Appendix A. Epidemiology

Publication Details

This section will discuss the incidence and prevalence of several neurological disorders that result in dysphagia. This section also discusses the diagnosed occurrence of dysphagia in general, and within patient populations suffering from each of these neurological disorders. The purpose is to determine, using these rates in conjunction with vital statistics data, approximately how many people in the United States may be affected by dysphagia and resulting complications and morbidities (found in the Burden of Illness section). This information allows us to determine which disease causes the most individuals to suffer from dysphagia, and with this information, determine which patients should receive the most diligent management.

We have limited the scope of our review of the epidemiological literature on neurological disorders to studies conducted in the United States because there are often cultural differences in the epidemiological rates of many diseases. We have also limited the evidence base to studies published within the past 10 years because the rates of many diseases have been found to change over time. Studies on the occurrence of dysphagia within groups of patients suffering from particular diseases were not limited to U.S. populations, nor to the past 10 years, because while the rates of the diseases themselves appear to vary by culture or time, there is no reason to believe that the symptomatology of these diseases would differ among cultures or change over time.

Epidemiological Methods

Table 7 in the Epidemiology section of this report shows a summary of the findings for the incidence and prevalence of various neurologic diseases, and the diagnosed occurrence of dysphagia within each of these diseases. Estimates are wide-ranging, and, when examining the literature, there is not always an obvious reason for these variations. Such differences can be created most readily by differences in populations (ethnic makeup, age-distribution, sex-ratio); as a result, if a study does not show age-, sex- and race-adjusted results, the findings will reflect only that population in which the study was performed, rather than the U.S. population as a whole. Many of the epidemiology studies considered here did indicate age- and sex-adjusted findings; however, few, if any, included a race-mix adjustment. This unfortunately causes limitations in data generalization from such respected disease epidemiology centers as the Mayo Clinic, which samples from a 96 percent Caucasian population.

Another sampling method that can affect results is the residence of the subject. A common method of disease surveillance involves review of hospital records. The limitations of this method are obvious-a hospital population does not necessarily represent the state of health in the population as a whole (usually leading to an overestimate of disease prevalence), and patient charts may include miscodings or omissions leading to miscounting of diseases. Door-to-door surveys suffer from some of the same problems. Very often, these community interviews will miss people who are in institutions, such as nursing homes or hospitals, thus underestimating the rates of diseases.

The method used to confirm the disease is another important consideration. In hospital record review, it often isn't known whether the admitting diagnosis was confirmed on examination, or by what method. For example, a patient may be admitted with a diagnosis of acute stroke, but it may be the case that this is not confirmed on CT examination. The patient's record, however, may still include the DRG code for acute stroke. Any epidemiological study that does not impose diagnostic requirements for disease confirmation runs similar risks. Some studies on major neurological diseases may identify patients through associations devoted to that disease, plus clinicians in the area who treat such diseases. Suspected cases are then examined by a study physician, and only included as a case if certain diagnostic criteria are met. This diagnostic confirmation is really the only way to absolutely control prevalence estimates; however, this example included a way of identifying patients that may be flawed. When examining the epidemiology of a disease in a defined geographical area, many researchers survey local physicians and associations to locate possible cases. This requires the assumption that all cases would have contacted the association or the relevant specialist within that defined area. It therefore requires a very isolated population. Such places are rare in the United States, and there is always the possibility that a patient traveled to a major medical center in the nearest metropolitan area for treatment, or has not sought treatment at all.

If epidemiology is determined retrospectively (through case review), the accuracy (as mentioned above) and extensiveness of the records are important, especially in prevalence studies. When examining annual incidence, it may be fairly easy to contact patients and physicians to obtain information about recent health status. However, when prevalence is of interest, accurate records are needed that extend back several years. Many hospitals do not keep such extensive records, and their records are generally kept by admission DRG code rather than by patient; this allows for the probability of double-counting some patients.

An alternative to this is offered by the Mayo Clinic, which has an extensive medical record linkage system that tracks cases by patient rather than by diagnosis or admission. This system tracks both inpatient and outpatient medical visits in the Olmstead County Minnesota area, and has records back to 1917. This makes it one of the most valuable and often-used tools for epidemiological research in the United States. Its only limitation is a largely white, upper-middle class population whose rates of disease may not be generalizable to the U.S. population as a whole.

