Abstract
Background: We conducted a prospective, population-based study of ALS in the Republic of Ireland for the 3-year period 1995 to 1997.
Methods: To ensure complete case ascertainment, multiple sources of information were used, including consultant neurologists, neurophysiologists, primary care physicians, and the Irish Motor Neuron Disease Association. The El Escorial diagnostic criteria for ALS were applied to all cases enrolled on the register and each patient was regularly followed up during his or her illness.
Results: Between January 1, 1995, and December 31, 1997, 231 patients were diagnosed with possible, probable, or definite ALS, including 133 men (57.6%) and 98 women (42.4%). The average annual incidence rate was 2.1 per 100,000 person-years (95% CI, 1.8 to 2.4), and 2.8 per 100,000 person-years for the population older than 15 years (95% CI, 2.4 to 3.1). The incidence rate was higher for men, being 2.5 per 100,000 person-years (95% CI, 2.0 to 2.9), than for women, at 1.8 per 100,000 person-years (95% CI, 1.5 to 2.2), and increased with age for both sexes. The median age at onset was 64.2 years for men and 67.8 years for women. On December 31, 1996, the crude prevalence was 4.7 per 100,000 of the total population (95% CI, 4.0 to 5.5), and 6.2 per 100,000 for the population older than 15 years (95% CI, 5.3 to 7.1). Adjusting to the 1996 Irish population as standard, the incidence of ALS in Ireland during the 3-year study period is the third highest reported to date.
Conclusions: There was a trend toward a higher incidence of ALS in the northwestern region of Ireland, although the numbers of cases involved were small and further study is required.
Amyotrophic lateral sclerosis (ALS), a degenerative disorder of unknown cause affecting motor neurons, is characterized by a combination of upper and lower motor neuron signs with progression. With the exception of two prospective, population-based studies from Scotland1 and Washington,2 incidence and prevalence studies of ALS have been based on retrospective data from selected population subgroups, such as patients referred to neurologic units of general hospitals3 or major medical institutions,4 hospital discharge data,5 cases identified at autopsy and mortality statistics,6 or cross-sectional survey data.7 These retrospective studies report the crude incidence of ALS to vary from 0.6 to 2.6 per 100,000 of the population and the prevalence from 1.6 to 8.5 per 100,000 of the population.8,9 The purpose of this study was to document the incidence and prevalence of ALS in the Republic of Ireland during the period from January 1, 1995, to December 31, 1997. This is the first study of ALS based on review of the complete medical records of patients diagnosed with ALS in a large, well-defined, and homogenous community.
Methods.
The Republic of Ireland is 70,273 square kilometers (27,132 square miles) in area. The 1996 national census estimated the Irish population to be 3,626,087, of whom a relatively high proportion (41%) was younger than 25 years.10 Ireland is divided into eight medical administrative regions (health boards) and there were 10 consultant neurologists and two consultant neurophysiologists in employment in the state during the study period. Medical services in Ireland are largely hospital-based and, in general, primary care physicians do not exercise a gatekeeper effect. Few patients seek medical care outside of the state and the provision of free or heavily subsidized public medicine ensures that the entire population at risk of ALS is likely to visit a neurologist at some stage during their illness. There is one designated ALS outpatient clinic in Ireland—the National Neuroscience Center for Ireland—run by one of the authors (O.H.) at Beaumont Hospital, Dublin. This clinic receives referrals from primary care physicians from all parts of the country for further evaluation and ongoing management in addition to quaternary referrals from other Irish neurologists.
Case ascertainment.
Population-based epidemiologic research is possible in the Republic of Ireland because medical care is practically self-contained within the community and is delivered by a small number of health care providers and neurologists. The strength of this study lies in the existence of a complete register of all patients with ALS in Ireland. This ongoing register of all newly diagnosed ALS cases was established on January 1, 1993, but case ascertainment was considered to be incomplete for the first 2 years. Case ascertainment in the Irish ALS Register is 100% complete since January 1, 1995. The Register is maintained at the National Neuroscience Center for Ireland.
