WELLINGTON – The world is currently in the grip of a diabetes epidemic. A recent major study by Majid Ezzati and colleagues from Imperial College London and Harvard University found that the number of adults with type 2 diabetes increased from an estimated 153 million in 1980 to 347 million in 2008. The number could be 370 million today.
Every region of the world is affected, although the epidemic is growing most rapidly in Oceania and least rapidly in East Asia. Globally, the type 2 diabetes epidemic has been growing in lockstep with rising obesity levels.
This is not surprising – an increase in body fat and a decrease in physical activity are the direct causes of type 2 (as opposed to type 1) diabetes. In fact, much of the health effects of obesity and physical inactivity are mediated through diabetes.
These health effects are serious. Diabetes already is the major cause of kidney failure, blindness, and lower-limb amputation in many countries, and a major cause of heart attacks and strokes.
Despite this, surveillance of diabetes remains relatively undeveloped throughout the world, even in high-income countries. Public-health surveillance is “the ongoing systematic collection, analysis, interpretation, and dissemination of health data for the purpose of preventing and controlling disease” – in short, information for action.
There is nothing in this definition that restricts surveillance to communicable diseases, yet in practice this has generally been the case. The reasons are not hard to find. Communicable disease outbreaks occur over days to weeks (or at the most, months); the danger is “clear and present”; and prevention and control generally requires intervention by the state – the quarantine of victims, tracing and immunization of contacts, or elimination of environmental sources of the infectious agent.
The situation regarding chronic diseases like diabetes is very different. The epidemic happens silently over years or decades; the danger is either not recognized or not considered avoidable; and action is often seen as the responsibility of the individual (lifestyle modification) or health-care system (pharmaceutical prescription), rather than the state.
Yet effective chronic disease surveillance can save lives. If disease trends are monitored, along with patients’ responses to treatment and the population’s exposure to risk factors, the success or failure of policies designed to prevent or control chronic diseases can be evaluated, resource allocation can be rationally prioritized, and the public can be kept fully informed of the risks that they face.
Recognizing this, in December 2005 the New York City Board of Health mandated the laboratory reporting of test results for glycosylated haemoglobin (HbA1c) – a biomarker for diabetes and a key indicator of blood glucose control – thereby creating the world’s first population-based diabetes registry. Mandatory laboratory reporting of HbA1c results (along with basic demographic data) for a defined population (New York City residents) allowed New York’s Department of Health to monitor trends in diabetes prevalence, assess testing coverage, and examine health-care use and glycemic control of residents living with diabetes.
Beyond these population-based surveillance functions, the registry was able to support patient care by ensuring that individual health-care providers and patients were made aware of elevated or rising HbA1c levels. Both the patient-support function and the surveillance function required use of a unique patient identifier, so that letters could be mailed to patients and tests from the same patient could be linked over time.
In 2009, Thomas Frieden and colleagues from the New York City Board of Health reviewed the registry’s first four years of operation and concluded that it was performing well. Getting all laboratories to report regularly and completely, however, proved challenging and not all health-care providers and patients proved willing to participate.
The Board of Health’s initiative has been widely praised as exemplifying the application of classical communicable-disease surveillance-and-control tools to a chronic disease. Other commentators, however, have criticized the registry for potentially compromising patient confidentiality and privacy, and even for disrupting the relationship between patients and their doctors.
While these criticisms may or may not be justified, it is probably true to say that the New York City diabetes registry, though highly innovative, is at best an interim solution. Rather than relying on laboratory reporting of a single biomarker, an ideal chronic-disease surveillance system would extract all necessary data directly from the patient record.
Any diagnosis of diabetes, or subsequent monitoring of disease progression, requires a medical consultation and hence an entry into the patient record – and so into the practice’s patient-management information system. Logically, the surveillance system should operate by extracting the entire subset of data required for surveillance purposes from each health-care provider’s patient management information system (“front-end capture”).
This data would then be securely transferred (electronically) to a suitable data warehouse. After appropriate cleaning (checking for missing data, correcting coding errors), and anonymizing, the data would be available for access and querying. Given appropriate statistical analysis and careful interpretation, useful reports could be generated for surveillance purposes and, if desired, for patient-care support as well (using encrypted unique patient identifiers to preserve confidentiality of personal information).
In view of the rising burden of diabetes and other chronic diseases throughout the world, urgent attention must be devoted to strengthening surveillance systems for noncommunicable diseases at all levels – from local practices to global institutions.