The statistical measures are important as well. To compare the results from various studies, we must be able to evaluate exactly what each study is measuring. Thus, if a study appears to measure prevalence, it is important to determine if they measured it using a point prevalence method (i.e., all cases that exist on a particular point prevalence date), or a range prevalence method (all cases during the years 1988-1996). These two different methods are not easily comparable, and the latter may actually be more suited to determining average annual incidence. Similar problems occur when examining studies on annual incidence; some report an incidence per year, while others report an incidence per person-year. These methods also may not be validly comparable.

Given the possibilities of study limitations as discussed above, the best study or studies about each neurologic disorder will be described below, with explanations about why that particular study (or studies) is superior to the others located in the published literature.

Epidemiology of Dysphagia in the U.S. Population

Dysphagia can include any swallowing problem originating in the mouth, pharynx, or esophagus (see the section entitled Dysphagia Defined for further discussion). Because we are focusing on oropharyngeal swallowing disorders, we will parse out any statistics pertaining specifically to these disorders whenever possible when reviewing the epidemiology literature in this section.

Studies on the epidemiology of dysphagia in the United States (without limitation to specific disorders or diseases) are shown in Evidence Table 1. Again, it must be stressed that these studies do not report definitive statistics on the epidemiology of dysphagia; rather, they report on the diagnosed occurrence, which will vary depending on the diagnostic tool(s) used.

In surveys of hospital patients (a commonly used tool, but biased because the patients surveyed are not representative of the population-at-large), dysphagia has been diagnosed in 12.5 percent (Groher and Bukatman, 1986) to 16.6 percent (Layne, Losinski, Zenner et al., 1989) of people of all ages, fairly consistent findings. However, swallowing function may gradually deteriorate as a function of aging (Shaker and Lang, 1994), and it is also a symptom of several neurological disorders more common among the elderly than the rest of the population, including stroke, Alzheimer's disease, Parkinson's disease, and motor neuron disease (to be discussed below). Therefore, the prevalence of dysphagia should be higher in an elderly or nursing home population than in the general population.

However, research investigating this has yielded mixed results. Rates for dysphagia in the elderly have ranged from 14.3 percent (Baum and Bodner, 1983) to 34.1 percent (Layne, Losinski, Zenner et al., 1989). The lowest rate was a survey of community-dwelling elderly, while the highest was a survey of elderly in a nursing home unit of a hospital. Diagnostic methodology was similar. Thus, differences in this rate are probably largely accounted for by the functional level of the population. Because of this increased incidence in nursing home patients, and because of their general tendency to experience increased morbidity from debilitation, it is especially important to ensure that dysphagia is diagnosed and treated correctly in these patients.

Epidemiology of Dysphagia Originating from Neurologic Conditions

A list of neurological causes of oral-pharyngeal dysphagia is shown in Table 21.

Table 21. Neurologic Causes of Oropharyngeal Dysphagia in the Elderly.

Table

Table 21. Neurologic Causes of Oropharyngeal Dysphagia in the Elderly.

No information was found in a Medline search on the occurrence of dysphagia in bulbar poliomyelitis, muscular dystrophy, hyperthyroidism or hypothyroidism, peripheral neuropathy, or on the overall prevalence of these disorders in the U.S. population. Therefore, we will focus on those relatively common disorders in the elderly for which there are data on both the overall incidence or prevalence of the disorder and on the diagnosed occurrence of dysphagia in individuals with that disorder, including stroke (cerebrovascular accident, or CVA); Parkinson's disease, Alzheimer's disease, amyotrophic lateral sclerosis (ALS), and a few other less common degenerative neurologic diseases. Neoplasms are also not considered in this report.

Epidemiology of Stroke in the United States

Six published studies were located that reported the annual incidence rate1 of CVA in the United States using various measurement techniques (Barker and Mullooly, 1997; Brown, Whisnant, Sicks et al., 1996; Kittner, White, Losonczy et al., 1990; McGovern, Burke, Sprafka et al., 1996; Modan and Wagener, 1992; Sacco, Boden-Albala, Gan et al., 1998). Prevalence rates cannot be examined for CVA, because it is an event rather than an ongoing disorder. The prevalence of cerebrovascular disease, on the other hand, can be examined; however, we are primarily interested in dysphagia that occurs shortly after an acute accident. Cerebrovascular disease overall was reported in the 1995 National Nursing Home Survey as the primary admitting diagnosis in 2.4 percent of elderly individuals (Dey, 1997). It is the third leading cause of death among individuals 65 and older (Department of Health & Human Services, 1997 ).

