Biology never ceases to surprise: preventing cancer of the ovary by tubal ligation

One might have thought that ovarian cancer can be prevented by removing the ovaries – say at the time of hysterectomy – and that obstructing or removing the fallopian tubes would not, by itself, reduce the risk of ovarian cancer. These hypotheses are based on the plausible assumption that cancer arises in the ovaries, just as breast cancer arises in the breast. However, it now appears that ovarian cancer may arise in the fallopian tubes, at least in a substantial proportion of cases. A Swedish record linkage study shows that fallopian tube ligation is associated with a 30% reduction in the incidence of ovarian cancer.[1] Confounding by high fertility do I hear you say? Apparently not as this has been adjusted for. Caused by preventing access of carcinogens ascending the reproductive tract? Probably not, since removal of the fallopian tube provides even stronger protection against ovarian cancer, than does ligation which leaves the ovarian end of the tube in situ. Implausible hypothesis reminiscent of transubstantiation? Apparently not, since cells with the appearance of ovarian cancer have been harvested from fallopian tubes, and the molecular signature of many ovarian cancers suggests a fallopian tube provenance.

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Falconer H, Yin L, Grönberg H, Altman D. Ovarian cancer risk after salpingectomy: a nationwide population-based study. J Natl Cancer Inst. 2015; 107(2): dju410.
Advertisements

Integrated Care

A recent BMJ paper [1] from Stephen Shortell and collaborators from the King’s Fund discusses the design of models of care to improve integration between hospital and community, and between health and social care – an old chestnut. They have a taxonomy of integrated care models that I represent like this:

DC - Integrated care Fig1

Model 1 seems to locate responsibility for integration mostly with community providers, while model 2a evokes a structural solution in which hospital and community providers work in an organisation straddling hospital and community. The article comes down in favour of model 2b and gives successful examples from America (where quality has improved at reduced cost).[2][3][4][5] The article emphasises the importance of integrated computer care records, almost saying this is a necessary ingredient.

On this latter point, the CLAHRC WM Director begs to differ – as argued in a previous post, a patient-held paper record has considerable advantages over ‘all singing and all dancing’ IT systems. He does agree, however, with the idea of an integrated record (not necessarily computer-based), the authors’ emphasis on ‘clinical integration’, and the need to win the hearts and minds of service providers.[6] CLAHRC WM is involved with two grant applications to help develop the tacit skills needed to care for patients with multiple morbidities, and different needs and preferences, across multiple types of care provider in different locations.

— Richard Lilford, CLAHRC WM Director

References:

  1. Shortell SM, Addicott R, Walsh N, Ham C. The NHS five year forward view: lessons from the United States in developing new care models. BMJ. 2015; 350: h2005.
  2. Centers for Medicare and Medicaid Services. Factsheet: Medicare ACOs continue to succeed in improving care, lowering cost growth. 2014.
  3. McWilliams JM, Chernew ME, Landon BE, Schwartz AL. Performance differences in year 1 of pioneer accountable care organizations. N Engl J Med. 2015. [ePub].
  4. Song Z, Rose S, Safran DG, Landon BE, Day MP, Chernew ME. Changes in healthcare spending and quality four years into global payment. N Engl J Med. 2014; 37: 1704-32.
  5. Markovich P. A global budget pilot project among provider partners and Blue Shield of California led to savings in first two years. Health Aff. 2012; 31: 1969-76.
  6. Curry N, Ham C. Clinical and service integration. The route to improved outcomes. London: King’s Fund, 2010.

Calling All Systematic Reviewers

The CLAHRC WM Director is provoked by the ever increasing – indeed, exponentially increasing – number of articles returned by standard literature searches. At this rate, screening all the articles identified by a typical search will be all but impossible within two decades. Some form of systematisation is necessary.

A start has been made in clinical research through the creation of the McMaster Premium LiteratUre Service (PLUS) database of pre-appraised clinical studies. PLUS is generated by manually reviewing 120 clinical journals for high-quality articles using a reproducible selection process. A paper comparing PLUS with 89 recent Cochrane reviews,[1] found that while PLUS contained fewer articles, restricting searches to PLUS did not change the conclusions of any of the Cochrane reviews included in the sample.

The PLUS database is a start, but:

  1. it still relies on manual review;
  2. it is confined to clinical research.

A method is urgently required to:

  1. improve coding of topic and study type;
  2. automate compilation of bibliographies;
  3. cover health and social care as a whole.

Literature retrieval processes will be radically different in two decades – they will have to be.

