Chlamydia Vaccine

Chlamydia, is the most common bacterial sexually transmitted infection (STI), with around 218,000 new cases in the UK in 2018, accounting for 49% of all new STI diagnoses.[1] Worldwide it is estimated that there are 131 million new cases each year. Although it can be treated with antibiotics, around 75% of people with chlamydia do not show any signs of infection, and left untreated it can lead to infertility. Therefore there is a need for a preventive measure. The first ever clinical trial looking at a vaccine for genital chlamydia was recently published in Lancet Infectious Diseases.[2]

The authors conducted a phase 1 RCT in 35 women, randomly assigning them to receive one of two versions of a new chlamydia vaccine (CTH522:CAF01 or CTH522:AH) (N=15 in each arm) or a saline placebo (N=5). Thirty-two women completed the study, being given an injection at the start of the study and then again 1 and 4 months later. This was followed by intranasal administrations at 4.5 and 5 months. No serious adverse events were reported, and there was no significant differences in the incidence of reactions at the injection-site (P=0.0526), or in reactions to intranasal administration (P=1.000) when comparing either vaccine to placebo. Analyses showed that both versions of the vaccine induced an immune response in all participants, compared to none in the placebo group. When comparing the two vaccines, one (CTH522:CAF01) showed more promising results including antibodies detected earlier, higher levels of IgG antibodies, an enhanced mucosal antibody profile, and a more consistent cell-mediated immune response profile. With both vaccines appearing to be safe and tolerable, further clinical trials will hopefully be forthcoming.

— Peter Chilton, Research Fellow


  1. Public Health England. Sexually transmitted infections and screening for chlamydia in England, 2018. Health Protection Report. 2019; 13(19).
  2. Abraham S, Juel HB, Bang P, et al. Safety and immunogenicity of the chlamydia vaccine candidate CTH522 adjuvanted with CAF01 liposomes or aluminium hydroxide: a first-in-human, randomised, double-blind, placebo-controlled, phase 1 trial. Lancet Infect Dis. 2019.

Ban the Term Animal ‘Model’

I have always been somewhat bemused by the term ‘animal model’ in research. As an animal lover and admirer, I have always bridled at the harsh denigration of animals to mere ‘models’ for our species.

Recently I read about the Arizona Cancer Evolution center, which compares and contrasts findings across the animal kingdom as a whole to learn generalisable lessons.[1] One interesting example concerns Peto’s paradox. This paradox turns on the observation that larger animals do not have higher cancer rates than humans, despite having many more cells. The resolution to the paradox comes from the finding that very large animals have a higher proportion of DNA repair and apoptosis genes. These genes help reduce somatic mutations or their effects, and compensate for the greater a priori risk.

Using animals as mere ‘models’ for humans is not only speciesist, but non-scientific. I hope we can get rid of this patronising and scientifically limiting term once and for all.

— Richard Lilford, CLAHRC WM Director


  1. Tollis M, Boddy AM, Maley CC. Peto’s Paradox: how has evolution solved the problem of cancer prevention? BMC Biol. 2017; 15: 60.

WASH 2: Thoughtful Analysis of the Three Great Recent WASH Trials 

We have reported previously on two of the above three trials, conducted in Kenya and Bangladesh.[1] A third similar trial has now been reported from Zimbabwe.[2] All three trials show that nutritional interventions reduce stunting. However, other trials have shown that nutritional interventions, while reducing stunting, do not improve cognitive outcomes.[3]

The water quality, sanitation and handwashing (WASH) interventions have no consistent beneficial effect on childhood diarrhoea, as mentioned in the previous article. Similarly, behavioural interventions are only helpful when they are very intense and even then the effects are not sustained. The WASH interventions do not reduce stunting. This is in contrast to observational studies that consistently show a correlation between WASH interventions and reduced stunting. When the investigators analysed the control group they reproduced the findings from the observational studies; WASH conditions were strong independent risk factors for poor linear growth in children. Inconsistency between the experimental and observational studies supports the conclusion that the observational studies are likely confounded by unmeasured factors in households that correlate with WASH conditions.

