Dr Kevin Fernando explains how SIGN 154 offers evidence-based guidance on glucose control therapies for people with type 2 diabetes

fernando kevin

Dr Kevin Fernando


20160729 independent content logo with text (rgb)

Read this article to learn more about:

  • cardiovascular complications and mortality in people with type 2 diabetes in Scotland

  • why and how Scottish Intercollegiate Guidelines Network (SIGN) updated its guidance on type 2 diabetes therapies in 2017

  • the current risk/benefit profiles of different drug classes.

Key points

GP commissioning messages

After reading this article, ‘Test and reflect’ on your updated knowledge with our multiple-choice questions. Earn 0.5 CPD credits.

The Scottish Diabetes Survey 20161 revealed that 4.8% of the Scottish population (257,728 people) were registered with type 2 diabetes (T2D).

Of these, 87.2% were overweight or obese, 17.2% were current smokers, and 92.8% had a glycated haemoglobin (HbA1c) recorded in the previous 15 months, 58.6% with a result of less than 58 mmol/mol. Importantly, and relevant to new key clinical recommendations in SIGN 154, 9.7% of those with T2D had previously had a myocardial infarction (MI).

State of the nation 2015: the age of diabetes2 published by Diabetes Scotland informs us that, tragically, around 2000 people with diabetes die early every year, predominantly from vascular complications. Since 2008, there has been a 20% increase in the number of individuals with diabetes recorded as having a MI, and a 39% increase in those undergoing coronary revascularisation. Furthermore, there has been a 19% increase in those individuals recorded as having a stroke, and an 82% increase in those with established end-stage renal disease (ESRD).2

Scottish Intercollegiate Guidelines Network (SIGN) diabetes guidelines

Scottish Intercollegiate Guidelines Network (SIGN) 116: Management of diabetes3 was originally published in 2010; since then considerable new clinical trial data have been published focusing on cardiovascular (CV) outcomes of diabetes drugs, rather than simply their glucose-lowering effects. Given this significant new pool of evidence, and to accompany the development of the updated Scottish Diabetes Prescribing Strategy, SIGN 154: Pharmacological management of glycaemic control in people with type 2 diabetes4 was published in November 2017.

The section in SIGN 116 on the pharmacological management of glycaemic control in people with T2D was updated and republished as a stand-alone guideline, SIGN 154. SIGN 116 continues to be current, although the material relating to glucose lowering therapies for people with T2D has been removed and replaced by SIGN 154.

SIGN 154 specifically addresses the pharmacological treatment of glucose-lowering in those with T2D and discusses the risks and benefits of the main classes of oral and injectable glucose-lowering agents, including insulins.

SIGN 154 also offers an updated algorithm to guide the choice of first-, second-, and third-line therapies (see Figure 1). It usefully differentiates therapies by their efficacy, any cardiovascular (CV) benefit, risk of hypoglycaemia, impact on weight, main adverse effects, and prescribing restrictions in renal impairment. Intensification of therapy is recommended after 3–6 months and should be guided by the patient profile. Drugs should be continued at each stage if either individualised HbA1c target is achieved or HbA1c falls >5.5 mmol/mol. Note the algorithm does not apply in severe renal or liver impairment.

Importantly, the section within SIGN 116 discussing optimal targets for glycaemic control in T2D was not updated due to a lack of significant new evidence. SIGN 154 reiterates that an HbA1c target of 53 mmol/mol in those with T2D is reasonable to reduce the risk of microvascular and macrovascular complications. A target of 48 mmol/mol may be appropriate at diagnosis; however HbA1c targets should be individualised to ensure that the benefits of treatment outweigh any harms, particularly the risk of hypoglycaemia.

Similarly, SIGN 154 refers back to SIGN 116 for recommendations regarding lifestyle and it will still be necessary to refer to SIGN 116 for recommendations on the management of diabetes complications.

Some key clinical recommendations from SIGN 154 are shown in Box 1.

Box 1: Key recommendations4

The following were highlighted by the guideline development group as the key clinical recommendations that should be prioritised for implementation.

Targets for glycaemic control

  • An HbA1c target of 7.0% (53 mmol/mol) among people with type 2 diabetes is reasonable to reduce the risk of microvascular and macrovascular disease. A target of 6.5% (48 mmol/mol) may be appropriate at diagnosis. Targets should be set with individuals in order to balance benefits with harms, in particular hypoglycaemia and weight gain.

