Pending lab test results at the time of discharge: a 3-year - PowerPoint PPT Presentation

Pending lab test results at the time of discharge: a 3-year retrospective comparison of paper versus electronic test ordering in three ED Nasir Wabe a , Ling Li a , Gorkem Sezgin a , Maria Dahm a , Elia Vecellio b , Robert Lindeman b , Johanna Westbrook a and Andrew Georgiou a a Centre for Health Systems and Safety Research, AIHI, Macquarie University, Sydney, Australia; b NSW Health Pathology, Prince of Wales Hospital, Sydney, Australia

Delivering safe and effective test result communication, management and follow-up Establishment of effective test result management systems in hospitals • Identifying the existing modes of communication, responsibility and accountability • Qualitative interviews, ethnographic observations Harnessing health information technologies (IT) • The effect of health IT on test result management • Controlled pre- and post- studies utilising data analytics to measure the impact on test requesting, follow-up and patient outcome. Enhancing the contribution of consumers • What do consumers want? What is needed to establish patient-centred health care? • Qualitative studies to explore patient and consumer views.

Introduction

Introduction • Lab testing is one of the key drivers in the clinical decision making . • In ED, lab tests are commonly ordered to establish a working diagnosis .  In US, lab test are ordered for 42% of ED visits (CDC 2015).  In one Australian hospital, 56.4% of all ED visits were ordered lab tests (Vecellio 2014).

Introduction… • Ideally, a test result should be available as early as possible and prior to patient discharge. • However, evidence from systematic reviews has suggested that that 1.0%-75% of test results are pending at discharge (Callen 2011).

Introduction… • What are the consequences of pending test results? • Pending test results can have major health consequences (Dalal 2014, Callen 2011).  Less likely to be followed-up and therefore may lead to: o Missed or delayed diagnosis, o Increased hospital revisits and o Adverse drug reactions

Introduction…

Introduction… http://simsrad.net.ocs.mq.edu.au/login?url=https://search.proquest.com/docview/2039368518?accountid=12219

Introduction… • So what is the role of EMR? o Allow clinicians to order tests and receive results electronically. o Provides a platform to streamline laboratory workflow.

Introduction… • Paper-based systems : o Can be more time-consuming to process within the laboratory and therefore may delay test result availability. • To date, studies assessing the relationship between test order types (paper-based or electronic) and the rate of pending test results at patient discharge are lacking.

Aim of the study • To determine the rate of pending test results at time of discharge from ED, and compare the rate for paper-based and electronic orders. • We hypothesised that tests ordered electronically would be less likely to be pending at discharge compared to those ordered using paper- based system.

Methods

Methods • Setting : o 3 hospital EDs in NSW, Australia: one large metropolitan hospital with over 500 beds [Hospital A] and two smaller regional hospitals [Hospital B and C]. • Population : o All patients attending the study EDs between Jan 2014 and Dec 2016. o Only patients who were treated in ED and eventually discharged from ED were included.

Methods… • Dataset : o Linked LIS and ED data o Chemistry and haematology tests were LIS EDDC selected. o POCT and add-ons were excluded.  An add-on is an additional test that is carried out on a specimen that has already been delivered to the lab.  The process of add-on testing does not follow the same process as a new test order.

Methods… • Statistical Analysis : o Descriptive statistics were reported as appropriate. o Pearson’s Chi -squared test was used to compare the rates of pending tests for paper-based and electronic orders. o As a subgroup analysis, the top five most frequently ordered tests were selected.

Results

Results • A total of 71,466 ED visits fulfilled the inclusion criteria:  55.9% were female  the median age was 49 years.  Most visits occurred at Hospital A (38.7%) while presentations to Hospitals B and C were roughly the same (30.5% each).

Results… Paper-based 2.8% (n=10,007) Total ED visits Total tests n=71,466 n=357,476 EMR 97.2% (n=347,469) • The rate of paper-based test orders were:  roughly the same for chemistry (2.8%) and haematology (2.9%).  slightly varied by study hospital.