Evidence Table 2 shows six published studies on CVA incidence in the United States. Incidence rates for all people (all races, ages, and both sexes) were reported by two studies, and range from an estimate of 145 per 100,000 people per year (Brown, Whisnant, Sicks et al., 1996), to 289.7 per 100,000 people per year (Modan and Wagener, 1992). The lower rate comes from the Mayo Clinic records, and therefore records all strokes, whether hospitalized or not; this provides a more accurate reflection of incidence rates than a hospital study, which produced the higher estimate listed above. However, as mentioned before, since the Mayo Clinic represents a largely Caucasian population, this may skew results, as African-Americans and Hispanics have been found to demonstrate higher rates of stroke (Sacco, Boden-Albala, Gan et al., 1998).

A sex discrepancy, in which women suffered fewer strokes than men, was noted in all studies reporting these statistics. Within each of these studies, there is a consistent trend that the stroke risk ratio for women to men is approximately 1:1.5; this seems to indicate that men have a 50 percent higher risk of having a stroke than do women.

Age-specific annual incidence rates for elderly people aged 65 to 74 were available from four studies shown in Table 22. Rates among both men and women varied more than three-fold. Two studies (McGovern, Burke, Sprafka et al. 1992; Barker and Mullooly, 1997) were hospital records surveys; these two studies may be not be accounting for strokes treated outside a hospital system.

Table 22. Annual Incidence of Stroke per 100,000 Individuals Age 65 to 74 Years.

Table

Table 22. Annual Incidence of Stroke per 100,000 Individuals Age 65 to 74 Years.

Incidence rates for individuals 85 years and older were available from two of these studies, and are shown in Table 23. Of particular interest is the Brown study, in which the rates of women were considerably higher than those of men (Brown, Whisnant, Sicks et al., 1996). Interestingly, these same researchers also found that the mean age of onset was 69 years for men and 77 years for women; the combination of these two findings might suggest that women simply have delayed onset in stroke incidence compared with men rather than an actual lower rate.

Table 23. Annual Incidence of Stroke per 100,000 Individuals Age 85 Years and Over.

Table

Table 23. Annual Incidence of Stroke per 100,000 Individuals Age 85 Years and Over.

To summarize, the annual incidence of stroke in the United States has been estimated to be between 145 and 290 per 100,000 individuals. Caucasians have a lower incidence than African-Americans and Hispanics, and women have lower incidence than men, except among the oldest elderly. Incidence increases with age: populations age 65 to 74 show incidence of 515 to 2,315 per 100,000; those age 85 and over range from 1,522 to 2,574 per 100,000. Rates vary depending on the study methodology (hospital survey versus medical records/physician survey) as well as population differences (age, sex, race, and place of residence). It therefore appears that both environmental and genetic influences affect risk for a CVA.

Epidemiology of dysphagia in stroke patients

The location and severity of a stroke lesion will determine its effects on swallowing function. Hemispheric and brainstem strokes are both associated with dysphagia; however, unilateral hemispheric strokes are less likely to cause problems because there is bilateral representation in the brainstem (Morrell, 1992). Also, posterior lesions that do not affect the motor cortex may not lead to dysphagia (Logemann, 1983c).

Brainstem and anterior cortical strokes most often lead to dysphagia characterized by a delay in triggering of the swallow reflex. Dysfunction of the cricopharyngeal muscle is also common (Logemann, 1983a; Morrell, 1992). In cortical strokes, lingual and pharyngeal paresis may be unilateral, thereby allowing for compensation by turning the head toward the paretic side while swallowing (Mendez, Friedman, and Castell, 1991).

Because dysphagia is so often a temporary result of an acute stroke, we have limited our discussion to those studies that specifically examined the occurrence of dysphagia shortly after an acute stroke event using the water swallow test, Bedside examination (BSE), videofluoroscopic swallowing study/modified barium swallow (VFSS/MBS), or fiberoptic endoscopic examination of swallowing/fiberoptic endoscopic examination of swallowing and sensory test (FEES/FEESST). Any studies that did not define their diagnostic method, tested patients for dysphagia later than 7 days post-stroke, included transient ischemic attacks (TIAs), or that did not test 100 percent of the consecutive stroke population for dysphagia were excluded. This eliminated 8 out of 19 studies identified (Berg and Mor, 1995; Chua and Kong, 1996; Gresham, 1990; Hamdy, Aziz, Rothwell et al., 1997; Robbins, Levine, Maser et al., 1993; Selley, Roche, Pearce et al., 1995; Taub, Wolfe, Richardson et al., 1994; Teasell, Bach, and McRae, 1994). Evidence Table 3 shows the study results from 13 worldwide studies of the diagnosed occurrence of dysphagia after a CVA.

