Red X iconGreen tick iconYellow tick icon

Student: Dru Norriss
Supervisors:Phil Hider & Maggie Wilson [Dept Public Health & General Practice and Pegasus Health]
Sponsor: Pegasus Health

Introduction and background

Glucose (or “blood sugar”) is the energy molecule of the body, which is distributed through the blood and taken up by cells to be used to perform work. In diabetes, the body loses its ability to control blood glucose levels correctly, leading to higher than normal levels and eventually complications involving all parts of the body.

In New Zealand, the number of people estimated to have diabetes is predicted to increase from 115,000 to over 160,000 in the next two decades.

About half of those with diabetes are thought to be undiagnosed. A large proportion of these people will have no symptoms, as the most common form of diabetes, type 2 diabetes (~90% of all diabetics), has a gradual and insidious onset.

The predicted large increase in prevalence is concerning because of the many complications that can result from diabetes. Among other things, being diabetic increases the risk of heart attack, stroke, can cause premature blindness, and is the leading cause of non-traumatic amputations as a result of damage to microscopic blood vessels.

Type 2 diabetes is a progressive disease. Treatment initially involves lifestyle changes (such as changes to diet, exercise, and smoking habits), then oral medications, and eventually insulin therapy. All these treatments aim to maintain blood glucose levels close to normal values.

Diabetes is also a good candidate for disease screening. Screening is when a health provider or organization performs tests or investigations on all or part of a population to find evidence of a certain disease before it presents with any obvious signs or symptoms (at which point it would normally be diagnosed). Patients can then be treated earlier on and possibly prevented from developing more severe disease and complications later on. For a disease to be a good candidate for screening it needs to meet certain criteria. Some examples are below:

  • The diseases needs to be an important health problem
  • There should be treatments available for the disease
  • The disease should be able to be easily diagnosed
  • There needs to be a time where the disease present but not active or obvious

Even though diabetes meets all the above conditions, it is still difficult to say that screening for it would be a good idea. This is because there is harm and cost associated with every individual screened, most of who will not have diabetes anyway. There is currently a lot of debate internationally around diabetes screening but forthcoming research hopes to answer some of these questions more fully.

Current New Zealand recommendations are to screen all adult Europeans >50 years, all non-Europeans >40 years, and those in either group 10 years earlier if they are considered to be particularly at-risk of developing diabetes.

In contrast, a group called the United States Preventive Services Task Force (USPSTF), which is in charge of making recommendations on disease prevention for the U.S., released a new document in June 2008 regarding diabetes screening. The USPSTF concluded that there is not enough evidence to say if the benefits of screening for diabetes in adults without symptoms or complications suggesting diabetes outweigh the risks and cost of screening them.

They do however suggest that screening hypertensive adults (patients with high blood pressure) is likely to be beneficial overall, and should occur every 3 years. This is because if a person with high blood pressure is known to have diabetes too, the treatment target for their blood pressure reading should be even lower than the target for people with just high blood pressure and no known diabetes.


The aim of this project was to see how many patients that go to Pegasus Health GPs (“general practitioners” or “family doctors”) have been screened for diabetes, and how many haven't. GPs were then given lists of patients that haven't been screened so that they can consider screening them in the future.


To do this, database queries were designed to pull the information out of GP databases automatically and then to use database and spreadsheet software to work out the numbers screened and not screened in the last 3 years.

In total, 59 GPs from 27 different practices participated. 22 practice databases were access remotely, and 5 practices were visited physically to collect the data.

Adults were considered to have sustained hypertension if they had 3 or more blood pressure readings of greater than 135/80mmHg between 1st November 2005 and 31st October 2008. Patients were considered to have been screened for diabetes if they had any lab inbox reports with the subject “Carbohydrate Master”, “Diabetes Profile” or “Glucose/Glycated Proteins”. These are tests used for diabetes screening that measure blood glucose (“blood sugar”) levels and the effects that those levels are having on haemoglobin, the oxygen-carrying molecule present in blood.


3259 adults were identified as having sustained hypertension. Of these, 2801 were identified as having been screened. This represents 85.95% screened. The GP with the lowest proportion screened had 56.96% patients screened, and the GP with the highest proportion had 100% Patients screened. 458 hypertensive patients who had not had any screening test were identified and their NHI numbers fed-back to their GPs to be considered for screening.

Conclusions and discussion

A screened proportion of 85.95% shows that overall the current practice of the GPs that participated is reasonably close to the USPSTF guidelines. There is considerable variation in proportion of patients with screening tests between GPs, with the lowest proportion being 56.96%.

These results need to be considered carefully though, as there are some limitations to the study design. Some of these include the definition of “consistently hypertensive” not being a standard one, and reliance of GPs manually coding patient hypertensive values and diabetic status within their software. Uncertainties like this can make it difficult to know how accurate results are.

Collection of data remotely proved acceptable and effective, with it being possible for the majority of practices. It saves considerable time over physically visiting practices, and would be an appropriate method for future projects both at Pegasus Health and at any other IPA with the appropriate IT infrastructure.

Back to top