Metformin

  • Metformin should be considered as the first-line oral treatment option for people with type 2 diabetes.

Sodium glucose co-transporter 2 inhibitors

  • In individuals with type 2 diabetes and established cardiovascular disease, SGLT2 inhibitors with proven cardiovascular benefit (currently empagliflozin and canagliflozin) should be considered.

Glucagon-like peptide-1 receptor agonists

  • For individuals with type 2 diabetes and established cardiovascular disease, GLP1RAs with proven cardiovascular benefit (currently liraglutide) should be considered.

HbA1c =glycated haemoglobin, haemoglobin A1c; SGLT2=sodium glucose co-transporter 2; GLP1RAs=glucagon-like peptide-1 receptor agonists

Pharmacological treatment options in people with type 2 diabetes

Metformin

SIGN 154 reminds us that metformin remains optimal first-line oral therapy for those with T2D.

Metformin is a moderately potent glucose-lowering agent with proven CV benefit from the seminal United Kingdom Prospective Diabetes Study4,5 (UKPDS). It tends to cause weight reduction and overall has a low risk of hypoglycaemia.4

The main adverse effect of metformin is gastrointestinal disturbance (predominantly diarrhoea) and it should be used with caution in those with moderate renal impairment. SIGN 154 signposts the British national formulary (BNF) (see bnf.nice.org.uk/drug/metformin-hydrochloride.html)6 and the relevant summary of product characteristics (SPC) for metformin to guide correct dosages at different stages of renal impairment.

Sulphonylureas (SUs)

SIGN 154 and also NICE Guideline 28 recommends that practitioners consider sulphonylureas (SUs) as first-line oral therapy in those who are intolerant of, or who have contraindications to, metformin.4,7 Sulphonylureas can also be considered in people presenting with osmotic symptoms such as thirst or polyuria, as add-on, second-line therapy to metformin and may also be useful in triple therapy.4

Sulphonylureas are potent glucose-lowering agents but there remains uncertainty regarding their CV safety profile, which indicates no proven CV benefit and possible CV harm.8 This question will hopefully be resolved during 2019 on publication of the CAROLINA9 randomised controlled trial (RCT), which is investigating the CV safety of linagliptin versus glimepiride.4

Sulphonylureas cause weight gain and overall have a high risk of hypoglycaemia; they should be used with caution in those vulnerable to the effects of hypoglycaemia, such as the elderly.4

SIGN 154 recommends careful monitoring when using SUs in the context of mild-to-moderate renal impairment owing to a greatly increased risk of hypoglycaemia in these patients. Sulphonylureas should be avoided in severe renal impairment. SIGN 154 again directs practitioners towards the BNF and relevant SPCs for guidance on appropriate dosages in renal impairment.4

Thiazolidinediones (TZDs)

SIGN 154 suggests practitioners consider pioglitazone (the only remaining thiazolidinedione [TZD] licensed for use in the UK), usually as dual or triple therapy, in those with T2D.

Pioglitazone is a moderately potent glucose-lowering agent with probable CV benefit (subgroup analyses from the PROactive trial10 demonstrated a reduction in non-fatal and fatal MI and recurrent stroke) but it is contraindicated in those with heart failure.4

Pioglitazone causes weight gain (more so than SUs) and overall has a low risk of hypoglycaemia.4

SIGN 154 warns us that pioglitazone is associated with a number of adverse effects including peripheral oedema, increased fracture risk, and a dose-dependent increase in the risk of bladder cancer.4 During 2011, the Medicines and Healthcare Products Regulatory Agency (MHRA) warned that pioglitazone should not be used in those with active bladder cancer, previous bladder cancer, or uninvestigated haematuria.11

SIGN endorses the use of pioglitazone at chronic kidney disease stage 3A; however, it can also be used in ESRD with no dose titration required.12 It should not be used in hepatic impairment.4,12

Dipeptidyl peptidase-4 inhibitors (‘gliptins’ or ‘DPP-4 inhibitors’)

SIGN 154 recommends that practitioners consider gliptins (DPP-4 inhibitors), usually as dual or triple therapy, in people with T2D.4

Gliptins are mildly potent glucose-lowering agents with no proven CV benefit but have an established CV safety profile from three high-quality CV outcome trials (CVOTs): SAVOR-TIMI 5313, EXAMINE,14 and TECOS15.

