Correspondence with the National Institute for Health and Clinical Excellence

This page lists correspondence with the National Institute for Health and Clinical Excellence. On 3 December 2012 BMJ editor in chief Dr Fiona Godlee wrote to NICE's chair, Professor Sir Michael Rawlins. In her letter she says: "Now that serious doubts have been raised about the evidence behind claims for oseltamivir’s effectiveness and safety, I am asking you to withdraw approval for oseltamivir until NICE has received and reviewed the full clinical trial data and those anonymised data are available for independent scrutiny."


All rapid responses

Rapid responses are electronic letters to the editor. They enable our users to debate issues raised in articles published on Although a selection of rapid responses will be included as edited readers' letters in the weekly print issue of the BMJ, their first appearance online means that they are published articles. If you need the url (web address) of an individual response, perhaps for citation purposes, simply click on the response headline and copy the url from the browser window.

Go slow: exceed your limits, but heed your limitations and need for patience.

Competing interests: No competing interests

18 December 2014
Hugh Mann
New York, USA
Click to like:

Dear Editor,

It was with interest that I read this RCT comparing atraumatic and cutting needles, as part of a patient safety project I undertook. Spinal needles fall into 2 categories - atraumatic with blunt tips that part dural fibres, and cutting needles. There is evidence to suggest that the use of cutting needles is associated with an increased incidence of post dural puncture headaches (PDPH) (1&2). This RCT found a reduced incidence of headache using atraumatic needles (3). However cutting needles are still used as standard to perform diagnostic lumbar puncture in our institution.

As part of a service quality improvement project we reviewed our diagnostic lumbar puncture practise between October 2012 and May 2013. Patients were identified by CSF samples sent to the lab and as a consequence 33 sets of notes were reviewed. In addition we identified all epidural blood patches performed over the last 3 years. We collected denominator data for spinals, epidurals and lumber punctures over the same period to define blood patch intervention rate for each procedure.

22G Quincke (cutting) needles were used on the wards, although needle type was not always recorded. Aseptic technique was varied and poorly documented. Overall no immediate complications in this small group were reported. Epidural blood patch rates post spinal anaesthetic (where 25G atraumatic needles are used as standard) were found to be around 1 in 1000 compared with around 1 in 100 post diagnostic lumbar puncture.

Clearly there may be confounding factors within the lumbar puncture group, however, there is a significant difference in requirement for blood patch. We have introduced a program of education using group discussion and a mannequin based skills training. In addition we are introducing standard packs for spinals with atraumatic needles, a change we feel is long overdue.

Dr Ilana Delroy-Buelles
ST6 Anaesthetics

Dr R Green
Consultant Anaesthetist

1) S.Halpern and R.Preston. Postdural Puncture Headache and Spinal Needle Design. Metaanalyses. Anaesthesiology 1994. 81:1376-1383
2) D.K.Turnbull and D.B.Shepherd. Post-dural puncture headache: pathogenesis, prevention and treatment. BJA 2003. 91(5): 718-29
3) S.R.Thomas, D.R.S.Jamieson and K.W.Muir. Randomised controlled trial of atraumatic versus standard needles for diagnostic lumbar puncture. BMJ 2000. Volume 321:986-90

Competing interests: No competing interests

18 December 2014
Ilana Delroy-Buelles
ST6 Anaesthetics
Dr R Green, Consultant Anaesthetist
Click to like:

Expanding globesity continues to present many novel challenges to society, from combusting crematoria (1) to wider trolleys and longer endoscopes in hospitals (2). As we approach New Year, with its cyclically futile outbreak of good intentions, we wondered if the article in the Xmas BMJ (3) might unwittingly provide yet another excuse to defer impending dietary restraint - this time on account of saving the planet?

Altruistic adiposity by selflessly putting one's own health at risk to save countless others from the ravages of climatic, not calorific, tsunamis. After all, if, by losing 10 Kg of fat, this then led to the liberation of 8.4 Kg of carbon dioxide (volatile carbon as CO2.), with its attendant environmental impact, would we not all be much better off by remaining somewhat plumper but cooler?

