Diagnosis of Diabetes


CONSIDER a diagnosis of diabetes in a patient with:

  • thirst and polyuria
  • unexplained weight loss or tiredness
  • pruritus vulvae, balanitis or recurrent 'UTI's'
  • recurrent infections
  • blurring of vision (usually an osmotic effect and not permanent)
  • discoloured or ulcerated feet
  • hypertension, ischaemic heart disease or stroke
  • obesity, with diagnosis of arterial disease or family history of diabetes.

In such patients, it is useful to perform preliminary screening investigations i.e. random plasma glucose measurement and urinalysis for presence of glucose and ketones. The diagnosis of diabetes has important medical and legal implications for the patient.

Therefore a diagnosis of diabetes should not be based solely on finding:

  • glycosuria
  • raised capillary blood glucose (finger prick sample)
  • elevated haemoglobin A1c (HbA1c) result.

The World Health Organization published revised guidelines on the diagnosis of diabetes, which were accepted for use by Diabetes UK from 1st June 2000.


Algorithm for Diagnosis of Diabetes


 1.    Classical symptoms (e.g. polyuria, polydipsia, unexplained weight loss) plus one of the following:

  • random plasma venous glucose concentration ≥11.1 mmol/L or
  • fasting venous plasma glucose concentration  ≥7.0 mmol/L or
  • venous plasma glucose concentration  ≥11.1 mmol/L (2 hour sample in OGTT)

 2.    No symptoms i.e. incidental finding of glycosuria or hyperglycaemia

  • Diagnosis should not be based on a single venous plasma glucose measurement
  • Additional testing on another day with a value in the diabetic range is essential (using either fasting, random or samples taken 2 hours following glucose load)
  • If fasting or random values are not diagnostic, the 2-hour value should be used

3.    If ketonuria is present with:

  • Severe symptoms i.e. vomiting and dehydration, urgent hospital admission is required.
  • Milder symptoms and weight loss discuss patient urgently with the diabetes team for consideration of insulin therapy.

For Urgent Hospital Contact Numbers view Nearest Specialist Clinic


Diagnostic Criteria not met?


If fasting glucose is 6.1 – 6.9 mmol/L, consider performing an Oral Glucose Tolerance test (see later)

 Oral Glucose Tolerance Test

Back to the top Back to the top of this page


The Role of the Oral Glucose Tolerance Test (OGTT)


An OGTT need only be considered to establish a diagnosis of diabetes if blood glucose values fall into an equivocal range (e.g. FPG >6.0 but an OGTT is not necessary if the diagnostic criteria for diabetes are present.

Perform OGTT after at least 3 days of unrestricted diet (> 150g CHO daily)

Fast patient overnight (8-14 hours, water allowed) and rest during the test.

Samples at times other than 0 and 2 hours are not necessary for diagnosis.

Diagnostic interpretation of OGTT is different in pregnancy, see Handbook section on pregnancy

The Role Of The Oral Glucose Tolerance Test (Ogtt)

Notes on Oral Glucose Preparations (OGTT)

  • For oral glucose tolerance testing, the standard dose for an adult is one pouch of Rapilose OGTT Solution (300ml/75g anhydrous glucose).  However, it can be easily adjusted to paediatric applications based on body mass.
  • The dose for children that weigh less than 43kg is 7ml (1.75g anhydrous glucose) per kg of body weight.  The total children's dose should not exceed 75g.
  • Each 300ml pouch of Rapilose OGTT Solution contains exactly 75g anhydrous glucose, which is the adult dose recommended by the World Health Organisation in oral glucose tolerance testing.
  • Rapilose OGTT Solution is produced in a ready to drink format in a 300ml aluminium foil pouch with a tamper evident twist off cap.
  • Rapilose OGTT Solution is a pleasant tasting orange flavoured drink that is non-carbonated and contains no colour additives.  It is also gluten, lactose, fat, caffeine and alcohol free.
  • Polycal® liquid (neutral or orange) is used in hospital setting.  113mL of Polycal® should be mixed with water to administer a total volume of 200 to 300 mL. (considerably cheaper than Rapilose® OGTT).

*IFG & IGT have an increased risk of future diabetes

  • Advise on healthy eating, regular exercise and avoidance of obesity
  • Check FPG annually
  • Treat co-existing coronary risk factors aggressively, as are at increased risk of developing cardiovascular disease.

