Painkillers and hearing loss in women?

@HealthMed Researchers from Boston in the USA have reported that pain-killer use increases risk of hearing loss in women. Sounds worrying but worth looking at what the study involved and asking the obvious questions: are the findings credible and can they be generalised beyond the population in the report?

This was a prospective observational study of around 62,000 women in the US Nurses Study, the overwhelming majority of whom were of Caucasian origin. Information was collected on use of three types of painkiller - ibuprofen, acetaminophen (paracetamol in the UK), and aspirin. Nurses were following for 22 years on average.

What did they find? The rate of reported hearing loss from all causes was high at one in six over the entire population in the study. For those using painkillers, there was no link with hearing change with occasional use (once or less per week) or with aspirin whatever the reported frequency if use. For ibuprofen or acetaminophen (paracetamol), regular use for 2 days a week or more was linked to increased risk of hearing loss. The highest relative reported increase in risk was 1in 4 (24%) excess hearing loss linked to use of ibuprofen on at least 6 days a week. The authors state they are unaware of the specific mechanisms linking hearing loss to these medicines in their study.

What potential weaknesses are there in the study?
These include:
-   Noise exposure is an important cause of hearing loss; and of headaches, an obvious reason to be taking painkillers. There was no information available on noise exposure as a confounding cause of hearing loss. The authors acknowledge that is an important flaw in their study.
- Hearing loss was self-reported and severity of hearing loss not measured.
- No information was given on the dose of pain-killers used.
- Taking painkillers regularly from 2 days a week to every day suggests underlying medical problems which may themselves have had an impact on hearing. Co-morbidity accounting for the pain-killer use was not clear.
- Oddly there was no consistent increase in risk of reported hearing loss with increasing numbers of days a week of painkiller use: similar hearing loss reported with paracetamol 6 or more days a week, as for 2 days a week but more hearing loss when taken taken 4 or 5 days a week
- In this study aspirin was not associated with hearing loss but would be expected to be based on previous studies, including studies in men. This raises questions about the accuracy of data collection.

Are the findings generalisable to people outside the study group?
- These were mainly Caucasian origin women who were nurses working in the USA. It is unclear whether similar findings would have occurred in other occupational groups, ethnic groups or other countries. However there are biochemical mechanisms by which all these drugs, including aspirin, could affect the hearing system. Therefore it would be premature to suggest that aspirin would be a safe alternative painkiller with regard to hearing.

Further research is needed to explore whether these findings are due to painkillers or co-incidental, and if confirmed, to find out who is at particular risk. For the present, any risk appears relatively low and linked in this study only to women taking selected painkillers (ibuprofen or paracetamol) for several days each week over long periods. Any long-term use of pain-killers should be discussed with a medical advisor or pharmacist. People who have unexplained hearing loss and are currently on these medicines at least 2 days a week for long periods should seek advice from their medical advisor, pharmacist or hearing consultant.

Anti-depressants, benzodiazepines, newer sleeping medicines and risk of motor accidents?

@HealthMed: The following comments were in response to a new paper linking use of anti-depressants, benzodiazepines and newer sleeping medicines to increased risk of motor vehicle accidents.

These comments were part of the briefing on this paper from the Science Media Centre.

In a paper published in the British Journal of Clinical Pharmacology, Tsai and colleagues from Taiwan have reported a ~40-70% relative increase in risk of non-fatal driving accidents in Taiwan linked to prescription of a wide range of anti-depressants, benzodiazepines and newer Z-drug sleeping medicines. In their study, around 5200 people who had had motor accidents were compared with around 31,000 people with no history of accidents. The medicines appeared possibly linked to from 1 in 20, to 1 in 70 accidents, depending on the drug type. Obvious questions for pharmacologists, health professionals and the public are whether these findings are credible, and if so, generalisable to other parts of the world.

This Taiwan report was based on a case-control study – more open to bias than randomized trials. Although the researchers matched controls for age, gender and year of the accidents, they were unable to rule out differences in patterns of driving or in alcohol intake between the groups. Alcohol amplifies any impairment in concentration resulting from effects of drugs that act on the brain; and ethnic Taiwanese are commonly are intolerant of alcohol, due to genetic lack of the enzyme needed to break down alcohol.

It is also possible that the increased accident risk resulted from effects of underlying psychological or psychiatric disorders for which the drugs were prescribed e.g. through poorer concentration or indirectly through additional effects of sleep disturbance caused by the conditions under treatment, independent of the drugs. In addition the researchers noted that those who had accidents also had a higher burden of a range of non-psychiatric diseases, also recognized as increasing risk of accidents.

What do these results mean for populations in other countries? Several of the drugs implicated in Taiwan are not commonly used in other countries and there are several sources of bias indicating need for caution in interpreting the study. However other studies have stressed the need for caution when driving when using these types drugs. When these drugs are prescribed, patients should seek advice about risks of driving from their physician or pharmacist. In any event, patients on these drugs should avoid any alcohol when driving.

More information: Chia-Ming Chang and colleagues. Psychotropic Drugs and Risk of Motor Vehicle Accidents: a Population-based Case-Control Study; British Journal of Clinical Pharmacology; DOI: 10.1111/j.1365-2125.2012.04410.x

Prevalence of alcohol and other drugs in fatally injured drivers. Joanne E. Brady, Guohua Li.  Addiction. Published online August 20, 2012

Teenagers, IQ risk and cannabis: cause and effect, bias or other explanations?

@HealthMed My first contact with study of the science of cannabis was as a medical student in Aberdeen undertaking a summer project in pharmacology: a new researcher, Roger Pertwee, was interested in the effects of bioactive ingredients of cannabis on brain regulation of temperature and muscle function.  Research since then by Pertwee and many others has identified the importance of endogenous chemicals, endocannabinoids, that are part of normal functioning of the brain and other parts of the body, for example in modulating mood and stimulating appetite. Abnormalities in endocannabinoid pathways have also been implicated in a wide range of medical conditions, including inflammation, obesity, cancer, cardiovascular disease as well as mood disorders. 

Now this week we have a report illustrating a new facet of potential harm from cannabis. In a long-term follow up study by scientists in Otago, at Duke University, USA and King's College London, lead by Madeline Meier (Duke), IQ and reported cannabis use was monitored over the 25 years from age 13 to 38 in the Dunedin Birth Cohort. Of particular interest, their study of over 1000 subjects included teenagers in whom IQ was checked before any cannabis exposure. Around 1 in 7 reported being regular cannabis users, 1 in 20 doing so at least weekly before the age of 18.

What did the investigators find? With no cannabis history, there was a small fall in IQ. However recurrent cannabis use was associated with an 8 point decline in IQ, comparable to that seen in early dementia. Importantly, this decrease in IQ was particularly marked when cannabis use began during teenage years. A further concern was that stopping cannabis use did not lead to recovery of the IQ loss. Commentary on the results has ranged from concluding that cannabis is harmful in teenagers but safe in adults, to more cautious notes that adolescent brains appeared more vulnerable to cannabis, without providing carte blanche for longer term safety of cannabis use in adults …

Are these fair interpretations? For further discussion of this long-term, prospective observational study, important caveats in its interpretation, and its potential implications, see my discussion posted on The Independent blogs site.