Tag Archives: Director’s Choice – From the Journals

Diagnosis, Diagnosis, Diagnosis

A new study from Ann Arbor confirms the crucial importance of diagnosis.[1] This study, based on analysis of no less than 63,000 settled inpatient malpractice claims shows that 22% were diagnosis-related. Moreover, the diagnosis-related claims were about twice as likely as other types of settled claim to result in permanent harm or death. Of course, malpractice claims represent only a small portion of all safety incidents. Nevertheless, this paper should be assimilated with other evidence, such as the high proportion of misdiagnoses at post mortem, even among thoroughly scrutinised patients who died in intensive care.[2] Some diagnostic errors result from a lack of vigilance, but diagnosis is hard when a condition is uncommon, or when a common condition presents in an uncommon way. And beware the unknown unknowns. At one time the BMJ produced a list of ‘easily missed diagnoses’, such as myasthenia gravis and Hirschsprung’s disease. Some eminently treatable diseases that are easily missed were listed in a previous blog [3] – Addison’s disease, mastocytosis, Lyme disease, Kawasaki’s disease, leaking aneurisms, and right-sided endocarditis, for example. You might want to add your own. TB, the great mimicker; drug overdose / interactions; and nutritional deficiencies are often missed.

— Richard Lilford, CLAHRC WM Director

References:

  1. Gupta A, Snyder A, Kachalia A, Flanders S, Saint S, Chopra V. Malpractice claims related to diagnostic errors in the hospital. BMJ Qual Saf. 2018; 27: 53-60.
  2. Winters B, Custer J, Galvagno SM, et al. Diagnostic errors in the intensive care unit: a systematic review of autopsy studies. BMJ Qual Saf. 2012; 21: 894-902.
  3. Lilford RJ. Diagnosis, Diagnosis, Diagnosis. Richard Lilford’s Friday Blog. February 15, 2013.
Advertisements

Vaccine in Pill Form

Storage of vaccines is an issue faced in a number of low- and middle-income countries due to the need for constant refrigeration. Some sites may have intermittent power or outages, they may simply not have the storage space for the necessary number of vaccines, or there may be difficulty in maintaining a cold temperature during transportation. A recent proof-of-concept study by Miles, et al.[1] may offer hope in the future though. The team based at Cardiff University produced a prototype oral polypeptide vaccine that was able to provide protection against the ‘flu in a mouse model. Previous polypeptide vaccines had poor stability, which hindered transport and thus therapeutic potential. For this vaccine the researchers created highly stable antigenic ‘mimics’ that did not exist in nature. It was stable in both human serum and gastric acid, was able to stimulate and prime an immune response specific to human ‘flu that was as effective as the standard biological equivalent, and remained immunogenic after being administered orally. Although there is still quite a way to go, it is worth keeping an eye on development of these encouraging findings.

— Peter Chilton, Research Fellow

Reference:

  1. Miles JM, Tan MP, Dolton G, et al. Peptide mimic for influenza vaccination using nonnatural combinatorial chemistry. J Clin Invest. 2018.

Relationship Between Health Workers and Preventable Disease

In this 2011 study from Chile,[1] Carla Castillo-Labordee regresses the proportion of health workers per 10,000 people in a country against the burden of preventable disease as measured in disability-adjusted life years (DALYs). She finds a negative association for health service personnel in general and doctors in particular. That is to say, the greater the proportion of doctors, the lower the burden of preventable disease. The effect was observable across diseases as a whole, communicable disease, and in maternity care. However, the effect was not observed specifically for trauma or non-communicable chronic disease. The associations were non-linear and, as for the effect of per capita GDP and health, strong effects are only seen at low baseline levels of the explanatory variable. The author sought to disentangle general socioeconomic development from human resources by controlling for the former. But I don’t think this is really possible since not only are interactions between these variables highly likely, but they may be entangled on the same causal chain. In short, they may be underestimating the effect of human capital inputs to health systems.

