Questions
- This past week has been about questions, or how to ask them.
- Having a background as a clinician, it has usually been my patients who have sparked my curiosity to learn some more, find out about a strange disease or a mysterious symptom. Through conversation, observation, much reading, chats with knowledgeable colleagues questions and hypotheses come up quite easily: it is part of clinical practice: an endless detective game, evidence-based but also intuitive because after all, for as many sprains and broken bones one can see, not one is the same - or should I say, not one is experiences in the same way.
- Even in the clinic room, my most precious skill is asking questions, and this is something that any patient-facing clinician would probably say. A story is told, I take the cues, probe, validate, diagnosis. I ask, and there and then I get an answer, from the right person at - hopefully - the right time. Perhaps my most cherished compliment is when a patient, after I finished asking my questions regarding the origin of her neck pain and giving her my 2p of advice goes ‘You’re like Poirot!’. Yessss.
- Anyhow, this is not the case with longitudinal data.
- The UK has a wealth of data from longitudinal studies - studies that follow the health of certain individuals (selected often to represent the general population) for a long period of time, sometimes their whole life from the day they were born. At given intervals, data is collected - perhaps questionnaires, health measures, or information about their social circumstances.
- These studies provide invaluable information to understand how health evolves along the life span - they are the birds-eye-view on the whole of life that clinicians dream of but that time-restraints in the clinic room never allow, nor do the filtering effect of memory and of narrative.
- I tried asking the data my question, but it didn’t answer. Horror.
- My question concerned some aspects of menopause that the study in question had addressed but only to a point, and although thousands of people had taken part I couldn’t ask my follow-up question, so the data remained silent.
- ‘Know thy dataset before asking questions’ was the moral of the week, because there are limits to what the data can tell you.
- This is not a great realisation for a data analyst but for a clinician it really is! It is a different way of asking questions, that relies on knowing on what can be answered. You are not dealing with humans anymore, but with human answers to a question that someone else has asked.
- Another interesting thing for me will be to see how, in studies that have been ongoing for decades, the data that is collected reflects the scientific priorities and beliefs of the time. In a sense, life course studies force researchers to tackle hands on scientific legacies of the past. I’d never had to deal with this directly as a clinician, because the ethos in clinical practice is to follow the most up-to-date evidence. But I feel I will enjoy this different kind of work.
- Thank you for making it to the end! I’m switching off for the week and hope you’re all having a lovely cooped up Saturday. Stay warm and safe!
Don't miss what's next. Subscribe to Modern Nature: