Yet Again RCTs Fail to Confirm Promising Effects of Dietary Factors in Observational Studies

This time the dietary factors were Omega-3 fatty acids and various trace elements, and the condition of interest was cognitive decline with age.[1] The interventions were tested in a factorial RCT and the mean age of patients was 71. To be sure, cognitive function declined on average with the passing of the years, but did so equally, irrespective of whether patients had omega-3, other nutritional supplements, both, or neither. Loss to follow-up was low, the study was large (n=3,741), and follow-up averaged five years. It is possible, at least in theory, that any effect of nutrition on cognition unfolds over decades not years. It is also possible that foods rich in the various nutrients tested contain other factors that, singly or in combination, are responsible for the beneficial effects seen in observational studies. How will we ever know? Here is a thought – RCTs of nutritional factors should be thought of as testing the hypothesis that they are harmful – what some call equivalence studies. Then, so long as the nutritional factor of interest is not harmful and is good to eat, it should be included in a balanced diet.

— Richard Lilford, CLAHRC WM Director


  1. Chew EY, Clemons TE, Agrón E, et al. Effect of Omega-3 Fatty Acids, Lutein/Zeaxanthin, or Other Nutrient Supplementation on Cognitive Function. The AREDS2 Randomized Clinical Trial. JAMA. 2015; 314(8):791-801.

Very Different Results from RCT and Observational Studies?

The CLAHRC WM Director lives only a few doors down from the Edgbaston golf course. The club house is a fine Georgian building and was the home of William Withering – a member of the Lunar Society and the person who discovered the cardiac-stimulating drug digitalis (digoxin). Foxglove plants, the natural source of digitalis, still grow in profusion. The CLAHRC WM Director prescribed this medicine frequently when working as a junior doctor on the medical wards. However, the drug fell from grace when many observational studies showed that use of the medicine was associated with an increased risk of death in patients with heart failure. However, a recent meta-regression [1] from Birmingham, London and Melbourne showed that the more care that was taken to reduce the risk of bias, the smaller the estimated increase in mortality, ending up with RCTs showing a neutral effect. The source of bias is obvious – doctors prescribe the medicine for their sickest patients, and the observable prognostic factors pick up only a proportion of the increased risk. The residual prognostic factors are subtle clues that experienced doctors can sense in a tacit way.

The accompanying editorial [2] uses these data to rubbish observational evidence, rather spectacularly missing a more subtle point – the greater the care taken in observational evidence, the more the risk of bias can be mitigated. Further mitigation is possible by adjusting for bias, using the method of Turner et al.,[3] thereby reducing point estimates and widening confidence limits into credible limits. We are considering using digoxin as an examplar of the method.

As for William Withering’s medicine; well it appears not to increase death rates after all, while both the observational and RCT evidence suggests that it reduces the need for admission. Two hundred and thirty years after his discovery, the scientific principles of the Englightenment that Withering espoused continue to refine our understanding of the medical uses of digoxin.

— Richard Lilford, CLAHRC WM Director


  1. Ziff OJ, Lane DA, Samra M, et al. Safety and efficacy of digoxin: systematic review and meta-analysis of observational and controlled trial data. BMJ. 2015; 351: h4451.
  2. Cole GD & Francis DP. Trials are best, ignore the rest: safety and efficacy of digoxin. BMJ. 2015; 351: h4662.
  3. Turner RM, Spiegelhalter DJ, Smith GC, Thompson SG. Bias modelling in evidence synthesis. J R Stat Soc Ser A Stat Soc. 2009;172(1):21-47.

Caring in a Fallen World

In the previous News Blog we discussed the plight of the homeless and other groups at the margins of society, such as migrant and traveller populations. We did not mention sex workers. Sex workers are exposed to all sorts of risk, especially sexually transmitted infections (STIs) and gender-based violence.

What is to be done? One approach is to tackle the problem at source and reduce, or even eliminate, prostitution. Supply-side measures, such as ‘raid and rescue’ are a failure both in terms of reducing prostitution and arguably for moral reasons as well. Demand-side interventions, such as penalising customers are also notoriously unsuccessful. We await with interest the results of criminalising all payments for sex in Sweden – though it appears not to have worked, despite government claims.[1] [2]

In the meantime, we live in a fallen world and the plight of sex workers should not be ignored – to “look the other way” is to show a profound lack of compassion reminiscent of the Pharisees of the New Testament. So what can be done while we wait for a Nordic utopia?

The CLAHRC WM Director was privileged to attend a talk on the subject during a recent conference organised for the Malawi-Liverpool Wellcome Trust Annual Scientific Meeting at the Makokola Retreat, Malawi. Frances Cowan gave a talk that synthesised scientific rigour and moral inspiration.