— Richard Lilford, CLAHRC WM Director

References:

  1. Castillo-Labordee C. Human resources for health and burden of disease: an econometric approach. Hum Res Health. 2011; 9: 4.

The Affordability of Care – Hard to Measure but Increasingly Important

Traditionally epidemiologists who worked on the relationship between wealth and disease were concerned with the effect of the first on the second. But, of course, disease can affect wealth, and economists like Jeffrey Sachs spotted the resulting circularity: poverty -> disease -> more poverty -> more disease. Increasingly, clinicians have started to worry about the catastrophic costs of disease and my colleague Bertie Squire from Liverpool School of Tropical Medicine is searching for treatment pathways to mitigate the financial consequences of recurrent tuberculosis. The Oregon experiment, reported in your News Blog,[1] shows that the most obvious benefit from extending insurance coverage to the un-insured lies in reducing the incidence of catastrophic loss.

Catastrophic loss:Events whose consequences are extremely harsh in their severity, relating to one or more losses such as bankruptcy, total loss of assets, or loss of life.” (The Law Dictionary, 2017).

An important question then, is how generous can publically financed insurance be? Or, to put the question another way, how can the affordability of health care be measured? This is a rather different question to that of the affordability of a particular new technology – a question of its Incremental Cost Effectiveness Ratios. This is because HTA is designed to determine the upper bound on ‘affordability’, while the fiscal question of affordability as a whole is concerned with total expenditure.

A paper in a recent issue of JAMA proposes an approach based on the total health costs divided by the median household income.[2] This might be a useful rule of thumb, but it is beset by problems, as pointed out in two leading articles.[3] [4] One such problem arises from the observation that some of the costs of health care / insurance premiums likely come out of household incomes – companies would probably pay employees more if it were not for the insurance premiums – so there is some double counting going on. More fundamentally, affordability cannot be inferred simply by the proportion of expenditure going on health care. One could argue, for instance, that the richer the country (higher the per capita GDP), the greater should be the expenditure on health. One way to get at the affordability construct would be to examine the cost of health care as a proportion of money left over after subtracting the ‘essentials’ of housing, food, clothing and transport to and from school / work. Another would be to calculate the effect of health care costs on how many families tip over into bankruptcy or teeter on the edge thereof. Unaffordability would still vary by type of family and type of insurance system, especially in a variegated health system like that in the USA. A simple number, like proportion of GDP spent on health, can only give a very coarse-grained idea of the consequences of increasing or decreasing the proportion of resources dedicated to health care. It is also important to consider the effect of high health care costs on the broader economy. There is always a danger that, absent price signals, the allocation to health will exceed what can be justified in terms of the benefit realised. That is to say that, given information asymmetries, health care will be driven more by provider than consumer needs.

— Richard Lilford, CLAHRC WM Director

References:

  1. Baicker K, Taubman SL, Allen HL, et al. The Oregon Experiment – Effects of Medicaid on Clinical OutcomesN Engl J Med. 2013; 368: 1713-22.
  2. Emanuel EJ, Glickman A, Johnson D. Measuring the burden of health care costs on US families: the Affordability Index. JAMA. 2017; 318(19): 1863-4.
  3. Antos J, Capretta JC. Challenges in Measuring the Affordability of US Health Care. JAMA. 2017; 318(19): 1871-2.
  4. Reinhardt U. What Level of Health Spending Is “Affordable?” JAMA. 2017; 318(19): 1869-70.

Do Poor or Rich People Consult Health Services Most Often in Low- and Middle-Income Countries?

