From hospital discharge to primary care: transition gaps

A 78-year-old woman is discharged after five days in hospital for pneumonia. She goes home with a paper summarizing the admission, a new prescription and the advice to contact the health unit. At the appointment, a week later, the family doctor finds that two old medicines were stopped without explanation, that a new one was added and that a test collected on the last day of the admission still had no result when she left. Nobody knows who will look at that result.

Hospital discharge is one of the riskiest passages in care. In a few days, the person goes from an environment watched all the time to their own home, and responsibility passes from the hospital team to primary care. This text shows where this transition usually fails, what research says about errors and adverse events after discharge, what a good discharge summary needs to contain and how to organize the first appointment in primary care.

Key points

• Errors linked to the discontinuity between hospital and primary care are frequent and are associated with readmission.
• A significant share of people have adverse events after discharge, most linked to medicines and many preventable.
• Medicine discrepancies in the discharge summary are common, especially in older people taking many medicines.
• Tests pending at discharge often bring results that require action, and often nobody is responsible for them.
• For the family doctor, the accuracy of the summary, the medicine changes and the dialogue with the hospital are what matter most.
• The first appointment after discharge needs to review medicines, pending items and warning signs, not just happen early.

Summary

• Why discharge is a fragile point
• Adverse events after discharge
• Medicines: where most errors happen
• Tests pending at discharge
• The discharge summary
• The first appointment after discharge
• The role of the team
• Situations that call for more care
• Frequently asked questions
• Final checklist

Why discharge is a fragile point

A study of 86 patients discharged from a general medicine service found at least one error linked to discontinuity of care in about half of them (Moore, 2003). The authors described three types of error: medicine continuity, test follow-up and investigation (Moore, 2003).

The test follow-up error happened when a result pending at discharge was not recognized afterwards. The investigation error happened when an outpatient test suggested by the hospital team was not done (Moore, 2003). This last type was the most worrying: patients with an investigation error had a 6.2 times higher chance of readmission within three months (Moore, 2003). In the service studied, discharge summaries were not even sent to the primary care doctor (Moore, 2003).

The fragility of discharge has several causes. The admission changes the treatment, often quickly. The hospital team knows the acute illness well, but little about the person's life. The family doctor knows the person, but did not follow the admission. And the information that links the two depends on a document written in a hurry, on a day full of other tasks.

The person is also more vulnerable. They have come out of an acute illness, may be tired, confused or afraid, and need to understand a new treatment. The text on delirium in the hospitalized older adult recalls that confusion can persist after discharge and get in the way of understanding the instructions.

There is also a question of time. Discharge is usually decided on the same day it happens, with the bed already expected by another patient. The summary is written among other tasks, the guidance to the patient is given standing up, near the door, and the doubts that come up at home at night have no one to turn to. None of this is an individual failure of those working in the hospital; it is the design of the process that leaves little room for a careful passage.

Adverse events after discharge

A study of 328 general medicine patients followed after discharge found 76 with at least one adverse event, about one in four (Forster, 2004). Most events were linked to medicines, and about half were considered preventable or mitigable (Forster, 2004).

Another study from the same group followed 400 patients and found 45 with an adverse drug event after discharge, about one in ten (Forster, 2005). The risk increased with the number of prescriptions, and lack of monitoring was a frequent cause (Forster, 2005). A practical finding: when side effects had been explained to the patient, the risk was lower (Forster, 2005).

These studies show that much of the harm after discharge does not come from the illness that led to admission, but from the new treatment and the lack of follow-up. This shifts the focus: the question is not only whether the person is well enough to go home, but whether someone will follow up on what changed.

For primary care, this means that the first days and weeks after discharge deserve special attention. New dizziness, a fall, confusion that does not improve, tiredness greater than expected or a change in appetite may be effects of the new treatment, and not of the illness that led to admission. Asking about this actively, instead of waiting for the person to bring up the complaint, helps find the problem early.

Medications: where most mistakes happen

A study of 933 patients aged 75 or older who used five or more medications found medication discrepancies in the discharge summary of 353 of them, about four in ten (Caleres, 2020). The most common discrepancy was the unintentional inclusion of a medication (Caleres, 2020). The risk factors were the use of medication packages prepared by the pharmacy for each time of day, the number of medications and discharge from a surgical service (Caleres, 2020).

Medication reconciliation is the systematic comparison between what the person used before, what they used in the hospital and what they should use afterwards. A review of 10 studies showed that electronic reconciliation reduced the proportion of medications with a discrepancy (relative risk 0.55), but did not reduce the number of patients with at least one discrepancy (Mekonnen, 2016). Omissions were the most common type of error (Mekonnen, 2016).

A trial with 3,491 patients tested an electronic reconciliation tool at discharge. There were fewer discrepancies in the intervention group, 437 versus 1,029, but there was no reduction in adverse drug events or readmissions (Tamblyn, 2019). Changes to medications the person was already using at home were a risk factor for adverse events (Tamblyn, 2019).

These results suggest that correcting the list is necessary, but not enough. The person needs to understand what changed and why, and someone needs to follow up on the effects. The most common discrepancies in practice include:

• a medication stopped in the hospital that is mistakenly resumed at home;
• a new medication with no clear indication of duration;
• an old medication omitted with no record of the reason;
• two versions of the same treatment, one from the old prescription and one from the new;
• a change in timing or method of use that did not reach the patient;
• a treatment that needed a follow-up test with no date scheduled.

Automatic prescription renewal after discharge is another trap. The text on why automatic renewal is a clinical decision shows how a treatment started for a temporary reason can become permanent without anyone deciding that.

Pending tests at discharge

Many patients leave the hospital with tests still awaiting results. A systematic review estimated that about three to four in ten of these tests bring potentially actionable results (Darragh, 2018). The 9 intervention studies included tested summary templates with a specific field for pending tests, audit with feedback and email alerts (Darragh, 2018).

The specific fields improved documentation in 3 of 6 studies, and alerts and audit improved physicians' knowledge of the results (Darragh, 2018). No study measured whether these interventions reduced harm to patients (Darragh, 2018).

A qualitative study with 27 professionals from 12 services investigated why test results keep getting lost even with electronic alerts (Zimolzak, 2022). Among the themes found were resident turnover, incidental findings, difficulty tracking alerts, outdated patient contact data, referrals, professionals covering for colleagues and the lack of definition about who is responsible for pending results at discharge (Zimolzak, 2022).

The central point is responsibility. A pending result needs an owner: whoever ordered it, whoever discharges the patient or whoever follows up afterwards. If this is not written down, each person assumes another will look at it. The report on diagnostic error includes failure to communicate a diagnosis among the forms of error, and a forgotten result is a direct example of that.

Pending tests also have an emotional side for the patient. Many people leave the hospital thinking that, if no one called, the result was normal. Explaining at discharge that there are results still to come, who will look at them and how the person will be notified avoids this false reassurance. In primary care, asking directly whether any test result is still awaited is a simple habit that closes a known gap.

The discharge summary

The discharge summary is the main instrument of the transition. Focus groups with 18 family physicians showed three themes: the importance of the summary, the role of the family physician and the need for dialogue (Caleres, 2018). Participants described themselves as the center of a web and, at the same time, as the final station where everything converges (Caleres, 2018).

For these physicians, accurate summaries help the work, while flawed summaries erode trust in the hospital (Caleres, 2018). A phone call from the hospital physician before discharge, in the most complex cases, was highly valued (Caleres, 2018).

A survey with 36 hospital resident physicians and 42 family physicians compared what each group considers a good summary (Yemm, 2014). Accuracy was the priority for both groups (Yemm, 2014). But there was a difference in medication changes: 13 family physicians pointed to them as the most important item, against 4 residents (Yemm, 2014). And 13 residents reported not having received adequate training to write summaries (Yemm, 2014).

A useful summary for primary care usually contains:

• the main diagnosis and the relevant secondary diagnoses;
• what was done during the admission, in a few lines;
• the complete list of medications at discharge, with what was started, stopped or changed and the reason for each change;
• the pending tests, with who is responsible for checking them;
• the investigations recommended after discharge, with a deadline;
• the warning signs explained to the patient;
• who to contact in case of doubt.

Summary element	What happens when it is missing	Source
Medication changes with reason	Discrepancies and adverse events after discharge	Caleres, 2020; Tamblyn, 2019
Pending tests and responsible person	Actionable results without follow-up	Darragh, 2018
Recommended investigations	Tests not done and more readmission	Moore, 2003
Explanation to the patient about side effects	More adverse drug events	Forster, 2005
Channel of contact with the hospital	Loss of trust and rework	Caleres, 2018

The language of the summary also matters. A text full of abbreviations and acronyms may be clear to whoever wrote it, but confusing to whoever receives it in another service, and almost incomprehensible to the patient who carries it along. A plain-language version for the person and the family, with what changed in the medications, the warning signs and who to contact, complements the technical summary.

The first appointment after discharge

It is common to recommend that the person be seen in primary care soon after discharge. The recommendation makes sense, but the research shows that the timing of the appointment, on its own, does not solve it. A study of 3,661 patients, of whom 707 were readmitted within 30 days, found no association between an appointment with the primary care physician in the first 7 days and the chance of readmission (hazard ratio 0.98) (Field, 2015).

What matters, then, is what happens at the appointment. Studies on errors and adverse events point to three tasks: reviewing the medications, checking the pending items and confirming that the person understood the plan (Moore, 2003; Forster, 2005).

The medication review should be done with the person and, if possible, with the boxes in hand. Comparing the discharge prescription, the previous list and what the person is actually taking is the most direct way to find discrepancies. When there are many complaints and many pending items, the text on how to prioritize what is severe, urgent, important and can be postponed helps to organize the appointment.

Pending items include tests without results, recommended investigations and specialist appointments. If the summary does not bring this information, it is worth contacting the hospital. The family physicians in the qualitative study pointed out that dialogue with the hospital is part of the process, not an extra (Caleres, 2018).

Finally, the appointment is the moment to review the warning signs and agree on the follow-up. The text on planned reassessment applies here: the plan after discharge needs a date and a criterion for changing.

A simple sequence for this appointment can help. First, ask how the person is and what they understood about the hospitalization. Then review the medications with the boxes in hand. Next, check pending tests, recommended investigations and scheduled appointments. Finally, agree on warning signs, follow-up and who to contact if in doubt. Recording everything in the medical record ensures that the next person on the team to see the patient knows what was done.

The role of the team

The transition does not depend on physicians alone. Nurses, pharmacists and community health workers can contact the person soon after discharge, check the medications at home and identify problems before the appointment. The text on conditions for task transfer to work discusses what is needed for this division to be safe.

In the hospital, nursing also has a central role. The assessment of vital signs and the escalation during the hospitalization help to define whether the person is in fact stable for discharge. The final guidance to the patient, often given by nursing, is the moment when side effects can be explained, which reduced the risk of adverse drug events in one of the studies (Forster, 2005).

Communication between teams is the thread that links everything. The text on diagnosis as teamwork shows that the same logic applies within a service and between services.

The pharmacist, when part of the team, can have a central role in reviewing medications after discharge. Comparing the previous list, the discharge prescription and what the person actually takes is detailed work, and a second person looking at the same list finds errors that the first one missed. Omissions are usually the hardest to notice without an item by item comparison.

Situations that call for more care

Some discharges are riskier than others. Older people with many medications have more discrepancies (Caleres, 2020), and those who receive more prescriptions have more adverse drug events (Forster, 2005). Discharges from surgical services were also associated with more discrepancies (Caleres, 2020).

The ability to understand the plan also matters. When there is doubt about the person's capacity to make decisions, the text on what decision-making capacity is helps to decide who needs to take part in the guidance. A family member or caregiver present at discharge can make a difference.

Discharges on request or against medical advice call for extra attention, because planning is often interrupted. The text on risk reduction at discharge shows that even in these situations it is possible to organize a minimal transition.

People with serious illnesses and a limited prognosis also benefit from a careful transition. Discharge can be the moment to resume or start the conversation about goals of care, as discussed in the text on why the advance planning conversation matters.

The caregiver is often the one who organizes the medications at home, takes the person to appointments and notices the first changes. Including them in the discharge guidance and in the first appointment, with the person's consent, reduces the chance that an important change gets lost. It is also worth asking who will help in the first days, because a person who lives alone and leaves the hospital frail may need closer contact with the team.

People with little support network, difficulty reading, several illnesses or many new medications are the ones who benefit most from early contact with the team after discharge. A phone call in the first days, a home visit or an early appointment can be the difference between noticing a mistake in time and finding it during a new admission. Each service can define, with its resources, who goes on this priority list.

When the discharge summary does not arrive, it is worth contacting the hospital, asking for the document and, in the meantime, piecing the story together with the person, the family and the prescriptions and medication boxes they brought. Clearly recording the missing summary in the chart helps the team request the document later and keeps another professional from assuming it was read.

Frequently asked questions

What are the most common mistakes in the discharge transition?

Errors in medication continuity, in follow-up of pending tests and in recommended investigations that are not carried out (Moore, 2003).

Are adverse events after discharge preventable?

Many are. About half of the adverse events after discharge were considered preventable or mitigable (Forster, 2004).

Does electronic medication reconciliation solve the problem?

It reduces discrepancies, but in a large trial it did not reduce adverse events or readmissions (Tamblyn, 2019). The conversation with the patient and follow-up remain necessary.

What should be done about tests pending at discharge?

Record in the summary which ones they are and who is responsible for checking them. About three to four in ten bring potentially actionable results (Darragh, 2018).

Does seeing the patient in the first week reduce readmission?

In a large study, the appointment within the first 7 days, on its own, was not associated with fewer readmissions (Field, 2015). The content of the appointment seems to matter more than the timing.

What does the family physician most want in the discharge summary?

Accuracy and clarity about medication changes (Yemm, 2014). A channel for dialogue with the hospital is also valued (Caleres, 2018).

Final checklist

• At discharge, list the medications started, stopped and changed, with the reason for each change.
• Record the pending tests and who is responsible for checking them.
• State the investigations recommended after discharge, with a timeframe.
• Explain to the patient the expected side effects and the warning signs.
• At the first appointment, compare the discharge prescription with what the person actually uses.
• Check the pending items and contact the hospital when the summary does not bring what is needed.

References

1. Caleres G, Modig S, Midlöv P, Chalmers J, Bondesson Å. Medication Discrepancies in Discharge Summaries and Associated Risk Factors for Elderly Patients with Many Drugs. Drugs Real World Outcomes. 2020. PMID 31834621. DOI 10.1007/s40801-019-00176-5.
2. Caleres G, Strandberg EL, Bondesson Å, Midlöv P, Modig S. Drugs, distrust and dialogue -a focus group study with Swedish GPs on discharge summary use in primary care. BMC Fam Pract. 2018. PMID 30045692. DOI 10.1186/s12875-018-0804-8.
3. Darragh PJ, Bodley T, Orchanian-Cheff A, Shojania KG, Kwan JL, Cram P. A Systematic Review of Interventions to Follow-Up Test Results Pending at Discharge. J Gen Intern Med. 2018. PMID 29352419. DOI 10.1007/s11606-017-4290-9.
4. Field TS, Ogarek J, Garber L, Reed G, Gurwitz JH. Association of early post-discharge follow-up by a primary care physician and 30-day rehospitalization among older adults. J Gen Intern Med. 2015. PMID 25451987. DOI 10.1007/s11606-014-3106-4.
5. Forster AJ, Clark HD, Menard A, et al. Adverse events among medical patients after discharge from hospital. CMAJ. 2004. PMID 14757670.
6. Forster AJ, Murff HJ, Peterson JF, Gandhi TK, Bates DW. Adverse drug events occurring following hospital discharge. J Gen Intern Med. 2005. PMID 15857487. DOI 10.1111/j.1525-1497.2005.30390.x.
7. Mekonnen AB, Abebe TB, McLachlan AJ, Brien JA. Impact of electronic medication reconciliation interventions on medication discrepancies at hospital transitions: a systematic review and meta-analysis. BMC Med Inform Decis Mak. 2016. PMID 27549581. DOI 10.1186/s12911-016-0353-9.
8. Moore C, Wisnivesky J, Williams S, McGinn T. Medical errors related to discontinuity of care from an inpatient to an outpatient setting. J Gen Intern Med. 2003. PMID 12911647. DOI 10.1046/j.1525-1497.2003.20722.x.
9. Tamblyn R, Abrahamowicz M, Buckeridge DL, et al. Effect of an Electronic Medication Reconciliation Intervention on Adverse Drug Events: A Cluster Randomized Trial. JAMA Netw Open. 2019. PMID 31539073. DOI 10.1001/jamanetworkopen.2019.10756.
10. Yemm R, Bhattacharya D, Wright D, Poland F. What constitutes a high quality discharge summary? A comparison between the views of secondary and primary care doctors. Int J Med Educ. 2014. PMID 25341221. DOI 10.5116/ijme.538b.3c2e.
11. Zimolzak AJ, Shahid U, Giardina TD, et al. Why Test Results Are Still Getting "Lost" to Follow-up: a Qualitative Study of Implementation Gaps. J Gen Intern Med. 2022. PMID 33907982. DOI 10.1007/s11606-021-06772-y.