ournal of Infection and Public Health (2016) 9, 749—756
he impact of automating laboratoryequest forms on the quality ofealthcare services
ajed Al Dogethera,∗, Yahya Al Muallema,owafa Househ (Supervisor)a,c, Basema Saddika,ohamed Khalifab
College of Public Health and Health Informatics, King Saud Bin Abdulaziz Universityor Health Sciences, National Guard Health Affairs, Riyadh, Saudi ArabiaKing Faisal Specialist Hospital and Research Center, Jeddah, Saudi ArabiaThe Research Chair of Health Informatics and Promotion, King Saud University, Saudirabia
eceived 23 June 2016; received in revised form 3 August 2016; accepted 1 September 2016
Summary In recent decades, healthcare organizations have undergone a sig-nificant transformation with the integration of Information and CommunicationTechnologies within healthcare operations to improve healthcare services. Var-ious technologies such as Hospital Information Systems (HIS), Electronic HealthRecords (EHR) and Laboratory Information Systems (LIS) have been incorporatedinto healthcare services. The aim of this study is to evaluate the complete-ness of outpatients’ laboratory paper based request forms in comparison with aelectronic laboratory request system. This study was carried out in the labora-tory department at King Abdulaziz Medical City (KAMC), National Guard HealthAffairs, Riyadh, Saudi Arabia. We used a sample size calculator for comparingtwo proportions. We estimated the sample size to be 228 for each group. Anylaboratory requests including paper and electronic forms were included. We cat-
egorized the clarity of the forms into understandable, readable, and unclear. A total of 57 incomplete paper forms or 25% were identified as being incomplete. For electronic forms, there were no incomplete fields, as all fields were manda-tory, therefore, rendering them complete. The total of understandable paper-basedlaboratory forms was 11.4%. Additionally, it was found that the total of readablewas 33.8% and the total for unclear was 54.8%, while for electronic-based forms,
∗ Corresponding author at: King Saud Bin Abdulaziz University for Health Science (KSAU-HS), Ministry of National Guard — Healthffairs (MNG-HA), Saudi Arabia.
there were no unclear forms. Electronic based laboratory forms provide a morear, and understandable format than paper-based laboratory
findings, KAMC should move toward the implementation ofory request forms for the outpatient laboratory department.dulaziz University for Health Sciences. Published by Elseviered.
deitfopp
alTtslortrtasimiscsicwcdt[ttnlThwawritten data-entry of serology requests is a pro-
Today, many hospitals have implemented or areplanning to implement information technology sys-tems and solutions to improve the quality ofservices provided to patients. In Saudi Arabia,the uptake of technology has lagged compared tomore industrialized nations. Some Saudi institu-tions are leading, while others are lagging in theimplementation of Hospital Information Systems.For example, King Faisal Specialist Hospital andResearch Center (KFSH-RC), in Saudi Arabia hasreached stage 6 for the Electronic Medical RecordAdoption Model (EMRAM), while other hospitals stilluse paper records.
One healthcare domain that has benefited fromthe use of information technology has been the lab-oratory department. Although the literature showsthat the use of information technology can enhancethe process of healthcare delivery, many hospi-tals in Saudi Arabia continue to use paper-basedforms when ordering lab tests [2—4]. The processof requesting lab investigations for outpatients usu-ally occurs manually through paper-based forms.The requesting physician fills the paper form andhands the form to the patient in order to deliverthe hardcopy manually to the laboratory reception-ist. Afterwards, the lab technician draws the bloodsamples from patients, attaches the paper form tothe acquired samples, and sends them to the cen-tral lab for further study and analysis (See Fig. 1).
Many clinicians cannot be expected to stay up-to-date with every complex test and diagnosticprocedure outside their specialty. Furthermore,the overcrowded environment at the lab receptionslows down the workflow and influences the qualityof clinical care provided to healthcare customers[1,5]. Appropriate implementation and the use oflaboratory electronic form test-requesting systemscan help overcome many of the aforementionedchallenges [2,3,5].
Research studies have been conducted on thecomputerization of Laboratory Information Systems
and their impacts on the organizational work-flow [2—6]. Other studies investigated the impactof incomplete data and its influence on patient
csh
iagnosis and treatment [3,4]. Based on the lit-rature, laboratory workflow processes can bemproved with the adoption of electronic labora-ory forms as compared to the use of manual paperorms [2,5]. Some studies focused on the influencef missing crucial clinical parameters on the inter-retation of laboratory results and the reporting ofatient diagnosis [1,5].
