Screening for dementia is but a first step that can lead to a dementia diagnosis, and it is performed before an early diagnosis of dementia. However, calls for early detection of dementia may resonate well with screening: the value attributed to earliness is present in both cases.Footnote 3 At the same time, national screening boards typically do not recommend screening for dementia [see, for example, 15,24].Footnote 4 Even so, because of the widely-shared ambition to achieve early detection, various forms of intervention have been proposed, and a wide variety of conceptualizations of these interventions has been put forward. To complicate things further, the same test to assess cognitive impairment can be used both in case-finding and in screening (such as the Mini Mental State Examination and the Abbreviated Mental Test Scale) [5, 26].
Population-Based Screening Versus Screening Tests
While there is a consensus about the value of diagnosing dementia early, arguments are often raised against screening for dementia. Boustani et al., for instance, state that while some screening tests have “reasonable accuracy for detecting mild to moderate dementia”, uncertainties remain about the potential harms and benefits of population-based screening for dementia, and, most importantly, about the effectiveness of treatments for those whose dementia would be detected this way . The lack of evidence concerning the potential harms and benefits of a population-based screening program for dementia is also central to why national bodies in the UK and the United States have advised against this form of screening .
Ashford et al. are an exception in these discussions. They argue for the introduction of a population-based screening program for dementia [2, 3]. This makes these authors particularly interesting for our analysis: What makes this argument possible in the face of the wide criticism of screening for dementia?
Central to the argument Ashford et al. put forward is a sharp distinction between the initial screening test and later diagnostic procedures—a distinction that Wilson and Jungner did not make. Based on this distinction, Ashford et al. state that “it is legitimate to insist that screening tests be properly evaluated”, and that screening should not be “asked to bear the responsibility for negative consequences associated with a lack of available clinical expertise, supervision, and counselling once dementia is identified” . They propose a “new operating definition of dementia screening” (ibid.), in which the term “screening” refers only to the initial test. Negative implications of later diagnostic procedures, such as false-positive diagnoses, overdiagnosis or any negative effects of clinical treatment decisions, are not reason enough to advise against screening. In their words: the “absence of empirical data on the specific impact of screening on patient outcomes is not sufficient to justify a decision to recommend against it” (ibid.).
In narrowing down the definition of what screening is, conceptually delimiting it to the application of the test, Ashford et al. open up the possibility to distance themselves from and thereby re-evaluate the generally accepted screening criteria that state how screening should be performed. This allows them to argue that not only should the criteria be limited to the initial testing, but they should also be adapted to the particular circumstances of the disease in question, in their case dementia. Through these steps, they argue that population-based screening for dementia is justified.
Whereas Ashford et al. are among the few scholars who explicitly argue for this “new” understanding of screening, their suggestion ties into a conceptual lack of clarity. We note a tendency here: the line between the terms “screening” and “screening test” (or “screening tool” or “screening instrument”) is frequently blurred. While screening assessments in national settings include much more than the assessment of various screening tests, several examples can be found in the literature in which screening test assessment is conflated with screening assessment. This is most evident in publications that discuss and compare the efficacy of various screening tests without taking into account their long-term harms and benefits, later subsequent diagnostics, possible treatments, and so on. Even so, these assessments are used to indicate the viability of screening in general [3, 16, 22, 36]. While it is clearly the case that the various screening tests must be assessed with regard to their accuracy, specificity and sensitivity, this use of the concepts (in which screening test assessment is a form of screening assessment) may cause the phenomenon of screening to become primarily associated with the application of particular instruments. If this occurs, a re-evaluation of screening assessment criteria may be justified.
Case-Finding, Cognitive Impairment Assessment and Opportunistic Screening
While Wilson and Jungner equated the concept of screening with case-finding, a distinction between the two concepts has arisen in recent years. The UK “case-finding” scheme for dementia exemplifies this. From 2013, all persons above the age of 75 who are admitted to a hospital in England, unplanned and for more than 72 h, have been asked a “case-finding question” about their memory, after which a test intended to help identify cognitive impairment may follow . Even though this practice has several characteristics of screening (as defined by the UK National Screening Committee), the most notable of which is that it targets “members of a defined population, who do not necessarily perceive they are at risk of, or are already affected by, a disease or its complications”, it is termed “case-finding” in the official terminology .