A number of studies have examined environmental faecal contamination and intestinal colonisation. These show that the WASH interventions did not significantly or sufficiently decontaminate the environment (see previous report). This means that more thorough interventions are required; presumably proper indoor lavatories and clean water piped into households. We simply have not yet found the tipping point when it comes to the intensity and coverage required to provide sanitary living conditions for poor people in poor countries. This is bad news. Many of the large gains in infant mortality have been achieved through relatively inexpensive interventions, such as a vaccination and impregnated mosquito nets. We seem to have run up against a more intractable problems when it comes to the eradication of childhood diarrhoea.

— Richard Lilford, CLAHRC WM Director


  1. Lilford RJ. Important New Data on WASH and Nutritional Interventions from Kenya and Bangladesh. NIHR CLAHRC West Midlands News Blog. 18 May 2018.
  2. Pickering AJ, Null C, Winch PJ, et al. The WASH Benefits and SHINE trials: interpretation of WASH intervention effects on linear growth and diarrhoea. Lancet Glob Health. 2019; 7(8):e1139-46.
  3. Lilford RJ. Nutritional Interventions for Childhood Stunting in Slums. NIHR CLAHRC West Midlands News Blog. 21 June 2019.

Mortality Associated with Proton Pump Inhibitors

Proton pump inhibitors (PPI) are very widely used in society. Moreover, their use is often prolonged, lasting years if not decades. These agents have an important effect upon a fundamental biological mechanism: the transport of positively-charged ions across cell membranes. It is therefore a matter of great importance to investigate the potential risks of widespread and prolonged use of an agent with such a profound effect on the body’s metabolism.

It is already known, or at least frequent investigations have found, that, in addition to an increased risk of diarrhoea, there is an association between PPI use and all-cause mortality. A recent paper in the BMJ has investigated this matter further.[1] The study was based on a cohort of US Veterans. The investigators compared all-cause and specific-cause mortality in patients taking PPIs with those who were not, and also examined the strength of any association by duration of use. They also examined for possible reverse causality by looking at outcomes only after a considerable duration from the start of therapy. They examined for selection bias by comparing outcomes among people taking PPIs with those taking other medicines for upper abdominal symptoms.

The study replicated the findings of previous studies, confirming an association between PPI use and all-cause mortality. In fact there was an extra 45 deaths per thousand people taking PPIs (95% confidence interval 28.20–61.40). Kidney disease and cancer of the upper gastrointestinal tract were significantly increased. Risks increased in proportion to duration of use.

Of course, such an observational study cannot prove causality beyond all reasonable doubt. However, the finding that the results were not consistent with reverse causality and that they were unlikely to be due to selection bias, makes a causal explanation highly plausible. The increasing association of deaths with duration of use is of particular practical importance. Since the burden of evidence favours a causal explanation between increasing duration of use and risk of premature death, people taking these medicines should be weaned off them as soon as possible.

— Richard Lilford, CLAHRC WM Director


  1. Xie Y, Bowe B, Yan Y, Xian H, Li T, Al-Aly Z. Estimates of all cause mortality and cause specific mortality associated with proton pump inhibitors among US veterans: cohort study. BMJ. 2019; 365: l1580.

WASH 1: Explaining the Results of Clinical Trials: A Superb Study of Mediating Variables

The disappointing results of water quality, sanitation and handwashing (WASH) interventions in low-income countries have been reported previously in this News Blog.[1][2] We suggested that the intensity of the interventions and their coverage, had not been sufficient, to decontaminate the environment in these trials. We posited a type of herd effect, whereby a tipping point would be reached if the interventions were sufficiently intensive and their coverage sufficiently wide.[3]

I can now report on a lovely study that was nested into one of the randomised trials. In this study water, hands, food, soil and flies were examined in both the intervention and control clusters in the trial.[4] What the study does is explain why the trials showed null results. While the water delivered did improve in terms of faecal contamination, the hands, food, soil and flies remained equally contaminated across both the control and intervention groups. It is becoming increasingly clear that interventions to reduce childhood diarrhoea are simply going to have to be more intensive and hence, unfortunately, more expensive than those deployed hitherto. We also need a better index of contamination. Diarrhoea is a very nonspecific measure of risk. Dysentery is too rare to use as an outcome in clinical trials, though it is much more specific. Measuring contamination along the causal chain, as in this brilliant study, is time intensive and expensive. In collaboration with our partners at the icddr,b in Bangladesh, we are working on a possible alternative measure, based on the profile of gastrointestinal tract bacteria in the early childhood population.

