The Good Dialysis Index


Introduction to the Good Dialysis Index

  At a pre-conference session of the Annual Dialysis Conference, Seattle, Washington, 6th March 2010 – ‘Successful Home Haemodialysis’ – I presented a concept that I had been working on for some time … the ‘Good Dialysis Index’ (GDI). This was the first public presentation of the GDI.  I have since been asked to present the same talk at the Asian Pacific Congree Seoul (2010), the Annual Australian Amgen Nephrology Symposium in Melbourne (2011) and the Australian and New Zealand Home Therapies Conference in Sydney (2012) – along with a number of other local or institution-specific presentations of the concept. 

 A GDI is not really a novel or new approach. I am aware that others before me (including Bob Lindsay in the British Medical Journal in the late 1960s!), and many who likely will follow, have played around with dialysis ‘scoring systems’. Perhaps it is a reflection of the dissatisfaction of some – myself clearly included – with the blinkered mathematics of dialysis ‘adequacy’ … Kt/V … that has prompted different thinking on what should and does constitute … good dialysis.

 While the interest shown from many at the conference has been both heartening and gratifying, it remains a ‘work-in-progress’ and may continue to evolve over time. However, I have decided to add it to this site for comment, trial application and consideration by any who may chance upon it here.

 I would welcome feed-back … positive or negative … through the ‘Contact Us’ option at this site.

 While some may wish to bring it to the attention of their dialysis teams, it remains to be validated. However, there has been sufficient interest at the Seattle ADC from dialysis systems and groups with very large patient numbers against whom it can be tested that a more rigorous evaluation may yet prove possible. Meanwhile, it is just an idea, a proposal, an untested hypothesis – and any who might apply it to themselves or their patients must be mindful of that.

 I do not pretend it to be perfect, but it is an attempt to ‘hear the patient speak’ …


Why, then, the need for a GDI?

 I believe we need additional factors, beyond Kt/V and beyond small molecular clearance, to properly evaluate ‘good dialysis’.

        We need to value-add to dialysis while Kt/V stays as  a ‘lowest common denominator’.

         Good dialysis, for each patient, should ensure the best of:

                          Fluid, salt, volume and blood pressure control

                          The prevention or regression of LV hypertrophy and the maintenance of good LV function

                          Simple, balanced iron and anemia management

                          The control and normalization of calcium, phosphate, the Ca++ x PO4 product and PTH

                          An unrestricted, palatable, healthy diet and fluid intake

                          Optimal middle molecule clearance

                          Down-regulation of inflammatory stimuli and the markers of inflammation – albumin and CRP

                          Removal of protein-bound uremic toxins (eg: p-cresol) which, though not currently removed by standard, non-protein-leaky hemodialysis membranes, will become a ‘future factor’


Good patient management must also ensure:

              The abolition of dialysis related symptoms

              Medication simplicity

              And an acknowledgement that patients have

      Lives to live

      Families to enjoy

      Jobs to (hopefully) return to

      A rehabilitation road to tread


Good health-management principles should also aim for:

                  Minimization of patient co-morbidity

                              Reduction in hospitalization

                              Control/reduction in burgeoning health-care expenditure


If we want to achieve ‘Good Dialysis’, we must:

                 1.  Let the patient ‘speak’

                 2.  Be mindful of the patients’ desired outcomes

                 3.  Include a balance of patient-derived information

                 4.  Use empathetic as well as systematic evaluation  


            At the end of the day, we must intend:


      Patient survival … but a ‘good’ survival which is patient-goaled

 and ..

     •      If and when the end is close, sensitively-managed dialysis withdrawal, palliation and as ‘good’ a death as possible


ALL these – and more – are the stuff of good dialysis


     How, then, should we ‘express’ good dialysis?

                                    •      By more mathematical formulae?

… I don’t think so

      If naiveté was good enough for Scribner

… I can accept that

      Maybe some ‘touchy-feely’ stuff isn’t so bad

… Medicine is still ‘an art’


 What follows should underpin the review of EVERY patient at EVERY clinic visit


 The GDI ‘score’ permits the serial monitoring of ‘good dialysis’ in each patient


 Each parameter scores an equal ‘1’

                   ·       There is, intentionally, no ‘weighting’ factor

 ·       This ensures maximum simplicity.

 ·       Though some favor relative ‘weighting’

 And … if all factors matter, then each factor matters, independent of a ‘weighting’


The potential advantages of the Good Dialysis Index include:

             1.  Serial follow-up may allow earlier identification of patient deterioration

             2.  Dialysis patient review is standardized

             3.  With long-term follow-up, there is a potential for:

                             •      Cross-modality benchmarking of ‘modality effect’

                             •      Cross-unit benchmarking of ‘unit effect’




Dialysis Key Performance Indicators (KPI)



If ‘0’, what intervention best addresses this







Patient-Directed Questions





Are you currently feeling ‘pretty good’ about things?               





Are your diet and fluid intake free of major restrictions?





Is your blood access is a native arteriovenous fistula?             





Is your dialysis free of cramp, nausea and headache … etc?                                  





Are you part or full-time employed or retired by choice?              





Have you remained out of hospital for the last 3 months?





 Process-directed Questions





≥7 sessions/each 2 wks … (all inter-dialytic breaks <48hrs)





≥18 total hrs /each wk … (within any frequency regimen)





2 mth Kt/V ≥1.3 or PRU ≥70% (or PRU ≥50% if ≥5 x HD/wk)





Mean inter-dialytic weight gains ≤2.5% dry weight





Mean pre-HD systolic BPs  ≥105  and  ≤150





No intra-dialytic saline has been used in last 4 wks





Laboratory-directed Questions





2 mth Hb ≥110 and ≤125 g/L





2 mth T’Sat ≥20% and ≤45%  ±  Ferritin ≥300 and ≤500 µg/L





2 mth pre-HD Ca x PO4 product  ≤4.0 (if SI) or ≤50 (if US)





2 mth calcium in normal range and PTH  ≥2 to ≤4 x normal





2 mth albumin ≥35 g/L





2 mth CRP is normal for the local laboratory





Chart-directed Questions





No anti-hypertensive medications are necessary





No phosphate binder medications are necessary








16-20   =  good dialysis


10-16   = can significantly improve  dialysis


<10      = poor or unacceptable dialysis



Click here for an easy to print version of the GDI

***Hint: Select "Fit to Page" in the Page Layout section of Printing Preferences***


This sheet is designed to be filled in at each clinic visit and will serve:

1.      As part of the Patient Medical Record

2.      As a serial reflection of patient progress, service performance and interventional identification

It remains un-validated … but ongoing application may help serve as that validation over time




Authored by Prof John Agar. Copyright © 2012
Nocturnal Haemodialysis Program, Barwon Health.
All rights reserved. Revised: July 1st 2012