One study focused on determining the categorynd regularity of errors when providing data onaboratory request forms at a hospital in Nigeria.he analysis included an assessment of the applica-ion forms, to determine incorrect or incompleteections of the application form, and the regu-arity or frequency of errors. Most of the datamitted and/or regularly repeated in the labo-atory forms was patient age and their location,he name of attending physician, and informationegarding patient’s gender. Further, audited labora-ory request forms did not have enough informationbout the diagnosis or the type of the clinicalample. The authors emphasized the dangers ofncomplete laboratory request forms that includedisdiagnosis and mismanagement of patients lead-
ng to deteriorating health among patients. Theyuggest that laboratory request test forms should beompleted to avoid issues in the healthcare systemsuch as misdiagnosis, repeated laboratory test andmproper treatment procedures [8]. Another studyollected examples of medical error cases whichere a result of missing names, times, and medi-al record numbers. These inadequacies of missingata lead to incorrect diagnosis and an increase inhe number of medical errors within the hospital3,4]. One study compared handwritten laboratoryest-requests with electronic Laboratory Informa-ion Systems. The authors identified the types andumbers of errors that existed in handwritten sero-ogy test requests received in outpatient clinics.he results showed that the written request formsad 67 out of 627 errors where 51 of these errorsere transcription faults while 10 were associ-ted with abbreviations. The study concluded that
ess that is prone to many mistakes. The authorsuggest the use of electronic ordering because itas the potential to eliminate handwritten and
T
th
wrmamtrticfcwceal
oteaddmrccdsmfqeo[
iit
urfifitepipiiactc
iivtifrpl
tspteMwaElEi
he impact of automating laboratory request forms
ranscription errors and improve data accuracy inospital information [10].
Another study evaluated pre-analytical errorsith inadequacies in the completion of laboratory
equisition forms. The study involved an assess-ent of original laboratory request forms received
t the department of clinical biochemistry for fouronths. The evaluation included a manual inspec-
ion of the presence of pre-fixed criteria. 56,000equisition forms were assessed, and it was foundhat the most inappropriately filled parameter wasnformation regarding specimens, treatment, andlinical data, which were missing in almost all theorms reviewed. Nevertheless, the relevant clini-al notes were clearly stated in 74.6% of the formshile the patient’s information was mostly filledorrectly. The findings emphasized the need tonforce and implement policies that would enhanceccuracy and compliance with the necessities ofaboratory request form completion.
In 2006, Plebani conducted a study focusedn laboratory errors that occur frequently duringhe delivery of laboratory testing. Most of theserrors occurred during the pre-analytical and post-nalytical stages of the laboratory test. The authoriscusses that these errors interfere with clinicaliagnosis. The author discusses that the mistakesade in the Total Testing Process (TTP) are labo-
atory related and may have been caused by poorommunication, actions taken by nurses or physi-ians involved in the testing process, or poorlyesigned testing processes. Furthermore, evidencehows that lab information is partially used, whichay contribute to further errors. The study is
ocused on providing a description of the most fre-uent and risky pre-, intra-, and post-analyticalrrors and provides advice on the practical meth-ds of measuring and reducing the risk of mistakes9].
In 2015, Muluberhan conducted a study exam-ning the significance of laboratory request formsn assisting the performance of all laboratory testso the benefit and satisfaction of all laboratory
tbl
Figure 1 Laboratory paper fo
751
sers. The authors assessed the content of emptyequest forms and evaluated the completeness oflled information on medical laboratory requestorms, and the communication of results to usersn two different hospitals. The study showed thathe standard of request forms was weak consid-ring that essential information required was notrovided by the requester, — leaving many gapsn the provided information. This affected therovision of clinical advice based on the limitednformation available on the request forms whichncreased potential errors. The study confirms thatll required parts of the request forms should beompleted to provide sufficient information neededo establish laboratory diagnosis, enhance patientare, save time and financial resources [7].
In 2013, Wiwanitkit found that significant errorsn the request forms were mainly due to incompletenformation and the use of non-standard abbre-iations. Many errors were also observed duringhe collection of specimens, diagnosis, and patientdentification. The study concludes that the majorault of the laboratory requests was incompleteequest form writing and suggests that medicalersonnel should ensure accuracy in specimen col-ection and writing request forms [11].