The data sources on newly diagnosed cases of ALS, both inpatient and outpatient, are as follows: consultant neurologists, neurophysiologists, neuropathologists, neurosurgeons, primary care physicians, and the Irish Motor Neuron Disease Association. Furthermore, patients referred directly by primary care physicians to the National Neuromuscular Clinic are enrolled on the Register once ALS is confirmed. All consultant neurologists working in Ireland collaborate in recruitment for the Irish ALS Register. The vast majority of ALS patients are reviewed by a neurologist either at diagnosis or at some stage during their illness for advice on symptomatic management. Once patient consent is obtained, clinical details are forwarded to the Register in the form of a discharge summary. Occasional patients, especially those older than 75 years, may be referred directly to a general physician or a geriatrician. However, most of these patients undergo neurophysiologic testing as part of their diagnostic workup, and both consultant neurophysiologists currently working in Ireland collaborate with enrollment on the Register. In some instances, information regarding the diagnosis of a previously unknown patient is obtained from the Irish Motor Neuron Disease Association. In these cases, the patient’s attending physician is contacted by the ALS research team to confirm the diagnosis and to obtain permission to include the patient’s details on the Register. The multiple sources employed ensure complete case ascertainment for all ALS patients diagnosed in Ireland. In all cases, informed consent is obtained from the patients before inclusion on the Register.
Using this database, we identified all Irish residents in whom ALS was diagnosed during the 3-year period from January 1, 1995, to December 31, 1997. The prevalence of ALS was determined from those patients known to be alive on December 31, 1996. The medical records and results of neurophysiologic tests of patients referred for inclusion on the Irish ALS Register were reviewed. In most instances, clinical records of the patients are reviewed on an ongoing basis; therefore, details of medical care provided to the majority of ALS patients were available for study.
Diagnostic criteria.
Diagnostic criteria applied are as outlined at the 1990 World Federation of Neurology consensus conference at El Escorial, Spain.11 These criteria are based on clinical evidence of a progressive disorder with a characteristic combination of upper and lower motor involvement in the same body regions. Certain exclusion criteria apply (such as known ALS mimic syndromes) and neurophysiologic testing is used to confirm the diagnosis. Individuals younger than 15 years are excluded owing to the clinical overlap with genetic motor neuron diseases resembling ALS. All patients fulfilling the diagnostic criteria undergo confirmatory neurophysiologic studies before inclusion of their data on the Register.
Patients with motor neuron diseases other than ALS are also referred to the Register. Other incidence and population-based studies of ALS have included these cases in the generation of incidence and prevalence rates.5,12 To facilitate comparison with other studies, all cases of progressive bulbar palsy (PBP), progressive muscular atrophy (PMA), and primary lateral sclerosis (PLS) are included among the incident cases for this study.
Residency criteria.
In addition to fulfilling diagnostic criteria, patients for the incidence study must have established residency in the Republic of Ireland for at least 1 year before diagnosis, thus excluding those who might have migrated into the area for diagnosis or treatment. Patients for the prevalence study were required to be residing in the state on December 31, 1996, and to have established residence in Ireland for at least 1 year before that date.
Data analysis.
The entire population of the Republic of Ireland older than 15 years is considered to be at risk of developing ALS. Thus, incidence rates were calculated with the observed number of cases divided by the age- and gender-specific person-years of observation estimated from 1996 census data for Ireland. Rates were also calculated separately for individual health board regions and for urban and rural areas. The denominator for the calculation of prevalence rates is the Irish population on December 31, 1996. Ninety-five percent confidence intervals (95% CI) were estimated from the cumulative Poisson distribution and a one-sample z-test was used looking for regional variations.13 Statistical significance indicates a p value less than 0.05 unless otherwise stated.
Data are stored on a computerized Microsoft (Seattle, WA) Access version 7 database,14 allowing for organized retrieval of data. Statistical analyses were performed using SAS (Cary, NC) version 6.11.15 Separate paper files are maintained for each case.
Results.
