Saturday, June 26, 2004
BMJ 2004;328:1552-1554 (26 June), doi:10.1136/bmj.328.7455.1552
Charles Bonnet syndrome—elderly people and visual hallucinations
Charles Bonnet syndrome
Charles Bonnet syndrome is a less frequently diagnosed but rather common cause of complex visual hallucination. Its prevalence in patients with visual impairment varies from 10% to 15%.1 The condition is named after the Swiss naturalist and philosopher Charles Bonnet. He reported the hallucinations of Charles Lullin, his 89 year old otherwise healthy and cognitively sound grandfather, who was blind owing to cataract and yet vividly saw men, women, birds, and buildings.1 3
Diagnostic criteria and clinical features
Though no universally approved diagnostic criteria for the syndrome exist, the core features are the occurrence of well formed, vivid, elaborate, and often stereotyped visual hallucinations in a partially sighted person who has insight into the unreality of what he or she is seeing. There should not be any feature of psychosis, impaired sensorium, dementia, intoxication, metabolic derangement, or focal neurological illness.3-5 The syndrome occurs most commonly in elderly people, probably because of the prevalence of visual impairment in this group. The common conditions leading to the syndrome are age related macular degeneration, followed by glaucoma and cataract. These hallucinations, which are always outside the body, may last from a few seconds to most of the day. They may persist for a few days to many years, changing in frequency and complexity. They have no personal meaning, and many patients can voluntarily modify them or make the image disappear if they close their eyes. The imagery is varied and may include groups of people or children, animals, and panoramic countryside scenes.1 3 5
Naturalist and philosopher Charles Bonnet described the condition in 1760
The syndrome can occur in people with normal vision.6 Some have argued that diagnosis of the syndrome does not exclude or require eye disease or brain lesions and that it could even be due to lesions that are not associated with the visual system.5 Reduced or absent stimulation of the visual system (deafferentation hypothesis) leading to increased excitability of the visual cortex is one of the hypotheses.1 7
Course, prognosis, and treatment
The course, prognosis, and treatment vary with the nature of the visual dysfunction. Removal of a cataract or recovery of vision leads to improvement. Other patients find relief when the eye disease progresses to total blindness.8 Some have suggested that the syndrome can even be an indication of early dementia9; this hypothesis needs to be validated. Treatments with drugs remain unsatisfactory, with only anecdotal evidence for the efficacy of atypical antipsychotics and anticonvulsants.1 Non-pharmacological interventions, such as increasing the lighting at home and reducing social isolation by encouraging interpersonal contact, are helpful.1
Clinical correlate
Our patient provides a classic example of Charles Bonnet syndrome. The clinical scenario and the nature of hallucinations are typical. The possibility of oxprenolol causing hallucination was remote as he had been taking the drug for 10 years whereas the hallucinations had been present for only a few weeks. Timely diagnosis and explanation, along with reassurance about the relatively benign nature of the condition, provided immeasurable relief. He still gets occasional hallucinations but considers them as "one of those things you have to put up with."
Conclusion
As well as being common in partially sighted people, Charles Bonnet syndrome occurs in 1.85-3.5% of psychogeriatric patients who have been referred to psychiatrists by adult physicians, general practitioners, and ophthalmologists for visual hallucinations.1 10 11
Doctors are unfamiliar with the syndrome as a possible diagnosis.1 12 "Near misses" have been reported, in which patients were almost confined to mental health institutions.13 Given the prevalence of partial visual impairment, the number of people in the community, especially elderly people, who do not report the symptoms for fear of being labelled as mentally unwell or demented must be substantial. Clinicians must therefore be aware and ask elderly people with visual impairment whether they have hallucinations. Firm reassurance that the syndrome is not related to mental illness is in itself a major relief to an elderly person burdened already with failing vision, social isolation, and other medical problems.
Not all elderly people presenting with visual hallucinations have dementia
Comment: Absolutley!
Charles Bonnet syndrome—elderly people and visual hallucinations
Charles Bonnet syndrome
Charles Bonnet syndrome is a less frequently diagnosed but rather common cause of complex visual hallucination. Its prevalence in patients with visual impairment varies from 10% to 15%.1 The condition is named after the Swiss naturalist and philosopher Charles Bonnet. He reported the hallucinations of Charles Lullin, his 89 year old otherwise healthy and cognitively sound grandfather, who was blind owing to cataract and yet vividly saw men, women, birds, and buildings.1 3
Diagnostic criteria and clinical features
Though no universally approved diagnostic criteria for the syndrome exist, the core features are the occurrence of well formed, vivid, elaborate, and often stereotyped visual hallucinations in a partially sighted person who has insight into the unreality of what he or she is seeing. There should not be any feature of psychosis, impaired sensorium, dementia, intoxication, metabolic derangement, or focal neurological illness.3-5 The syndrome occurs most commonly in elderly people, probably because of the prevalence of visual impairment in this group. The common conditions leading to the syndrome are age related macular degeneration, followed by glaucoma and cataract. These hallucinations, which are always outside the body, may last from a few seconds to most of the day. They may persist for a few days to many years, changing in frequency and complexity. They have no personal meaning, and many patients can voluntarily modify them or make the image disappear if they close their eyes. The imagery is varied and may include groups of people or children, animals, and panoramic countryside scenes.1 3 5
Naturalist and philosopher Charles Bonnet described the condition in 1760
The syndrome can occur in people with normal vision.6 Some have argued that diagnosis of the syndrome does not exclude or require eye disease or brain lesions and that it could even be due to lesions that are not associated with the visual system.5 Reduced or absent stimulation of the visual system (deafferentation hypothesis) leading to increased excitability of the visual cortex is one of the hypotheses.1 7
Course, prognosis, and treatment
The course, prognosis, and treatment vary with the nature of the visual dysfunction. Removal of a cataract or recovery of vision leads to improvement. Other patients find relief when the eye disease progresses to total blindness.8 Some have suggested that the syndrome can even be an indication of early dementia9; this hypothesis needs to be validated. Treatments with drugs remain unsatisfactory, with only anecdotal evidence for the efficacy of atypical antipsychotics and anticonvulsants.1 Non-pharmacological interventions, such as increasing the lighting at home and reducing social isolation by encouraging interpersonal contact, are helpful.1
Clinical correlate
Our patient provides a classic example of Charles Bonnet syndrome. The clinical scenario and the nature of hallucinations are typical. The possibility of oxprenolol causing hallucination was remote as he had been taking the drug for 10 years whereas the hallucinations had been present for only a few weeks. Timely diagnosis and explanation, along with reassurance about the relatively benign nature of the condition, provided immeasurable relief. He still gets occasional hallucinations but considers them as "one of those things you have to put up with."
Conclusion
As well as being common in partially sighted people, Charles Bonnet syndrome occurs in 1.85-3.5% of psychogeriatric patients who have been referred to psychiatrists by adult physicians, general practitioners, and ophthalmologists for visual hallucinations.1 10 11
Doctors are unfamiliar with the syndrome as a possible diagnosis.1 12 "Near misses" have been reported, in which patients were almost confined to mental health institutions.13 Given the prevalence of partial visual impairment, the number of people in the community, especially elderly people, who do not report the symptoms for fear of being labelled as mentally unwell or demented must be substantial. Clinicians must therefore be aware and ask elderly people with visual impairment whether they have hallucinations. Firm reassurance that the syndrome is not related to mental illness is in itself a major relief to an elderly person burdened already with failing vision, social isolation, and other medical problems.
Not all elderly people presenting with visual hallucinations have dementia
Comment: Absolutley!
BMJ 2004;328:1514 (26 June), doi:10.1136/bmj.328.7455.1514-b
Oestrogen doesn't protect mental function in older women
Janice Hopkins Tanne
New York
The "oestrogen only" treatment arm of the women's health initiative memory study shows that oestrogen does not protect against mild cognitive impairment or dementia and slightly increases the risk, say two reports published in JAMA ( 2004;291: 2947-58, 2959-68)[Abstract/Free Full Text]. Mild cognitive impairment strongly predicts dementia.
Previous studies have suggested that oestrogen might prevent cognitive decline, said Lon Schneider, professor of psychiatry, neurology, and gerontology at the University of Southern California in Los Angeles, who commented on the studies in an editorial ( JAMA 2004;291: 3005-7[Free Full Text]). He told the BMJ that the finding which showed oestrogen might increase the risk of cognitive impairment and dementia was unexpected. "The studies show that giving oestrogen or oestrogen and progestin [progestogen] to women over 65 in hopes of delaying dementia or increasing cognitive function is misplaced. They should not be getting oestrogen."
Comment: another nail in the coffin for HRT
Oestrogen doesn't protect mental function in older women
Janice Hopkins Tanne
New York
The "oestrogen only" treatment arm of the women's health initiative memory study shows that oestrogen does not protect against mild cognitive impairment or dementia and slightly increases the risk, say two reports published in JAMA ( 2004;291: 2947-58, 2959-68)[Abstract/Free Full Text]. Mild cognitive impairment strongly predicts dementia.
Previous studies have suggested that oestrogen might prevent cognitive decline, said Lon Schneider, professor of psychiatry, neurology, and gerontology at the University of Southern California in Los Angeles, who commented on the studies in an editorial ( JAMA 2004;291: 3005-7[Free Full Text]). He told the BMJ that the finding which showed oestrogen might increase the risk of cognitive impairment and dementia was unexpected. "The studies show that giving oestrogen or oestrogen and progestin [progestogen] to women over 65 in hopes of delaying dementia or increasing cognitive function is misplaced. They should not be getting oestrogen."
Comment: another nail in the coffin for HRT
BMJ 2004;328:1509-1510 (26 June), doi:10.1136/bmj.328.7455.1509
Is epidural injection of steroids effective for low back pain?
The evidence is equivocal, but clinical experience favours its use in some patients
What then are the implications for clinical practice? Epidural injection therapy has not yet been shown to be effective, nor has it been shown to be ineffective.12 Side effects are relatively minor, and a tendency exists towards an outcome favouring injection therapy. On the basis of our longstanding clinical experience we suggest that epidural steroid injection may have a role in specific clinical situations. Low back pain that has not resolved within three months leads to greater long term morbidity.12 Epidural injection therapy may provide a useful adjunct to recovery in patients whose symptoms have extended beyond three months in the absence of recognised indicators of chronicity ("yellow flags"),4 and who may have radicular symptoms. The evidence for and against epidural steroid injection should be clearly explained to allow patients to make an informed choice regarding their treatment.
Comment: another treatment that has eqivocal evidence, but clinically helpful. Difficult to practice EBM when evidence is not there yet.
Is epidural injection of steroids effective for low back pain?
The evidence is equivocal, but clinical experience favours its use in some patients
What then are the implications for clinical practice? Epidural injection therapy has not yet been shown to be effective, nor has it been shown to be ineffective.12 Side effects are relatively minor, and a tendency exists towards an outcome favouring injection therapy. On the basis of our longstanding clinical experience we suggest that epidural steroid injection may have a role in specific clinical situations. Low back pain that has not resolved within three months leads to greater long term morbidity.12 Epidural injection therapy may provide a useful adjunct to recovery in patients whose symptoms have extended beyond three months in the absence of recognised indicators of chronicity ("yellow flags"),4 and who may have radicular symptoms. The evidence for and against epidural steroid injection should be clearly explained to allow patients to make an informed choice regarding their treatment.
Comment: another treatment that has eqivocal evidence, but clinically helpful. Difficult to practice EBM when evidence is not there yet.
BMJ 2004;328:1507 (26 June), doi:10.1136/bmj.328.7455.1507
Doll et al document the steady attenuation of adverse effects among quitters. Quitting at any age confers benefit compared with persistent smoking, but early quitting is necessary to approximate death rates among never smokers. Nonetheless, the authors estimate that even a 60 year old smoker could gain at least three years of life expectancy by stopping.
Comment Good editorial on paper in this weeks BMJ
Doll et al document the steady attenuation of adverse effects among quitters. Quitting at any age confers benefit compared with persistent smoking, but early quitting is necessary to approximate death rates among never smokers. Nonetheless, the authors estimate that even a 60 year old smoker could gain at least three years of life expectancy by stopping.