Many studies discussed the implementation andhe use of electronic laboratory request forms andhow a clear advantage in the use of electronic overaper-based laboratory request forms. However,here have been few studies comparing paper andlectronic-based laboratory forms in Saudi Arabia.ost of available studies evaluated completeness,ithout measuring clarity, where clarity is defineds data that is understandable, readable, and clear.specially in Saudi Arabia, where English is a secondanguage, the possibility of unclear data entry intonglish laboratory forms becomes a risk for higherncidence of medical errors.
The main objective of this study is to evaluatehe completeness of outpatient laboratory paper-ased request forms in comparison with electronicaboratory request system, is to provide an in-depth
rm workflow at KAMC-RD.
lflcl((SBa
C
Awnrrs
ptc
E
TcrtcpraRIo
D
Daf
R
TfAf
752
overview of the data quality, accuracy and com-pleteness on both systems. Furthermore, there aresome secondary objectives, which are: (1) To eval-uate the completion of information on laboratoryexamination request paper forms in comparison toelectronic form, (2) To evaluate the clarity of therequesting physicians information in comparison toelectronic forms, (3) To evaluate the completion ofall required fields on the paper forms in comparisonto electronic forms and (4) To evaluate unautho-rized modification of information on paper forms incomparison to electronic form.
Methods
Study setting
The study was carried out in the laboratory depart-ment at King Abdulaziz Medical City (KAMC),National Guard Health Affairs. The laboratorydepartment consists of four units, anatomicalpathology, clinical chemistry, immunopathology,and toxicology. KAMC is metropolitan hospital thatis based in Saudi Arabia with a combined capac-ity of approximately 1500 beds. Since 2013, KAMCbegan to enter lab requests into the LaboratoryInformation System, for inpatients only. For outpa-tients, it remains a paper-based form. Our studycompares the inpatient electronic request formswith the paper-based outpatient forms.
Sample
We used a sample size calculator for comparingtwo proportions. Using a 95% confidence interval,we estimated the sample size to be 228 for eachgroup. Any laboratory requests including paperand electronic forms were included. When col-lecting the total 456 electronic and paper forms, allforms were selected randomly by using a random-izer website. Furthermore, we introduced exclusionand inclusion criteria. The study included labora-tory request forms in the inpatient and outpatientsetting within KAMC-RD and excluded: (1) Labpaper-form requests from outside KAMC-RD; (2)Interns and less than one-year trainees; (3) visitingor locum staff, to ensure a consistent awarenesslevel of staff on both paper and electronic forms.
Data extraction and collection
This study assessed hand-written outpatient lab-
oratory request forms for a three-month periodbeginning on March 1, 2015 and ending on May 31,2015. The method of collecting data was to acquirethe information from laboratory forms at the
aifw
M.A. Dogether et al.
aboratory reception area. All laboratory requestorms from different clinics are transferred to theaboratory. It took an average of 10 min to apply theoding scheme to each of the laboratory forms. Theaboratory form contained ten primary sections:1) physician information; (2) patient information;3) diagnosis; (4) Hematology; (5) Chemistry; (6)erology; (7) Microbiology; (8) Flow cytometry; (9)lood Bank; (10) Miscellaneous. See Appendix A for
sample lab form.
lassification of errors
researcher independent from the laboratory, butorking within the hospital as a biomedical engi-eer, audited the contents of the handwrittenequest and electronic forms associated with eachequest and classified the errors into categories ashown in Table 1.
A secondary researcher audited the work of therimary researcher to improve the reliability ofhe results. Any disagreements were discussed untilonsensus was reached.
thical considerations
he study was approved by KAIMRC and no informedonsent was required, as the researchers were onlyeviewing paper and electronic forms. We ensuredhat none of the patients were identified and thatare was taken in retrieving and returning any of theaper-based forms used. No patient or physician-elated information was collected. Data retrievalnd analysis began after receiving Institutionaleview Board (IRB) approval from King Abdullahnternational Medical Research Center in Novemberf 2015.
ata analysis
ata analysis was carried out by using descriptivend the chi-square test through statistical packageor social sciences (SPSS) version 21.
esults
able 2 shows the total number of errors recordedor both the electronic and paper-based forms.ny field within the electronic and paper-basedorm that was completely missing was counted
s an error, and the whole form was flagged asncomplete. A total of 57 (25%) incomplete paperorms were identified. For electronic forms, thereere no incomplete forms, as all the fields were
The impact of automating laboratory request forms 753
Table 1 Coding scheme.