— Richard Lilford, CLAHRC WM Director


  1. Lilford RJ. Important New Data on WASH and Nutritional Interventions from Kenya and Bangladesh. NIHR CLAHRC West Midlands News Blog. 18 May 2018.
  2. Lilford RJ. Childhood Diarrhoeal Diseases – Update of the Famous Wolf Review. NIHR CLAHRC West Midlands News Blog. 19 October 2018.
  3. Lilford RJ, Oyebode O, Satterthwaite D, Melendez-Torres GJ, Chen Y-F, Mberu B, et al. Improving the Health and Welfare of People Who Live in Slums. Lancet. 2017; 389: 559-70.
  4. Ercumen A, Pickering AJ, Kwong LH, et al. Do Sanitation Improvements Reduce Fecal Contamination of Water, Hands, Food, Soil, and Flies? Evidence from a Cluster-Randomized Controlled Trial in Rural Bangladesh. Environ Sci Technol. 2018; 52(21): 12089-97.


Aborting Migraine: The CLAHRC WM Director Experiments on Himself

One evening recently, about half an hour after a workout in the gym, I suddenly found that I could not read due to multiple scotomas (blind patches). Although I have suffered intermittent migraine for most of my life, I was surprised. This surprise was because I had not had a migraine for over five years. In fact, the scotomas were bilateral and central, while my migraines are almost always unilateral and peripheral. I therefore thought I might be having a stroke and decided to check for facial asymmetry. My face looked okay until I smiled. At this point it looked rather asymmetrical. However, no further localising symptom developed; I guess I have a slightly lopsided smile! The aura then changed to a more typical right temporal hemianopia. I realised, with relief, that this was just another migraine.

I sat on my bed staring into the wardrobe mirror and worked out the anatomy. Scotomas on the right temporal side meant ischaemia (not enough blood) in my left occipital cortex. The thought occurred to me that anaesthetists hyperventilate people during brain surgery to reduce the concentration of carbon dioxide in the blood, which, in turn, reduces blood flow to the patient’s brain. Likewise blood flow can be increased by increasing blood CO2 by rebreathing into a bag.

So I went and fetched the plastic bag I use to carry fluids through the security checkpoint at airports, and rebreathed for a while to increase the CO2 in my blood. The result was absolutely miraculous. After about 90 seconds, the scotomas shrank and disappeared. I thought it might recur but it did not and I could read perfectly.

I thought I might be off to Stockholm and immediately punched ‘rebreathing in migraine’ into Google in my phone. Alas and alack! I am after the fact as per usual. According to the first review article I found the beneficial effect of hypercapnia (increasing CO2 in the blood) on a migraine attack is well described.[1]

So why isn’t rebreathing more widely recommended? I don’t know, but suspect doctors fear being sued if someone should come to harm while having, say, a haemorrhagic stroke or heart attack. However, I think that a thorough risk assessment of rebreathing should be carried out. This assessment could be followed by a RCT of rebreathing in young people with classical migraine who are fully informed of the risks. The incidence of migraine is higher among doctors than the general population. Doctors are not a vulnerable group, as far as the relevant informational asymmetries are concerned. So, how about a trail among doctors? Then, if the results are positive, the method could be promoted more broadly with very careful explanation. The promotional materials would need very clear instructions on risks, benefits and precautions.