Frances first described the Sonagachi Project in Calcutta.[3] This project (which was first implemented in 1991 as the Sexually Transmitted Diseases [STD] / HIV Intervention Project [SHIP]) , was designed to empower female sex workers to act collectively. For example, they taught each other how to manage relationships with clients and insist on the use of condoms. They also worked to destigmatise sex work – redefining sex work as employment and articulating a set of rights for sex workers, including their right to an education, good health and freedom of movement.

This project prevented the spread of HIV, with the prevalence remaining static at about 2% in Sonagachi, while rising alarmingly across the rest of India. The Sonagachi experience has been rolled out in many countries with support from the Bill and Melinda Gates Foundation.[4]

Next, Frances gave an account of a systematic review and meta-analysis of 22 studies (encompassing 30,325 sex workers) of community empowerment programmes.[5] The authors found that community empowerment-based approaches to addressing HIV were significantly associated with reductions in HIV and other STIs, and with increases in condom use. However, there are a number of barriers to implementation, such as funding constraints, legislation and discrimination. Busza and colleagues [6] explain in more detail how such programmes work, involving community cohesion among the workers, the development of a sense of self-worth and autonomy, and hence the collective power to gradually take over the programme from the instigators. In essence the instigator intervenes to kick-start a programme that then becomes self-sustaining.

Lastly, Frances described her own work replicating and extending this model in the unpropitious context of Zimbabwe. She trained female sex workers to become, not just peer educators, but also to develop para-legal skills so that they could support colleagues to seek redress against violence (including violence meted out by the police who are supposed to protect citizens). As in the Sonagachi programme, meetings which were facilitated by the instigator in the first instance were later taken over by peers from the ‘sisterhood’. It was important to let the participants select their own goals. For example, many wanted help in negotiating the best price from customers. However, the relationship built up in meeting this requirement enabled instruction to be provided on health issues, such as safe use of condoms.

Yes, it would be better if the world could be sanitised of evil and women would never be placed in a situation where they need to sell sexual services. But, in the meantime, much can be done in a practical sense to improve the lot for people who, through no fault of their own, find themselves in a position where prostitution is perceived as their best option.

— Richard Lilford, CLAHRC WM Director


  1. Dodillet S & Östergren P. The Swedish Sex Purchase Act: Claimed Success and Documented Effects. International Workshop: Decriminalizing Prostitution and Beyond: Practical Experiences and Challenges. 2011 Mar 3-4; The Hague, the Netherlands.
  2. Jordan A. The Swedish Law to Criminalize clients: A failed experiment in social engineering. Issue Paper 4. Washington, DC: American University Washington College of Law. 2012.
  3. Jana S, Basu I, Rotheram-Borus MJ, Newman PA. The Sonagochi Project: a sustainable community intervention program. AIDS Educ Prev. 2004; 16(5):405-14.
  4. Avahan – The India AIDS Initiative. The Business of HIV Prevention at Scale. New Delhi, India: Bill & Melinda Gates Foundation. 2008.
  5. Kerrigan D, Kennedy CE, Morgan-Thomas R, et al. A community empowerment approach to the HIV response among sex workers: effectiveness, challenges, and considerations for implementation and scale-up. Lancet. 2015; 385(9963): 172-85.
  6. Mtetwa S, Busza J, Davey C, Wong-Gruenwald R, Cowan F. Competition is not necessarily a barrier to community mobilisation among sex workers: an intervention planning assessment from Zimbabwe. BMC Public Health. 2015; 15(1):787.

Health of the Homeless and Inclusive Health Services

My friend and colleague Aidan Halligan died recently. Like myself he was a Professor of Obstetrics and Gynaecology who then branched out to pursue an eclectic career. We both did a stint in the Civil Service; in Aidan’s case as Deputy Chief Medical Officer. We also shared an interest in the homeless. Aidan established a charity for homeless people (Pathway). The application form for a CLAHRC included a section where applicants were invited to apply for an additional £1 million, and we proposed research into health care for the homeless. Although this supplementary application was unsuccessful, CLAHRC WM continues to pursue a research project in this field and we are delighted to be collaborating with Pathway.

The homeless are quite a hard client group to define precisely. Perhaps, rather than trying to narrow down the definition, it is more useful to broaden it out. The concept can be expanded to include many who occupy the margins of society, irrespective of whether they are homeless in the literal sense – they will include many people recently incarcerated, those with severe drug abuse, and nomadic populations. The difficulties they encounter overlap with those of economic migrants and asylum seekers, although their psychological assets are different – descending to the bottom is not the same as starting there. I suspect that in many instances facilities are closed to migrants, while they are open to the ‘homeless’ – just not accessed.