Another example of a “screening-like” practice for which the term “screening” is not used is the “detection of cognitive impairment” that takes place as a part of the Medicare Annual Wellness Visit (AWV) in the US. The AWV is offered to all US citizens above the age of 65. During the visit, cognitive function is assessed in a procedure in which the physician observes and interviews the patient together with a knowledgeable informant (family member, friend, caregiver, other) about potential cognitive concerns. “If appropriate”, the regulations issued by the Centers for Medicare & Medicaid Services (CMS) state, “use a brief validated structured cognitive assessment tool” . However, the regulations do not provide any guidance about how to determine when the use of a cognitive assessment tool is appropriate .Footnote 5 Despite this uncertainty, this practice, just as the UK “case-finding” scheme, has several of the features of screening: it targets a predefined, pre-symptomatic population and involves control questions and/or the use of a brief test instrument. Even so, in the official terminology, the cognitive assessment that takes place during the AWV is referred to as “detection of cognitive impairment” or “cognitive impairment assessment” [10, 12].
So why is the term “screening” not used in these cases? A simple answer is that screening for dementia is not supported by the national health authorities in the two countries [18, 24]. It would therefore probably be next to impossible to introduce such a program, especially a population-based screening program, on a national scale. In both countries, however, the pressure to improve (early) detection and diagnosis of dementia has been high. This has caused politicians to act and launch programs such as those described above [19, 34]. However, in order to do so without contradicting the recommendation of the national health authorities, these programs cannot be called screening. They must be given other names.
What takes place here is a conceptual distancing from the concept of screening. Rather than arguing for a redefinition of what screening is, as Ashford et al. do, the message is that we are concerned with a completely different what, namely that of “case-finding” or “detection of cognitive impairment”. A deeper understanding of this conceptual approach can be gained by viewing it through the relationship between the what and the how. Due to the intimate bond between the what and how of screening, a distance from the concept of screening may enable a distancing from the demanding assessment criteria so closely associated with it. In other words, giving the practice another name may make it possible to introduce screening-like practices without the need to abide by the established criteria.
A generally accepted definition of case-finding has not been established, nor have any generally accepted assessment criteria for the process been laid down . The what and how of case-finding are thus more “open” than those of screening. As a result, a wide variety of practices can be labelled “case-finding”, and it is possible to distinguish several forms of it. As Ranson et al.  express it: “there is much ambiguity around what it [case-finding] means, particularly with respect to how it differs from screening”. They suggest that case-finding should be understood as:
an offer of a brief, opportunistic investigation to identify possible signs or symptoms of dementia, initiated by a clinician during consultation with a patient at high risk of dementia on the basis of clinical judgment that an initial dementia enquiry is appropriate and is likely to benefit the patient .
Ranson et al. suggest that two distinguishing features be used to define case-finding. First, case-finding does not, as screening does, target a predefined population, but is applied on a case-by-case basis. Second, the determination of whether case-finding should be applied or not is the result of an individual clinician’s clinical judgment about whether the procedure “is appropriate and is likely to benefit the patient”, not the result of identifying a particular population to whom the test is offered .
With Ranson et al., the possibility of introducing screening-like practices by naming them “case-finding” vanishes. In their conceptualization, the what of case-finding is filled with a particular content, one that clearly differentiates it from screening. A central aspect of this distinction is the disassociation of case-finding from one of the taken-for-granted whats of screening: its population-based character. Case-finding is instead connected to another what: the application of a test on a case-by-case basis, according to a particular clinician’s clinical judgment.
