Abstract
Keywords
ADHD is a relatively common behavioral syndrome beginning in childhood. Most sources suggest that ADHD affects between 3% and 7% of children today, although some estimates place the rate as high as 15% (Barkley & Brown, 2009; Schmidt & Petermann, 2009). Presenting symptoms include inattention, increased distractibility and forgetfulness, difficulty with sustained mental effort, executive deficits such as poor organization, and/or excessive activity, restlessness, and impulsivity (American Psychiatric Association, 2000). Children with ADHD often have difficulties with school performance and demonstrate increased levels of comorbid oppositional defiant disorder and mood disorders (Schmidt & Petermann, 2009).
It is now widely recognized that ADHD symptoms can persist into adulthood. What is unknown is the extent to which such symptoms may persist in late life, when a different types of memory problems and dementia typically arise. When patients present to memory clinics, the diagnosing of dementia is sometimes complicated by the presence of premorbid disorders, such as ADHD, which adversely affect cognitive functioning. Because patients with ADHD may have difficulty attending to specific stimuli, their encoding of information may be ineffective or incomplete. The deficits in executive functioning that are frequently associated with ADHD can further affect performance on neuropsychological assessment (Roth & Saykin, 2004).
In our memory clinic (Madison Veterans Affairs [VA] Hospital), we have occasionally encountered patients who reported lifelong difficulties with attention and memory, and whose background included undocumented school difficulties and frequent job changes. We have been challenged to distinguish between long-standing attention, memory, and executive problems versus those of recent onset, which may signify a neurodegenerative disorder.
Determining symptom etiology is complex in geriatric patients. According to the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text revision; DSM-IV-TR; American Psychiatric Association, 2000), attention difficulties must be apparent by the age of seven to meet criteria for the diagnosis of ADHD. Current recommended methods of diagnosis suggest using several sources of information, including the clinical interview and history, collateral input from family members who can confirm early attention problems, data from a neuropsychological evaluation, school records indicating attention difficulties in educational settings, physical examination, and a family history (Adler & Cohen, 2004; Weiss & Murray, 2003; Wender, 1998). Diagnosing ADHD is more complex in geriatric populations for several reasons. Typical sources of information may be unavailable to elderly patients (e.g., collateral information, decades-old elementary school records). Moreover, little is known about the presentation of ADHD in elderly populations, such as whether and how the syndrome may change or even remit as individuals age (Seidman, 2006). Yet, preliminary evidence indicates that in some patients, lifelong ADHD symptoms may develop into unique and cumulative patterns as patients age (Wetzel & Burke, 2008).
Results from a recent study examining ADHD symptom presentation in dementia patients and healthy controls suggested that patients with Lewy body dementia (LBD) demonstrated significantly higher levels of childhood impulsivity and hyperactivity than either Alzheimer’s disease patients or healthy controls. The authors hypothesized a common pathophysiological substrate in the relatively lower levels of dopamine and norepinephrine evident in patients with both LBD and ADHD. However, as the study participants were demented, and information regarding childhood symptoms was provided by an informant who may not have known the subjects in childhood, results must be considered preliminary at this time (Golimstok et al., 2010).
Attention difficulties are frequently seen in patients who develop cognitive impairment later in life. Disorders such as LBD (Kraybill et al., 2005), stroke (Welsh-Bohmer & Warren, 2006), traumatic brain injury (Okie, 2005), delirium (Rabin, Wishart, Fields, & Saykin, 2006), and Alzheimer’s disease (Cohen, Malloy, Jenkins, & Paul, 2006) may involve deficits in attention. In addition, as attention is an important component in memory processes, patients with impaired attention capacity are likely to develop memory deficits. A trained neuropsychologist using a comprehensive neuropsychological battery would likely be able to differentiate between attention and memory difficulties due to ADHD and those due to acquired cognitive dysfunction such as dementia. The pattern of deficits in ADHD usually involves a decreased rate of learning (Seidman, Biederman, Weber, Hatch, & Faraone, 1998), whereas dementia typically involves a rapid rate of forgetting (Welsh-Bohmer & Warren, 2006). However, the brevity of many memory-screening batteries and the overlap in symptoms may result in uncertainty regarding etiology. When geriatric patients present to memory clinics with cognitive impairments, the possible presence of ADHD may thus become a confounding variable. Given the paucity of data regarding the occurrence and implications of ADHD in late life, we conducted a cross-sectional study to assess the extent to which the staffs of memory clinics consider this issue to be concerning and how they address it.
