The CDR scale scores dementia severity on a five-point system: 0 means no impairment, 0.5 indicates very mild or questionable dementia, 1 represents mild dementia, 2 means moderate dementia, and 3 signals severe dementia. Each score corresponds to specific losses in memory, orientation, judgment, and daily functioning, and the shift from one score to the next marks a meaningful decline in a person’s ability to live independently. For example, a person scored at CDR 0.5 might consistently forget where they put their keys and partially recall recent conversations, but they still manage their own hygiene and finances. A person at CDR 1, by contrast, has crossed a threshold where memory loss actively interferes with everyday life and independent functioning begins to erode. Developed in 1982 at Washington University School of Medicine by Charles P.
Hughes, Leonard Berg, and colleagues, the Clinical Dementia Rating scale has become the standard instrument for staging Alzheimer’s disease in research settings worldwide. It was later refined by John C. Morris and revised in 1993 to sharpen the distinctions between severity levels. Beyond the global score, a companion metric called CDR-SB (Sum of Boxes) provides a more granular, continuous measure of decline and has become the primary endpoint in major Alzheimer’s drug trials. This article breaks down what each CDR score actually looks like in practice, explains how CDR-SB scoring works, and covers how these numbers are shaping the latest treatment research.
Table of Contents
- What Does Each CDR Score Mean for Alzheimer’s Patients and Their Families?
- CDR 2 and CDR 3 — What Moderate and Severe Scores Look Like in Daily Life
- CDR-SB Scoring — A More Sensitive Measure of Alzheimer’s Progression
- How CDR-SB Is Used in Alzheimer’s Drug Trials — and What the Results Mean
- Who Administers the CDR and Why It Cannot Be Self-Scored
- CDR 0.5 — The Most Debated Score on the Scale
- The Future of Dementia Staging Beyond the CDR
- Conclusion
- Frequently Asked Questions
What Does Each CDR Score Mean for Alzheimer’s Patients and Their Families?
The CDR evaluates cognitive and functional performance across six domains: Memory, Orientation, Judgment and Problem Solving, Community Affairs, Home and Hobbies, and Personal Care. Each domain is scored individually, and memory is treated as the primary category. If at least three of the five secondary categories match the memory score, the global CDR equals the memory score. The first five domains use a five-point ordinal scale (0, 0.5, 1, 2, 3), while Personal Care uses a four-point scale with no 0.5 option in traditional scoring. This structure means the global CDR is not a simple average. It is a weighted judgment that privileges memory decline as the central marker of disease progression. At CDR 0, a person is cognitively normal with no memory complaints and full independence in personal care, social functions, and daily activities. At CDR 0.5, the picture is more nuanced and often the most difficult stage for families to interpret. A person at this level shows consistent slight forgetfulness with partial recollection of events, sometimes described as “benign” forgetfulness.
Their intellectual interests and home activities may have slipped slightly, but social and work performance remain largely intact. Personal care is not affected. The challenge here is that CDR 0.5 can describe someone in the earliest stages of Alzheimer’s or someone experiencing normal age-related cognitive changes. The distinction often requires longitudinal monitoring rather than a single assessment. CDR 1, mild dementia, is where the clinical picture sharpens. Memory loss becomes moderate and is more pronounced for recent events, directly interfering with everyday activities. A person at this stage may forget appointments, repeat questions within the same conversation, or lose track of recent news they read that morning. The critical marker in the CDR framework is that a shift from 0.5 to 1 in any domain represents a loss of independence in that domain. This is often the point where families notice that a parent who once managed household finances or cooked complex meals can no longer do so without help.
CDR 2 and CDR 3 — What Moderate and Severe Scores Look Like in Daily Life
CDR 2, moderate dementia, describes a person with severe memory loss where only highly learned material is retained. Orientation is impaired, meaning the person may not reliably know the date, the season, or sometimes where they are. Judgment and problem-solving abilities are significantly compromised. There is no independent function outside the home, and the person requires assistance with personal care tasks such as dressing and hygiene. At this stage, a person might still recognize close family members and recall childhood memories or deeply ingrained skills, but the ability to form new memories or navigate unfamiliar situations has largely disappeared. CDR 3, severe dementia, represents the most advanced stage on the scale. Memory loss is extreme, and disorientation is pervasive. The person cannot function autonomously in any capacity and must be closely monitored or supervised at all times.
Full assistance is required for all personal care. At this stage, communication is often fragmentary, and the person may not recognize family members consistently. However, it is worth noting a limitation of the CDR at this level: the scale’s ceiling at 3 means it does not differentiate between degrees of severity within advanced dementia. A person who can still speak in short phrases and one who is entirely nonverbal both receive the same global score. For caregivers managing day-to-day care, this lack of granularity can feel inadequate, and supplemental assessments are often used alongside the CDR in late-stage care planning. The progression from CDR 2 to CDR 3 is not always linear or predictable. Some individuals plateau at a moderate stage for months or even years, while others decline rapidly. The CDR captures a snapshot of current functioning, not a trajectory, which is why repeated assessments over time are essential for understanding an individual’s disease course.
