Prevalence of Parkinson\'s disease in junín, Buenos Aires province, argentina

Mowmenl Disorders Vol. 12, No. 2, 1997, pp. 197-205 0 1997 Movement Disorder Society

Prevalence of Parkinson’s Disease in Junin, Buenos Aires Province, Argentina Mario 0. Melcon, “Dallas W. Anderson, TRodolfo H. Vergara, and $Walter A. Rocca Department of Neurology, Regional Hospitul, Junin, Buenos Aires Province, Argentina; *Biometry and Field Studies Branch, National Institute of Neurological Disorders and Stroke, Bethesda, Maryland, U.S.A.; ?Department of Health and Social Work, Sanitarj District I l l , Junin, Buenos Aires Province, Argentina; and $Department of Health Sciences Research, Mayo Clinic and Mayo Foundation, Rochester, Minnesota, U.S.A.

Summary: We investigated the prevalence of Parkinson’s disease (PD) in a South American city: Junin, Buenos Aires Prov-

100,000 population. The age-specific prevalence was consistently higher in men than women, and it increased with advancing age for both sexes. In addition to prevalence figures, we present tallies related to clinical features of PD, as well as tallies related to other subtypes of parkinsonism. Key Words: Epidemiology-Parkinsonism-Parkinson’s diseasePrevalence.

ince, Argentina. At dwellings systematically selected, the case finding involved household screenings and neurological examinations (i.e., a two-phase survey approach). Only persons 40 years of age or older were eligible (N = 7,765). There were 51 cases of PD identified, yielding a crude prevalence of 656.8 per

We report findings on prevalence and clinical features from one of the first major household surveys of Parkinson’s disease (PD) in a South American population. The survey was conducted in Junin, a city of -75,000 residents, located in an agricultural region within the Province of Buenos Aires, Argentina. The residents, almost all of whom were white, had ready access to health-care services (e.g., general and specialty care were conveniently available, free of charge, at the regional hospital, which was located within the city). The Junin survey made use of household screening questionnaires and neurological evaluations, and, in many respects, was patterned after a survey conducted in Copiah County, Mississippi (1,2).

dykinesia, rigidity, and impaired postural reflexes. For parkinsonism, at least two of the cardinal signs must have been present. The subtypes of parkinsonism were: drug-induced parkinsonism, parkinsonism in vascular disease, other parkinsonism, unspecified parkinsonism, and idiopathic parkinsonism fie., PD). Drug-induced parkinsonism required the intake of neuroleptic drugs (e.g., butyrophenones, phenothiazines) or antidopaminergic agents (e.g., cinnarizine, flunarizine) at any time in the 6 months preceding symptom onset. There must have been a negative history for the parkinsonian signs before the first use of these drugs. Parkinsonism in vascular disease required clear clinical evidence of at least two of the following: hypertension, emotional incontinence and pseudobulbar palsy, broad-based rigid gait, definite history of stroke in the course of the illness, widespread pyramidal signs, and abrupt onset with stepwise progression of symptoms. Other parkinsonism consisted of parkinsonism secondary to nervous system infection, severe head trauma, brain tumor, dementia, or other neurological diseases that possibly affected the basal ganglia. This type of parkinsonism also included cases with associated features or Parkinson-plus syndromes. Unspecified parkinsonism was parkinsonism for

METHODS PD was defined in the context of parkinsonism, and four cardinal signs were considered: resting tremor, bra~~~

..

Received May 31,1995, and in revised form April 1, 1996. Accepted May 13, 1996. Address correspondence and reprint requests to Dr. D. W. Anderson at National Institutes of Health, National Institute of Neurological Disorders and Stroke, Federal Building, Room 7C-16, 7550 Wisconsin Avenue, MSC 9135, Bethesda, MD 20892-9135, U.S.A. This work was presented in part at a meeting of the World Federation of Neurology, Research Group on Neuroepidemiology, Bethesda, Maryland, U.S.A., May 6, 1994.

