Human Subject: An Investigational Memoir

Previous chapter | Contents | Next chapter | References | Contact


17. The Sins of the Mother

Il est bien certain que pour les questions d'application immédiate à la pratique médicale, les expériences faites sur l'homme sont toujours les plus concluantes.

In order to get Emily’s PsychoPharm records, I needed for her to sign a release form. She was happy to have the records given or sent directly to me, which indicated either apathy or a great deal of trust. Or maybe she knew that the PsychoPharm staff would bungle the request.

A week after I turned in the form, I was sitting in one of the officettes, waiting for my next interrogator, when a study coordinator walked in. She smiled and handed me a stack of photocopies that looked suspiciously familiar. It was my own medical record, with the request for Emily’s records attached.

“This isn’t what I asked for,” I said.

“Oh?” She looked at the release form.

“No, I was asking for my daughter’s records. See? That’s her name and date of birth.” I pointed to the top of the form. When the study coordinator still seemed puzzled, I explained, “She was here about 8 or 9 years ago.”

“Ohhhh.” She looked a little less puzzled. “We thought it was kind of strange . . . ”

It would indeed have been strange for me to fill out a release form to have my own records released to me in person, especially since I’d already asked for and obtained most of those records. And I had explained to the receptionist that the request pertained to a study that my daughter participated in many years ago. Anyway, the study coordinator said they would try again to process the request.

Several weeks later I had received nothing in response to my request, so I sent a follow-up email to PsychoPharm. I got no response, so a few weeks after that I emailed the chief study coordinator. She wrote back, apologizing for the delay and explaining that they’d had “some turnover at the front desk.” (That would explain a lot, I thought. You just can’t get good pastries these days.) She said it was likely that we would need to submit another signed release. I sent her Emily’s summer address—at a fishing lodge in Pelican, Alaska—so that they could mail her the form.

Six more weeks passed. Emily called, from the Pelican city dump (the only place in town with cell phone reception), and said she hadn’t gotten the form yet. So I sent another email, and a nice man from Regulatory Affairs/IT replied three days later, apologizing for the delay again and saying that he would mail the form to Emily right away. I was surprised that someone from his department would be performing this kind of clerical function, but he obviously knew what he was doing, because a week later the form arrived in Pelican, and two weeks after that the records arrived in my mailbox.

The first page in the packet was a medication history. This was followed by a progress note that stated, “Screen fail for study due to GAD.” In other words, Emily had failed to qualify for the depression study, because what she had was generalized anxiety disorder, not depression. The next item was a psychiatric evaluation that concluded by recommending Emily as “an excellent candidate for the Adolescent GAD Study.” The evaluation had been written by Dr. K, the same psychologist that I had seen occasionally during my PsychoPharm days, and it was co-signed by PsychoPharm’s director.

So Emily’s memory had been correct: She had been found unsuitable for the depression study, but was deemed an ideal candidate for the GAD study, which, she recalled, needed more subjects.

There was just one problem: The envelope contained no records pertaining to the GAD study. There was only a lengthy (60-plus page) diagnostic instrument called the “Kiddie-SADS-Present and Lifetime Version,” or K-SADS-PL, which consisted of a series of questions designed to assess the child’s current and past symptoms in 21 categories, including Depressive Disorders, Panic Disorder, Anorexia Nervosa, Conduct Disorder, Encopresis (fecal incontinence), and Alcohol Abuse. At the end of the questionnaire was a supplement on “Overanxious Disorders,” which Dr. K had completed. But that was it. There was no information about the study for which Emily had qualified.

I wrote back to the nice man who had sent us the release form, asking if perhaps a section of the file had been omitted. At first I hesitated to write, almost deciding to call him instead, because everyone knows that you don’t send sensitive information by email. But, as you’ll learn in chapter 21, I’ll do almost anything to avoid making a phone call. To justify my decision, I remembered cases of people listening in on, and even recording, other people’s cell phone conversations (that’s the only kind of phone I have). Surely radio waves wafting to and from cell towers offered no more security than the electrical impulses propelling my message to PsychoPharm.


Like most such diagnostic tools, the K-SADS-PL is copyrighted, but it’s one of those rare instruments that can be downloaded and freely used. The only requirements are that the user be a not-for-profit institution or be using it “in an IRB approved research protocol.” Presumably the GAD and depression studies had both been IRB approved, so that even though PsychoPharm was not a nonprofit, it still had the right to use the instrument. Since the patient copy didn’t include the instructions, it was helpful to find a copy to consult online.

The K-SADS-PL begins with an “unstructured” Introductory Interview involving the parent and child together. The ostensible purpose of this session is to obtain factual information about the child’s current complaint(s), health history, relationships, etc. The real goal, however, is to build rapport with both parent and child. The instructions say this section “should never be omitted” (emphasis theirs), so presumably even if you already have the necessary information, you go through the motions in the service of building rapport. Those psychologists—up to their old deceptive tricks again!