Building further on these ideas about CO2 and migraine, I remembered that mountain sickness is caused by hypocapnia (too little CO2 in the blood). The mechanism is over-breathing (as I had doubtless been doing during the exercise that preceded my migraine). Mountain sickness can be treated with a medicine called acetazolamide. So I punched ‘acetazolamide and migraine‘ in the search engine. It turns out that this medicine is an effective treatment for migraine attacks. A trial of its use as prophylaxis had to be stopped early because of side-effects, but there was no evidence of reduced frequency of attack at the point where the trial was brought to a close. In any event, the effects on change in blood CO2 levels in precipitating and aborting a migraine seem reasonably well established.

I would value readers’ opinions.

— Richard Lilford, CLAHRC WM Director 


  1. Fuglsang CH, Johansen T, Kaila K, Kasch H, Bach FW. Treatment of acute migraine by a partial rebreathing device: A randomized controlled pilot study. Cephalalgia. 2018; 38(10): 1632-42.

Multiple Micronutrient Supplementation During Pregnancy and Birth Outcomes in Low-Income Countries

I thank John Ovretveit (Karolinksa Institutet, Sweden) for bringing this important Lancet article to my attention.[1]

The study is a meta-analysis of individual patient data from 17 RCTs in low- and middle-income countries. The use of individual patient data enables multiple sub-group analyses to be carried out.

This paper confirms some things we already know. For example, iron reduces pre-term delivery and low birth weight. The effect is much greater in anaemic woman in both low- and high-income countries. Iron and folic acid combined have a more powerful effect on reducing these outcomes than either iron or folic acid alone.

This paper shows that including at least one further micronutrient has additional benefit. It also shows that multiple micronutrients reduce neonatal mortality, but only among girls. There is no solid biological explanation for this observation. Consistent with the underlying hypothesis, starting micronutrients early in pregnancy, and high adherence were associated with larger effect sizes. Importantly, no harms were demonstrated for micronutrients.

I recently summarised evidence regarding micronutrients for adults in high-income countries.[2] Here there was no evidence of benefit and evidence of harm with high doses of mineral supplementation. I would still be cautious about liberal use of mineral supplements in pregnancy, with the exception of iron.

— Richard Lilford, CLAHRC WM Director


  1. Smith ER, Shankar AH, Wu LS, et al. Modifiers of the effect of maternal multiple micronutrient supplementation on stillbirth, birth outcomes, and infant mortality: a meta-analysis of individual patient data from 17 randomised trials in low-income and middle-income countries. Lancet Glob Health. 2017; 5(11):e1090-100.
  2. Lilford RJ. Stop Taking Those Supplements: Just Stop. NIHR CLAHRC West Midlands News Blog. 7 June 2019.

Why You Need to Know About Primary Care Networks

Amidst the constant organisational tinkering and occasional wholesale reform that is the NHS structure, it would be easy to note the emergence of a new construct without seeking to fully understand what it is and what it does. Often, by the time the new structural entity has determined its role and embedded this within the wider landscape of service provision, there are already plans in place to supersede it.

Primary care has been particularly prone to this in recent years with Vanguards, Pioneers, Federations, Multispecialty Care Providers to name but a few. So Primary Care Networks (or PCNs), which became legal entities at the beginning of July, may understandably have failed to arouse your interest. If so, then read on.

So what exactly are they? PCNs aim to address the long-standing challenge in primary care of economy of scale. By providing services at a greater scale it is hoped they will help to address issues around workforce, the integration of non-acute services and increasingly ageing buildings from which care is delivered. These networks will mainly cover populations of 30-50,000 people (with some exceptions at both ends of the scale for particularly rural or populous areas). Almost all GPs have signed up to create around 1,300 networks across England and, whilst some do span boundaries, most are within existing Clinical Commissioning Group (CCG) footprints. They provide the locality building blocks for Primary Care against the backdrop of 191 CCGs, a number which continues to decrease with mergers.

As referenced earlier, the longevity of PCNs is always likely to be an initial query. However, they look to have a strong future, at least in the medium term. They are heavily backed by strategy and policy (they are a core component of the NHS Long Term Plan), and are financially wedded to the core GP contract, itself backed by much of the funding announced for primary care within the Long Term Plan.

