(Flickr user Images Money)

“HIV drug goes from $13.50 a tablet to $750 overnight”

“Cancer drugs doubled to $10,000 per month”

“Senate blames greed for $1,000 Hepatitis C drug”

Recent headlines like these have stirred debate over rising drug prices. But none of them tell us about a drug's real value. Does the drug cure a deadly disease or just add a few weeks to a patient’s life? Does it work for all patients or just for some? These are important – but frequently overlooked -- questions when considering the value of a new drug and whether its price is justified.

As a physician, I want scientific innovations to continue bringing new game-changing treatments to my patients. The arrival of Gleevec several years ago drastically improved the outlook for patients with chronic myelogenous leukemia. CML changed from a rapidly fatal disease to a chronic disease similar to diabetes. Hepatitis C can now be cured with drugs such as Sovaldi, avoiding subsequent development of liver cancer and the need for liver surgery and transplantation. These drugs are truly miraculous, and as a society, we should be prepared to pay high prices for such innovations. High financial reward incentivizes scientists and the drug industry to search for major breakthroughs.

But what about new drugs that are not nearly as effective? Numerous new medicines have been approved in recent years to manage stage 4 solid cancers, such as colon, lung, pancreas and kidney. With a few exceptions, these drugs prolong patients’ lives by only a few months. However, almost all of them cost more than $10,000 per month. This leaves insurance companies saddled with large bills for drugs providing only a minimal benefit, a cost ultimately passed on to average Americans in the form of high premiums.

These drugs are significantly overpriced. If we want to pay for value, prices must be linked to benefit. My team of researchers recently published a framework in JAMA Oncology demonstrating how a drug’s price should be established based on its benefit. We used the example of Portrazza, a drug recently approved for stage 4 squamous lung cancer. It extends life by six weeks for people that will eventually succumb to their disease. Using a sophisticated economic model, we demonstrated that it should not cost more than $1,700 per month.

Some people suggest that the pharmaceutical industry is able to self-regulate in order to provide appropriate prices. Our study provided a litmus test to see if this was true. We made the value of Portrazza available to the manufacturer, Eli Lilly, prior to FDA approval. However, the company subsequently announced the price at $11,430 per month. This is more than six times the upper limit of what we estimated was a value-based price.

Our model considers a year of life to be worth between $50,000 and $200,000, with an adjustment to incorporate quality of life. It uses that value to calculate a price range for a drug, given how much time it adds to patients' lives. It is no easy task to estimate the monetary value of life. But in a world where payers have finite resources, we cannot avoid this question. The range that we used was a very generous range, given that in the UK, in order for a drug to gain approval, it must cost less than £30,000 ($43,100) per quality-adjusted year.

Some say drugs should be priced highly to compensate pharmaceutical companies for their investment in developing medications that never reach the marketplace. I disagree. Prices should be based on benefit alone. Investors, mathematicians, and researchers can analyze the efficacy of a drug prior to making the large investment needed for a phase 3 trial. If a drug is expected to be minimally beneficial, and thus warrants a low price, companies could choose not to make the substantial investment needed for the phase 3 trial.

Prominent lung cancer experts are not excited by the level of efficacy provided by Portrazza. There are already suggestions they may not recommend the drug in major clinical guidelines followed by doctors around the world, such as those developed by the National Comprehensive Cancer Network. Perhaps a more appropriate price may increase enthusiasm.

Presidential candidates Hillary Clinton and Sens. Bernie Sanders and Marco Rubio have publicly recognized the drug-pricing problem. This is not a partisan issue – getting value for money on health care matters to us all. But we can’t just force prices lower, as that will kill future investment in innovative drugs. Value-based pricing is the better solution, providing a balance between incentivizing innovation and controlling costs.

More from Wonkblog:

How an $84,000 drug got its price: "Let's hold our position ... whatever the headlines"

Why treating acne and rashes is getting so expensive

Why drug spending is so high, in three charts