“I knew his personal history through his group-therapy sessions. I think part of what motivated him was his deep shame at what went on in the past. I’m not sure how much of his struggle you were party to.”

“In all honesty, we never met. I found out we were related through a complicated set of circumstances, but he was gone by then.”

“He was a lost soul and I wish we’d done better by him.”

“When did he enroll in the study?”

“Might have been March. He was hospitalized for acute alcohol poisoning. The social worker referred him after he’d been through detox and rehab. This was meant as long-term support.”

“So you felt he would benefit?”

“That was the hope. Do you know anything about the study?”

I shook my head, not wanting to interrupt the flow. He was already more forthcoming than I’d had reason to hope.

“We’re looking at the use of three drugs in combination, one of which, Glucotace, we’re especially interested in. When Terrence came into the program, we explained we were running a random double-blind study and couldn’t guarantee which group he’d be assigned to. One group is given the drug. The control group receives a placebo.”

“That’s actually one of the questions I had for you,” I said. “He seemed to think the medication made him sick.”

Dr. Reed said, “So I was told. He was convinced he was on Glucotace. This was week one when we were barely under way. Fifty-fifty chance of it. Subjects aren’t told what medication they’re taking. Those of us conducting the study also operate in the dark. That’s what the term ‘double-blind’ refers to.”

“Really. You didn’t know what he was taking?”

“I do now. I’m not sure how familiar you are with research strategies. Since I designed the study and applied for the grant, I have a rooting interest in the outcome, as you might imagine. If we both knew he was on Glucotace, it might influence the questions I asked and the answers he gave. Even if I knew what he was taking and he didn’t, the outcome could be affected. We’re all suggestible in one way or another. A patient taking a placebo might actually get better because that’s the nature of the beast. What we believe affects our physiology. If we’re anxious, our heart rate goes up. If we feel safe, our respiration slows.”

“I’ve experienced that myself,” I said.

“In a clinical trial, our job is to render an unbiased account of test results. Some of this is based on blood work and other screening procedures, but some of what we track are the subjective reports from the patients themselves.”

“If he complained, would you have looked into it?”

“Of course. Absolutely. As you may know, Terrence did complain and we undertook a review. He was seen by three other clinic doctors, and all of us kept a close eye on his lab values and conducted regular physical examinations. If he’d suffered serious side effects, the symptoms would have surfaced.”

“So there weren’t any?”

“I wish I could reassure you, but he had a problem showing up for his appointments. Completely hit or miss.”

“Because of the medication?”

“This was a personality issue. We’re meticulous about outpatient monitoring, but we’re dependent on compliance. Without that, the numbers mean nothing.”

“Was he terminated or did he leave of his own accord?”

“We were forced to sever the relationship. We had no choice. I felt bad about it. He was a good man and we gave him every chance to straighten up. He couldn’t seem to manage it.”

“He had a friend named Charles Farmer.”

“I remember Charles. Same thing with him. He was another of those hard-core cases. Charles showed up for one of his exams so high, he could hardly stand. I have no idea what he was on, but I terminated him that day. Without reliable feedback, we might as well abandon ship. We can’t have patients taking meds we know nothing about. Even something as innocent as vitamins or nutritional supplements can skew results.”

“Were you alarmed that both men died within months of their enrollment?”

“Of course. That’s why I contacted the coroner’s office. I was concerned there was an underlying disease process hampering his recovery.”

“And was there?”

“Nothing any of us could pinpoint, including Dr. Palchek’s postmortem. Subjects aren’t always honest about their medical histories. A participant with unsuspected health issues could be at increased risk of an adverse event. We screen as rigorously as we can, but ultimately we can’t pick up every warning sign, especially if a patient is hiding something.”