All good so far, but why might we in NIHR CLAHRC and ARC West Midlands be so interested in PCNs? Well, one of their key remits is to improve the integration of primary and community care services, and there has been an active effort to include social care and voluntary and third sector organisations and to increase social prescribing. Here it speaks to much of our planned work agenda around the management of long-term conditions, and around the acute interfaces of care within both physical and mental health. There will also be funding to help GPs increase their use of digital health, which will be interesting to consider alongside our evaluative work on electronic prescribing, patient-accessed health records and virtual outpatient consultations.

PCNs also potentially provide very nice-sized clusters for implementation. The opportunity to break down, for instance, two CCGs in to seven or eight PCNs in order to conduct an intervention is an extremely attractive one. The more multidisciplinary approach to primary care delivery, which will undoubtedly include an element of skill substitution, will be ripe for evaluation that ARCs will be well placed to address.

Whether PCNs will deliver all that they promise remains to be seen. However, they are an interesting development, look as though they will have longevity, and we hope will offer a new range of evaluative opportunities for service delivery research.

— Paul Bird, Head of Programmes (engagement)

Collaboratives to Systemise Learning in Health Care Quality Improvement

Improving quality of care is an enduring challenge for all health care organisations. As a result, health care organisations carry out Quality Improvement (QI) projects (which often incorporate some form of service evaluation and/or audit) [1,2] and they have become routine activities in most hospitals. The efforts put into these projects, and the learning derived from them, tend to be localised to the institution in which they are carried out. In addition, as changes introduced and outcomes observed are confined to individual organisations, establishing causal-relationships is a problem; absent a control group it is difficult to rule out temporal trends and regression to the mean.[3] Further, there is no way to examine the transferability of findings from individual QI projects to the system at large.

Considering the difficulties in introducing and sustaining successful interventions faced by health care QI communities, there is a strong imperative to find a better way to harness the experience from the potentially large volume of evidence generated from these isolated QI projects. Lack of coordination of these activities entails two glaring dangers. First, the QI projects that do get to see the light of day are a highly biased subset of all that can be published – when did the BMJ and HSJ Awards for Quality Improvement go to a study with a null result? Second, there is a risk that lessons that can be learned from success or failure are not captured.

Creating a collaborative to coordinate QI efforts from different institutions and to systemise the process of developing, implementing and evaluating QI projects seems to be an obvious solution. Such collaboratives emerged from North America in late 1980s and are gaining popularity internationally.[4] QI collaboratives often focus on specific issues or specialities (e.g. the Michigan Keystone ICU Project for reducing catheter-related bloodstream infections),[5] although generic approaches that can be adapted for different service issues/topics have been developed, such as the collaborative model from the famous Breakthrough Series of the US Institute for Health Improvement (IHI).[6] QI collaboratives to tackle specific service issues have shown promising results in improving health care quality based on a recent systematic review,[4] though somewhat ironically, the major threats concerning the validity and generalisability of individual QI projects mentioned above seem to be an equal concern for the QI collaborative literature.

The NHS has not missed the bandwagon of utilising QI collaboratives to improve service quality, with a variety of approaches being trialled through different initiatives. These span the full range from those systematically and formally created, such as the national Patient Safety Collaboratives Programme established by the Academic Health Science Networks and NHS Improvement (NHSI); the partnership between five NHS trusts and the Virginia Mason Institute also facilitated by the NHSI; and the Q Community programme being experimented by the Health Foundation; through to those which have grown organically through grass roots social movements, such as the Academy of Fabulous Stuff. This variety in deployment of QI collaboratives raises interesting questions of what approach and elements would be most effective for what contexts. Members of CLAHRC WM, Dr Nicola Burgess and Professor Graeme Currie from the Warwick Business School, are currently funded by the Health Foundation to evaluate the partnership with Virginia Mason Institute and we will follow their findings with great interest.

As with any other evaluations, systematic collection of data in a standardised format is essential for building solid evidence. This requires good infrastructure with a continuous commitment of resources, which are not easily achievable particularly when they involve multiple organisations. Here in the West Midlands we propose that the Academic Health Science Network and our future Applied Research Centre will be in an ideal position to facilitate this effort, with the leadership and input from our newly formed Health Data Research Centre based in University Hospitals Birmingham NHS Foundation Trust. We propose to work closely with Sustainability and Transformation Partnerships and with the custodians of local databases. We are also developing links with national organisations, such as NHS Improvement, Healthcare Quality Improvement Partnership, and the NIHR HS&DR Programme.

