All the world wants stages

[This expands on the idea of PSP stages from my last post, so you’ll want to read that one first.]

“Does PSP have stages?” is a question frequently posed by patients justifiably concerned about how far along they are in the degenerative course of the illness and what new symptoms might lie ahead. 

Many people have become familiar, however reluctantly, with the widely used TNM staging system for cancer, where T refers to the size and extent of the primary tumor, N the extent of spread to lymph nodes and M the presence or absence of metastasis to other organs.  Each is assigned a value and summed to generate a stage from I to IV.  For some types of cancer, a letter is appended to denote additional detail.  Complicated but useful. 

Parkinson’s disease takes a simpler approach.  The Hoehn-Yahr Scale, published in 1967, has five stages: 1)  symptoms only on one side of the body, 2) symptoms on both sides or in the face, voice or trunk but no balance problem, 3) balance problem that does not require assistance, 4) balance and/or gait problem requiring assistance, 5) confinement to bed or wheelchair most of the time.  Notice that only the laterality and gait/balance are considered here.  Still, the H-Y Scale is very useful and popular.  The paper by Hoehn and Yahr presenting their scale remains far the most widely-cited publication on Parkinson’s disease.  (A little Internet snark, if you’ll excuse me: My former chairman and mentor, Roger Duvoisin, actually did most of the work on the scale before reporting for duty as a medical officer in the Navy, headed for Vietnam, at which point his two senior colleagues wrote up the paper without him.)

PSP is a complicated disease, with dozens of symptoms that can be very roughly lumped into four main areas: parkinsonism (meaning stiffness, slowness and problems with speech and swallowing), loss of mental function (including both cognitive and behavioral issues), impaired eye movement, and balance problems.  In creating a staging system for PSP, one could follow the cancer model, assigning a rating to each disease feature, summing those, and then defining each stage as a specific range of that total.  Or, one could use the Parkinson’s model, relying on just one feature of the disease that’s easy to evaluate and important to the patient’s daily function.

Now let’s consider the purpose of a staging system.  Its main virtue is convenience.  Ideally, it shouldn’t require any imaging or lab tests and should be usable by any clinician.  If patients and caregivers can apply it, that would be a plus. 

A staging system, like any diagnostic test, should have both validity and reliability, and yes, there’s a difference.  There are multiple subtypes of validity that we need not discuss here.  But in general, validity is the degree to which an accurate answer to the question actually measures what it purports to measure.  For example, if I want to know how severe your PSP is and I only ask about your bladder function, the validity for assessing PSP overall would be low.  But if you know your bladder symptoms well and communicate that information to me accurately, the question would have high reliability.   The opposite sort of example is if I try to assess the severity of PSP by measuring the number of neurofibrillary tangles in the brain.  That would be a highly valid way to assess PSP, but the ability of the available imaging techniques or spinal fluid tests to actually do that is not good enough just yet, meaning that their reliability as a measure of PSP is inadequate.

The staging system that my colleagues and I provisionally devised for PSP and used in our prognostic study described in the August 9 post uses an approach similar to cancer’s TNM system.  It uses only information obtainable from the PSP Rating Scale scores.  It considers only swallowing and gait/balance, as those two issues are the most closely related to long-term complications from malnutrition, aspiration, falls and immobility. 

We assessed the validity of the staging system by showing that stage parallels the same patients’ total PSPRS scores almost exactly.  That’s called criterion validity.  But the proposed staging system still needs to be tested for multiple other kinds of validity as well as for reliability.

Just FYI, here’s how to calculate the stage using our proposed system:  First, rate the following four items from the PSP Rating Scale:

3.  Dysphagia for solids by history

0  Normal; no difficulty with full range of food textures

1  Tough foods must be cut up into small pieces

2  Requires soft solid diet

3  Requires pureed or liquid diet

4  Tube feeding required for some or all feeding

13.  Dysphagia for half a glass of water

0 None

1 Single sips, or fluid pools in mouth or pharynx, but no choking/coughing

2 Occasionally coughs to clear fluid; no frank aspiration

3 Frequently coughs to clear fluid; may aspirate slightly; may expectorate frequently rather than swallow secretions

4 Requires artificial measures (oral suctioning, tracheostomy or feeding gastrostomy) to avoid aspiration

26.  Gait without assistance if possible

0  Normal

1  Slightly wide-based or irregular or slight pulsion on turns

2  Must walk slowly or occasionally use walls or helper to avoid falling, especially on turns

3  Must use assistance all or almost all the time

4  Unable to walk, even with walker; may be able to transfer

28.  Sitting down without using hands

0  Normal

1  Slightly stiff or awkward

2  Easily positions self before chair, but descent into chair is uncontrolled

3  Has difficulty finding chair behind him/her and descent is uncontrolled

4  Unable to test because of severe postural instability

Then total the four scores.  Stage 1 is 1-4 points, Stage 2 is 5-8 points, Stage 3 is 9-12 points, Stage 4 is 13-15 and Stage 5 is the full 16.

Reasonably simple, but it takes training and experience to administer the items accurately and there’s a whole list of little rules and tips that I’ve published in my book but didn’t include here.  I’ll continue to test the validity of the system using a larger dataset and I may fool around with other schemes.  I’ll keep you posted.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s