Machine learning is a potential gamechanger in the healthcare field. Despite the optimistic projections both IT and healthcare professionals have made, many healthcare organizations are still reluctant to makes significant investments in artificial intelligence initiatives. They tend to remain skeptical of some of this technology, despite the growing body of evidence supporting its usefulness.
Healthcare providers are more likely to recognize the potential of machine learning if they are aware of specific benefits that medical software yields, especially when leveraging machine learning technology. One of the most compelling reasons healthcare providers should consider the merits of artificial intelligence is that it can make a big difference in solving the problem of elevated medical error rates.
Professor Thomas Davenport of Babson College and Deloitte managing director Ravi Kalakota addressed some of the benefits of machine learning in healthcare in their paper The potential for artificial intelligence in healthcare. They stated that one of the biggest opportunities is in the field of precision healthcare.
In healthcare, the most common application of traditional machine learning is precision medicine “ predicting what treatment protocols are likely to succeed on a patient based on various patient attributes and the treatment context.2 The great majority of machine learning and precision medicine applications require a training dataset for which the outcome variable (eg onset of disease) is known; this is called supervised learning, the authors write.
However, there are other benefits as well. They can collectively help solve the medical error concerns raised by many healthcare providers.
Machine learning could be necessary to solve the crisis of high medical error rates
Medical error rates are a serious concern in the United States. Data shows that there are around three errors for every 1000 treatments. When weighing both the financial and social costs, this can be very concerning.
The cost of medical errors is around $20 billion every year. Approximately 100,000 people die as a result of medical errors each year, which makes them the third leading cause of death in the United States.
Healthcare organizations have been under growing pressure to tackle the problem with medical error rates. The Patient Protection and Affordable Care Act introduced a number of initiatives to encourage hospitals and clinics to lower medical error rates. However, the effectiveness of these reforms has been called into question, because they work by using other local providers as benchmarks. If the majority of healthcare providers were not making a strong effort to reduce their respective error rates, then peer organizations might have a little incentive to take similar initiatives.
Part of the reason that many healthcare providers have been reluctant to tackle growing concerns about error rates is that the problems might seem too complex to tackle. This is why machine learning solutions might appear to be worthwhile investments for healthcare organizations.
How can healthcare providers use machine learning to address medical error rate concerns?
Machine learning offers a number of promising possibilities for healthcare providers trying to address medical error rate issues. Some of these opportunities are highlighted below.
Providing better healthcare documentation through smart records
Most of the discussion about the benefits of machine learning in healthcare center around improving and streamlining front line services. However, machine learning can be at least as beneficial in the realm of medical record-keeping.
Quotient Health and other technology firms are using machine learning to improve medical records. Smart records can be stored more cost effectively. However, the real long-term cost benefits of using smart records are due to the potential reduction in medical errors.
Healthcare providers are less likely to make serious mistakes, such as prescribing medications that patients are allergic to if they have accurate medical records. Smart records are a potential solution.
Improving medical diagnostics
One significant reason that medical errors arise is that healthcare providers don’t properly diagnose disorders. A Cambridge startup has made significant strides in improving medical diagnostics through machine learning. They raised over $15 million in 2017 to bring their technology to market.
Improving the vetting process for healthcare providers
Healthcare organizations have already started to use machine learning to take a better look at the creditworthiness of potential customers. The same practices can be used to do more thorough background checks on potential employees. They will be able to identify physicians and other healthcare practitioners that could be at risk of causing medical errors. Although the ethics of this might be called into question and the practices will need to be tweaked to avoid discrimination accusations, the potential benefits are still significant.