Comment Good editorial on paper in this weeks BMJ
BMJ 2004;328 (26 June), doi:10.1136/bmj.328.7455.0-c
Smokeless tobacco use affects babies
Mothers' consumption of smokeless tobacco during pregnancy decreases babies' birth weight and gestational age at birth. Analysing the use of smokeless tobacco in 1217 pregnant women from Mumbai, Gupta and Sreevidya (p 1538) found that the decrease in birth weight was independent of gestational age, sex of baby, and maternal characteristics, and the risk for preterm delivery was increased. The authors say that mothers' use of smokeless tobacco should receive specific attention as part of routine prenatal care.
Comment: tobacco in any form can affect the unborn child
Smokeless tobacco use affects babies
Mothers' consumption of smokeless tobacco during pregnancy decreases babies' birth weight and gestational age at birth. Analysing the use of smokeless tobacco in 1217 pregnant women from Mumbai, Gupta and Sreevidya (p 1538) found that the decrease in birth weight was independent of gestational age, sex of baby, and maternal characteristics, and the risk for preterm delivery was increased. The authors say that mothers' use of smokeless tobacco should receive specific attention as part of routine prenatal care.
Comment: tobacco in any form can affect the unborn child
BMJ 2004;328 (26 June), doi:10.1136/bmj.328.7455.0
What happens to doctors who smoke?
Men born between 1900 and 1930 who smoked cigarettes 50 years ago, and who continued to smoke, died on average 10 years younger than lifelong non-smokers. Fifty years after the publication of their original study, which we republish in this issue (p 1529), Doll and colleagues (p 1519) analysed the mortality of 34 439 male British doctors in relation to their smoking habits. They found that more than half of the young doctors in the 1950s who continued to smoke cigarettes were killed by their habit. Among men born in 1920, prolonged smoking tripled age specific mortality, but stopping smoking at age 50 halved the hazard, and stopping at age 30 avoided almost all of it.
Comment: just as was thought, smoking kills but never too late to stop
What happens to doctors who smoke?
Men born between 1900 and 1930 who smoked cigarettes 50 years ago, and who continued to smoke, died on average 10 years younger than lifelong non-smokers. Fifty years after the publication of their original study, which we republish in this issue (p 1529), Doll and colleagues (p 1519) analysed the mortality of 34 439 male British doctors in relation to their smoking habits. They found that more than half of the young doctors in the 1950s who continued to smoke cigarettes were killed by their habit. Among men born in 1920, prolonged smoking tripled age specific mortality, but stopping smoking at age 50 halved the hazard, and stopping at age 30 avoided almost all of it.
Comment: just as was thought, smoking kills but never too late to stop
Saturday, June 19, 2004
BMJ Career Focus 2004;328:266-268 (19 June)
Writing CVs and handling job interviews
The first part of this article discusses how Sara wrote her curriculum vitae (CV). The second describes how Sara handled the job interview. Sara, her CV, and the doctors named in it are fictitious. Names of known institutions are used to maintain a sense of reality.
Writing a CV
Sara wanted a CV that would get her interviews for SHO posts in psychiatry. She read about writing CVs,1 talked with junior doctors and sought advice from her clinical attachment supervisor.
Sara had a choice of formats—the targeted, the chronological, and the functional. In addition to personal details, a targeted CV names the target job and summarises work history; a chronological CV lists work history from the first to the last job. A functional CV highlights functional skills and achievements and summarises work history from the current or most recent post back to first. Sara, like most doctors, used the functional format to write her CV.2
She followed the "Usual format for medical CVs" (box), made notes on all items listed, typed, edited, and arranged them to produce a document (figure) that summarises her personal, educational, and career details. She can use her CV to inquire about, or apply for jobs, training positions and research grants, and as a source of sorted information for application forms.
Sara Kirabo was looking for an SHO post in paediatrics
She showed the CV to her attachment supervisor and was advised to sort out a few points, and to list her publications in a standard format. They also discussed a cover letter to accompany the CV when applying for jobs.2
Sara knew that when a recruiter started reading her CV, she would have less than a minute to convince them of her suitability for a post. She wrote a CV that is concise, attractive, and easy to skim in search of important information. It was printed on good quality, white, 80 g/m2 paper. It would not be the weight or cost of paper on which the CV was printed, but the content and quality of her writing that would impress consultants when they shortlisted for interviews.
Sara avoided excess prose and tables and presented the information mostly in note form. Long paragraphs would not allow key points to stand out, and tables would not highlight past duties and achievements. Sara's CV emphasised her achievements, but contained no lies:
The GMC's professional conduct committee found a doctor guilty of serious professional misconduct after she made false claims about her academic achievements and employment history. The committee suspended her registration for 12 months.3
Sara wrote the CV in an active language showing that she is an active and decisive individual: she was careful with grammar and spelling, used a spellchecker, and then read the CV carefully. Sara was afraid that bad spelling and poor grammar might lead to rejection of her application.
Let us briefly clarify a few items in "Usual format for medical CVs" that puzzled Sara as she wrote her CV.
The heading announces the subject of the document and avoids the need for a front sheet naming the doctor described in the CV. Sara's CV is headed: Dr S Kirabo MB ChB.
Sara's contact details are adequate for the recruiter to telephone, email, or send her a letter. Her personal identity does not include her marital status, spouse's nationality or size of her family (she is married and has two young kids). She could have included her hepatitis B status.
Sara's qualification is MB ChB. When she gets other degrees or membership of the royal college of psychiatrists she will add them to her CV. Her pass at PLAB tests is not a qualification; so she listed it separately.
Usual format for medical CVs
Heading (initials, surname, one qualification)
Contact details
Postal address
Telephone
Email
Personal identity
Names (first name(s), surname)
Date of birth
Nationality
Sex
Career summary
Qualifications, dates, institution, and location
Professional organisations (GMC, royal medical college, MDU, etc)
Career plan and how the job will help you achieve it
Summary of skills and achievements
Education (dates, institution, course, prizes, other achievements)
Work history (date, position, employer, location, duties, and achievements)
Courses, conferences attended
Presentations at meetings
Clinical audit, research
Publications in journals or books
Additional information—useful skills (information technology, languages, etc) that do not fit elsewhere
Hobbies—leisure interests and activities.
Referees—names and contact details of two people
When she passes part 1 MRCPsych she will list it separately too till she gets the full membership. She will not include failed exams.
Sara's career plan was concise and described how the job would help her achieve it. It was on the first page of her CV.
Sara summarised skills she gained from her career. In future when she applies for a specialist registrar (SpR) post she will summarise her skills and achievements to highlight what the employer would get by appointing her to the post.
The education entry highlights leadership responsibilities held at high school and university to indicate Sara's leadership potential.
Under work experience Sara listed posts she had held and her duties and achievement in each. Her clinical attachment is not an employment post. She listed it separately under "clinical attachment."
She has no need to hide or minimise any aspect of her medical career. All legitimate medical work, including national service, should be included in a CV. Interviewers like to ask questions about atypical careers.
Sara accounted for significant gaps in her career. She had heard that some consultants scrutinise CVs for career breaks and discard those with long unexplained gaps.
Sara described her research activities, listed conferences attended and presentations made.
She followed the Vancouver style when listing her publications.4 All publications, be they journal articles, abstracts in conference proceedings, book chapters, or whole books, qualify for listing in a CV whether done abroad or in the United Kingdom.
Sara listed her hobbies to show that she has a life outside medicine. She gave names and contact details of two referees. After settling in her new job she will ask two suitable consultants for permission to name them as referees. References from consultants a doctor has worked with are more influential than those from supervisors of clinical attachments.
The length of Sara's CV—three pages—was determined by jobs, audits, research, presentations and publications done, and courses and conferences attended. Two to three pages should suffice for most house officers and junior SHOs. A senior SHO applying for an SpR post may need four to five pages.2
Sara did not pay a commercial company to write her CV. Such companies do not write CVs, but type, edit, and print (produce) documents from information supplied by clients. They print the document on unnecessarily expensive paper and charge a lot of money for the service. By following "Usual format for doctors' CVs" Sara crafted her CV and asked her supervisor's secretary for advice on getting it professionally typed and printed. Because she asked politely, the secretary typed and printed the CV free of charge. She could have directed Sara to one of several commercial typists nearby. Sara recently brought her an ornamental basket from Uganda.
Handling job interviews
An interview is an occasion when individuals or groups meet and talk with each other, with one side asking questions and the other answering them. In a job interview an applicant (interviewee) is seen and questioned by an employer or their representative to determine the applicant's suitability.
Although interviewees strive to convince interviewers on their suitability, interviewers ask themselves a few questions about each candidate:
Is she the most suitable? Did Sara have the right personality and drive? Was it the right job for her?
Will she do the job? If interviewers felt that although capable, Sara would not willingly perform her duties, they would not appoint her.
Will she fit in? People want somebody who will fit in nicely. Sara would be appointed if interviewers felt she would fit into the team—character, drive, dependability, initiative, congeniality, etc.
Pre-interview inquiries
After sending out 60 applications, Sara was shortlisted for interview by two mental hospitals—one in Yorkshire and the Maudsley in south London. The interviews were on the same date. Her supervisor recommended attending the London interview.
Sara did not have to visit the Maudsley. She could have rung and spoken with the incumbent of the post she would be interviewing for. But as she did not know London, she visited the hospital, noted the route and where the interview would be held. She met people in the post, some nurses, and one consultant.
Preparing for interview
Sara sought advice from junior doctors at Watford Hospital. One registrar advised her to read her CV before the interview. "Interviewers will have copies of your CV" he said, "and will question you on its content." Two SHOs who had recently succeeded at job interviews took Sara through a mock interview and explained how they had handled questions like:
Why did you apply for this job?
What are your career objectives?
What are your weaknesses/strengths?
Tell us about a clinical audit you have done
Tell us about an interesting case you have handled recently
What is the most challenging clinical situation you have met?
Presenting for interview
What did Sara look like (appearance, confidence), and sound like (voice, language)? Critical decisions (trust and distrust, like and dislike, etc) are often made after brief encounters—interviews. Only qualified candidates would be interviewed. So, it would not be qualifications but the impression she created by her appearance, confidence, speech, and demeanour that would determine the outcome.
Sara dressed for a professional interview and not a party. She wore smart, clean, and socially appropriate clothes—not a T shirt or jeans. She could have worn trousers, a shirt, and a jacket, but she felt more comfortable in a skirt, a shirt, and a matching jacket. She took her hepatitis B serology status, eligibility for limited registration, and passport with her.
Demeanour
At the human resources department Sara made her inquiries politely and waited calmly. On entering the interview room she was introduced to three consultants and a human resources officer. She shook hands with a moderately firm grip and made direct eye contact with panel members as they were introduced to her. She sat in the chair indicated by the chairman, made herself comfortable, and waited for questions.
She spoke clearly, concisely, and simply; neither whispering nor shouting. She watched interviewers' faces to see if they heard and understood her answers. She did not draw attention to her weaknesses nor argue with interviewers. She smiled and laughed appropriately.
Sara answered the questions asked. She gave concise but informative answers confidently and politely. If a question was not clear she asked for an explanation or for it to be repeated. When she did not know the answer, she said so confidently. She did not use jargon or abbreviations unless she was sure the interviewers knew them too. When an interviewer was speaking she did not interrupt, but listened and answered the question that followed the speech.
Sara was asked about her work in health education and counselling fellow students. She gave succinct but full answers. Interviewers were impressed by her understanding of the HIV pandemic.
Questions about the past
Sara was aware that she might be asked about past failures. She was prepared to talk about one case that did not go well. The cause of mishap had been identified and lessons learnt. She had reflected on the incident and could discuss it confidently and rationally stressing the lessons and not the failure. She was not asked.
Finally, she was asked if she had any questions. She thanked the panel for the opportunity, said she had talked with doctors and nurses on the ward, and had no questions. She was told a decision would be made later that afternoon and that she could wait for the result if she wished. She said she would wait. She left the interview room confidently and politely.
At 5.30 pm., three of the 10 interviewees were offered jobs. Sara was among the successful doctors. She accepted the job offer.
Further reading
Turya EB. Your career after PLAB: survival tools for young doctors. Manchester: Edukom, 2003.
McErin S. Writing the medical CV. Manchester: Edukom, 2004.
