Chapter 9
The email from the NIH arrives on a Tuesday in early November, which is appropriate because every significant thing that has happened to me in the last three months has apparently decided that Tuesday is the day.
I’m at my desk in my campus office, grading midterms for my Advanced Biostatistics course — the one my graduate students covered while I was in the recovery facility, the one I took back last week with a sleeve on my left arm and a shoulder that still doesn’t lift past a hundred and ten degrees.
The office is small and overcrowded with books in a way that I used to apologize for and now find comforting.
It smells like coffee and printer toner and the particular must of journals that have been stacked on shelves for too long. It smells like mine.
I click it open with my right hand while my left hand holds the coffee mug steady on the desk.
A small act of coordination that would have been impossible six weeks ago and is now just a thing I do — hold a mug, click a mouse, live a life in the space between what my body was and what it’s becoming.
Dear Dr. Merrifield,
We are pleased to inform you that your application, “Adaptive Sequential Dosing Models for Oncology Clinical Trials: A Bayesian Framework for Real-Time Dose Optimization,” has been selected for funding under the R01 Research Project Grant mechanism.
I read the sentence three times. The words are clear and plain and they mean what they say: the grant was funded.
My grant. The one I designed and wrote and submitted and agonized over and talked about at a dinner on a deck that burned down.
The one Rowena remembered at that dinner.
The one Langston said he’d extrapolated.
The one I applied for because adaptive dosing methodology is important and because I believed the research would change how oncology trials are conducted and because I am a scientist and this is what scientists do.
The funding amount: two hundred and seventy-five thousand dollars a year for three years.
The review score: in the top fifth percentile.
The reviewer comments, attached as a PDF I will read later: uniformly positive, noting “exceptional methodological rigor” and “immediate translational applicability.”
The grant is mine. Not Merrifield’s. Not the foundation’s.
Not a committee seat given to me because I married the right name.
This is Dr. Calla Merrifield’s grant, earned by her research, reviewed by her peers, funded on its merit.
Every cent of it came from work I did with my own mind while living in a marriage that gave me four percent of its attention.
I sit at my desk and I hold the coffee mug and I feel something I haven’t felt since the fire: pride that belongs entirely to me. Pure and unshared and not contingent on anyone else’s approval or calendar or presence at a gala. Mine.
I text Auden: Got it.
Her reply comes in four seconds: THE GRANT?
Then: I’m coming over tonight. I’m bringing champagne. The real kind.
The return to teaching happens in stages.
The first week back, I lecture from a stool because standing for ninety minutes makes my left side ache in ways I don’t have the energy to manage while explaining maximum likelihood estimation.
By the second week I’m standing for the first forty-five minutes and sitting for the rest. By the third week — this week — I stand for the whole session and only notice the ache when I stop moving.
The students treat me the way I want to be treated: as their professor.
Not their object lesson, not their cautionary tale, not the woman whose story they might have heard through the department’s inexhaustible gossip network.
They ask about the midterm curve. They argue about confidence intervals.
They complain about the homework. When I write equations on the whiteboard with my right hand, they copy them down without glancing at the arm I’m not writing with.
“The critical region for a two-sided test at alpha equals point-oh-five is bounded by the absolute value of the z-statistic exceeding one-point-nine-six,” I say, drawing the bell curve and shading the tails.
“The region defines your rejection criteria. You decide, before you see the data, what threshold of evidence would cause you to reject the null hypothesis. This is important. You set the boundary first. The data arrives second.”
I pause. The room is quiet. Thirty-two faces watching me — graduate students, mostly, a few advanced undergrads who shouldn’t be here but are because the department’s scheduling is, as always, a controlled experiment in chaos.
“Never move the boundary after you’ve seen the data,” I say. “That’s called data dredging, and it will get your paper retracted and your career shortened. You draw the line. The data falls where it falls. You interpret honestly.”
I am not talking about statistics.
I’m not not talking about statistics, either.
The line between the two is thinner than it’s been in weeks, and I let it stay thin, because standing in front of a classroom full of people who care about my methodology and not about my marriage is the most alive I’ve felt since the fire.
Here, I am Dr. Merrifield. Here, the name means the woman who designed the adaptive dosing framework that three oncology centers are already citing. Here, the name is mine.
After class, a student named Park approaches the lectern. She is a second-year PhD candidate, quiet, meticulous, the kind of student who turns in homework that looks like a published paper.
“Dr. Merrifield, I read your Montreal paper over the break,” she says. “The one on crossover contamination in dose-escalation designs.”
“And?”
“I think there’s an extension. A way to apply the framework to immunotherapy dosing, where the dose-response curve is non-monotonic.” She sets a printed page on the lectern — her own preliminary analysis, typeset, with references. “Would you be willing to look at it?”
I look at the page. The analysis is clean.
The intuition is correct. The student has read my work — not the abstract, not the summary, the actual paper — and has seen something I hadn’t considered.
A twenty-four-year-old in her second year of a PhD program has engaged more deeply with my research than my husband did in five years of marriage.
“Yes,” I say. “Come to my office hours Thursday. Bring your dataset.”
She nods and leaves and I stand at the empty lectern with her page in my hand and I think about how Langston said he’d extrapolated my methodology and I think about how this student actually read it, and the contrast is a small, specific pain that I file away with the others.
The lecture invitation arrives the following week. The New England Biostatistics Consortium is hosting its annual symposium in Cambridge, and they want me to deliver the plenary address. Not a panel. Not a breakout session. The plenary. My grant, my methodology, my name on the marquee.
“This is a big deal,” Auden says. We’re in my apartment.
She’s cross-legged on the couch with a glass of wine — the good terrible wine, our wine — and I’m at the desk reading the invitation for the third time.
“Three hundred people minimum. The department chair from Harvard attends every year. So does the head of the FDA’s biostatistics division. ”
“I know, Auden.”
“I’m saying this because the last time you had a room of three hundred, you gave it up for a gala.”
The words land. She doesn’t say them with malice — Auden doesn’t do malice, she does precision — but they land anyway, because they’re true and because the truth of them is exactly what makes this invitation feel like more than a professional accolade.
It feels like a correction. A recalibration.
The universe handing back the three hundred people I gave away in Montreal and saying: here, try again.
“I’m going,” I say.
“Obviously you’re going.”
“Not ‘obviously.’ Last year I would have checked Langston’s calendar before accepting.”
“This year you don’t have a Langston calendar to check.”
“No. I don’t.”
She raises her glass. “To the plenary.”
I raise mine. “To the plenary.”
We drink. The wine is as bad as always. It tastes like freedom.
The symposium is on a Saturday in late November.
I wear a blazer that covers the compression sleeve on my left arm, not because I’m hiding it but because the sleeve is beige and the blazer is navy and the contrast is distracting, and I want people looking at my slides, not my arm.
I spend twenty minutes at the bathroom mirror in the conference center, getting the blazer to sit right over the bulk of the sleeve, adjusting the cuff so it doesn’t ride up when I gesture.
My left hand can hold a laser pointer now.
Not steadily — it trembles if I hold it too long — but steadily enough for fifty minutes of slides.
The auditorium is full. Not thirty people — three hundred.
The room has tiered seating and good lighting and the kind of microphone system that makes your voice sound like it belongs there.
I stand at the podium and I look out at three hundred faces and I feel the vertigo of being seen — not as Langston Merrifield’s wife, not as the woman from the fire, not as the Merrifield Foundation’s advisory committee member, but as Dr. Calla Merrifield, the researcher whose adaptive dosing framework is being cited in half the oncology trial design papers published this year.
“Good morning,” I say. “I’m going to talk about how we change minds in the middle of an experiment.”