• We treat communication as a soft skill. Data show it is a safety system with failure points at handoffs, discharges, and language mismatches.¹
• Design the critical conversations. Surgical timeouts, structured handoffs, teach-back at discharge, and reliable escalation paths help reduce harm and confusion.⁴
• Bridge literacy and language gaps. Professional interpreters, plain language, pictograms, and shared decision aids improve comprehension and adherence.⁵
• Align transparency with continuous learning. Communication-and-resolution programs and standard cultural competence frameworks build trust and safer follow-through.

When communication is clear, care flows. When it is not, little snags can become big problems. A patient can leave the hospital confident, or return in 72 hours because a dosing instruction was unclear. A discharge phone call can prevent confusion, or arrive after pills have been taken the wrong way. A meta-analysis of discharge communication found lower readmissions and higher adherence when teams used structured approaches.⁷ Health literacy tools make that structure concrete. Teach-back turns a monologue into a loop by asking patients to explain the plan in their own words.⁴ Patient decision aids translate risks and benefits so that choices align with patients’ values.⁵

Language access is essential. Studies across settings show that trained professional interpreters and bilingual clinicians raise quality and reduce errors for patients with limited English proficiency. When communication breaks down inside the care team, incident reports illuminate the causes, from omitted information to unclear goals. None of this is news on the ground. Clinicians work through interruptions and heavy cognitive load. Patients juggle fear, unfamiliar terms, and complex instructions. The opportunity is to focus on a short list of critical conversations and redesign them using methods that are simple, easy to teach, and trackable.

Healthcare runs on handoffs. Shift changes, consults, admissions to the floor, transfers to the operating room (OR), weekend coverage, and discharge to home all require accurate information transfer and a shared mental model. The reality of many units feels different. Teams move fast, interruptions are constant, and handoffs vary by person and by day. Root-cause patterns recur: missing allergies or pending tests, ambiguous plans, and similar-sounding names in a busy ward. 

At the bedside, nurses and physicians often lack a single, shared template. Situation, Background, Assessment, and Recommendation (SBAR) was created to give every conversation a structure that works under time pressure. The evidence base shows moderate gains in safety when SBAR is used consistently, especially for phone communications and escalations.⁸ Simulation-based training helps students and new clinicians internalize the cadence before encountering real-world risks.⁹ Teams that run SBAR drills report more concise calls, quicker responses, and fewer missed cues. Still, uptake drifts when leadership attention wanes or when new staff miss the initial training.

Team culture matters as much as the tool. TeamSTEPPS bundles closed-loop communication, briefs, huddles, and debriefs to ensure that information is heard, confirmed, and acted upon. Studies among early-career nurses and across service lines show improved teamwork climate and communication, with signals on patient outcomes when programs are maintained.¹⁰ Still, many hospitals conduct one-time trainings without integrating these habits into the electronic health record (EHR) and daily workflow. A huddle that depends on one champion often fades once that person rotates.

Digital channels add a new twist. Secure messaging improves reach, yet high message volume is now linked to higher odds of wrong-patient ordering errors.¹¹ The problem is the total cognitive load at peak times and the ease of firing off a quick request without a shared patient context. Inboxes overflow, attention splits, and identity checks get sloppy. When clinicians cover multiple teams or when patient names are similar, the risk increases.

Patients and families see the effects. They sit through shift changes hearing language they do not understand. They leave with a stack of papers and a pharmacy bag that doesn’t match what they thought they heard. They try to reach someone with a question and end up in a phone tree. They return to the emergency department (ED) with a preventable problem. Clinicians notice it too. They write careful notes and still field panicked calls because the plan was not explained in a way that stuck with the patient.

Under stress, humans default to heuristics. Without prompts and visible checklists, busy teams skip steps, or they complete them in their heads without saying them aloud. A simple timeout that forces each person to state their role and the critical details takes seconds and saves lives in the OR.¹² The same principle applies to medication reconciliation and discharge. If there is no teach-back and no matching of dosing tools to instructions, the risk of error remains high.¹³

Challenge 1 is predictability. The same failure modes appear across wards and across institutions. The fix is to decide which conversations are mission-critical, and then make their safe version the norm every time, for every patient, with scripting and visual cues that fit the reality of a busy shift. That requires standard work, prompts inside the record, and measurement that focuses leaders on fidelity. When this focus is sustained, handoffs, escalations, and discharge can feel different within a single quarter.

