Tag Archives: BJGP

Qualitative research and the BJGP

Kath photo

Kath Checkland (@khcheck) is a GP and a Professor at the University of Manchester. She is a passionate advocate of the value of qualitative research, and is a member of the BJGP editorial board.

On Saturday, the British Medical Journal published an open letter, signed by 75 senior academics (of which I was one), calling for the journal to rethink its current stance on the publication of qualitative research. The letter was prompted by the publication on Twitter of an extract from a rejection letter stating: ‘I am sorry to say that qualitative studies are an extremely low priority for The BMJ. Our research shows that they are not as widely accessed, downloaded, or cited as other research.’

This stance represents a change in policy for the BMJ. Not only has the journal published some important and influential qualitative papers (for example see Gabbay et al 2004), but it was also at the forefront of promoting the use of qualitative methods in health research, publishing a number of highly-regarded series of ‘education and debate’ papers exploring the value of qualitative methods and providing a superb introduction to a variety of methods for the novice researcher (Pope and Mays 1995, Pope et al 2000, Pope and Mays 2009). In what would seem to represent a significant change in policy, an editorial accompanying the letter argues that: ‘qualitative studies are usually exploratory by their very nature and do not provide generalisable answers’, and puts them together with case reports, surveys of self-reported practice etc as studies less likely to ‘change clinical practice and help doctors make better decisions’.

The ‘rapid responses’ to the letter indicate the depth of support for the stance taken by the signatories. Qualitative and quantitative researchers have added their voices, highlighting the absurdity of an editorial position which selects articles by study design rather than by quality, importance of the topic or relevance of the findings. In their original claims and in their response to the letter the BMJ editorial board demonstrate a number of misunderstandings about the nature of research more generally, not only about that done using qualitative methods. These include:

  1. An apparent belief that quantitative studies can provide definitive answers to healthcare problems that will straightforwardly ‘change practice’. Some of those responding to the BMJ article have pointed out the weakness of this position. All scientific results are provisional, and clinicians changing their practice on the basis of a single randomised trial would be rightly accused of being reckless. Furthermore, the approach espoused would seem to equate statistical generalisability with real world generalisability.
  2. A failure to appreciate the complexity of service delivery. Recent MRC guidance re-emphasises the fact that even when a trial result is confirmed and replicated, many important questions still remain (Moore et al 2015). To what extent was the study population highly selected? How might the findings translate in the real world? For other study designs such as quantitative observational studies, further research is needed to explore causation and to try to delineate the relevant contexts in which particular effects might occur. Implementation research is a huge field, and it is rarely the case that the most important studies use quantitative methods. Understanding how human beings behave within healthcare organisations is vital if clinical services are to be improved, and understanding how patients appreciate and value treatments and types of services, as well as different aspects of their lives is essential if overall care is to be improved. Such research topics are not trivial or minor, and it simply is not the case that research which explores these issues can never be generalisable, of high impact or change practice. It is also not the case that BMJ readers are unlikely to find such research interesting to read or valuable in shaping their practice.
  3. A misunderstanding of the nature of qualitative evidence, confusing a method with methodology. While there is undoubtedly a place, as many responses to the letter have highlighted, for qualitative elements exploring aspects of trials of complex interventions, such research represents a tiny proportion of the possible applications of qualitative techniques. Not only does qualitative research answer different types of questions (‘how, why and in what circumstances?’ rather than ‘what?’), but it also embodies a wide range of philosophical approaches and theoretical perspectives. Just because research is not statistically generalisable does not mean that its findings cannot have wider relevance. Theoretical generalisation means that research speaks to a wider body of theory which applies beyond the immediate context, and research using qualitative methods may use theory to generate findings which are highly generalisable. For example, the work done by May and colleagues on Normalisation Process Theory (May 2009) provides a well-researched and widely applicable framework within which to think about how change occurs and how it can be supported in clinical practice. One could argue that such work is more likely to generate sustained and embedded change in clinical practice than any meta-analysis or trial.

In the 1970s and 1980s social scientists engaged in the so-called ‘paradigm wars’ in which those of different theoretical persuasions debated (sometimes hotly) their different approaches, beliefs and assumptions. In general, within social science such wars have long been over, with scholars embracing different paradigms and world views respecting one another’s work and working together without rancour. It is sad to see the BMJ apparently reopening those debates.

