Incidence of amyotrophic lateral sclerosis in three counties in western Washington state

Except for certain endemic areas in the South Pacific, ^[1] reports of incidence rates for amyotrophic lateral sclerosis (ALS) have been uniform throughout the world with variation in rates thought to be due to underascertainment. ^[2] In 1992, Chancellor and Warlow ^[3] reviewed the incidence rates for ALS from nine surveys, including one study from the United States, ^[4] that they judged to have complete or near complete case ascertainment. The crude incidence rates ranged from 0.6 per 100,000 per year in Sardinia, Italy, to 2.6 per 100,000 per year in Varmland county, Sweden. Standardizing the rates to the Scottish population, ^[3] Chancellor and Warlow found that the differences among incidence rates were statistically significant, suggesting a wider geographic variation in rates than previously reported. In all studies, the incidence rates for ALS were higher in men than women. Few studies have addressed the question of race, but at least one reported lower mortality rates from ALS for nonwhites than whites. ^[5] As part of a population-based case-control study of risk factors for ALS, we measured its incidence in King, Pierce, and Snohomish counties in western Washington state.

Methods.

Patients eligible for the study were residents of one of the three counties, 18 years of age or older, and diagnosed by a neurologist as having ALS during the period April 1, 1990 through March 31, 1995. Patients with progressive muscular atrophy or predominantly bulbar symptoms were included in the study. Patients with primary lateral sclerosis or any other diagnosis were excluded. Based on the 1990 U.S. Census for the three counties, the total population is approximately 2.5 million, 51% of whom are women, 71% are under the age of 45 years (69% for the U.S. population), and 14% are nonwhite (20% for the U.S. population).

Patients were identified through several sources. Neurologists, physiatrists, neurosurgeons, neuroradiologists, hospices, local ALS support services, the Muscular Dystrophy Association, and medical examiners were initially informed of the purpose of the study, and they received monthly letters reminding them to notify us, by telephone, of any newly diagnosed patients with ALS. Approximately 350 letters were sent each month to these health care providers. The ALS patients who were recruited during the first 4 years of the incidence study were also invited to participate in a case-control risk factor study and a follow-up study through a letter sent to them by their referring physician. The 95% CIs for crude and age-specific rates were calculated based on the Poisson distribution. ^[6] The 95% CIs for age-adjusted rates were calculated as suggested for standardized rates. ^[6] These studies were approved by the Human Subjects Committee at the University of Washington.

Results.

During the 5-year surveillance period, 235 patients who met the eligibility criteria were referred to the ALS study, including 127 men (54%), mean age 57 years (range 25 to 87 years), and 108 women (46%), mean age 65 years (range 31 to 85 years). Ten patients (4.3%) were nonwhite and 33 (14%) were under the age of 45 years. The crude incidence for ALS was 1.8 per 100,000 per year (95% CI, 1.3 to 2.4). The incidence rate, adjusted to the age distribution of the 1990 total U.S. population, was significantly higher for men at 2.1 per 100,000 per year (95% CI, 1.3, 2.9) than for women at 1.9 per 100,000 per year (95% CI, 1.1, 2.7) (p < 0.05). For both men and women, the incidence rates increased with age Table 1, Figure 1. The number of nonwhite ALS patients was small, comprised mostly of African-Americans. The age-adjusted incidence rate for nonwhite men was 0.74 per 100,000 per year (95% CI, 0.00, 1.96) and for nonwhite women 0.53 per 100,000 per year (95% CI, 0.00, 1.91).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Incidence of ALS in men and women for western Washington state from April 1990 to March 1995.

Discussion.

The age-adjusted incidence rates (per 100,000 people per year) for three counties in western Washington state are comparable with age-adjusted incidence rates from a Scotland study ^[7]: 2.1 (95% CI, 1.3, 2.9) for men and 1.9 (95% CI, 1.1, 2.7) for women in Washington and 2.2 (95% CI, 1.6, 2.7) for men and 1.7 (95% CI, 1.2, 2.1) for women in Scotland. Investigators have conducted only two other incidence studies of ALS with near complete ascertainment in the United States, in Minnesota ^[4] and in Texas. ^[8] Compared with our study, the similarly adjusted incidence rates in Minnesota were higher in men (2.9) but lower in women (1.6). In Texas, the rates were lower for both men (1.2) and women (1.2). In our study, the age-adjusted incidence rates for women are higher than in other population-based studies as are the age-specific incidence rates for men and women under 45 years of age.

Chancellor and Warlow ^[3] compared the age- and gender-specific rates from eight of nine surveys that were judged to have near complete ascertainment. Because of methodologic differences between studies, they restricted their comparisons to the age range 45 to 74 years. We included the results from our study and those from Scotland and Texas. In Table 2, we present the results of these 11 surveys, all age-adjusted to the 1990 U.S. population for those 45 to 74 years old. For ALS, as for MS, ^[3,8] rates tended to be higher in studies from northern latitudes than southern latitudes (for men, the Spearman rank correlation coefficient was 0.80, p = 0.01; for women, 0.50, p = 0.12; for both combined, 0.62, p = 0.03). In three U.S. and one Finnish studies, the differences in rates between men and women are less striking than in the other seven studies. Such differences may reflect methodologic issues, especially case definition and ascertainment, or differences in the prevalence of etiologic risk factors in certain populations. These considerations may also explain the lower incidence rates for nonwhites in Washington state compared with those reported in the survey from Texas. ^[8]

One age group that poses problems for all incidence studies of ALS is the group 80 years of age and older. We found, as have others, ^[3,7-9] that the incidence rates in both men and women increase with age but decline in the oldest age group. The decline may be due to underascertainment in these older people. ALS can be more difficult to recognize in this group and, even if suspected in a patient, may not lead to a consultation with a neurologist. ^[7] A shorter survival with ALS and competing risk of death from other diseases in the elderly may also contribute to underascertainment. The increasing incidence of ALS over time, suggested by some studies of mortality, ^[3,9-10] may simply reflect fewer competing risks and improved overall life expectancy. ^[10] Mortality data from different birth cohorts indicate that cohorts born after 1900 show a continuous increase in mortality from ALS with age. ^[9] If the decreased incidence rate in the oldest age group is an artifact, the importance of age may be even greater than suggested by current studies.

The incidence rates found in our study are consistent with other studies from northern latitudes. By applying our rates to the 1990 U.S. population, we estimate that about 4,700 people would be diagnosed with ALS in the United States each year. The number would be smaller if the lower rates in the southern latitudes are considered. Although descriptive epidemiologic studies, such as this one, are important as an indicator of the magnitude of the problem, analytic epidemiologic studies are essential to identify etiologic risk factors, modification of which could reduce the risk of developing the disease.

Acknowledgments

We thank the neurologists in King, Pierce, and Snohomish Counties and the Muscular Dystrophy Association of Washington State. We also thank Lorene Nelson, PhD, epidemiologist at Stanford University, for her invaluable contribution.