To test this hypothesis we used contemporary UK data for CO2 production from fossil fuels, from the National Health Survey (from PHE (4)), and from the Office for National Statistics. If we now make a series of approximations and assumptions, as follows

(a) 83% of UK population is >15 years of age which is 52.3 million people.
(b) For every 10 Kg of fat lost, 8.4 kg (84%) is volatile carbon as CO2. See (3).
(c) The average man in England in 2013 was 5ft 9in (175.3cm) tall and weighed 13.2 stone (83.6kg). BMI of 27.3 but their ideal weight was 75 Kg (BMI 24.5). Average weight loss needed therefore is 8.6 Kg per man.
(d) The average woman in England weighed 11 stone (70.2kg) and was 5ft 3in tall (161.6cm). BMI 27.Ideal weight 64 Kg (BMI 24.5). Average weight loss needed 6.2 Kg per woman.
(e) Weight loss down to ideal weight over 12 months (at a very modest monthly rate, and perfectly achievable)
(f) 50% men and 50% women
(g) A grand national total of 387 million Kg of fat lost over the year.
(h) Which equates to 325 million Kg of volatile CO2 generated - this is 0.325 million metric tonnes of CO2
(i) Total carbon fuel burning-related emissions of volatile CO2 in 2013 was 464.3 million metric tonnes of CO2
(j) So the additional amount of volatile CO2 produced in this really most unlikely "mass contraction and shrinkage" annus mirabilis would be just an additional 0.07% of the carbon fuel contribution to CO2 emissions in that same year.

It seems to us therefore that the excuse of remaining portly as a noble gesture to help protect the planet is just plain old hot air. Thinner waistlines will not lead to more wastelands. Tightening the belt will not destroy the veld. So, come January 1st, let's all now plan to get into some serious fat loss secure in the knowledge that we can help ourselves without harming others. Just one more wafer-thin mince pie........

David Goldsmith and Eric Heymann


(1) Last accessed 18.12.14

(2) endocrine-today/%7B39256858-e05c-4804-90ba-b5318394a4be%7D/health-care-community-responding-to-obesity-epidemic. Last accessed 18.12.14

(3) BMJ 2014;349:g7257

(4) _ for_england. Last accessed 18.12.14

Competing interests: No competing interests

18 December 2014
David J Goldsmith
Eric P Heymann
Guy's and St Thomas' Hospitals London
Guy's and St Thomas' Hospitals London
Click to like:

Acronyms are where native speakers of Greek and Latin languages take their revenge on the English speakers. Many nice sounding acronyms - used for studies or scientific organizations - have weird meanings in older languages. I can recall studies like VIGOR (in Latin, liveliness) Vioxx Gastrointestinal Outcomes Research; MINGO (in Latin, doing a pee) Medical Interpreter Network of Georgia and Oregon; EMET (in Latin, vomit) Electro-Mechanical Engineering Technology; and ADE (Latin God of Death) Adverse Drug Events.

I enjoyed the article and the idea.

Competing interests: No competing interests

18 December 2014
Marica Ferri
not relevant in this case
Praça Europa 1 - Lisbon
Click to like:

Lewis has confirmed my assertion that it is very difficult to question or remove a diagnosis. The science of undiagnosis is under-developed and a subject much in need of further study1 Is a mantra like ‘once a diabetic always a diabetic’ a useful axiom for doctors to live by, or evidence of a closed mind?

I have had three patients in recent years who have all fulfilled WHO criteria for the diagnosis in terms of two fasting blood sugars > 7.0 mmol/l. Despite this, all three were of a normal weight and subsequent HbA1c levels were never > 40mmol/mol over several years and without any medication. What are we to make of such a situation? The diagnoses were not made erroneously, yet the passage of time has surely proven them to be wrong. We must remember that our diagnostic criteria are an arbitrary distinction between what is normal and what is not, meaning that some people will be wrongly classified as diabetic – and that the tighter we make these criteria the more such false positives will result.

With regard to all the undiagnosed patients with dementia, we should certainly be mindful of the need to identify people who are unsupported due to the lack of a diagnosis, but we should question the scientific basis of such figures as I have argued in this piece and elsewhere.2 Alistair Burns is not an independent figure since the Government has given him the task of raising diagnosis rates prior to the general election – arguably a more significant conflict of interests than any link with industry. The need to achieve this aim was behind the ill-thought out £55 incentive scheme.

As I wrote recently3, I believe the best way to improve the care for those with dementia – both the diagnosed and the undiagnosed is not to set targets for diagnosis, but to improve the care of those affected by dementia so that GPs have a truly effective service to offer to our patients and their families.