CDR-SB Scoring — A More Sensitive Measure of Alzheimer’s Progression
While the global CDR score places a person into one of five broad categories, the CDR Sum of Boxes (CDR-SB) provides a finer-grained picture by summing all six individual domain scores. The resulting number ranges from 0 to 18, with higher values indicating greater impairment. This continuous scoring system picks up smaller increments of change that the global CDR misses, which is why it has become the preferred measure in clinical research. Staging cutoffs for the CDR-SB, established by a Texas Alzheimer’s Research Consortium study published in JAMA Neurology, map onto the global CDR categories as follows: a CDR-SB of 0 corresponds to normal cognition (CDR 0), scores of 0.5 to 4.0 correspond to very mild impairment (CDR 0.5), scores of 4.5 to 9.0 indicate mild dementia (CDR 1), scores of 9.5 to 15.5 correspond to moderate dementia (CDR 2), and scores of 16.0 to 18.0 indicate severe dementia (CDR 3). To put this in concrete terms, consider a person who scores 0.5 in Memory, 0 in Orientation, 0.5 in Judgment, 0.5 in Community Affairs, 0 in Home and Hobbies, and 0 in Personal Care. Their CDR-SB would be 1.5, placing them in the very mild range.
Six months later, if their Memory rises to 1 and Home and Hobbies rises to 0.5, their CDR-SB would climb to 2.5. The global CDR might still read 0.5, but the CDR-SB has captured a meaningful worsening that the broader score obscured. This sensitivity makes CDR-SB especially valuable for tracking early-stage decline, when changes are subtle and families and clinicians need to know whether the disease is progressing or stable. However, the CDR-SB is not without limitations. Because it sums across domains, two people with the same CDR-SB score can have very different functional profiles. A score of 6 might reflect moderate impairment across all domains or severe impairment in memory and judgment with intact personal care. Context always matters when interpreting the number.
How CDR-SB Is Used in Alzheimer’s Drug Trials — and What the Results Mean
CDR-SB has become the primary endpoint in major Alzheimer’s disease-modifying drug trials, largely because its continuous scale can detect treatment effects that the global CDR cannot. The two most prominent recent examples are lecanemab (marketed as Leqembi) and donanemab (marketed as Kisunla), both amyloid-targeting antibodies. In the CLARITY AD Phase 3 trial, which enrolled 1,795 participants with early Alzheimer’s disease, the adjusted mean change in CDR-SB at 18 months was 1.21 points with lecanemab compared to 1.66 points with placebo. That difference of negative 0.45 points represents a 27 percent slowing of decline. In the TRAILBLAZER-ALZ 2 trial, donanemab slowed CDR-SB progression by negative 1.2 points versus a weighted control at three years in early-start participants. These numbers are small in absolute terms, which has fueled ongoing debate about clinical meaningfulness.
A 2025 analysis attempted to translate these statistics into practical impact: for a patient with a baseline CDR-SB of 2, lecanemab could extend independence in instrumental activities of daily living by approximately 10 additional months, while donanemab in patients with low to medium tau burden could extend independence by approximately 13 months. The tradeoff that families and clinicians must weigh is real. These treatments carry risks, including brain swelling and microbleeds, and they require regular infusions and MRI monitoring. A 27 percent slowing of decline is not a cure or a reversal, and for some patients the risks may outweigh the benefits. But for others, particularly those in the earliest stages with the most independence left to preserve, those additional months of functional autonomy carry enormous personal value. The CDR-SB is the yardstick by which these decisions are increasingly measured.
Who Administers the CDR and Why It Cannot Be Self-Scored
The CDR requires a trained clinician to administer through semistructured interviews with both the patient and a reliable informant, typically a family member or primary caregiver. This dual-interview design is deliberate. People in the early stages of dementia often lack awareness of their own deficits, a phenomenon called anosognosia. Without corroboration from someone who observes the person daily, the assessment would systematically underestimate impairment. Conversely, some patients are acutely aware of and distressed by their cognitive lapses, and their self-reports may overstate the functional impact. The informant interview helps calibrate both directions. This requirement means the CDR cannot be reliably administered at home through a questionnaire or online tool.
Families who encounter CDR scoring rubrics online and attempt to self-assess a loved one should understand that their results will not carry clinical weight and may be inaccurate. The scoring algorithm itself involves judgment calls about which domain scores should drive the global rating, and these calls require training in the instrument’s conventions. The CDR is primarily used in research settings internationally to stage dementia severity and track progression, and its adoption in routine clinical care remains uneven. Many community neurologists and primary care physicians use simpler screening tools like the MMSE or MoCA for initial assessment and reserve the CDR for formal research protocols or specialist referrals. A practical warning: if a clinician or research team provides a CDR score, always ask whether they used the standard CDR global score or the CDR-SB, as the numbers mean very different things. A report that says “CDR of 4” is a CDR-SB score, not a global CDR score, since the global scale only goes to 3. Misinterpreting one for the other can cause unnecessary alarm or false reassurance.