--------------------------------------------------------------------------------
Elitham B Turya, consultant in child health
Writing CVs and handling job interviews
The first part of this article discusses how Sara wrote her curriculum vitae (CV). The second describes how Sara handled the job interview. Sara, her CV, and the doctors named in it are fictitious. Names of known institutions are used to maintain a sense of reality.
Writing a CV
Sara wanted a CV that would get her interviews for SHO posts in psychiatry. She read about writing CVs,1 talked with junior doctors and sought advice from her clinical attachment supervisor.
Sara had a choice of formats—the targeted, the chronological, and the functional. In addition to personal details, a targeted CV names the target job and summarises work history; a chronological CV lists work history from the first to the last job. A functional CV highlights functional skills and achievements and summarises work history from the current or most recent post back to first. Sara, like most doctors, used the functional format to write her CV.2
She followed the "Usual format for medical CVs" (box), made notes on all items listed, typed, edited, and arranged them to produce a document (figure) that summarises her personal, educational, and career details. She can use her CV to inquire about, or apply for jobs, training positions and research grants, and as a source of sorted information for application forms.
Sara Kirabo was looking for an SHO post in paediatrics
She showed the CV to her attachment supervisor and was advised to sort out a few points, and to list her publications in a standard format. They also discussed a cover letter to accompany the CV when applying for jobs.2
Sara knew that when a recruiter started reading her CV, she would have less than a minute to convince them of her suitability for a post. She wrote a CV that is concise, attractive, and easy to skim in search of important information. It was printed on good quality, white, 80 g/m2 paper. It would not be the weight or cost of paper on which the CV was printed, but the content and quality of her writing that would impress consultants when they shortlisted for interviews.
Sara avoided excess prose and tables and presented the information mostly in note form. Long paragraphs would not allow key points to stand out, and tables would not highlight past duties and achievements. Sara's CV emphasised her achievements, but contained no lies:
The GMC's professional conduct committee found a doctor guilty of serious professional misconduct after she made false claims about her academic achievements and employment history. The committee suspended her registration for 12 months.3
Sara wrote the CV in an active language showing that she is an active and decisive individual: she was careful with grammar and spelling, used a spellchecker, and then read the CV carefully. Sara was afraid that bad spelling and poor grammar might lead to rejection of her application.
Let us briefly clarify a few items in "Usual format for medical CVs" that puzzled Sara as she wrote her CV.
The heading announces the subject of the document and avoids the need for a front sheet naming the doctor described in the CV. Sara's CV is headed: Dr S Kirabo MB ChB.
Sara's contact details are adequate for the recruiter to telephone, email, or send her a letter. Her personal identity does not include her marital status, spouse's nationality or size of her family (she is married and has two young kids). She could have included her hepatitis B status.
Sara's qualification is MB ChB. When she gets other degrees or membership of the royal college of psychiatrists she will add them to her CV. Her pass at PLAB tests is not a qualification; so she listed it separately.
Usual format for medical CVs
Heading (initials, surname, one qualification)
Contact details
Postal address
Telephone
Personal identity
Names (first name(s), surname)
Date of birth
Nationality
Sex
Career summary
Qualifications, dates, institution, and location
Professional organisations (GMC, royal medical college, MDU, etc)
Career plan and how the job will help you achieve it
Summary of skills and achievements
Education (dates, institution, course, prizes, other achievements)
Work history (date, position, employer, location, duties, and achievements)
Courses, conferences attended
Presentations at meetings
Clinical audit, research
Publications in journals or books
Additional information—useful skills (information technology, languages, etc) that do not fit elsewhere
Hobbies—leisure interests and activities.
Referees—names and contact details of two people
When she passes part 1 MRCPsych she will list it separately too till she gets the full membership. She will not include failed exams.
Sara's career plan was concise and described how the job would help her achieve it. It was on the first page of her CV.
Sara summarised skills she gained from her career. In future when she applies for a specialist registrar (SpR) post she will summarise her skills and achievements to highlight what the employer would get by appointing her to the post.
The education entry highlights leadership responsibilities held at high school and university to indicate Sara's leadership potential.
Under work experience Sara listed posts she had held and her duties and achievement in each. Her clinical attachment is not an employment post. She listed it separately under "clinical attachment."
She has no need to hide or minimise any aspect of her medical career. All legitimate medical work, including national service, should be included in a CV. Interviewers like to ask questions about atypical careers.
Sara accounted for significant gaps in her career. She had heard that some consultants scrutinise CVs for career breaks and discard those with long unexplained gaps.
Sara described her research activities, listed conferences attended and presentations made.
She followed the Vancouver style when listing her publications.4 All publications, be they journal articles, abstracts in conference proceedings, book chapters, or whole books, qualify for listing in a CV whether done abroad or in the United Kingdom.
Sara listed her hobbies to show that she has a life outside medicine. She gave names and contact details of two referees. After settling in her new job she will ask two suitable consultants for permission to name them as referees. References from consultants a doctor has worked with are more influential than those from supervisors of clinical attachments.
The length of Sara's CV—three pages—was determined by jobs, audits, research, presentations and publications done, and courses and conferences attended. Two to three pages should suffice for most house officers and junior SHOs. A senior SHO applying for an SpR post may need four to five pages.2
Sara did not pay a commercial company to write her CV. Such companies do not write CVs, but type, edit, and print (produce) documents from information supplied by clients. They print the document on unnecessarily expensive paper and charge a lot of money for the service. By following "Usual format for doctors' CVs" Sara crafted her CV and asked her supervisor's secretary for advice on getting it professionally typed and printed. Because she asked politely, the secretary typed and printed the CV free of charge. She could have directed Sara to one of several commercial typists nearby. Sara recently brought her an ornamental basket from Uganda.
Handling job interviews
An interview is an occasion when individuals or groups meet and talk with each other, with one side asking questions and the other answering them. In a job interview an applicant (interviewee) is seen and questioned by an employer or their representative to determine the applicant's suitability.
Although interviewees strive to convince interviewers on their suitability, interviewers ask themselves a few questions about each candidate:
Is she the most suitable? Did Sara have the right personality and drive? Was it the right job for her?
Will she do the job? If interviewers felt that although capable, Sara would not willingly perform her duties, they would not appoint her.
Will she fit in? People want somebody who will fit in nicely. Sara would be appointed if interviewers felt she would fit into the team—character, drive, dependability, initiative, congeniality, etc.
Pre-interview inquiries
After sending out 60 applications, Sara was shortlisted for interview by two mental hospitals—one in Yorkshire and the Maudsley in south London. The interviews were on the same date. Her supervisor recommended attending the London interview.
Sara did not have to visit the Maudsley. She could have rung and spoken with the incumbent of the post she would be interviewing for. But as she did not know London, she visited the hospital, noted the route and where the interview would be held. She met people in the post, some nurses, and one consultant.
Preparing for interview
Sara sought advice from junior doctors at Watford Hospital. One registrar advised her to read her CV before the interview. "Interviewers will have copies of your CV" he said, "and will question you on its content." Two SHOs who had recently succeeded at job interviews took Sara through a mock interview and explained how they had handled questions like:
Why did you apply for this job?
What are your career objectives?
What are your weaknesses/strengths?
Tell us about a clinical audit you have done
Tell us about an interesting case you have handled recently
What is the most challenging clinical situation you have met?
Presenting for interview
What did Sara look like (appearance, confidence), and sound like (voice, language)? Critical decisions (trust and distrust, like and dislike, etc) are often made after brief encounters—interviews. Only qualified candidates would be interviewed. So, it would not be qualifications but the impression she created by her appearance, confidence, speech, and demeanour that would determine the outcome.
Sara dressed for a professional interview and not a party. She wore smart, clean, and socially appropriate clothes—not a T shirt or jeans. She could have worn trousers, a shirt, and a jacket, but she felt more comfortable in a skirt, a shirt, and a matching jacket. She took her hepatitis B serology status, eligibility for limited registration, and passport with her.
Demeanour
At the human resources department Sara made her inquiries politely and waited calmly. On entering the interview room she was introduced to three consultants and a human resources officer. She shook hands with a moderately firm grip and made direct eye contact with panel members as they were introduced to her. She sat in the chair indicated by the chairman, made herself comfortable, and waited for questions.
She spoke clearly, concisely, and simply; neither whispering nor shouting. She watched interviewers' faces to see if they heard and understood her answers. She did not draw attention to her weaknesses nor argue with interviewers. She smiled and laughed appropriately.
Sara answered the questions asked. She gave concise but informative answers confidently and politely. If a question was not clear she asked for an explanation or for it to be repeated. When she did not know the answer, she said so confidently. She did not use jargon or abbreviations unless she was sure the interviewers knew them too. When an interviewer was speaking she did not interrupt, but listened and answered the question that followed the speech.
Sara was asked about her work in health education and counselling fellow students. She gave succinct but full answers. Interviewers were impressed by her understanding of the HIV pandemic.
Questions about the past
Sara was aware that she might be asked about past failures. She was prepared to talk about one case that did not go well. The cause of mishap had been identified and lessons learnt. She had reflected on the incident and could discuss it confidently and rationally stressing the lessons and not the failure. She was not asked.
Finally, she was asked if she had any questions. She thanked the panel for the opportunity, said she had talked with doctors and nurses on the ward, and had no questions. She was told a decision would be made later that afternoon and that she could wait for the result if she wished. She said she would wait. She left the interview room confidently and politely.
At 5.30 pm., three of the 10 interviewees were offered jobs. Sara was among the successful doctors. She accepted the job offer.
Further reading
Turya EB. Your career after PLAB: survival tools for young doctors. Manchester: Edukom, 2003.
McErin S. Writing the medical CV. Manchester: Edukom, 2004.
--------------------------------------------------------------------------------
Elitham B Turya, consultant in child health
BMJ 2004;328 (19 June), doi:10.1136/bmj.328.7454.0-f
Treat minor burns effectively
Most minor burns can be safely managed in primary care. In the third article of our ABC of burns (p 1487), Hudspith and Rayatt give some important tips on how to treat patients with minor burns. The first aid given can influence the final cosmetic outcome: stopping the burning process, prompt cooling with tepid tap water, covering the burn (cling film is ideal), and keeping the patient warm is paramount. Burns should be kept clean, but routine use of antibiotics should be discouraged, say the authors. If the burns have not healed within two weeks, refer the patients to a burn surgeon.
comment: good advice, remember the cling film
Treat minor burns effectively
Most minor burns can be safely managed in primary care. In the third article of our ABC of burns (p 1487), Hudspith and Rayatt give some important tips on how to treat patients with minor burns. The first aid given can influence the final cosmetic outcome: stopping the burning process, prompt cooling with tepid tap water, covering the burn (cling film is ideal), and keeping the patient warm is paramount. Burns should be kept clean, but routine use of antibiotics should be discouraged, say the authors. If the burns have not healed within two weeks, refer the patients to a burn surgeon.
comment: good advice, remember the cling film
BMJ 2004;328 (19 June), doi:10.1136/bmj.328.7454.0-d
Nursing home effect may explain high mortality
Looking after patients who live in nursing homes, and who are more likely to die than patients in the community, may explain excessively high mortality in general practice. Mohammed and colleagues (p 1474) analysed the case mix and results of two general practitioners flagged up by the Shipman inquiry as having higher mortality than expected, and they found that the GPs were looking after a high proportion of patients living in nursing homes, which explained their results. GPs with high mortality merit a proper investigation for credible causes, say the authors.
Comment: good commom sense, numbers often lie and looking into such figures can reveal the truth
Nursing home effect may explain high mortality
Looking after patients who live in nursing homes, and who are more likely to die than patients in the community, may explain excessively high mortality in general practice. Mohammed and colleagues (p 1474) analysed the case mix and results of two general practitioners flagged up by the Shipman inquiry as having higher mortality than expected, and they found that the GPs were looking after a high proportion of patients living in nursing homes, which explained their results. GPs with high mortality merit a proper investigation for credible causes, say the authors.
Comment: good commom sense, numbers often lie and looking into such figures can reveal the truth
Saturday, June 12, 2004
How the pharmaceutical industry influences the medical literature
Read
Key points
How a trial is funded can affect its outcome or conclusion2-5
In one recent survey, most authors of clinical practice guidelines did not declare their ties with the drug industry6
Editorials give clinical experts a chance to express their opinion. There is evidence that this opinion may be influenced by whether the author has ties with the drug industry7
Some of the articles you read may have been ghost written by writers employed by the drug industry8 9
A scenario to think about
You do two sessions a week as a clinical assistant in your local accident and emergency department. The department is reviewing its policy on treating patients with acute stroke. You've been asked to review international recommendations on the use of thrombolysis for acute stroke.
