like to thank Husain Cader, Sufian Masroor, and Rizwan Farid for their help with the manuscript.