Method
Study Population: Selection of Memory Clinics to Survey
We selected memory clinics from two sources, including all Department of VA medical centers with a Geriatric Research, Education, and Clinical Center (GRECC) except our own, and VA medical centers that reported the presence of a dementia clinic. In addition, we conducted an Internet search for memory clinics within the United States, verifying contact information with the national yellow pages. In all, we sent surveys to 165 memory clinics across the country.
Survey Questionnaire
The survey comprised 21 questions regarding whether and how memory clinics screen for ADHD, and how they address ADHD as a confounding variable. In addition, the survey solicited estimates of the frequency with which these clinics identify and diagnose patients with ADHD. Based on conservative estimates of adult ADHD in the general population at rates of 1% to 5% (Barkley, 1997), clinics were asked to rate the frequency of cases of ADHD in memory clinics as either rarely (1%-2%), occasionally (3%-5%), or frequently (6%-10%). The survey also inquired how they address ADHD as a confounding variable when patients with attention problems present in memory clinics. Finally, the survey inquired about the composition of memory teams in these clinics, and the types of clinical settings and catchment areas involved. A copy of the survey can be found in the appendix.
Results
We obtained an initial response rate of approximately 25% of clinics surveyed. We then mailed the survey out again, requesting clinicians who had not yet had the opportunity to respond to consider completing and returning the questionnaire. This resulted in an additional 17% response rate for a total of 42% (n = 62) response rate. In an attempt to gain fuller disclosure of actual clinical practice, all responses were anonymous; we did not seek or obtain any identifying information from responding clinics. Altogether, of the 165 surveys mailed, 12 were returned unopened, 2 were returned stating that no memory clinic existed at that location, and 2 were returned blank with no explanation. Sixty-two were returned, of which 57 were fully completed, whereas 5 did not respond to questions regarding how often they saw cases of ADHD.
Sources of responding hospitals and clinics included VA, university or university-affiliated clinics, community, private, and public. Eight hospitals and clinics reported dual affiliations (Table 1). Memory clinics that responded reported staffing to include the following: physician, social worker, neuropsychologist, psychologist, neuropsychology technician, registered nurse, occupational therapist, speech pathologist, rehabilitation therapist, and physician’s assistant (Table 2). Twenty-four clinics were in urban catchment areas, 5 in rural areas, and 29 in mixed rural and urban areas. One clinic did not respond to this question. As responses to the survey were anonymous, we were unable to make correlations between our specific clinics or types of clinics and response rates or particular clinical procedures. The following analyses include the 57 fully responding clinics.
Sources of Responding Hospitals and Clinics
Note: VA = veterans affairs. Some hospitals carry dual affiliations, such that numbers summate to greater than 100%.
Staff of Memory Clinics Reported
Note: Most clinics had more than one staff member, such that numbers summate to greater than 100%.
Approximately one half of the memory clinics that responded reported some contact with ADHD patients, either in evaluating patients with a prior diagnosis of ADHD or specifically making that diagnosis in clinic. Sixty percent (n = 34) of responding clinics reported having seen previously diagnosed cases of ADHD, whereas 40% (n = 23) reported not having seen cases of ADHD. More than half of respondents reported diagnosing ADHD for the first time in their memory clinics (see Table 3).
Contact With ADHD Patients in Memory Clinics
Note: MCI = mild cognitive impairment. Percentages have been rounded.
Approximately two fifths of responding clinics reported cases of ADHD with comorbid late-onset cognitive impairment. Patients with comorbid mild cognitive impairment (MCI) and ADHD were reported by a total of 24 of 57 clinics. A very similar number of respondents (n = 22) reported having seen cases of comorbid dementia and ADHD (see Table 3).
Methods of identifying patients with ADHD varied. As summarized in Table 4, although a quarter of clinics that responded reported using two or more methods to identify patients with ADHD, the majority used either one or none. The most frequent methods of identification were history and neuropsychological evaluation. Only three clinics reported accessing a collateral informant. These results are depicted in Table 5.