Gliptins are weight-neutral and overall have a low risk of hypoglycaemia. They are generally well tolerated with few significant adverse effects.4 Notably, the above three CVOTs did not demonstrate an increased risk of pancreatitis with gliptins,13–15 which was previously highlighted as a concern by the MHRA during 2012.16

Gliptins can be used at all stages of renal impairment albeit at reduced doses. Linagliptin does not require dose titration in renal impairment. SIGN 154 once again signposts the BNF (see bnf.nice.org.uk/treatment-summary/type-2-diabetes.html) and relevant SPCs for guidance on appropriate dosages of gliptins at different stages of renal impairment.4

Sodium glucose co-transporter 2 (SGLT2) inhibitors

SIGN 154 suggests that sodium glucose co-transporter 2 (SGLT2) inhibitors are considered as add-on therapy to metformin in people with T2D. Sodium glucose co-transporter 2 inhibitors can also be used as triple therapy; however dapagliflozin and pioglitazone should not be combined4 due to licensing restrictions.

Sodium glucose co-transporter 2 inhibitors are moderately potent glucose-lowering agents. Uniquely, empagliflozin and canagliflozin have proven CV benefit from two recent high-quality CVOTs investigating patients at high risk of CV disease (EMPA-REG OUTCOME17 and CANVAS,18 respectively). Empagliflozin demonstrated significant reductions in CV death, all-cause death, and hospitalisation for heart failure compared with placebo, while canagliflozin demonstrated a significant reduction in a primary composite CV endpoint as well as a reduction in hospitalisation for heart failure. Both drugs also demonstrated a small but significant reduction in the progression of renal disease.

Driven by these two landmark CVOTs, a key new clinical recommendation in SIGN 154 is to preferentially consider an SGLT2 inhibitor with proven CV benefit (i.e. currently only empagliflozin and canagliflozin) in those with T2D and established CV disease.4

Sodium glucose co-transporter 2 inhibitors can lead to weight loss due to their glucosuric effect and overall have a low risk of hypoglycaemia due to their insulin-independent mode of action.4

The adverse effect profile of SGLT2 inhibitors is similar to the symptoms of T2D; commonly reported are symptoms of genital mycotic infections and also urinary tract infections.4,19

SIGN 154 reminds us that the MHRA issued a drug safety update during 2016, warning about reports of diabetic ketoacidosis (DKA) with SGLT2 inhibitor use.20 It is a rare adverse effect (between 1:1000 and 1:10,000 patients affected) with certain individuals more likely to develop this complication (those with low endogenous insulin reserves, those with increased insulin requirements [e.g. with acute illness, or having surgery, or who misuse alcohol] or conditions that lead to reduced oral intake or severe dehydration).

Sodium glucose co-transporter 2 inhibitors should therefore be part of ‘sick day guidance’ and be stopped temporarily alongside other drugs such as metformin or angiotensin-converting-enzyme (ACE) inhibitors during acute dehydrating illness or major surgery.4,21 In several cases, the DKA was associated with normal or near-normal glucose levels—euglycaemic DKA. If DKA is suspected, the MHRA advises checking for urine or blood ketones even if blood glucose levels are normal.20

An unexpected finding of the CANVAS18 CVOT was a doubling of the risk of lower-limb amputations (predominantly toe amputations) with canagliflozin compared with placebo. The incidence of lower limb amputation was 7 per 1000 patient-years in the canagliflozin 100 mg group and 5 per 1000 patient-years in the canagliflozin 300 mg group, compared with 3 per 1000 patient-years in the placebo group.18 This triggered an EMA review22 of all SGLT2 inhibitors, which concluded that the risk:benefit ratio of SGLT2 inhibitors remains favourable, but did mandate that a warning should be included in the product information of all SGLT2 inhibitors to reflect this finding. Furthermore, CANVAS18 also found an increased risk of fractures with canagliflozin, and the US Food and Drug Administration (FDA) recommends a fracture risk assessment when prescribing canagliflozin.23