Pick the handful of moments where communication changes outcomes, then standardize them so teams can execute under pressure. Start with three priorities.

First, the operating room. The World Health Organization (WHO) Surgical Safety Checklist turned a ritual into a lifesaving habit. Teams pause to confirm identity, site, and plan, and to anticipate risks together. Complications and mortality dropped when teams used the checklist with fidelity, and that effect has been replicated in diverse settings.¹² The lesson is portable: run a micro-timeout before high-risk procedures outside the OR, for example bedside line placements, conscious sedation, or thrombolysis decisions. The habit is the point. Everyone speaks, critical details are said out loud, and the team hears the same plan.

Second, the transition from hospital to home. Teach-back works when it is specific and paired with the right tools. In a randomized ED trial, teach-back for low health literacy patients reduced decisional conflict and improved understanding.¹³ Build a short discharge script that covers the diagnosis in plain language, the top three actions at home, the exact dosing with matching tools, and a safety net for what to do if things change. Place a visible checklist in the EHR so nurses and clinicians can see the steps and document completion with a single click—no hunting through tabs. Reinforce with a next-day call that targets the biggest failure points.

Third, identity and escalation. Use the wrong-patient Retract-and-Reorder measure to track and reduce wrong-patient orders. It’s a practical way to surface meaningful signals in EHR data and pinpoint where identity checks fail.¹⁴ Combine this with a standard escalation script that names the patient and the concern, states the recommendation, and requests a specific timeframe for response. Train for respectful assertiveness so junior staff can escalate without friction across hierarchies.

Embed these practices into daily habits. Morning briefs set the plan. Mid-shift huddles catch risks as they emerge. End-of-day debriefs capture one thing to keep and one thing to change. TeamSTEPPS skills help turn this into muscle memory for new staff, while senior clinicians model brevity, clarity, and closed loops.¹⁰ Place laminated cards at workstations, and embed one-click templates in the EHR that prefill SBAR fields into a message or progress note. The goal is to reduce effort at the moment effort feels hardest.

Patients are partners in this system. Invite a family member to the bedside discharge discussion if the patient agrees. Use a single sheet with pictograms and milliliter-only dosing for pediatric medications, along with a syringe that matches the volume. That simple change significantly reduces major dosing errors in trials.¹³ Align pharmacy labeling with the instructions patients heard in the room. Put the callback number in large font on the first page and staff the line with people who can solve problems.

Measure what you standardize. Unit leaders can post weekly run charts for handoff fidelity, percent of discharges with teach-back, and wrong-patient order events. Celebrate consistent performance, even when it feels routine. Safety improves when the right words are said in the same reliable way, every time.

Many patients leave visits smiling and uncertain. They don’t want to appear difficult. They worry about slowing the line. They plan to call later and do not. Parents in pediatrics face even harder tasks: deciphering dosing, juggling school and work, and navigating portals that do not match their reading level. Adults in every specialty meet new terms at speed. The result is silent confusion that turns into missed doses, double doses, and preventable returns.

Language is the first barrier. Patients with limited English proficiency fare worse when ad hoc interpreters stand in for professionals. Reviews show that trained professional interpreters and bilingual clinicians improve satisfaction, quality, and outcomes.¹⁵ Still, many hospitals rely on family members, untrained staff, or phone lines that are difficult to access during night shifts. Clinical teams often hesitate to call because they fear slowing down. That small delay is outweighed by the reduction in errors and the increase in trust when patients feel heard in their own language.

Health literacy is the second barrier. Well-meaning instructions can confuse even highly educated patients, especially under stress. Discharge education designed for busy caregivers needs pictures, chunked information, and a quick loop to confirm understanding. A randomized clinical trial in pediatrics showed that a literacy-informed bundle with pictogram-based instructions, show-back, and structured teach-back reduced dosing errors and improved knowledge.¹⁶ Dosing tools matter, too. In trials, parents made fewer mistakes when given a syringe that matched the prescribed volume and when labels use milliliters only.¹⁷

Information quality inside the record is the third barrier. Patients who read their notes spot mistakes. When those errors involve medications or diagnoses, the risk of harm rises. Surveys of patients reading open notes found frequent errors, highlighting a communication opportunity.¹⁸ Inviting patients to flag and fix mistakes helps every downstream conversation, since the next clinician starts from a cleaner record.

