Rationale and impact

These sections briefly explain why the committee made the recommendations and how they might affect practice. They link to details of the evidence and a full description of the committee's discussion.

Diagnosing hypertension

Recommendations 1.2.1 to 1.2.5 and 1.2.8

Why the committee made the recommendations

Overall, there was limited new evidence on the accuracy of different methods of measuring blood pressure. Most of the studies identified were small, and the populations and protocols for measurement varied making interpretation difficult. However, the committee agreed that it was important to focus on the evidence from these more recent studies (post-2000) because the evidence should reflect the current use of electronic sphygmomanometers, which have replaced mercury-based sphygmomanometers.

The evidence did not show that changing the current blood pressure thresholds for clinic measurement or home blood pressure monitoring (HBPM) would improve diagnostic accuracy compared with ambulatory blood pressure monitoring (ABPM), so the committee agreed the 2011 thresholds for diagnosis should be retained. The committee noted that these are in line with most international guidance.

Limited evidence suggested that clinic blood pressure measurement is less accurate than HBPM or ABPM when used to diagnose hypertension. The committee members acknowledged that these findings were in line with their clinical experience and agreed that clinic blood pressure measurement alone would not be an adequate method to diagnose hypertension.

The committee discussed repeat clinic blood pressure measurements when there is a difference in blood pressure between arms and noted that clinical practice varied. Based on their experience and knowledge, the committee members agreed that a cut-off of 15 mmHg would be more suitable than 20 mmHg, which was specified in the 2011 recommendations. This is in line with recent evidence that suggests a small difference in arm blood pressure is associated with an increased risk of cardiovascular events, possibly due to vascular damage.

ABPM correlates well with invasive blood pressure measurement and can identify both white-coat and masked hypertension. Based on the evidence in the previous guideline and the committee's experience and knowledge, it was agreed that ABPM remains the gold standard for the accurate measurement of blood pressure in primary care. ABPM has therefore been retained as the preferred method for the diagnosis of hypertension. In addition, economic evidence obtained by updating the health economic model for the 2011 guideline confirmed that ABPM is still likely to be the most cost-effective method for diagnosis, even with the inclusion of new data for improved accuracy of home and clinic measurement.

The evidence showed that validated HBPM is an accurate method of diagnosing hypertension for people in sinus rhythm. The committee's experience in clinical practice supported this, and the committee agreed that it is a suitable alternative when ABPM is unsuitable or not tolerated. The committee noted that the British and Irish Hypertension Society maintains a list of validated blood pressure devices for home use.

The evidence did not suggest that there were any benefits of adding telemonitoring to HBPM. Therefore, the committee agreed that it could not make a recommendation on telemonitoring for the diagnosis of hypertension.

How the recommendations might affect practice

The recommendations reinforce current good practice. However, the committee noted that implementation of the 2011 recommendations on ABPM has been challenging and that there is still variation in practice. A change in practice and additional resources and training will be needed in areas where there is currently no access to ABPM devices. However, ABPM was found to be the most cost-effective method of diagnosis, and it is anticipated that the long-term benefits of accurate diagnosis and treatment (such as avoiding over diagnosis and unnecessary treatment) will outweigh any initial costs.

Full details of the evidence and the committee's discussion are in evidence review A: diagnosis.

Return to recommendations

Relaxation therapies

Why the committee deleted the recommendation on relaxation therapies

The evidence on relaxation therapies was limited to a single small study. The study suggested some benefit in reducing angina and myocardial infarction, but it also suggested an increase in stroke. The committee agreed that the study was not adequate to assess the effectiveness of these therapies or to make a recommendation.

The 2011 guideline stated that relaxation therapies could reduce blood pressure, but it did not recommend their routine use in practice. The committee noted that this was based on evidence for reducing blood pressure only, and there was no evidence of a direct benefit to people with hypertension, such as improving quality of life or reducing cardiovascular events. The committee agreed there was insufficient evidence of benefit to recommend that people pursue this option themselves and agreed to remove this recommendation. It is not the intention of the committee to stop people from trying relaxation therapies if they wish to, but to make people aware that there is less evidence for benefit of this intervention compared with other lifestyle interventions or pharmacological treatment. The committee agreed that the clinical focus for non-pharmacological treatment of hypertension should be on encouraging people to make lifestyle changes, such as taking regular exercise and maintaining a healthy weight.