Consistent with current licensing, all currently available SGLT2 inhibitors cannot be prescribed if estimated glomerular filtration rate (eGFR) is <60ml/min/1.73m2. After treatment with an SGLT2 inhibitor has been initiated, if eGFR subsequently drops below 60ml/min/1.73m2, dapagliflozin should be stopped and dose reductions are recommended for empagliflozin and canagliflozin. Further details can be found in the BNF and relevant SPCs.24–26

Glucagon-like peptide-1 receptor agonists (GLP1RAs)

SIGN 154 recommends that practitioners consider glucagon-like peptide-1 receptor agonists (GLP1RAs) in people with a body mass index (BMI) greater or equal to 30 kg/m2 (or ethnicity-adjusted equivalent), in combination with oral diabetes drugs or basal insulin (or both) as third- or fourth-line therapy. Furthermore, GLP1RAs should be considered as alternatives to insulin for those in whom combination oral therapies have been ineffective.4

Glucagon-like peptide-1 receptor agonists are potent glucose-lowering agents. Notably, liraglutide has proven CV benefit from the high-quality LEADER27 CVOT. Liraglutide demonstrated significant reductions in death from CV causes and all-cause death compared with placebo, as well as a small but significant reduction in the progression of renal disease. ELIXA28 and EXSCEL29 were two other CVOTs that investigated the GLP1RAs lixisenatide and exenatide (given once daily and once-weekly, respectively); these trials did not demonstrate any CV benefit compared with placebo but did establish the CV safety of these drugs.

Driven by the LEADER CVOT,27 a key new clinical recommendation in SIGN 154 is to preferentially consider a GLP1RA with proven CV benefit (i.e. currently only liraglutide) in those with T2D and established CV disease.

Glucagon-like peptide-1 receptor agonists can lead to significant weight reduction and overall have a low risk of hypoglycaemia.4

The adverse effect profile of GLP1RAs comprises mainly gastrointestinal disturbance (commonly nausea, vomiting, and diarrhoea).4 Reassuringly, the above three CVOTs27–29 did not observe any statistically significant increase in cases of pancreatitis or pancreatic cancer, both of which had previously been highlighted as possible concerns with GLP1RAs.30,31 Note that the LEADER CVOT did observe a significantly increased incidence of acute gallstone disease with liraglutide.27

Glucagon-like peptide-1 receptor agonists can be used in severe renal impairment; however, advice for use and dose adjustment varies between the different GLP1RA agents. Further details can be found in the BNF and relevant SPCs.4

Insulin

SIGN 154 provides some helpful guidance to practitioners considering insulin when combination oral therapy is ineffective and GLP1RAs are not indicated (BMI <30 kg/m2), not tolerated, or are contraindicated.4

Metformin should be continued when insulin is initiated; however, practitioners should consider reducing or stopping SUs and assess the benefits and risks of continuing other agents on an individual basis.4

Once-daily bedtime neutral protamine Hagedorn (NPH) (isophane) insulin is recommended when adding insulin to metformin. Analogue insulins should be considered on an individual basis according to risk of hypoglycaemia.4

The dose of bedtime basal insulin should be titrated against fasting blood glucose levels. If the mutually agreed HbA1c target is not achieved, consider introducing prandial insulin injections.4

Information for healthcare practitioners and patients

SIGN 154 usefully provides a checklist (Section 12.1) of information for all classes of diabetes drugs (including insulin) that healthcare professionals (HCPs) can use when discussing glucose-lowering in people with type 2 diabetes and their carers. This information can also be used to guide the development of locally produced information materials. See Box 2 for a list of suggested further useful sources of information for HCPs, those with diabetes, and their carers.4

Box 2: Sources of further information for healthcare practitioners, people with diabetes, and their carers4

Conclusion

SIGN 154 is a welcome and pragmatic addition to the suite of guidelines produced by SIGN. SIGN 154 provides a user-friendly, patient-centred algorithm, which allows the individualisation of treatment choices in T2DM taking into account a number of important patient characteristics. Moreover, key clinical recommendations in SIGN 154 incorporate the latest published CVOT data; CVD remains the leading cause of death in those with T2D, and these new recommendations will help drive improved outcomes especially in those with T2D and established CVD.

