Based upon literature findings, African Americans are associated with a high prevalence of dementia, yet they are less likely to be reported and diagnosed appropriately according to the Alzheimer’s Association. Our findings suggested a similar pattern that nonwhite patients are less likely to be diagnosed with dementia/AD, but the association was not significant.
With regard to cholinesterase inhibitor therapy, older studies have published that minorities in the United States receive fewer mental health services compared to whites. Moreover, a more recent study using the NAMCS survey comparing rates from outpatient mental health treatment showed that Hispanics and blacks had lower visit rates per 1,000 for drug therapy than whites (48.3 and 73.7 vs. 109.0; pO.OOOl and pO.Ol, respectively). In our study population, a relatively few number of visits made by minority patients had a diagnosis of dementia/AD status and this could perhaps reflect underutilization or lack of appropriate diagnosis among minority patients in the ambulatory care setting.
As compared to the higher prevalence of dementia reported, 6-10% from the literature, our findings show that less than 1% of the ambulatory visits were identified with dementia/AD status. Although our findings from the NAMCS database report visit-based prevalence instead of patient-level prevalence, a few reasons for the apparent discrepancy can be addressed. The first reason is related to truncation bias, a limitation of using the NAMCS database. As is common with other databases, the NAMCS only includes up to three diagnoses. Therefore, if a patient visit was related to common clinical conditions other than dementia, it is possible that the dementia diagnosis would be excluded. Although our study definitions could have incorporated other proxy information, such as mentions, to identify dementia/AD status, it would have deviated from our study’s primary goal, which focused solely on the treatment patterns subsequent to a formal diagnosis with dementia/AD. The second reason for the inconsistency in the prevalence results could be related to care setting. As elderly patients with dementia may be frequently observed in long-term care and nursing home settings, the prevalence of dementia in the ambulatory care setting may be lower than the overall prevalence of dementia counting both noninstitution-alized and institutionalized patients. Therefore, the low prevalence observed from our study may reflect variation of dementia by setting.
Of the four cholinesterase inhibitor that are approved by the FDA, galantamine, and donepezil tablets were prescribed in our sample for the treatment of AD. However, tacrine was not prescribed at all, perhaps because of its association with hepatotox-icity. Donepezil was the most prescribed of the cholinesterase inhibitor among the visits; this finding is consistent with other studies that showed that is the most common agent prescribed among patients diagnosed with dementia.
Our study showed that slightly less than half of office visits with a dementia diagnosis had a cholinesterase inhibitor prescription. According to the
latest practice guidelines from the American Academy of Neurology, cholinesterase inhibitor should be considered in patients with mild-to-moderate AD as clinically significant changes in cognition, behavior and functioning have been detected. Our findings on the modest prevalence of cholinesterase inhibitor prescription may provide only a snap-shot view of current practice. It may reflect that patients with dementias other than AD are not prescribed with cholinesterase inhibitor, or routine prescriptions are not provided to patients with conditions such as mild cognitive impairment, early AD, severe AD and other types of dementias in which no adequate controlled trials demonstrated pharmacological efficacy. In fact, our subgroup analysis showed that AD visits were about twice as likely to be associated with cholinesterase inhibitor prescription (OR=2.28; 95% confidence interval of 1.05,4.95; p=0.034), supporting adherence to the current practice guidelines. Although these cholinesterase inhibitors have shown cognitive improvement over placebo, other potential explanations for this rather conservative use can be based on uncertainty as to whether the drugs slow the progression of the disease, delay nursing home placement or change the mortality rate as suggested by Rabins.
Early detection of many clinical conditions is one of the most critical steps for better outcomes of the diseases. Practitioners and clinicians in ambulatory care settings are in a unique position as early symptoms can be communicated to them by patients or family members. According to Richards and Hen-dries, therapy intervention for dementia should involve families in order to provide support aid in long-term care and legal decision-making for the patient prior to the onset of diminished capacity. Early detection of dementia among patients with mild cognitive impairment was discussed in recent practice guidelines. According to the guidelines, patients with mild cognitive impairment should be recognized and monitored for cognitive and functional decline as the individuals have a high risk of progressing to dementia or AD. Therefore, continuous monitoring of patients with mild cognitive impairment is an important practice in ambulatory care settings.
Not only early detection, but also referral of a potential dementia patient to a specialist is another important process to initiate appropriate treatments. Although the study did not explore if visits recorded with dementia were referred by specialists, our findings suggest that the referral of dementia/AD patients from primary care physicians to specialists has occurred as psychiatrists and neurologists are the ones who predominantly provide ambulatory care services for demented patients and are most likely to prescribe cholinesterase inhibitor. fosamax medication
Using NAMCS data introduces several limitations. Each record of NAMCS represents a visit, not an individual patient. Therefore, estimated prevalence of dementia or cholinesterase inhibitor prescription should be interpreted accordingly. In addition, clinical and social factors potentially affecting patterns of cholinesterase inhibitor prescription or dementia diagnosis due to personal preferences or drug coverage status of the patient cannot be controlled. As briefly mentioned earlier, the possibility of truncation in drug records among visits with high prescribing volume cannot be ignored as NAMCS drug data contain up to six medications. In our study population, about 90% of visits received 0-5 prescribed medications. Therefore, about 10% of visits were associated with a possibility of bias due to truncated drug records.
In summary, considering the high prevalence of dementia/AD among the elderly in the United States, the study findings suggest more efforts should be given to detect dementia in the ambulatory care setting and to treat AD patients with cognitive-enhancing agents after the formal diagnosis. In addition, our findings demonstrate that physicians specializing in psychiatry and neurology predominantly provide ambulatory care services for dementia patients.