Error type Definition Example
Understandable The form provides a decent knowledgeabout written fields
Found patient with anemia, sampleblood for CBC
Readable The form is clear to read in hand writingand provides no knowledge
Vit D and Basic Screen
Unclear The form is unreadable and cannotextract information
The jdshncsl (unclear handwriting)
Table 2 Completion of forms (overall errors rate).
Record type Complete Incomplete Total
Paper form 171 57 228% Complete paper 75% 25% 100%Electronic form 228 0 228
mporotc
dstfIwamAu1tbboatv
bPc
D
TraflintcttiwttMetitIp
lhigher rate of data completeness the systems could
% Complete electronic 100% 0% 100%
andatory, therefore, rendering all forms com-lete. Using the Chi-Square test, we found anverall significant association between the type ofecord (electronic vs. paper) and the completionf the forms (�2 (1) = 273.600, P ≤ 0.001), showinghat electronic forms were more complete signifi-antly.
For clarity of the data content, and to assessata quality, we categorized the variable into threeections: understandable, readable, and unclear. Ifhere is no missing content in any of the fields in theorm, we categorized the form as understandable.f there were two fields with some missing contentithin the form, we categorized the form as read-ble. If there were three or more fields with someissing content we categorized the form as unclear.s shown in Table 3, we found that the total ofnderstandable paper-based laboratory forms was1.4%, the total of readable forms was 33.8% andhe total of unclear forms was 54.8%. For electronicased forms; there were no unclear electronic-ased laboratory forms. However, there was a totalf 21.1% of the forms that were understandable and
total of 78.9% that were readable. By runninghe chi-square to measure the association betweenariables, the result shows a significant association
baf
Table 3 Relation of record type to clarity filed.
Type of record Clarity filed
Understandable
PaperCount 26
Percent 11.4%
ElectronicCount 48
Percent 21.1%
etween type of record and clarity �2 (2) = 355.676, ≤ 0.001, showing that electronic forms were morelear and of higher quality significantly.
iscussion
he study revealed that that paper-based labo-atory forms are less clear, less understandablend less complete than electronic-based laboratoryorms. Although this finding is intuitive and simi-ar to other findings within the field, this findings new for Saudi Arabia. The first line of commu-ication between the patients and the clinician ishe laboratory form. Effective design and properompletion of laboratory request forms is essen-ial to get a better performance from all laboratoryests and benefit diagnostics. We found many miss-ng fields, and more frequently incomplete fields,hen paper-based forms were used while switching
o electronic-based forms would greatly improvehe quality of care, efficiency, and productivity.oreover, there are several advantages in usinglectronic-based system including but not limitedo the following: (1) Enhancing the clarity of clin-cal diagnosis; (2) Improving access to patientests ubiquitously; (3) Reducing duplication; (4)mproving clinical decision support, especially withatients who suffer from allergies.
From the patient’s perspective, using electronicaboratory forms has several benefits. Due to the
e linked to improved diagnosis, which can bettributed to the complete filling of the laboratoryorms within an electronic environment [1].
Readable Unclear Total
77 125 22833.8% 54.8% 100%180 0 22878.9% 0% 100%
lacsptcbt
R
Wasstttmc
C
Tutmiimtmmspisiamtdmla
F
754
When the diagnosis is reliable, the treatment ofthe patients becomes easier since the system cangive predictions of what the ailments could be andthe best treatment associated with the diagnosisdata attained. Errors associated with manual datahandling are eliminated when EHRs have been putin place in the diagnosis process. Studies show thatmany of the errors are a result of the different pro-fessionals involved before the final result can beachieved. When an error occurs in the initial stagesof the data analysis, it is transferred to final stageaffecting the outcomes but with the EHR all thetask is done by a single tool and results released.As compared to the manual system, the EHR dataprocessing is more efficient. When the process isefficient, then the patients are treated faster andthis means the process of decision making is madeby doctors faster [3].
On the other hand the doctors use the EHR toquickly link data transfer from one department toanother within the healthcare facilities and out-side. The data transfer is done electronically whichsaves the space and time used when data is on tran-sit. The faster data analysis and diagnosis enablesdoctors to serve many patients within a limited timeinterval leading to increasing productivity of a givenhealthcare facility. Due to the automated aspectsof EHR, the errors avoided can improve medicalpractices workflow. Electronic-based forms in labo-ratories allow easier and more coordinated care incomparison to paper-based forms. The coordinatedfunctionalities of the systems are a result of imple-menting the integration of artificial intelligenceand algorithms leading to a more comprehensivedecision-making systems for quality healthcare ser-vices. The spaces used in file storage can be utilizedfor other functions in the facilities since the EHRsystems use minimal space. One study discussesthat confidentiality and completeness of the datais highly catered to since damage due minor humanfactors such as dirt does not affect hospital recordswhile in electronic form [12].