The committee agreed that further research would be useful to determine whether relaxation therapies are a clinically effective treatment for hypertension in terms of reducing cardiovascular events or improving quality of life (see research recommendations). They also noted that a larger study would be needed to obtain meaningful results.

How this might affect practice

Relaxation therapies were not recommended for routine use in the 2011 guideline, and they are not used in current practice for the management of primary hypertension in adults. The 2011 recommendation advised that people may try them as part of their treatment to reduce blood pressure, but committee consensus was that uptake has been low. Therefore, current practice will not be affected by the removal of the 2011 recommendation.

Full details of the evidence and the committee's discussion are in evidence review H: relaxation therapies.

Return to recommendations

Starting antihypertensive drug treatment

Recommendations 1.4.9 to 1.4.14

Why the committee made the recommendations

The evidence suggested that antihypertensive drug treatment was effective at reducing cardiovascular events in people with a clinic blood pressure of 160/100 mmHg or more (stage 2 hypertension).

A large study also suggested there was benefit of treating people with stage 1 hypertension. However, other studies in people with a low cardiovascular risk did not identify a benefit of treatment, and the committee agreed that the benefit of treatment across different cardiovascular risk groups was uncertain. The evidence was used to develop an economic model to compare the cost effectiveness of antihypertensive treatment with no treatment in people with stage 1 hypertension at different levels of cardiovascular risk. For people aged 60, the model showed that treatment was cost effective at a 10-year cardiovascular risk level of 10%, but there was some uncertainty at around 5% risk. Further analysis showed that it was cost effective to offer antihypertensive treatment to people aged 40 and 50 with stage 1 hypertension at a 5% risk and aged 70 and 75 at a 10% or 15% risk. QRISK was specified as the risk tool because it is recommended by NICE for risk calculation and most likely to be used in practice.

Taking into account the evidence and the results of the model, the committee were confident that people under 80 with stage 1 hypertension and a cardiovascular risk above 10% should have a discussion with their healthcare professional about starting antihypertensive treatment, alongside lifestyle changes, and that this would be a clinically and cost-effective use of NHS resources. The committee also agreed that antihypertensive treatment should be considered for people under 60 with a risk below 10%, with the degree of uncertainty in treating people at low risk reflected in the strength of the recommendation.

The committee members were mindful of the additional population that would be affected by lowering the threshold and aware that the decision to start drug treatment would depend on the person's preferences and their individual risk of cardiovascular disease. The recommendations highlight the importance of discussing the person's preferences for treatment and encouraging lifestyle changes.

Some studies investigated the benefits of treating hypertension in people with lower cardiovascular risk or people with blood pressure below 140/90 mmHg. However, some of these studies were not directly relevant because they included a high proportion of participants with chronic kidney disease and previous cardiovascular events. For this reason, several studies could not be used to inform the recommendations. For details of these studies see evidence review C: initiating treatment.

The committee discussed the lack of evidence to inform a threshold for starting treatment in people aged under 40. It was agreed that this is an important area for future research and the research recommendation was carried forward from the previous guideline (see research recommendations).

The committee agreed that there was no evidence to suggest that thresholds for starting treatment should be different in people with type 2 diabetes. The previous recommendations for people with type 2 diabetes (in NICE's guideline on type 2 diabetes in adults) suggested starting antihypertensive drug treatment if lifestyle interventions alone did not reduce blood pressure to below 140/80 mmHg or 130/80 mmHg in the presence of kidney, cerebrovascular or eye disease. However, this was based on evidence from 2 small studies in which the participants did not have hypertension. Further evidence for lower treatment thresholds in people with type 2 diabetes was limited within this review, with the committee aware of some evidence to suggest that lower blood pressure thresholds did not reduce the rate of cardiovascular events in people without additional risk factors. The committee therefore agreed that there was insufficient evidence to recommend a different threshold for starting treatment for this subgroup.

There was no evidence identified on thresholds for people aged over 80, and no prior recommendation for this age group with hypertension below stage 2, therefore the committee agreed that the threshold for starting treatment in people aged over 80 should be consistent with the target for treatment in this population (150/90 mmHg or lower).

