This site is intended for health professionals only

How we diagnose sleep apnoea in primary care

Before we introduced our new service, patients with suspected obstructive sleep apnoea (OSA) were referred to the respiratory support and sleep centre at our secondary care provider, Papworth Hospital NHS Foundation Trust near Cambridge, and could wait up to a month to be seen. Some patients were not referred at all and others were referred inappropriately.

We did have a conversation with secondary care about our plans, and they weren’t too keen on it initially.

It was a kind of ‘specialist’ thing. They argued that you shouldn’t have access to the sleep clinic directly from primary care and had concerns around what would happen if GPs missed things.

But with NICE figures suggesting there are a million undiagnosed cases of OSA, I thought it had to happen – untreated OSA can increase the risk of hypertension, heart attack, stroke, obesity and type 2 diabetes. Under PBC I was clinical lead for cardiology, and so I have a long-standing interest in the subject area. I am also passionate about bringing care and management of conditions closer to the patient.

What we did

One surgery in the commissioning group purchased the same model of overnight pulse oximeter as is used at the sleep clinic at Papworth, as well as the software used for interpretation of the results.

The oximeter was bought in 2009 from existing PBC ‘savings’. Back then, we had notional budgets but could make savings year on year. If we wanted to access those savings we had to put in a business plan to NHS Cambridgeshire, which we did – and it was approved.

We had a patient, who used to be a practice manager, liaise with Papworth about sourcing the machine. The oximeter cost about £430 and the software cost £405.

We start with a patient with suspected OSA being assessed by their GP.

If the assessment indicates OSA, we then run through the Epworth Sleepiness Scale questionnaire – a standard diagnostic tool for sleep disorders which is available online.

Higher Epworth scores distinguish patients with primary snoring from those with OSA and increase with severity of the condition.

Once the Epworth questionnaire is completed – and depending on the results – the oximeter can usually be loaned out to the patient straight away.

The patient returns the machine the next day and the data is transferred to the practice computer via USB, after which a report is available to be obtained and interpreted by the GP. We refer to this three-part procedure as the ‘triple lock’. Only confirmed cases after the triple lock are referred to the sleep clinic for treatment. The three-part assessment is effectively what previously took place across primary and secondary care – but now all done by the GP. Just because you have a fancy piece of kit doesn’t mean you don’t have to go through all the rest of the assessment.

The questionnaire is the same one as was used in secondary care, so the extra element is the GP’s clinical assessment – the application of GP common sense, if you like.

Generating the report is quite straightforward. It only takes a minute to plug it into my computer. And it makes it a simple matter to assess – basically, if the green line drops below the red line, I refer.

The service is currently just for patients at my practice – Alconbury and Brampton surgeries – but there are negotiations going on to extend it to other local practices and involve Papworth.

Secondary care now feel the change is welcome. A few times I’ve referred and they haven’t even bothered to repeat the test, which I take as a good sign.


A total of 47 patients have used the service since its inception in July 2009. All were assessed within two working days – compared with waits of up to a month to be seen at the sleep centre.

Of the 47 patients assessed, 12 had suspected OSA and were referred to the sleep clinic. This equates to about 25%, compared with 100% previously. The other 35 – some 75% – were not referred.

Of the 12 who were referred, 11 had the diagnosis confirmed while the other patient was diagnosed as having restless leg syndrome – so it was still an appropriate referral to the sleep centre.

In summary, numbers of referrals to the sleep clinic have fallen markedly and are now more appropriate. The results also demonstrate that the service is beneficial for the local health economy.

A first appointment at the sleep clinic costs £160, so the 35 patients who were not referred saved a total of £5,600 over the two years. Other practices in our commissioning group are now interested in adopting the service and the savings are set to increase.

Patients have also benefited through the rapid access to diagnostics. As the oximeter is at their local GP surgery, patients do not have to travel to get it fitted.

Previously, they faced a round trip of over 20 miles. It is also fitted by a familiar person – their own GP or a local healthcare assistant.

The service has been extremely well received by patients. We’ve had no negative feedback at all. Comments have included: ‘An excellent service,’ and ‘What a good idea!’.

I believe the positive reaction has been helped by involving patients in the concept from the very beginning.

The idea was first floated in July 2008 during the annual general meeting of our patient participation group, where it received overwhelming support.

The service has also meant better awareness of the condition among both patients and staff. By raising its profile and improving access to diagnosis and

eventual treatment, we can improve outcomes in those conditions and help ameliorate risk.

The Future

Some local practices are already interested in taking part in an extended version of the service, and we’re at the early stages of negotiation for that. Papworth are interested in helping us extend the service too – in fact, it has realised that savings can also be made in its QIPP plan, so added the service into that. We had our first teleconference about this last month.

It’s basically just a question of logistics and how we organise it. One potential way under consideration would be to adopt a hub-and-spoke model with the hospital at the centre.

In a rural area like ours we might have to locate the service in market towns – and perhaps have a machine in each town.

But because they’re quite cheap – and depending on what budget we have – we might even be able to have one machine in each practice.

It could easily be adopted elsewhere too. Three practices from across the country have already contacted us about doing the same thing in their localities – one practice in London, one in the North West and one in the Peterborough area.

Dr Malav Bhimpuria is chair of Hunts Care Partners and a GP in Huntingdon, Cambridgeshire.