To suggest, as the BMJ editorial does, that qualitative work can rarely be of interest to a general medical audience, and that it represents a niche approach which should be published in ‘more specialised’ journals does a disservice to the journal’s many readers who understand and value the diverse contributions which research using qualitative methods may offer, as well as to its own honourable history.

Here at the BJGP we take a different view, and would encourage readers of the BMJ to diversify their reading to explore some of the challenging, highly pertinent and practice-changing research that we publish. In the last 6 months alone we have reported: useful evidence for GPs trying to support those bereaved by suicide ( et al 2016); recommendations as to how practices should support patients to self-monitor their anti-coagulation therapy (Tompson et al 2015); and evidence from a realist review delineating when and how digital communication methods might support access to care for marginalised patients (Huxley et al 2015). For readers, these three examples — selected at random — all address significant clinical issues, and have the potential to change important aspects of clinical care. For researchers, we offer the opportunity to have your study tested and commented on by knowledgeable reviewers skilled in a wide variety of social science disciplines and approaches, providing an invaluable service for those we reject as well as those we publish.

Top 10 most read BJGP research articles published in 2015

16Jan_Top10_research_2015_BJGP_smThese are the top 10 most read research articles based on full text downloads from bjgp.org.

1. Child obesity cut-offs as derived from parental perceptions: cross-sectional questionnaire.
http://bjgp.org/content/65/633/e234

Parental perceptions and clinical definitions of child obesity are known to diverge; however, the extent of the discrepancy has not been documented. This study characterises parental classifications of obesity and identifies sociodemographic characteristics that predict misclassification. Also, BMI centile cut-offs for weight status are established as derived from parental perceptions.

2. Does mindfulness improve outcomes in patients with chronic pain? Systematic review and meta-analysis.
http://bjgp.org/content/65/635/e387.full

This current review looks at management of non-malignant chronic pain as a whole, includes only randomised controlled trials, and uniquely focuses on humanistic outcomes such as pain acceptance and perceived pain control. These are of particular relevance with this self-help technique, as well as clinical and economic outcomes.

3. Help seeking for cancer ‘alarm’ symptoms: a qualitative interview study of primary care patients in the UK.
http://bjgp.org/content/65/631/e96.full

The Model of Pathways to Treatment highlights the importance of understanding patient appraisal and decision-to-consult processes for improving earlier diagnosis. Little is known about how people make decisions about visiting their GP for potential cancer symptoms in everyday life, without a researcher-imposed cancer perspective. This is the first qualitative, community-based study to assess how people respond to cancer ‘alarm’ symptoms outside of the cancer context. The results not only highlighted the importance of people’s interpretations of symptoms, but also their sense that they had to limit their demands for GP advice, both to preserve their self-image and to avoid uncomfortable interactions with the GP if they were seen as time wasters. The findings highlight potential avenues to promote prompt help seeking.

4. Simplified sleep restriction for insomnia in general practice: a randomised controlled trial.
http://bjgp.org/content/65/637/e508.full

Insomnia is a common health problem seen in general practice. Cognitive behavioural therapy for insomnia (CBT-I) is effective but its use has been limited by the time and expense required for delivery. Sleep restriction, or restricting the time in bed, is one component of CBT-I. This trial found that simplified sleep restriction advice, delivered in two GP consultations, is a practical, effective intervention for chronic primary insomnia.

5. Cannabis, tobacco smoking, and lung function: a cross-sectional observational study in a general practice population.
http://bjgp.org/content/65/631/e89.full

Cannabis smoking is associated with increased respiratory symptoms, but evidence of adverse effects on lung function is sparse. This study provides the first UK data on the impact of cannabis smoking on the prevalence of respiratory symptoms and chronic obstructive pulmonary disease in a general practice population. Although many adverse effects appeared attributable to tobacco, evidence of some additional adverse effects of cannabis was found.

6. Physician associates and GPs in primary care: a comparison.
http://bjgp.org/content/65/634/e344.full

Physician associates (previously known as physician assistants) are a new professional group in UK general practice, and evidence is required on their outcomes and costs. For patients attending for same-day or urgent appointments, PAs attended a younger patient group who present with less medically acute problems and fewer long-term conditions, compared to those attended by GPs. After adjusting for case-mix, there was no difference between PA and GP consultations in the rate of investigations, referral to secondary care, prescriptions issued, or the rate of patient re-consultation for the same or a closely related problem within 14 days. Patients report high levels of satisfaction with PA and GP consultations. The average PA consultation was longer than with a GP, although costs per consultation with a PA were lower.

7. Reduction in self-monitoring of blood glucose in type 2 diabetes: an observational controlled study in east London.
http://bjgp.org/content/65/633/e256.full

In many people who have type 2 diabetes but are not treated with insulin, self-monitoring of blood glucose (SMBG) is an often unnecessary task that confers no benefit, while being at very high, and increasing, cost. This is the first study to describe a systematic and major reduction in unnecessary use of SMBG. In people on metformin or no treatment, SMBG use was reduced from 29.6% to 6.0%; in all non-insulin users the rate fell from 42.8% to 16.5%. If replicated nationally, this would avoid unnecessary testing in 340 000 people and save £21.8 million that is currently spent on diabetes prescribing.

8. Patients’ online access to their electronic health records and linked online services: a systematic review in primary care.
http://bjgp.org/content/65/632/e141

Online services have been successfully piloted by large American organisations (such as Kaiser Permanente) but this success has not been replicated in the UK. This study found that patient satisfaction improved through enabling better self-care. Clinicians had concerns about the additional burden and workload from online access but evidence found their fears were only partly realised. There is a lack of good quality research in these areas in the UK; especially research that measures impact (such as health outcomes).

9. Patient information materials in general practices and promotion of health literacy: an observational study of their effectiveness.
http://bjgp.org/content/65/632/e192.full

Patient information leaflets (PILs) provide information to patients to encourage participation in their health care. Research evidence has variously concluded that PILs do affect patient health outcomes, but that many are poorly written. This study shows that less than 25% of PILs in general practice meet recommended reading-level guidance, and that most would be too complex for 43% of the English population. Less than 10% of the PILs covered managing illness or health promotion.

10. Provision of medical student teaching in UK general practices: a cross-sectional questionnaire study.
http://bjgp.org/content/65/635/e409.full

Demand for GP care is rapidly increasing, but the supply of GPs is problematic. Currently, the number of GPs going into training is also problematic. Plans to expand GP training to 50% of medical graduates have been proposed. Teaching medical students in general practice increases the probability that they will opt for a career in general practice and so appropriate undergraduate experience will be important for workforce planning. Since the late 1960s GP teaching of medical students has increased to occupy 13% of the clinical teaching in UK medical schools. This study suggests that this growth has stopped, and may be in decline. Effective financial mechanisms may help to alleviate this, but central intervention may be required to deliver it.

Altmetrics: now available for BJGP articles

Altmetrics_promo_homepage_newerThe world of scholarly publishing is changing rapidly, partly in response to digital publication, and also with more focus on the dissemination and implementation of published research. Traditional bibliometrics, such as the impact factor, have been used to measure aggregated citation rates as a proxy measure of journal quality. There is now more interest in looking at article-level and author-level metrics.

Peer-review publication is one component of an ‘ecosystem’ of dissemination, which includes, for example, citations, news and media coverage, discussion on social media and websites, and inclusion in practice guidelines. These new metrics – ‘altmetrics’ – defined as anything that is not a citation, can be captured in a number of ways.

The BJGP has launched the Altmetric donut, a colourful, arresting image which depicts the various media which have paid attention to a given article, with a numerical score reflecting the number of ‘mentions’. The Altmetric buttons, appearing within the ‘Info’ tab of each article, are not substitutes for traditional bibliometrics, but we think will become a useful addition to understanding how research results ‘get out’ and are incorporated into practice.

Quality indicators for child health in the UK

Peter_Gill_Peter Gill is a paediatric resident at The Hospital for Sick Children in Toronto, Ontario and an Honorary Fellow at the Centre for Evidence-Based Medicine at the University of Oxford. Follow Peter on Twitter @peterjgill

In the December 2014 issue of the British Journal of General Practice, several colleagues and I published a set of paediatric quality indicators for UK primary care.[1] The paper represents the main findings from my doctoral thesis completed under the supervision of Prof David Mant and Anthony Harnden at Oxford University. It is exciting to see the paper in print (it provided a morale ‘boost’ after working a stretch of nights) accompanied by a thoughtful editorial.[2] But having returned to the ‘coal face’, I am reminded of the integral role of quality indicators in clinical practice.

Why develop indicators?