The committee discussed the additional risks of starting treatment in older people, particularly those who are frail or have multiple comorbidities. Based on their expertise and experience, they agreed that the use of clinical judgement should be highlighted in decision making for people with frailty or multimorbidity, and that it should apply to people of any age. The committee agreed that a number of factors should be considered when discussing treatment options in this group and noted that healthcare professionals should refer to NICE's guideline on multimorbidity for further advice.

How the recommendations might affect practice

The recommendations will have a significant impact on practice because more people will now be eligible for treatment. It is difficult to predict the extent of the impact because there is variability in how the 2011 recommendation with a threshold of 20% is being implemented in practice. However, it is believed, based on some recently published UK data, that potentially around 50% of people with stage 1 hypertension and risk below 20% are already being treated with antihypertensive drugs (Sheppard et al. 2018).

People with stage 1 hypertension should already be monitored every year, but reducing the threshold will increase the number of people being prescribed antihypertensive drugs and increase staff time and consultations involved in starting and monitoring their drug treatment. However, there will be a reduction in cardiovascular events resulting in savings, although it is acknowledged that the costs and savings may fall in different sectors of the NHS.

Full details of the evidence and the committee's discussion are in evidence review C: initiating treatment.

Return to recommendations

Monitoring treatment and blood pressure targets

Recommendations 1.4.15 to 1.4.22

Monitoring treatment

Why the committee made the recommendations

The committee agreed that there was not enough evidence to strongly recommend HBPM for monitoring treatment in adults with hypertension. The evidence on monitoring was limited, with relatively small studies comparing different combinations of HBPM (with or without telemonitoring and with or without pharmacist input), pharmacy monitoring and clinic monitoring. It suggested that people had improved blood pressure control with HBPM with telemonitoring, with or without pharmacy input, compared with clinic monitoring, and the greatest blood pressure reduction was achieved with pharmacist input. However, the evidence was insufficient for the committee to make a recommendation.

The committee decided to retain the 2011 recommendation on using clinic blood pressure, but also agreed that the updated guideline should support home monitoring for people who wish to use it. The committee discussed the importance of patient choice and agreed that home monitoring should be an option, if it is suitable and the person is willing and motivated to use it. HBPM is already widely used in practice, especially for people with a white-coat effect. The committee agreed this would be reflected in the recommendation supported by the evidence and consensus opinion. Based on their experience, the committee agreed that training and advice would be needed for people using HBPM to ensure that people take measurements correctly and know when to contact their healthcare professional if they are not achieving their target blood pressure.

The 2011 guideline included a recommendation for further research for the best method of monitoring hypertension in people with atrial fibrillation. No evidence was identified in the updated reviews to inform recommendations for this group and therefore the committee agreed that this research recommendation should be retained to inform future updates of the guideline (see research recommendations).

The committee agreed they could not make a recommendation on telemonitoring because the evidence was not sufficient to show a clear benefit and the studies were inconsistent in the telemonitoring methods used.

How the recommendations might affect practice

The recommendations reflect current practice, so there should be no change in practice. They will encourage appropriate and suitable training to be given so that both people with hypertension and their healthcare professionals are confident that blood pressure is being measured properly using home monitoring devices.

Full details of the evidence and the committee's discussion are in evidence review B: monitoring the response to treatment.

Return to recommendations

Blood pressure targets

Why the committee made the recommendations

No evidence was identified to determine whether cardiovascular risk or blood pressure targets should be used. The committee agreed that in the absence of evidence the focus should be on blood pressure targets, based on their expertise and experience of current practice.

The evidence for blood pressure targets showed that there were both benefits and harms associated with a lower clinic systolic blood pressure target of 120 mmHg compared with 140 mmHg in people with primary hypertension without type 2 diabetes. Although the evidence suggested some benefit in reducing mortality and cardiovascular events, the lower blood pressure target was associated with a greater risk of harms, such as injury from falls and acute kidney injury. The committee agreed that the long-term implications of these adverse events were unclear and that further research is needed.

This evidence came from the SPRINT trial, which was a large study undertaken in the US. The committee discussed concerns about the population included in the study and the applicability to UK practice of the methods used. The study used automated blood pressure devices with a time delay and an isolated rest period, which is not common practice in the UK. The committee considered that the use of these devices would lead to lower blood pressure readings than in routine UK clinical practice. They also had concerns that some medicines were stopped when blood pressure targets were achieved, which may have had an impact on the results. The committee also discussed concerns about applicability of the population, for example, the participants had high cardiovascular risk levels including many with pre-existing cardiovascular disease or renal impairment and were already receiving treatment before the study started. These concerns made the evidence difficult to interpret and use to inform the recommendations. Further details of the committee's discussion of this study is included in evidence review D: targets.