Dr Kevin Fernando

GP partner, North Berwick; committee member of the Primary Care Diabetes Society; GPwSI in Diabetes and Medical Education

Key points

  • SIGN 154 updates the pharmacological treatment of glucose-lowering in those with T2D and also contains recommendations on glycaemic targets in T2D. Lifestyle recommendations, and the management of complications are retained in SIGN 116
  • An HbA1c target of 53 mmol/mol in those with T2D is reasonable to reduce the risk of microvascular and macrovascular complications; however, HbA1c targets should be individualised to balance benefits against harms
  • Intensification of therapy is recommended after 3–6 months and should be guided by patient profile
  • Medications should be continued at each stage if either individualised HbA1c target is achieved or HbA1c falls more than 5.5 mmol/mol in 3–6 months
  • Metformin remains optimal first-line oral therapy for those with T2D
  • In those with T2D and established CVD, SGLT2 inhibitors with proven CV benefit should be preferentially considered
  • In those with T2D and established CVD, GLP1RAs with proven CV benefit should be preferentially considered
  • Once-daily bedtime NPH (isophane) insulin is recommended when adding insulin to metformin. Continue metformin but consider stopping or reducing SUs
  • SIGN 154 provides a useful checklist of information about all classes of diabetes drugs that healthcare practitioners can use when discussing glucose-lowering therapies with people with T2D and their carers.

T2D=type 2 diabetes; HBA1c =glycated haemoglobin, haemoglobin A1c; SGLT2=sodium glucose co-transporter 2; CVD=cardiovascular disease; CV=cardiovascular; GLP1RA=glucagon-like peptide-1 receptor agonists; NPH=neutral protamine Hagedorn; SU=sulphonylurea

Commissioning messages

written by Dr David Jenner, GP, Cullompton, Devon

  • Local commissioning organisations in Scotland should incorporate this updated SIGN guidance into local formularies, whereas in England and Wales relevant NICE guidance and technology assessments should be used to inform these
  • Local formularies should clearly identify the licensed indications for each drug and the licensed indications for their use in combination with other agents:
    • these drug indications are now very complex so an illustrative clinical algorithm can be helpful, made available on an app for smartphones or desktop computers to allow for rapid updates
  • Some anti-diabetic drugs are expensive, so acquisition costs should also be identified
  • Increasingly, newer drugs are demonstrating cardiovascular disease outcome benefits and these should be clearly identified
  • Given the complexity of this therapeutic area and rapid appearance of new evidence and drugs, commissioners should provide regular education sessions on diabetes therapies for GPs and nurses to support their formulary guidance.

SIGN=Scottish Intercollegiate Guidelines Network


Now Test and reflect: view our multiple choice questions

g logo gls turquoise

Read more about this topic with the Guidelines summary of SIGN 116 and 154 on Management of type 2 diabetes

