Answers descend, questions ascend

Which are more valuable, questions or answers? Philosophers seem to prefer questions. Questions—Why are we here? Is there a God? What is love?—last for ever. Answers are usually provisional and often wrong. Doctors, in contrast to philosophers, seem to prefer answers. Patients come with questions. Doctors answer them. Maybe that's the origin of the bias that has resulted in medical journals being full of answers. The time has come, I'm convinced, to pay much more attention to questions.

Think of a journal as an information system. Such systems are designed to answer questions, but where are the questions? We have no systematic way of collecting questions from our readers. We should collect questions, preferably in real time when doctors are caring for patients, sort them, and commission answers that will no doubt give rise to more questions. At the moment we have a machine for answering questions but no afferent loop to supply questions. Soon we will have.

John Ely and others prompt these thoughts on questions with their attempt to produce a taxonomy of questions that arise as doctors care for patients (p 429). They see four uses for their taxonomy: to help authors produce clinically relevant material by answering real questions; to produce better tools for answering questions; to identify areas of knowledge where answers are lacking; and to set priorities for clinical research. Douglas Carnall in his Website of the Week reflects on current attempts to build a machine that will answer the questions of patients and doctors (p 458). There isn't much within medicine, but AskJeeves (www.ask.com) allows you to ask questions in natural language.

The British government has built a machine—NHS Direct—that will answer patients' questions and help them decide whether they need to see a doctor. It mostly seems to do well, but Kate Farrer and others show how it got one answer badly wrong (p 446). A mother rang about a 6 month old boy who was unwell with vomiting and increasing lethargy. Because the boy was drinking well and frequently wetting his nappies she was advised three times to keep him at home. Unfortunately he was eventually admitted with severe diabetic ketoacidosis. One of the most disturbing parts of the story is that the doctors could not find a way to give feedback to NHS Direct. It may be a machine that cannot learn.

Perhaps the biggest question for health policy people in Britain is whether the NHS can survive. Allyson Pollock observes that the creation of large numbers of intermediate care beds, as proposed in the new national plan, may lead to another large chunk of the NHS falling off into the private sector (p 393). It may follow long term care and dentistry.