Evidence from a smaller study also showed some benefit of lowering clinic systolic blood pressure targets to 130 mmHg. However, the committee noted that the study was based on people already receiving treatment and that it lacked information on adverse events.

The committee agreed that there was no evidence to suggest that blood pressure targets should be different in people with type 2 diabetes. Evidence for lower targets in people with type 2 diabetes was also limited, with some evidence to suggest that lower blood pressure targets did not reduce the rate of cardiovascular events. Previous recommendations for people with type 2 diabetes (in NICE's guideline on type 2 diabetes in adults) suggested a blood pressure target below 130/80 mmHg in the presence of target organ damage such as kidney, cerebrovascular or eye disease. The committee noted that the evidence behind this recommendation was based on 2 small studies in people without hypertension. They also had concerns about the relevance of the study design. The committee were also aware of trial data showing less benefit in populations with type 2 diabetes with fewer additional risk factors. The committee therefore agreed that there was insufficient evidence to recommend a different blood pressure target for this subgroup. It was noted that people with later-stage chronic kidney disease are covered by other NICE guidelines.

Overall, the committee agreed that the evidence was unclear and insufficient to determine whether a lower target would be beneficial and whether it would outweigh the associated harms. Therefore, the 2011 clinic blood pressure target of 140/90 mmHg for adults under 80 years was retained and applies to people with or without type 2 diabetes. The corresponding HBPM and ABPM targets were also retained at 135/85 mmHg. The recommendations emphasise the importance of achieving and maintaining a level consistently below the person's blood pressure target, whether this target be based on clinic, HBPM or ABPM.

Based on their experience, the committee members felt that people with postural hypotension are at risk of adverse events if a sitting or lying blood pressure is used for monitoring, because this measurement would overestimate daytime blood pressure and result in overtreatment. For example, a patient with a sitting systolic blood pressure of 140 mmHg might have a much lower blood pressure when standing and be at an increased risk of falls if treated based on their sitting blood pressure. The committee decided to recommend that 3 groups who are at risk of postural hypotension (people over 80 years, with type 2 diabetes and with symptoms of postural hypotension) should have their standing blood pressure measured, and their treatment modified accordingly if they have postural hypotension. The standing blood pressure should be used for future monitoring.

The committee noted that there was a lack of evidence for blood pressure targets in people aged over 80 years. Based on their experience the committee members agreed to retain the recommendation from the 2011 guideline, which was based on the only large, outcome-based randomised controlled trial in this age group. The committee also agreed that different blood pressure targets might be needed for people who are frail or have other conditions because they may have an increased risk of adverse events and less to gain from the long-term benefits of stricter targets. The committee decided it was not possible to define a blood pressure target for all possible clinical scenarios, and so recommended that clinical judgement should be used to agree an achievable target for each individual after a discussion about the possible risks and benefits. The committee agreed that further research in this area would be helpful and developed a research recommendation to inform future guidance for older people.

How the recommendations might affect practice

The recommendations should reinforce current good practice. However, the new recommendations place more emphasis on maintaining blood pressure consistently below the blood pressure targets. As a result this could lead to a higher use of antihypertensive drugs and an increase in consultations to maintain target blood pressure. For people with type 2 diabetes and target organ damage (not covered by other guidelines), the slightly higher target blood pressure compared to that recommended previously may reduce adverse events and may lead to fewer appointments and reduced drug use.

Full details of the evidence and the committee's discussion are in evidence review D: targets.

Return to recommendations

Step 1 treatment

Recommendations 1.4.29 to 1.4.37

Why the committee made the recommendations

The committee reviewed the evidence for starting treatment for primary hypertension with a single antihypertensive medicine compared with starting with 2 antihypertensive medicines at once (dual therapy). Additionally, the committee reviewed the evidence on whether specific subgroups of people with hypertension might benefit from starting on dual therapy, for example people with type 2 diabetes, older people, or those of particular family origins.