The instructions also say: “In the introductory interview and throughout the K-SADS, interviewers are encouraged to use language generated by the child and/or parent when querying about symptoms (e.g., ‘For how long did you fell bummed?’).” I think they mean “feel,” not “fell,” but in the 21st century kids are probably no more likely to feel it than to fell it.

After the introductory interview, the interviewer meets with the child and the parent separately to ask each of them the hundreds of questions (several per symptom) in the “screen interview.” For each symptom there are thus three ratings: Child, Parent, and Summary. Some of the questions have three possible answers: no information (0), no (1), or yes (2). But most offer four options: no information (0), symptom not present (1), “subthreshold levels of symptomatology” (2), and “threshold criteria” (3). The thresholds for each symptom are described in detail.

This was at a time when Emily rarely spoke to her parents, so I was astounded to see that for every question, without a single exception, we had been in complete agreement. For instance, to the questions about arguing with adults and disobeying rules, Dr. K had indicated that we both said that neither of these behaviors was present; that is, the number ‘1’ was circled for both of us. I found it very hard to believe that I would have given that answer, and I was even more skeptical about my alleged answer to the questions about stealing. How could I have known for sure whether Emily had stolen anything in the past year? She had a history of petty thievery as a child, but surely I would have said that I had no information about her recent behavior. And yet Dr. K had given the rating of 1—not present—for both of us.

Only in the section called “Overanxious/Generalized Anxiety Disorder” were Emily’s symptoms rated at threshold level (a rating of 3), and they were at that level for each of the four symptoms: Unrealistic Worry About Future, Somatic complaints, Marked Self-Consciousness, and Marked Feeling of Tension/Unable to Relax. The ratings indicated that I, in whom Emily never confided, was fully aware that she felt self-conscious most days of the week.

It seemed so unlikely that a teenager and her despised mother would agree completely on these psychological matters that I began to think the whole interview process was rigged. It looked as if, from the start, the plan had been to make sure that Emily qualified for the GAD study and not the depression study. This apparent underhandedness greatly diminished my respect for Dr. K, but it seemed in keeping with the impression I’d gotten of PsychoPharm: The staff will do whatever it takes to fit subjects into the slots that needed filling.

The Psychiatric Evaluation contained at least one error, stating that Emily had lived for two years in a state she’s never set foot in, but what bothered me more was the frequent use of the word “denied,” as in “Emily denied any other psychiatric difficulties” and “Emily denied any alcohol or other substance use.” In my book (or anyone’s if it’s a dictionary), to deny something means to refuse to admit to it. This implies that you have been confronted with an allegation and are stubbornly asserting its falsehood. I pictured Emily cringing in a corner while Dr. K, whom I had previously thought to be mild-mannered and sympathetic, stood over her, shouting, “Have you or have you not used heroin?” But then I shouldn’t hold this choice of words against a particular practitioner, as it’s probably the recommended lingo for psychological interrogations.

There was only one drug listed in the medication history, and it was from an episode that I’d completely forgotten. A year before we went to PsychoPharm, Emily had complained of insomnia to our family doctor, who had prescribed nefazodone. That’s the antidepressant with very few side effects (see chapter 11), which was unfortunately pulled from the market due to possible liver damage. Emily only took it for three weeks (she didn’t like something about the way it made her feel), which wasn’t long enough for it to have any effects, good or bad. “Inadequate trial,” Dr. K had written, revealing her awareness that all drug-taking is experimental, whether as part of a structured clinical trial or with a prescription from one’s doctor.

Of course, nefazodone was never tested on children for the treatment of insomnia, depression, or anything else, but that didn’t keep doctors from prescribing it (see chapter 20 for more about off-label drug use). Physicians are only now beginning to design clinical trials for treating insomnia in children and adolescents. In 2006, a group of experts in the treatment of pediatric insomnia published their recommendations for design of such studies, asserting that “simply prescribing a reduced dosage of a drug approved for adult usage or using drugs indicated for other conditions is not a satisfactory solution.” They also noted that insomnia is frequently a symptom of mood and anxiety disorders in both children and adults. In fact, 75 percent of children with major depressive disorder have sleep problems. (Mindell et al., 2006)


The nice man at PsychoPharm wrote back, very apologetic, to say that he’d accidentally requested the wrong file from storage. He said that he would request the right one and mail me a copy as soon as it arrived. About a month later, after just one more badgering email, I received two fat envelopes documenting my daughter’s treatment at the hands of Big Pharma.