The homeless have it rough, even when they are not asleep – their mortality rate is not just 1.4 or 1.7 as for many medical conditions such as diabetes, but many times greater than other people.[1] [2] [3] The life expectancy of a homeless person in a high-income country is probably worse than that of a person of similar age living in a Nairobi slum.

But homelessness is not common – it is not a ‘public health problem’ in the sense of diabetes, depression, dementia, cancer, and so on. However, the homeless occupy the extreme left end of the inequalities distribution and, as hinted earlier, their personal ‘assets’ are low. ‘Tough love’ may be good in many circumstances, but not for the homeless. It is not always easy to imagine the plight of another human, and when the person is unclean and aggressive the well of compassion can dry up. But, as suggested in a previous post, it is up to us whether we let it stay dry or replenish it. Aidan had the imagination to at least glimpse what it might be like at the ‘extreme left’ and so replenished his well. So, yes, we should be prepared to pay more per unit of health gain for homeless people. How much more is for society to decide according to a collective ‘social welfare function’.

But good intentions are not enough. Interventions for people who are homeless have been published recently.[4] Co-ordinating social and different types of medical care is key to improved health and wellbeing for the homeless, as it is for other groups of people with complex needs (acute medical illness, multi-morbidity). This can be achieved with different combinations of case-management, critical time interventions, and assertive outreach. It also appears to be the case that getting the person into a permanent dwelling is critically important and improves access to, and effectiveness of, other services – hence the ‘housing first’ concept. Such services need to permit alcohol, while trying to address the problem, else the person will end up back on the street. Of course, those who don’t drink also don’t want to share dwellings with those who do. I never said this was going to be easy!

What is CLAHRC WM going to do about it then? Inspired by Toby Lewis, Chief Executive of Sandwell and City Hospital in Birmingham, we plan an intervention targeted on the A&E department. Emergency care is the sharp-end of homeless care and a case picked up here and managed across specialities and across the hospital/community interface could add real value. No more patients discharged onto the street in dressing gown and slippers. At least, that is our hypothesis. Our nascent plan is to pilot the intervention at Sandwell and City hospital and then, if it appears successful, roll it out – yes under a step wedge evaluation![5] We are not solipsistic, but seek collaborators at home and abroad.

Aidan Halligan was speaking to an orthopaedic surgeon who had done a brilliant knee operation on a homeless person. When Aidan asked how the patient had managed back in the community, the surgeon was unable to give an answer. “You are a good surgeon,” Aidan said, “But when you help homeless people, not just their knees, you will be a great surgeon.”

— Richard Lilford, CLAHRC WM Director


  1. Hwang SW. Mortality among men using homeless shelters in Toronto, Ontario. JAMA. 2000; 283(16):2152-7.
  2. Barrow SM, Herman DB, Córdova P, Struening EL. Mortality among homeless shelter residents in New York City. Am J Pub Health. 1999;89(4):529-34.
  3. Hibbs JR, Benner L, Klugman L, et al. Mortality in a cohort of homeless adults in Philadelphia. N Engl J Med. 1994; 331(5):304-9.
  4. Hwang SW & Burns T. Health interventions for people who are homeless. 2014. 384:1541-7.
  5. Hemming K, Haines TP, Chilton PJ, Girling AJ, Lilford RJ. The stepped wedge cluster randomised trial: rationale, design, analysis, and reporting. BMJ. 2015;350:h391.

Smokeless Tobacco

News blog readers will have seen our recent blog post on the dangerous areca nut that is often chewed with tobacco. Siddiqi and colleagues have carried out a massive survey of the use of “smokeless tobacco” worldwide, yielding data from 113 countries.[1] Smokeless tobacco is heavily consumed in South and South East Asia, and some African and Nordic countries. The authors conducted an extensive meta-analysis of the effects of oral tobacco – a task made complex by the many different preparations. For example, versions used in Nordic countries, South Africa and North America seem innocuous, whereas versions used in Asia are highly carcinogenic, even after controlling for alcohol and smoking. Of course, much of the toxicity may come not from the tobacco, but from the areca nut that often accompanies it and which releases large amounts of cancer and fibrosis-inducing copper.

— Richard Lilford, CLAHRC WM Director


  1. Siddiqi K, Shah A, Abbas SM, Vidyasagaran A, Jawad M, Dogar O, Sheikh A. Global burden of disease due to smokeless tobacco consumption in adults: analysis of data from 113 countries. BMC Medicine. 2015. 13:194.

Look Out for ‘p-hacking’

Psychologists do it, biologists do it, and even squeaky clean economists do it. How can you find out whether a group of scientists are guilty of making lots of comparisons and then selecting positive results for publication? You can draw a graph using all the p values in all the papers, or all the p values relating to the central question in all the papers. You might find it looks like this:

DC - Bullshit Detectors Fig 1

That would be a reassuring result. But what about if it looked like this?