Back to the top Back to the top of this page


Diagnostic Tests for Diabetes - NHS Tayside Guidance February 2013


Colleagues may be aware of the debate about the potential use of HbA1c as a diagnostic test for Type 2 Diabetes. This has been featured in the medical literature, February 2013 edition of the Journal of the Royal College of General Practitioners.  While there are some potential benefits, in particular in relation to convenience, there are also considerable risks and challenges with any such change away from the current standard diagnostic tests of serum glucose measurement +/- glucose tolerance testing (as per the algorithm in the Tayside Diabetes Handbook).


There are significant numbers of situations where use of HbA1c as a diagnostic test is clinically inappropriate, leading to the risk of wrong diagnosis. All clinicians will need to carefully consider this list of ‘exclusions’ and ‘cautions’ whenever they wish to perform a diagnostic test. This is a clinical governance issue.


The cost to the laboratories of an HbA1c measurement is up to 25 times greater than that of a serum glucose measurement. A change to the use of HbA1c as the diagnostic test of choice will therefore have major financial implications for Lab services.


Work is ongoing at both local and National levels to work through these risks and issues.


In the meantime, we would ask that colleagues continue to use blood glucose measurement +/- glucose tolerance testing as the routine diagnostic tests for Type 2 Diabetes.


Dr Alistair Emslie-Smith, Lead Clinician, NHS Tayside Diabetes MCN.

Dr Bill Bartlett, Joint Clinical Director Diagnostics, NHS Tayside.


Patients Presenting at Diagnosis with Osmotic Symptoms


If patient has presented with osmotic symptoms (polyuria/polydipsia or weight loss) please follow specific guidance in Management of Newly Diagnosed Patients with Osmotic Symptoms.


Diabetes ‘Resolved’ or ‘In Remission’


It has become apparent that the term ‘diabetes resolved’ is being used both within SCI-Diabetes and in Primary Care IT systems to code patients after pancreatic or islet cell transplants, or after successful bariatric surgery or other intensive weight reduction, where these treatments have begun to normalise hyperglycaemia. The ‘diabetes resolved’ code is also being used to ‘correct’ administrative errors, where a patient has been wrongly coded as having diabetes, and diagnostic errors, where a patient has been wrongly diagnosed as having diabetes and in patients with Gestational Diabetes whose post-partum glucose profile has returned to normal.


All such patients, with a ‘diabetes resolved’ code on SCI-Diabetes, are AUTOMATICALLY REMOVED from the DRS retinal screening register. This is having the detrimental effect that some patients who DO require retinal screening fail to receive appropriate recall.


Abnormal glucose metabolism leading to hyperglycaemia defines the disease of diabetes mellitus, but hyperglycaemia exists on a continuum. The levels of fasting glucose (at or above 7mmol/l) and HbA1c (at or above 48 mmol/mol or 6.5%) that are used to define the diagnosis of diabetes are chosen because they are the levels associated with the diabetes specific complication of diabetic retinopathy. Levels of glucose below these diagnostic values but above “normal” levels (i.e. fasting glucose between 6-6.9mmol/l and HbA1c of 42-48 mmol/mol or 6-6.4%) have been defined to indicate people who are at increased risk of developing type 2 diabetes. Levels of fasting glucose below 6mmol/l and HbA1c levels below 42 mmol/mol (6%) are defined as within the normal range. Successful management of diabetes with lifestyle and/or medication or transplant or bariatric surgery may result in glucose levels below those diagnostic of diabetes but, hitherto, it has not been clarified as to whether this should be termed good diabetes control, remission, resolution or cure?


Published Evidence

Very little scientific or actuarial evidence is available to guide this debate.   A report from a group of diabetes experts, convened under the auspices of the American Diabetes Association, (ADA) was published in 2009 (Buse et al 2009) Consensus was difficult to achieve in some areas and the published recommendations are not the official position of the ADA.


A report on the results from diabetic retinopathy screening in 119 people with diabetes after bariatric surgery showed that 18% developed new diabetic retinopathy or worsened pre- existing retinopathy. The authors conclude that the rapid improvement in glycaemic control following bariatric surgery is well recognised but the potential for worsening of diabetic retinopathy is less commonly perceived. It is vital therefore to continue retinal screening in people with diabetes after bariatric surgery. (Varadharn et al 2012)


A very recent publication on the association of an intensive lifestyle intervention with remission of type 2 diabetes (Gregg et al 2012) uses the term remission. It suggests that an intensive lifestyle intervention may be associated with a partial diabetes remission in a subset of patients with type 2 diabetes, particularly those whose diabetes is of short duration, who have lower HbA1c levels at entry and who have substantial weight loss or fitness change.