Some limited evidence from a single study showed that initial dual therapy may reduce cardiovascular events in people with hypertension and type 2 diabetes, but the committee members were disappointed that more comprehensive data were not available. The committee discussed the benefits of optimising treatment for hypertension early and agreed that this can substantially improve quality of life. However, there was not enough evidence to determine confidently the benefits or harms of starting treatment with dual therapy. In response to the lack of available evidence, the committee developed a research recommendation to determine if particular subgroups would benefit from starting dual therapy, to inform future guidance.

In the absence of compelling new evidence on step 1 dual therapy, the committee agreed that the previous recommendations for step 1 treatment should be retained (with minor changes for clarity), because they were based on robust clinical and cost-effectiveness evidence. One exception to this was the 2006 recommendation for considering beta-blockers in certain groups of younger people. The committee discussed this recommendation and agreed that beta-blockers are rarely used as step 1 antihypertensive treatment in current practice and there is no established relationship between beta-blocker use in primary hypertension and a reduction in cardiovascular events. For these reasons, the committee decided that the recommendation should not be retained. The committee noted that this is consistent with most international guidelines.

This update of the guideline also updates and replaces the section on blood pressure management from NICE's guideline on type 2 diabetes in adults. That guideline recommended that adults with type 2 diabetes of any age should start on an angiotensin converting enzyme (ACE) inhibitor as step 1 treatment (except women with a possibility of becoming pregnant and people of black African or African–Caribbean family origin). The committee discussed the evidence for this and agreed that it was sufficient to support and retain this recommendation. The committee agreed it should be broadened to include the choice of an ACE inhibitor or an angiotensin-2 receptor blocker (ARB; also referred to as A-type drugs), because they are now cost equivalent, and the committee also agreed they are clinically equivalent.

For people of black African or African–Caribbean family origin with type 2 diabetes, the previous recommendation was to offer step 1 dual therapy with an ACE inhibitor and either a diuretic (D-type drug) or a calcium channel blocker (CCB; C-type drug). However, these recommendations were based on monotherapy studies and when the committee looked at this evidence alongside the new dual therapy evidence review, they concluded that it was insufficient to recommend starting dual therapy in any subgroup of people with type 2 diabetes. The committee noted that people with type 2 diabetes who are older or are of black African or African–Caribbean family origin may not achieve their target blood pressure on ACE inhibitor or ARB monotherapy and may need to start step 2 drug therapy in the short term.

How the recommendations might affect practice

Overall, the recommendations for step 1 treatment reflect current practice for people who do not have type 2 diabetes. For people of black African or African–Caribbean family origin who have type 2 diabetes, the recommendation to start antihypertensive monotherapy rather than dual therapy may result in an extra clinical appointment if the dose needs to be adjusted. However, it may also reduce potential harms from initial overtreatment of blood pressure.

Full details of the evidence and the committee's discussion are in evidence review E: step 1 treatment.

Return to recommendations

Step 2 and 3 treatment

Recommendations 1.4.38 to 1.4.42

Why the committee made the recommendations

No evidence for step 2 or step 3 treatment was identified that was relevant to determining the best sequence for step 2 and step 3 antihypertensive treatment. Some of the studies available on drug treatments for hypertension were not included in this review because they were designed to inform step 1 treatment. Others did not reflect UK clinical practice. For details of these studies see evidence review F: step 2 and step 3 treatment.

Based on evidence from the previous version of the guideline and their clinical expertise, the committee members agreed to retain the same choice of drugs from the 2011 guideline, which reflect current best practice. The committee agreed that, in the absence of evidence of which treatment(s) are most effective for step 2 or step 3, the recommendation should be to offer any of these treatments based on an individualised approach informed by risks and benefits of each treatment and the person with hypertension's preference.

The committee noted that the changes to the step 1 recommendations for some people with type 2 diabetes do not necessitate a change in the step 2 recommendations since the same options for combination treatment at step 2 are available.

The committee agreed that the choice of drug should be discussed and agreed with the person, based on the person's step 1 treatment, the risks and benefits of each treatment option, and taking into account the person's preferences and other clinical factors. The updated recommendations reflect this, giving the choice of possible treatment options. A patient decision aid has been developed to support healthcare professionals and people with hypertension to discuss their treatment options and make informed decisions.

How the recommendations might affect practice

The recommendations are unlikely to alter current practice. The options for drug treatment remain the same and most step 2 or 3 treatment decisions are already based on an individualised approach.

