Part I: Compare the two detection methods for active detecting

detecting Parkinson’s disease

In this study, we first demonstrate an evaluation of the relative efficacy of the active

and passive detection methods of PD screening in a massive screening program which

provided a natural experimental design (random assignment). The random screening in

2001 seemed to balance the population. The active method detected approximately

1.8-fold of the PD cases of the passive method. The active method detected more early H-Y

stage (stage I and II) PD cases than did the passive method. The active method reduced

49% of PD cases diagnosed at H-Y stage III or higher, compared to the passive method.

The method used to detect PD has been considered to account for the large variation in the

estimates of IPD prevalence and incidence of many epidemiological studies.^8-14In one systematic review of Parkinson’s disease in Asia,⁹the prevalence ranged from 35.8 to 68.3

per 10⁵person-years in record-based studies and ranged from 51.3 to 176.9 per 10⁵

person-years in door-to-door surveys. In door-to-door surveys, the standardized incidence

rates were 8.7 per 10⁵person-years; in record-based studies, it ranged from 6.7 to 8.3 per

10⁵person-years. These discrepancies are due to different case-finding methods, different

r ac c c c ct t t t t ti i i i i iv v v v ve e e e e

95

age distributions in the population, different diagnosis criteria, and different genetic and

environmental factors. Although door-to-door survey seemed to yield higher prevalence

and incidence rates than record-based studies, no study has directly elucidated the

differences between case-finding methods and none of the previous studies compared

different methodologies of determining PD prevalence. Our study directly proved that the

active method is able to detect approximately 1.8-fold the IPD cases of the passive

method. In addition, our pseudo-experimental design was able to overcome the potential

discrepancies of previous epidemiological studies such as age distributions, genetic, or

environmental factors.

Two previous door-to door surveys estimated the prevalence of IPD in Taiwan. Liu et al. found that the prevalence rates of Parkinson’s disease in Kinmen was 587 per 10⁵

person-years among those aged 50 years or older in a single-phase door-to-door survey by

neurologists.³⁴A two-stage door-to door survey in Ilan county, Taiwan found a crude IPD

prevalence rate of 367.9 per 10⁵person-years and an incidence rate of 30.1 per 10⁵

person-years among subjects aged 40 person-years or older.³³The age-adjusted prevalence rate for all age

groups was 130.1 per 10⁵person-years and the age-adjusted incidence rate was 28.7 per

10⁵person-years after being adjusted to 1970 US census in Ilan county. The crude

prevalence rate was 1,520 per 10⁵person-years in our study of adults aged 40 years or

older. The age-adjusted prevalence rate was 552.5 per 10⁵person-years. If compared to the erenenenenenttttt gegegegegegegegeneenenenenenenenenetititititititititicccc c ccccananananananananand d d

hii hi hi

hgggghgheererererererererpppppppprererererererer vavavavavavavavalelelelelelelelencnnncncncnnce eeee eee

cidaaaaateteteteted edd d d ththththe

96

previous results from Ilan county, our active method detected four times the PD cases of

those found in the two-stage door-to door survey.

Early detection of PD is important because a previous study found that subjects in

H-Y stage I may have the same life expectancy as the general population.¹⁸Early detection

and treatment of PD may increase life expectancy.^{3, 93-95}Liou et al. found that PD cases

detected early showed a 74% reduction in the incidence of stage III or greater PD and a

26% reduction in mortality.¹⁸Our result suggest that the active detection method identified

more stage I and II PD cases than did an examination of the health insurance claims record

(80.4% vs. 61.5%, p=0.04). We also showed that the active detection method could reduce

49% of the incidence of PD at H-Y stage III or greater at diagnosis.

The active screening method is more time consuming and requires more resources

than the passive detection method. However, delayed diagnosis of PD may result in rapid

progress in H-Y stage and much greater medical costs to deal with the many complications

that accompany the progression of this disease.^{85, 96, 97}The relative cost effectiveness and

benefit of the active and passive detection methods need further evaluation.

Although we demonstrated that the active method detected almost two times the PD

cases of the passive method, the present study had several limitations. First, the

participants in our study were adults who attended a community-based integrated

screening program rather than a nationally representative sample. The incidence and s ttttthehehehehePPDDDDDDDDDcacacacacacacacasesesesesesesesesesssssssssofofofof f

d thatatatatatsssssububububu jejejejej ctcsssssinininininHHHHH

H-97

prevalence rates may not represent those of the national population. However, the aim of

the study was to compare two methods of PD detection, not determine national rates.

Secondly, we did not review all the H-Y stages of the PD cases due to resource limitations

and sampled 65 of 233 PD cases to determine H-Y staging data through the medical

records. However, our sample was sufficient to suggest that active detection is superior in

detecting early stage PD.