You discover that in 1996 the US Food and Drug Administration approved the use of the thrombolytic drug alteplase (tissue plasminogen activator) in ischaemic stroke. In its guidelines on emergency cardiovascular care, the American Heart Association initially classified this treatment as an "optional" ("class IIb") intervention. But in 2000 the association upgraded its advice, giving the very highest support ("definitely recommended" or "class I") to the use of this drug in acute stroke.10
You then read an article in the BMJ that points out that most of the panellists who made the recommendation to upgrade the advice had ties with Genentech, the US manufacturer of alteplase. The article also points out that Genentech contributed over $11m (£7.8m) to the American Heart Association in the decade before its recommendation on alteplase.10
So what's the problem?
Well, it's impossible to know for sure whether Genentech influenced the association's recommendation. The association's president strongly denied that this had happened.11 But I would argue that there is surely the appearance of potential bias in these clinical guidelines in that the company funded most of the panellists and gave enormous financial donations to the association.
In this module, I discuss some of the ways in which certain drug companies might be influencing what you're reading.
The drug industry's influence on clinical trials
The influence of drug industry funding on trial results and conclusions
There is evidence that the funding source of a trial affects the outcome or conclusion of the trial.
In a systematic review, Lexchin and colleagues looked at 30 studies that compared research sponsored by the drug industry with research that was funded in other ways. The studies sponsored by pharmaceutical companies were four times as likely to have outcomes favouring the sponsor than the studies funded by other sources2
In a second systematic review, Bekelman and colleagues reviewed all English language original, quantitative studies on the extent, impact, and management of financial conflicts of interest in biomedical research. They found a significant association between industry sponsorship and pro-industry conclusions (odds ratio, 3.60; 95% confidence interval, 2.63-4.91). Industry sponsorship was also associated with restrictions on publication and data sharing3
Davidson analysed 107 controlled clinical trials and classified them as favouring either a new therapy or a traditional therapy, and as being supported by a drug company or not having industry support. There was a significant association between the source of funding and the outcome of the study. Trials funded by the manufacturer of a new drug were significantly more likely to favour the new drug than traditional treatment4
In a systematic review of 44 trials of cancer drugs, only 5% of industry sponsored studies reached unfavourable conclusions about the company's drugs compared with 38% of studies with non-profit funding - a significant difference5
The influence of the drug industry on trial design
Bodenheimer has examined some of the ways that drug companies can choose study designs to make their products appear as enticing as possible.12 As two BMJ editors said, "Many tricks can be used to give companies the results they want."1 Here are some examples.
The new drug is tested against a placebo
Instead of against the standard evidence-based treatment.13 The World Medical Association Declaration of Helsinki deems it unethical to give trial patients a placebo if an evidence based treatment is available.14
The new drug is tested against an insufficient dose of a competitor drug
Rochon and colleagues reviewed 56 industry-sponsored randomised controlled trials of non-steroidal anti-inflammatory drugs for arthritis.15 Every trial found that the trial sponsor's drug was equal to or better than the competitor drug. But in half of the trials the dose of the sponsor's drug was higher.
To make the new drug's side effect profile look favourable, the drug is tested against a higher dose of the competitor drug
This may have happened with trials of new antidepressants, where the selling point is not that they are more effective but that they are less toxic.13
Testing the new drug in a young healthy population, rather than in the population that will receive it
Why would a company want to do this type of trial? The answer is that the incidence of side effects of a drug is usually higher in an elderly population with disease than in a younger healthy group. So by doing the trial in young healthy people, the side effect profile of the drug seems more favourable. For example, non-steroidal anti-inflammatory drugs are often used in elderly people because of the high prevalence of musculoskeletal disorders, especially in women. And yet in the major trials evaluating these drugs, only 2.1% of patients were older than 65 years and less than 0.1% were older than 75 years.16
Using surrogate rather than clinical end points
Surrogate end points are generally laboratory values (such as serum cholesterol) or physical signs (such as blood pressure) used as substitutes for clinically important end points (such as cardiac deaths). Using surrogate end points generally allows drug companies to do smaller, cheaper studies than using clinical end points.17 Although an intervention may exert a favourable outcome as judged from its effect on the surrogate end point, it may not improve - or it may even compromise - health. For example, the Cardiac Arrhythmia Suppression Trial (CAST) study found that class Ic antiarrhythmic drugs suppressed premature ventricular complexes (a surrogate end point), yet mortality was unexpectedly increased.18
The influence of the drug industry on data analysis
The way in which a trial is analysed can influence the trial results. Some drug companies sometimes selectively report only the more favourable analyses. For example, Melander and colleagues looked at placebo controlled studies of selective serotonin reuptake inhibitors. They found that many studies selectively reported only the most favourable analyses and ignored the less favourable ones.19
Drug industry control over trial publication
Company-sponsored trials with disappointing or negative results are less likely to be submitted for publication than those with positive results.20 This is known as publication bias and it could have awful consequences for patients.
Here are just a few examples of companies trying to suppress the publication of unfavourable data:
In 1987, the manufacturer of a new preparation of levothyroxine (Synthroid) contracted with a researcher at the University of California to compare the new drug with existing thyroid preparations. In 1990, the researcher found that the new drug was no better than cheaper alternatives. The manufacturer refused to allow the results to be published21
A company called the Immune Response Corporation contracted with a University of California researcher to carry out a randomised controlled trial of a new immune system modulator to treat HIV infection. The researcher found that the modulator was ineffective. The company tried to suppress publication of the study22
A randomised trial of the class I antiarrhythmic drug lorcainide in patients with acute myocardial infarction, carried out in 1980, remained unreported for 13 years despite the increased death rate in the lorcainide group.23 The use of class I antiarrhythmic drugs generally increased through the 1980s. "At the peak of their use in the late 1980s," said Ian Chalmers of the Cochrane Centre, "it has been estimated that class I antiarrhythmic drugs given to people with heart attacks were causing between 20 000 and 70 000 premature deaths every year in the United States alone. This yearly total of deaths is of the same magnitude as the total number of Americans who died in the Vietnam war."24
Another way in which the drug industry might influence publication is duplicate publication of favourable trials (that is, a single trial is published many times in many different journals). By changing the names of the authors or the language (say from English to French), or by adding a few extra data, such duplicate publication can often go unnoticed.
One problem with covert duplicate publication is that it leads to overestimation of a drug's efficacy. For example, while undertaking a systematic search for published randomised trials of ondansetron as a postoperative antiemetic, Tramèr and colleagues found that 14 out of 84 trials (17%) were duplicates.25 Data from 11 980 patients receiving ondansetron appeared in the literature although the data were originally from only 8645 patients (72%). The number needed to treat to prevent vomiting with intravenous ondansetron (4 mg) compared with placebo was 9.5 in 16 reports which were never duplicated, but only 3.9 in three reports which were duplicated. So a reader unaware of the duplicate publication would have overestimated the antiemetic efficacy of ondansetron.
Drug industry influence on clinical practice guidelines
Clinical practice guidelines give a summary of the best evidence and make recommendations on what to do. Expert clinicians usually write them. These guidelines probably affect the practice of large numbers of doctors. So any influence that a drug company might exert on guideline authors may be transmitted many times over to readers. We would all hope, therefore, that guideline authors are very explicit about their ties with industry.
Yet one study found that most guideline authors did not declare their industry ties. Choudhry and colleagues surveyed 192 authors of 44 guidelines endorsed by North American and European societies on common adult diseases published between 1991 and July 1999. One hundred authors (52%) provided usable responses, representing 37 of the 44 different guidelines.6
Eighty-seven percent of authors had some form of interaction with the pharmaceutical industry. Fifty-eight percent had received financial support to perform research and 38% had served as employees or consultants for a pharmaceutical company. In published versions of the guidelines, however, only in two cases did the authors declare their ties with industry.
Yet some authors did believe that industry ties could have influenced the recommendations:
7% thought that their own relationships with the pharmaceutical industry influenced the recommendations
19% thought that their coauthors' recommendations were influenced by their relationships.
Drug industry influence on editorials
Editorials give clinical experts a chance to express a stated opinion and this opinion may be influenced by whether the author has ties with industry.
You may remember that, in the 1990s, there was a very intense debate about whether calcium channel antagonists increased the risk of death when given to patients with coronary artery disease.26 27 Stelfox and colleagues searched the English language medical literature published from March 1995 to September 1996 for articles examining the controversy about the safety of these drugs.7 Authors who supported the use of calcium channel antagonists were significantly more likely than neutral or critical authors to have financial relationships with manufacturers of these drugs (96% vs 60% and 37% respectively; P<0.001). Supportive authors were also more likely than neutral or critical authors to have financial relationships with any pharmaceutical manufacturer, irrespective of the product (100%, vs 67% and 43%, respectively; P<0.001).
Ghost writing
The New York Times and the Lancet have recently discussed how certain drug companies have written editorials themselves and then published them under the name of a respected clinician (who may not have even seen the piece before it went to print).8 9 This is known as the "ghost-guest syndrome." The ghost is the unnamed industry writer and the guest is the prestigious named author.
The New York Times story about the marketing of the drug gabapentin (Neurontin) illustrates this syndrome.8 In the US, the Food and Drug Administration has only approved Neurontin for two limited indications - seizure control in patients with epilepsy who are already on one drug and post-herpetic neuralgia. There is no good evidence to support its use for other neurological conditions. The New York Times reported that Warner-Lambert, the makers of Neurontin, hired two marketing firms to ghost write articles claiming the benefits of Neurontin for many unapproved uses. The newspaper said that the company then paid physicians $1000 to act as "guest" authors.8
Further reading and resources
BMJ theme issue: Time to untangle doctors from drug companies. Available at: http://bmj.bmjjournals.com/content/vol326/issue7400 (Accessed 20 April 2004.)
Editorial. The invisible hand of the marketing department. Can Med Assoc J 2002;167:5. Available at http://www.cmaj.ca/cgi/content/full/167/1/5
Liberati A, et al. Which guidelines can we trust? West J Med 2001;174:262-5.
Als-Nielsen B, et al. Association of funding and conclusions in randomised drug trials. JAMA 2003;290:921-8.
Royal College of Physicians: The relationship between physicians and the pharmaceutical industry. J R Coll Physic Lond 1986;20:235-42.
http://www.nofreelunch.org
This is a US-based non-profit group that serves as "a source of information, inspiration, and assistance for those who are trying to rid their practices, institutions, and colleagues of promotional influence."
Abbasi K, Smith R. No more free lunches. BMJ 2003;326:1155-6.
Lexchin J, Bero LA, Djulbegovic B, Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ 2003;326:1167-70.
Bekelman JE, Li Y, Gross CP. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. JAMA 2003;289:454-65.
Davidson RA. Source of funding and outcome of clinical trials. J Gen Intern Med 1986;1:155-8.
Friedberg M, et al. Evaluation of conflict of interest in economic analyses of new drugs used in oncology. JAMA 1999;282:1453-7.
Choudhry NK, Stelfox HT, Detsky AS. Relationships between authors of clinical practice guidelines and the pharmaceutical Industry. JAMA 2002;287:612-7.
Stelfox HT, et al. Conflict of interest in the debate over calcium-channel antagonists. N Engl J Med 1998;338:101-6.
Drug company sued for promoting drugs in exam rooms. New York Times, 15 May 2002.
Larkin M. Whose article is it anyway? Lancet 1999;354:136.
Lenzer J. Alteplase for stroke: money and optimistic claims buttress the "brain attack" campaign. BMJ 2002;324:723-9.
Faxon D. Setting the record straight. Available at: http://bmj.bmjjournals.com/cgi/eletters/324/7339/723#21263 (Accessed 20 April 2004.)
Bodenheimer T. Uneasy alliance: clinical investigators and the pharmaceutical industry. N Engl J Med 2000;342:1539-44.
Smith R. Medical journals and pharmaceutical companies: uneasy bedfellows. BMJ 2003;326:1202-5.
