Screening for dementia: why it helps, and how we can do better

dementia medical school research
By Siva M.

The dementia spectrum is enormous, spanning a complex range of cognitive, pathophysiological, and social factors which vary from person to person. We pour billions annually into research—trying to get to root causes and fund clinical trials—but for most conditions, we’re far away from a cure. Treatment is usually focused on symptom management and therapy: trying to slow the decline and helping patients maintain a high quality of life.

In this setting, early diagnosis is crucial. The sooner someone is diagnosed, the sooner they can work with their doctor to start on a treatment plan, and the easier it’ll be to navigate daily life going forward. For well-studied diseases like Parkinson’s and Alzheimer’s, research has shown astonishing differences in treatment effectiveness based on the stage of diagnosis. The problem is that early warning signs—memory troubles, periods of confusion, changes in mood or energy—can be subtle and hard to pinpoint until the disease progresses to a more advanced stage. The goal of routine screening is to pick up on these signs before it gets to that point.

Gold standard screening tests: MoCA and MMSE

The Montreal Cognitive Assessment (MoCA) and Mini-Mental State Examination (MMSE) are the two standard screening tests used in hospitals everywhere to check for signs of cognitive impairment. They are scored out of 30 points and are intended to take about 10 minutes to administer. (In practice, they can often take much longer, up to 30 minutes.) Take a look at the MMSE, pictured below:

Screen Shot 2023-06-24 at 2.23.09 PM

There are several reasons why this sort of test is very useful. It’s quantitative and standardized—so you can compare numerical scores between different patient groups as well as track an individual patient’s progress over time—and it also differentiates between different cognitive/motor domains (e.g. spatial function, language, memory). The test is easy to score, and hospitals use standardized score cutoffs to determine whether to refer someone for further treatment.


Despite their upsides, tests like the MoCA and MMSE have some very clear weaknesses. Here, we’ll talk about a few important ones.

Lack of Change.

The MoCA and MMSE were developed in 1992 and 1975, respectively. Though there have been some adjustments made since, both tests have remained relatively unchanged over the past decades. While this can be a good thing (it allows for comparisons between tests taken many years apart), the inflexibility can lead to the test becoming outdated in light of new research and clinical knowledge.


These cognitive exams can take the majority of the time in a typical 45-minute appointment with one’s physician. That leaves less time for things like screening for motor issues, conversation, and creating a plan—all which could be just as useful.

‘Summary’ Scoring.

Each cognitive domain (e.g. “memory”) is represented by only one or two test items, which may not truly capture the tester’s underlying ability in that cognitive domain. Additionally, though individual scores per test item are calculated, only the total score typically goes into the medical record (and is used for determining whether further treatment is needed). Combined with natural test-retest variability in overall score, this can make it difficult to identify where exactly declines are occurring.

Lack of Adaptivity.

These tests are non-adaptive, meaning that everyone gets the same test items regardless on how they performed on previous portions of the test. If a particular task is way too easy (or way too hard) for the tester, it won’t be very helpful in assessing whether their true ability has stayed the same or declined over time. In contrast, an adaptive design calibrates the difficulty of each item to the tester’s ability (as estimated by performance on earlier portions of the test), ensuring that the response to each question is actually informative.

A New Approach

In the past few years, there’s been a new wave of digital screening tests that aim to overcome the shortcomings of their paper-and-pencil counterparts. Backed by rigorous data analysis combined with clinical experience, these tools may offer a level of adaptivity and precision leagues beyond that of conventional exams. Instead of arbitrarily assigning whole-number point values to test items, performance scores on each subtest can be calculated using mathematical techniques like item response theory (IRT) and maximum likelihood estimation. Administered on a computer or tablet, they can be made into games—rather than stuffy paper/pencil tests—allowing for an enjoyable experience which also saves time. Below is an example of one such project, the Boston Adaptive Assessment Tools, which I’ve been fortunate to be a part of developing.

Screen Shot 2023-06-24 at 2.24.57 PM

Though these digital screening tools still need to overcome a number of hurdles—technological anxiety, validation and norming, scaling—they represent an exciting new approach that could help improve the diagnosis and treatment of dementia worldwide. Personally, I’m quite excited to see where this field goes in the coming years!

A first-generation US college student, Siva studied Mathematics and Molecular/Cellular Biology at Harvard College, graduating in three years with highest honors in field and Phi Beta Kappa distinction. He is now a MD candidate at Harvard Medical School.


academics study skills MCAT medical school admissions SAT college admissions expository writing English MD/PhD admissions writing LSAT GMAT strategy GRE physics chemistry math biology graduate admissions ACT law school admissions interview prep test anxiety language learning academic advice MBA admissions premed personal statements homework help career advice AP exams creative writing MD study schedules Common Application summer activities test prep history computer science philosophy organic chemistry secondary applications supplements economics PSAT admissions coaching grammar mathematics research law statistics & probability psychology 1L ESL CARS SSAT covid-19 dental admissions legal studies logic games reading comprehension USMLE engineering Spanish calculus parents Latin verbal reasoning DAT PhD admissions case coaching excel mentorship political science AMCAS French Linguistics MBA coursework Tutoring Approaches academic integrity chinese medical school Anki DO English literature Social Advocacy admissions advice algebra astrophysics biochemistry business classics diversity statement genetics geometry kinematics mental health presentations quantitative reasoning skills study abroad time management work and activities IB exams ISEE MD/PhD programs STEM adjusting to college algorithms art history artificial intelligence athletics business skills careers cold emails data science functions gap year international students internships letters of recommendation logic mechanical engineering poetry resume revising science social sciences software engineering tech industry technical interviews trigonometry 2L 3L AAMC Academic Interest DMD EMT FlexMed Fourier Series Greek Health Professional Shortage Area Italian Lagrange multipliers London MD vs PhD MMI Montessori National Health Service Corps Pythagorean Theorem Python Sentence Correction Step 2 TMDSAS Taylor Series Zoom acids and bases amino acids analysis essay architecture argumentative writing art art and design schools art portfolios biomedicine brain teaser campus visits cantonese capacitors capital markets cell biology central limit theorem centrifugal force chemical engineering chess chromatography class participation climate change clinical experience community service constitutional law consulting cover letters curriculum dementia demonstrated interest dental school dimensional analysis distance learning electric engineering electricity and magnetism enrichment escape velocity european history executive function finance first generation student freewriting fun facts genomics graphing harmonics health policy history of medicine history of science hybrid vehicles hydrophobic effect ideal gas law induction infinite information sessions institutional actions integrated reasoning intern investing investment banking lab reports linear algebra linear maps mandarin chinese matrices mba medical physics meiosis microeconomics mitosis music music theory neurology neuroscience office hours operating systems organization pedagogy phrase structure rules plagiarism potential energy pre-dental proofs pseudocode psych/soc qualifying exams quantum mechanics