Workload and channel overload amplify these barriers. Clinicians toggle between the bedside, the EHR, pagers, and secure messages. Recent research links higher secure messaging volume to increased wrong-patient ordering errors during inpatient care.¹¹ That is a signal to slow the message flood, tighten identity checks, and move complex decisions back to real-time conversations. Patients struggle with channel overload too: portals, printed instructions, and automated calls often give conflicting answers. Families pick the closest one and hope it’s right.

Cultural context shapes communication, too. A serious diagnosis lands differently across families and traditions. Some want direct probabilities; others need time and space before numbers can land. The National Culturally and Linguistically Appropriate Services (CLAS) Standards exist for a reason: they ask organizations to deliver care that is effective, equitable, understandable, and respectful by default.¹⁹ Yet many teams have never seen these standards made concrete on their unit.

Finally, incentives are often misaligned. Throughput metrics reward speed over clarity, and documentation requirements reward volume over usefulness. Staff who take two extra minutes to run a teach-back or call an interpreter often do so on personal conviction. Challenge 2 is to treat comprehension as a clinical outcome and to make the right move the easy move at the point of care.

Start with a simple rule: if a patient’s preferred language is not English, use a trained professional interpreter or a bilingual clinician. The evidence is robust and spans settings, specialties, and outcomes.⁶ Normalize interpreter use by removing friction. Put a quick-dial interpreter button on every phone. Equip rounding teams with dual-handset mobiles. Build EHR prompts that default to documenting interpreter use whenever language preference is set to non-English. Recognize and celebrate units that achieve high interpreter-use rates for key encounters, such as consent, discharge, goals of care, and new medications.

Make health literacy visible. Train every clinician and nurse in plain-language techniques, and then verify understanding with teach-back. The method is straightforward: explain the plan in short chunks, ask the patient to teach it back in their own words, clarify any gaps, and repeat for the top three actions. A randomized ED study involving patients with low health literacy showed that teach-back improves understanding at discharge.¹³ A pediatric trial added pictograms, dosing demonstrations, and show-back, reducing medication errors and improving knowledge.¹⁶ Build these elements into your standard discharge packet. Use pictures for timing and amounts. Use milliliters only. Provide a syringe that matches the dose.

Decision quality improves when patients can clearly see their options. Patient decision aids turn risk into visuals and values into choices, leading to more informed, values-congruent decisions without increasing anxiety. The updated Cochrane review covers hundreds of studies and shows consistent gains in knowledge and clarity.⁵ Equip clinics with one or two high-yield decision aids per specialty, such as imaging for low back pain, options for early-stage breast cancer, or anticoagulation for atrial fibrillation. Store them in the EHR so they’re a click away during the visit and available on the portal.

Clean the record with patients. Invite patients to review their notes and medication lists, then make it easy to submit corrections. Evidence shows patients find real errors, and many are safety relevant.¹⁸ Equip staff to fix errors the same day. Publish a simple “How to Fix Your Record” box on the portal landing page, and print it on the discharge sheet.

Tame channel overload. Reserve secure messaging for simple, low-risk requests. Route complex decisions to scheduled calls or visits with the right people present. Pair messaging limits with identity safeguards: use two-patient identifiers in message templates and embed a required field that pulls the patient banner into the thread. Monitor wrong-patient orders using the Retract-and-Reorder measure to see if safeguards are working.¹⁴

Support culture and equity with a clear standard. The National CLAS Standards offer a blueprint for culturally and linguistically appropriate services.¹⁹ Translate these high-level standards into concrete, unit-level behaviors. For example, require documentation of language preference at registration. Track interpreter use for key encounters. Provide community health workers who can reinforce instructions at home in the preferred language. Equip front desks with clear signage about interpreter availability.

Measure what matters. Post monthly data on interpreter use for key encounters, percentage of discharges with teach-back, dosing tool matching for pediatric medications, and correction cycle time for patient-reported record errors. Leaders who treat comprehension as a safety metric will reduce returns, decrease errors, and build stronger trust.