CDR 0.5 — The Most Debated Score on the Scale
CDR 0.5 occupies an uncomfortable middle ground. It was originally labeled “questionable dementia,” and that label captures the genuine clinical uncertainty at this stage. A person at CDR 0.5 may be experiencing the earliest manifestations of Alzheimer’s disease, or they may be showing cognitive changes that will never progress to dementia. Research has shown that a substantial proportion of individuals rated CDR 0.5 do go on to develop dementia within five years, but others remain stable or even revert to CDR 0 at subsequent assessments.
For families, hearing that a loved one is at CDR 0.5 can feel like being told something is wrong without being told what to do about it. What makes CDR 0.5 clinically important is that it has become the entry criterion for most Alzheimer’s prevention and early treatment trials. The lecanemab and donanemab trials enrolled participants at CDR 0.5 and CDR 1, meaning this is precisely the population for whom new treatments are being developed and tested. If a person receives a CDR 0.5 rating, it does not mean dementia is inevitable, but it does mean they may be eligible for interventions that are most effective when started early. This is a case where the score’s ambiguity is actually a call to action rather than a reason to wait and see.
The Future of Dementia Staging Beyond the CDR
The CDR has been the gold standard for over four decades, but the field is evolving. Blood-based biomarkers for amyloid and tau, which were experimental just a few years ago, are now entering clinical practice and may eventually allow dementia staging based on biological progression rather than observed symptoms alone. Digital cognitive assessments, wearable devices, and remote monitoring technologies are also being explored as ways to track decline continuously rather than through periodic in-office evaluations.
None of these approaches are likely to replace the CDR entirely in the near term. The strength of the CDR lies in its holistic assessment of how disease affects a person’s actual life, not just their biology or their performance on a timed test. But as Alzheimer’s treatment moves toward earlier intervention, possibly before symptoms are obvious to anyone, staging tools will need to evolve accordingly. The CDR-SB’s sensitivity to small changes has already proven its value in this direction, and future refinements may extend the scale’s resolution at both the mildest and most severe ends of the spectrum, where its current precision is weakest.
Conclusion
The CDR scale translates the complexity of Alzheimer’s disease into a structured framework that clinicians, researchers, and families can use to understand where a person stands and what to expect. Each score, from 0 through 3, corresponds to specific losses across memory, orientation, judgment, daily functioning, and personal care, with the CDR-SB providing a more sensitive continuous measure for tracking change over time. These scores are not abstract numbers.
They describe whether a person can manage their own finances, remember recent conversations, navigate their neighborhood, or dress without help. For families navigating a diagnosis, the most important step after receiving a CDR score is to discuss it thoroughly with the assessing clinician. Ask what the score means for your loved one’s current level of independence, what changes to watch for that might signal progression, and whether the score makes them eligible for clinical trials or new treatments. The CDR is a tool, not a verdict, and understanding it clearly is one of the most practical things a caregiver can do.
Frequently Asked Questions
What is the CDR scale and who developed it?
The Clinical Dementia Rating scale was developed in 1982 at Washington University School of Medicine by Charles P. Hughes, Leonard Berg, and colleagues. It was refined by John C. Morris and revised in 1993. It assesses cognitive and functional performance across six domains and produces a global score from 0 (no impairment) to 3 (severe dementia).
What is the difference between the CDR global score and CDR-SB?
The CDR global score places a person into one of five categories (0, 0.5, 1, 2, or 3). The CDR-SB sums all six individual domain scores for a continuous score ranging from 0 to 18, providing a more sensitive measure of change over time. The CDR-SB is the primary outcome measure in most current Alzheimer’s drug trials.
Can I score the CDR at home for a family member?
No. The CDR requires administration by a trained clinician through semistructured interviews with both the patient and a reliable informant. Self-administered versions will not produce valid results and should not be used for clinical decision-making.
Does a CDR score of 0.5 mean my loved one has Alzheimer’s?
Not necessarily. CDR 0.5 indicates very mild or questionable cognitive impairment. Some people at this stage do progress to dementia, but others remain stable or return to normal cognition. However, CDR 0.5 is the stage at which most early-intervention Alzheimer’s treatments are tested, so it may be worth discussing clinical trial eligibility with a specialist.
How much do new Alzheimer’s drugs slow decline as measured by CDR-SB?
In Phase 3 trials, lecanemab slowed CDR-SB decline by 27 percent (a difference of 0.45 points at 18 months), while donanemab slowed progression by 1.2 points versus control at three years. A 2025 analysis estimated that for a patient with a baseline CDR-SB of 2, these treatments could extend independence in daily activities by approximately 10 to 13 additional months.
How often should the CDR be administered?
In research settings, the CDR is typically administered every 6 to 12 months to track progression. The frequency in clinical practice depends on the physician’s judgment and the patient’s stage. More frequent assessments may be appropriate during periods of suspected change or when monitoring treatment response.