In a previous news blog [1] we argued that many poor people in cities within low- and middle-income countries (LMICs) bypass local facilities in slums and proceed directly to outpatient departments at local hospitals. They are clearly prepared to trade convenience and time to obtain what they perceive as a superior service. Now it turns out that poor people are, at least in some places, heavier users of health services than rich people. This is because they are less healthy than richer people, do I hear you say? Well, it would appear not. In their outstanding study in New Delhi, Das and Sánchez-Páramo [2] found that poor people consult more frequently for time-limited single episode diseases than richer people. And they do so despite out-of-pocket payments. What is going on here? Their behaviour is contrary to economic theory, whereby the marginal utility of a service should be less attractive to people with smaller disposable incomes, and this should be particularly the case when utility gain is small. This counter-intuitive finding reminds me of the observation that very poor people will pay for radios or television even while they are still hungry, in direct contradiction of Maslow’s hierarchy of needs.

The authors speculate the reasons for high consulting rates for short duration conditions. They consider the possibility that many short duration conditions are really manifestations of chronic diseases causing short-term complications; boils and thrush in patients with diabetes, etc. This seems unlikely because the gradient in consultation rates between rich and poor people is the same among young and old people, despite the steep gradient in prevalence of chronic disease by age. The authors think the most likely explanation lies in information asymmetry – the poor are less likely to buy their own medicines and this could reflect lower knowledge levels. I have another explanation – that low income and high consultation rates reflect the same latent variable. Some people, according to this theory, have generally less resilient personalities, find it harder to obtain and retain jobs, and are more likely to seek solace in consultation. This is a hard hypothesis to test in cross-sectional studies, but it could be studied in longitudinal studies or possibly using Mendelian randomisation. No-one should interpret any confirmatory findings as victim-bashing – rather the reverse; it provides society an opportunity to demonstrate its caring attitude.

— Richard Lilford, CLAHRC WM Director

References:

  1. Lilford RJ. A Heretical Suggestion. NIHR CLAHRC West Midlands News Blog. 9 February 2018.
  2. Das J & Sánchez-Páramo C. Short but not Sweet: New Evidence on Short Duration Morbidities from India. Policy Research Working Paper 2971. Washington, D.C.; World Bank Development Research Group. 2003.

What Do You Think When You Hear ‘Scientist’?

If you have spent much time in universities you may have seen various stickers or leaflets raising awareness of campaigns that support women in STEM fields (science, technology, engineering and mathematics). There has been a push in recent years to get more girls and women into STEM subjects. Fifty-two percent of those who graduated in STEM disciplines in 2014 were female.[1] This varies widely between disciplines though, with females making up around 80% of graduates in subjects allied to medicine or veterinary sciences, but only around 15% in computer science or engineering and technology. While the gender balance of all STEM graduates are roughly equal, this is not reflected in employment however, with figures suggesting around 23% of employees in UK STEM industries are female,[1] while data from the UNESCO Institute for Statistics less than 30% of scientific researchers worldwide are female.[2]

Does the future hold more promise? A meta-analysis by Miller and colleagues looked at fifty years worth of studies where school children were asked to draw a scientist and examined the genders depicted.[3] They found that over time the percentage showing female scientists has increased – from 0.6% in data collected in 1966-77 to around 40% in 2015. However, when looking at the age of children (in studies since the 1980s) they found that while there was roughly equal representation of male and female scientists among 5 and 6 year olds, by the age of 7-8 years significantly more men were drawn. In the drawings made by girls only, the switch from predominantly female to male depictions happened around 10-11 years. Perhaps with an increase in female representation in STEM roles, especially in public, then young girls might be more likely to see themselves in such a field and thus increase the proportion in the workplace. Equally more needs to be done to emphasise gender equality at these key developmental milestones.

— Peter Chilton, Research Fellow

References:

  1. WISE Campaign for Gender Balance in Science, Technology & Engineering. Women in STEM workforce 2017. 24 October 2017.
  2. UNESCO Institute for Statistics. Women in Science. Fact Sheet No. 43. March 2017.
  3. Miller DI, Nolla KM, Eagly AH, Uttal DH. The Development of Children’s Gender-Science Stereotypes: A Meta-analysis of 5 Decades of U.S. Draw-A-Scientist Studies. Child Development. 2018.

