Reconciling Medications

Safe Practice Recommendations

and

Implementation Strategies

Medication Safety Facts

Medication errors account for more than 7,000 deaths annually

Approx. two out of every 100 patients admitted to the hospital will experience a preventable adverse drug event

Over 12% of patients with an ADE within 2 weeks of discharge

“Reconciling Medications”

A systematic process to reduce the number of medication events occurring at interfaces of careCreating the most complete and accurate list possible of all home medications for each patient and then comparing that list against the physician’s admission, transfer, and/or discharge orders. Discrepancies are brought to the attention of the physician and, if appropriate, changes are made to the orders. Any resulting changes in orders are documented.

RESEARCH:

Errors that are the result of an omission are often not reported as errors, although they may result in an adverse outcome for the patient.

They may manifest themselves as: Unexplained elevated lab values

Due to inaccurate dosing Missed medications

Readmissions due to: Doubling up of medications Missed medications at discharge Contraindication to unknown OTC or herbal meds

Problem identified

Info on patients’ home meds not being systematically collected; in multiple places in the chart, often incomplete

Poor or inadequate processes to compare list of pre-admit medications to orders

Research study demonstrated that over half of all hospital medication errors occurred at the interfaces of care[Rozich, Resar 2001]

Medication errors based on chart review

56%

21%

14%

9%

Interface Errors

Drug Distribution orDocumentation ErrorsAllergy Documentation

Others

Source: Luther Midelfort Hospital -- Mayo Health System chart review

“We found that the list of medications that details current drug use was either nonexistent or wrong more than 85% of the time” [Rozich/Resar 2004, p.8]

Examples of errors

No orders for needed home meds Missed or duplicate doses from inadequate

records of frequency/last administration time Surgeon inadequately addressing meds for

chronic conditions Failure to restart meds at transfers Doubling up (brand/generic combinations,

formulary substitutions)

Unintended medication discrepancies at admission

Studies show over half of patients have discrepancies between home medications and medications ordered at admission, many with potentially serious results 54% of patients; 39% potentially serious [Cornish Arch Intern Med 2005] More than half; 59% could have caused harm if the error continued after discharge [Gleason Am Jnrl H-Sys Pharm 2004] 

More evidence on impact:Johns Hopkins Surgical ICU

Dramatic reduction in medication errors resulted from reconciling: Baseline: 31 of 33 (94%) of patients with MD

changing orders when discrepancies brought to their attention

By week 24, nearly all medication errors in discharge orders eliminated

As a result of routine reconciling, average of 10 orders per week are changed

[Pronovost, 2003]

The

Reconciling Process

THE PROPOSED SOLUTION

A process to obtain the best home medication list possible through a defined resource list and active review of the patient’s medical history.

Patient Pharmacy Family Patient’s Med List PCP VNA

Utilize strategic interviewing practices.

Ask open ended questions to obtain info on OTC meds & herbals.

Reconciling process: admissionGetting the home med list (at intake)

Interviewing strategies to promote accuracy Input from patient/family/alternative sources Outreach: patients arrive with accurate list

Writing medication orders Goal: work from accurate home med list

Identify and reconcile discrepanciesOrder (no omissions, no duplicates, right med/dose/

frequency/route)

Communicate (to next level of service)

BECOMING A STATEWIDE INITIATIVE The Massachusetts Hospital Association in

collaboration with the Massachusetts Coalition for the Prevention of Medical Errors reviewed evidence of medication reconciliation to determine:

Importance – How much can we impact safety? Feasibility – Is this a doable process? Measurability – Can we monitor our progress?

Statewide advisory board voted to accept this initiative!

Getting started

1) Initiate leadership dialog – resource commitment, regular reporting channels

2) Form a multidisciplinary team

3) Risk assessment/baseline measurement

4) Aim statement, timeline

5) Pick pilot unit

6) Begin testing

Define Aim / Obtain Baseline Measure

Aim: To reduce the rate of unreconciled medications

at admission by 50 % within 9 months. Measure:

Baseline measurement of 20 charts, subsequent measures performed on 30 charts per month for the first 3 months after implementation of form. Evaluate the frequency of the measure after the first three months.

