Abstract
We assessed the prevalence of Parkinson's disease (PD) in a general elderly population in the Netherlands.The study formed part of the Rotterdam Study, a population-based door-to-door study, and included 6,969 persons 55 years of age or older living in a suburb of Rotterdam, the Netherlands. All participants were examined, and those who either had at least one possible cardinal sign of parkinsonism at the neurologic screening, reported that they had PD, or were taking antiparkinsonian drugs were invited for further evaluation. The prevalence of PD in this population was 1.4% (1.2% for men, 1.5% for women). Prevalence increased with age, and prevalence figures were 0.3% for those aged 55 to 64 years, 1.0% for those 65 to 74, 3.1% for those 75 to 84, and 4.3% for those 85 to 94. The corresponding age-specific figures for men were 0.4%, 1.2%, 2.7%, and 3.0%, and for women, 0.2%, 0.8%, 3.4%, and 4.8%. Among 95- to 99-year-old women the prevalence was 5.0%. Twelve percent of the subjects with PD were detected through the screening and had not been diagnosed previously.
NEUROLOGY 1995;45: 2143-2146
Parkinson's disease (PD) is among the most frequent chronic neurodegenerative diseases in the elderly. However, prevalence estimates of PD vary widely, ranging from 10 to 405 per 100,000 population. [1] This variation may be due to differences in case-finding procedures, in diagnostic criteria, and in the age distribution of populations. We present the results of a population-based PD prevalence survey among 6,969 subjects living in Rotterdam, the Netherlands, in which we examined each individual neurologically.
Methods.
Study population.
This study forms part of the Rotterdam Study, a population-based cohort study on prevalence, incidence, and determinants of diseases in the elderly. [2] The four main areas of interest of the Rotterdam Study are neurologic diseases, ophthalmologic diseases, locomotor diseases, and cardiovascular diseases. The first cross-sectional survey started in 1990 and was completed in June 1993. All inhabitants of Ommoord, a suburb of Rotterdam, the Netherlands, who were 55 years of age or older, either living independently or institutionalized, were invited to participate in the study by mail and were contacted by telephone 2 weeks later. Names and addresses were drawn from the municipal register, which is reliable, complete, and updated weekly. Of the 10,275 eligible subjects, 7,983 (78%) agreed to participate and signed informed consent statements. All participants were interviewed at home and most were subsequently examined at a research center. The number who visited the research center decreased slightly, to 7,129 (69%), due to refusal, disease, or death. In this prevalence study we included the 6,969 subjects (68%) who had a neurologic screening examination or who were using antiparkinsonian medication or reported that they had PD.
Case-finding and diagnostic procedures.
To assess the prevalence of PD we used a two-phase design. In the first phase, all participants were asked about previous diagnosis of PD, and any drug use was coded according to the Anatomical Therapeutic Chemical (ATC) classification index. [3] In addition, every participant was neurologically examined by one of the study physicians. All subjects who either used antiparkinsonian drugs (ATC code N04), reported that they had PD, or had at least one possible cardinal sign of parkinsonism (ie, resting tremor, rigidity, bradykinesia, or impaired postural reflexes) at the neurologic screening examination were invited for further evaluation in a second phase.
In the second phase, those who screened positive were examined by a neurologist or a neurologist-intraining. A structured clinical work-up, comprising the motor examination of the Unified Parkinson's Disease Rating Scale (UPDRS), [4] a neurologic examination, and standardized history taking, was used to establish the diagnosis of parkinsonism and to classify this parkinsonism. We used diagnostic criteria agreed upon in EUROPARKINSON (European Community Concerted Action on the Epidemiology of Parkinson's Disease). [5] Parkinsonism was diagnosed as ``definite'' if at least two of the cardinal signs were present in a subject not taking antiparkinsonian drugs, or if in a subject treated with antiparkinsonian medication one or more signs, documented by medical history, had improved by treatment. Parkinsonism was classified as ``possible'' if only one sign was present in an untreated subject or if a specifically treated patient did not report benefit from treatment and had only one cardinal sign. Possible parkinsonism is not included in the prevalence figures. All newly diagnosed patients with definite parkinsonism were reevaluated by a second neurologist.
PD was defined in a person with definite parkinsonism by exclusion of all other possible causes of parkinsonism. Parkinsonism associated with other causes included (1) drug-induced parkinsonism (ie, following use of neuroleptics or other antidopaminergic drugs in the 6 months before onset of symptoms and with no history of parkinsonism); (2) parkinsonism related to cerebrovascular disease (ie, with a clear time relationship between the cerebrovascular event and onset of atypical parkinsonism, preferably supported by neuroimaging, usually without tremor); (3) parkinsonism associated with dementia; (4) parkinsonism in multiple system atrophy or progressive supranuclear palsy; and (5) other parkinsonism. Included in this last category were subjects with more than one possible cause or with no clear time relationship between the possible cause and the parkinsonism, as well as those subjects in whom all other possible causes of parkinsonism could be excluded but who had not shown any progression over more than 15 years in the course of the disease and who did not respond to antiparkinsonian drugs.
Some subjects who screened positive could not be evaluated in the second phase for a variety of reasons, such as refusal, disease, or death. For these subjects we obtained additional information from medical records of neurologists and general practitioners. They had to meet the same inclusion and exclusion criteria for parkinsonism and PD as those who could be examined.
In demented patients the neurologic screening examination was often difficult to conduct or to interpret. Therefore, irrespective of the results of the screening, we reviewed the medical records of demented institutionalized participants to check for a previous diagnosis of PD or secondary parkinsonism.
Data analysis.
