A person with diabetes has problems with their body responding to insulin or making the hormone. Insulin is produced by the pancreas and is needed by body cells to take up sugars (glucose) from the blood. The glucose is an important source of energy for the body.

Type 2 diabetes is the most common type of diabetes. The risk of developing the disease increases with age, being overweight and lack of physical activity. People with diabetes are often at risk of cardiovascular disease because of high blood pressure (hypertension), high blood cholesterol (LDL cholesterol) and increased blood clotting. They are more likely to have heart poor blood circulation (including peripheral vascular disease) that in the long term can damage the eyes, kidneys, feet and cause heart disease and a stroke.

A reduction in blood glucose levels and of blood pressure is likely to reduce the risk of diabetes complications. For some people medication with oral hypoglycaemic agents or insulin is unavoidable for controlling their blood glucose levels.

Self-management skills are needed to enable people to manage any medication and improve their health long term, as changes in diet and lifestyle are a vital part of their health care. Providing people with effective ongoing education and emotional and psychological support helps them to have the knowledge, skills, attitudes and motivation to effectively manage the diabetes.

A large study in the UK showed that people with diabetes were less likely to have eye, kidney and foot (microvascular) complications if they rigidly controlled their blood glucose levels. Tightly controlling their blood pressure as well also reduced complications. Long-term management of diabetes and the associated cardiovascular risk factors is, therefore, very important. Your doctor uses glycated haemoglobin (HbA1c) measurements to follow blood sugar levels over a two to three month period. This is possible because glucose binds to haemoglobin, the oxygen-carrying protein in red blood cells. If blood glucose levels stay high the haemoglobin is changed over the life of the red blood cells (red blood cells are constantly being replaced, half of them every two months).

We present some of the evidence from Cochrane systematic reviews about complementary and alternative treatments related to people with diabetes. This evidence comes from carefully researched reviews of information about clinical trials done to evaluate medical treatments. Studies are only included in these reviews if they meet pre-defined criteria.

There are Cochrane reviews available for four possible therapies.  


The risk of developing type 2 diabetes increases with being overweight, lack of physical activity and as ages. The majority of people who develop type 2 diabetes are insulin resistant. They develop high blood sugars (hyperglycaemia) when their bodies can no longer support the compensatory high insulin levels needed to control blood glucose.

People with insulin resistance may also have atherosclerosis, high triglycerides and irregularities in fat (lipid) metabolism (dyslipidaemia), glucose intolerance, high uric acid levels (hyperuricaemia), high blood pressure and polycystic ovary syndrome.

Insulin resistance is measured by impaired glucose tolerance and increased fasting glucose levels are associated with the insulin resistance. It is an intermediate metabolic stage in the development of diabetes. This stage includes individuals with fasting glucose levels equal or greater than 110 mg/dl but lower than 126 mg/dl or after an oral glucose tolerance test with 2 hour values equal or greater than 140 mg/dl but lower than 200 mg/dl.

What is known

The cells in the body need zinc both for structure and for producing energy. It is an essential trigger for many biochemical reactions and for protein production and is a part of more than 200 enzymes. Zinc has a role in the synthesis and action of insulin. The human body does not produce zinc on its own, so it must be obtained from outside sources. The mineral zinc can be found in both animal and plant food sources, but the richest source of zinc comes from animal food sources.

What the synthesised research says

There is currently no evidence to support people with insulin resistance taking a zinc supplement for the prevention of type 2 diabetes mellitus.

How it was tested

Only one study from Brazil met the pre-defined inclusion criteria of this review. Fifty-six non-diabetic adults with normal glucose tolerant obese women (aged 25 to 45 years, body mass index 36.2 ± 2.3 kg/m2) were sequentially assigned to zinc supplement (30 mg per day) or non-active placebo for four weeks. Plasma insulin concentrations, as a measure of insulin resistance, were the same with both interventions.

Measures of body mass, size, skinfold thickness and constitution; triglycerides, cholesterol and lipoprotein cholesterol; energy and nutrient intake (proteins, carbohydrates, lipids and zinc) were not clearly different for women receiving zinc supplementation compared to placebo.

Side effects and general limitations

Four weeks were not sufficient to assess a long-term process like the development of glucose intolerance and diabetes.

The daily recommended oral dose of zinc is 12 mg for women and 15 mg for men (of zinc sulphate from 30 to 200 mg) per day. Amounts of two gram per day or more can cause gastrointestinal irritation and vomiting.