Many of these studies were hospital surveys; this may be an advantage; such studies interview patients just after the acute CVA because dysphagia is often most severe immediately after the event (Barer, 1989; Smithard, O'Neill, England et al., 1997; Smithard, O'Neill, Parks et al., 1996). This allows us to assess specifically the occurrence of dysphagia across studies without the confound of time variations.

Of studies conducted exclusively in a hospital setting, the reported occurrence of dysphagia among stroke patients ranged from 19.4 percent (Nilsson, Ekberg, Olsson et al., 1998) to 90 percent (Kidd, Lawson, Nesbitt et al., 1993). Diagnosis in the Kidd study was made on the basis of videofluoroscopic study; this method often increases the number of positive tests because it detects silent aspiration as well as symptoms that are not detected by and may never affect the patient. This is supported by the fact that these same authors assessed the same 60 patients using a water swallow test, and only 25 patients (41.7 percent) were diagnosed dysphagic on the basis of coughing, choking, or wet voice. There is an overall trend in the hospital-based literature that diagnosis made using results of VFSS result in higher prevalence rates [65.5 percent (Daniels, Brailey, Priestly et al., 1998); 74.6 percent (Daniels, McAdam, Brailey et al., 1997); and 90 percent (Kidd, Lawson, Nesbitt et al., 1993)] than those studies using other methodologies such as observation and BSE [19.4 percent (Nilsson, Ekberg, Olsson et al., 1998); 29.4 percent (Barer, 1989); 30 percent (Axelsson, Asplund, Norberg et al., 1989); 31.5 percent (Smithard, O'Neill, England et al., 1997); 41.7 percent (Kidd, Lawson, Nesbitt et al., 1993); 45.1 percent (Gordon, Hewer, and Wade, 1987); 49.6 percent (Smithard, O'Neill, Parks et al., 1996); 59.0 percent (DePippo, Holas, and Reding, 1994)].

One study did not limit its patient pool to hospitalized patients (Wade and Hewer, 1987). Wade and Hewer (1987) reviewed all records of stroke patients for whom evaluation of swallowing function (as measured by a water swallow test) was available within the first 7 days after the stroke event. They did not report overall swallowing disorders or dysphagia, but rather reported occurrence of individual symptoms: 14.2 percent choked, 6.2 percent displayed an abnormal swallowing pattern, and 22.6 percent demonstrated an abnormally slow swallow. It is impossible to determine the overall rate of swallowing disorders from this study because many patients may have shown more than one symptom; however, this study does provide a portrait of the typical symptoms experienced by a stroke patient.

The occurrence of dysphagia in a stroke population may differ depending on where the stroke occurred (left hemisphere, right hemisphere, both, or brainstem). It used to be thought that dysphagia occurred primarily after brainstem strokes, but more recent evidence suggests that it occurs at a similar rate after hemispheric strokes (Lugger, 1994). Of the studies discussed above, most of those that reported lesion location had a patient population made up largely of hemispheric stroke patients (85 to 100 percent of patient pool) (Barer, 1989; Daniels, Brailey, Priestly et al., 1998; Gordon, Hewer, and Wade, 1987; Kidd, Lawson, Nesbitt et al., 1993; Nilsson, Ekberg, Olsson et al., 1998). The reported occurrence of dysphagia in these patient populations ranged from 19.6 percent (Nilsson, Ekberg, Olsson et al., 1998) to 65.5 percent (Daniels, Brailey, Priestly et al., 1998). No studies exclusively examined a brain-stem stroke population.

There is consistent evidence that dysphagia is a transient symptom for many stroke patients. Barer (1989) followed 357 hemispheric stroke patients admitted to a hospital within 48 hours of the event. Swallowing was assessed with a water swallow test. At day 1 assessment, 105 of 357 patients demonstrated swallowing impairment (29 percent). This decreased steadily until, at 6 months, only 1 out of 248 (0.4 percent) assessed still demonstrated swallowing dysfunction (Barer, 1989).

The rate of swallowing impairment was related to the age of the patients at initial assessment: while 23 percent of those patients under age 70 were swallowing impaired, 36 percent of those over age 70 were. This was a significant difference (p<0.01). This difference disappeared by 1 week, however. The authors suggest that this may be due to the high mortality rate of the elderly patients with dysphagia, leading to a decrease in statistical power of the findings (Barer, 1989).