When something goes wrong, words matter. Patients deserve to hear what happened, why it happened, what will be done to address the consequences, and how the organization will prevent a recurrence. Many clinicians want this too. They worry about blame, legal exposure, and career risk. They are often unsure who should speak, what to say, and how to support families over time. Without a clear approach, conversations happen late, inconsistently, or not at all. Trust erodes. Learning stalls.

Communication-and-resolution programs (CRPs) were created to solve this problem. The model is simple. Disclose. Explain. Apologize when appropriate. Investigate with rigor. Share findings. Offer fair resolution. Improve the system. Health systems and liability insurers have implemented CRPs with encouraging results, yet real-world fidelity varies widely.²⁰ Some programs focus on claims management and lose the patient-safety engine. Others underinvest in training leaders to handle hard conversations. A strong CRP makes the transparent path the default and supports staff and patients through it.

Why include this in a piece on everyday communication? Because post-event communication sets the tone for every interaction that follows. Families decide whether to continue care with the same health system based on how that first conversation feels. Staff decide whether to report near-misses based on whether past events led to fair learning or quiet blame. Leaders decide whether to fund training based on what they track. If CRP processes are fuzzy, families hear mixed messages, and teams learn the wrong lessons.

Communication after harm requires cultural competence too. Serious conversations happen within the frame of a family’s language, beliefs, and expectations. The National CLAS Standards ask organizations to make communication understandable and respectful by default.¹⁹ That includes offering interpreters for post-event meetings, preparing materials in the preferred language, and honoring the role of elders or faith leaders when families request it.

Ethical clarity helps clinicians. A guided script provides the sequence. Acknowledge the event. Offer empathy without hedging. Explain the known facts. State what will happen next and when. Invite questions. Schedule a follow-up by default. Document thoroughly and share the plan across the team to keep messaging consistent. Provide support for the clinicians involved. They are often affected too, and unsupported staff communicate poorly.

Systems need feedback loops. Organizations with strong CRPs connect event analysis to frontline training. If a delayed handoff contributed, the next month’s safety brief focuses on handoff reliability. If a dosing instruction confused multiple families, discharge templates change within weeks. If wrong-patient orders spike on a ward, leaders reinforce identity checks and reduce message overload. Accountability rests with the system, not with individuals acting alone.

Families notice when the story changes from week to week. That happens when investigations stall or when units lack a shared narrative. Set a standard timeline for key steps, with named roles. Choose a single point of contact for the family, then keep appointments and call back when promised. A clear communication plan reduces anxiety for everyone.

Transparency does not weaken defense against frivolous claims. Programs that communicate openly and fairly report lower costs and faster resolution than business-as-usual as well as better relationships and learning.²⁰ The point is not litigation strategy. The point is humane care and reliable improvement after harm.

Make communication after harm a core clinical skill. Train leaders, attendings, residents, nurses, and risk teams together using real cases. Use a short script, practice in simulation, then coach on the job. Provide a pocket card with the sequence for an initial conversation and a support contact number. Add a simple checklist to the EHR that reminds clinicians to invite an interpreter when needed, schedule follow-ups, and document agreements.

Adopt or strengthen a CRP with meaningful governance. Publish the policy in plain language for staff and families. Define what constitutes an event that triggers the CRP. Name the people who lead the response and the timeline for each step. Include patient and family advisors in the design and review of cases where appropriate. Measure timeliness, completion of steps, satisfaction, and safety actions taken after reviews. Share aggregate findings at grand rounds and town halls so learning spreads.

Connect CRP actions to daily communication habits. If an event involves a handoff failure, refresh SBAR use across the unit with a brief huddle script and visible prompts. If the event involves a confusing consent, add a short decision aid and teach-back to that consent flow. If the event involves language mismatch, track interpreter use for similar encounters and aim for near-universal use. Momentum builds when post-event learning changes the next day’s practice.

Bring cultural competence to the center. Use the CLAS Standards as a unit-level checklist.¹⁹ Confirm language preference at registration. Offer interpreters for every high-stakes conversation. Prepare materials that match the family’s reading level and language. Identify trusted community partners who can support understanding outside the hospital. Equip social workers and community health workers to reinforce plans in homes and clinics. These steps honor families and reduce miscommunication in the moments that count.

Be transparent with data. Post deidentified run charts on event response times, completion of family meetings within set windows, and the rate of safety actions taken after reviews. Celebrate improvements and explain setbacks. Leaders who make progress visible build trust and invite staff into solutions.