1. Getting the home med list

What have we learned? Adopt standardized form Share responsibilities, ordering prescriber

accountable... crew resource management principles

Validate with the patient Don’t let perfection be the enemy of the good

1. Getting the home med list

Who? Shared responsibilities, always someone with sufficient expertise:

RN who completes the initial admission history Pharmacist/pharmacist technician

[Michels/Meisel 2003; Gleason/Groszek 2004]

MD if reconciling form not complete when ready to write orders

1. Getting the home med list

What? Current home meds Include OTCs & herbals Dose, frequency, time of last dose Optional: route, source of information, compliance,

purpose Many building collection of patient allergies into the

process

2. Using home list when writing orders

What have we learned? Make highly visible Provide access at point when orders are written Have reconciling form serve as an order sheet.

benefits and issues...

Project phasing

Pilot testing: identify changes, measure to know if the changes are an improvement

Implementation: take a successful change and build it into the way the entire pilot population/pilot unit does their work

Spread: replicating a change/package of changes beyond the pilot unit into other parts of the organization

Maintain the gains

3. Identifying, reconciling discrepancies

Who? Generally nursing assigned responsibility of

comparing the home list to the admit orders, identifying variances, and reconciling all differences

Pharmacist involvement can be productive, especially for organizations with decentralized pharmacy

Need strategy for handing off any unresolved differences at shift change

Implementation Strategies

Resource requirements

During testing/implementation phase Make explicit allocation for those with patient care

responsibilities Managers need to pay attention to workloads; don’t

assign tests to someone overloaded Ongoing

Build into regular workflows Collecting home history IS time consuming; some

have added resources to support that (e.g. pharmacy techs)

Post Team Members- Encourage Input

Contact any of the following Medication Reconciliation PI Team members to answer any of your questions:

Melissa Bartick, MD - X9335 Jennifer Fexis, Quality - X9406 Darlene Civita, RN ICU- X9350 Vicky Casto, RN ACU - X9335 Deb Wilkinson, RPh - X9363

Tips for engaging MDs

Personal appeals from VP of Medical Affairs and/or Chiefs of services

Trial with key leaders on each unit; get their input via “hallway consultations” not meetings

Identified “Ambassadors” from engaged hospitalists; they then educated others

Developed into CME risk program MDs from key committees (P&T, Medical Records)

Chief Medical Resident on the team, with responsibility to report back to other residents

Baseline risk assessment

Chart review Institution-wide

Mini-FMEA, flow charting existing processes

Do in conjunction with initial tests of change

Just-enough measurement/analysis Don’t get bogged down here!!

Mission

Every patient will receive all medications they have been taking at home unless they are held/discontinued by their caregiver(s) and all new medications as ordered -- correct drug, dose, route, and schedule.

The goal of reconciling is to design a process that will ensure the most accurate patient home medication list available, thus reducing the number of medication events upon admission, transfer and discharge

Choosing where to start

Use risk assessment process Willing volunteers At admission logical place

Pros & cons: Med vs Surg units Some success starting @ transfer: ICU,

CCU, telemetry units Probably not ED

Start small, focus on one unit

Small tests... 1 unit, 1 RN, 1 MD, 1 patient Add more staff, more shifts, refining process and form Keep testing on that one unit until you refine the

process and can show that it works (test on all shifts, patients coming in as direct admits, from ED, transfers, etc)

Pilot unit

1) Mini-team including nurse managers, front-line nurses, MD champion

2) Project introduction, staff education

3) Baseline measurement for the unit4) Pick reconciling form to test (steal shamelessly...)

5) Begin testing

Piloting a reconciling form

Testing; avoid forms committees... Simple vs complex

Reconciling status Orders: continue, change, d/c, hold Optional: data sources, purpose/indication,

date/time of last dose, amt of non-compliance Columns for reconciling at discharge? Signature lines

Get support of your CEO; cannot do it without leadership at the top

Use data (to motivate, to know if changes are leading to improvement)

Strong representation from leadership of the 3 key stakeholder groups: MD, RN, pharmacy

Start small

Fundamental ingredients...

Culture...

Core issues of teamwork and communi-cation... organizational culture matters

Changing the way people do work; every time you try to change behavior, it’s only natural to be met with resistance Recognize that this is HARD;

Difficult task: but not impossible Unit briefings/pharmacy rounding

Challenges and barriers

Time and resources “How can we find the time to do this?”

