Reading the CDR Sum of Boxes Score

What does the CDR Sum of Boxes actually measure?

It measures six things, then adds them up. A clinician rates memory, orientation, judgment and problem solving, community affairs, home and hobbies, and personal care. Each of those six ratings is a box. The sum of the six is the CDR Sum of Boxes, a total running from 0 to 18, where a higher number means greater impairment 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate..

A box is one of the six rated areas of the Clinical Dementia Rating; the sum of boxes is simply their total.

The parent instrument is older than the total. The clinical dementia rating was introduced in 1982 as a global rating device for a prospective study of mild senile dementia of the Alzheimer type, and it separated older adults across a wide range of cognitive function, from healthy to severely impaired 2Ref 2Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL (1982).A New Clinical Scale for the Staging of Dementia.The origin and purpose of the Clinical Dementia Rating: introduced in 1982 as a clinician-administered global rating device for a prospective study of mild senile dementia of the Alzheimer type, validated as separating older adults across a wide range of cognitive function from healthy to severely impaired, and intended as a staging instrument rather than a screening or diagnostic test.. It is administered by a clinician, and it was built to stage — to describe where someone sits — rather than to screen or to diagnose 2Ref 2Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL (1982).A New Clinical Scale for the Staging of Dementia.The origin and purpose of the Clinical Dementia Rating: introduced in 1982 as a clinician-administered global rating device for a prospective study of mild senile dementia of the Alzheimer type, validated as separating older adults across a wide range of cognitive function from healthy to severely impaired, and intended as a staging instrument rather than a screening or diagnostic test..

The three boxes at the front of that list are cognitive. The three at the back — community affairs, home and hobbies, personal care — are about function: what a person still does, and how much help it takes. Half of the total, in other words, is a description of daily life rather than of recall.

What does a given total mean on the staging scale?

Published cutoffs map the total onto the familiar dementia staging scale. In 1,577 participants — 110 healthy controls, 202 with mild cognitive impairment, and 1,265 with probable Alzheimer disease — analysis on a derivation sample produced ranges that were then confirmed on a separate validation sample 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate.. Those ranges are what most clinics use when they translate a sum into a stage.

Agreement between the stage derived from the sum and the global rating was high, with kappa values of 0.86 to 0.94, and 93.0% of participants fell within these staging categories 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate..

93.0% of the 1,577 participants landed inside the published bands 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate..

One caveat travels with the table. That cohort was overwhelmingly probable Alzheimer disease 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate.. If the diagnosis in your family is vascular dementia, Lewy body dementia, or frontotemporal dementia, the same total may not describe the same picture, and how the clinician reads it is a fair question to raise in any dementia staging assessment.

Why report a sum instead of one stage number?

Because the sum keeps detail that a single stage rounds off. One assessment can produce both numbers, and the 2008 study behind the cutoffs above treated the Sum of Boxes as a staging metric in its own right rather than as a byproduct of the global rating 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate.. The bands themselves show why that matters: global CDR 1.0 covers every total from 4.5 to 9.0 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate..

Two people can both be called stage 1 and still sit four and a half points apart on the sum 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate..

That spread is the practical argument for asking for the sum and not only the stage. Movement inside a band — a box that went from needing prompting to needing hands-on help — can change what a week looks like at home without changing the stage at all. A family planning around the stage alone would never see it happen.

The three boxes that are not about memory

Three of the six boxes rate everyday function rather than cognition: community affairs, home and hobbies, and personal care 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate.. This is the half of the score most families recognise without being told — the volunteer shift that quietly stopped, the workshop untouched for a year, the shirt buttoned into the wrong holes. It is also the half that tends to move first in the paperwork of ordinary life.

Everyday function is measured on its own elsewhere, and the instruments do not convert into one another. The Vineland Adaptive Behavior Scales, for instance, are normed from birth to age 90 and score communication, daily living skills, and socialization on the same mean-100 metric as an IQ score, where higher means better function 3Ref 3Pepperdine CR, McCrimmon AW (2018).Test Review: Vineland Adaptive Behavior Scales, Third Edition (Vineland-3) by Sparrow, S. S., Cicchetti, D. V., & Saulnier, C. A..That everyday adaptive function is measured by its own separate norm-referenced instrument: the Vineland-3 spans birth to age 90 and scores communication, daily living skills and socialization, with domain and composite scores on a mean-100 metric where higher indicates better adaptive functioning.. A number from a scale like that is not a CDR Sum of Boxes, and quoting one in place of the other creates confusion that takes an appointment to undo.

That distinction matters once the number leaves the clinic. If long-term care insurance is in the picture, it is worth asking the insurer which assessment its file actually wants before another appointment is booked to produce the wrong one.

Does a change in the score mean the dementia is progressing?

Not by itself, and there is no published number that settles it. The study that established the staging cutoffs reports neither a minimal important change nor a measurement-error estimate 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate.. No validated threshold exists for how many points of movement count as real decline rather than a harder day, a different person supplying the daily-life detail, or a different clinician doing the rating.

What can be read with more confidence is a crossing. A total that moves from 8.0 to 10.0 has crossed out of one published band and into the next 1Ref 1O'Bryant SE, Waring SC, Cullum CM, et al. (2008).Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study.The CDR Sum of Boxes as a staging metric in its own right: the six box scores (memory, orientation, judgment/problem solving, community affairs, home and hobbies, personal care) that are summed to produce it, the 0-to-18 range and higher-is-worse direction, the derivation-and-validation cohort of 1,577 participants (110 controls, 202 MCI, 1,265 probable Alzheimer disease), the cutoff ranges mapping the sum to global CDR 0.5, 1.0, 2.0 and 3.0, the kappa 0.86-0.94 agreement and the 93.0% classification figure, and the fact that the paper reports no minimal important change and no measurement-error estimate.. A total that moves from 8.0 to 8.5 has not. Both are worth mentioning at the next appointment, but only one of them changes the stage.

A single half-point shift between two visits is not, on its own, evidence of anything.

The more useful question at a follow-up is not what the number is now. It is which boxes moved, and in which direction.

What the score cannot tell you

It cannot diagnose, and it cannot forecast. The Clinical Dementia Rating was built as a staging instrument, a way to place someone along a range running from healthy to severely impaired, and not as a screening test or a diagnostic one 2Ref 2Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL (1982).A New Clinical Scale for the Staging of Dementia.The origin and purpose of the Clinical Dementia Rating: introduced in 1982 as a clinician-administered global rating device for a prospective study of mild senile dementia of the Alzheimer type, validated as separating older adults across a wide range of cognitive function from healthy to severely impaired, and intended as a staging instrument rather than a screening or diagnostic test.. A total of 5.5 does not name the disease causing it, and nothing in the score sets a date for anything that comes next.

It does not decide where someone should live, either. Two people at the same total can need entirely different arrangements depending on who else is in the house, whether there are stairs, whether nights are safe, and how much help is already arriving each week.

The distinction that matters most early on is mild cognitive impairment vs early dementia, and that is a clinical judgement rather than an arithmetic one. If nobody has said plainly which of the two applies, that is the question to bring to the next visit, along with what the dementia diagnosis itself rests on.

Questions worth asking about your own number

Families often ask how doctors determine dementia stage, and the honest answer is that the number is the summary rather than the method. These questions get at the method. Each one is reasonable to ask of any clinician who hands over a report with a total on it and no explanation attached, and none of them require knowing anything technical first.

• Which of the six boxes carried the points, and which were rated at zero?
• Is this a global CDR stage, a Sum of Boxes, or both — and which number is written where in the chart?
• Who supplied the day-to-day information the rating was based on?
• What would you expect to have changed by the next visit, and what would surprise you?
• Where does this number travel from here: the chart, an insurer, a care plan?

That last question is worth asking plainly. A total recorded today can be quoted back months later, in a benefits review or in the admission agreement for a residential setting, by someone who never met the person it describes.