World Medical Association Declaration of Helsinki. Available at: http://www.wma.net/e/policy/b3.htm (Accessed 20 April 2004.)
Rochon PA, et al. A study of manufacturer-supported trials of nonsteroidal anti-inflammatory drugs in the treatment of arthritis. Arch Intern Med 1994;154:157-63.
Rochon PA, et al. Reporting of age data in clinical trials of arthritis. Deficiencies and solutions. Arch Intern Med 1993;153:243-8.
Sobel BE, Levine MA. Medical education, evidence-based medicine, and the disqualification of physician-scientists. Experimental Biol Med 2001;226:713-6. Also available at: http://www.ebmonline.org/cgi/content/full/226/8/713
Epstein AE, Bigger JT Jr, Wyse DG, Romhilt DW, Reynolds-Haertle RA, Hallstrom AP. Events in the cardiac arrhythmia suppression trial (CAST): mortality in the entire population enrolled. J Am Coll Cardiol 1991;18:14-9.
Melander H, Ahlqvist-Rastad J, Meijer G, Beermann B. Evidence b(i)ased medicine - selective reporting from studies sponsored by pharmaceutical industry: review of studies in new applications. BMJ 2003;326:1171-3.
Dickersin K. How important is publication bias? A synthesis of available data. AIDS Educat Prevent 1997;9:15-21.
Rennie D. Thyroid storm. JAMA 1997;277:1238-43.
Andreopoulos S. The unhealthy alliance between academia and corporate America. West J Med 2001;175:225-6.
Cowley AJ. The effect of lorcainide on arrhythmias and survival in patients with acute myocardial infarction: an example of publication bias. Int J Cardiol 1993;40:161-6.
Yamey G. Scientists who do not publish trial results are "unethical." BMJ 1999;319:939.
Tram?r MR, Reynolds DJ, Moore RA, McQuay HJ. Impact of covert duplicate publication on meta-analysis: a case study. BMJ 1997;315:635-40.
Psaty BM, Heckbert SR, Koepsell TD, et al. The risk of myocardial infarction associated with antihypertensive drug therapies. JAMA 1995;274:620-5.
Furberg CD, Psaty BM, Meyer JV. Nifedipine: dose-related increase in mortality in patients with coronary heart disease. Circulation 1995;92:1326-31.
Read, reflect, respond module
Drug companies' influence
How the pharmaceutical industry influences the medical literature
Reflect
How to reflect
Reflecting isn't just about closing your eyes and having a think. To really reflect you should ask yourself these questions:
What do I think this learning module was about?
Can I apply it in my work?
What barriers am I likely to come across?
How will I manage these barriers?
How will I know if I'm doing things better?
Review these questions and answers to help you reflect on the issues discussed in this module.
I can't believe that ghost writing is common. Surely it's an isolated rarity?
There hasn't been a lot of good research on the prevalence of ghost writing so it is hard to say exactly how common this phenomenon is.
Flanagin and colleagues surveyed the corresponding authors of 809 articles (editorials, reviews, and original research) in six well known US medical journals.28 They found that 11% of the articles had ghost authors - individuals who had made substantial contributions to the research or writing of the article but who were not named as authors. It is unclear from the study whether these ghost authors were paid by industry, and this has been the subject of great debate.9 29
Newspapers have often reported stories about industry sponsored ghost writing in the medical literature. For example, on 23 May 1999 the Dallas Morning News reported that Wyeth-Ayerst Laboratories, maker of the weight-loss drug dexfenfluramine (Redux), had "paid ghost writers for articles promoting obesity treatments and then used prominent researchers to publish the work [in journals] under their names". Two of the Wyeth-supported articles were published in peer reviewed medical journals before Redux production was halted.9
Drummond Rennie, one of the editors of JAMA, has said: "The practice [of ghost writing] is well known, scandalous, and outrageous. It is a perfect illustration of deceptive authorship practices for commercial reasons."9
And what do ghost writers themselves say? Here's what one anonymous ghost writer told the Lancet:
"I recently had my first and last experience as a "ghost writer" for a medical communications company. I agreed to do two reviews for a supplement to appear under the names of respected 'authors.' I was given an outline, references, and a list of drug company approved phrases. I was asked to sign an agreement stating that I would not disclose anything about the project. I was pressured to rework my drafts to position the product more favourably, and was shown another company produced review as an example - it read like bad promotional writing. I asked the company to reduce my fee and rewrite the drafts themselves."9
Isn't it naive to think that drug companies will want to publish trials showing their new drug is no better than an existing one?
I don't think this is a question of naiveté but of ethics. Failure to publish the results of randomised controlled trials is arguably a form of scientific and ethical misconduct.24 There are many reasons why patients, clinicians, and researchers should have access to these results, such as:
The results may have important implications for patient safety
Researchers systematically reviewing the literature should have unfettered access to all the data
To prevent costly duplication of research
To foster multicentre research collaboration.
One way to ensure that all trials are published, regardless of the results, is by having an international register of all clinical trials.24 An example of an online register is Current Controlled Trials (http://www.controlled-trials.com).
Given that drug companies fund clinical trials, isn't it inevitable that they will influence the way in which the study is conducted, analysed, or published?
It isn't inevitable. Many researchers sign contracts with their funders that give the researchers the right to handle the data and publish the trial results without the funder's involvement.
As an example of how it is perfectly possible to receive sponsorship for a study without the sponsor interfering with the study's conduct or publication, here is a statement from a systematic review published in the Lancet30:
"Role of the funding source:
The sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report."
How much can we trust drug advertisements?
There is good evidence to show that many drug advertisements are misleading.
Villaneuva and colleagues assessed all advertisements for antihypertensive and lipid-lowering drugs published in six Spanish medical journals in 1997 that had at least one bibliographical reference.31 In 45 out of 125 promotional claims in the advertisements, the reference did not support the promotional statement:
Twenty claims that were not supported by the reference recommended the drug for a patient group other than that studied in the trial
In another 15 claims, the advertisement transferred results from studies in high risk groups to the general patient population
Four claims focused on the benefits of the advertised drug for specific populations (such as elderly people or patients with diabetes) that had been either excluded or not analysed in the study referenced
One claim transferred results that had been obtained in vitro or in animals to human beings
In 10 advertisements the publicity exaggerated the results reported
In nine advertisements, false statements were made, corresponding mostly to claims of risk reduction that had not been significant in the research referenced
In six advertisements, the work referenced had no relation to the promotional claim.
Wilkes and colleagues assessed all 109 full page advertisements from 10 leading medical journals.32 The authors were able to find four-fifths of the references cited in the advertisements. They then sent the advertisements and the references to specialist reviewers, asking them to evaluate the advertisements using Food and Drug Administration criteria for acceptable drug advertising.
In a third of cases two or more reviewers disagreed with the advertiser's claim that the drug was the "drug of choice." In 40% of advertisements the reviewers thought that information on efficacy was not balanced with that on side effects and contraindications. Overall, reviewers would not have recommended publication of 28% of the advertisements and would have required major revisions in a third.
I recently read a drug advertisement in a medical journal. The advertisement seemed to me to grossly exaggerate the effects of the drug. What should I do?
Complaints about the promotion of medicines should be submitted to the Director of the Prescription Medicines Code of Practice Authority, 12 Whitehall, London SW1A 2DY. The telephone number is 020 7930 9677.
You could also contact the editor of the medical journal to share your concerns. Medical journal editors should not publish misleading advertisements. In its guidelines for medical editors on good publication practice, the Committee on Publication Ethics says: "Advertisements that mislead must be refused, and editors must be willing to publish criticisms, according to the same criteria used for material in the rest of the journal."33
Finally, you can get advice from the Advertising Standards Authority (http://www.asa.org.uk), the United Kingdom's independent, self regulatory body for non-broadcast advertisements, sales promotions, and direct marketing in the UK.
I read an article reporting the result of a randomised controlled trial in a journal and it reminded me of another article I had read in the past. I found the original article and it quickly became obvious that the two articles were essentially reporting the same data. What should I do?
In the interests of patient safety, and of scientific integrity, I would argue that you should contact the editors of the two journals that have published the same data, saying that you are worried that this is a case of duplicate publication. They can then investigate this possibility further, and they can also refer the case to the Committee on Publication Ethics (http://www.publicationethics.org.uk).
It may turn out not to be duplicate publication. Perhaps, for example, the authors were totally upfront about their previous published work when they submitted it to the second journal and the editor of the second journal felt that the submission was different enough to warrant publication.
In its guidelines on good publication practice, the Committee on Publication Ethics says that duplicate (or redundant) publication occurs when, "two or more papers, without full cross reference, share the same hypothesis, data, discussion points, or conclusions."33 The guidelines also say:
Published studies do not need to be repeated unless further confirmation is required
Previous publication of an abstract during the proceedings of meetings does not preclude subsequent submission for publication, but full disclosure should be made at the time of submission
Republication of a paper in another language is acceptable, provided that there is full and prominent disclosure of its original source at the time of submission
At the time of submission, authors should disclose details of related papers, even if in a different language, and similar papers in press.
What medical journals should I read? Which are the least biased?
Almost all medical journals carry drug advertisements and publish articles written by authors with ties to industry. Here are a few pointers to help you work out whether the journal you are reading is a drug company marketing tool or a source of valuable, evidence based information:
There should be a very clear distinction between the advertisements and the editorial content - the articles should not read like "advertorials"
A drug advertisement should be placed many pages away from any editorial content that mentions that drug. Advertisers often want to know what is to be published in a journal so they can position their advertising alongside editorial material favourable to their products. Unfortunately, many journals seem to sell advertising space on this basis13
Authors (and ideally the journal's editors) should explicitly declare their competing interests, including their ties with industry, so that you can then make a judgment as to whether you think an article is biased
Journals sometimes publish supplements sponsored by drug companies, and you should be aware that papers in supplements may be of poorer quality than those published in the main journal.34 35 A supplement often reports the papers that were presented at a symposium. Richard Smith, editor of the BMJ, says: "If a journal is willing to publish every paper presented at a symposium that was funded by a single company and that dealt with one drug, then it can charge a substantial fee. Often these papers will be set pieces by, to be crude for a moment, 'paid industry hacks' and will have been published many times."13
There are a few journals that don't carry drug advertisements at all. An example is the Drugs and Therapeutics Bulletin (http://www.dtb.org.uk/dtb/index.html), published by the Consumers' Association. The bulletin says that it is "wholly independent of industry, advertising, Government, regulatory authorities, and the medical establishment."
Abbasi K, Smith R. No more free lunches. BMJ 2003;326:1155-6.
Lexchin J, Bero LA, Djulbegovic B, Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ 2003;326:1167-70.
Bekelman JE, Li Y, Gross CP. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. JAMA 2003;289:454-65.
Davidson RA. Source of funding and outcome of clinical trials. J Gen Intern Med 1986;1:155-8.
Friedberg M, et al. Evaluation of conflict of interest in economic analyses of new drugs used in oncology. JAMA 1999;282:1453-7.
Choudhry NK, Stelfox HT, Detsky AS. Relationships between authors of clinical practice guidelines and the pharmaceutical Industry. JAMA 2002;287:612-7.
Stelfox HT, et al. Conflict of interest in the debate over calcium-channel antagonists. N Engl J Med 1998;338:101-6.
Drug company sued for promoting drugs in exam rooms. New York Times, 15 May 2002.
Larkin M. Whose article is it anyway? Lancet 1999;354:136.
Lenzer J. Alteplase for stroke: money and optimistic claims buttress the "brain attack" campaign. BMJ 2002;324:723-9.
Faxon D. Setting the record straight. Available at: http://bmj.bmjjournals.com/cgi/eletters/324/7339/723#21263 (Accessed 20 April 2004.)
Bodenheimer T. Uneasy alliance: clinical investigators and the pharmaceutical industry. N Engl J Med 2000;342:1539-44.
Smith R. Medical journals and pharmaceutical companies: uneasy bedfellows. BMJ 2003;326:1202-5.
World Medical Association Declaration of Helsinki. Available at: http://www.wma.net/e/policy/b3.htm (Accessed 20 April 2004.)
Rochon PA, et al. A study of manufacturer-supported trials of nonsteroidal anti-inflammatory drugs in the treatment of arthritis. Arch Intern Med 1994;154:157-63.
