9.3 Special-Population Prescribing
Key Takeaways
- Pediatric doses are milligrams per kilogram of the stated liquid concentration — never assume an adult tablet splits into a safe, even pediatric dose.
- Classic pediatric avoid-list: tetracyclines in young children (permanent tooth staining, with limited disease-specific exceptions you should not over-claim), fluoroquinolones not first-line, and aspirin during viral illness (Reye syndrome).
- FDA PLLR replaced letter categories with narrative Pregnancy, Lactation, and Females and Males of Reproductive Potential sections — read the story, not a remembered A/B/C/D/X grade.
- Known teratogens include ACE inhibitors/ARBs, isotretinoin, valproate, methotrexate, tetracyclines, warfarin, and live vaccines; preferred pregnancy examples include labetalol/nifedipine/hydralazine, insulin, doxylamine-pyridoxine, and many beta-lactams.
- For lactation, check LactMed (Hale is a conceptual risk framework) rather than inventing pump-and-dump rules; in geriatrics, start low, go slow, dose to eGFR, and count anticholinergic burden.
ANCC scores eight age groups (infant through frail elderly) against every drug class on the outline. Special-population prescribing is not a courtesy chapter. It is how the same ACE inhibitor is first-line in a 52-year-old with diabetic CKD and an absolute mistake in the second trimester. This section is the lifespan filter you apply after Sections 9.1 and 9.2.
Pediatrics: milligrams per kilogram, not half an adult tablet
Write pediatric systemic drugs as mg/kg/dose (or mg/kg/day divided), then convert to mL of the concentration in the bottle. Amoxicillin 400 mg/5 mL is not the same liquid as 200 mg/5 mL. Parents who "give one teaspoon" of the wrong bottle under- or overdose. Recalculate when weight changes. Recalculate when the pharmacy substitutes a different concentration. Show the mL on the script and in the after-visit summary.
Never assume an adult tablet is safely divisible. Unscored film-coated or extended-release tablets do not split into equal doses. Even a scored 10 mg tablet may not be licensed as a 5 mg pediatric product. If a liquid or a chewable exists, use it. If you must split, only scored immediate-release tablets and only when a pediatric reference supports that exact split.
Tetracyclines in young children. The classic exam fact is permanent staining of developing teeth (and effects on bone) when tetracyclines are given during tooth formation, generally taught as a reason to avoid the class in young children and in pregnancy. Do not over-claim the exception. Short-course doxycycline is the recommended treatment for suspected Rocky Mountain spotted fever and some other tick-borne illnesses even in young children, because untreated RMSF kills and a brief doxycycline course is not the same as months of tetracycline for acne. That is a disease-specific rescue, not permission to use doxycycline as a first-line ear-infection drug in a 4-year-old.
Fluoroquinolones are not first-line in children. Musculoskeletal concerns (cartilage, tendinopathy) plus a better-studied pediatric alternative for most community infections keep FQs off the first line. Exceptions exist (some complicated urinary infections, anthrax, certain resistant organisms) and belong to a specialist plan, not to a well-child Friday add-on.
Aspirin during a viral illness in children and teenagers is linked to Reye syndrome. Do not treat influenza or varicella fever with aspirin. Use acetaminophen or ibuprofen at pediatric doses when an antipyretic is indicated, and say aspirin out loud so the parent does not reach for a leftover adult bottle. Pepto-Bismol (bismuth subsalicylate) is a salicylate cousin — do not hand-wave it as "not aspirin."
Other pediatric Implementation habits: weight-based liquids for most antibiotics and antipyretics; avoid OTC cough-and-cold combinations in young children; codeine and tramadol are the wrong pediatric analgesics (Section 9.1). If the stem gives only an adult tablet strength and a toddler's weight, the correct move is often "do not write that tablet."
Pregnancy: narrative labels, not letter grades
The FDA Pregnancy and Lactation Labeling Rule (PLLR) replaced the old A / B / C / D / X letter categories with narrative subsections: Pregnancy (including labor), Lactation, and Females and Males of Reproductive Potential. Letters are gone from new labels because a "C" hid everything from "no data" to "real animal harm." If an item still waves a letter, treat it as outdated packaging. Read the narrative: human data, animal data, dose timing, and what to monitor.
Known teratogens and pregnancy-avoid drugs you must recognize on sight:
| Drug / class | Why it is on the never-or-almost-never list in pregnancy |
|---|---|
| ACE inhibitors and ARBs | Fetal renal dysgenesis, oligohydramnios, skull hypoplasia — especially mid-to-late pregnancy; avoid the class |
| Isotretinoin | Craniofacial, cardiac, and CNS anomalies; iPLEDGE exists because of this |
| Valproate | Neural-tube defects and later neurodevelopmental harm; not a casual mood stabilizer in a person who can become pregnant |
| Methotrexate | Abortifacient and teratogen; washout and folate belong in the rheumatology/dermatology plan |
| Tetracyclines | Fetal bone and tooth effects |
| Warfarin | Fetal warfarin syndrome (nasal hypoplasia, stippled epiphyses) and CNS bleeding risk |
| Live vaccines (MMR, varicella, live attenuated influenza, others) | Theoretical and observed fetal-infection risk; do not give in pregnancy |
Preferred primary-care examples in pregnancy (not a complete obstetric formulary):
- Hypertension: labetalol, nifedipine, and hydralazine are the agents you should be able to name. ACE inhibitors, ARBs, and usually atenolol are the wrong list. Methyldopa appears in older stems; current practice prefers labetalol or nifedipine for chronic hypertension in pregnancy when a drug is needed.
- Diabetes: insulin is the preferred pharmacologic therapy. Do not start an ACE inhibitor for "renal protection" in a pregnant patient with diabetes — that is a teratogen, not tertiary prevention.
- Nausea and vomiting of pregnancy: doxylamine plus pyridoxine is the first-line prescription pairing you should know.
- Many infections: beta-lactams (penicillins, cephalosporins) are the workhorse. Nitrofurantoin and some other agents have trimester-specific caveats; do not treat pregnancy UTI from a nonpregnant pocket card.
If a chronic drug is a teratogen and the patient could become pregnant, Planning includes contraception or a preconception switch (valproate to a safer antiseizure plan with neurology; ACE inhibitor to labetalol or nifedipine before conception when possible). Implementation is stopping the ACE inhibitor the day you document a positive pregnancy test and replacing blood-pressure control the same visit.
Lactation: look it up; do not invent pump-and-dump
LactMed (NIH) is the source to name on the exam and in clinic. Hale's lactation risk categories are a conceptual framework you may see in textbooks; they are not a license to quote a pump-and-dump interval you cannot source. Most common beta-lactams, ibuprofen, and acetaminophen at usual doses are compatible with breastfeeding; that does not mean every drug is. Radioactive iodine, some antineoplastics, and a short list of others are incompatible. Codeine and tramadol are poor lactation choices because of 2D6 ultrarapid-metabolizer infants (Section 9.1).
Do not invent a 24-hour pump-and-dump rule for every medication. Timing relative to feeds, infant age and prematurity, relative infant dose, and oral bioavailability in the infant all matter. When you do not know, you open LactMed or call the pharmacist — you do not tell a parent to discard milk for a weekend because it "sounds safer."
Geriatrics: start low, go slow, dose the kidney, count anticholinergics
Start low, go slow is not timid medicine. It is how you avoid orthostasis from an alpha blocker, hypoglycemia from a sulfonylurea, and delirium from diphenhydramine. You still reach a target (blood pressure, A1c appropriate to frailty, pain control) — you just do not start the 45-year-old dose on day one.
Renal dosing is the highest-yield geriatric calculation. Muscle mass falls, creatinine stays "pretty," eGFR does not. Gabapentin, many DOACs, metformin, renally cleared antibiotics, lithium, and morphine-6-glucuronide all accumulate. Recalculate when the patient has vomiting, a new diuretic, or an NSAID you should not have added.
Anticholinergic burden stacks: oxybutynin plus diphenhydramine plus amitriptyline plus olanzapine plus meclizine. The older brain pays in confusion, constipation, dry mouth, urinary retention, blurred vision, and falls. Prefer non-anticholinergic alternatives (mirabegron rather than high-dose oxybutynin when appropriate; a second-generation antihistamine rather than diphenhydramine; not amitriptyline as a sleep drug). Beers flags many of these (Section 9.2); the FNP still individualizes.
Vignette. A 7-year-old has suspected RMSF after a tick-endemic camping trip: doxycycline now, not "wait until age 8 so the teeth are safe." A 29-year-old at 24 weeks with chronic hypertension on lisinopril: stop the ACE inhibitor today, start labetalol or nifedipine, and teach why. A breastfeeding parent with streptococcal pharyngitis: penicillin or amoxicillin is a LactMed-friendly default, not a weekend of discarded milk. An 88-year-old with eGFR 32 and insomnia: not diphenhydramine, not a full-dose adult gabapentin, and not an NSAID for the "ache" that will finish the kidney.
Match the molecule to the age group on the TCO, not to the dose you memorized in the adult hospital.
A parent asks the FNP to prescribe "half of an adult 500 mg tablet" of an unscored, film-coated antibiotic for a 14-kg toddler. What is the correct Implementation response?
Which statement about tetracyclines in children is accurate enough for FNP-BC?
A 32-year-old at 18 weeks of gestation has chronic hypertension previously treated with lisinopril. Which plan matches preferred pregnancy prescribing?
A breastfeeding parent needs an antibiotic and asks how long to pump and dump. What is the best FNP response?