The new file contained, in addition to everything I’d gotten in the first shipment, multiple instances (from different dates) of a variety of instruments for rating a child’s anxiety level. There was the K-SADS-GAD (for General Anxiety Disorder), the CARS (Children’s Anxiety Rating Scale), and something called SCARED: Self-report for Childhood Anxiety Related Disorders. There were so many different forms and questionnaires, and the file was so disorganized, that I eventually gave up on trying to piece it all together. But in virtually every case, whether the questionnaire had a 3- or 7-point scale, Dr. K had indicated that we were all three in complete agreement on the severity of Emily’s symptoms. The only exception was the SCARED, which happened to be the only questionnaire that we had filled out ourselves.

The SCARED consists of two parts, one for the parent and one for the child. Each form has a series of statements, with three possible ratings for each: not at all true (0), sometimes or somewhat true (1), and very true or often true (2). There was some overlap between the psychologist-administered and the self-completed forms, so it was easy to find discrepancies between what Dr. K had claimed we said and what we really thought. Here are a couple of examples:

On the psychologist-completed forms, we were always in complete agreement. It seemed pretty clear that the facts, as recorded by Dr. K, had been “fixed around” the diagnosis (to borrow a phrase from British foreign intelligence).

The SCARED did show a few areas of mother-daughter agreement. For instance, we both knew that Emily didn’t mind at all being away from her family, and we both knew that she never worried that something bad might happen to her parents. In fact, she probably hoped that both her parents would die so that she could live with someone else. (This is pure speculation, of course, because such fantasies weren’t included on any of the rating instruments, but from the age of 3 Emily had exhibited occasional matricidal tendencies: When asked to identify her favorite part of the movie Bambi, she had answered, with no hesitation, “The mother, dead.”)

There were some other items on which I had managed to guess her feelings correctly (with a three-point scale, the odds were pretty good), such as “When my child gets frightened, he/she feels like things are not real.” and “My child is afraid to be alone in the house.” I rated both of these as somewhat or sometimes true, even though I had no idea how she really felt.

Some of our points of disagreement gave me twinges of retroactive guilt for being so oblivious to my daughter’s suffering. I had indicated that Emily never felt dizzy when she was frightened; she had rated this statement as very true. I rated her as not at all afraid of having panic attacks; she rated herself as very afraid. But considering the rancorous relationship we had at the time, I think I did pretty well.

As far as I could tell, the physical exam was not rigged to make her appear healthier than she was, but only the ECG was abnormal. It revealed a condition called sinus bradycardia, which basically means that she had a very low heart rate. According to an article in the eminently authoritative Wikipedia, sinus bradycardia is “not necessarily a bad symptom.” That reassured me enough to keep me from accusing PsychoPharm of withholding vital health information. For all I know, they did alert me to this symptom at the time, and being a negligent, self-absorbed parent, I chose to ignore it.

The study medication was buspirone, just as Emily remembered, and the sponsor was Bristol Myers Squibb, manufacturer of the brand-name drug BuSpar. A few years later the company was found by the FDA to have illegally obstructed entry into the market of generic versions of some of its drugs, including BuSpar.

After she began taking the drug (or placebo), we returned for five more visits. Each time, various tests were administered to see if she’d gotten any less anxious. At the end of six weeks, her illness had improved only slightly —she had progressed from “markedly” to “moderately” ill. The good news was that she’d had no noticeable adverse events. Of course, we don’t know for sure that she was taking any drug at all, or how much of it she took, but it seems likely that she’d gotten the real deal, just not enough of it or not for long enough.

All the rating instruments were administered again at the end of the study. There were a few symptoms on which Dr. K noted a 1 instead of a 2, e.g., crying spells and temper tantrums. In other areas Emily seemed to have gotten worse, like being afraid to go to school and to be home alone. For the most part, everything was about the same. On the SCARED, however, there were some interesting changes. I totaled up all the 0’s, 1’s, and 2’s on all four of our questionnaires, and here’s what I found:

So even if the drug had no noticeable benefit, even if my urging Emily to participate in the study had caused her temporary hatred of me to grow exponentially, and even if, as she now claims, the drug caused her lasting harm, at least the experience resulted in greater understanding, on my part, of the pain my daughter had endured in junior high school.

I was curious about the overall results of the study, now that I knew of at least one subject who had shown hardly any improvement, so I looked for a published report. In an article about treating anxiety disorders in children, I found one sentence that said Bristol-Myers Squibb had reported to the FDA that their study had not shown buspirone to be safe or effective in children. Because the company hadn’t been required, or even encouraged, to register the trial, most researchers or clinicians never knew about it, so they would never look for a report of the results. And because there was no published report, a 2002 book was able to cite a lack of evidence to the contrary in advocating the use of buspirone for GAD in children and adolescents (Rosenberg, Davanzo, & Gershon, 2002).


Previous chapter | Contents | Next chapter | References | Contact