A lactation consultant told me recently that, in her unit, they bet over lunch about what the InfantRisk Center is going to say when they call. Half and half is the running favorite. She said it with affection, but also with a question in her voice. If you're going to say the same thing every time, why are we calling?
That conversation has stuck with me, because it points at something we need to be clearer about. When we recommend "half breast milk, half donor milk" or a 1:3 ratio in the NICU or any other conservative starting point, we are not handing you the final answer. We are handing you a place to begin. The road keeps going from there, and you and the family are the ones walking it. And the number is not a default, even when it sounds like one. We build it from the drug, the dose, the rest of the regimen, and the baby you described to us. It lands on half and half a lot because careful answers tend to land in the same neighborhood.
What we mean by "half and half"
Before we go further, it's worth pinning down what we actually mean by half and half, because this is where a lot of callers get tripped up. We mean that half of the baby's total daily intake is mom's milk, and half is donor milk (or formula when necessary). Mom keeps taking her medication exactly as prescribed. Half and half describes what goes into the baby.
There are a couple of ways to deliver that ratio:
Mixing each bottle. Mom's milk and formula or donor milk are combined in the same bottle, half and half, so every feed is a blend. If mom is pumping, this is usually the simplest approach.
Alternating feeds. One feed is all mom's milk (at the breast, or a full bottle of expressed milk), the next is all stored mom’s milk, formula, or donor. This can be easier to chart in a NICU or hospital where feeds are scheduled, and it's the natural fit when mom is nursing directly rather than pumping.
These two are not interchangeable for every medication, and it's worth being deliberate about which one you pick. We usually recommend mixing each bottle. Alternating feeds usually works fine when a drug has a long half-life and sits at a fairly steady concentration in milk, because the baby gets about the same exposure whichever feed it lands on. But when a drug reaches a high peak and clears quickly, such as many fast acting opioids or some beta blockers, an alternating all-mom's-milk feed can hand the baby an undiluted dose right at that peak, which is what half and half is meant to soften. Mixing dilutes every feed, so it blunts those peaks. When peak exposure is the concern, mix the bottle; save alternating for the long-half-life, steady-state drugs.
Whichever method you use, stay consistent, so you have a stable comparator when it's time to decide whether to advance.
The same idea applies to other ratios we recommend. A 1:3 in the NICU means 25% mom's milk and 75% formula or donor. In a mixed bottle that's a true 1:3 blend at every feed; with alternating feeds it's one feed of mom's milk for every three of formula. The same peak-versus-steady logic holds, so for a high-peak drug the mixed bottle is the safer way to reach that ratio. If an infant could tolerate a full feed of mom’s milk for a high-peak drug, mixed feedings might not be necessary.
Why we start conservative
We answer about 11,000 helpline calls a year. Every call is different. Some are from a seasoned NICU team with two neonatologists, a pharmacist, and a lactation consultant in the room. Some are from a single overwhelmed prescriber in a rural clinic who has not managed a lactating patient on this medication before. We can't see who's on the other end of the line, and we don't know how much monitoring capacity the family has at home.
So we start at an exposure we're comfortable with when we can't see the rest of the picture. Low and up is easier to defend than high and back down, especially on day one, when nobody has any infant data yet.
If the team has more experience than the average caller, then you have room to move. We want you to move. When the lactation consultants in Maryland told me, "I think we could do more than half and half on a lot of these," my answer was: then do more. If you have the clinical context to push further safely, that's not us disagreeing with you. That's the system working.
One more piece of honesty: even within our team, different clinicians have slightly different risk tolerances. If you called us twice on the same case, you might hear a starting point a step higher or lower. Both would be defensible. Two pharmacists looking at the same incomplete picture won't always draw the line in the same place, and that's true of consultative practice anywhere. What it means for you is that the specific number you got from us is one reasonable conservative starting point, not the only one.
How the maternal regimen shapes the starting point
People sometimes assume our recommendation is a fixed answer per drug, that sertraline gets one answer and lithium gets another. It isn't. In practice, at least four things move the recommendation: the age of the baby, how sick or fragile the baby is, the dose or doses involved, and whether mom is on one medication or several (and at what doses). The starting point we give is shaped by all of them together, not just the L rating of any one medication.
A rough rule of thumb, though every case is its own:
- One medication at a standard dose: Full feeds is usually reasonable if the L rating supports it.
- One medication at a high dose: We might still say full feeds if the RID stays low and the drug doesn't have significant CNS or respiratory effects. If it does hit those systems, we'll often recommend a partial start.
- Several medications at standard doses: Overlapping mechanisms start to matter. Two L3s that both sedate aren't the same as one L3 that sedates.
- Several medications at high doses: We usually start at half and half or 1:3, because the combined exposure and the overlapping mechanisms compound in ways the individual monographs can't fully weigh.
One more factor shapes the starting point: whether mom took the medication during pregnancy. That cuts in two directions at once. On one side, a baby exposed in utero could be born with a load of drug already on board and a still-immature ability to clear it, which argues for more caution in the early days. On the other side, the fact that she was treated through pregnancy usually means the risk-benefit already landed in favor of treatment, that the exposure was judged worth the payoff, which argues for staying the course rather than pulling back hard at birth. We weigh both, and neither one automatically wins. The kinetics of the drug often drive these evaluations.
This is why our recommendations vary case to case, even for medications that would look identical to the app on their own. The app is answering the question, 'what is the risk profile of this drug?' The call center is answering the question, 'what is the risk profile of this specific regimen for this specific baby?' Those aren't the same question.
Treat it like any other titration
The framing that helps most clinicians I talk to is this: our recommendation is a starting dose. You already know how to titrate.
Think about heparin. You don't pick a dose and walk away. You start, you check the aPTT, you adjust. Think about insulin drips, or any psych med taper, or starting an ACE inhibitor. The opening dose is informed by guidelines, but the actual dose is informed by the patient in front of you.
Partial breastfeeding works the same way. Start at the conservative ratio. Watch the infant for whatever short-term outcome you'd be worried about: sedation, feeding changes, irritability, GI symptoms, weight tracking. If the baby looks the same as before, go up. A term, healthy 4-month-old can usually tolerate a faster step-up than a fragile NICU baby. The signal you're watching for is the same signal you'd watch for with any titration. Did anything change? If yes, why, and is it the drug? If no, keep going. And write it down as you go: where you started, what you were watching for, what you saw, and what the family decided. Nobody remembers the details six months later, and that note is what the decision rests on if anyone asks.
Monitoring in a compromised baby: sorting drug from disease
The titration model above assumes you can see the drug's effect on the infant. That's straightforward in a healthy term baby. It is harder in a compromised one, because the baseline isn't stable and many of the signals you'd otherwise watch for (apnea, sedation, feeding difficulty) can look identical to features of the underlying illness. Here's how we think about it:
Establish the baby's baseline first. Before you step up, know what the baby is doing at the current ratio. Frequency and severity of apnea and bradys, level of respiratory support, feeding effort and tolerance, tone, arousal at cares. That's your comparator. Without it, any change after a step-up is ambiguous.
Step up in small increments. For a fragile baby, a step of roughly 20% of total intake is one reasonable increment, small enough that if something changes, you can plausibly attribute it to the change. Larger steps can be reasonable in more stable situations. None of these numbers are thresholds. Pick a step big enough to make progress and small enough to safely adjust for anything might happen next.
Time your look. After a step-up, watch through at least one full drug exposure cycle at the new ratio before making a decision. Often, 24 hours is used. A single bad shift right after a step-up isn't a signal unless it persists.
Watch for the right signals. Check the anticipated infant adverse effects and watch for them. Worsening apnea or bradys beyond the baby's established pattern. New hypotonia not explained by other events. Decreased feeding effort or poor arousal at cares. A change in weight trajectory. These matter more than any single observation. If you see something worrisome, back down and observe for changes again.
Don't attribute the wrong things to the drug. If the baby's course is already dynamic, extubation, sepsis workup, other medication changes, hold at the current ratio through the change and step up during a stable window. Otherwise you'll be sorting through too many variables at once. Baseline variability the baby had before the step-up doesn't become drug effect just because you happened to advance the day before.
Trust your team's clinical eye. The neonatology nurses and providers at the bedside know this baby's pattern in a way no monograph can capture. If they're comfortable, you have more room to advance than the algorithm suggests. If they're uneasy, that's real data even without a specific finding. Pause the ramp and watch.
Advancing from half and half: what we tell callers
When callers ask us 'how long do we stay at half and half?', the honest first answer is that it depends on the situation, because there are really two kinds. For a small category of medications, half and half (or a lower ratio) isn't a temporary starting point at all. It's the plan, and it stays where it is. For everything else, it's titratable: a place to begin and move up from. So the first question is which kind of case you're in, and then, if it's titratable, how fast to advance.
Any guidance we gave you on the call is a reference point, not a rule. Sometimes we'll say 'advance at 72 hours if baby is tolerating,' or 'stay at half for a week and reassess.' That's our best read from the other side of a phone. But you're the one who can see the baby, the situation, the family's capacity, and how hard this mom is fighting to breastfeed, and on all of that you almost always know more than we do. If what you're seeing at the bedside points a different way than the timeline we floated, trust what you're seeing. Our guidance is there to inform your judgment, not replace it.
Advance gradually. The practical version most of our nurses use is to move up one feeding at a time. If a family has the bandwidth, one feeding a day is a common pace, though there's nothing magic about a day: swap one more feed to mom's milk, hold there, watch, then add another. Any slow, stepwise increase works. The point is to change one thing at a time so that if something shifts in the baby, you can tell what caused it. A fragile baby gets smaller, slower steps (see the monitoring section above); a healthy term baby can move faster.
If you're an outpatient family or a caller without much monitoring capacity, a reasonable default is to hold at the ratio we recommended until you think it's appropriate to reassess, and to call us back if you'd like help. There's no penalty for staying put a little longer. And if the baby is stable and doing well, you don't need to call back for “permission.” We're always glad to be a second set of eyes, though.
If the maternal regimen changes, you can call back. A new medication added, a dose increased, or a medication discontinued all change the exposure math, and the previous starting point may no longer apply. Same goes if something changes on the baby's side, a new diagnosis, a significant clinical change, or a new symptom that might be drug-related.
When "InfantRisk said" becomes a problem
A while back, a mom called us about her 4-month-old. She'd been doing half and half since birth on a relatively high antidepressant dose, the baby was thriving, she wanted to increase to full breastfeeding, and her pediatrician was not comfortable because at the time of delivery we had suggested the hospital team start at half and half.
That's the failure mode I want to head off. Our recommendation at delivery was for that moment, with that team, with no infant data yet. Four months later, that family had months of experience about how this baby tolerates this exposure. Four months of infant observation is information, and it belongs in a shared conversation.
So please: if the baby is doing well, say that out loud in the conversation. A number we gave you at delivery shouldn't outlast the evidence that replaced it. Moving off it is the treating team's call with the family, not ours, and you don't have to run it by us first, though we'd love to hear how it went, because that's how we recalibrate (more on that in a moment).
When the app is enough, and when a call adds value
We've been asked a lot recently whether teams should be calling us on every case where mom is on a lactation-relevant medication, or whether the app is enough. Both are reasonable resources, and they're built to complement each other. Roughly, this is how we split them.
The app is designed to answer:
- A healthy term infant with a mother on a single medication at standard doses. The routine case.
- Follow-up questions on infants who are tolerating an established maternal regimen without concern.
- Confirming a monograph rating and reading the underlying data yourself.
Call us for cases where you need more support:
- Polypharmacy with overlapping mechanisms. Multiple L3s that both cause sedation, or both act on the CNS or respiratory drive, are not the same as one L3 on its own, and the app cannot dynamically combine them.
- Unusually high maternal doses.
- Cases of maternal adverse effects. If mom is altered, there is more suspicion for higher than average drug presence in blood and then milk.
- Preterm or NICU infants on respiratory support, particularly when the maternal medication has known CNS or respiratory effects.
- Recent in-utero exposure combined with ongoing lactational exposure, where you want a starting point that accounts for what is already on board.
- Cases where your team wants a second set of eyes before advancing past a starting point that feels conservative. We're happy to reason through it with you.
A single L3 in an otherwise stable NICU baby is a judgment call. If the maternal dose is standard, the app monograph is likely enough. If the team is uncertain, call us. An individualized assessment is worth the ten minutes on the phone.
The exception: drugs we say to avoid entirely
There is a small category of medications we flag as not compatible with breastfeeding, or as requiring a defined interruption: specific chemotherapy agents, specific radioactive isotopes, and a small number of other agents in defined clinical contexts. These are agent-by-agent determinations rather than class-wide ones, and most drugs within those categories are not on the list. Those recommendations don't move much, because they're not based on a conservative starting estimate. They're based on the pharmacology of the individual agent.
Everything else, the "start at half and half" calls, the "consider holding for X hours" calls, the polypharmacy calls, those are designed to be moved. If we wanted to give you the final answer, we would say "do not breastfeed." When we instead say "start here," it is because there is room.
The short-term and long-term asymmetry, and where mom’s voice matters
There's one place this titration model gets hard, and you should know about it so you can name it for families. Short-term outcomes are watchable. Sedation, feeding issues, irritability, you'll see them, and you can back off. Long-term outcomes, subtle effects on neurodevelopment, for example, in a baby chronically exposed to a maternal psych medication, don't show up in the next visit. You can't titrate to a signal you can't see. But…Don’t forget about the known benefits of breastfeeding when weighing the unknown risks of infant exposure. Don’t forget the costs of an untreated mother.
In those situations, nobody, not us, not you, not the prescriber, has the data the family deserves. What we can do is have an honest conversation about it: here are the short-term effects we would monitor, here is the long-term question we cannot fully answer, here is what we know about exposure during pregnancy versus exposure through milk, here is what the family values, here is what they are willing to accept. The clinical recommendation in that situation isn't ours to make alone. It is a team conversation with the family in the center.
Please call us back
The single most useful thing a clinician can do for us, and for the next family who calls in with this drug, is to call back and tell us what happened. We went up to three-quarters at week two, the baby was fine. Or: We tried full feeding and saw sedation, dropped back to half, it resolved. That feedback loop is how our recommendations get better over time. The recommendations themselves rest on published pharmacokinetic and milk-concentration data, relative infant dose, and monograph review. What field reports add is the clinical experience those sources cannot supply, and they let us update on a faster cycle than a guideline revision allows.
If you don't call back, we don't learn. And the next family stays at the conservative starting point a little longer than they had to.
A working analogy
The ER doesn't call poison control on every overdose. They call when they need extra help, a complicated co-ingestion, a drug they don't see often, a clinical picture that doesn't add up. The rest of the time, they handle it themselves, with poison control as a backstop they can reach if they need it.
That's the relationship we want with you. If the situation is complicated, or the medication is unusual, or you want a second set of eyes on the risk-benefit, please call: 806-352-2519, Monday through Friday, 8:00 a.m. to 3:00 p.m. Central. If you've got it handled and the baby is telling you it's fine, that call is yours to make with the family. Just document what you saw and why you moved. You have the patient in front of you, and we don't.
We are here to give you a leg up, not to unnecessarily stand in your way.
Alicia Nelson RN BSN IBCLC
Sally Bain RN BSN
Katie Boatler RN BSN
Kaytlin Krutsch PhD PharmD MBA BCPS
The InfantRisk Center is part of Texas Tech University Health Sciences Center. We provide evidence-based consultation on medications, environmental exposures, and infant safety during lactation. The MommyMeds app (for parents) and InfantRisk Center HCP app (for clinicians) are curated by lactation pharmacology specialists and updated as new research is published. The InfantRisk Center provides drug information and risk assessment based on the information reported by the caller. We do not examine patients, do not have access to the medical record. Our guidance is informational and is intended to support, not replace, the judgment of the treating clinician, who remains responsible for the care of the patient. Parents and caregivers reading this article should discuss any change in feeding or in medication with their own clinician before making it.