Rochon PA, et al. Reporting of age data in clinical trials of arthritis. Deficiencies and solutions. Arch Intern Med 1993;153:243-8.
Sobel BE, Levine MA. Medical education, evidence-based medicine, and the disqualification of physician-scientists. Experimental Biol Med 2001;226:713-6. Also available at: http://www.ebmonline.org/cgi/content/full/226/8/713
Epstein AE, Bigger JT Jr, Wyse DG, Romhilt DW, Reynolds-Haertle RA, Hallstrom AP. Events in the cardiac arrhythmia suppression trial (CAST): mortality in the entire population enrolled. J Am Coll Cardiol 1991;18:14-9.
Melander H, Ahlqvist-Rastad J, Meijer G, Beermann B. Evidence b(i)ased medicine - selective reporting from studies sponsored by pharmaceutical industry: review of studies in new applications. BMJ 2003;326:1171-3.
Dickersin K. How important is publication bias? A synthesis of available data. AIDS Educat Prevent 1997;9:15-21.
Rennie D. Thyroid storm. JAMA 1997;277:1238-43.
Andreopoulos S. The unhealthy alliance between academia and corporate America. West J Med 2001;175:225-6.
Cowley AJ. The effect of lorcainide on arrhythmias and survival in patients with acute myocardial infarction: an example of publication bias. Int J Cardiol 1993;40:161-6.
Yamey G. Scientists who do not publish trial results are "unethical." BMJ 1999;319:939.
Tramèr MR, Reynolds DJ, Moore RA, McQuay HJ. Impact of covert duplicate publication on meta-analysis: a case study. BMJ 1997;315:635-40.
Psaty BM, Heckbert SR, Koepsell TD, et al. The risk of myocardial infarction associated with antihypertensive drug therapies. JAMA 1995;274:620-5.
Furberg CD, Psaty BM, Meyer JV. Nifedipine: dose-related increase in mortality in patients with coronary heart disease. Circulation 1995;92:1326-31.
Flanagin A et al. Prevalence of articles with honorary authors and ghost authors in peer-reviewed medical journals. JAMA 1998;280:222-4.
Phillips SG, Carey LA. Comment in: Lancet 1999;354:1563.
Fazel S, Danesh J. Serious mental disorder in 23 000 prisoners: a systematic review of 62 surveys. Lancet 2002;359:545-50.
Villanueva P, Peiro S, Librero J, Pereiro I. Accuracy of pharmaceutical advertisements in medical journals. Lancet 2003;361:27-32.
Wilkes MS, Doblin BH, Shapiro MF. Pharmaceutical advertisements in leading medical journals: experts' assessments. Ann Intern Med 1992;116:912-9.
Committee on Publication Ethics. Guidelines on good publication practice. Available at: http://www.publicationethics.org.uk/cope1999/gpp/gpp.phtml#gpp (Accessed 20 April 2004.)
Rochon PA, Gurwitz JH, Cheung M, Hayes JA, Chalmers TC. Evaluating the quality of articles published in journal supplements compared with the quality of those published in the parent journal. JAMA 1994;272:108-13.
Cho MK, Bero LA. The quality of drug studies published in symposium proceedings. Ann Intern Med 1996;124:485-9.
Competing interests:
GY used to work for the BMJ Publishing Group. BMJ Publishing Group journals, such as the weekly BMJ, carry drug advertisements.
Read, reflect, respond module
Drug companies' influence
Comment:
good module from BMJlearning re: probity and drugs company, did not know about ghost writing and the "sneaky" ways in which data is published or suppressed to make the drug look better than is.
Read
Key points
How a trial is funded can affect its outcome or conclusion2-5
In one recent survey, most authors of clinical practice guidelines did not declare their ties with the drug industry6
Editorials give clinical experts a chance to express their opinion. There is evidence that this opinion may be influenced by whether the author has ties with the drug industry7
Some of the articles you read may have been ghost written by writers employed by the drug industry8 9
A scenario to think about
You do two sessions a week as a clinical assistant in your local accident and emergency department. The department is reviewing its policy on treating patients with acute stroke. You've been asked to review international recommendations on the use of thrombolysis for acute stroke.
You discover that in 1996 the US Food and Drug Administration approved the use of the thrombolytic drug alteplase (tissue plasminogen activator) in ischaemic stroke. In its guidelines on emergency cardiovascular care, the American Heart Association initially classified this treatment as an "optional" ("class IIb") intervention. But in 2000 the association upgraded its advice, giving the very highest support ("definitely recommended" or "class I") to the use of this drug in acute stroke.10
You then read an article in the BMJ that points out that most of the panellists who made the recommendation to upgrade the advice had ties with Genentech, the US manufacturer of alteplase. The article also points out that Genentech contributed over $11m (£7.8m) to the American Heart Association in the decade before its recommendation on alteplase.10
So what's the problem?
Well, it's impossible to know for sure whether Genentech influenced the association's recommendation. The association's president strongly denied that this had happened.11 But I would argue that there is surely the appearance of potential bias in these clinical guidelines in that the company funded most of the panellists and gave enormous financial donations to the association.
In this module, I discuss some of the ways in which certain drug companies might be influencing what you're reading.
The drug industry's influence on clinical trials
The influence of drug industry funding on trial results and conclusions
There is evidence that the funding source of a trial affects the outcome or conclusion of the trial.
In a systematic review, Lexchin and colleagues looked at 30 studies that compared research sponsored by the drug industry with research that was funded in other ways. The studies sponsored by pharmaceutical companies were four times as likely to have outcomes favouring the sponsor than the studies funded by other sources2
In a second systematic review, Bekelman and colleagues reviewed all English language original, quantitative studies on the extent, impact, and management of financial conflicts of interest in biomedical research. They found a significant association between industry sponsorship and pro-industry conclusions (odds ratio, 3.60; 95% confidence interval, 2.63-4.91). Industry sponsorship was also associated with restrictions on publication and data sharing3
Davidson analysed 107 controlled clinical trials and classified them as favouring either a new therapy or a traditional therapy, and as being supported by a drug company or not having industry support. There was a significant association between the source of funding and the outcome of the study. Trials funded by the manufacturer of a new drug were significantly more likely to favour the new drug than traditional treatment4
In a systematic review of 44 trials of cancer drugs, only 5% of industry sponsored studies reached unfavourable conclusions about the company's drugs compared with 38% of studies with non-profit funding - a significant difference5
The influence of the drug industry on trial design
Bodenheimer has examined some of the ways that drug companies can choose study designs to make their products appear as enticing as possible.12 As two BMJ editors said, "Many tricks can be used to give companies the results they want."1 Here are some examples.
The new drug is tested against a placebo
Instead of against the standard evidence-based treatment.13 The World Medical Association Declaration of Helsinki deems it unethical to give trial patients a placebo if an evidence based treatment is available.14
The new drug is tested against an insufficient dose of a competitor drug
Rochon and colleagues reviewed 56 industry-sponsored randomised controlled trials of non-steroidal anti-inflammatory drugs for arthritis.15 Every trial found that the trial sponsor's drug was equal to or better than the competitor drug. But in half of the trials the dose of the sponsor's drug was higher.
To make the new drug's side effect profile look favourable, the drug is tested against a higher dose of the competitor drug
This may have happened with trials of new antidepressants, where the selling point is not that they are more effective but that they are less toxic.13
Testing the new drug in a young healthy population, rather than in the population that will receive it
Why would a company want to do this type of trial? The answer is that the incidence of side effects of a drug is usually higher in an elderly population with disease than in a younger healthy group. So by doing the trial in young healthy people, the side effect profile of the drug seems more favourable. For example, non-steroidal anti-inflammatory drugs are often used in elderly people because of the high prevalence of musculoskeletal disorders, especially in women. And yet in the major trials evaluating these drugs, only 2.1% of patients were older than 65 years and less than 0.1% were older than 75 years.16
Using surrogate rather than clinical end points
Surrogate end points are generally laboratory values (such as serum cholesterol) or physical signs (such as blood pressure) used as substitutes for clinically important end points (such as cardiac deaths). Using surrogate end points generally allows drug companies to do smaller, cheaper studies than using clinical end points.17 Although an intervention may exert a favourable outcome as judged from its effect on the surrogate end point, it may not improve - or it may even compromise - health. For example, the Cardiac Arrhythmia Suppression Trial (CAST) study found that class Ic antiarrhythmic drugs suppressed premature ventricular complexes (a surrogate end point), yet mortality was unexpectedly increased.18
The influence of the drug industry on data analysis
The way in which a trial is analysed can influence the trial results. Some drug companies sometimes selectively report only the more favourable analyses. For example, Melander and colleagues looked at placebo controlled studies of selective serotonin reuptake inhibitors. They found that many studies selectively reported only the most favourable analyses and ignored the less favourable ones.19
Drug industry control over trial publication
Company-sponsored trials with disappointing or negative results are less likely to be submitted for publication than those with positive results.20 This is known as publication bias and it could have awful consequences for patients.
Here are just a few examples of companies trying to suppress the publication of unfavourable data:
In 1987, the manufacturer of a new preparation of levothyroxine (Synthroid) contracted with a researcher at the University of California to compare the new drug with existing thyroid preparations. In 1990, the researcher found that the new drug was no better than cheaper alternatives. The manufacturer refused to allow the results to be published21
A company called the Immune Response Corporation contracted with a University of California researcher to carry out a randomised controlled trial of a new immune system modulator to treat HIV infection. The researcher found that the modulator was ineffective. The company tried to suppress publication of the study22
A randomised trial of the class I antiarrhythmic drug lorcainide in patients with acute myocardial infarction, carried out in 1980, remained unreported for 13 years despite the increased death rate in the lorcainide group.23 The use of class I antiarrhythmic drugs generally increased through the 1980s. "At the peak of their use in the late 1980s," said Ian Chalmers of the Cochrane Centre, "it has been estimated that class I antiarrhythmic drugs given to people with heart attacks were causing between 20 000 and 70 000 premature deaths every year in the United States alone. This yearly total of deaths is of the same magnitude as the total number of Americans who died in the Vietnam war."24
Another way in which the drug industry might influence publication is duplicate publication of favourable trials (that is, a single trial is published many times in many different journals). By changing the names of the authors or the language (say from English to French), or by adding a few extra data, such duplicate publication can often go unnoticed.
One problem with covert duplicate publication is that it leads to overestimation of a drug's efficacy. For example, while undertaking a systematic search for published randomised trials of ondansetron as a postoperative antiemetic, Tramèr and colleagues found that 14 out of 84 trials (17%) were duplicates.25 Data from 11 980 patients receiving ondansetron appeared in the literature although the data were originally from only 8645 patients (72%). The number needed to treat to prevent vomiting with intravenous ondansetron (4 mg) compared with placebo was 9.5 in 16 reports which were never duplicated, but only 3.9 in three reports which were duplicated. So a reader unaware of the duplicate publication would have overestimated the antiemetic efficacy of ondansetron.
Drug industry influence on clinical practice guidelines
Clinical practice guidelines give a summary of the best evidence and make recommendations on what to do. Expert clinicians usually write them. These guidelines probably affect the practice of large numbers of doctors. So any influence that a drug company might exert on guideline authors may be transmitted many times over to readers. We would all hope, therefore, that guideline authors are very explicit about their ties with industry.
Yet one study found that most guideline authors did not declare their industry ties. Choudhry and colleagues surveyed 192 authors of 44 guidelines endorsed by North American and European societies on common adult diseases published between 1991 and July 1999. One hundred authors (52%) provided usable responses, representing 37 of the 44 different guidelines.6
Eighty-seven percent of authors had some form of interaction with the pharmaceutical industry. Fifty-eight percent had received financial support to perform research and 38% had served as employees or consultants for a pharmaceutical company. In published versions of the guidelines, however, only in two cases did the authors declare their ties with industry.
Yet some authors did believe that industry ties could have influenced the recommendations:
7% thought that their own relationships with the pharmaceutical industry influenced the recommendations
19% thought that their coauthors' recommendations were influenced by their relationships.
Drug industry influence on editorials
Editorials give clinical experts a chance to express a stated opinion and this opinion may be influenced by whether the author has ties with industry.