Anchor everything in respect. Families want to hear the truth, to know what will change, and to feel seen. Clinicians want to learn without fear. A strong communication culture meets both needs. CRPs provide a framework.²⁰ CLAS provides a compass. When teams use both, patients experience care that feels consistent and humane, even when things go wrong.

Communication tools require fidelity and context. SBAR and checklists help only when teams actually use them and when leaders defend the time to do so.¹² Secure messaging is valuable, and overload increases the wrong-patient risk without identity safeguards and channel discipline.¹¹ Teach-back is effective, but rushed use that skips comprehension checks reduces benefit.¹³ Interpreters improve care, but partial implementation leads to false confidence.⁶ Decision aids improve knowledge and values alignment, and they work best when clinicians engage with them during the visit.⁵ CRPs foster trust and learning when leadership treats them as mission-critical.²⁰ The theme is consistent execution and measurement at the points where miscommunication causes harm.

Healthcare communication improves when we focus on a few critical conversations and make them reliable. Handoffs, escalations, timeouts, discharge, consent, and post-event conversations change outcomes when done with structure, prompts, and respect. The challenges are clear. Handoffs drift under workload. Literacy and language gaps often go unnoticed. Transparency after harm is uneven. The opportunities are practical. Standardize the steps, fit them into the day, and measure fidelity.

The actions in this piece apply across care settings. An OR timeout breeds the same habit of shared understanding that makes a bedside procedure safer. A teach-back loop in the ED builds the muscle that prevents an outpatient dosing error. A respectful post-event conversation strengthens trust and invites families to keep engaging with care. Leaders can set a tone where communication is treated as clinical work with protocols, coaching, and feedback.

Start with one unit and one conversation. Make the safer version easy to do and hard to skip. Post the metrics. Thank the people who make it work on the busiest days. The benefits reach patients and staff immediately. Fewer avoidable returns. Fewer wrong-patient errors. More confidence at the bedside. When teams speak the same language and families truly understand the plan, outcomes improve and the work feels better.

Increasing Digital Health Literacy

Practical strategies to make patient-facing tools clearer and easier to use, with takeaways you can integrate into portals, consent flows, and discharge materials to boost understanding and follow-through.

How Social Norms Reduced Missed Hospital Appointments by 31.7%

A short case study on message design in the UK National Health Service (NHS) that cut no-shows using simple descriptive norms—a useful example for crafting reminders and escalation scripts that keep identities clear and appointments on track.