Neither Child Nor Adult: the Science, Sociology and Literature of Adolescence

Adolescents often get a bad press. The teenage years can be very troubled and, as a parent I can say, troubling. For a rich, tender and insightful view News Blog readers may have encountered ‘My Brilliant Friend’ by Elena Ferrante.

The period of adolescence seems to be widening: at one end, puberty is occurring ever earlier; at the other, the latest neurophysiology findings suggest that brain maturation continues to the age of 24.[1][2]

The topic of adolescence is heavily featured in the most recent edition of Nature.[3] The argument is made that the topic has been relatively under researched and that we, as a society, should adopt a more positive attitude to this crucial phase of life. While childhood sets many of the patterns for later life, adolescence  is also a crucial formative stage, especially as far as mental health is concerned.

This has been long recognized at CLAHRC WM, where we have a thriving scientific programme focusing on psychoprophylaxis, and increasing portion of our work is carried out in schools.

— Richard Lilford, CLAHRC WM Director

References:

  1. Patton GC, Olsson CA, Skirbekk V, et al. Adolescence and the next generation. 2018; 554: 458-66.
  2. Ledford H. Who exactly counts as an adolescent? 2018; 554: 429-31.
  3. Dahl RE, Allen NB, Wilbrecht L, Suleiman AB. Importance of investing in adolescence from a developmental science perspective. 2018; 554: 441-50.

Does Pet Ownership Make Us Healthier?

Like many of us, I love animals, and am crazy about both dogs and cats. It is almost hard to describe how much I love them. But is this all very good for me? I mean, dogs can bite, cats scratch, and all animals can transmit infections. On the other hand, they are psychotropic; cuddly, warm and attentive. Some evidence suggests that people who keep pets are healthier than those without.

However, a rather dismal paper in the BMJ puts paid to all of that.[1] According to an analysis of a large cohort study of aging individuals, those who own pets do no better than those who do not. They are no stronger, happier or otherwise healthier than people without pets. Still, I would like to live with a cat or dog.

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Batty GD, Zaninotto P, Watt RG, Bell S. Associations of pet ownership with biomarkers of ageing: population based cohort study. BMJ. 2017; 359: j5558.

A Very Clever Study Using Instrumental Variable Analysis

In observational studies, adding a third or fourth anti-hypertensive medication leads to rapidly diminishing marginal improvements in blood pressure and an increase in side effects. However, this finding could originate from a type of selection bias called bias by indication; the harder cases get more treatments. To get around this problem an Instrumental Variable (IV) analysis was carried out on a dataset generated by a randomised trial of intensive versus standard blood pressure target.[1]

In the IV analysis, where randomised group was the instrument, the incremental effect of adding a further anti-hypertensive was maintained across all orders of increment. Not only was there an improvement in blood pressure, but also a reduction in cardiovascular events. There was only a small increase in adverse events. This proves that there is a bias by indication in observational studies, which reduces the apparent marginal benefit and increases marginal harms. This is a neat study and a fine example of IV analysis.

— Richard Lilford, CLAHRC WM Director

Reference:

  1. Markovitz AA, Mack JA, Nallamothu BK, Ayanian JZ, Ryan AM. Incremental effects of antihypertensive drugs: instrumental variable analysis. BMJ. 2017; 359: j5542.

Interruptions Lead to Errors

I thank Julian Bion for drawing my attention to an interesting paper by Joanna Westbrook and colleagues in BMJ Quality and Safety.[1] In this meticulous study, 36 emergency physicians were shadowed for over 120 hours. When physicians experienced interruptions or were tired they made many more prescription errors than when they were not interrupted or were refreshed. This study corroborates the existing psychological literature and supports efforts to try to create a calm and ordered environment in which physicians can do complex tasks.

— Richard Lilford CBE, CLAHRC WM Director

Reference:

  1. Westbrook JI, Raban MZ, Walter SR, Douglas H. Task errors by emergency physicians are associated with interruptions, multitasking, fatigue and working memory capacity: a prospective, direct observation study. BMJ Qual Saf. 2018.