Sunday, 3 December 2017

NHS Homeopathy #3 - Analysis GP Prescribing data

A previous post discussed the now closed Items which should not be routinely prescribed in primary care: A Consultation on guidance for CCGs, the petition created by supporters of homeopathy and how wrong the supporters of homeopathy were about the Consultation.

It also discussed some of the workings of the NHS.

Decline in Prescribing
One thing that it very clear from even the most cursory analysis of NHS England GP prescribing data is that the prescribing of homeopathic medicines has declined enormously. This article from the Nightingale Collaboration clearly demonstrates this. There seem little point in duplicating its content - please do read it.

What the article does not explain is why there has been such a dramatic decline. Unfortunately, this is an impossible question to answer. It is possible to speculate on potential reasons and they fall into a number categories -

  1. Decreased Patient Demand - It has been suggested that to some degree prescribing of homeopathic medicines is patient driven ie patients specifically ask for them. The numbers of patients asking could have declined. This could be due to demographic change or it could be due to changes in public attitudes. Although there seems to have been a drop in the number of lay homeopaths in recent years, this is not a good proxy for patient demand for homeopathy.
  2. Typical GP Behaviour - There's an acronym that used to appear on GP notes - TEETH - "tried everything else, try homeopathy". How widespread this attitude was in the past is unclear. Typical GP attitudes may have changed. The prescribing of placebo is seen by some as unethical. In recent years opponents of homeopathy within the medical profession have become increasingly vocal. As discussed in the previous post, CCGs have increasingly recommended not prescribing homeopathic medicines but it is difficult to guage what impact this has had directly - also the decline started way before the re-organisation that created CCGs. It is know that there were locally argeed forumlaries in some areas prior to this but they are difficult to get hold of.
  3. True Believers - GPs who are enthusiasts for homeopathy are not exactly common. Various numbers have been suggested for the number of GPs with "training" in homeopathy but it is impossible to determine how many. It may well be the case that there are fewer of them now. If the historical high levels were associated mainly with prescribing by "true believers", as they leave the NHS GP practice (retire or move into private practice) and fewer younger GPs are becoming "true believers", this could well account for the decline in prescribing. It is worth pointing out that no NHS GP could get away with 100% prescribing of homeopathic medicines. That's only possible in the private sector and even it would carry considerable risks. Although CCGs have no power to prevent such a thing, peer pressure and patient disatisfaction would make it impossible. Even "true believers" may have changed their behaviour as a result.
  4. Economics - As mentioned in the previous post, it makes economic sense for many patients and NHS England that GPs suggest to patients which homeopathic medicine to buy themselves rather than write a prescription. Changes in prescription charges vs cost of homeopathic medicines could drive GPs to increasing to suggest rather that prescribe. It is also the case that GP perception of increased financial pressures on NHS England could drive such a change. If this were the case the ratio of prescriptions for patients exempt from charges should have changed over time,
Even if were possible to obtain data to confirm or refute these speculations, it would not be an easy or cheap process. Given that the cost of GP prescribing of homeopathic medicines is so small and still seems to be decline there seems little point in doing so. It could cost NHS England more to do the analysis that the actual cost of the products!

Demographic Analysis
Along side the Items which should not be routinely prescribed in primary care: A Consultation on guidance for CCGs, NHS England also published an equality analysis. Much of it is irrelevant to homeopathic medicines but contains some key demographic details.

In 2016, there were 1541 (63.2%) prescriptions dispensed to female patients and 889 (36.8%) to male patients. It is not number of unique patients, so it is not possible to derive an average number of prescriptions, prescription items or cost per patient. That would have provided an insight into the type of conditions (acute vs chronic) being prescribed for.

In terms of age bands - 

Age BandNumber of PrescriptionsPercentage of Total
Under 1835914.71%
18 to 3027311.19%
31 to 4438615.82%
45 to 6463526.02%
65 and over78732.25%

It is worth noting that those under 16, between 16 and 18 in full time education and those over 60 are automatically exempt from prescription charges.