We calculated age- and sex-specific prevalence figures of parkinsonism and PD for the study population that underwent a complete first-phase screening. The prevalence figures are presented in 10-year age groups and are presented separately for men and women. To determine whether the overall age-adjusted prevalences of men and women differed significantly, we performed a logistic regression analysis in which gender and age were entered as determinants in the model.
Results.
Of the 6,969 participants, 653 (9.4%) screened positive and were invited for evaluation in the second phase. Table 1 shows the age and sex distributions of the study population and the percentages of subjects who screened positive. In the second phase, 499 subjects who screened positive (76%) were clinically examined. Of the subjects evaluated, we found 89 subjects to have parkinsonism, and in 69 of these this parkinsonism was attributed to PD. In 128 of the 154 subjects who screened positive who could not be further examined, we obtained a sufficient amount of other medical information on which to base the diagnosis. This revealed another 40 subjects with parkinsonism, of whom 28 met the diagnostic criteria for PD. Therefore, we identified a total of 129 subjects with parkinsonism, of whom 97 were suffering from PD. The other diagnoses consisted of parkinsonism associated with dementia (9/129); drug-induced parkinsonism (3/129); parkinsonism related to vascular disease (1/129), multiple system atrophy (2/129), or progressive supranuclear palsy (1/129); and other parkinsonism (16/129). Table 2 presents the age- and sex-specific prevalence figures for parkinsonism and PD in the total study population. The prevalence increased with age, even in the highest age categories. The prevalence of PD did not differ significantly between men and women.
Table 1. Age and sex distributions of the study population and percentages of subjects screening positive*
Table 2. Prevalence* (%) of any parkinsonism (PS) and Parkinson's disease (PD) in the Rotterdam Study
Twelve subjects with PD (12%) were newly diagnosed through the study protocol; 11 of them were 70 years of age or older. Of the 75 subjects who reported that they had PD, 12 (16%) turned out to be not affected by the disease. Seven of those participants had never been diagnosed with PD, and five subjects with a previous diagnosis of PD were excluded because they did not fulfill the inclusion criteria for PD used in this study; one of these patients had a nonprogressive hemiparkinsonism, two patients did not have parkinsonism according to the UPDRS, and the two other patients had more than one possible cause for their parkinsonism. In addition, we classified one subject with a previous diagnosis of multiple system atrophy and two subjects with a previous diagnosis of progressive supranuclear palsy as having PD since no other neurologic symptoms were present to justify the previous diagnoses. Of all patients with PD, 73 (75%) used antiparkinsonian medication.
Discussion.
We found a consistent and rapid increase in the prevalence of both parkinsonism and PD with age, with no leveling off in the highest age categories and no significant gender difference.
The overall response in our study was good, but nonresponse may have led to a certain underestimation of the prevalence. Indeed, in the Rotterdam Study, the prevalence of PD was higher among those who screened positive who could not be examined in the second phase than among those who could. Furthermore, we classified parkinsonian participants with atypical parkinsonism or with no clear time relationship between the possible cause and the parkinsonism as having ``other parkinsonism.'' This may also have resulted in some underestimation of the prevalence of PD in our study.
Prevalences of PD, reported in various studies, show wide variation. [1] Many of these studies were based on existing medical records. [6-12] Differences in case-finding procedures, in diagnostic criteria, and in accessibility and level of medical services may hamper comparison. Some of these studies dated from before levodopa treatment was available, and the advent of this treatment may have influenced survival of parkinsonian patients [13,14] and thus the prevalence. Moreover, patients who failed to seek medical attention for their parkinsonism were not systematically included in these studies. PD is a neurodegenerative disease with an insidious onset, and parkinsonian symptoms are likely to remain unrecognized or to be misclassified or accepted as part of a normal aging process. This also leads to underestimation of the prevalence and might explain the lower prevalence estimates [6-12] than that reported in our study, as well as the decrease in prevalence of PD in the higher age categories reported in some studies. [9-11]
Not until 1985 were there published results of population-based prevalence studies of PD. All population-based studies showed an increase in the prevalence with age, even in the highest age category. [15-21] Although the methodology used in these studies should limit the underestimation of prevalence mentioned above, the age-specific prevalence estimates for the age category 70 to 80 years varied from 0.6% in China [15] to 1.3% in France, [21] 1.8% in our study, and 2.0% in Sicily. [16] The prevalence estimates of parkinsonism and PD in our study are among the highest reported thus far and are in concordance with the results of the Sicilian and French studies, which, together with the Chinese study, were the only ones to administer a direct screening instrument to each individual, including a brief neurologic examination by a physician. The discrepancy between the results of the Chinese study and those of the studies conducted in Rotterdam, Sicily, and France could be due to differences in survival of PD patients, in classification, and in etiologic factors.
Although some prevalence studies indicated higher prevalences for men [7,8,15,20] or for women, [9,10] we, like some other investigators, [11,12,17,21] found no significant gender differences in prevalence.
The screening instrument used in this study enabled us to detect 12 subjects with PD (12%) who had not been diagnosed before. Remarkably, overall prevalence figures in our study were similar to those in the Sicilian and French studies, in spite of a lower percentage of newly diagnosed subjects in the Rotterdam Study (12% versus 35% and 27%), which may be due to differences in referral habits and in accessibility of medical services.
In conclusion, we found an increase in the prevalence of parkinsonism and PD with age, even in the highest age categories. A substantial portion of patients with PD were previously undetected, especially among the oldest.
Acknowledgments
We gratefully acknowledge the participants in the Rotterdam Study and the assistance of the Rotterdam Study coworkers, Coby van den Heuvel and Rene Vermeeren.
- Copyright 1995 by Advanstar Communications Inc.
REFERENCES
Letters: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Use of Whole-Genome Sequencing for Mitochondrial Disease Diagnosis
Dr. Robert Pitceathly and Dr. William Macken