Volume 37, Issue 6 p. 216-218a
Policy Profile
Free Access

The billion pound challenge: deprescribing, diabetes and diets

Mark Greener BSc(Hons), MRSB,

Mark Greener BSc(Hons), MRSB

Medical Correspondent

Search for more papers by this author
First published: 30 November 2020


Low calorie diets don't just reduce blood glucose sufficiently to, in many cases, place type 2 diabetes into remission, they also offer the opportunity to withdraw or reduce the dose of drugs for diabetes and some comorbidities. Mark Greener here examines the potential for deprescribing in people with type 2 diabetes, the issues facing health care professionals and the outstanding questions.

Diabetes care in England passed a landmark during 2018, when the NHS bill for glucose-lowering drugs exceeded £1 billion. Almost 1-in-20 prescriptions issued by GPs are for diabetes treatments,1 some of which cause potentially serious adverse events. Meanwhile, increasing evidence suggests that losing sufficient weight to put type 2 diabetes (T2D) into remission can reduce the need for glucose-lowering medicines and drugs for comorbidities.2-4 ‘Deprescribing of medicines tends to be the most important factor when patients decide to follow a low calorie diet,’ [LCD] comments Huda Latif, Diabetes Specialist Pharmacist at Luton Clinical Commissioning Group.

Increasing evidence

Patients in the intervention group of the Diabetes Remission Clinical Trial (DiRECT) replaced all their meals with a formula liquid diet providing 825–853kcal/day for three months, which could be extended up to five months. During the next two to eight weeks, the group receiving the LCD (defined as 800–1200kcal day; a very LCD is less than 800kcal a day), also called Total Diet Replacement (TDR), reintroduced food following a structured diet and attended fortnightly until end of food reintroduction. Thereafter, patients made monthly visits to maintain weight loss.3 So, TDR is one of the three integral stages that make up the DiRECT programme.

Patients discontinued oral antidiabetic drugs and certain antihypertensives, which were reintroduced according to the research protocol and national guidelines.3 ‘Before starting the TDR, we stopped all drugs for diabetes, partly because of the risk of hypoglycaemia, but also as the research was designed to detect diabetes remission,’ explains Alison Barnes, Senior Research Dietitian on DiRECT. ‘This was prudent as in most cases we saw a rapid drop in glucose levels within a week of starting the TDR, which reflects the decline in liver fat. However, not everyone in the study was able to achieve physiological remission, for a number of reasons. In a real-life setting on a larger scale, careful monitoring and a deprescribing approach to diabetes medications would maximise the benefits of a DiRECT-style intervention.’

DiRECT enrolled 306 people (the intention to treat analysis included 149 in each group) with T2D, BMIs of 27–45kg/m2, aged 20–65 years. All had been diagnosed with T2D within the last six years and none were prescribed insulin therapy. Mean body weight declined by 10.0kg at 12 months in the TDR group compared with 1.0kg among controls. Mean HbA1c fell by 0.9% overall (by 17mmol/mol in those achieving remission) with a halving of diabetes medications in the intervention group, and rose by 0.1% despite increased medication use by controls. After 12 months, 24% of the TDR group lost at least 15kg compared with none of the controls. The intervention increased the likelihood of diabetes remission almost 20-fold (odds ratio [OR] 19.7): 46% and 4% of the TDR and control group respectively.3

At 24 months, 36% of the TDR group and 3% of controls remained in remission.5 ‘Weight regain was associated with loss of remission. HbA1c was 17mmol/mol lower in those who maintained remission and was higher than baseline only in those who never achieved remission,’ Ms Barnes adds.

‘DiRECT shows that people with short duration T2D who are able to lose a significant amount of weight using whatever dietary approach suits them have a very good chance that their T2D will go into remission,’ says Ms Barnes, who is also Lecturer in Human Nutrition & Dietetics at Newcastle University and a Diabetes UK Clinical Champion. ‘The benefits often emerge rapidly, which is very motivating for individuals wanting to improve their health. Hunger seems to subside within a few days of starting the TDR and participants often report having more energy, as surprising as that may seem. But it's important to prepare people, such as offering advice about removing temptations from the home environment, finding sources of social support and counselling about potential side effects, such as constipation. In addition, following TDR offers the opportunity for people to change their relationship to food. We make about 200 food choices a day. Some of these, such as dipping into the biscuit jar, are often subconscious, triggered by availability or habit. Following a structured programme helped many of our study participants break these unhelpful behavioural patterns.’

