A missed detection of prodromal dementia may be the worst enemy of a timely diagnosis.

To face the challenge of the under-diagnosis of dementia [1] and to meet the crucial need for its timely diagnosis [2], many countries have adopted a stepwise case-finding diagnostic strategy [3], [4]. The first step usually consists of nonspecialist screening of cognitive impairment in a primary care setting. If it is positive, the patient is referred to a secondary care service where a full evaluation is carried out; if it is negative, the patient is generally referred to a follow-up screening some time later [4]. This strategy probably represents a great advancement in the diagnosis of dementia, and its efficacy is supported by several lines of evidence indicating that cognitive screening can considerably improve the identification of dementia [5]. Nonetheless, it should be noted that a nonspecialist first evaluation has the potential to be rather inaccurate. It can definitely result in false negatives as well as in false positives. For instance, a recent study addressing the effect of screening for cognitive impairment in elderly veterans has shown that many individuals who were found to be normal at a brief cognitive test were recognized as having dementia at a subsequent full evaluation [6]. Conversely, a widespread cognitive screening tool (i.e., the Montreal Cognitive Assessment) was found to yield many false positives when used to detect dementia [7]. A certain number of false negatives may be acceptable for a case-finding strategy [4] and false positives would be an economic, but not a diagnostic issue, as overdiagnosis would likely be amended at the second step of the full evaluation. However, to maximize the benefits of a timely diagnosis of dementia, it is currently believed that it should indeed be made at a prodromal (i.e., at mild cognitive impairment, MCI) stage [8]. At this stage, we would expect the first nonspecialist evaluation to be more inaccurate, especially in terms of false negatives, as prodromal signs of dementia can be subtle and difficult to recognize and require more extensive cognitive examination [9]. Moreover, a missed detection could have worse consequences on the diagnostic process at a prodromal rather than at a more advanced stage of dementia. In fact, it is likely that patients with mild-to-moderate dementia unidentified at the first screening would ask for a new referral very soon, because of significant cognitive and/or behavioral disturbances causing distress to themselves and their families. On the other hand, patients with prodromal dementia going undetected on first screening might not feel the need for a new referral for a long time, because their subtle or selective disturbances are probably less distressing. Thus, an inaccurate first detection of prodromal dementia can significantly delay its recognition and become the worst enemy of a timely diagnosis. In sum, the current diagnostic strategy does not seem to be able to adequately cope with the challenge of a timely diagnosis of prodromal dementia and could thus result in the loss of a whole range of potential benefits: for patients (e.g., improving quality of life), caregivers (e.g., developing appropriate care plans), health care professionals (e.g., providing patients and families with important information), and society (e.g., reducing health care costs) [3], [8]. It is likely that in a near future, a more accurate diagnosis of prodromal dementia will be possible in primary care when reliable biomarkers of the disease (e.g., low Aß42 in the cerebrospinal fluid) or other attractive advanced tests (e.g., the cholinergic stress test) [10] will be routinely available in a clinical setting. In the meanwhile, we believe that it is possible to improve the current strategy by incorporating a new diagnostic stage between the nonspecialist screening and the specialist full evaluation. We envisage an intermediate stage—we may call it a “1.5 stage of full detection”—where a “frontline” dementia specialist (i.e., a behavioral neurologist, a neuropsychologist, a geriatrician, an old age psychiatrist, or an advanced-practice nurse) would cooperate side by side with the primary care doctor, by performing additional skilled evaluations of the patient's cognitive, affective, and behavioral status. The implementation of this new model in countries whose health care system is organized in primary and secondary care would require creating new shared spaces where a close collaboration between generalists and specialists may be achieved (e.g., district memory clinics) and converting a number of specialists to full-time consultants for primary care services. In conclusion, there is current emphasis on the role of primary care services in the diagnosis and management of dementia, and this is certainly important, but it should be borne in mind that we urgently need dementia specialists to assist primary care physicians in the assessment of dementia at a prodromal stage. Yet, dementia specialists are incredibly few [2], and it takes time to train them. The clock is ticking…