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Hemens BJ, Haynes RB. McMaster Premium LiteratUre Service (PLUS) performed well for identifying new studies for updated Cochrane reviews. J Clin Epidemiol. 2012; 65(1): 62-72.

 

Use of WHO Surgical Check List

We thank Mary Dixon-Woods for drawing our attention to an interesting article on the use of the fabled WHO surgical checklist.[1] Interesting because the topic is important and because the authors used a step wedge, cluster, experimental design as they introduced the intervention across different surgical specialities in two Norwegian hospitals. Step wedge designs need to avoid pitfalls of all cluster studies related to interaction between intervention and willingness to be recruited. The neat way out of this conundrum is to use routinely collected data and enter everyone in the cluster. That was done here. It is important to control for systematically later time periods in the intervention ‘cells’ of the step wedge and, again, the authors did so. So what did they find in this procedurally satisfactory study? A large and statistically significant intervention effect was observed. This is in keeping with many, but not all, previous studies of the conventional checklist.

My problem lies in the underlying hypothesis; as a previous surgeon I find the theoretical basis for the checklist unconvincing. In other words I start from a sceptical prior that is reluctantly being pulled towards a more optimistic estimate. Also, I fret over publication bias in the social science/service delivery literature, as discussed in a previous post. All the same a sceptic like me cannot ignore these positive results. So how may they work. Firstly, the word “checklist” may be a misnomer. It may just be a convenient focus around which to engender a positive and professional personal and team approach. This could explain why it sometimes works and sometimes does not. The idea here would be that it can’t work when: 1) attitudes are totally hostile, or 2) practice is already very good so there is little headroom for improvement. In that case it would be like any behavioural intervention – it will work among those who are receptive to improvement, but not yet improved. It is also possible that use of the checklist, even in a tokenistic way, will be effective in the very long term. Here I rely on the theory of cognitive dissonance.[2] People who start with ritualistic tokens of compliance are inclined to either stop complying or move their attitudes towards their outward actions, if I understand correctly. Comments welcome.

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Haugen AS, Søfteland E, Almeland S, et al. Effect of the World Health Organization Checklist on Patient Outcomes: A Stepped Wedge Cluster Randomized Controlled Trial. Ann Surg. 2015; 261 (5): 821-8.
  2. Festinger L. A Theory of cognitive dissonance. Stanford, CA: Stanford University Press, 1957.

Health care in a parallel world: the Birmingham screwdriver

Imagining health care in a parallel world can reveal a lot about the health care system we enjoy in this one.

Counterfactual narratives have long been popular. Livy speculated about a confrontation between Rome and Alexander the Great, had the latter chosen to expand his empire westwards instead of eastwards.[1] Kingsley Amis wrote about a world where the reformation failed and Roman Catholicism continued to dominate Europe for centuries.[2] An alternate health service seems a minor alteration in comparison. What would it look like?

I recently took part in a workshop on type 2 diabetes in adolescents and young adults. Most of the speakers were medical researchers, and the audience clinicians. I was the public health afterthought. The talks focused on pathophysiology of type 2 diabetes, speculating on whether South Asians might exhibit a distinct illness trajectory to Europids. This effortlessly morphed into a speculative discussion of genetics. The medical academics were fascinated, leaning forward on their seats, vying with each other to interject. A single question about whether South Asians and White British might possibly have different lifestyles was brushed aside. The essential genetic homogeneity of the human species compared to its great ape cousins was ignored,[3] (see also our previous blog).

Although irrelevant to patients, pathophysiology and genetics fascinate doctors because they are the core of our undergraduate professional training. Pharmacological treatments predominate our therapeutic thinking because they are the logical response to pathophysiology. Doctors enjoy a near monopoly on prescribing and it is the defining and distinguishing feature of the profession. As the profession is a key influencer of the health services and research agendas, the ability to deliver the right drugs to the right patients is a central preoccupation of the health care system and the understanding of pathophysiology in order to develop and test drugs dominates the research agenda. From my background reading on the public health aspects of diabetes I learned that only 16.7% of newly diagnosed type 2 diabetics are offered structured diabetes education and only 3.6% attend.[4] How could an important and effective intervention be afforded such a low priority?