DiRECT enrolled people from North-East England and Scotland. But the findings translate to other populations. The open-label DIADEM-I study, performed in primary care and community settings in Qatar, analysed 147 people aged 18–50 years diagnosed with T2D in the previous three years and used the same programme structure as DiRECT. After a year, patients randomised to the TDR showed a mean 11.98kg reduction in body weight compared with 3.98kg among controls with 15% and 1% respectively losing at least 15% of their body weight. Patients randomised to the TDR were about 12-fold more likely than controls to attain diabetes remission (HbA1c <6.5% [<48mmol/mol], 61% and 12% respectively; OR 12.03) and normoglycaemia (HbA1c <5.7% [<39mmol/mol] and no diabetes medication for ≥3 months; 33% and 4% respectively; OR 12.07).4

Despite a marked reduction in weight, some cases of T2D may be too advanced for TDR to induce remission. Ms Barnes notes that some patients may have relatively advanced T2D on enrolment. ‘In DiRECT, certain patients, such as those taking at least two drugs for diabetes before starting the TDR and those with higher HbA1c, were less likely to achieve remission even if they lost weight,’ she remarks. ‘This may suggest that they had fewer functional beta-cells remaining than those who entered remission. These patients may be physiologically unable to achieve remission. Medication use can still be reduced, although it's likely that some will still be needed. However, weight loss has many other health benefits, including a reduced risk of cardiovascular disease and certain cancers. So, it's still important to encourage and support weight loss.’

‘Patients need to be aware that the LCD programme does not “cure” diabetes, but they have a chance to achieve and maintain remission, eliminating the progression of complications and risks associated with diabetes,’ Mrs Latif says. ‘I envision a life to my patients that is disease free with good quality as we age gracefully.’

Opportunities for deprescribing

‘When patients adopt a LCD, they lose weight dramatically and most of my patients are delighted,’ adds Mrs Latif. ‘This, in turn, reduces their blood glucose levels and HbA1c. Patients whose HbA1c declines to <48mmol/mol would be eligible for deprescribing, especially if they are taking medication that increases the risk of developing hypoglycaemia, such as sulphonylureas or meglitinides.’ The savings could be considerable. According to NHS Digital, the net ingredient cost for diabetes drugs at the clinical commissioning group level was between £208 and £587 per patient on the Quality and Outcomes Framework diabetes register in 2018–19.

DIADEM-I confirmed that following programmes including TDR offered numerous opportunities for deprescribing. At baseline, 91% and 88% of the TDR and control groups received antidiabetic medication. After 12 months, 6% of the TDR group took antidiabetic medications compared with 81% of controls. Blood pressure and lipid levels also improved more in the TDR group. So, at baseline, 34% and 29% of the TDR and control groups received antihypertensives respectively, compared with 16% and 35% after 12 months. Moreover, 31% and 40% of the TDR and control groups received lipid-lowering medicines respectively, compared with 26% and 76% at 12 months.4

‘It naturally follows that a significant reduction in HbA1c levels necessitates the need to take patients off their long-term oral T2D medication as their clinical condition improves,’ says Dr Farhan Rabbani, a GP in Wallington Surrey and Medical Director of The Diabetes Reversal Company. ‘Significant weight loss often improves other metabolic markers including hypertension and hypercholesterolaemia. However, in DiRECT, some patients had to restart their antihypertensives when they began eating normally, so close monitoring is essential.’ Mrs Latif adds that close monitoring is especially important when food is gradually re-introduced.

Tread cautiously

While LCDs offer the prospect of deprescribing, health care professionals need to tread cautiously, counselling, monitoring, motivating and supporting patients. ‘I found out that engaging patients on the LCD programme was not difficult. Indeed, almost 95% were keen to start as soon as possible,’ Mrs Latif reports. ‘I explained in detail what the programme involves and the health benefits driven by opportunities to deprescribe. I always stress that I will follow patients in every step of the way, either through face-to-face consultations or remotely. I reassure patients that deprescribing specific agents, especially those that induce hypoglycaemia and diabetic ketoacidosis, is clinically safe and important. Deprescribing will not put patients at an increased risk of developing complications.’

Dr Rabbani stops medications, such as glinides and sulphonylureas, that can induce hypoglycaemia in patients starting a LCD. ‘In other diet plans, which have a gentler rate of weight loss, clinicians will need to take a more nuanced approach to deprescribing. The same principles apply to antihypertensives where there can be a risk of hypotension if medication is not reduced at a rate to match decreasing blood pressure as weight normalises. Lipid-lowering medicines carry a lower risk, as in general, high or low cholesterol levels will not cause immediate symptoms.’

In addition, health care professionals may need to consider drug interactions and other risks, Mrs Latif comments. For example, some drugs (such as beta-blockers) can mask hypoglycaemic symptoms, while some antidiabetic medications may interact with other agents. She says that health care professionals should always consider pharmacokinetic properties before referring eligible patients.

‘Sodium-glucose co-transporter-2 inhibitors [SGLT2i] can enhance diabetic ketoacidosis, which indicates discontinuing therapy is paramount prior to programme initiation,’ Mrs Latif says. ‘On the other hand, continued therapy with metformin, DPP4 inhibitors, thiazolidinediones [glitazones] and GLP-1 receptor agonists pose no risk to patients following a LCD; however, regular reviews are required to determine appropriateness of therapy if continued. As there is no risk of hypoglycaemia, there is no clinical need to stop these drugs as long as the health benefits outweigh the risks, after engagement on shared decision making with the patient.’