Smithard et al. (1997) followed 121 consecutive stroke patients admitted to a hospital within 24 hours of a stroke. They followed the patients for up to 180 days. At admission, 61 (51 percent) demonstrated a compromised swallow using a BSE method of assessment. At day 180, 8 out of 73 assessed (11 percent) demonstrated dysphagia. Overall, 6 cases out of 61 original cases (9.8 percent) were known to persist to 6 months. Some cases developed after the initial assessment (24 total); all but 2 were transient (Smithard, O'Neill, England et al., 1997; Smithard, O'Neill, Parks et al., 1996).

In conclusion, dysphagia is a symptom experienced by many stroke patients just after the event, occurring in 20 to 90 percent of patients, depending on the diagnostic methodology. It is impossible to determine which of these studies provides the most accurate rate, because, as mentioned before, dysphagia is not a defined disease, but a cluster of symptoms that may not occur all the time and the diagnosis of which may be influenced by subjective judgment of the clinician. A median of similar studies provides the best overall estimate: hospital studies using videofluoroscopy had a median diagnosis rate of 74.6 percent (Daniels, McAdam, Brailey et al., 1997). Those using BSE or a water swallow test had a median of 41.7 percent (Kidd, Lawson, Nesbitt et al., 1993). The difference between these two estimates is accounted for both by silent aspiration and physiologic dysfunction that may not affect the patient's health. If it were found that all of the cases of dysphagia diagnosed with videofluoroscopy caused significant disability to the patient, then it might be warranted to test all acute stroke patients for dysphagia. However, there is no evidence that this is the case.

Limited evidence suggests that dysphagia occurs more often in older stroke victims. It appears that hemispheric strokes are commonly characterized by dysphagia, despite commonly held beliefs that dysphagia usually occurs only after a brainstem stroke. Dysphagia is most often a transient symptom of acute stroke, with a large majority of cases spontaneously recovering within 6 months.

Epidemiology of Parkinson's Disease in the United States

Parkinson's disease was reported as the admitting diagnosis in 2.1 percent of all elderly admitted to nursing homes in the 1985 National Nursing Home Survey conducted by the U.S. Department of Health and Human Services (Hing, Sekscenski, and Strahan, 1989). Three other studies on prevalence rates of Parkinson's disease in the United States were found in the medical literature and are shown in Evidence Table 4.

The methods of these three studies are fairly distinct from one another. Lilienfeld et al. (1990), who reported the lowest rates, reviewed acute-care hospital discharges over a five year period (excluding individuals over 75 years of age) (Lilienfeld, Sprafka, Pham et al., 1991). Mayeux et al. (1995) established a community registry by polling health maintenance organizations, hospitals, and nursing homes; they interviewed everyone (except those over 75 years old) suspected of having Parkinson's disease for confirmation (Mayeux, Marder, Cote et al., 1995). Mitchell et al. (1996), who reported the highest rates, focused solely on nursing home residents, examining ICD-9 diagnoses (Mitchell, Kiely, Kiel et al., 1996). Their rates ranged from 45 per 100,000 (Lilienfeld, Sprafka, Pham et al., 1991) to 2,250 per 100,000 (Mitchell, Kiely, Kiel et al., 1996) when only primary Parkinson's disease is considered (Parksinson's disease not secondary to any other disorder). It is clear from these studies that the elderly experience more Parkinson's disease, and those in nursing homes have an even higher rate. One study (Lilienfeld, Sprafka, Pham et al., 1991) reported that women have a lower rate than men (45 versus 65 per 100,000).

Epidemiology of dysphagia in Parkinson's disease

Seven studies were identified that analyzed the diagnosed occurrence of dysphagia or swallowing problems in patients with Parkinson's disease (see Evidence Table 5 for details on study characteristics and results). Study designs of particular interest were those that withheld levodopa therapy (L-dopa) during the day of dysphagia testing, and those that reported disease severity on widely used scales (such as Hoehn and Yahr). Estimates of the occurrence of dysphagia in patients with PD range from 23 to 81 percent. These rates may be affected by duration and severity of disease, whether patients were taking medication at the time of testing, and the method of measuring dysphagia. Most of these researchers examined patients with mild to moderate Parkinson's disease (stage 1 to 3 out of 5 on Hoehn and Yahr Scale). Three studies reported on the relationship between severity of illness and severity of dysphagia; two studies (Fuh, Lee, Wong et al. 1997; Wintzen, Badrising, Roos et al. 1994) reported no correlation when severity of disease was measured using the Hoehn and Yahr scale. One study (Coates and Bakheit, 1997) reported a strong correlation between both severity and length of illness and severity of dysphagia, when severity of illness was measured using the United Parkinson's Disease Rating Scale (UPDRS) to assess disease severity and the Chicago Assessment Scale to assess level of swallowing dysfunction.