In another world a profession of health educators is in the ascendant. The profession dominates the provision of health care. Clearly the most important intervention for anyone developing a chronic disease is structured education. This conveys factual information about prognosis, life skills, confidence, and self-efficacy. The first intervention follows diagnosis. It serves an anthropological, as well as an educational, purpose, marking a life transition into a new state. Ongoing education reinforces skills, builds knowledge, and addresses the disease progression. Alongside service delivery, a vigorous research agenda constantly refines the educational interventions. New educational materials are developed. Innovative modes of delivery test new communication technologies, gamification (the use of game thinking and mechanics in non-game contexts to engage users), and virtual learning communities. Patients become co-producers of educational interventions. Stratified education is emerging where psychometric testing and preference elicitation allows patients to be matched to the most appropriate educational intervention. The primary outcomes of health care are the same: quality of life and length of life. The process measures by which we mark our progress are very different: self-efficacy, knowledge, and measurable skills replace physiological parameters. Even the typology of disease might change, with categories defined by the type of educational intervention as much as by pathophysiology.

What does this tell us? Sometimes they are so ingrained, we can’t see our own assumptions. The French call this déformation professionnelle. To a man with a hammer, everything looks like a nail. Which is why a hammer was referred to as the Birmingham screwdriver.

— Tom Marshall, Co-Director CLAHRC WM, Prevention and Detection of Diseases

References:

  1. LiviuS T. The History of Rome (book IX, sections 17–19). English Translation by Rev. Canon Roberts. New York, NY: E.P. Dutton and Co. 1912.
  2. Amis K. The Alteration. London: Jonathan Cape. 1976.
  3. Prado-Martinez J, Sudmant PH, Kidd JM, et al. Great Ape Genetic Diversity and Population History. Nature. 2013; 499: 471-5.
  4. The Healthcare Quality Improvement Partnership (HQIP). National Diabetes Audit 2012 – 2013. Report 1: Care Processes and Treatment Targets. Leeds: Health & Social Care Information Centre. 2014

Assessing Publication Bias in Social Sciences – a Critically Important Paper from Science

Publication bias means that null results do not make it into the public domain. Assessing publication bias is straightforward in subjects where all studies have to be registered in advance – clinical trials for example. But there is little evidence on publication bias in service delivery / health services research. The CLAHRC WM Director suspects that this lack of evidence arises because much social science literature is observational rather than experimental, and it is so hard to collect convincing evidence on publication bias among such studies. There is no registry of studies; the original hypothesis may not correspond to comparisons reported; many studies might not be written up; and the investigators may evaluate a large number of associations so that results do not neatly dichotomise into significant or null. In addition, the famous funnel plot may be less likely to signal bias than is the case for much clinical research. This is because the association between sample size and risk of publication bias is less likely to hold when the size of the sample is limited more by the size of the database than the cost of recruiting individual participants. These problems were overcome in an interesting article that studied the destiny of 249 grant-funded (peer review) studies conducted within a single ongoing data collection survey over a ten year period.[1] Most of the studies consisted of an evaluation of modifications of the survey instrument (questionnaire) used to populate the survey database. The results show a massive effect. Studies with a positive result (as judged by the author) were much more likely to be written up and, if written up, much more likely to be published. The fact that the source studies were all based on a single database removes (or at least strongly mitigates) bias due to interaction between study topic and probability of a positive result.

These results reinforce the CLAHRC WM Director’s weariness to accept positive results of association studies, such as those that relate patient perception of care to standardised morality rates. Such results feed into the prevailing meta-narrative, in this case that organisational culture determines the quality of the full range of front line services. A null result is less likely to survive peer review under such circumstances. The paper cited here interviewed holders of grants based in the database, and found that they were disheartened by null results and often did not bother to submit them, anticipating that they would be rejected. They are right to be pessimistic since null results were less likely to be accepted when submitted, in keeping with the natural human tendency to reject studies that do not fit with prevailing or preconceived ideas.[2] [3]

What do we recommend? Only studies where the protocol has been published should be considered for publication, and they should all be published provided the protocol was adhered to. The clinical research world has tightened up its act. It is high time for the service delivery world to stop claiming scientific exceptionalism and adhere to the standard tenets of good scientific practice that hark back to Francis Bacon.[4]

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Franco A, Malhotra N, Simonovits G. Publication bias in the social science. Science. 2014; 345(6203): 1502-5.
  2. Lord CG, Ross L, Lepper MR. Biased Assimilation and Attitude Polarization: The Effects of Prior Theories on Subsequently Considered Evidence. J Pers Soc Psychol. 1979; 37(11): 2098-109.
  3. Kaptchuk TJ. Effect of interpretive bias on research evidence. BMJ. 2003; 326: 1453-5.
  4. Bacon F. Novum Organum. 1620.

