Big data, the doctor and musculoskeletal medicine

Big data is big business. You type a few words into a search engine and an algorithm determines which hints and advertisements you will see based on the data the database already has. That will also happen in the healthcare sector. Or no: that has been happening already for a long time. There are also computers that make diagnoses based on big data. By entering endless data from patients in a database and comparing them with what an individual patient enters, it is calculated what is going on and what the steps to take are. Are the days of the doctor counted?

To diagnose with a conversation

Research has shown that 70-80% of the diagnoses can be made by an interview with the general practitioner. This number is lower in the hospital (emergency department: 20%). A diagnosis can be missed due to the complexity of the problem, additional problems that are distracting from the actual problem, bias of the doctor or sometimes, unfortunately, an incorrect conversation. Big data do not suffer from this and will definitely score better. Perhaps the biggest importance of big data is in the stage prior to illness: predicting disease on the basis of DNA profile, history, lifestyle and more and advising how to deal with it. That too is a reality and literally provides valuable information to health insurers, for example.

To diagnose with physical examination

Physical examination adds a few percentages to the general practitioner about what can be achieved with the interview. This is probably more favorable in the hospital, although I do not have any numbers ready (emergency department + 20%). Big data can come far with limited resources: measuring blood pressure, pulse frequency, photo of the skin, cardiogram, blood sugar level and with the advancement of technology more and more. I imagine that the person in question is putting on a kind of harness with various sensors mounted on it. It is already commonplace with smartwatches.

To diagnose with musculoskeletal medicine

What big data can not do, and probably also is not possible in the short term, is to carry out a proper investigation of the musculoskeletal system (back, shoulder). A brief explanation. Big data is based on statistics: summarizing a lot of data in a common denominator. An individual does not necessarily meet the common denominator. The characteristics of the musculoskeletal system such as degree of range of movement are very individual: what is different for one person is normal for another person. This also applies, to a lesser extent, to blood tests and other health parameters. So the interpretation of the data in physical examination is difficult for a computer. This problem can be overcome by impacting and measuring the individual from birth (or earlier). The interpretation by the computer also depends, of course, on what is being introduced. How good are the step-by-step plans or algorithms that we have made people and carry the computer with?

Treatment with big data

As long as the diagnosis is correct, big data can mean a lot: information (your complaint will resolve by itself), advice (use a day paracetamol), instruction (note posture change), exercise (with video as an example), referring (the letter is already written) and perhaps also an injection (if you trust it enough). Part of the course of musculoskeletal medicine is also: motivating, mobilizing in various variations, orthomanipulation, manipulation. I do not see a computer doing that easily. Human contact is also necessary to deal with calamities or deviations from the rule.

Conclusion

A major blow to efficiency in health care will be made by using big data as long as commercial motives are in the background. Quality depends strongly on (the quality of) data from scientific research and how that is interpreted before it is introduced into the computer. Big data, especially in musculoskeletal medicine, will not be able to cope with anything and a consultation of, for example, a musculoskeletal doctor is required for a complete trajectory when it comes to examination and treatment of the musculoskeletal system.

References

Hampton JR et al. (1975). Relative contributions of history-taking, physical examination and laboratory investigation to diagnosis and management of medical outpatients. BMJ 31:486-489.

Paley L et al. Utility of clinical examination in the diagnosis of emergency department patients admitted to the department of medicine of an academic hospital. Arch Intern Med. 2011 Aug 8;171(15):1394-6.

Mayer-Schonberger V, Cukier K. De big data-revolutie: hoe de data-explosie al onze vragen gaat beantwoorden. 2013 Maven Publishing.