Those researchers who measured dysphagia using videofluoroscopy (VFSS or MBS) (three studies) found rates of dysphagia varying from 63.2 to 81.1 percent (Bushmann, Dobmeyer, Leeker et al., 1989; Fuh, Lee, Wang et al., 1997; Wintzen, Badrising, Roos et al., 1994). Questionnaires (3 studies) yielded rates from 22.9 to 76.9 percent (Coates and Bakheit, 1997; Hartelius and Svensson, 1994; Singer, Weiner, and Sanchez-Ramos, 1992). Most researchers reported rates in the 60 to 80 percent range. The mean age and duration of illness seem comparable in most of these studies and thus probably do not affect the variation of the rates. However, whether the patients were receiving L-dopa at the time of the study may have affected results. It is unclear whether severity of swallow dysfunction correlates with severity of Parkinson's disease.

Because L-dopa appears to affect swallow function in some patients (Bushmann, Dobmeyer, Leeker et al., 1989; Fuh, Lee, Wang et al., 1997), it appears that the best study would test patients while not taking the medication. Two studies did this (Bushmann, Dobmeyer, Leeker et al., 1989; Fuh, Lee, Wang et al., 1997), and found rates of 75 percent and 63.2 percent in groups of 20 and 19 patients, respectively. Both used MBS. There is no apparent advantage to either of these studies, so the mean will be used in the calculation of burden of illness, 69.1 percent.

Epidemiology of Alzheimer's Disease

Patients with Alzheimer's disease and other degenerative brain disorders made up 2.7 percent of all nursing home admissions in the 1985 National Nursing Home Survey (Hing, Sekscenski, and Strahan, 1989) (This was classified separately from senile dementia.) Two other epidemiological studies were found for the United States, one which looked at the population in general through review of clinical records, and one which focused on the African-American community in a cross-sectional interview study (Beard et al. 1991), using records from the Mayo Clinic record linkage system, identified all medical cases of Alzheimer's disease on prevalence date January 1, 1980. They reported age- and sex-adjusted prevalence rates of 259.8 per 100,000 overall and 2,634.1 per 100,000 people age 65 and older. Age-specific information indicated that this rate rose from 236/100,000 in those age 65-69, to 18,018/100,000 in those aged 85 and older (Beard, Kokmen, Offord et al., 1991). The Mayo Clinic serves a 96 percent Caucasian community; however, this was the only study on the overall population, and therefore is the study we will use for calculation of burden of illness.

Epidemiology of dysphagia in Alzheimer's disease patients

One study was located that examined the prevalence of swallowing disorders in Alzheimer's patients (see Evidence Table 7 for details). Horner et al. (1994), in a prospective case series, examined 25 patients age 55 and over with moderate or severe Alzheimer's disease using videofluoroscopy. Twenty-one out of 25 patients (84 percent) showed some sort of swallowing abnormality on videofluoroscopy. Six of these patients with abnormalities showed evidence of aspiration (24 percent of all patients). The most prevalent abnormality was delayed reflex initiation, followed by hesitancy of oral preparation and deficient pharyngeal clearance (Horner, Alberts, Dawson et al., 1994). This study suggests that an overwhelming majority of Alzheimer's patients experience swallowing problems; however, it is only one study and included only a small patient pool. No evidence is available about the occurrence of dysphagia in patients with mild Alzheimer's disease; it is therefore currently not possible to determine if there is a correlation between severity of Alzheimer's disease and severity of dysphagia.

Epidemiology of Dysphagia in Other Degenerative Neurologic Diseases

Several other less common neurologic diseases contribute to the prevalence of dysphagia in the elderly population. Discussed here only are those for which we could find published evidence that dysphagia is a common side effect: multiple sclerosis (MS), progressive supranuclear palsy (PSP), motor neuron disease (MND) including ALS, also known as Lou Gehrig's disease, and Huntington's disease. Those epidemiological studies examining the incidence or prevalence of these neurological disorders in the U.S. population are shown in Evidence Table 8; those studies examining the occurrence of dysphagia within each of these disorders are shown in Evidence Table 9.

Multiple sclerosis

The onset of MS usually occurs between the second and fifth decades (McFarlin and McFarland, 1982a) but because of its slow, insidious progression over several years, many elderly also suffer from it. It is characterized by the idiopathic demyelination of the white matter of the central nervous system. When certain cranial nerves are affected by this demyelination, dysphagia will occur.