— Yen-Fu Chen, Associate Professor

— Richard Lilford, CLAHRC WM Director

— Paul Bird, CLAHRC WM Head of Programme Delivery (Engagement)


  1. Jones B, Vaux E, Olsson-Brown A. How to get started in quality improvement. BMJ. 2019;364:k5408.
  2. NHS Health Research Authority Research Ethics Service. Defining research. 2017.
  3. Chen Y-F, Hemming K, Stevens AJ, Lilford RJ. Secular trends and evaluation of complex interventions: the rising tide phenomenon. BMJ Qual Saf. 2016; 25: 303.
  4. Wells S, Tamir O, Gray J, Naidoo D, Bekhit M, Goldmann D. Are quality improvement collaboratives effective? A systematic review. BMJ Qual Saf. 2018; 27: 226-40.
  5. Pronovost P, Needham D, Berenholtz S, et al. An Intervention to Decrease Catheter-Related Bloodstream Infections in the ICU. N Engl J Med. 2006;355(26):2725-32.
  6. Institute for Healthcare Improvement. The Breakthrough Series: IHI’s Collaborative Model for Achieving Breakthrough Improvement. IHI Innovation Series white paper. Boston: Institute for Healthcare Improvement; 2003.

Pitfalls of Participatory and Action Research

With the new craze for participatory and action research here are some dangers and some ways to avoid them.

“Research with people, not on people”; “no research about us, without us”. We have all heard these strap lines. They are irenic and they signal virtue. As it turns out, the CLAHRC WM Director agrees with them.

So, what’s the problem? Danger arises if the slogans are taken to mean that all scientific observations have to be “participatory”, and to crowd out use of objective measurements. Objectivity is an important tenet of science. Human perceptions are famously fallible.[1-3] It is for this reason that the canon scientific includes many measures to minimise bias and thereby avoid error. Indeed, that is the purpose of the scientific method; the avoidance of error. Important principles here include the use of blinding/masking and use of observers who are independent of any interventions.

On the other hand, research which is done only by numbers, and which does not include the views of stakeholders, is impoverished. For example, participant observations may provide an early warning of likely problems or that an intervention needs to be modified. So, we need to combine both approaches – welcome to mixed methods research.

The issue then, is not whether research can be entirely participatory or objectively detached, but how to combine the subjective and objective elements.

Two points stand out. First, they must be independent pieces of work, as far as data collection is concerned. The findings can be brought together at a later stage, but data collection should be independent. If members of the public participate in data collection then they must follow the principles laid down in the protocol (see previous News Blog for article on this point).[4] The second point, is that some sort of participant inquiry should be part and parcel of most interventions, because this enables them to be adapted to local circumstances. But such inquiry should be considered as part and parcel of the intervention, not of its summative evaluation. This ‘intra-mural’ research could be considered as a “formative evaluation”. Any summative evaluation should stand outside of the formative evaluation; it should be as objective as possible. Both formative and summative evaluations may contain qualitative and quantitative elements, but the former are likely to predominate in formative evaluations while the latter are more predominant in summative evaluations. This argument is laid out in more detail elsewhere.[5]

— Richard Lilford, CLAHRC WM Director


  1. Paley J & Lilford RJ. Qualitative Methods: an Alternative View. BMJ. 2011; 342: d424.
  2. Daston L & Galison P. Objectivity. New York: Zone Books; 2007.
  3. Mayo DG. Error and the growth of experimental knowledge. Chicago: University of Chicago Press; 1996.
  4. Lilford RJ. Patient and Public Involvement in Data Collection. NIHR CLAHRC West Midlands News Blog. 18 August 2017.
  5. Lilford RJ, Foster J, Pringle M. Evaluating eHealth: How to Make Evaluation More Methodologically Robust. PLoS Med. 2009; 6(11): e1000186.