The testing process in The testing process in primary care: primary care:

Safety and quality implications Safety and quality implications for improving health care for improving health care

Nancy C. Elder, MD, MSPHUniversity of Cincinnati Department of Family Medicine

Funded by AHRQ K-08HS013914-04, 2005-2010

How I got interested in the testing How I got interested in the testing processprocess

2001: Me, new to town, week 1 in new medical practice

21 year old woman, no period for 3 months, scared she may be pregnant!! (she is not)

Review of medical chart (paper)◦ 6 months ago TSH = 29. (hypothyroidism)

Dr.’s note to MA: call patient and have her rtc. MA’s note: tried to call patient, no answer.

◦ 3 visits in intervening 6 months for colposcopy and F/U of abnormal pap smear. None mention TSH

Me: “We screwed up”Reality: Lots of mistakes like this occur, but how

many, causes, outcomes, interventions to improve are unknown

The testing process in primary The testing process in primary carecare

T es tO r d er ed

tes tin gim p lem en ted

T es tr es u ltstr ac k ed

r es u ltsr e tu rn to

o f f ic e an dc lin ic ian

r es p o n s e totes t r es u ltsb y c lin ic ian

tes t r es u ltsd o c u m en ted

an d f iled

P atien tn o tif ied o ftes t r es u lts

P a tien tm o n ito r ed

th r o u g hf o llo w u p

T es tp er fo r m ed

If you can't describe what you are doing as a process, you don't know what you're doing.” W. Edward Demings

Are testing process errors really Are testing process errors really that important to quality?that important to quality?

Frequent?◦Error reports, interviews and observations,

chart reviewsAdverse events and consequences?

◦Error reports, chart reviewsHinder progress toward patient centered

medical home and similar reforms?◦Look at testing process steps relationship to

PCMH characteristics

““Frequency” and types of testing Frequency” and types of testing process errorsprocess errors

Error reports from family physicians and staff◦AAFP National Research Network (NRN) reporting

studies (Dovey, 2002, Phillips, 2006, Hickner 2008) 14 – 25% of ALL physician and staff reported errors

were related to testing Testing process errors break down:

◦ordering tests (12.9%), ◦ implementing tests (17.9%), ◦ reporting results to clinicians (24.6%), ◦clinicians responding to results (6.6%), ◦notifying patient of results (6.8%), ◦general administration (17.6%), ◦communication (5.7%), ◦charting or filing (14.5%)◦other categories (7.8%)

““Frequency” and types of testing Frequency” and types of testing process errorsprocess errors

Observations and interviews with family physicians and staff (Elder, 2006, Elder, 2008, Elder 2009)◦18 focus groups of family physicians and staff

identified problems with all steps in the testing process. Underlying contributing factors included not following procedures, inadequate systems, lack of standardization communication problems.

◦4 family medicine offices in SW Ohio overwhelming depend on individuals to work around testing process problems

““Frequency” and types of testing Frequency” and types of testing process errorsprocess errors

Chart reviews (ongoing)◦In 261 test results in 8 offices in SW Ohio:

74% had a clinician’s interpretation, 70% of patients were notified 53% of abnormal results had follow up plans

◦In 11 urban CHC offices in Chicago, only 61% of abnormal results for pap smears, mammograms, INRs and PSAs had appropriate follow up documented.

Interviews with patients (ongoing)◦Most patients have experienced results not

received, not timely and/or not understandable.

Adverse events and consequences Adverse events and consequences from testing process errorsfrom testing process errors

Error reports from family physicians and staff (Hickner, 2006)◦Adverse consequences included

time lost and financial consequences (22%), delays in care (24%), pain/suffering (11%) and adverse clinical consequence (2%).

◦18% of events resulted in some patient harm

Chart review (Ongoing)In 11 urban CHC offices, more abnormal mammograms

and INRs (70%)had documented follow up than did abnormal pap smears and PSAs (55%).

PCMH hindered by testing process PCMH hindered by testing process errorserrors

PCMH: Quality of CarePCMH: Quality of Care

Most common breakdowns in diagnostic process in closed malpractice claims (Phillips, 2004)◦55% failure to order appropriate test◦45% failure to create a proper follow up plan◦37% incorrect interpretation of a test result

Testing process steps: ◦ordering ◦ interpretation ◦follow up

PCMH: Early identification of PCMH: Early identification of health problemshealth problems

Process of care failures in breast cancer diagnosis (Weingart, 2009) Failure of patients to complete ordered tests a common

factor in breast cancer diagnostic delays

Testing process step◦Tracking

PCMH: Fewer unnecessary testsPCMH: Fewer unnecessary tests

Missing clinical information during primary care visits (Smith, 2005)◦Clinicians reported missing laboratory results in

6.1% of all visits and radiology results in 3.8%.◦59.5% felt these missing results resulted in delayed

care or additional services, including repeating tests.

Testing process steps:◦Tracking◦Documentation◦Patient notification

PCMH: Higher patient satisfactionPCMH: Higher patient satisfaction

Patient preferences for notification of normal laboratory test results: a report from the ASIPS Collaborative (Baldwin, 2005)◦ Privacy, responsive and interactive feedback, convenience,

and timeliness with detailed information are critical for patient satisfaction

Effect of providing information about normal test results on patients' reassurance: randomised controlled trial (Petrie, 2007)◦ Providing patients with information about normal test

results before testing can improve rates of reassurance and reduce the likelihood of future reports of chest pain.

Testing Process steps◦ Ordering◦ Implementing◦ Patient notification

BUT…test result OUTCOMES are BUT…test result OUTCOMES are what really matter, right?what really matter, right?

Quality performance measures reliant on testing outcomes◦HEDIS measures◦Ambulatory Care Quality Clinical Performance

Measures for Ambulatory Care◦Pay for Performance

So why study the testing PROCESS?

Importance of improving testing Importance of improving testing PROCESSPROCESS

Use of process measures to monitor the quality of clinical practice (Lilford, 2007)◦most suitable management tool for judging and

rewarding quality ◦Clinical outcomes are likely to be affected by

factors other than the quality of care◦Outcome measures provide insufficient

information about how to improve◦Assessment of process encourages universal

improvement rather than focusing on outliers

Summary: Testing process Summary: Testing process implications for improving health implications for improving health care care

Testing process errors are frequent and occur across all process steps

Adverse events and harm have been associated with testing process errors

Poorly functioning testing processes hinder practices from achieving PCMH standards

Studying processes is appropriate to monitor and reward health care quality

The Future…The Future…

What is necessary to improve testing process safety and quality?◦Adoption of technology AND a culture of safety!

Improving which steps give the most “bang for the buck?”◦Identified errors of implementation and patient

notification associated with harm and/or adverse events!

◦Follow up of abnormal results most often missing, but rarely identified by staff and clinicians!

What interventions at what step will bring the most improvement to the testing process?◦??????????

Thanks!Thanks!

“We should work on our process, not the outcome of our processes.”

W. Edward Demings

Nancy C. Elder, MD, MSPHeldernc@fammed.uc.edu