197

198

M . 0. MELCON ET AL.

which clinical information was insufficient to reach an etiologic classification. PD was diagnosed by ruling out the other subtypes of parkinsonism already mentioned. A further requirement was that at least one of the cardinal signs had to be either resting tremor or bradykinesia. Eligibility for the Junin survey was linked to the prevalence day January 1, 1 99 1. Persons were eligible if, on that date, they were 40 years of age or older, and they resided in Junin. Excluded were soldiers and college students who lived temporarily within Junin, inmates at a local prison, and residents of the one chronic-care institution serving Junin and the surrounding region. (See Discussion for more on the chronic-care institution.) Data collection was in two phases. In phase 1, a twostage systematic sample of blocks and dwellings from within Junin determined the households to be included (3). To accomplish the sampling, the survey area was divided into 2 1 sectors and the residential blocks within each sector were numbered sequentially. The blocks with odd numbers (i.e., 1, 3, 5, . . .) were selected. There were 690 of these blocks, and, within each, the dwellings were numbered sequentially. The dwellings with odd numbers were selected, yielding 5,839 of them for the sample. Trained but medically unsophisticated interviewers visited these dwellings and performed household screenings to identify persons who possibly had parkinsonism (or other major neurological disorders). The screening was a face-to-face interview in Spanish, using a verbatim questionnaire (i.e., questions were read word for word to the respondent, in the order of appearance on the form). One responsible adult in the family (usually the wife/mother) answered about each family member living in the same household. The screening questions for parkinsonism are given in the Appendix, in both Spanish and English. The Spanish version had been refined in a series of pilot investigations previously reported ( 3 ) . The final investigation of sensitivity and specificity involved visits to households that contained persons who had previously come to medical attention. The status of these persons with respect to parkinsonism was known to the senior author (M.O.M.), but not to the interviewers who conducted the screening. The sensitivity estimate for parkinsonism was 24 of 24 (1 OO.O%), and the specificity estimate was 16 of 20 (80.0%) (3). Briefly, the screening addressed the following: whether anyone had PD; tremor in hands, legs, head, or jaw; permanent muscular rigidity in arms or legs (not because of swelling in the arms, legs, or joints, nor because of arthritis or rheumatism); difficulty standing up from a sitting position (not because of severe pain in the legs); slow walking (compared with others of same

Movement Disoi-ders, Vol. 12, No. 2 , 1997

age, and not caused by an operation on hip, knee, or leg); difficulty to stop walking or running (because legs will not obey); slow dressing (slower than before); clumsy fingers (when buttoning clothing); difficulty in writing (handwriting smaller and jerkier than before); voice problems (speaking in a low and muffled voice, difficult to understand). Satisfying any of these items was sufficient to be screened as positive. The persons screened as negative were not evaluated further. In phase 2, project neurologists examined the persons screened as positive in phase 1, and made preliminary diagnostic determinations. Typically, the examination would last 3 0 4 5 min, with about half that time devoted to history taking and some clinical measurements (e.g., pulse, blood pressure). An assessment was made of the cranial nerves, motor function, coordination, tendon reflexes, gait and station, sensory perception, and higher cortical functions. Also, a specific evaluation was made for PD. This evaluation included, in so far as was possible, a review of medication history. On the basis of the aforementioned clinical definitions, the senior author (M.O.M.) made the final diagnostic determinations for the persons examined and those few who were not. He relied on examination findings and whatever useful medical information he could obtain about each person. This information pertained mainly to signs and symptoms, history, and medicines, and was obtained through contacts with family members and health-care providers. In this article, we use the term “prevalence” to mean the number of persons with PD in the sample divided by the number of persons in the sample. Both the numerator and denominator relate to the prevalence day January 1, 199 1. The investigation is descriptive in that prevalences and other results refer to the sample of Junin residents screened in phase 1. However, on the basis of the systematic sampling that was used, it is reasonable to assume that the prevalences obtained also refer to all Junin residents (or appropriate subgroups of residents) who satisfied the eligibility criteria (see also the Discussion section). Extrapolation beyond Junin has no statistical basis (4) and is not recommended. Additional details about the survey personnel, the survey methods, and the pilot investigations have been published elsewhere (3).

RESULTS A total of 5,839 households in Junin were systematically selected for this survey, of which 5,648 (96.7%) were screened in phase 1. The 191 nonparticipating households consisted of 1 that had ceased to exist and 190 that had refused to cooperate. Most refusals came from households that did not want to be bothered by the

PREVALENCE OF PARKINSON’S DISEASE

199

ism by project neurologisty. These 27 persons consisted of the following: 12 of 51 with PD (9 men, 3 women) (23.5%); 7 of 8 with drug-induced parkinsonism (3 men, 4 women) (87.5%); 3 of 7 with parkinsonism in vascular disease (2 men, 1 woman) (42.9%); I of 1 (man) with other parkinsonism (100.0%); 4 of 6 with unspecified parkinsonism (2 men, 2 women) (66.7%). The 12 persons whose PD was first diagnosed during the survey ranged in age from 52 to 82 years. They all displayed resting tremor as a cardinal sign; none displayed impaired postural reflexes. Seven had PD for at least 3 years. Despite access to free medical care in Junin, 7 of the 12 persons had refused to see a doctor (reportedly because of negative experience with doctors in the past). In regard to the remaining 5 persons, 4 were misdiagnosed by their family physicians and 1 was neglected by relatives. The latter individual was depressed and physically disabled from an earlier accident. Of the 51 persons with PD, 15 (29.4%) showed two cardinal signs, 21 (41.2%) showed three cardinal signs, and 15 (29.4%) showed four cardinal signs. Considering each cardinal sign individually, 47 persons (92.2%)

survey. Some refusals were passive in the sense that cooperation was promised but it never materialized. The 5,648 screened households contained 17,049 persons, including 7,765 persons (45.5%) who were 40 years old or older. Of the 7,765 persons, 213 (2.7%) were screened as positive for parkinsonism, and 200 of the 213 persons (93.9%) were examined (10 persons refused, 1 was unavailable, 2 were deceased) in phase 2. One of the authors (M.O.M.) obtained adequate medical information for the 13 persons not examined (see the Methods section); that is, he was able to determine whether each had parkinsonism. We identified 73 persons (41 men, 32 women) with parkinsonism (Table 1). PD was, by far, the most frequent subtype seen, affecting 51 persons (69.9%). The remaining 22 persons were classified as follows: 8 with drug-induced parkinsonism (4 taking neuroleptic drugs, 4 taking antidopaminergic agents); 7 with parkinsonism in vascular disease; 1 with other parkinsonism; and 6 with unspecified parkinsonism. Referring again to the same 73 persons, 27 (17 men, 10 women) (37.0%) were first diagnosed for parkinson-

TABLE 1. Cases of parkinsonism, by subtype and sex, and cases Of’Parkinson’s disease, population at risk, and prevalence (per 100,000 populution), by age and sex: Junin, Buenos Aires Province, Argentina, January 1 , 1991“ Sex Men Parkinsonism subtypes Parkinson’s disease Drug-induced parkinsonism Parkinsonism in vascular disease Other parkinsonism Unspecified parkinsonism Total Parkinson’s disease Age (years) 4 W 9 No. of cases Pop. size Prevalence 50-59 No. of cases Pop. size Prevalence 60-69 No. of cases Pop. size Prevalence 70-79 No. of cases Pop. size Prevalence 80+ No. of cases Pop. size Prevalence Total No. of cases Pop. size Prevalence

28 4 4 lh 4 41 0 92 1 0.0 3 886 338.6 8 960 833.3 9 494 1,821.9 8 137 5,839.4 28 3,398 824.0

Women 23 4 3 0 2 32 0 1,084 0.0 0 1,076 0.0

6 1,238 484.7 12 722 1,662.0 5 241 2,024.3 23 4,367 526.7

Total (%)

51 (69.9) 8 (11.0) 7 (9.6) l(1.4) 6 (8.2) 73 (100.0) 0 2,005 0.0 3 1,962 152.9 14 2,198 636.9 21 1,216 1,727.0 13 384 3,385.4 51 7,765 656.8

Pertains to persons 40 years old or older. secondary to dementia.

” Parkinsonism

Movement Disorders, Vol 12, No. 2 , 1997

M. 0. MELCON ET AL.

200

showed resting tremor, 39 (76.5%) showed bradykinesia, 46 (90.2%) showed rigidity, and 21 (41.2%) showed impaired postural reflexes. In terms of duration, 2 persons (3.9%) had PD for 20 years or more, 8 (15.7%) had the condition for 10-19 years, and 41 (80.4%) had the condition for