Full details of the evidence and the committee's discussion are in evidence review F: step 2 and step 3 treatment.

Return to recommendations

Step 4 treatment

Recommendations 1.4.43 to 1.4.49

Why the committee made the recommendations

No evidence on step 4 treatment was identified that could be used to formulate new recommendations. However, the committee reviewed the 2011 recommendations and agreed that they should be retained and updated to reflect current best practice.

The committee discussed the importance of confirming resistant hypertension before starting step 4 treatment. Based on their clinical experience and knowledge of best current practice, the committee members agreed that a recommendation to highlight this would help prevent overtreatment and ensure that people receive the right care.

Despite the lack of evidence formally reviewed, the committee discussed the recommendation based on their clinical experience, taking the 2011 recommendations into account. The committee agreed that although the evidence for spironolactone did not meet the criteria for inclusion in the updated review for the guideline because the key study had a very short follow up and did not report any of the cardiovascular outcomes specified in this review protocol, the use of an aldosterone antagonist is now common clinical practice. Therefore, there was no reason to suggest that this recommendation should be changed.

In the 2011 guideline, high-dose thiazide diuretics were recommended as a potential step 4 treatment in people with high blood potassium levels. The committee felt that there was a lack of evidence for this approach and noted that the studies did not show an improvement in cardiovascular outcomes at higher doses, albeit in people without resistant hypertension. The committee agreed that the recommendation for considering alpha- or beta-blockers should be retained based on significant clinical experience of their safe and effective use and because adding a further drug is likely to have a greater effect on blood pressure than increasing the thiazide diuretic dose.

How the recommendations might affect practice

The recommendations represent current good practice and so should not change practice. High-dose thiazide diuretics are not commonly used as step 4 therapy and so removing this should not change practice.

There might be a small reduction in step 4 treatment with more thorough checks to confirm resistant hypertension. However, this may also result in an increase in blood pressure measurements to appropriately confirm resistant hypertension where this is not already being done.

Full details of the evidence and the committee's discussion, including information on key studies from the 2011 guideline that were not included in this update, are in evidence review G: step 4 treatment.

Return to recommendations

Identifying who to refer for same-day specialist review

Recommendations 1.5.1 to 1.5.2

Why the committee made the recommendations

There was no evidence identified to inform recommendations on this topic. The committee reviewed the 2011 recommendations and agreed that they should be updated by consensus based on their clinical expertise. In particular they agreed it would be helpful to clarify which features warranted same-day referral, which would need further investigation and when repeat blood pressure measurement should be taken.

The committee noted that it can be difficult to differentiate between accelerated hypertension and severe hypertension. They discussed the advantages and disadvantages of broader criteria for same-day referral, which would increase referrals to hospital but reduce the risk of missing people who need urgent treatment. The committee decided it would be beneficial to add some emergency symptoms to the existing recommendation, which will help healthcare professionals to decide when to refer.

Based on their experience, the committee members agreed that some people with severe hypertension could be receiving unnecessary treatment because the 2011 guideline recommended treatment based on severe hypertension alone. The committee agreed that this could be prevented if investigations for target organ damage were carried out quickly before offering treatment in people with severely raised blood pressure and no other symptoms of concern. The committee also agreed that checking blood pressure again within 7 days in people with no target organ damage would ensure that people with severe hypertension are followed up and offered suitable treatment.

The committee agreed that further research is needed in this area, particularly for people with extreme hypertension (220/120 mmHg or higher) or emergency symptoms. The committee members developed a research recommendation to help inform future recommendations on same-day specialist assessment.

How the recommendations might affect practice

The emergency symptoms listed in the recommendation may lead to more referrals to hospital. However, people with emergency symptoms will benefit from urgent treatment because accelerated hypertension can be fatal if untreated.

There may be some additional resource use from doing target organ damage tests more quickly and re-measuring blood pressure within 7 days. However, the number of people started on treatment immediately may be reduced because of undertaking investigations first.

The population with severe hypertension is very small, and the proportion with severe hypertension and additional symptoms that suggest accelerated hypertension is even smaller; therefore, resource impact is unlikely to be substantial.

Full details of the evidence and the committee's discussion are in evidence review I: same-day specialist review.

Return to recommendations

  • National Institute for Health and Care Excellence (NICE)