Caring for children is an important part of UK general practice yet several studies have demonstrated that care quality can be improved. However, only 3% of Quality Outcomes Framework (QOF) markers relate to children and there is no set of child-relevant indicators for UK primary care. For years, the call to develop and integrate child health indicators into QOF has fallen on deaf ears despite evidence that leaving out indicators probably has negatives consequences for care quality. Therefore, we sought to develop a set of quality indicators for children and adolescents which cover a range of paediatric care and reflect existing UK evidence-based national guidelines.

What did we do?

In just over 2000 words, the BJGP paper outlines how a set of 35 indicators were developed; it looks deceiving simple. In reality, the study is the culmination of five years of work with invaluable contributions from many (including funding bodies). First, we selected priority areas [3] after searching the evidence [4], evaluating unplanned hospital admissions [5] and interviewing GPs [6]. Second, we reviewed NICE and SIGN national guidelines relevant to children in primary care before translating key guideline recommendations into quality indicators. Third, a UK-wide panel of GPs with a special interest in child health assessed the validity and implementation of indicators using the RAND methodology.

We believe the 35 indicators have high levels of clinical support, reflect national guidelines, and could be feasibly implemented in the UK. Rather than narrowly focus on easily measurable aspects of care, we address broader determinants of child health outside of clinical guidelines (e.g. child safeguarding and professional development) and prompt GPs to critically reflect on their actions (e.g. rationale for antibiotic prescribing).

How can the indicators improve quality?

There are many ways in which the care quality can improve with implementation of the indicators. For example, the indicators may lead to the earlier diagnosis of easily missed conditions such as Coeliac disease and Type 1 Diabetes which may improve quality of life and reduce complications. In 2008, there were 6300 paediatric medication related safety incidents; the prescribing indicators may improve medication safety. Appropriate evidence-based management of common conditions such as asthma, may reduce emergency department visits and secondary care referral. Each indicator, after appropriate piloting in general practice, could potentially play an important role to improve care quality and identify which areas require increased attention.

What next?

The NHS is an exemplary model of a health system with robust primary care. But, at the risk of being cliché, any system is only as strong as its weakest link. Quality indicators, whether linked to QOF or used as an audit tool, must cover the full range of UK general practice. The on-going restructuring of the NHS may only further marginalise the care of children and adolescents whose interests are often poorly represented.

These quality indicators provide a starting point. They are a set of tools for clinicians working with children. What we need now is for individuals to pick up the tools and get to work. In the US, primary care indicators were signed into law by President Obama in 2009 through the Children’s Health Insurance Program Reauthorisation Act. Why are there not similar initiatives in the UK? Bold steps are needed to implement paediatric indicators in UK primary care.

[bctt tweet=”BJGP Blog: Bold steps needed to implement paediatric indicators in UK primary care.”]

The paper is Open Access and available at http://bjgp.org/content/64/629/e752

 

References

1.     Gill PJ, O’Neill B, Rose P, Mant D, Harnden A. Primary care quality indicators for children: measuring quality in UK general practice. Br J Gen Pract. 2014 Dec;64(629):e752-7. doi: 10.3399/bjgp14X682813.

2.     Dowell A, Turner N. Child health indicators: from theoretical frameworks to practical reality? Br J Gen Pract. 2014 Dec;64(629):608-9. doi: 10.3399/bjgp14X682585.

3.     Gill PJ, Hewitson P, Peile E, Harnden A. Prioritizing areas for quality marker development in children in UK general practice: extending the use of the nominal group technique. Fam Pract. 2012 Oct;29(5):567-75.

4.     Gill PJ, Wang KY, Mant D, Hartling L, Heneghan C, Perera R, Klassen T, Harnden A. The Evidence Base for Interventions Delivered to Children in Primary Care: An Overview of Cochrane Systematic Reviews. PLoS One. 2011; 6(8): e23051. doi:10.1371/journal.pone.0023051

5.     Gill PJ, Goldacre MJ, Mant D, Heneghan C, Thomson A, Seagroatt V, Harnden A. Increase in emergency admissions to hospital for children aged under 15 in England, 1999-2010: national database analysis. Arch Dis Child. 2013 May;98(5):328-34. doi: 10.1136/archdischild-2012-302383.

6.     Gill PJ, Hislop J, Mant D, Harnden A. General practitioners’ views on quality markers for children in UK primary care: a qualitative study. BMC Fam Pract. 2012 Sept 14;13(1):92 doi:10.1186/1471-2296-13-92.