Beletate V, El Dib RP, Atallah AN. Zinc supplementation for the prevention of type 2 diabetes mellitus. Cochrane Database of Systematic Reviews 2007, Issue 1. Art. No.: CD005525. DOI: 10.1002/14651858.CD005525.pub2.


People with diabetes are asked to control their diet by eating less saturated fat and refined carbohydrates, and to limit alcohol intake. Drugs may be used to help increase the release or effectiveness of insulin, slow digestion of carbohydrates, or to change the fats in the blood.

What is known

Studies of world populations show that fats and oils from the sea may be beneficial to health. Animal fat is rich in saturated fatty acids. Vegetable and marine oils are rich in polyunsaturated fatty acids. Supplements of fish oils in the diet may, therefore, have beneficial effects in type 2 diabetes.

What the synthesised research says

Taking fish oil lowered blood plasma triglyceride levels. This was most evident in people with high levels of triglycerides at the beginning of a trial. High doses of fish oil and fish oil taken for longer than two months also led to greater reductions in triglyceride levels. 
Overall, there was no change in plasma cholesterol concentration or any increase in HDL concentration. A slight increase in LDL cholesterol was seen under these conditions but this can occur with drugs that lower triglycerides.
Fasting glucose or sugar bound to the hemoglobin of red blood cells was not clearly changed (increased). This would have indicated loss of ‘glycemic control’. No other potential adverse effects were reported.

How it was tested

The researchers made a thorough search of the medical literature and found 18 randomised trials that randomly assigned people with type-2 diabetes to adding fish oil (3 to 18 grams per day) or adding a salt solution or vegetable oil to their diet. A total of 823 participants were involved with the majority being male, aged between 55 and 65 years and with established diabetes. Few had complications of diabetes. The trials lasted four to eight weeks or three to six months, average 12 weeks, which is short duration for a chronic condition.

Side effects and general cautions

No controlled trials looked at long-term effects of taking fish oil on heart attacks, strokes or blood vessel changes or of fish oil compared with or in addition to drug treatments.


Farmer A, Montori V, Dinneen S, Clar C. Fish oil in people with type 2 diabetes mellitus. The Cochrane Database of Systematic Reviews 2001, Issue 3. Art. No.: CD003205. DOI: 10.1002/14651858.CD003205.


Once a person has been diagnosed with diabetes he or she is confronted with the challenge of changing lifestyle. They may need to lose weight reduction, adapted their diet and exercise more. For some people medication with oral hypoglycaemic agents or insulin is unavoidable for controlling their blood glucose levels. Control of both blood glucose and blood pressure control is important to prevent complications from diabetes and serious cardiovascular events such as a heart attack or stroke. People go through a process of seeking, receiving and following treatment and advice. Identifying diabetes, strict treatment recommendations, follow up, self-care and self-management, and following a treatment plan are all important in helping a person with diabetes.

Whether a person develops small blood vessel changes that can lead to eye and kidney problems depends a lot on blood glucose levels and how long someone has had diabetes; while the risk of disease of large blood vessels (veins and arteries) depends on age, gender and genetic factors, lifestyle (including nutrition, exercise, smoking) as well as high blood sugars and fats (lipids). This means that it is important that the person keeps to their treatment and care plan (called adherence or compliance). Compliance can be describes as the extent to which a person's behaviour in terms of taking medication, following a diet, or making lifestyle changes agrees with their medical and healthcare advice.

Known ways of encouraging people to keep to treatment plans include giving short-term plans, medication that requires fewer doses per day and is in easy-to-use packaging, is not too expensive, giving information and opportunities for education that are appropriate for the individual, sending out reminders for appointments and tests, and taking measures to ensure satisfaction with services and their healthcare.

What the synthesised research says

Blood glucose control (glycaemic control) was improved to some degree with attention to keeping to treatment. This was measured as an overall slight decrease in people’s glycated haemaglobin levels.

The trials also noted positive effects for people with diabetes keeping their appointments, attending eye clinics attendance, collecting prescription refills and having an improved knowledge of diabetes.

Taking oral medication once daily seemed to be effective in encouraging people to take their medication.

How it was tested

The researchers made a thorough search of the medical literature and found 21 studies containing data on 4,135 people with type 2 diabetes. Fourteen of these trials involved randomly assigning people to recieving interventions aimed at improving adherence to treatment or usual care, five were controlled trials including before and after studies, and one was a population (epidemiological) study.