During the 3-year study period, 231 Irish residents were diagnosed as having possible, probable, or definite ALS. Of these, 9 (3.9%) had familial ALS based on detailed family history. Altogether, there were 172 patients with ALS residing in the community on the prevalence day of December 31, 1996. Of these 172 cases, 104 (60.5%) represented new cases diagnosed during the study period, whereas the remaining 68 (39.5%) were diagnosed before the start of the study period on January 1, 1995. Thirty-two additional cases carrying a diagnosis of ALS were referred for inclusion on the Register, but were ultimately rediagnosed with an ALS mimic syndrome and excluded from this study.
Of the 231 incident cases, only one 68-year-old woman had clinical and neurophysiologic features typical of PBP without clinical evidence of anterior horn cell involvement at time of death. There were two cases of PMA: one autopsy-proven case in a 74-year-old woman and one 57-year-old man with slowly progressive symptoms of 5 years’ duration. PLS was diagnosed in only one case—a 67-year-old woman who had developed symptoms 5 years before with consistently negative electromyography. All five of these cases were included in the incidence figures, although they are defined as “suspected” or “possible” ALS cases according to the El Escorial criteria. On the prevalence day, there was one patient with PMA and one with PBP residing in Ireland, both of whom were diagnosed before 1995 and therefore were not included in the incident cases.
Figure 1 details the age- and gender-specific incidence rates and the crude incidence rates with 95% CIs. Based on the 231 newly diagnosed cases of ALS in Ireland, the average annual crude incidence rate was 2.1 per 100,000 person-years (95% CI, 1.8 to 2.4). The average annual crude incidence rate for the population older than 15 years was 2.8 per 100,000 person-years (95% CI, 2.4 to 3.1). With single cases, the crude incidences of PBP and PLS were estimated at 0.01 per 100,000 person-years, and 0.02 per 100,000 person-years for PMA. There was a preponderance of men among the ALS incident cases: 133 versus 98 women. The male and female age-adjusted incidence rates in the population older than 15 years were 3.3 (95% CI, 2.7 to 3.8) and 2.3 (95% CI, 1.9 to 2.8), respectively. The average annual incidence for ALS was consistently higher for men than for women in all age groups, with the exception of the 45 to 49 age group in which the difference was reversed (see figure 1). Furthermore, in both men and women, rates generally increased with advancing age, reaching a peak at 80 to 84 for men and 70 to 74 for women and declining rapidly thereafter. Median age at diagnosis of ALS was 64.2 years (range, 33.0 to 87.0) for men and 67.8 years (range, 36.1 to 92.5) for women.
Figure 1. Age- and gender-specific incidence rates of ALS in Ireland 1995–1997.
The manner in which these incident cases presented is given in table 1. Eighty (34.7%) of the 231 ALS cases presented with purely bulbar symptoms at onset, 129 (55.8%) initially presented with limb symptoms, and the remaining 22 (9.5%) had generalized onset. Of the 80 purely bulbar onset cases, 75 (93.7%) had dysarthria as their first symptom; only 5 (6.3%) presented with dysphagia or choking. Forty-four (44.9%) of the 98 women presented with purely bulbar symptoms at onset; the corresponding figure for men was 36 of 133 (27.1%). Lower limb onset was more common than upper limb onset in the Irish population, with 72 (55.8%) of the total 129 cases presenting with symptoms referable to the legs.
Clinical spectrum at presentation of ALS among Irish residents, 1995–1997 (n = 231)
The crude prevalence of ALS on December 31, 1996, was 4.7 per 100,000 of the population (95% CI, 4.0 to 5.5) and was 6.2 per 100,000 of the population older than 15 years (95% CI, 5.3 to 7.1). Prevalence rates were greater for men, with male and female prevalence rates of 7.9 (95% CI, 6.4 to 9.4) and 4.5 (95% CI, 3.4 to 5.7) per 100,000 population older than 15 years, respectively (figure 2). Thus, the male to female ratio was 1.8:1. Overall, the prevalence rate increased with advancing age, reaching a peak in the 65 to 69 age group for women and 80 to 84 for men and declining thereafter. The median age at diagnosis of ALS was 59.7 years (range, 31.0 to 91.6) for men and 62.7 years (range, 33.5 to 84.7) for women.
Figure 2. Average annual age- and gender-specific prevalence rates of ALS on December 31, 1996.