DC - Bullshit Detectors Fig 2

Such a result shows clustering of p values just below the conventional significance level of 0.05 and is strongly indicative of p-hacking.

In an important blog,[1] prominent statistician Uri Simonsohn shows that the first – satisfactory – pattern approximates to findings when all p values in all papers are analysed. But when the sample of p values in ‘enriched’ by selecting principal findings or those p values associated with a co-variate, then the signal emerges from the noise, as in the second figure.

— Richard Lilford, CLAHRC WM Director


  1. Simonsohn U. [41] Falsely Reassuring: Analyses of All P-Values. 24 Aug 2015. [Online].

Accuracy of the Recording of the Cause of Deaths in Hospitals is low

Prosaic as it might sound, the above article is unusually important [1] – the medical examiner system in England will now examine case notes of many deaths, and case note review is the only real method to assess the technical quality of medical care – so called trigger tools miss most of the important stuff. This systematic review finds that between half and three quarters of stated causes of death are wrong. This is not surprising given that implicit case note review has high measurement error. The gold standard here is an external systematic review, such as the proposed medical examiner system. Such a method still misses many of the real causes and post-mortem is the ultimate standard. The authors provide a lot of guidelines for future investigators proposing case note reviews.

— Richard Lilford, CLAHRC WM Director


  1. Rampatige R, Mikkelsen L, Hernandez B, Riley I, Lopez AD. Systematic review of statistics on causes of deaths in hospitals: strengthening evidence for policy makers. Bull World Health Organ. 2014;92:807-16.

Acute Loss in Cognitive Function Follows a Stroke

People who have had a stroke have lower cognitive ability than age-matched controls. But people with lower cognitive ability are predisposed to stroke. This paper [1] summarises a prospective study of no less than 23,000 volunteers. Lower cognition was indeed associated with increased risk of stroke, but the stroke itself occasions a further substantial drop – 1 point on a 6 point scale. Cognitive development continues to fall after the initial insult – perhaps through the “connectome,” as discussed in a previous blog.

— Richard Lilford, CLAHRC WM Director


  1. Gorelick PB, Nyenhuis D. Stroke and Cognitive Decline. JAMA. 2015;314:29-30.

Effect of Smoke-Free Legislation on Human Health

Two recent studies have considered the effect of smoke-free legislation bans on perinatal and child health. The first looks at smoke free legislation in England.[1] Their interrupted time series documents an immediate drop in stillbirths, low birth weight, and neonatal mortality following introduction of the legislation in July 2007. Interestingly, sudden infant death rates showed no change, perhaps because parents already avoided smoking at home?

The second paper is a systematic review of smoking bans in work places and/or public places, and at least one measure of health.[2] Eleven studies were identified, all interrupted time series. Again smoke free bans were associated with reductions in pre-term birth and hospital attendance for asthma, but this time no effect on birth weight was noted. The CLAHRC WM Director fancies himself as a libertarian. The limit on personal liberty is maternal risk of physical harm to others. He therefore supports smoke-free legislation.

— Richard Lilford, CLAHRC WM Director


  1. Been JV, Mackay DF, Millett C, Pell JP, van Schayck OCP, Sheikh A. Impact of smoke-free legislation on perinatal and infant mortality: a national quasi-experimental study. Sci Rep. 2015;5:13020.
  2. Been JV, Nurmatov UB, Cox B, Nawrot TS, von Schayck CP, Sheikh A. Effect of smoke-free legislation on perinatal and child health: a systematic review and meta-analysis. Lancet. 2014;383:1549-600.

Appendectomy: Not Mandatory in all Cases of Appendicitis

Unoperated appendicitis is not a death warrant. A recent editorial in JAMA provides some interesting evidence going back to the origins of modern surgery.[1] A very nice RCT is reported in the same issue of JAMA.[2] It shows no long-term health risks in the conservatively managed group, but over a quarter in the non-surgical group had an interval operation for ongoing symptoms. However, here is the risk – a point mentioned by neither the editorialist nor trial authors. The cases entered in the study were all carefully selected by CT scan. But in most parts of the world emergency CT scanning is not available. So it would not be possible to apply the many exclusions, including a plug of material, called an appendicolith, in the lumen of the appendix. So while these are important studies in high-income countries, it would be very hard, and arguably risky, to generalise to other countries.

— Richard Lilford, CLAHRC WM Director


  1. Livingstone E, Vons C. Treating Appendicitis Without Surgery. JAMA. 2015;313(23):2327-8.
  2. Salminen P, Paajanen H, Rautio T, et al. Antibiotic Therapy vs. Appendectomy for Treatment of Uncomplicated Acute Appendicitis. The APPAC Randomized Controlled Trial. JAMA. 2015; 313(23):2340-8.