References

  1. Scottish Diabetes Survey Monitoring Group. Scottish Diabetes Survey 2016. NHS Scotland. Available at: www.diabetesinscotland.org.uk/Publications.aspx?catId=1
  2. Diabetes Scotland. State of the nation 2015: the age of diabetes. Available at: www.diabetes.org.uk/In_Your_Area/Scotland
  3. Scottish Intercollegiate Guidelines Network. Management of diabetes. SIGN 116. Edinburgh: SIGN, 2010. Available at: www.sign.ac.uk/assets/sign116.pdf
  4. Scottish Intercollegiate Guidelines Network. Pharmacological management of glycaemic control in people with type 2 diabetes. SIGN 154. Edinburgh: SIGN, 2017. Available at: www.sign.ac.uk/assets/sign154.pdf
  5. No authors listed. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). UK Prospective Diabetes Study (UKPDS) Group. Lancet 1998; 352 (9131): 854–865.
  6. NICE website. BNF. Metformin hydrochloride. bnf.nice.org.uk/drug/metformin-hydrochloride.html (accessed 30 January 2017).
  7. NICE. Type 2 diabetes in adults: management. NICE Guideline 28. NICE, 2015. Available at: www.nice.org.uk/guidance/ng28
  8. Dwyer M. Time to be cautious about prescribing sulfonylureas? Br J Gen Pract 2015; 65 (631): 93–94.
  9. CAROLINA: Cardiovascular outcome study of linagliptin versus glimepiride in patients with type 2 diabetes. Available at: https://clinicaltrials.gov/ct2/show/NCT01243424
  10. Wilcox R, Bousser M, Betteridge D et al. Effects of pioglitazone in patients with type 2 diabetes with or without previous stroke: results from PROactive (PROspective pioglitAzone Clinical Trial In macroVascular Events 04). Stroke 2007; 38 (3): 865–873.
  11. Medicines and Healthcare products Regulatory Agency. Pioglitazone: risk of bladder cancer. Drug safety update Aug 2011, vol 5, issue 1: A1.
  12. eMC website. Pioglitazone hydrochloride. www.medicines.org.uk/emc/ingredient/2226
  13. Scirica B, Bhatt D, Braunwald E et al, for the SAVOR-TIMI 53 Steering Committee and Investigators. Saxagliptin and cardiovascular outcomes in patients with type 2 diabetes mellitus. N Engl J Med 2013; 369: 1317–1326.
  14. White W, Cannon C, Heller S et al, for the EXAMINE Investigators. Alogliptin after acute coronary syndrome in patients with type 2 diabetes. N Engl J Med 2013; 369: 1327–1335.
  15. Green J, Bethel M, Armstrong P et al, for the TECOS Study Group. Effect of sitagliptin on cardiovascular outcomes in type 2 diabetes. N Engl J Med 2015; 373: 232–242.
  16. Medicines and Healthcare products Regulatory Agency. Dipeptidylpeptidase-4 inhibitors: risk of acute pancreatitis. Drug safety update September 2012, vol 6, issue 2: A3
  17. Zinman B, Wanner C, Lachin J et al, for the EMPA-REG OUTCOME Investigators. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med 2015; 373: 2117–2128.
  18. Neal B, Perkovic V, Mahaffey K et al, for the CANVAS Program Collaborative Group. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med 2017; 377: 644–657.
  19. Monami M, Nardini C, Mannucci E. Efficacy and safety of SGLT2 inhibitors in T2DM: a meta-analysis of RCTs. Diabetes Obes Metab 2014; 16 (5): 457–466.
  20. Medicines and Healthcare products Regulatory Agency. SGLT2 inhibitors: updated advice on the risk of diabetic ketoacidosis. Drug safety update April 2016, vol 9, issue 9: 1.
  21. Scottish Government Model of Care Polypharmacy Working Group. Polypharmacy guidance (2nd edition). Scottish Government, March 2015. Available at: www.sehd.scot.nhs.uk/publications/DC20150415polypharmacy.pdf
  22. European Medicines Agency. SGLT2 inhibitors: information on potential risk of toe amputation to be included in prescribing information. EMA/118223/2017. April 2017.
  23. US Food and Drug Administration. FDA Drug Safety Communication 2015: FDA revises label of diabetes drug canagliflozin (Invokana, Invokamet) to include updates on bone fracture risk and new information on decreased bone mineral density. 9 October 2015.
  24. NICE website. BNF. Dapagliflozin. bnf.nice.org.uk (accessed 30 January 2017).
  25. NICE website. BNF. Empagliflozin. bnf.nice.org.uk (accessed 30 January 2017).
  26. NICE website. BNF. Canagliflozin. bnf.nice.org.uk (accessed 30 January 2017).
  27. Marso S, Daniels G, Brown-Frandsen K et al, for the LEADER Steering Committee on behalf of the LEADER Trial Investigators. Liraglutide and cardiovascular outcomes in type 2 diabetes. N Engl J Med 2016; 375: 311–322.
  28. Pfeffer M, Claggett B, Diaz R et al, for the ELIXA Investigators. Lixisenatide in patients with type 2 diabetes and acute coronary syndrome. N Engl J Med 2015; 373: 2247–2257.
  29. Holman R, Bethel M, Mentz R et al, for the EXSCEL Study Group. Effects of once-weekly exenatide on cardiovascular outcomes in type 2 diabetes. N Engl J Med 2017; 377: 1228–1239.
  30. ‘Dispatches’ programme. Diets, drugs and diabetes. Channel 4, Monday 10 June 2013 www.channel4.com/info/press/news/dispatches-investigation-raises-diabetes-drug-questions (accessed 8 February 2018).
  31. Gale E. Incretin therapy: should adverse consequences have been anticipated? Editorial. BMJ 2013; 346: f3617.