You may remember that, in the 1990s, there was a very intense debate about whether calcium channel antagonists increased the risk of death when given to patients with coronary artery disease.26 27 Stelfox and colleagues searched the English language medical literature published from March 1995 to September 1996 for articles examining the controversy about the safety of these drugs.7 Authors who supported the use of calcium channel antagonists were significantly more likely than neutral or critical authors to have financial relationships with manufacturers of these drugs (96% vs 60% and 37% respectively; P<0.001). Supportive authors were also more likely than neutral or critical authors to have financial relationships with any pharmaceutical manufacturer, irrespective of the product (100%, vs 67% and 43%, respectively; P<0.001).
Ghost writing
The New York Times and the Lancet have recently discussed how certain drug companies have written editorials themselves and then published them under the name of a respected clinician (who may not have even seen the piece before it went to print).8 9 This is known as the "ghost-guest syndrome." The ghost is the unnamed industry writer and the guest is the prestigious named author.
The New York Times story about the marketing of the drug gabapentin (Neurontin) illustrates this syndrome.8 In the US, the Food and Drug Administration has only approved Neurontin for two limited indications - seizure control in patients with epilepsy who are already on one drug and post-herpetic neuralgia. There is no good evidence to support its use for other neurological conditions. The New York Times reported that Warner-Lambert, the makers of Neurontin, hired two marketing firms to ghost write articles claiming the benefits of Neurontin for many unapproved uses. The newspaper said that the company then paid physicians $1000 to act as "guest" authors.8
Further reading and resources
BMJ theme issue: Time to untangle doctors from drug companies. Available at: http://bmj.bmjjournals.com/content/vol326/issue7400 (Accessed 20 April 2004.)
Editorial. The invisible hand of the marketing department. Can Med Assoc J 2002;167:5. Available at http://www.cmaj.ca/cgi/content/full/167/1/5
Liberati A, et al. Which guidelines can we trust? West J Med 2001;174:262-5.
Als-Nielsen B, et al. Association of funding and conclusions in randomised drug trials. JAMA 2003;290:921-8.
Royal College of Physicians: The relationship between physicians and the pharmaceutical industry. J R Coll Physic Lond 1986;20:235-42.
http://www.nofreelunch.org
This is a US-based non-profit group that serves as "a source of information, inspiration, and assistance for those who are trying to rid their practices, institutions, and colleagues of promotional influence."
Abbasi K, Smith R. No more free lunches. BMJ 2003;326:1155-6.
Lexchin J, Bero LA, Djulbegovic B, Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ 2003;326:1167-70.
Bekelman JE, Li Y, Gross CP. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. JAMA 2003;289:454-65.
Davidson RA. Source of funding and outcome of clinical trials. J Gen Intern Med 1986;1:155-8.
Friedberg M, et al. Evaluation of conflict of interest in economic analyses of new drugs used in oncology. JAMA 1999;282:1453-7.
Choudhry NK, Stelfox HT, Detsky AS. Relationships between authors of clinical practice guidelines and the pharmaceutical Industry. JAMA 2002;287:612-7.
Stelfox HT, et al. Conflict of interest in the debate over calcium-channel antagonists. N Engl J Med 1998;338:101-6.
Drug company sued for promoting drugs in exam rooms. New York Times, 15 May 2002.
Larkin M. Whose article is it anyway? Lancet 1999;354:136.
Lenzer J. Alteplase for stroke: money and optimistic claims buttress the "brain attack" campaign. BMJ 2002;324:723-9.
Faxon D. Setting the record straight. Available at: http://bmj.bmjjournals.com/cgi/eletters/324/7339/723#21263 (Accessed 20 April 2004.)
Bodenheimer T. Uneasy alliance: clinical investigators and the pharmaceutical industry. N Engl J Med 2000;342:1539-44.
Smith R. Medical journals and pharmaceutical companies: uneasy bedfellows. BMJ 2003;326:1202-5.
World Medical Association Declaration of Helsinki. Available at: http://www.wma.net/e/policy/b3.htm (Accessed 20 April 2004.)
Rochon PA, et al. A study of manufacturer-supported trials of nonsteroidal anti-inflammatory drugs in the treatment of arthritis. Arch Intern Med 1994;154:157-63.
Rochon PA, et al. Reporting of age data in clinical trials of arthritis. Deficiencies and solutions. Arch Intern Med 1993;153:243-8.
Sobel BE, Levine MA. Medical education, evidence-based medicine, and the disqualification of physician-scientists. Experimental Biol Med 2001;226:713-6. Also available at: http://www.ebmonline.org/cgi/content/full/226/8/713
Epstein AE, Bigger JT Jr, Wyse DG, Romhilt DW, Reynolds-Haertle RA, Hallstrom AP. Events in the cardiac arrhythmia suppression trial (CAST): mortality in the entire population enrolled. J Am Coll Cardiol 1991;18:14-9.
Melander H, Ahlqvist-Rastad J, Meijer G, Beermann B. Evidence b(i)ased medicine - selective reporting from studies sponsored by pharmaceutical industry: review of studies in new applications. BMJ 2003;326:1171-3.
Dickersin K. How important is publication bias? A synthesis of available data. AIDS Educat Prevent 1997;9:15-21.
Rennie D. Thyroid storm. JAMA 1997;277:1238-43.
Andreopoulos S. The unhealthy alliance between academia and corporate America. West J Med 2001;175:225-6.
Cowley AJ. The effect of lorcainide on arrhythmias and survival in patients with acute myocardial infarction: an example of publication bias. Int J Cardiol 1993;40:161-6.
Yamey G. Scientists who do not publish trial results are "unethical." BMJ 1999;319:939.
Tram?r MR, Reynolds DJ, Moore RA, McQuay HJ. Impact of covert duplicate publication on meta-analysis: a case study. BMJ 1997;315:635-40.
Psaty BM, Heckbert SR, Koepsell TD, et al. The risk of myocardial infarction associated with antihypertensive drug therapies. JAMA 1995;274:620-5.
Furberg CD, Psaty BM, Meyer JV. Nifedipine: dose-related increase in mortality in patients with coronary heart disease. Circulation 1995;92:1326-31.
Read, reflect, respond module
Drug companies' influence
How the pharmaceutical industry influences the medical literature
Reflect
How to reflect
Reflecting isn't just about closing your eyes and having a think. To really reflect you should ask yourself these questions:
What do I think this learning module was about?
Can I apply it in my work?
What barriers am I likely to come across?
How will I manage these barriers?
How will I know if I'm doing things better?
Review these questions and answers to help you reflect on the issues discussed in this module.
I can't believe that ghost writing is common. Surely it's an isolated rarity?
There hasn't been a lot of good research on the prevalence of ghost writing so it is hard to say exactly how common this phenomenon is.
Flanagin and colleagues surveyed the corresponding authors of 809 articles (editorials, reviews, and original research) in six well known US medical journals.28 They found that 11% of the articles had ghost authors - individuals who had made substantial contributions to the research or writing of the article but who were not named as authors. It is unclear from the study whether these ghost authors were paid by industry, and this has been the subject of great debate.9 29
Newspapers have often reported stories about industry sponsored ghost writing in the medical literature. For example, on 23 May 1999 the Dallas Morning News reported that Wyeth-Ayerst Laboratories, maker of the weight-loss drug dexfenfluramine (Redux), had "paid ghost writers for articles promoting obesity treatments and then used prominent researchers to publish the work [in journals] under their names". Two of the Wyeth-supported articles were published in peer reviewed medical journals before Redux production was halted.9
Drummond Rennie, one of the editors of JAMA, has said: "The practice [of ghost writing] is well known, scandalous, and outrageous. It is a perfect illustration of deceptive authorship practices for commercial reasons."9
And what do ghost writers themselves say? Here's what one anonymous ghost writer told the Lancet:
"I recently had my first and last experience as a "ghost writer" for a medical communications company. I agreed to do two reviews for a supplement to appear under the names of respected 'authors.' I was given an outline, references, and a list of drug company approved phrases. I was asked to sign an agreement stating that I would not disclose anything about the project. I was pressured to rework my drafts to position the product more favourably, and was shown another company produced review as an example - it read like bad promotional writing. I asked the company to reduce my fee and rewrite the drafts themselves."9
Isn't it naive to think that drug companies will want to publish trials showing their new drug is no better than an existing one?
I don't think this is a question of naiveté but of ethics. Failure to publish the results of randomised controlled trials is arguably a form of scientific and ethical misconduct.24 There are many reasons why patients, clinicians, and researchers should have access to these results, such as:
The results may have important implications for patient safety
Researchers systematically reviewing the literature should have unfettered access to all the data
To prevent costly duplication of research
To foster multicentre research collaboration.
One way to ensure that all trials are published, regardless of the results, is by having an international register of all clinical trials.24 An example of an online register is Current Controlled Trials (http://www.controlled-trials.com).
Given that drug companies fund clinical trials, isn't it inevitable that they will influence the way in which the study is conducted, analysed, or published?
It isn't inevitable. Many researchers sign contracts with their funders that give the researchers the right to handle the data and publish the trial results without the funder's involvement.
As an example of how it is perfectly possible to receive sponsorship for a study without the sponsor interfering with the study's conduct or publication, here is a statement from a systematic review published in the Lancet30:
"Role of the funding source:
The sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report."
How much can we trust drug advertisements?
There is good evidence to show that many drug advertisements are misleading.
Villaneuva and colleagues assessed all advertisements for antihypertensive and lipid-lowering drugs published in six Spanish medical journals in 1997 that had at least one bibliographical reference.31 In 45 out of 125 promotional claims in the advertisements, the reference did not support the promotional statement:
Twenty claims that were not supported by the reference recommended the drug for a patient group other than that studied in the trial
In another 15 claims, the advertisement transferred results from studies in high risk groups to the general patient population
Four claims focused on the benefits of the advertised drug for specific populations (such as elderly people or patients with diabetes) that had been either excluded or not analysed in the study referenced
One claim transferred results that had been obtained in vitro or in animals to human beings
In 10 advertisements the publicity exaggerated the results reported
In nine advertisements, false statements were made, corresponding mostly to claims of risk reduction that had not been significant in the research referenced
In six advertisements, the work referenced had no relation to the promotional claim.
Wilkes and colleagues assessed all 109 full page advertisements from 10 leading medical journals.32 The authors were able to find four-fifths of the references cited in the advertisements. They then sent the advertisements and the references to specialist reviewers, asking them to evaluate the advertisements using Food and Drug Administration criteria for acceptable drug advertising.
In a third of cases two or more reviewers disagreed with the advertiser's claim that the drug was the "drug of choice." In 40% of advertisements the reviewers thought that information on efficacy was not balanced with that on side effects and contraindications. Overall, reviewers would not have recommended publication of 28% of the advertisements and would have required major revisions in a third.
I recently read a drug advertisement in a medical journal. The advertisement seemed to me to grossly exaggerate the effects of the drug. What should I do?
Complaints about the promotion of medicines should be submitted to the Director of the Prescription Medicines Code of Practice Authority, 12 Whitehall, London SW1A 2DY. The telephone number is 020 7930 9677.
You could also contact the editor of the medical journal to share your concerns. Medical journal editors should not publish misleading advertisements. In its guidelines for medical editors on good publication practice, the Committee on Publication Ethics says: "Advertisements that mislead must be refused, and editors must be willing to publish criticisms, according to the same criteria used for material in the rest of the journal."33
Finally, you can get advice from the Advertising Standards Authority (http://www.asa.org.uk), the United Kingdom's independent, self regulatory body for non-broadcast advertisements, sales promotions, and direct marketing in the UK.
I read an article reporting the result of a randomised controlled trial in a journal and it reminded me of another article I had read in the past. I found the original article and it quickly became obvious that the two articles were essentially reporting the same data. What should I do?
In the interests of patient safety, and of scientific integrity, I would argue that you should contact the editors of the two journals that have published the same data, saying that you are worried that this is a case of duplicate publication. They can then investigate this possibility further, and they can also refer the case to the Committee on Publication Ethics (http://www.publicationethics.org.uk).
It may turn out not to be duplicate publication. Perhaps, for example, the authors were totally upfront about their previous published work when they submitted it to the second journal and the editor of the second journal felt that the submission was different enough to warrant publication.