The interventions included telephone follow ups by nurses, mailed education and information materials, appointment reminders and follow ups of missed appointments, home health aide visits, diabetes education programmes and use of a diabetes educators, pharmacy-based interventions and a patient participation programme.

General cautions, limitations of the clinical trials

Diabetes management involves more than taking medication for lowering blood glucose levels. People are asked to make lifestyle changes including taking regular exercise, reducing their weight, if they are overweight, and other cardiovascular risks. These are all part of their care plan. It may be easier to keep to some parts than others and therefore in improving long-term outcomes. 
How the clinical trials were carried out had some limitations and the numbers of participants were low. They also made the assumption that glycated haemaglobin levels are an effective measure of adherence to treatment.

The benefits were small, which mean that the interventions were not very effective or that larger numbers of people need to be studied in well-conducted clinical trials.


Vermeire E, Wens J, Van Royen P, Biot Y, Hearnshaw H, Lindenmeyer A. Interventions for improving adherence to treatment recommendations in people with type 2 diabetes mellitus. The Cochrane Database of Systematic Reviews 2005, Issue 2. Art. No.: CD003638. DOI: 10.1002/14651858.CD003638.pub2.


Large numbers of people from a wide range of ethnic and cultural groups and at all levels of society are affected by type 2 diabetes. Lifestyle, diet and decreased physical activity all contribute. People are also living longer; the most common type 2 diabetes is more commonly diagnosed over 40 years of age. Diabetes carries a risk of multiple, disabling, yet potentially preventable complications and greatly increases the chances of coronary heart disease and stroke. Any reduction in blood glucose levels and of blood pressure is likely to reduce the risk of complications.

Self-management skills are needed to enable people to manage any medication and improve their health long term as changes in diet and lifestyle are a vital part of their health care. By providing people with effective ongoing education and supporting them emotionally and psychologically they are more likely to be equipped with the knowledge, skills, attitudes and motivation to self manage.

What is known

Diabetes education dominated by a traditional model in which doctors, nurses, dietitians and other members of a healthcare team instruct people (patients) on a one-to-one basis gives active prescription of diet, medication and advice on healthy practices. It may not stimulate effective patient motivation and behaviour changes. The scarcity of time and resources has now led to group-based diabetes education programmes.

Patient-centred education offered to patients and their families can have its own beneficial effect on quality of life, helping to prevent avoidable complications and offering support. Patient-centred education is the close involvement of patients and carers in the planning of the education by soliciting the patient's opinions, concepts, ideas, feelings and questions, offering support, and allowing the patient to be involved in decision making.

Educational programmes are complex as the skills of the educator, where they take place, interaction and rapport between participants, and the number of hours involved all contribute to what a person gets out of the programme or course.  

What the synthesised research says

Group-based, patient-centred educational programmes for people with type-2 diabetes resulted in improved health outcomes.

The 11 trials included in this systematic review showed that group-based diabetes education programmes were of benefit to adults with type 2 diabetes. The evidence showed reduced glycated haemoglobin at four to six months (six trials, 924 people) and 12 months (eight trials). If additional group education sessions were provided on an annual basis, the improvements were maintained longer-term (two to four years) (two and one trial). Fasting blood glucose levels (one trial) and diabetes knowledge (four trials, 708 participants) also improved and were maintained over time.

Group education programmes reduce medication requirements (five trials, 654 people) so that for every five people attending a group-based education programme we can expect one person to reduce diabetes medication in about a year; and improve healthy living self-management skills (five trials), increase self-empowerment (one trial) and improve food-related aspects of quality of life (one trial).  These included freedom to eat or drink and enjoyment of food.

The incidence of diabetic retinopathy may be reduced, measured two to four years later.

Only small changes in blood pressure were seen four to six months bot not at one year; small reductions in body weight were only apparent at around one year.

There is no evidence to suggest that programmes delivered in either primary or hospital diabetes centres (secondary) care are better; nor that the programme is more effective if delivered by a physician, dietitian or nurse trained to deliver a diabetes education programme. Delivering different length group-based diabetes education programme to groups of 4 to 6 participants or 16 to 18 participants did not appear to alter the effectiveness of the course.