Rate of pending results • Electronic ( 6.6%) versus paper ( 9.7%): Diff = 3.1% ( P <0.001). Table 1 . Rate of pending test results at ED discharge. Test order Test Pending Test Result type Count Test Count % (95% CI) Paper 10,007 966 9.7 (9.1-10.2) Electronic 347,469 22,928 6.6 (6.5-6.7)

Rate of pending results by study EDs 1 5 . 0 The difference was significant for all hospitals * 1 2 . 5 ) * P e n d i n g t e s u l t ( % D i f f = 3 . 0 % * 1 0 . 0 D i f f = 2 . 7 % D i f f = 3 . 4 % 7 . 5 5 . 0 2 . 5 0 . 0 P a p e r E M R P a p e r P a p e r E M R E M R H o s p i t a l A H o s p i t a l B H o s p i t a l C Figure 1: Rate of pending test results at ED discharge by study ED, 2014-2016. * P <0.001

Rate of pending test results over time • Diff. was consistently significant. 1 5 . 0 ) • Did not change considerably over 1 2 . 5 * P e n d i n g t e s u l t ( % * * 1 0 . 0 D i f f = 2 . 0 % time for paper orders D i f f = 3 . 2 % D i f f = 3 . 7 % 7 . 5 • for electronic orders the rate 5 . 0 dropped from 8.0% in 2014 to 2 . 5 5.4% in 2016. 0 . 0 P a p e r E M R P a p e r P a p e r E M R E M R 2 0 1 6 2 0 1 4 2 0 1 5 Figure 2: Rate of pending test results by order types over time, 2014-2016. *P<0.001.

Rate of pending test results for top five tests 2 2 . 5 • Auto. Diff, EUC, LFT, 2 0 . 0 * ) CaMgPO4 and CRP= P e n d i n g t e s u l t ( % 1 7 . 5 D i f f = 4 . 0 % 1 5 . 0 represented 70% of the total 1 2 . 5 tests. 1 0 . 0 * 7 . 5 • 4 of 5 tests had lower rates of * * D i f f = 3 . 1 % 5 . 0 D i f f = 1 . 8 % D i f f = 0 . 4 % D i f f = 2 . 2 % pending test results. 2 . 5 0 . 0 P a p e r E M R P a p e r P a p e r P a p e r P a p e r E M R E M R E M R E M R A u t o . D i f f . E U C L F T C a M g P 0 4 C R P Figure 3: Rate of pending results type of tests, 2014-2016. *P<0.001.

Discussion and conclusion

Key findings • The major finding is that the rate of pending results was significantly lower for electronic orders than for paper orders. • This could be due to improved timeliness of laboratory workflow when the electronic system is used:  Shorter data entry time: o Georgiou et al . (2013) found that the median data entry time was 3 minutes shorter for electronic orders than paper order.  Shorter overall TAT. o In study by Thompson et al. (2004) the overall TAT decreased from a median of 148 to 74 minutes following the implementation of EMR. 24

Key implications • Pending tests are a risk factor for the failure to follow-up test results which can be a major source of harm for patients (Callen 2011).  Lower rates of pending test results with electronic ordering could lead to a reduction in potentially harmful outcomes that may arise from abnormal laboratory results arriving after patient discharge. 25

Limitations of the study • Information on the nature of the results (e.g. abnormal result returned post discharge) were not included.  We did not assess whether the results of the pending tests were potentially actionable . • Paper ordering has remained consistent at around 2-3% over the course of this study.  Further research into this area should explore why paper systems continue to be used. 26

Conclusion • Our findings highlighted an important benefit of ordering tests via the electronic system which can potentially improve patient safety. • Future research is needed to explore how pending test results are managed, including how abnormal test results arriving after discharge from ED are communicated to clinicians and patients . 27