Opioid misuse has risen to the level of national emergency. The facts are alarming. More people now die of drug abuse than car accidents in the US. Painkiller prescriptions by doctors are now sufficient for each American adult to have a bottle of pills. And, a US citizen dies about every 10 minutes from an opioid overdose. Perhaps the only thing more stunning than the facts behind this issue is how long it has taken to become considered a crisis.

Like many complex public health issues, the opioid crisis has many causes and owners. This makes it more challenging to solve as there isn’t just one “villain” responsible. There are issues of personal responsibility of the opioid user himself/herself, lack of familial support systems, and lack of education. There are legal causes – a lack of an effective regulatory environment and a continued failure to stem a supply of illegal and prescribed narcotics from getting into the wrong hands. The health care community has focused on pain management, been highly influenced by pharma companies who manufacture opioids, and doctors have willingly over-prescribed opioids (in a few years it is projected that there will be on average one opioid prescription for every person living in the US). And, “aggressive” would be too mild a term to describe how pharma companies have marketed these drugs to doctors and consumers.

I am sure that if you talk about the opioid crisis with others and ask what has caused it, “market research” would not be something that is apt to come up. But, our industry has played an important role in creating an environment in health care that is conducive to over-prescription of opioids. Let me explain.

In the late 1980’s a business trend called Total Quality Management became established in US businesses. It was largely a reaction to a perceived threat to US manufacturing from Japan, and it focused highly on statistical measurement. Put simply, TQM assumed that you can’t improve something if you can’t measure it. TQM first gained hold in US manufacturing (where it is still commonly employed). But, it wasn’t long until it spread like wildfire throughout all types of US businesses, including those in the service sector.

This was a boon for market research. The firm I first joined in market research was a custom research company and customer satisfaction measurement was a primary expertise. It was perfect timing – we were in a great position to conduct surveys supporting TQM efforts taking place in a wide variety of industries.

One of these was health care. The 80’s and 90’s were transitionary times in our health care system, as HMO’s and insurance companies became much more powerful and began to “manage” health care. They more tightly controlled which procedures would be reimbursed, and, for better or worse, exerted much greater oversight over the care that doctor’s provided to patients.

This was happening at the same time TQM became all the rage in business. So, insurance companies, hospital administrations, and regulators all began to insist on TQM measures in health care. One of the most important of these was the patient satisfaction survey.

Research companies responded. A few major players emerged, and the company I worked for became a mid-sized supplier in this area. We had an excellent approach and established a small team to work on it, which I eventually managed for a couple of years in the 90s. Patient satisfaction surveys blossomed, and are still in widespread use. I’d guess that if you have ever been to a hospital for a procedure, you’ve been asked to fill out a questionnaire shortly after your visit.

Health care providers, namely doctors, hated these surveys. With umpteen years of medical training behind them, why were they now being evaluated by what their patients thought of the quality of the care they provided? They resisted these surveys and still do.

At the time, we chalked this resistance up to the fact that no professional wants to be evaluated this way. I know I resisted when we started asking clients what they thought of our work and when the results were incorporated into my performance reviews.

We saw this doctor resistance as a misunderstanding of what we were measuring and also a misuse of our data by hospital administrators. In any service delivery, there are two contributors to outcomes: 1) the quality of the service being delivered and 2) the manner in which it is delivered. In many contexts, including health care, consumers should not be placed in a position to evaluate the former. Health care quality assessment is the province of experts, and peers and medically-trained supervisors are probably in the best position to evaluate it.

However, the latter (how the service is delivered) is best evaluated by patients. Our studies constantly showed that doctors understated the importance of these “bedside manner” measures. In our models, these softer issues dominated a patient’s willingness to comply with the physician’s instructions. And, what use is the doctor’s expertise if the patient doesn’t do what he/she says? (As an aside, the nursing profession loved these surveys, as the analyses often showed that the nurse was more important than the doctor in garnering patient compliance).

We were strong advocates for these surveys and how they empowered patients to become an important part of their own health care. Survey results would cause doctors to become better communicators with patients.

With 25 years of hindsight, I can now say that these surveys had a painful unintended consequence: they contributed to the opioid epidemic in this country.

Just as these surveys were taking hold in health care, two important things occurred:  1) the pharma industry developed new forms of opioid painkillers and marketed them aggressively to doctors, and 2) the medical profession adopted pain as a “fifth vital sign” (along with blood pressure, heart rate, respiratory rate, and temperature.) Prior, pain had been viewed as a symptom, and not an objectively measured sign.

This was a watershed moment for the opioid issue. Now, health care providers were constantly asking patients about their pain level (on a 0 to 10 scale or a smiley-face scale). Doctors focused on pain, and there was a widely held perception that pain was being undertreated in patients. This happened precisely at a time when new forms of opioids were available for prescription.

So, what does the patient satisfaction survey have to do with all this? Well, in most of the analyses we would do, a patient’s satisfaction with a doctor visit or procedure was always highly correlated with one item:  “did you feel better as a result of the appointment?” Reduction of pain and patient satisfaction were largely the same thing. We’d move past this in the analysis and tell clients that beyond pain relief, there were bedside manner measures that did matter and that were in their control to change. This is analogous to studies researchers do on consumer products. Price usually overwhelms our models. So, we mention that, and then give guidance about other things beyond price that you can work on.

Hospitals and doctors were being held accountable to the results of these surveys. They quickly learned that pain relief was paramount to how they would be evaluated. This put pressure on them to prescribe opioids for mild ailments. Prior to this time, opioids were mainly used short-term for acute cases and for patients recovering from surgery. Now, they were being prescribed for every-day ailments – toothaches, back pain, broken bones, etc.  At this time pharma companies were downplaying the potential of addiction to these drugs, marketing them heavily, and as a result prescriptions soared.

Of course, I wouldn’t want to overstate the contribution of the market research community to this national crisis. This is however a good example of the importance and unintended consequences of our work. There was a perfect storm of things brewing – the TQM fad, increased power of insurance companies, the development and marketing of new drugs, and a focus on pain management in health care.

I also don’t advocate that we end the patient survey. It provides important feedback to health care providers and I strongly feel that it has resulted in better communications between providers and patients. But, results shouldn’t be used so prominently in the evaluations of hospitals and their staff and the content of these surveys should shift from pain measures. It will take a lot of effort in a lot of directions to resolve the opioid issue, and our industry has an important role to play.