1. Coebergh, J. a, Wren, D. R. & Mumford, C. J. “Undiagnosing” neurological disease: how to do it, and when not to. Pract. Neurol. 14, 436–9 (2014).

2. Brunet, M. Dementia diagnosis targets: a problem of scale? The Guardian (2014). at

3. Brunet, M. Dementia: I don’t want a £55 “bribe” to diagnose patients. The Telegraph (2014). at

Competing interests: I am the author of the article and my interests have been published with the article.

18 December 2014
Martin D Brunet
General Practitioner
Binscombe Medical Centre
106 Binscombe, Godalming, GU7 3PR
Click to like:

Coursing along for hundreds of miles, the mountains called the Himalayas were formed millions of years ago. The name means “house of snow.” These highlands rose into shape when a satellite land fused into the continental mass of Asia. Originally an island out in the ocean, the Plate of India gravitated towards the Eurasian Plate of Asia over countless millenia. Ten million years ago, the Indian and Eurasian Plates collided to raise the highest ridge on the planetary surface – a crenellated two thousand miles of the Himalayas.

The mountains are therefore a visible line of fusion between the Indian subcontinent and the supracontinent of Asia. A pushing together of tectonic plates means that the peaks grow higher by an inch per year, the geophysics seen as tremors on the scratching graphs of the seismologists. On the foothills of the Himalayas grow cherries and walnuts and almonds. And amid the foliage may be found the jewelled eyes of the snow leopard, or the blunt movements of the isabelline bear.

For millenia, the wall-like mountainous range curbed the migration of humans from west to east and vice versa. West of the Himalayas is the top of India which – as a road into the subcontinent below – is a whirl of genealogical diversity with fibres that extend to Europe. Kings such as Alexander the Great galloped their horsemen across from the Mediterranean into the uppermost reaches of India, but were unwilling to climb the Himalayan barricades to enter the rice-growing mists of the Orient. The legacy of the invasions from the west is seen in the physicality of the peoples of north India, the east-west affinity also apparent in what is the Indo-European language system.

In the heart of the Himalayas is the nation of Nepal. To the east of these mountains is the region of Tibet which, until the 1950s, was a separate country managed by its own government. All this changed in 1950 when China rolled its tanks into Tibet on the assertion that the area once belonged to the dynasties of China. By dint of the size of the Chinese army the minority people of Tibet were sitting ducks in the encounter, and their province was annexed with ease.

Ethnologically, the Tibetans are a people in their own right because they have their own language. For a thousand years, the community has practised its own form of Buddhism under the spiritual leadership of a Dalai Lama (Ocean of Wisdom). The Dalai Lama is like the Pope in Roman Catholicism. Born in 1935, the current incumbent in the chair of the Dalai Lama is eighty years of age, and as a modern personality is a familiar presence in the media of the West, not least because of the conflict in his homeland.

With Tibet overrun by the territorial ambitions of China, in 1959 the Dalai Lama and his people could only flee westward into India. They were met by a sympathetic nation which established a sanctuary on the Himalayan foothills for the head monk and a government-in-exile. Introduced to his role in the age of Tibeto-Chinese strife, the Dalai Lama has lived most of his eighty years in exile. A populous community of Tibetan refugees is also quartered in the south Indian state of Karnataka.

In the Chinese raids, hundreds of Tibetan monasteries were smashed and desecrated, the military transgression leading to mass slaughter and human rights abuses which continue today. Symptomatic of these currents is that from 2009 onwards, Tibetan monks, attired identifiably in bindings of red, have set themselves alight in public and burnt to death in a parade of ongoing protests. The Tibetans left behind exist in a region where their language has been stifled in schools and replaced by the Mandarin of the Chinese. Thus beside the overtness of military action has convincingly brewed more insidious forms of cultural repression.

For the Tibetans, the Dalai Lama is their spiritual lodestar, the arch-representation of a way which harks back a thousand years. To the Chinese government in Beijing, the Lama is a thorn-in-the-side who is retarding the economic progress of their Tibetan province. Some commentators from the West, specifically those against all religion, have favoured the Chinese invasion because of its collateral effect of crushing Buddhism. Others have argued that due to trading relationships the West has been reluctant to broach the Tibeto-Chinese acrimony, where a huge nation is, by reputable accounts, tormenting a minority people. Reacting to international pressure in the 1980s, China relented by starting investment in Tibet and laid down lines of railway. However, the influx of the Han Chinese served only to further overwhelm the indigenous culture of the Tibetans.

Maintaining the tenets of Buddhism, the Dalai Lama continues his position of non-violence and asks for genuine autonomy for his region, accepting that Tibet can territorially remain an appendage of the People’s Republic of China. But over fifty years no resolution has been reached between the Tibetans and communist China, the latter accusing the former of backwardness and separatism.

Accompanying the uniqueness of the Tibetan language is a medical system which is integral to the land. Distinguishable from the ancient alchemies of India and China, the Tibetan approach considers the human body as a balance of related energies. A Tibetan physician studies the tongue, urine and the pulse as indicators of the static, excretal and kinetic aspects of a body. For instance, the urine is scrutinised to observe the size and shape of the bubbles that form on its surface. If, when stirred, there is a crackling sound from the bubbles a Tibetan physician makes a diagnosis of infection.

A social ecosystem which accumulated centuries of wisdom has medicines assembled from the minerals and herbs that are native to the Tibetan Plateau. The mango does not grow in the climate of Tibet and, as such, does not feature in the chemistry of these mixtures. The Tibetan people who are disconsolate, and exiled in the south of India, think of their homeland in the mountains when they plant mango trees in the hot soil of the Karnataka plains.

Competing interests: No competing interests

18 December 2014
Jagdeep Singh Gandhi
Consultant Ophthalmic Surgeon
Worcester Royal Eye Unit
Worcestershire Royal Hospital, Worcester, UNITED KINGDOM
Click to like:

The recent call by Singleton and Strang for an evidence-infused drug policy [1] does not substantially differ from other similar evidence-based pleas, a leitmotiv that has become the catch-phrase of the drug policy field [2]. Moreover, when they mention the conditions required for their approach success (e.g., openness to and seeking of evidence by politicians, policy makers and practitioners; open-minded engagement from the public and the media), they do not address the processes and strategies leading to such prerequisites.

Initiatives for surmounting blatant divergences between rhetoric of evidence-based-policy and reality of drug policies seem to be dismissed as non-existent or not relevant. A non idle or futile first step to ameliorate, at least partially, such dissociation must be to acknowledge that evidence is but one input in drug policy-making (i.e., multiple non-evidentiary influences must be accommodated). In this regard, some authors have emphasized that understanding the entanglement of the policing context and the drug policy-making process can help researchers to maximize the uptake of their work [3, 4].

The implementation of successful experiences from social-epidemiological knowledge translation literature [5] and from the research on use of social-science knowledge in public policy [6] is also highly recommended in the drug policy-making arena. Several of these non-conventional knowledge-translation approaches, together with those tailored specifically for the addiction field (e.g., addiction science advocacy [7]), will surely play a significant role in advancing the development of – quoting Ritter and Bammer’s words [3] – an evidence-informed drug policy.

[1] Singleton N, Strang J. What would an evidence based drug policy be like? BMJ 2014;349:g7493.
[2] Lancaster K. Social construction and the evidence-based drug policy endeavour. Int J Drug Policy 2014;25:948-51.
[3] Ritter A, Bammer G. Models of policy-making and their relevance for drug research. Drug Alcohol Rev 2010;29:352-7.
[4] Ritter A, Lancaster K. Illicit drugs, policing and the evidence-based policy paradigm. Evid Policy 2013;9:452-72.
[5] Murphy K, Fafard P. Knowledge translation and social epidemiology: Taking power, politics and values seriously. In O’Campo P, Dunn JR, eds. Rethinking social epidemiology: Towards a science of change. Springer, 2012. p. 267-83.
[6] Prewitt K, Schwandt TA, Straf ML, eds. Using science as evidence in public policy. National Academies Press, 2012.
[7] Polcin D. Addiction science advocacy: Mobilizing political support to influence public policy. Int J Drug Policy 2014;25:329-31.

Competing interests: No competing interests

18 December 2014
Joan Trujols
Clinical Psychologist
Unitat de Conductes Addictives, Hospital de la Santa Creu i Sant Pau, Institut d'Investigació Biomèdica Sant Pau (IIB Sant Pau), CIBERSAM
Sant Antoni Maria Claret 167, 08025 Barcelona, Spain
Click to like:

The first person to suggest a study of the relationship between Circumcision and AIDS seems to have been forgotten. Please see the paper (1)
Clinical epidemiology, not seroepidemiology, is the answer to Africa's AIDS problem.