Emergence of Antimicrobial Resistance is Much Greater in Low- than High-Income Countries

This recent article in Science [1] gives five evidence-based reasons for greater emergence of, and more devastating consequences from, antibiotic resistance in low-income countries (LICs) compared to higher income countries:

  1. LICs harbour a higher rate of extremely virulent organisms, such as those responsible for typhoid fever and tuberculosis.
  2. Antimicrobials are (even) less carefully regulated and are frequently available ‘over the counter’.
  3. Treatment for established infection is more often delayed and access to supportive care for life-threatening infections is less widely available.
  4. Laboratory testing for the infectious agent is less widely available and therefore ‘syndromal’ treatment based on suspicion, rather than confirmation of a bacterial cause for ill health, is understandably widespread.
  5. Routine dosing of animals to improve yields is more widespread.

Of course, increasing resistance to antibiotics in LICs impacts on high-income countries, and the article shows how molecular tracking of microbial genetics has enabled the dissemination of resistance genes to be mapped across the world.

The CLAHRC WM Director thinks that new molecular genetic techniques will help target antibiotics at the correct diagnosis, and also help in controlling the spread of resistant organisms. The technology is becoming ever cheaper, as discussed in a previous blog.

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Baker S. A return to the pre-antimicrobial era? Science. 2015; 347(6226):1064-6.

More Evidence for Short Doses of Antibiotics in Infection

The CLAHRC WM Director was always taught at Medical School to provide long courses of antibiotics (7–10 days) and try to maximise compliance throughout. This was thought to reduce the emergence of resistance. Hedrick et al.[1] add to the evidence that short courses are less likely to result in antimicrobial resistance. So what’s best for the patient – a fixed short course or wait until the patient ‘responds’? Surprisingly, no difference in outcome was observed in Hedrick’s study. This would explain why the physiological response to infection appears to outlast the effective phase for antibiotic effectiveness. Most of the symptoms of infection result from the immune response to infection, not the infection itself. The CLAHRC WM Director hypothesises that persistence of symptoms beyond the effective phase of antibiotic use is caused by the bodies ‘mopping up’ exercise, where the immune system is eradicating dead or damaged bacteria.

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Hedrick TL, Evans HL, Smoth RL, et al. Can we define the ideal duration of antibiotic therapy? Surg Infect. 2006; 7(5): 419-432.

Increased Weekend Mortality

The subject of increased weekend mortality for patients admitted over the weekend has been mentioned in this blog previously. This has been attributed to reduced availability of consultants over the weekend. However, a causal link between reduced consultant cover and worse outcomes is unproven. If consultant availability is the main factor behind the weekend effect, then we would expect to see a surge in mortality if consultant presence dropped over the working week. Jena et al. (2015) [1] studied mortality rates of patients admitted for myocardial infarction, cardiac arrest, or heart failure during normal times compared to periods when national cardiology meetings were taking place in the USA. A sizeable proportion of heart specialists down tools to attend the conference. Surprisingly, not only were adjusted 30-day mortality rates not increased, but they were lower among high-risk patients admitted during meetings compared to those admitted at other times. The American College of Cardiology responded by saying they were reassured that during dates of national meetings, patients received care that was no worse than normal. But is the lower risk really reassuring if you are a heart specialist! The CLAHRC WM Director reflects that the premise behind the paper may be wrong and sufficient specialists, or near specialists, stay behind to manage the acute service over the conference period.

— Samuel Watson, Research Fellow

Reference:

  1. Jena AB, Prasad V, Goldman DP, Romley J. Mortality and Treatment Patterns Among Patients Hospitalized With Acute Cardiovascular Conditions During Dates of National Cardiology Meetings. JAMA Intern Med. 2015; 175(2): 237-44.

An Argument to Give Family Physicians / GPs a Larger Role in Hospital Care

This article in the New England Journal of Medicine suggests a larger role for family physicians / GPs in hospital care.[1] It suggests that they should do rotations in which they work on the hospital staff. Alternatively, they should join ward rounds or visit patients in hospital. The suggested advantages are two-fold. Firstly, family doctors could influence care (for the better!) by advising hospital staff. Secondly, this contact would facilitate a smooth transition to the community after discharge. Aware of time constraints, they suggest virtual visits to the hospital through teleconferencing. The CLAHRC WM Director remains sceptical. Such an expanded hospital role for family doctors will be time-consuming even with teleconferencing, and the opportunity costs are not considered in the article.

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Goroll AH, & Hunt DP. Bridging the Hospitalist­–Primary Care Divide through Collaborative Care. N Engl J Med. 2015; 372(4): 308-9.