In its guidelines on good publication practice, the Committee on Publication Ethics says that duplicate (or redundant) publication occurs when, "two or more papers, without full cross reference, share the same hypothesis, data, discussion points, or conclusions."33 The guidelines also say:
Published studies do not need to be repeated unless further confirmation is required
Previous publication of an abstract during the proceedings of meetings does not preclude subsequent submission for publication, but full disclosure should be made at the time of submission
Republication of a paper in another language is acceptable, provided that there is full and prominent disclosure of its original source at the time of submission
At the time of submission, authors should disclose details of related papers, even if in a different language, and similar papers in press.
What medical journals should I read? Which are the least biased?
Almost all medical journals carry drug advertisements and publish articles written by authors with ties to industry. Here are a few pointers to help you work out whether the journal you are reading is a drug company marketing tool or a source of valuable, evidence based information:
There should be a very clear distinction between the advertisements and the editorial content - the articles should not read like "advertorials"
A drug advertisement should be placed many pages away from any editorial content that mentions that drug. Advertisers often want to know what is to be published in a journal so they can position their advertising alongside editorial material favourable to their products. Unfortunately, many journals seem to sell advertising space on this basis13
Authors (and ideally the journal's editors) should explicitly declare their competing interests, including their ties with industry, so that you can then make a judgment as to whether you think an article is biased
Journals sometimes publish supplements sponsored by drug companies, and you should be aware that papers in supplements may be of poorer quality than those published in the main journal.34 35 A supplement often reports the papers that were presented at a symposium. Richard Smith, editor of the BMJ, says: "If a journal is willing to publish every paper presented at a symposium that was funded by a single company and that dealt with one drug, then it can charge a substantial fee. Often these papers will be set pieces by, to be crude for a moment, 'paid industry hacks' and will have been published many times."13
There are a few journals that don't carry drug advertisements at all. An example is the Drugs and Therapeutics Bulletin (http://www.dtb.org.uk/dtb/index.html), published by the Consumers' Association. The bulletin says that it is "wholly independent of industry, advertising, Government, regulatory authorities, and the medical establishment."
Abbasi K, Smith R. No more free lunches. BMJ 2003;326:1155-6.
Lexchin J, Bero LA, Djulbegovic B, Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ 2003;326:1167-70.
Bekelman JE, Li Y, Gross CP. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. JAMA 2003;289:454-65.
Davidson RA. Source of funding and outcome of clinical trials. J Gen Intern Med 1986;1:155-8.
Friedberg M, et al. Evaluation of conflict of interest in economic analyses of new drugs used in oncology. JAMA 1999;282:1453-7.
Choudhry NK, Stelfox HT, Detsky AS. Relationships between authors of clinical practice guidelines and the pharmaceutical Industry. JAMA 2002;287:612-7.
Stelfox HT, et al. Conflict of interest in the debate over calcium-channel antagonists. N Engl J Med 1998;338:101-6.
Drug company sued for promoting drugs in exam rooms. New York Times, 15 May 2002.
Larkin M. Whose article is it anyway? Lancet 1999;354:136.
Lenzer J. Alteplase for stroke: money and optimistic claims buttress the "brain attack" campaign. BMJ 2002;324:723-9.
Faxon D. Setting the record straight. Available at: http://bmj.bmjjournals.com/cgi/eletters/324/7339/723#21263 (Accessed 20 April 2004.)
Bodenheimer T. Uneasy alliance: clinical investigators and the pharmaceutical industry. N Engl J Med 2000;342:1539-44.
Smith R. Medical journals and pharmaceutical companies: uneasy bedfellows. BMJ 2003;326:1202-5.
World Medical Association Declaration of Helsinki. Available at: http://www.wma.net/e/policy/b3.htm (Accessed 20 April 2004.)
Rochon PA, et al. A study of manufacturer-supported trials of nonsteroidal anti-inflammatory drugs in the treatment of arthritis. Arch Intern Med 1994;154:157-63.
Rochon PA, et al. Reporting of age data in clinical trials of arthritis. Deficiencies and solutions. Arch Intern Med 1993;153:243-8.
Sobel BE, Levine MA. Medical education, evidence-based medicine, and the disqualification of physician-scientists. Experimental Biol Med 2001;226:713-6. Also available at: http://www.ebmonline.org/cgi/content/full/226/8/713
Epstein AE, Bigger JT Jr, Wyse DG, Romhilt DW, Reynolds-Haertle RA, Hallstrom AP. Events in the cardiac arrhythmia suppression trial (CAST): mortality in the entire population enrolled. J Am Coll Cardiol 1991;18:14-9.
Melander H, Ahlqvist-Rastad J, Meijer G, Beermann B. Evidence b(i)ased medicine - selective reporting from studies sponsored by pharmaceutical industry: review of studies in new applications. BMJ 2003;326:1171-3.
Dickersin K. How important is publication bias? A synthesis of available data. AIDS Educat Prevent 1997;9:15-21.
Rennie D. Thyroid storm. JAMA 1997;277:1238-43.
Andreopoulos S. The unhealthy alliance between academia and corporate America. West J Med 2001;175:225-6.
Cowley AJ. The effect of lorcainide on arrhythmias and survival in patients with acute myocardial infarction: an example of publication bias. Int J Cardiol 1993;40:161-6.
Yamey G. Scientists who do not publish trial results are "unethical." BMJ 1999;319:939.
Tramèr MR, Reynolds DJ, Moore RA, McQuay HJ. Impact of covert duplicate publication on meta-analysis: a case study. BMJ 1997;315:635-40.
Psaty BM, Heckbert SR, Koepsell TD, et al. The risk of myocardial infarction associated with antihypertensive drug therapies. JAMA 1995;274:620-5.
Furberg CD, Psaty BM, Meyer JV. Nifedipine: dose-related increase in mortality in patients with coronary heart disease. Circulation 1995;92:1326-31.
Flanagin A et al. Prevalence of articles with honorary authors and ghost authors in peer-reviewed medical journals. JAMA 1998;280:222-4.
Phillips SG, Carey LA. Comment in: Lancet 1999;354:1563.
Fazel S, Danesh J. Serious mental disorder in 23 000 prisoners: a systematic review of 62 surveys. Lancet 2002;359:545-50.
Villanueva P, Peiro S, Librero J, Pereiro I. Accuracy of pharmaceutical advertisements in medical journals. Lancet 2003;361:27-32.
Wilkes MS, Doblin BH, Shapiro MF. Pharmaceutical advertisements in leading medical journals: experts' assessments. Ann Intern Med 1992;116:912-9.
Committee on Publication Ethics. Guidelines on good publication practice. Available at: http://www.publicationethics.org.uk/cope1999/gpp/gpp.phtml#gpp (Accessed 20 April 2004.)
Rochon PA, Gurwitz JH, Cheung M, Hayes JA, Chalmers TC. Evaluating the quality of articles published in journal supplements compared with the quality of those published in the parent journal. JAMA 1994;272:108-13.
Cho MK, Bero LA. The quality of drug studies published in symposium proceedings. Ann Intern Med 1996;124:485-9.
Competing interests:
GY used to work for the BMJ Publishing Group. BMJ Publishing Group journals, such as the weekly BMJ, carry drug advertisements.
Read, reflect, respond module
Drug companies' influence
Comment:
good module from BMJlearning re: probity and drugs company, did not know about ghost writing and the "sneaky" ways in which data is published or suppressed to make the drug look better than is.
BMJ 2004;328 (12 June), doi:10.1136/bmj.328.7453.0-d
Does eradicating H pylori improve gastric symptoms?
Eradication of Helicobacter pylori does not increase or reduce symptoms of gastro-oesophageal reflux. Harvey and colleagues (p 1417) randomised 1558 patients with H pylori infection to receive active treatment or placebo and found that, two years after treatment, eradication had no effect on the overall prevalence of heartburn or reflux, nor on pre-existing gastric symptoms. In participants who had isolated reflux, treatment prevented the development of heartburn.
also see http://bmj.bmjjournals.com/cgi/content/full/328/7453/1388. editorial
Comment: only do h lylori in those with reflux, but usually patients present with GORD once have symptoms of heartburn so you may have missed the boat!
Does eradicating H pylori improve gastric symptoms?
Eradication of Helicobacter pylori does not increase or reduce symptoms of gastro-oesophageal reflux. Harvey and colleagues (p 1417) randomised 1558 patients with H pylori infection to receive active treatment or placebo and found that, two years after treatment, eradication had no effect on the overall prevalence of heartburn or reflux, nor on pre-existing gastric symptoms. In participants who had isolated reflux, treatment prevented the development of heartburn.
also see http://bmj.bmjjournals.com/cgi/content/full/328/7453/1388. editorial
Comment: only do h lylori in those with reflux, but usually patients present with GORD once have symptoms of heartburn so you may have missed the boat!
BMJ 2004;328 (12 June), doi:10.1136/bmj.328.7453.0
Use NSAIDs for renal colic
In patients with acute renal colic, non-steroidal anti-inflammatory drugs (NSAIDs) should be the drug treatment of choice. Reviewing 20 trials including 1613 patients with renal colic, Holdgate and Pollock (p 1401) found that patients taking non-steroidal anti-inflammatory drugs had slightly less pain and were less likely to need additional analgesia than those taking opioids. Those taking opioids were more likely to have vomiting or other adverse events.
COMMENT: already knew to use NSaiDs for renal colic but heartening to know that does not need more opiod analgesia
Use NSAIDs for renal colic
In patients with acute renal colic, non-steroidal anti-inflammatory drugs (NSAIDs) should be the drug treatment of choice. Reviewing 20 trials including 1613 patients with renal colic, Holdgate and Pollock (p 1401) found that patients taking non-steroidal anti-inflammatory drugs had slightly less pain and were less likely to need additional analgesia than those taking opioids. Those taking opioids were more likely to have vomiting or other adverse events.
COMMENT: already knew to use NSaiDs for renal colic but heartening to know that does not need more opiod analgesia
Friday, June 04, 2004
BMJ Career Focus 2004;328:226 (5 June)
Growing your CV
Every doctor, no matter what age and career stage, should continue to develop their curriculum vitae, as Elitham Turya explains
Growing your CV
Every doctor, no matter what age and career stage, should continue to develop their curriculum vitae, as Elitham Turya explains
BMJ Career Focus 2004;328:225 (5 June)
Writing a winning CV
Sam McErin outlines a successful formula
comment: always good to know what a good CV should be like
Writing a winning CV
Sam McErin outlines a successful formula
comment: always good to know what a good CV should be like
BMJ 2004;328:1372-1375 (5 June), doi:10.1136/bmj.328.7452.1372
Hands-on guide to questionnaire research
Administering, analysing, and reporting your questionnaire
Petra M Boynton, lecturer in health services research
Comment: good aticle for research using questionnaires
Hands-on guide to questionnaire research
Administering, analysing, and reporting your questionnaire
Petra M Boynton, lecturer in health services research
Comment: good aticle for research using questionnaires
BMJ 2004;328 (5 June), doi:10.1136/bmj.328.7452.0-e
Patients with chronic fatigue syndrome are not treated properly
General practitioners are ambivalent about chronic fatigue syndrome and may not be treating patients according to best evidence. Raine and colleagues (p 1354) interviewed 46 general practitioners in England and found that some hold negative stereotypes for chronic fatigue syndrome, but not for irritable bowel syndrome. This antipathy was due to the lack of bodily location, patients failing to conform to the work ethic and traditional "sick role," and disagreement with patients over causes and management. Referral for mental health interventions was unlikely because doctors were unfamiliar with them or thought them unavailable or unnecessary.
Comment: needs more training to try and overcome the stereotype thinking. Perhaps if there was clearer management strategy/clinic may change opinions
Patients with chronic fatigue syndrome are not treated properly
General practitioners are ambivalent about chronic fatigue syndrome and may not be treating patients according to best evidence. Raine and colleagues (p 1354) interviewed 46 general practitioners in England and found that some hold negative stereotypes for chronic fatigue syndrome, but not for irritable bowel syndrome. This antipathy was due to the lack of bodily location, patients failing to conform to the work ethic and traditional "sick role," and disagreement with patients over causes and management. Referral for mental health interventions was unlikely because doctors were unfamiliar with them or thought them unavailable or unnecessary.
Comment: needs more training to try and overcome the stereotype thinking. Perhaps if there was clearer management strategy/clinic may change opinions