The incidence rates of ALS in the eight Irish health board regions are compared in figure 3. The national average annual incidence rate was 2.8 per 100,000 population over 15 years of age (95% CI, 2.4 to 3.1). The average annual incidence rate for the population over 15 years old was 4.4 per 100,000 person-years for the northwestern health board. This rate was statistically different from the national average annual crude rate (two-sided z test = 2.13, p = 0.017) and none of the other health boards had a similar statistically significant high incidence rate. When age- and gender-adjusted to the 1996 Irish population, the northwestern incidence rates remained high (3.9 per 100,000 person-years) in comparison with the other health boards and the national incidence rate, but failed to reach statistical significance (expected cases in the area = 15.1, actual cases in the area = 21, z test = 1.5, p = 0.07). When comparisons are made between multiple regions and the national mean incidence rate, chi-square = 3.74, p = 0.81. This area of the country—consisting of Donegal, Sligo, and Leitrim counties—is geographically isolated from the main population centers of Ireland and has a lower population density than the rest of the country. Similar to the incidence rates, only the northwestern health board had a prevalence rate for the population over 15 years old (12.0 per 100,000 population over 15 years old) that was statistically higher than the national prevalence rate of 6.2 per 100,000 (z test = 2.9, p = 0.002). Age and gender adjustment of the northwestern prevalence data to the 1996 Irish population strengthened this statistical significance (expected cases = 11.4, actual cases = 19, z test = 2.25, p = 0.01; chi-square = 8.84 and corresponding p = 0.26). There did not appear to be any difference in incidence rates of ALS between rural versus urban areas. However, the number of cases was small when subdivided.
Figure 3. Distribution of ALS in Ireland according to health board, 1995–1997. IR = average annual crude incidence rate per 100,000 population older than 15 years. National average annual incidence rate = 2.8 per 100,000 population older than 15 years. *Statistically significant z value = 2.13, p = 0.017.
Discussion.
This study is the third prospective, population-based study of ALS1,2 and the first to identify all ALS cases in an entire country. It is also the first epidemiologic study of ALS to employ the El Escorial criteria for diagnosis, thus allowing for more standardized comparison with future epidemiologic studies. Mortality studies are inherently less accurate than register-based studies because of possible underreporting on death certification and are further complicated by differences in methodology of death certification between countries and over time.16 Population-based register studies are particularly useful in the study of ALS as they facilitate 1) prospective evaluation of progression of disability and 2) application of diagnostic investigations such as neurophysiologic testing to clarify the diagnosis.17 The continuous collection of clinical data over the course of the illness also facilitates establishing any possible cause and effect relationship, and prospective studies are recognized as being more accurate than retrospective studies in measuring the incidence of uncommon diseases such as ALS.18
Comparison of crude rates between countries is fraught with difficulties because of potential differences in population structure. Therefore, for meaningful comparison, the crude incidence rates should be adjusted by standardizing to a selected standard population. In this study, the incidence rates of the studies being compared were adjusted to the 1996 Irish population using the direct method of adjustment.19 We compared the incidence rates for the 45 to 74 age band of 13 incidence studies of ALS, including: the current Irish study (1995 to 1997); eight studies judged to have the most complete case ascertainment17; two recent population-based studies from Scotland1 and western Washington state2; an incidence study from Texas20; and data from a mortality study of the Irish population that reported on two distinct time periods (1968 to 1977 and 1978 to 1987)6 (table 2). Only the Rochester, MN4 and middle Finland21 studies showed higher incidence rates; furthermore, the difference with Rochester was small. Country standardized incidence rates were considered to be significantly different from the Irish ALS incidence rate if the 95% CI of that country did not overlap the Irish 95% CI. Only Israel (1959 to 1974),12 Sardinia (1965 to 1974),8 and Ireland (1968 to 1977)6 were statistically significantly lower than the Irish rate (1995 to 1997). The apparent reduced incidence in all three studies, which ended in the early 1970s, most likely reflects underascertainment due to relative lack of neurophysiologic support for diagnosis and referral patterns.
