The Hospital Changed Mom's Medications

Nothing went wrong in a dramatic sense. Medication lists come apart in the hospital for ordinary, structural reasons, and understanding them makes the discrepancies less frightening and easier to describe on the phone.

A hospital pharmacy stocks its own formulary, so a home medication is often swapped for a therapeutic equivalent it carries. Medicines are deliberately held during a procedure or an acute illness — and the order to restart them is a separate step that sometimes never happens. Doses get adjusted in response to labs drawn that morning. Several specialists write orders on the same patient across a five-day stay. And the home list the hospital started from may itself have been wrong on the way in, since medication histories taken at admission are a well-documented weak point ^[5].

Reconciliation is formally the hospital's responsibility, not yours. The Joint Commission's National Patient Safety Goals require accredited organizations to maintain and communicate accurate medication information — to find out what the patient is taking, compare it against what is being prescribed, give the patient written information, and tell them to bring an up-to-date list to every visit ^[7]. That requirement has been in force in one form or another since 2005 ^[5]. Errors still reach roughly half of discharged patients ^[1]. Your check at the kitchen table is not paranoia or second-guessing the team. It is the last layer of a system that is known to leak at exactly this seam.

Every difference between the old list and the new one falls into one of five categories. Naming the category is most of the work, because it tells you what question to ask and who to ask it of.

The work is clerical, and it goes faster than it looks. Do it at a table with everything in front of you, before the first full day of dosing at home is behind you.

If you have twenty minutes rather than an hour, spend them on the drug classes most often implicated when something goes wrong after discharge. In the systematic review, the adverse drug events reported after discharge most commonly involved antibiotics, diabetes medicines, pain medicines, and cardiovascular drugs — particularly blood pressure medicines and anticoagulants ^[1]. Adverse drug events are the most common post-discharge complication overall ^[6], and in one classic study of patients discharged from a general medical service, 19 percent had an adverse event within about three weeks, two-thirds of which were drug-related ^[3].

Three specific patterns are worth looking for by name. A duplicate: the same drug appearing twice under a brand name and a generic name, or a swap where the old bottle is still in rotation. A blood thinner discrepancy of any kind — dose, frequency, or which agent — because the margin for error is narrow. And an insulin or diabetes medicine whose dose changed in the hospital, where eating patterns were nothing like they will be at home.

The rule that governs all of it: verify, do not adjust. Do not start, stop, split, or skip a prescription because a piece of paper and a pill bottle disagree. Take the discrepancy to the prescriber or a pharmacist and let them make the call.

Remote caregiving turns a two-hour job into a week of phone tag, mostly because nobody can see the same thing at the same time. Texted photos of pill bottles are not a system; they are a slideshow that ends up scattered across four phones.

Three moves help. First, get the discharge summary yourself rather than relying on relay — through the patient portal with proxy access, or a records request. Second, split the roles explicitly: one sibling owns the worksheet and keeps it current, one owns the phone calls. Two people calling the same office about the same drug produces two different answers. Third, keep one shared source of truth that everyone reads instead of a group chat that everyone skims. A caregiver workflow for medical records covers the wider version of this system, and organizing medical records without the overwhelm is where the discharge folder itself should end up.

One more practical note: fill the new prescriptions at a single pharmacy if you can. A pharmacy's interaction check only sees what it has dispensed, which is why keeping the records in one place is a standing recommendation for older adults on multiple medicines ^[9]. It is the closest thing to a free second opinion available the week after a hospital stay.

The reconciliation is not finished until one clean list exists and everyone is working from it. That usually happens at the first post-discharge visit. Medicare's transitional care management rules give a useful sense of the intended tempo: contact with the patient or caregiver within two business days of discharge, a face-to-face visit within seven days for higher-complexity cases and within fourteen days otherwise, with medication reconciliation completed no later than the date of that visit ^[12]. If nobody has offered that visit, ask for it.

Bring the worksheet with every UNCLEAR row still marked, and bring the bottles if you can carry them. Ask the three questions that matter most: which of these does she keep taking indefinitely, which have stop dates, and is there anything on this list she no longer needs. Then ask for the reconciled list in writing and send it to the pharmacy and any specialist who prescribes for her. Questions to ask before a doctor appointment and what to bring to a specialist visit both help you get more out of the fifteen minutes, and an appointment prep packet puts the current list in your hand at the door. If remembering the conversation afterward is the hard part, recording the visit with permission means the instructions survive the drive home.

Keep the dated versions rather than overwriting them. A hospitalization is the single biggest medication event most people experience, and seeing what started, stopped, and changed around that date is exactly what a personal health timeline is for.

It is worth being straight about the evidence. Discrepancies after discharge are common and measurable: in a study of older adults, 14.1 percent had at least one medication discrepancy at home, and 14.3 percent of those patients were rehospitalized within 30 days compared with 6.1 percent of patients without one ^[2]. A 2024 study of older ICU survivors found unintentional discrepancies in 68 percent of patients somewhere across their care transitions, 49 percent at discharge specifically, and roughly twice the risk of an emergency department visit within 30 days among those affected ^[4]. Those are associations observed in specific populations, not a promise about any one person.

And AHRQ is candid that medication reconciliation programs alone have not been shown to reduce readmissions or adverse events after discharge ^[5]. So the honest claim for the evening you are about to spend is narrower than a health outcome: this catches discrepancies, and it produces a specific question for a clinician who can act on it. That is a real contribution, and it is one nobody else in the system is positioned to make — because you are the only person who can see both the paperwork and the pill cabinet.