MS was recorded in 0.6 percent (600 per 100,000) of all nursing home admissions documented by the 1985 National Nursing Home Survey (Hing, Sekscenski, and Strahan, 1989). Three other studies were identified that examined prevalence of MS in the United States. Two of these studies were investigations of possible cluster areas of MS (Helmick, Wrigley, Zack et al., 1989; Hopkins, Indian, Pinnow et al., 1991), and therefore are not appropriate for determining a U.S. generalizable burden of illness.

In Olmstead County, Minnesota, Wynn et al. (1990) examined patient records of the Mayo Clinic records linkage system, and discovered a sex- and age-adjusted prevalence of 170.8 per 100,000 on prevalence date January 1, 1985. This rate was significantly higher for women (231.9) than for men (79.8). Of those age 65 and older, the overall prevalence was 265.7 per 100,000, again, higher for women (369.1) than for men (89.7) (Wynn, Rodriguez, O'Fallon et al., 1990). Even though the Mayo Clinic records are very thorough, these figures may not be generalizable to the entire population, because of a latitudinal gradient of MS prevalence that has been discovered in North America (McFarlin and McFarland, 1982b), in which more southern areas have higher rates.

We identified one study that examined the diagnosed occurrence of swallowing disorders in individuals with MS. Hartelius and Svensson (1994) sent a questionnaire to members of the Multiple Sclerosis section of the Association for the Neurologically Disabled in Sweden. A total 278 questionnaires was sent and 203 returned (response rate of 73 percent). Of respondents, 68 percent were females, consistent with the observation that women are at higher risk of MS than men (Wynn, Rodriguez, O'Fallon et al., 1990); 33 percent reported that mastication and swallowing were more difficult than prior to disease onset; 13 percent reported having difficulty swallowing liquid at least fairly often; 16 percent reported problems swallowing solids at least fairly often; and 27 percent reported choking on food or drink at least fairly often (Hartelius and Svensson, 1994). An overall prevalence rate of dysphagia was not possible to determine given the self-report nature of this study. The occurrence of dysphagia reported in this study may be unrealistically high because of self-selection bias.

Motor neuron disease

MND is a family of neurodegenerative diseases in which there is a selective progressive depletion of motor neurons (Kirshner, 1989). There are three common subtypes of MND: ALS, the most common type, involves both the upper and lower motor neurons of the pyramidal tracts; progressive muscular atrophy (PMA) is characterized by motor neuron depletion in the anterior horn of the spinal cord; and bulbar palsy (BP) mainly affects the lower cranial nerve nuclei (Leighton, Burton, Lund et al., 1994). Progression of this disease is rapid and therefore life expectancy is generally short (less than 5 years).

The epidemiology of motor neuron disease in the United States was investigated in one published study, and ALS specifically investigated in two others. Lilienfeld et al. (1991) examined hospital discharge data for a 5-year period in a large metropolitan area; the average annual age-adjusted incidence of MND was estimated at 6 per 100,000 in 1984, the most recent year covered. For individuals age 65 to 74, men showed an incidence of 25 per 100,000, and women, 18 per 100,000 (Lilienfeld, Sprafka, Pham et al., 1991). The prevalence rate of MND was not calculated. These incidence rate estimates may be low because of the focus on hospital discharges.

Two studies looking specifically at ALS reported annual incidence rates of 1.14 (Annegers, Appel, Lee et al., 1991) and 1.8 (McGuire, Longstreth, Koepsell et al., 1996) per 100,000. In both studies, women appeared to have a slightly lower incidence than men. Incidence rates were higher among the elderly; Annegers et al. (1991) reported that individuals age 75 and over had a rate of 4.01 per 100,000; McGuire et al. (1996) reported higher rates, at 7.18 per 100,000 men, and 5.54 per 100,000 women. The Annegers study results came from surveys of neurologists, many of whom refused participation; therefore, these rates may be unrealistically low.

We identified two published studies that investigated the diagnosed occurrence of swallowing disorders in MND. Reported dysphagia rates were 51.1 percent (Leighton, Burton, Lund et al., 1994) and 71 percent (Mayberry and Atkinson, 1986). Dysphagia rates for ALS reported by Leighton were 29 percent; this study is likely superior to that of Mayberry et al. because it does not suffer from self-selection bias as a result of survey methods.