Having made this distinction, Ranson et al. surprisingly go on to reconnect case-finding with screening by arguing that “the criteria for assessing evidence for screening proposals […] also apply to case-finding” . Although case-finding is clearly distinct from screening, it should be assessed using the same standards as the latter, they argue. Thus, while disassociating case-finding from the what of population-based screening, Ranson et al. reconnect it with the how specified by the screening criteria. In their assessment of case-finding for dementia against the (UK NSC) screening criteria, they conclude that it does not fulfil the criteria and should currently not be offered .Footnote 6
Yet another distinction is also sometimes made: between “passive” and “active” case-finding [see e.g., 30, 31]. Mate et al. define “passive” case-finding as an activity in which “patients are evaluated for dementia because they or a caregiver bring a memory/cognition concern to their GP, or because their GP raises the issue based on their clinical judgement” . “Active” case-finding, on the other hand, they state, “specifies an opportunistic dementia assessment of ‘at-risk’ patients based on a number of factors including vascular risk factors, Parkinson’s disease and learning disabilities, in addition to subjective memory complaint” . The difference between active and passive case-finding, according to them, lies in the level at which the decision to test is made: if this is a clinical level the case-finding is passive, while if an “at-risk” population has been identified on the level of the health system the case-finding is active, and allows for the testing of individuals before they experience symptoms or show signs of disease. Mate et al.'s definition of passive case-finding is close to Ranson et al.’s definition, and results in a clear distinction from screening. This clear distinction (between the whats) leads Mate et al., unlike Ranson et al., to propose that passive case-finding should not be assessed against the same criteria as screening . Their conclusion is that passive case-finding is already taking place in primary care, and that the benefits of introducing a screening program in its place would not outweigh the potential harm associated with the latter . In other words, they argue that a difference in the whats should entail a difference in the hows.
Two contrasting effects of the definition of active case-finding proposed by Mate et al. can be identified. On the one hand, the definition makes it more difficult for them to uphold the distinction from screening—since active case-finding, in their use of the term, retains some of the population-based characteristics associated with screening. For this reason, Ranson et al. would likely conceptualize active case-finding as a form of screening, while the UK Department of Health can argue that this is the form their case-finding scheme takes. On the other hand, the definition helps set active case-finding apart from screening. Rather than a test that is systematically offered to everyone in the predefined group, active case-finding is offered when the opportunity arises. This opportunistic character is a defining feature also of passive case-finding. Above, we saw how Ranson et al. defined case-finding as an “opportunistic investigation” .
Here another conceptual complication arises, namely, how to distinguish case-finding from “opportunistic screening”. The term “opportunistic screening” is an example of a concept that is used to distinguish between different types of screening. When it is used, the population-based character that is ordinarily implicit in definitions of “screening” comes to the forefront. As Speechley et al. have shown, however, the concept of opportunistic screening is often used synonymously with case-finding in the scientific literature .Footnote 7 Why this is so becomes evident when we look more closely at some of the definitions of opportunistic screening that are offered. Opportunistic screening is generally defined in two slightly different ways. In some contexts, the emphasis is put on the actor asking for or offering the test, as in the definition used by the New Zealand National Screening Unit, which states that opportunistic screening “happens when someone asks their doctor or health professional for a check or test, or a check or test is offered by a doctor or health professional” . In other contexts, the emphasis is put instead on the setting in which the test is administered, as in the definition used by Ashford et al., in which opportunistic screening is understood as an activity applied “to individuals who for other reasons [than screening] come to a setting where screening might occur” . Although less detailed, these definitions have much in common with the conceptualizations of passive and active case-finding discussed above. The definition of opportunistic screening used by the New Zealand National Screening Unit overlaps with the definition of passive case-finding, since the decision to administer the test is taken by the actors involved in a clinical encounter. The definition of opportunistic screening proposed by Ashford et al. overlaps with the definition of active case-finding, since the testing is a result of a pre-existing structure of a particular healthcare institution, or of the healthcare system in general.
We see that it is difficult to separate the whats and the associated hows of case-finding and opportunistic screening. The choice between them and other concepts is ethico-politically charged. While “screening” might not be a proper term for the “case-finding” scheme that is used in the UK, since the testing is not offered universally to all citizens above the age of 75, evading comprehensive assessments by choosing the term “case-finding” instead of a concept that contains the term “screening” (such as “opportunistic screening”) is also questionable. We discuss below the ethico-political implications of such conceptualizations, and end the article by proposing a set of recommendations.