There’s been lots of coverage around NHS England’s announcement they will be restricting 17 procedures in order to cut costs.
Understandably, this has caused concern. Social media is full of people saying this is simply more Tory cost-cutting, these changes are ideologically motivated, or they will mean people won’t get the healthcare they need.
But how worried should you be about this announcement? Here’s a list of every option on NHS England’s list. They are proposing the first four move to ‘by-request only’ while the remaining 13 will have to meet certain criteria to be performed.
- Snoring surgery – a 2009 review by the National Institue for Health Research (NIHR), the largest clinical trial funder in the UK, found there was no difference between surgical and non-surgical methods to treat snoring.
- Dilatation and curretage (D&C) for heavy periods – NICE already recommends doctors don’t offer this procedure as there are better alternatives.
- Knee arthroscopies for osteoarthirtis – The Cochrane Collobaration are the finest systematic reviewers in the world, and look at every piece of research on a topic before making reccomendations about treatments. Two separate reviews suggest this offers zero benefit.
- Injections for back pain – a Cochrane review suggests no benefit, though there may be specific groups which responds to specific injection therapies.
- Breast reduction – the availability of this surgery already varies widely across the country. Some areas don’t fund it at all, and in the past, I’ve had to tell a woman she was unlikely to get surgery due to local criteria.
Most of the concerns I’ve seen online have been surrounding women whose breasts are causing them back pain. Only two randomly controlled trials have been conducted on breast reduction, but a Finnish review looking at these, and lesser quality trials, suggests there are significant quality of life improvements from surgery.
- Removal of benign skin lesions – this definition is very broad, and looking for evidence relating to surgery for every benign skin lesion would probably take me as long as my medical degree did. I’d guess NHS England is proposing to restrict these as doing so won’t kill anyone, or do much long term harm.
I’d expect lesions causing significant physical or mental difficulties (e.g. restricting eye opening, blocking a nostril, large and distressing facial lesions) to meet the criteria for treatment.
- Grommets for glue ear – a Cochrane review suggests very limited benefit to this procedure, but warns the research is low to very low quality. They also mention widespread pneumococcal vaccination may impact the success rate of grommets, and new and high-quality RCTs are needed to assess all whether grommets are useful.
- Tonsillectomy – while having your tonsils and/or adenouds out used to a bit of a rite of passage, this is no longer the case. A Cochrane review described tonsil removal as having ‘modest impact’, with the most severely affected children getting a small benefit, which had to be weighed up against a small but significant chance of major bleeding.
- Haemorrhoid surgery – while the non-surgical treatments of haemorrhoids (e.g. banding, injection therapy, infrared therapy) have higher recurrence, they treat most mild cases sufficiently according to a 2015 review. I struggled to find research directly comparing surgical and non-surgical methods. Similarly to the treatment of heavy periods, I’d expect surgical treatments to remain as a second line therapy once less invasive alternatives have been tried, due to their reduced costs and complications.
- Hysterectomy for heavy periods – as with dilation and curretage, NICE already says this shouldn’t be used as first line treatment, and only used as second line in certain circumstances.
- Removal of chalazia (a specific type of benign lesion on the eyelid) – up to 80% resolve by themselves with conservative management. There is very limited evidence looking at their management, but I imagine this would be treated similarly to benign skin lesions – if they cause problems (e.g. infection, visual disturbance), then they qualify for surgery.
- Removal of bone spurs for shoulder pain – a Lancet review found no clinically significant differences whether bone spurs were removed or not
- Carpal tunnel syndrome (CTS) release – a review from the American Association of Orthopaedic Surgeons state surgery and steroid injections/splinting give goood benefits in CTS, but surgery offered better outcomes at 12 months. Currently, NICE recommend trying conservative management for 6 weeks before referring for surgery. Severe CTS (e.g. paralysis, gangrene) should be managed much more urgently, with emergency surgery a possibility.
- Dupuytren’s contacture release – Dupuytren’s is a condition where the fingers become permanently bent towards the palm. Surgery is very effective, with debate among exactly which approach is best. Unfortunately, the condition tends to recur despite surgery.
- Ganglion excision – ganglions are benign rubbery lumps found most commonly on the back of the wrist. They are also known as Bible Bumps, watch this video to see why:
The evidence surrounding surgery is mixed. This 2014 review is very thorough, and essentially boils down to: surgery can be very good, but it’s necessity is questionable. It seems likely gangions are being included for similar reasons to benign skin lesions and chalazion.
- Trigger finger release – a Cochrane review from Feberuary 2018 states they can’t be sure surgery is superior to steroids in the long-term, and has increased pain. They are uncertain about the risk of adverse events and nerve or blood vessel injury. There are no studies reporting hand function or patient satsifaction. A mixed bad to say the least.
- Varicose vein surgery – a 2017 review suggests similar outcomes from surgery, and non-surgical therapies.
Out of the 17 procedures listed:
- one (breast reduction surgery) seems to be completely inappropriately listed, with the only good review I could find supporting its use
- one (Dupuytren’s release) where the procedure is very good, but almost inevitably needs repeating. There is likely to be debate about when it is economical for the NHS to pay for the treatment.
- three (ganglion, chalazia and benign skin lesion removals) seem likely to be included because they are rarely serious enough to require treatment
- five (snoring surgery, knee arthroscopies for osteoarthritis, injections for back pain, shoulder bone spur removal, and varicose vein surgery) have good quality evidence showing no benefits.
- five (D&C for heavy periods, grommets, tonsillectomy, hysterectomy for heavy periods, haemorrhoid surgery, CTS release, and trigger finger release) either have very mixed evidence, or there’s evidence a conservative approach prior to surgery is best.
Looking at the whole picture, it’s clear most of these procedures either have no benefit, or have conservative alternatives worth trying first. All treatments have risks, and doctors always have to weigh up whether those risks are worth the benefits the treatment brings.
If there are no benefits, it’s obvious the risks aren’t worth it. If there are conservative approaches which are just as, or nearly as good, then it’s usually worth going for them first. By my reckoning, 12 of the conditions fit these descriptions.
The five remaining ones seem to be listed on mostly economic grounds.
Dupuytren’s surgery is very effective, but given recurrence is very likely, it seems sensible to wait until its definitely necessary before operating. From a patient’s point of view, repeated operations are annoying, though not as annoying as reduced hand function. From a much colder economic view, when is it worth paying for an operation we’ll possibly need to repeat? Personally, I’d say ‘most of the time’ but I can see why those controlling the NHS’ budget would want to get a more specific answer.
The final three (removal of ganglions, chalazion, and benign skin lesions) raise an interesting question. These conditions rarely cause serious harm, and rarely cause any symptoms. Obviously if they cause either, we should be treating them, but what if they aren’t? Should the NHS be treating things which aren’t causing problems? Clearly, NHS England think the answer is yes.
So overall, is this a cost-cutting exercise? Yes. But it’s also an exercise in clinical improvement. Five of these procedures have no reason to be offered as they offer no benefit. Seven have conservative alternatives worth trying first. Three are things you can easily argue don’t really even need treating at all.
Only two of the procedures seem like they’ve been listed just for economic reasons – Dupuytren’s and breast reduction. Restricting either seems puzzling, but especially so in the case of the latter. Breast reduction is already restricted in many areas, and it’s hard to see what further restrictions can occur without causing harm to patients. I’ll attempt to keep an eye out for NHS England’s consultation on these procedures, and write something about their final decisions.
This is the first blog post I’ve written in 18 months. Constructive feedback very much welcomed.