Number of Methods Used to Identify Presence of ADHD
Types of Methods of ADHD Identification
Conclusion
More than half of responding clinics (60%) reported contact with ADHD patients in their memory clinics, whereas less than half (40%) reported no contact with ADHD patients at all. As ADHD is reported in the adult population at rates of 1% to 5% (Faraone, Spencer, Montano, & Biederman, 2004; McCann & Roy-Byrne, 2000), this study raises the question of why ADHD may not be seen or recognized as frequently in memory clinics. Possible reasons are speculative, as these issues have not yet been systematically examined, but may be as follows. ADHD may be a less prevalent disorder in late life because its symptoms diminish over time (Halperin, Trampush, Miller, & Newcorn, 2008). The ADHD clinical profile may increase in heterogeneity with age, rendering it more difficult to recognize in older patients (Schmidt & Petermann, 2009). The validity of DSM-IV-TR criteria may be reduced in this age group (Simon, Czobor, Balint, Meszaros, & Bitter, 2009). Its symptoms may be misattributed to age-related cognitive decline (Smith & Rush, 2006). Patients with ADHD may be less likely to survive into old age due, for example, to a higher prevalence of risk-taking behaviors (Barkley, 2004) or comorbid disorders (Roth & Saykin, 2004). Patients with ADHD may be incarcerated or be more likely to present in addiction-treatment centers (Seidman et al., 1998). Based on our own experiences, we have identified additional possible reasons that ADHD may go unrecognized. First, data may be inaccessible that could confirm the diagnosis (e.g., old school records). Older patients with ADHD may also be inappropriately managed or referred by their physicians. In addition, ADHD behaviors may have been better tolerated and therefore may have been diagnosed less frequently in older cohorts than in current younger ones. The data from this study also raise the possibility that ADHD is underrecognized and underdiagnosed in the geriatric population.
Although 55% of responding clinics reported making new diagnoses of ADHD in their clinics, the reported means by which they made these identifications raises concerns. Of the 57 responding clinics, 38 indicated the number of methods used to identify ADHD. Approximately 44% (n = 17) of those clinics making new diagnoses of ADHD utilized at least two of the recommended methods for assessing symptoms, but more than half reported utilizing only one. Only two cited the use of an informant in the diagnostic process. One clinic reported using the attention scale from the Cognistat. This alone, however, would likely not constitute an adequate method of assessing ADHD as a potentially confounding factor, as this scale (as well as the Cognistat as a whole) includes very few items related to ADHD (Mueller, Kiernan, & Langston, 2001).
Although slightly more than one half of responding clinics reported newly diagnosing cases of ADHD, and approximately 40% reported seeing cases of ADHD comorbidly with both MCI and dementia, less than one third (30%) described attempts to address ADHD as a confounding variable utilizing more than one method to establish the diagnosis. These preliminary data suggest that in memory clinics that responded, ADHD symptomatology may not adequately be taken into account as part of the premorbid baseline against which late-onset cognitive impairment needs to be measured. These results raise additional concerns, as the literature shows that a multimodal approach is the most effective approach to assessing ADHD (Adler & Cohen, 2004; Weiss & Murray, 2003; Wender, 1998). Thus, it seems likely that some memory clinics may currently be unprepared to assess the significance of ADHD symptoms in the context of late-life cognitive impairments.
This study has a number of significant limitations, including its small sample size. As only 62 clinics responded, it is possible that responding clinics were self-selected, with those identifying ADHD patients more likely to respond. Other possible reasons for poor response rate include limited time on the part of respondents, or limited knowledge or interest in ADHD in older adults. The small sample size precluded making correlations between clinic composition and location, and methods of diagnosis and treatment. A substantial proportion of clinics that responded (25%) were associated with the VA system. As elderly patients in VA memory clinics are disproportionately male, and as the diagnosis of ADHD is made much more frequently in males, the collected data may be somewhat skewed. Another limitation may involve participants’ time constraints. Because some questions were open ended rather than multiple choice, participants may have underregistered methods of diagnosis actually utilized. Taken together, these considerations limit the generalizability of the results, and conclusions should therefore be regarded in a preliminary light. In addition, given the current absence of supportive evidence in the literature, our assumptions regarding the confounding role of ADHD symptoms in memory evaluations are speculative at this time and await corroboration.
The results from this study provide preliminary evidence regarding the need for greater awareness of late-life ADHD in memory clinics. The data highlight the importance of considering ADHD as a potential contributor to cognitive problems in late life as well as the need for more systematic approaches. These data also suggest that ADHD in the more general geriatric population may be underidentified. Given the burgeoning population of older adults in many countries, more attention needs to be devoted to studying the assessment of ADHD in late life and how this may complicate the diagnosis of late-life cognitive disorders.
Footnotes
Appendix
An earlier version of this manuscript was presented as a poster at the Gerontological Society of America Annual Meeting, San Francisco, CA, on November 16, 2007.
The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: William S. Middleton Memorial Veterans Hospital, Geriatric Research, Education, and Clinical Center (GRECC). This is GRECC manuscript No. 2008-02.