Roles and responsibilities “It’s not my job” “I’m not going to sign that form”

Data collection Need data... but don’t let data collection delay

testing, overwhelm

Medication ChecklistHere’s how patients can help the ‘medication reconciliation’ process:

Keep an updated list of all medications including herbals, vitamins and OTC. Including dosage and reason for taking the drug

Include all allergies and describe reaction Include immunization history Take the list to all doctor visits and medical testing labs, as

well as pre-assessment visit for admission or surgery and all hospital visits including ER

When you leave the hospital, be sure to update your list with new medications and ask if any medications are duplicated

Keep this list in with you at all times

Staff education

Include staff ed rep on your team Create simple template clarifying the steps to be taken to

complete reconciling Lead off with examples of errors from your own hospital Use front line staff from pilot unit to educate staff on

subsequent units Build into orientation, ongoing staff ed Publish your data and progress in your organizations

newsletter

Measurement

Just-enough measurement

Core measure Percent Medications Unreconciled

Orders changed, “great catches”, stories Measures linked to each test, for example:

% patients with reconciling form in chart RN/MD assessments of process

Spread: % patients on units w/ reconciling Context of institution-wide ADE reduction

# Medications Unreconciled(per 100 Admissions)

0

50

100

150

200

250

300

1/5/

98

1/12

/98

1/19

/98

1/26

/98

2/2/

98

2/9/

98

2/16

/98

2/23

/98

3/2/

98

3/9/

98

3/16

/98

3/23

/98

3/30

/98

4/6/

98

4/13

/98

4/20

/98

4/27

/98

5/4/

98

5/11

/98

5/18

/98

5/25

/98

6/1/

98

6/8/

98

6/15

/98

6/22

/98

6/29

/98

7/6/

98

Admission reconciliation

Transferreconciliation

Dischargereconciliation

Luther Midelfort Implementation Impact

Baseline data collection

GOAL: Identify current safety risks

How complete is info on patient’s pre-admission meds? How hard to find? In multiple places?

How often are home meds omitted from admit orders? not re-started after transfer, at discharge? duplicate therapies at discharge?

Example: Why is it Needed?

In a chart review of our admit orders, we found an average of over 4 discrepancies per patient, with omitted medications the most significant error.

Source: University of Kansas Hospital

Terry Rusconi [2003]

Collecting your data

No. Admitting Medications Dose (1)Frequency

(2) Route (3)

Data Source

Medication List Documented on - List all that apply

Do Elements

of List Match?

Y or N or ? (4)

Is discrepa

ncy intentional? Y or No or ?

(5)

Admitting Medication Orders

Dose (6)Frequency

(7)Route (8)

Are Admitting Meds

Addressed By MD? (9)

Y or N or ?

Comments

1

2

3

4

5

6

7

8

9

10

11

12

Total (1) BlanksTotal (2) Blanks P = patient V = VNA 100 - ED sheet Total (9) N or ?Total (3) Blanks F = family N = Nursing home 200 - RN admission Total (6) BlanksNumber of Meds Rx - RX bottle C = Pharmacy 300 - H & PE Total (7) Blanks

H = History 400 - PAT form Total (8) Blanks M = MD office 500 - None Total ordered meds

Lists Documented On:Data Source: Total (5) N or ?

12345678910111213141516171819202122232425262728293031323334353637383940414243444546

A B C D E F G

Chart Review

Total Admission Medications

Total Blanks (1) (2) (3)

Number of Discrepancy's (5)

Number of Admitting Meds Not Addressed by

MD (9)Total Ordered Medications

Total Blanks (6) (7) (8)

1 5 0 2 1 5 02 1 3 1 1 4 03 13 10 2 2 11 34 10 12 0 0 9 05 2 2 0 0 7 06 10 30 0 0 9 07 13 13 2 0 13 38 7 3 0 0 8 09 9 0 3 0 8 0

10 10 14 2 1 10 211 13 4 0 0 19 1912 10 0 0 0 11 013 0 0 8 8 8 014 12 23 4 5 9 015 11 16 10 1 10 1016 15 6 11 0 18 217 10 14 10 7 3 418 9 4 3 1 12 619 11 15 6 10 6 020 2 2 1 1 5 0