It would be useful to have had this information over time but...

In depth (but quick and dirty) GP Prescribing data Analysis
It is possible to download NHS England GP prescribing data. Data is released on a monthly basis. Unfortunately, the files generated are very law. Typically, they contain 4 million rows of data. It would a mammoth task to download the available historical data and the data would have to imported into a proper database. This would be beyond the casual user.

Fortunately, the OpenPrescribing initiative has done that hard work all ready. It provides a simple to use tool to explore the data in a number of ways. It can drill down from CCG level to GP practice level. Also, it's possible to download the data to to manipulate in ways that they tool can't do, although it is tricky to deal with. This has been done. The resulting analysis is very interesting and seems to disprove some of the contentions of UK homeopathy groups as well as misreporting by the media.

As OpenPrescribing plan to publish a more academic analysis, this analysis will be somewhat cursory.

Some of the terms used require some explanation. Prescription is obvious. Items is the effectively the number of medicines ordered on a prescription. It is surprising that more than one homeopathic medicine is sometimes prescribed - this goes against the tenets of classical homeopathy. Actual cost is the cost to NHS England to re-imburse community pharmacies for the cost of medicines. It does not include the £1.25 per item dispensing fee or the potential £20 "specials" fee. It's not possible to determine the value of these fees but they could be considerable. 

Looking at CCG level, for the period 2012-07 to 2016-06 (neatly 5 years), those CCGs with more than 1% of the total spend are -

Total Result £556,994.37 Percentage
NHS BRISTOL CCG £93,603.72 16.81%
NHS WEST KENT CCG £44,747.14 8.03%
NHS LIVERPOOL CCG £34,222.06 6.14%
NHS HERTS VALLEYS CCG £14,570.72 2.62%
NHS SHEFFIELD CCG £10,649.43 1.91%
NHS DORSET CCG £10,269.83 1.84%
NHS SOMERSET CCG £9,846.58 1.77%
NHS HARINGEY CCG £8,896.19 1.60%
NHS LEWISHAM CCG £7,455.12 1.34%
NHS KERNOW CCG £6,773.46 1.22%
NHS WILTSHIRE CCG £6,618.03 1.19%
NHS SHROPSHIRE CCG £6,457.52 1.16%

The OpenPrescribing tool did not like all of those CCGs selected - CCGs up to Lewisham were input. Unfortunately, when drilling down to Practice level, OpenPrescribing data downloads do not provide CCG. Not that this matters.

Below is a list of those Practices within the selected list of CCGs. Only those with spend over £3k in the period are are shown.

BLACKTHORN £30,653.53

Helios Medical Centre makes up for 12% of total spend in the period, which is extraordinary. It falls under Bristol CCG whose local formulary does not include homeopathic medicines. Bristol CCG will be well aware of the prescribing via reporting. Although the data does not drill down to which GP is prescribing, the staff list suggests that Dr Frank Mulder and Dr Joan Platford could be responsible for the majority of the prescriptions. Both have links to Anthrosophy and anthroposophic medicine. There would appear to be no or little change in the value of prescriptions over time.

St. Luke's Medical Centre no longer exists. It was forced to close due inability to recruit staff. It closed back in July 2015. If it had not, its spend would have easily exceeded Helios. Although it did close, it re-opened in another guise St. Lukes Therapy Centre. Stroud is a hotbed of Anthroposophy and so is the Centre. Whilst it is true that many practices are having difficulties recruiting GPs, there is a question of whether GPs would be comfortable working with anthroposophic enthusiasts.

The Belmont Surgery is now Wadhurst Medical Group. Its staff list turns up Dr. Andrew Sikorski. Sikorsi's LinkedIn profile shows him to be an integrative medical consultant. He is a GP, not a Consultant according to the GMC register. Skorski's "treatments" include Psionic Therapy whatever that is. Sikorski is quite well known for his pro-homeopathy comments in various places on the internet. The value of prescriptions for homeopathic medicines has fallen over time though. Back in 2012 andd 2013, values of over £1,000 per month were routinely seen. In 2017, £100 would be more typical. The reasons for the decline are unclear.

The Blackthorn Medical Centre has links to the Blackthorn Trust which offers anthroposophic medicine. This page on the practice website states that the Trust certainly was funded by the NHS at one point. Whether this is still the case is another matter. The staff list when compared to this newletter from the Blackthorn Trust suggests that both Dr Martin Gerhards and Dr Saskia Renkema are involved in anthroposophic medicine. The newsletter also sheds some light on prior funding of the Blackthorn Trust by the NHS. Similar to the Wadhurst Medical Group, there seems to have been a decline the value of prescriptions. 

Islington House Medical Centre is in Liverpool.It is unusual in that it was set up in 1937 and would appear to have been in the hands of the same family since then. The prescribing patterns seem very strange. Back in 2012 and 2013, typical prescribing value was roughly £180. This has declined In 2017, there were months when there was no prescribing of homeopathic medicines. But then there are these outliers -

date name items actual_cost Average item
2016-05-01 ISLINGTON HOUSE MEDICAL CENTRE 17 2821.52 165.97
2016-06-01 ISLINGTON HOUSE MEDICAL CENTRE 12 3003.42 250.29
2016-09-01 ISLINGTON HOUSE MEDICAL CENTRE 4 2709.86 677.47
2016-12-01 ISLINGTON HOUSE MEDICAL CENTRE 4 2709.74 677.44

There is no immediate explanation for this. There is no known homeopathic medicine that costs £165 a go, let alone £677. Miscoding could be one explanation. Liverpool CCG will be approached to see if they can shed any light on this.

Bedminster Family Practice used to list homeopathy as one of the services provided. Dr Nick Wilson ran a clinic according to older documents. However, it is no longer listed on their Clinics and Services page. The prescribing history is similar to Islingthon House in that spend was generally a low, showing decline but certain months were outliers, but more so.

date name items actual_cost Average item
2015-08-01 BEDMINSTER FAMILY PRACTICE 4 1,120.94 280.24
2016-01-01 BEDMINSTER FAMILY PRACTICE 6 1,591.95 265.33
2016-03-01 BEDMINSTER FAMILY PRACTICE 2 543.50 271.75
2016-04-01 BEDMINSTER FAMILY PRACTICE 2 998.89 499.45
2016-08-01 BEDMINSTER FAMILY PRACTICE 2 566.70 283.35
2016-09-01 BEDMINSTER FAMILY PRACTICE 5 1,689.73 337.95
2017-01-01 BEDMINSTER FAMILY PRACTICE 4 1,949.37 487.34
2017-03-01 BEDMINSTER FAMILY PRACTICE 2 594.40 297.20
2017-04-01 BEDMINSTER FAMILY PRACTICE 2 1,090.08 545.04
2017-05-01 BEDMINSTER FAMILY PRACTICE 2 594.25 297.12
2017-09-01 BEDMINSTER FAMILY PRACTICE 3 1,852.42 617.47

Again, there are questions about the average value of the items prescribed. Bristol CCG will be contacted.

One GP practice of interest is not in this list. One of the partners at Plas Meddyg Surgery is Dr Ralf Schmalhorst is a member of the Council of the Faculty of Homeopathy. One of the reasons the practice is not on the list is that the spend is trivial. However this may be explained by Dr Schmalhorst offering homeopathy on a private basis.

It is very difficult to draw any firm conclusions from the data and investigations except that -

  • GPs who are interested in Anthroposophical Medicine are amongst the highest prescribers of homeopathy
  • There are some very strange high item value

1 comment:

  1. I've 'played' with NHS prescribing data before now. And I've found miscoding before now. Decimal point in the wrong place usually. But I've also been operating at the specific product level where the proper price can be seen in prescriptions from other practices.