He wrote, "In other tribes in Africa where AIDS has proceeded from the introduction phase to the propagation phase and men have developed the disease data need to be collected on circumcision state (for men and women); abnormal sex practices".

1. Konotey-Ahulu FID. British Medical Journal, 1987, 20 June, 294, 1593-1594

Competing interests: No competing interests

18 December 2014
JK Anand
Retired doctor
Free spirit
Click to like:

Evidence-informed decision-making and evidence-informed public health requires a mix of knowledge – contextual, experiential, and scientific evidence all have something to contribute. Using a knowledge-into-action approach as an improvement process recognises the need to generate new knowledge, manage and translate this through review, critical appraisal and synthesis, and then apply it in order to determine effective policy and practice.

Developments to consider problem drug use from a public health perspective are informing current debates around drug policy. Championing the reduction of harms and a recovery agenda appropriately allows for this. For example, work in Scotland has focused on defining and attaining positive health and social outcomes for individuals with a set of core outcomes and indicators agreed at community planning level[1] . Furthermore, an evidence-informed Outcomes Framework for Problem Drug Use has recently been published[2] . In illustrating the pathways to achieving positive outcomes the framework outlines the desired direction of travel and available evidence of effective interventions with policy and practice notes relevant to Scotland. Crucially, the framework is not a definitive or prescriptive account of problem drug use, but is offered as a planning resource that can be adapted and reviewed according to local circumstance and needs assessment. It is hoped that where evidence may be lacking this can foster innovation and evaluation in order to contribute to the evidence base, share learning and advance knowledge of appropriate effective interventions.

Consistent with Singleton & Strang’s reflections (BMJ 2014:349:g7493) of the need for a policy-making environment akin to that of a ‘learning organisation’ it is intended that the outcomes framework supports and facilitates the generation, management, and utilisation of knowledge by decision-makers in order to infuse policy with evidence.

[1] see the Scottish Government website
[2] available online

Competing interests: No competing interests

18 December 2014
Elinor Dickie
Public Health Adviser
NHS Health Scotland
Click to like:

The causes of poor recruitment of medical graduates to general practice are disputed.1,2 Ironically, recruitment is likely to be further discouraged by the Royal College of General Practitioner's (RCGP) current media campaign to obtain increased funding for general practice accompanied by estimates of current GP consultation rates which greatly exceed available estimates from NHS England and Scotland. On 27th May 2014, Rebecca Smith, Medical Editor of the Daily Telegraph, citing Dr Maureen Baker, President of the Royal College of General Practitioners, stated “family doctors are seeing up to 60 patients a day and working round the clock in order to meet growing demand with less funding, medical leaders have warned... Between 2008 and 2012 the number of consultations carried out by GPs in England has risen from 300million to 340million”. On 30th May 2014, in a response from the RCGP to an article in The Times on GP workload by Alice Thomas, Dr Baker stated “it is no surprise that workloads for family doctors are ballooning, the majority of GP’s are now carrying out 40 to 60 patient consultations per day”. On 26th September 2014, Dennis Campbell, Health Correspondent of The Guardian, citing the RCGP in an article on GP waiting times stated “the average number of consultations undertaken by each GP has risen from 9,264 to 10,714 between 2005-06 and 2012-13.'' On 4th November 2014 on BBC News, Health Correspondent Nick Triggle again cited an estimated 340million GP consultations in 2013-14 in NHS England. On 23rd November 2014, in an article on rising GP workload in The Observer, Daniel Boffey, Policy Editor, cited Dr Baker as follows “our research has shown that doctors are routinely having to work 11-hour days and are making between 40-60 patient contacts a day. We now make 340million patient consultations a year, 40million more than 5 years ago”.