Sum 173 171 65 38 185 49Summary Statistics - Admit Orders

# discrepancy errors: Sum of (5) 65# patients: # charts reviewd: 20Discrepancy errors/100 admissions: 325Average error per patient 3.25

Summary Statistics- H & P

# reconciliation errors: Sum of (9) 38# patients: # charts reviewd: 20Reconciling errors/100 admissions: 190Average error per patient 1.9

Summary Statistics - Admit Orders

# dose, freq, route omission errors: Sum of (C) 171

Total admission medications: Sum of (B) 173

Admitting Omission errors?100 admissions 99

Average error per med 0.99

Summary Statistics - H & P

# dose, freq, route omission errors: Sum of (G) 49

Total ordered medications: Sum of (F) 185

Order Omission errors?100 admissions 26

Average error per med 0.26

Baseline: practical process

Multidisiplinary team of reviewers RN, MD, Pharm... QI rep to combine

Minimum 20 charts Institution-wide, random or stratify to ensure all

units represented Minimum stay of 3 days Can be fruitful to include re-admits

Find home meds and list on form Compare to admit orders Identify “unreconciled medications”

Ongoing data collection

Need frequent measurement on every unit where you are testing: monthly charts to display on unit

Process: easy for patients where the reconciling form has been completed; follow process used in baseline data collection when no reconciling form

DON’T CHEAT: Don’t skip patients without a reconciling form Don’t just look for home med list; the question is,

have the home meds been RECONCILED?

TIPS on collecting your data...

Share responsibilities, engaging implementers Limit sample: 20 charts Real-time review: patients on unit for 24 hours Establish rules for consistent treatment where judgment

required (omission or obvious hold or d/c based on patient condition; but strategy should encourage increased documentation by prescriber)

Set time limit (when unable to find home meds, use list from admit orders and indicate that all are unreconciled)

Share “Great Catches”: examples of orders changed, errors prevented

Beyond Admission

and

Longer-term Considerations

Reconciling at Transfer

Compare most recent med record (MAR) and home med list against transfer orders. Issues:

Access to reconciling form with home med history at point when new orders written

Need to modify reconciling form to add columns for reconciling at transfer?

Identifying responsibilities of both the transferring and the receiving unit

Embedding into workflow: Who writes transfer orders? When? Where?

Reconciling at Discharge

Patients especially vulnerable immediately post-discharge Over 12% of patients with an ADE within

2 weeks of discharge [Forster 2003] Address potential for doubling up based

on formulary substitutions or other brand/generic name confusions

Prohibit “resume home meds”!!! Verification of dosing instructions

Outpatient Settings

Applies to settings where the outpatient: may receive medication where patient's response to treatment might be

affected by medications they are on where a practitioner who can review and modify the

patient's medications is a part of the outpatient service

Examples include outpatient oncology services, GI laboratories, emergency department, urgent care clinics, certain imaging procedures.

Using as an order sheet

Proceed with caution, but efficiency gains Most MDs find it very helpful; makes their life

easier, decreases duplication Timing: 6-10 months into the process? Modifications to reconciling form:

Add MD signature line(s) Columns to indicate “continue” or “discontinue” Amendment form

Automation

If you can’t do it on paper,

don’t even try it in vapor First must have a stable process: adequate

testing of the form, implementation on multiple units

Careful design required; who enters info, who can update/change, may introduce new errors

Automation: John Hopkins ICU project Revised form to strike balance between

burden of data collection and comprehensiveness of medication information

Automated process after 48 weeks, paper forms converted to electronic form

Intervention now takes 20 minutes on admission and 20 minutes at discharge with minimal marginal costs

Better access to medication histories

Promote patients maintaining medication cards Provide in ED, at discharge Disease specific support groups Pharmacy medication review Senior center (file of life) Partner with PCPs, nursing homes, VNA,

health plans

Better access to medication histories

Interview strategies including increased use of open ended questions

Link medications to conditions, prescribing physicians

Checklists of OTCs/herbals and commonly missed meds

Leverage expertise of VNAS Shared databases