NHS Scotland Guidance

The expert consensus in NHS Scotland is that:

  • All patients who have a previous diagnosis of diabetes (with the exception of Gestational Diabetes – see below) should continue to receive screening for diabetic retinopathy.
  • The term ‘diabetes resolved’ (and Read Code 212H.00) should NOT be used or coded in SCI-Diabetes or Primary Care IT systems.
  • Use of the term ‘diabetes in remission’ in SCI-Diabetes will ensure that the person remains on the DRS recall Register. (Read Code C10P)


We recommend that these definitions and recommendations be used in NHS Scotland

  1. To correct administrative or diagnostic errors

Where a patient has been wrongly coded as having diabetes or wrongly diagnosed as having diabetes, the diagnosis on SCI-Diabetes should be changed using the Diagnostic Admin Form to ‘NOT DIABETIC’.


  1. People with diabetes and an HbA1c below 48 mmol/mol (6.5%)

If they are on any medication or are still having on-going procedures after surgery they are just “well controlled”. “Diabetes resolved” should not be used. All annual complication and risk factor screening should continue.


  1. People with diabetes and an HbA1c below 42 mmol/mol (6.5%) who have had bariatric surgery or pancreas or islet cell transplant, or very significant weight loss from dieting and are on no glucose lowering therapy or continuing surgical procedures

If glucose levels have remained below diagnostic criteria for diabetes for 1 year or more they may be considered as being in remission, but need continuing annual screening for risk factors and complications so “diabetes resolved” should not be used .


There is as yet no evidence to suggest if and when annual complication screening for people in this group could be altered to every 2 years or could cease. In view of this we recommend that until such evidence appears that annual complications screening should continue indefinitely.



There is evidence that the term “diabetes resolved” is being used inappropriately by clinicians.

This can result in people being no longer being recalled for diabetes risk factor or complication screening although they are at risk of such complications.


We recommend that where appropriate the term “diabetes in remission” be used and that “diabetes resolved” is not used.


This NHS Scotland Guidance is in line with recent Guidance published by NHS Diabetes in England.



Buse JB, Caprio S, Cefalu WT et al How should we define cure in diabetes Diabetes Care 2009 32: 2133-2135

Gregg EW, Chen H, Wagenknecht Le et al Association of an Intensive Lifestyle Intervention with Remission of Type 2 Diabetes JAMA 2012 308: 2489-2496

Varadhan L, Humphreys T, Walker AB et al Bariatric surgery and diabetic retinopathy: a pilot analysis ObesSurg 2012 22:515-516

Coding, Classification and Diagnosis of Diabetes 2011


Coding of Past Medical History of Drug-Induced Diabetes


Some patients in Tayside have had drug-induced diabetes secondary to steroids or other agents and then demonstrate glucose and HbA1cs in the non-diabetic range once the drug has been withdrawn.

It is known that all such patients remain at ‘high risk’ of subsequently developing Type 2 Diabetes and so they need to remain on a Practice recall register for annual glucose screening.


There is no definitive evidence to guide us on which of these patients require ongoing retinal screening. We recommend the following pragmatic approach:

  1. Any patient found to have any degree of retinopathy during the time that they had confirmed diabetes and who has subsequently reverted to sustained biochemical normoglycaemia after the withdrawal of the causative drug and any hypoglycaemic therapy, should remain on the DRS Retinal Screening Register. To ensure this, they should be coded on SCI-Diabetes and in the Practice System as ‘Diabetes in Remission’ (Read Code C10P.).


  1. Any patient known to have had drug-induced diabetes for a total of at least three years, who during that time was regularly screened and not found to have any degree of retinopathy and who has subsequently reverted to sustained biochemical normoglycaemia after the withdrawal of the causative drug and any hypoglycaemic therapy, should also remain on the DRS Retinal Screening Register. To ensure this, they should be coded on SCI-Diabetes and in the Practice System as ‘Diabetes in Remission’ (Read Code C10P.).


  1. Any patient known to have had drug-induced diabetes for a total of less than three years, who during that time was regularly screened and not found to have any degree of retinopathy and who has subsequently reverted to sustained biochemical normoglycaemia after the withdrawal of the causative drug and any hypoglycaemic therapy does not need to remain on the DRS Retinal Screening Register. They should be coded on SCI-Diabetes and in the Practice System as ‘Stress Induced Hyperglycaemia’ (Read Code R10D1).


  1. As the Read Code C10P (‘Diabetes in Remission’) is a diabetes code, these patients should also be ‘exception reported’ from the QOF Diabetes indicators on the grounds of 'not appropriate for review'.