How it was tested

The researchers made a thorough search of the medical literature and found 11 trials with 1532 participants carried out in the United States, United Kingdom, Austria, Argentina, Germany, Spain and Italy. One trial recruited Mexican Americans; another recruited 25% South Asians and 75% white people (Caucasians). Two other trials reported that 95% of participants were white people.

The smallest study included 36 participants and the largest study 314 participants. The proportion of men and women was roughly the same in each group with the exception of one trial that recruited only women. Eight of the eleven trials randomly assigned participants to receive group-based diabetes education programmes or routine treatment, which was individual appointments with a healthcare professional or team.

All trials recruited adults with type 2 diabetes and the mean age of participants was between 51 and 65 years. The age ranges were similar, with the lower age bracket being 30-35 years and the highest age bracket being 71-85 years in five trials.

The duration of diabetes was between six and nine years in seven trials; in one trial it was less than a year and in another trial participants were newly diagnosed.

General cautions, limitations of clinical trials

The programmes varied a lot in content, number and timing of sessions and length of follow up. Group-based diabetes education programmes varied in duration with the least intensive being three or four hours per year for two to four years. Eight trials described programmes that ranged from six to fifteen hours of group-based education over a period of between four weeks and 10 months with the most intense education programme being 52 hours over one year.

Not a lot of trials have been carried out.

Only one trial looked at using trained lay health workers and one medical assistants, with little evidence of effectiveness.

The participants in the clinical trials are likely to be committed and motivated to the management of their diabetes. They may also have received extra attention because they were in a clinical trial. More people may drop out of such programmes when they are offered to the general adult population with type 2 diabetes.

Young people with type 2 diabetes and pregnant women were not included as participants in the trials.


Deakin T, McShane CE, Cade JE, Williams RDRR. Group based training for self-management strategies in people with type 2 diabetes mellitus. The Cochrane Database of Systematic Reviews 2005, Issue 2. Art. No.: CD003417. DOI: 10.1002/14651858.CD003417.pub2.


Training courses in self management are a way of helping people with persistent health problems live with them. A key part of self-management for someone with diabetes often involves training them to be able to monitor their own blood glucose levels. They can then measure their levels and see how they change over a day and more and with different food and levels of exercise or activities.

The thinking is that self-monitoring blood glucose levels will help a person control their blood glucose levels. It also may make them more independent and ready to take on lifestyle changes, as well as encouraging them to take their medications and keep to treatment and care plans.

People who are using insulin can adjust their insulin doses and work out the best times to take them, to prevent extreme high and low glucose levels (improved glycaemic control);

Knowing how the blood glucose levels change can also help doctors or diabetes nurses to know when to change treatments. On the other hand, there is some concern that for some people self-monitoring may cause them anxiety, frustration or lead to physical discomfort as involves pricking their own finger. This may make it useful, especially for people not on insulin.

What the synthesised research says

Self monitoring of blood glucose levels might be effective in improving glycaemic control in people with type 2 diabetes who are not using insulin.

Only six eligible randomised controlled trials were found that could provide the information sought. Four trials found more improvement in glycated haemaglobin (HbA1c) control in people measuring their own blood glucose levels than in those who were not. In one high-quality recent study the glycated haemaglobin levels decreased 0.8% in those who were self monitoring and 0.6% in the control group.

The trials did not provide any evidence that self monitoring improves fasting blood glucose levels; number of serious low blood glucose (hypoglycaemic) episodes; quality of life in terms of personal satisfaction, impact on life, worry socially, with work or diabetes related; sense of wellbeing; or satisfaction with treatment. These outcomes were measured in only a few trials.

How it was tested

The researchers made a thorough search of the medical literature and found six eligible clinical trials. A total of 1285 people with diabetes were randomly assigned to self monitoring their blood glucose levels or usual care without self monitoring. They did this for at least 6 months. The trials were carried out between 1989 and 2005.

General cautions, limitations of the clinical trials

A limitation of three studies was that the researchers did not give standard instructions to the participants as to how to adjust their behavior and change their lifestyle and medication to modify their glucose values. Changes would be essential to show positive benefits of measuring blood glucose levels.

Furthermore, people who are less well educated (or English speaking) may have problems in interpreting the information given by the nurses in the trials.

The trials did not have a lot of participants; one included 113 people receiving self monitoring, another 345 people; while the other studies had groups ranging from 12 to 68 people.

