Does it work? When doctors need evidence

Healthcare professionals can generate important clinical questions for addiction research. Answering such questions by conducting a Cochrane review of evidence is a satisfying learning process and can contribute to drugs policy. This article summarises the experiences of an addiction medicine researcher conducting a Cochrane review, developing and evaluating a researcher-facilitated programme for medical student research activity in general practice.

photo credit: theconversation.net

One summer afternoon in 2010, an interview with a family physician in Dublin opened my eyes about talking therapies for drink problems among people who also used other drugs. “Does counselling work for these people?” the doctor asked.  “Yes”, I was absolutely convinced about it, but I had no evidence for my faith. Surprised by his interest, I sent him the only two studies on the topic that I knew of; never heard back from him.

I searched for more studies without success. Many studies on general population showed up in my internet search, but none for people who also used other drugs.
This made me doubt my beliefs. At that time, a national funding agency announced a call for Cochrane training fellowships. Cochrane collaboration hosts the largest database of systematic reviews to inform healthcare decisions. Cochrane reviews are the jaguars of medical evidence synthesis. The fellowship was a godsend. I could use the funding to learn from Cochrane gurus and answer the Dublin doctor’s question by making the most of all available literature. My supervisor introduced me to a Cochrane author, Dr Liam Glynn, who reviewed self-management strategies for high blood pressure. He agreed to mentor my fellowship. We booked the title for our review with the Cochrane Drug andAlcohol Review Group in Italy and started to work on it when we got the funding.
The review found very few studies, most of which didn’t have a control group or randomised patients without drink problems; we could not give any recommendations to doctors.
The next step in the quest for the answer, we approached patients with dual drug and alcohol problems and fed their ideas back to the experts. Expert consensus recommendations are standard in the absence of scientific studies. The group had to rely on semi-structured interviews with doctors and patients and “B class” evidence from my review. The result of their consensus was a manual for family doctors.
Having developed the manual, we tested its value to answer our original question: “Does it work?” The new pilot trial encourages doctors to ask people who use illicit drugs about alcohol and to help those with mild problems; severe problems are best treated by a specialist.  Sixteen general practices (GPs) in two deprived regions will be randomised to receive the manual-based training or to keep doing what they do. The latter group will be trained later.
When I finished my Cochrane training and review, it was time for me to give back and teach medical students because the fellowships worked on the pay-it-forward model. Equipping the new generation of doctors with critical literature review and appraisal skills was my contribution to the improvement of addiction healthcare delivery. The aim of our teaching project was to create and evaluate a training-through-research programme for medical students, facilitated by a seasoned researcher.
We offered online webinars, methodological advice, mentoring, and one-one interaction. Our medical school emailed all students and we randomly selected a handful needed for our research projects. Collaborators from biostatistics, psychiatry and public health aided the programme. The students presented their work at four conferences and wrote three academic papers for medical journals.
Teaching literature reviews to medical students was a rewarding learning experience. I learned that the quality and commitment of students varies; different expectations led to different work processes and outputs. Some students submitted their work in more finished stage than others; competing priorities precluded achievement of higher standards. The manuscript preparation, submission and publication processes were too long for short student projects, although some students persevered and remained involved until the end.
From a personal perspective, I still don’t know whether counselling works for drink problems in people who also use other drugs, but I’ve learned how to query the literature when doctors need evidence.

This post is based on our presentation at the INMED conference in Belfast, and o recent article in the Substance Abuse journal. References:
  • Klimas, J., & Cullen, W. (2014). Addressing a Training Gap through Addiction Research Education for Medical Students: Letter to editor. Substance Abuse. doi: 10.1080/08897077.2014.939802
  • Klimas, J., & Cullen, W. (2014). Teaching literature reviews: researcher-facilitated programme to support medical student research activity in general practice. Poster presented at the Annual scientific meeting of the Irish Network of Medical Educators, February 21, Belfast, NI.