1. The Joint Commission. (2024, August 22). Reducing handoff communication failures and inequities in healthcare. https://www.jointcommission.org/en-us/knowledge-library/news/2024-08-reducing-handoff-communication-failures-and-inequities-in-healthcare
2. Umberfield, E., Ghaferi, A. A., Krein, S. L., & Manojlovich, M. (2019). Using incident reports to assess communication failures and patient outcomes. Joint Commission Journal on Quality and Patient Safety, 45(6), 406–413. https://doi.org/10.1016/j.jcjq.2019.02.006
3. Agency for Healthcare Research and Quality. (2023). TeamSTEPPS 3.0: Team strategies and tools to enhance performance and patient safety. https://www.ahrq.gov/teamstepps-program/index.html
4. Agency for Healthcare Research and Quality. (2015). Use the Teach-Back Method. (Health Literacy Universal Precautions Toolkit, 2nd ed., Tool 5) [PDF]. U.S. Department of Health and Human Services. https://www.ahrq.gov/sites/default/files/wysiwyg/professionals/quality-patient-safety/quality-resources/tools/literacy-toolkit/healthlittoolkit2_tool5.pdf
5. Stacey, D., Lewis, K. B., Smith, M., Carley, M., Volk, R., Douglas, E. E., Pacheco-Brousseau, L., Finderup, J., Gunderson, J., Barry, M. J., Bennett, C. L., Bravo, P., Steffensen, K., Gogovor, A., Graham, I. D., Kelly, S. E., Légaré, F., Sondergaard, H., Thomson, R., Trenaman, L. & Trevena, L. (2024). Decision aids for people facing health treatment or screening decisions. The Cochrane Database of Systematic Reviews, 1(1), CD001431. https://doi.org/10.1002/14651858.CD001431.pub6
6. Flores G. (2005). The impact of medical interpreter services on the quality of health care: a systematic review. Medical Care Research and Review, 62(3), 255–299. https://doi.org/10.1177/1077558705275416
7. Becker, C., Gouskova, N., Choma, N. N., Newberry, S., Ritchie, C., & Sarkisian, C. (2021). Interventions to improve communication at hospital discharge: A systematic review and meta-analysis. JAMA Network Open, 4(8), e2122107. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2783547
8. Müller, M., Jürgens, J., Redaèlli, M., Klingberg, K., Hautz, W. E., & Stock, S. (2018). Impact of the communication and patient hand-off tool SBAR on patient safety: a systematic review. BMJ Open, 8(8), e022202. https://doi.org/10.1136/bmjopen-2018-022202
9. Yun, J., Lee, Y. J., Kang, K., & Park, J. (2023). Effectiveness of SBAR-based simulation programs for nursing students: a systematic review. BMC Medical Education, 23(1), 507. https://doi.org/10.1186/s12909-023-04495-8
10. Hassan, A. E., Mahmoud, M. H., & Mohamed, A. A. (2024). Evaluating the effect of TeamSTEPPS on teamwork perceptions and patient safety culture among newly graduated nurses. BMC Nursing, 23, 170. https://bmcnurs.biomedcentral.com/articles/10.1186/s12912-024-01850-y
11. Lou, S. S., Heo, H., Lam, M. B., et al. (2024). Secure messaging use and wrong-patient ordering errors among inpatient clinicians. JAMA Network Open, 7(12), e245115. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2827227
12. Haynes, A. B., Weiser, T. G., Berry, W. R., et al. (2009). A surgical safety checklist to reduce morbidity and mortality in a global population. New England Journal of Medicine, 360(5), 491–499. https://doi.org/10.1056/NEJMsa0810119
13. Griffey, R. T., Shin, N., Jones, S., Aginam, N., Gross, M., Kinsella, Y., Williams, J. A., Carpenter, C. R., Goodman, M., & Kaphingst, K. A. (2015). The impact of teach-back on comprehension of discharge instructions and satisfaction among emergency patients with limited health literacy: A randomized, controlled study. Journal of Communication in Healthcare, 8(1), 10–21. https://doi.org/10.1179/1753807615Y.0000000001
14. Agency for Healthcare Research and Quality. (2023). Automated Retract-and-Reorder measures to improve medication safety. https://digital.ahrq.gov/program-overview/research-stories/automated-retract-and-reorder-measures-improve-medication-safety
15. Karliner, L. S., Jacobs, E. A., Chen, A. H., & Mutha, S. (2007). Do professional interpreters improve clinical care for patients with limited English proficiency? A systematic review of the literature. Health services research, 42(2), 727–754. https://doi.org/10.1111/j.1475-6773.2006.00629.x
16. Carroll, A. R., Dempsey, C., Gal, T. S., et al. (2024). Health literacy–informed communication to reduce pediatric discharge medication dosing errors: A randomized clinical trial. JAMA Network Open, 7(1), e2351822. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2813929
17. Yin, H. S., Parker, R. M., Sanders, L. M., et al. (2017). Pictograms, units and dosing tools, and parent medication errors: A randomized study. Pediatrics, 140(1), e20163237. https://publications.aap.org/pediatrics/article/140/1/e20163237/37975
18. Bell, S. K., Delbanco, T., Elmore, J. G., Fitzgerald, P. S., Fossa, A., Harcourt, K., Leveille, S. G., Payne, T. H., Stametz, R. A., Walker, J., & DesRoches, C. M. (2020). Frequency and Types of Patient-Reported Errors in Electronic Health Record Ambulatory Care Notes. JAMA Network Open, 3(6), e205867. https://doi.org/10.1001/jamanetworkopen.2020.5867
19. U.S. Department of Health and Human Services, Office of Minority Health. (2013). National Standards for Culturally and Linguistically Appropriate Services in Health and Health Care. https://thinkculturalhealth.hhs.gov/clas/standards
20. Mello, M. M., Kachalia, A., Roche, S., et al. (2014). Communication-and-resolution programs: The challenges and lessons learned from six early adopters. Health Affairs, 33(1), 20–29.https://www.healthaffairs.org/doi/10.1377/hlthaff.2013.0828