While examining trends in NHS performance in NHS England and Scotland before and after Scottish devolution, I obtained data on consultation rates in general practice in both countries. GP consultation rates in NHS England are only available between 1996-97 and 2008-09 from a study carried out by the Q Research GP database in the University of Nottingham on behalf of the NHS Information Centre.3 This compiled data from 602 practices and about 10million patients. In a thirteen year period between 1995-96 and 2008-09 the crude consultation rate for general practitioners (practice consultations, home visits and telephone consultations) rose by 13% from 3.0 to 3.4 consultations per person year. The consultation rate for practice nurses rose by 138% from 0.8 to 1.9 consultations per patient year. Combined consultation rates for GP’s, nurses and other clinicians rose by 55% from 3.9 to 5.5 per patient year. No data on GP consultation rates in NHS England are available after 2008-09.

Allowing for the rise in population, these consultation rates are equivalent to a rise in GP consultations in NHS England of 20% from 145million in 1996-97 to 175million in 2008-09. Projected to 2012-13 at a continuing increase of about one per cent per year, consultations rise to 190 million in 2012-13,56% of the total claimed by the RCGP in that year.

Estimates of GP consultation rates in NHS Scotland are available from 2003-04 to 2012-13, derived from an 8% sample of all Scottish practices contributing to the Practice Team Information study carried out by the Information Services Division of NHS Scotland.4 Between 2003-04 and 2012-13, the estimated number of GP face-to-face consultations in NHS Scotland rose by 4% from 15.626million (95% CI 14.938-16.315) to 16.236million (95% CI 15.499-16.973). These rates are equivalent to GP consultation rates of 3.1 per person year in both 2003-04 and in 2012-13.

Since no data on GP consultation rates are available for NHS England after 2008-09, the author emailed the RCGP on 7th October requesting the source of the data given to the media for 2012-13. The RCGP replied that the estimates were obtained from a report commissioned from management consultants Deloitte entitled “Under Pressure. The funding of patient care in general practice”, published on 2nd April 2014 The report included only projections of all consultations taking place in general practice by a general practitioner, practice nurse or other clinicians between 2008-09 and 2017-18 from regression analysis based on the Q Research GP database between 1995-96 and 2008-09. The projection of 349 million consultations for 2012-13 in the Deloitte report is similar to the 340million consultations cited by the RCGP in comments to the media. Data on annual and daily GP consultations per capita in 2012-13 are not contained in the Deloitte report and appear to have been derived by the RCGP from the Deloitte projections. The RCGP's claim that the 2012-13 Deloitte projection represents GP consultations only appears to be a misrepresentation of the data; the derived annual and daily consultation rates appear similarly distorted. .

As noted above, the estimated 190million GP consultations projected for NHS England in 2012- 13 from the Q Research GP database were carried out by 36,294 Whole Time Equivalent GP’s and represent 5,300 consultations per GP annually. Assuming that each GP works a 5-day week with 6 weeks’ holiday and consults for 230 days annually, he/she would carry out about 23 consultations daily. In NHS Scotland, 16.2million consultations were carried out by 3,735 Whole Time Equivalent GP’s, equivalent to 4,347 consultations annually and about 19 consultations daily. These consultation rates are substantially lower than the 10,714 annual and 40-60 daily consultations claimed by the RCGP in 2012-13.

Assuming a normal distribution of average consultation rates, a small proportion of practices may experience average consultation rates of 40-60 daily. Seasonal and daily variations may also produce peaks in demand but average rates of this magnitude are unsupported by available data from NHS England and Scotland.

Attempts by the author to obtain further information on this issue from the RCGP have failed, apart from the provision of the source of the projected consultation rates in the Deloitte report. An unintended consequence of exaggerated claims of GP consultation rates by the RCGP may be to provide further disincentives to medical practitioners considering a career in general practice.

1. Wakeford R. Fire the Medical Schools Council if you want more GPs. BMJ 2014;349:g6245. (28 October)
2. Majeed A. The NHS, not medical schools, is responsible for the crisis in GP recruitment. BMJ 2014;349:g6967. (29 November)
3. Hippisley-Cox J, Vinogradova Y. Trends in consultation rates in General Practice 1995/1996 to 2008/2009: Analysis of the QResearch® database. Final report to the NHS Information Centre and Department of health. Leeds: The NHS Information Centre for health and social care; 2009. Available from:
4. GP Consultations / Practice Team Information (PTI) Statistics [internet]. Edinburgh: ISD Scotland; 2013 [cited 2014 Dec 17]. Available from:

Competing interests: No competing interests

18 December 2014
Matthew G Dunnigan
Retired consultant physician
Glasgow, G11 7HH
Click to like: