How to follow your doctor's plan

The plan usually arrives as an after-visit summary. CMS defines the clinical summary as an after-visit summary that provides a patient with relevant and actionable information and instructions, and its list of contents runs from the reason for the visit and the medication list to clinical instructions, future appointments, referrals, and tests still pending (CMS, Clinical Summaries, issued November 2014). Under that federal program, a practice met the measure by giving the summary within one business day for more than 50 percent of office visits (CMS, November 2014), so some visits end without one and it is fair to ask for it.

ONC describes the purpose in one line: easy access to your health records lets you monitor your health conditions better, understand and stay on track with treatment plans, and find and fix errors in your record (ONC, Patient Access Information for Individuals, last updated January 14, 2026). You also have the right to see your health records, get a copy, and have it sent to a specialist or another party you choose (ONC, January 14, 2026).

What the summary holds and how to sort its lines into tasks has its own page, the after-visit summary checklist. Getting ready for the visit itself is covered in the appointment preparation checklist.

The WHO defines adherence as the extent to which a person's behaviour (taking medication, following a diet, and/or executing lifestyle changes) corresponds with agreed recommendations from a health care provider (WHO, Adherence to Long-Term Therapies, 2003). Its headline figure is that adherence to long-term therapy for chronic illnesses in developed countries averages 50%, and that in developing countries the rates are even lower (WHO, 2003). The same report says increasing the effectiveness of adherence interventions may have a far greater impact on the health of the population than any improvement in specific medical treatments (WHO, 2003).

The CDC's Grand Rounds on medication adherence gives two figures for the steps after a prescription is written: approximately one in five new prescriptions are never filled, and among those filled, approximately 50% are taken incorrectly (CDC, MMWR, November 17, 2017). It puts the direct health care costs of nonadherence at approximately $100 billion to $300 billion (CDC, 2017).

The WHO also refuses to put the whole problem on the patient. It sorts the causes into five dimensions (social and economic factors, the health care team and system, the condition, the therapy, and the patient) and says the traditional misconception is that adherence is a patient-driven problem (WHO, 2003).

Five points show up in the research, each with its own owner.

• The words. Between 40% and 80% of medical information provided by healthcare practitioners is forgotten immediately, the more information presented the lower the share recalled, and almost half of what is remembered is incorrect (Kessels, J R Soc Med, 2003).
• The prescription. Of 195,930 electronic prescriptions in one community study, 78% were filled, and of the 82,245 written for new medications, 72% were filled (Fischer et al., J Gen Intern Med, 2010). New prescriptions for chronic conditions went unfilled most often, at 28.4% for hypertension, 28.2% for high cholesterol, and 31.4% for diabetes (Fischer et al., 2010). The CDC's rounded figure for the same step is one in five (CDC, 2017).
• The referral. Studies found that 25 to 50 percent of referring physicians did not know whether their patients had actually seen the specialist they were referred to (Mehrotra, Forrest and Lin, Milbank Quarterly, 2011). The review also found that primary care doctors often do not know what the specialist recommended (Mehrotra et al., 2011).
• The result. A systematic review of 19 studies found tests not followed up at rates from 6.8% to 62% for laboratory tests and from 1.0% to 35.7% for radiology, with missed cancer diagnoses among the consequences (Callen et al., J Gen Intern Med, 2012). All 19 were American studies, and the authors list the role of the patient among the parts of any fix (Callen et al., 2012).
• The dose. Among prescriptions that do get filled, approximately 50% are taken incorrectly with regard to timing, dosage, frequency, and duration (CDC, 2017).

Teach-back is the most specific published fix for the forgetting problem. AHRQ defines it as a way of checking understanding by asking patients to state in their own words what they need to know or do about their health, and says it can improve patient understanding and adherence and decrease call backs and cancelled appointments (AHRQ, Health Literacy Universal Precautions Toolkit, Tool 5, last reviewed April 2024).

AHRQ writes it for clinicians, and the rules transfer to anyone holding a plan. Questions that can be answered with yes or no, like "Do you understand?", are not teach-back, because patients are likely to answer yes whether they understand or not (AHRQ, April 2024). Information goes in small pieces with a check after each one, which AHRQ calls chunk and check (AHRQ, April 2024). Patients can look at handouts during a teach-back as long as they use their own words, and when it uncovers a misunderstanding the clinician explains again a different way and asks again (AHRQ, April 2024). One of AHRQ's sample questions is the one to answer before you leave the room: "When you get home, what will you tell your partner about this visit?"

The evidence is positive and uneven. A systematic review of 12 studies found teach-back improved disease-specific knowledge, adherence, self-efficacy, and inhaler technique, with a positive but inconsistent trend for self-care and readmissions, and limited evidence on quality of life and on how long knowledge is retained (Ha Dinh et al., JBI Database of Systematic Reviews, 2016).

Kessels names spoken advice as a weak way to deliver a plan: in most instances medical advice is spoken, and that is not a very successful method, while written information is better remembered and leads to better treatment adherence (Kessels, 2003). The same review says specific instructions beat general ones, and its example is that "You must take the next two weeks off" is recalled more accurately than "You need to get some rest" (Kessels, 2003). Instructions placed first are remembered better, and telling the patient the categories in advance (what is wrong, what tests come next, what to expect, the treatment, and what the patient must do) also improves recall (Kessels, 2003).

AHRQ's patient sheet puts the same point as an instruction: find out what to do next and ask for written instructions, brochures, videos, or web sites, and if you do not understand your instructions after you get home, call your doctor (AHRQ, Be More Engaged in Your Healthcare, last reviewed November 2024). AHRQ's medicine tool shows why precise wording matters, noting that "Take 1 pill in the morning and 1 pill at bedtime" is unambiguous, whereas "Take twice daily" could lead to errors (AHRQ, Tool 16, April 2024).

AHRQ asks clinics to give patients a list of medicines that includes all prescription and over-the-counter medicines as well as supplements, organized so it is easy to see when to take each one over the course of a day (AHRQ, Tool 16, April 2024). MedlinePlus asks the patient to keep the same record going after the visit: continue to keep a record of any symptoms and all of your medicines, and write down any new questions about your health, medicines, or treatment (MedlinePlus, September 15, 2024).

For the medicines themselves, AHRQ's published memory strategies are specific. Tie each dose to something you already do at the same time every day, such as meals or brushing teeth. Use a pill organizer filled from the medicine list. Enlist family members for reminders and pill boxes. Set alarms on a phone, watch, or voice assistant, or use a reminder app (AHRQ, Tool 16, April 2024). The CDC lists pillboxes and blister packs among its strategies for organizing regimens in clear and simple ways (CDC, 2017).

A lifestyle change gets the same treatment through AHRQ's action plan: one specific step that is small and realistic over a short time such as a week, a concrete start date, and a confidence score from 1 to 10 (AHRQ, Tool 15, April 2024). AHRQ reports that a confidence level of 7 or above increases the likelihood that the patient will carry out the plan, and that below 7 the plan should be revised until it feels doable (AHRQ, Tool 15, April 2024).

MedlinePlus lists what a provider may want after the visit (seeing a specialist, having a test, taking a new medicine, scheduling more visits) and says for the best results to take medicines as directed and go to any follow-up appointments (MedlinePlus, September 15, 2024). AHRQ's version for the patient is to make appointments to have tests done or see a specialist if you need to, and to call your doctor's office to find out test results and ask what you should do about them (AHRQ, November 2024).

The referral and result studies above are the reason to treat every pending item as open until you hear back. Primary care doctors often do not know whether a patient went to the specialist (Mehrotra et al., 2011), and failure to follow up test results is common enough that its reviewers call it an important safety concern which requires urgent attention (Callen et al., 2012). AHRQ's action plan tool also sets a time for follow-up and says that when patients have not kept to the plan, the job is to reassure them that this is common and build a plan they can achieve (AHRQ, Tool 15, April 2024).

The medicines have one rule that none of these sources bend: talk with your doctor or pharmacist before you stop taking any medicines that your doctor prescribed (AHRQ, November 2024).

I carry the plan your clinician gave you. Send me the summary, the printout, or the portal text and I keep it, read it back in plain words, and pull out the lines that ask something of you. The medicines, the new habit, the lab, the referral, the follow-up window: those go on one list, and I text you about each one when it's due and ask how it went. When a result was due and nothing came, I can pull up the office's number and hours so you can make the call. Before the next visit, ask me and I read back what you did, what you told me, and the questions you saved, so you walk in with the record in hand.

I never change the plan, explain what a result means, or decide whether a symptom is urgent. When a line is unclear or two instructions disagree, I help you write the question for the office, and the clinician answers it.

No clinician reviewed this page. It restates the CMS, ONC, AHRQ, MedlinePlus, WHO, and CDC documents and the published studies linked below, and those, and your own clinician's instructions, are the documents to trust over anything here.