Progressive supranuclear palsy

PSP is often mistaken for Parkinson's disease because of its Parkinsonian symptom characteristics. It has a late middle-age onset, and is characterized by a high density of neurofibrillary tangles and neuropil threads in the basal ganglia and brainstem (Litvan, Agid, Calne et al., 1996). Symptoms include bradykinesia and rigidity, severe gait difficulty, and pseudobulbar signs. It progresses more rapidly than Parksinson's disease (Golbe, Davis, Schoenberg et al., 1988) and does not generally respond to L-dopa treatment. Onset of these symptoms has been reported to begin at a mean age between 55 and 70 years (Litvan, Agid, Calne et al., 1996).

Two studies were located that examined incidence or prevalence of PSP; Golbe et al. (1988), in a survey and exam, determined a prevalence rate of 1.39 per 100,000 (Golbe, Davis, Schoenberg et al., 1988). Bower et al. (1997), using the Mayo Clinic records linkage system, calculated an annual incidence rate of 1.1 per 100,000; this rose to 5.3 per 100,000 for those age 50 and older (Bower, Maraganore, McDonnell et al., 1997). These are the only two studies we found in the published literature on PSP, and they are not comparable given that one study calculated prevalence and one calculated incidence.

The most frequent cause of death in PSP is pneumonia (Litvan, Agid, Calne et al., 1996); therefore, determination of the occurrence and characteristics of swallowing problems is important. Litvan et al. (1997) assessed 27 patients with PSP using a swallowing questionnaire and ultrasound imaging of swallowing. Results indicated that all PSP patients had at least one swallowing complaint; 15 (55.6 percent) complained of difficulty swallowing-5 had more difficulty swallowing solids than liquids, and 10 had more difficulty with liquids than solids. Five (19 percent) demonstrated aspiration (Litvan, Sastry, and Sonies, 1997). This was the only published study identified that investigated swallowing problems in PSP patients.

Huntington's disease

Huntington's disease is a hereditary neurodegenerative disease affecting the basal ganglia, causing movement disorders, dementia, and emotional impairment (Kagel and Leopold, 1992; Leopold and Kagel, 1985). Dysphagia is very common in these patients, characterized by lack of control of the rate of food intake, difficulty with liquids, bolus retention in the buccal recesses, and choking. Many of these problems are side effects of cognitive impairment rather than directly the result of neurologic or neuromuscular dysfunction, per se (Leopold and Kagel, 1985).

One study was located that examined the prevalence and incidence of Huntington's disease in a U.S. population. Kokmen et al. (1994), at the Mayo Clinic, examined records from 1950 to 1990 and identified 10 definite cases of Huntington's disease. The prevalence on January 1, 1990 was 1.9 per 100,000 overall (1.8 for women and 2.0 for men). The average annual incidence rate between 1970 and 1989 was 0.2 per 100,000 (0.3 for women and 0.1 for men) (Kokmen, Ozekmekci, Beard et al., 1994).

One study examined the diagnosed occurrence of dysphagia in Huntington's disease; Kagel and Leopold (1992) examined 35 patients with Huntington's disease (mean age 45.5) using a clinical questionnaire, assessment of feeding, and videofluoroscopic exam. Twenty-seven patients (80 percent) coughed or choked on both solids and liquids; six were symptomatic with food only, and two with liquid only. Thus, 100 percent of patients with Huntington's disease showed dysphagia of one kind or another. In the hyperkinetic variant of Huntington's disease (30 of 35 patients), 29 (96.7 percent) demonstrated oral dysphagia, 27 (90 percent) pharyngeal dysphagia, and 11 (36.6 percent) esophageal dysphagia during clinical assessment. A total of 100 percent of hypokinetic-Huntington's disease patients (five patients) demonstrated oral and pharyngeal swallowing dysfunction; none demonstrated esophageal dysphagia (Kagel and Leopold, 1992). Selection methodology of these patients was not described; therefore, there is a possibility that the researchers specifically selected patients with dysphagia for study.

Summary

Dysphagia is quite commonly diagnosed in all of these neurological patients. Estimates have been very high for stroke, Parkinson's disease, Alzheimer's disease, and Huntington's disease; however, fairly low estimates have also been reported for those diseases on which there was more than one study published. These variations indicate interstudy differences in diagnostic criteria and methodology. It is therefore impossible to determine if dysphagia is prevalent enough in any one of these diseases that 100 percent of these patients should be tested for dysphagia; a determination whether or not to do this would be based on clinical judgment. It must also be pointed out that dysphagia, per se, is not a risk factor for pneumonia; it is the aspiration of certain substances that leads to pneumonia. Dysphagia, on the other hand, may put a patient at risk for malnutrition if the patient is unable to safely swallow adequate amounts of food.

Footnotes

1

Defined as the total number of new cases of stroke diagnosed in a given year. This is abbreviated in this report as incidence rate.