Indeed, continuing certain antidiabetic medicines may be beneficial. ‘GLP-1 receptor agonists seem to show cardioprotective actions independently of their effect on blood glucose, increase satiety and slow gastric emptying,’ Mrs Latif points out. ‘So, GLP-1 receptor agonists can be continued during a LCD programme and their use reviewed during the maintenance phase.’

‘There is limited information and guidance on reducing or stopping insulin therapy, which must be done under specialist guidance,’ Dr Rabbani adds. ‘In general, a 10% drop in body weight will significantly improve a patient's HbA1c levels necessitating removal of some or all of their long-term medications. Using certain medications, such as metformin and some gliptins, long term increases the strain on renal function and the dosages of some medications need to be changed or stopped as renal function declines, because of, for example, advancing age or diabetic nephropathy,’ Dr Rabbani says. ‘Health care professionals should monitor patients closely with HbA1c testing at least every three months.’ The Diabetes Reversal Company uses a three-step process to support patients during deprescribing (Figure 1).

The Diabetes Reversal Company's 3-step process to support patients during deprescribing

Mrs Latif provides blood glucose monitors to patients taking agents that can predispose to hypoglycaemia. ‘I also advise patients taking antihypertensive agents to purchase a blood pressure monitor to record and monitor their blood pressure readings for review and to keep a weekly record of their weight and waist circumference,’ she says. ‘I've found that improvements in these measurements can motivate patients to adhere to the programme. Patients have full control of their health outcomes and are able to observe the clinical benefits.’

Mrs Latif reports that she referred one patient to a LCD programme who showed pre-prandial morning blood glucose levels of 16–19mmol/L. ‘He used to have large meal portions,’ she says. ‘Within a week of starting the LCD programme, driven by self-motivation, family support and professional advice, his readings were as low as 11mmol/L. He was very pleased. With further encouragement from his health care professionals, family and friends, he pushed himself to maintain even lower readings and to meet the targets we set together. During regular follow-ups, we agreed to deprescribe all medications within a year of LCD including antidiabetic and antihypertensive therapies.’

Outstanding questions

Dr Rabbani suggests that there are several outstanding questions regarding deprescribing in people with T2D who enter remission following LCD. ‘Currently, for example, NHS GPs and consultants have not received formal training in deprescribing. So, will they feel competent and safe to manage deprescribing?’, Dr Rabbani asks. ‘Will their medical indemnity cover deprescribing? What are the training requirements for clinicians to undertake deprescribing and how will this be delivered nationally? Will the expectation be on all GPs or consultants to support deprescribing or will we need sub-specialties to manage the process?’

In addition, Dr Rabbani notes that a patient's weight can fluctuate during the weight loss programme and afterwards as their compliance to the lifestyle changes varies. ‘How will we monitor these patients more closely and manage their medications if their weight increases leading to increased HbA1c levels?’, he asks.

Ms Barnes comments that long-term follow-up of the DiRECT and other TDR studies will help to determine, for example, to what extent remission impacts rates of diabetes complications. ‘Historically it was thought that beta-cell number inevitably declines over time,’ she says. ‘But weight loss changes the trajectory of T2D by regaining residual beta-cell function. How durable this effect can be and what impact the pre-existence of diabetes prior to remission may have on longer-term outcomes remain to be determined. For remission, the key is significant weight loss (at least 10kg for most people). This can be achieved with a variety of dietary approaches, so sharing information and finding what suits the individual is important. We also need to consider ongoing dietary management for individuals who don't achieve remission. When we know that beta-cell recovery has not been possible, low carbohydrate diets could offer a therapeutic strategy, reducing the pressure on remaining beta cells and lowering blood glucose irrespective of any effect on weight. This is often accompanied by a reduced need for glucose lowering medications.’

Studies also need to ascertain whether the benefits offered by LCD apply to different patient groups, such as the elderly and those from black and other ethnic backgrounds. ‘Patients older than 65 years of age are not eligible for LCD due to increased risk of hypoglycaemia and the higher prevalence of comorbidities, including renal and hepatic impairment, that can predispose certain patients to hypoglycaemia,’ Mrs Latif remarks. ‘This is particularly important in frail, elderly patients, who are at an increased risk of falls. Poly-pharmacy is very common in elderly people and medicine optimisation, which may involve deprescribing agents that can trigger hypoglycaemic episodes and other specific medicines when the treatment is not clinically relevant, is essential.’ ‘The benefits of weight normalisation are numerous and advantageous to all age groups and ethnicities and should, therefore, be encouraged,’ Dr Rabbani says. However, some patients, such as the elderly, need especially careful monitoring.

‘If there is one thing I've learnt over the years as a GP treating patients from cradle to grave, it's that the human body has an amazing ability to repair and self-heal given the right environment,’ Dr Rabbani concludes. ‘If patients are able to address their lifestyle and normalise their weight, they will not only feel better, they will also see improved clinical markers for their long-term conditions. From our extensive experience, the best thing about this is that they often then act as health advocates and inspiration to their family and friends. Getting healthy and feeling great is infectious!’


References are available online at www.practicaldiabetes.com.