Comparison of epidemiologic studies of ALS standardized to 1996 Irish population and using 45–74 age band
It is interesting to note that the majority of the 95% CIs of the most complete studies overlap to a greater or lesser extent (see table 2), which may suggest that there is minimal geographic and temporal variation in the epidemiology of ALS. Recently, evidence pointing to a uniform geographic and temporal distribution of ALS throughout the developed world has been called into question.22 It remains unclear if the reported regional differences represent a genuine variation in ALS incidence or merely a product of varying case ascertainment due to improved diagnostic accuracy. Furthermore, as the population distribution of developed countries grows older owing to increased life expectancy, Gompertzian interdisease competition may account for some of these perceived differences.23 Direct comparison of incidence rates is further impeded by the methodologic differences of the various studies. However, the application of the El Escorial criteria may standardize comparisons with future studies.
The trend of the age- and gender-specific incidence for the Irish population is similar to that reported in most other studies in that there is a rapid increase after 40 years of age, reaching a peak at 60 to 79 and declining rapidly thereafter.17 It has been suggested that this decrease in the 75 plus age group may reflect difficulties with diagnosis and case ascertainment in the elderly rather than a genuine decrease in incidence.16 Only one study, based in Rochester, MN,4 showed a continuous rise in incidence with age. In our study, the incidence rates for the 75 plus age band were based on 43 patients representing 18.6% of the total 231 incident cases. The corresponding figures for the Rochester study were calculated from only 12 cases (total 44 cases) and rates based on small numbers may be unreliable. In addition, the CI of this age group in the Rochester study did not exclude a decrease in this age group.
Forty-four percent of Irish patients presented with bulbar symptoms, which is higher than that reported by other studies.24,25 One possible explanation is a greater than expected number of bulbar cases among women older than 60 years. The cause for this is uncertain, but it may reflect the prospective design of the study allowing improved case ascertainment among this demographic group. Alternatively, the rate of bulbar onset disease may be higher in Irish women than elsewhere, but there is no other firm evidence to support this notion. Contrary to other studies,24,26 the lower limbs were a more frequent site of onset of symptoms than the upper limbs, accounting for 31.2% and 17.7%, respectively. One possible explanation for this finding is that upper motor neurons traveling to the lower limbs are more vulnerable to the underlying pathologic process of ALS by virtue of their greater length.27
The crude prevalence of ALS in Ireland is similar to prevalence rates reported by other studies.21,28 However, there was a sharp decline in the prevalence rate for women older than 69 years, which occurred to a lesser degree in incidence rates. This may reflect poor case ascertainment in this demographic group, as diagnosis can be obscured by concomitant diseases of the elderly.17 Alternatively, survival time in elderly women may be shortened owing to the higher proportion of elderly women presenting with bulbar onset disease. Bulbar onset disease is recognized to be commoner in this demographic group and has a correspondingly poor prognosis.26
Compared with other studies,29 our incidence of PMA and PBP is strikingly low. The degree of case follow-up and the prospective design of the Irish ALS Register may account for the low rates of these clinical variants. It is readily apparent that the majority of patients presenting with bulbar symptoms progress to develop limb symptoms and vice versa. Our findings support the view that it is uncommon for an individual presenting with either bulbar or limb symptoms not to proceed to generalized involvement by the time of death.30 At the moment, neither of the two living patients with PMA or PLS have developed upper or lower motor neuron signs after several years of follow-up, but these patients continue to be reviewed.
Our figures suggest that there is a possibly increased incidence of ALS in northwest Ireland. Similar pockets of increased incidence of ALS have been previously noted in Finland,31 England and Wales,32 and the Unites States.33 The reasons for the increased incidence in northwest Ireland are unclear, and may be explained by the small number of cases involved. Follow-up is needed to determine whether incidence of ALS in this area continues to be higher than average (at least for one additional 3-year period of ascertainment). Pending that, etiologic speculations are premature.
Acknowledgments
Supported by the Irish Motor Neuron Disease Association and the Irish Brain Research Foundation.
Acknowledgment
The authors gratefully acknowledge the assistance of all the consultant neurologists, neurophysiologists, and primary care physicians who collaborate in recruitment for the Irish ALS Register.
- Received May 20, 1998.
- Accepted October 24, 1998.
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.