There was also the possibility that any improvement in glycated haemaglobin levels was because of the increased attention healthcare providers gave in the trials. Some people may respond better than others to self monitoring, for example those who are newly diagnosed or their blood glucose levels poorly controlled.


Welschen LMC, Bloemendal E, Nijpels G, Dekker JM, Heine RJ, Stalman WAB, Bouter LM. Self-monitoring of blood glucose in patients with type 2 diabetes mellitus who are not using insulin. The Cochrane Database of Systematic Reviews 2005, Issue 2. Art. No.: CD005060. DOI: 10.1002/14651858.CD005060.pub2.


People with diabetes are often at risk of cardiovascular disease because of high blood pressure (hypertension), high blood cholesterol (LDL cholesterol) and increased blood clotting. Over time, people have an increased risk of heart disease, stroke, eye and kidney complications neuropathy that results in poor peripheral circulation as in the lower legs and feet. Dietary changes, exercise and behaviour modification are all recommended for people diagnosed with type-2 diabetes.

What is known

Physical activity uses energy and exercise prescriptions provide specific recommendations for the type, intensity, frequency and duration of physical activity with a specific objective, such as to increase fitness or health.

Aerobic exercise uses moderate repetitive cyclical movements where breathing provides sufficient oxygen, often involving walking, cycling, swimming and jogging. Resistance training uses muscular strength to move a weight or to work against a resistive load (for example exercise with free weights or weight machines). Resistance training builds muscles when performed regularly at a sufficient intensity. With progressive resistance training, the weights start light and increase over a period of time.

Individual choice of aerobic activities is likely to lead to a greater willingness to continue with a program as is progressively building up the levels of exercise. Mixed aerobic and resistance training sessions have become popular in recent years.

What the synthesised research says

Following an exercise program gave people better blood sugar (glycaemic) control than if they did not regularly exercise. Glycated haemoglobin levels, as a measure of blood glucose control, decreased overall by some 0.6% (13 trials, 361 participants) over a relatively short period of time (eight weeks to one year). Medication use also decreased.

The participants improved blood glucose control without weight loss, in the 10 studies (248 participants) reporting this. The people who exercised lost body fat content (visceral and subcutaneous, two trials, 40 participants) and the failure to lose weight with exercise is probably explained by the conversion of fat to muscle (one trial). 
Exercise improved the body's response to insulin (one trial) and decreased blood triglyceride lipids (six trials). No clear difference was found in blood levels of cholesterol (five trials) or systolic blood pressure (four trials). Fasting plasma glucose (nine trials) and insulin concentrations (seven trials) did not change.

These results show a need to engage people and making lifestyle changes. The level of reduction in glycated haemaglobin levels is comparable with those using medications such as metformin and sulphonylureas.

How it was tested

The researchers made a thorough search of the medical literature and found 14 relevant randomised controlled trials. Trials were from nine countries, including Australia. These included 377 participants comparing ongoing exercise against no exercise. The duration of the interventions in the studies ranged from eight weeks to one year.

The mean age of most groups was between 45 and 65 years and slightly more men than women participated.

The shorter lasting studies mainly involved progressive resistance training and the exercise prescription in longer studies tended to involve moderate aerobic exercise such as walking or cycling or controlled endurance training combined with muscle strength training.

Exercise interventions ranged from eight weeks duration (4) to one year (1).

The duration of supervised exercise sessions was generally one hour, ranging from 30 minutes for resistance training to two hours weekly for a Qi Gong program. The number of weekly sessions was generally three (for example, moderate intensity exercise such as walking three times per week at 60 to 79% of maximal oxygen uptake) but varied from one to seven. High intensity exercise regimes included progressive resistance training of three sets of ten to twelve repetitions of lifting weights that represented 70 to 85% of a maximum voluntary contraction load of a muscle group.

Among more recent studies, mixed aerobic and resistance training sessions were prescribed.

General cautions, limitations of the clinical trials

No trial reported diabetic complications over the time of the trial or adverse effects of exercise. Generally, the studies were well-conducted. A limitation was the small number of participants included in the analyses for adiposity, blood pressure, cholesterol, body's muscle and quality of life.


Exercise for type 2 diabetes mellitus. Cochrane Database of Systematic Reviews 2006, Issue 3. Art. No.: CD002968. DOI: 10.1002/14651858.CD002968.pub2 by Thomas DE, Elliott EJ, Naughton GA.

For more information

Refer to Cochrane Metabolic and Endocrine Disorders Group module.