Study limitations
Most of the paper-based forms used in the lab-oratory area focused on information filled in byclinicians so it was difficult to check for any alter-ation or if the exact test was requested for thepatient by reviewing all their medical records since
there is no laboratory authorization to do this.The shortage of published research on the topic;especially on the comparison between laboratorypaper-based forms and electronic forms is another
Fnn
M.A. Dogether et al.
imitation. From an economic perspective, there is financial impact to using paper-based forms thatould not be evaluated during this study. Furthertudies may provide a more exact measurement ofotential costs. Another important limitation washe MERS-Coronovirus Outbreak, most students whoollected data from KAMC-RD faced the hospitaleing temporarily closed, this caused a major delayo collect both paper and electronic-based forms.
ecommendations
e recommend that the outpatient laboratory unitt KAMC implement the electronic-based laboratoryystem. With the implementation of the best careystem across the hospital, the outpatient labora-ory unit should start to work with the best careeam and implement the system within the outpa-ient laboratory unit. Delaying the implementationay increase medical error rates and potentially
ause unwanted harm to patients.
onclusion
he study shows that paper-based forms are lessnderstandable, less complete, and less readablehan electronic-based forms. Important infor-ation required on the requisition forms was
ncomplete or missing. This could lead to low qual-ty reports, misdiagnosis of test outcomes thatay have harmful effects on patient administra-
ion, and might build the potential for futureistakes. Conversely, the provision of all infor-ation in an accessible clear form leads to the
trengthening of health care which is a benefitassed on the patient. Some of the benefits feltn the health care affect doctors and include fasterervice deliveries, completeness in the health data,ncreased number of patients treated per day,nd faster transfer of records from one depart-ents to another. The most crucial benefit is
he increase in accurate and fast generation ofiagnosis reports which enhance treatment andanagerial decisions. Incomplete information on
aboratory requisition forms can lead to misman-gement and misdiagnosis of patients.
unding
unding was provided by King Abdullah Inter-ational Medical Research Center. Funding grantumber SP15/070.
T
C
N
E
En
A
TSRI
he impact of automating laboratory request forms
ompeting interests
one declared.
thical approval
thical approval was granted by King Abdullah Inter-ational Medical Research Center. A
755
cknowledgment
he project was scientifically supported by Kingaud University, Deanship of Scientific Research,esearch Chairs and The Research Chair of Healthnformatics and Promotion.
ppendix A.
[
[
756
References
[1] Nutt L, Zemlin AE, Erasmus RT. Incomplete laboratoryrequest forms: the extent and impact on critical resultsat a tertiary hospital in South Africa. Ann Clin Biochem2008;45(5):463—6.
[2] Georgiou A, Prgomet M, Toouli G, Callen J, Westbrook J.What do physicians tell laboratories when requesting tests?A multi-method examination of information supplied tothe microbiology laboratory before and after the introduc-tion of electronic ordering. Int J Med Inform 2011;80(9):646—54.
[3] Idowu AA, Jeje OA. Incomplete laboratory request forms asa contributory factor to preanalytical errors in a Nigerianteaching hospital. Afr J Biochem Res 2011;5(3):82—5.
[4] Onyiaorah IV, Ukah CO, Anyiam DC, Odike MA, Oyeka IC.Effect of remedial measures on inadequacies in the com-pletion of laboratory request forms by clinicians. Clin Audit2012;4:9—14.
[5] Zemlin AE, Nutt L, Burgess LJ, Eiman F, Erasmus RT. Poten-tial for medical error: incorrectly completed request formsfor thyroid function tests limit pathologists’ advice to cli-nicians. SAMJ 2009;99(9):668—71.
[
Available online at www
ScienceD
M.A. Dogether et al.
[6] Van Walraven C, Naylor CD. Do we know what inappropriatelaboratory utilization is?: a systematic review of laboratoryclinical audits. JAMA 1998;280(6):550—8.
[7] Muluberhan A. Assessment of Medical Laboratory requestform completeness and noncommunicated result to clini-cians in the two public hospitals found in Hawassa city,Southern Ethiopia; 2015 (Doctoral dissertation, AAU).
[8] Oladeinde BH, Omoregie R, Osakue EO, Onifade AA. Evalu-ation of laboratory request forms for incomplete data at arural tertiary hospital in Nigeria; 2012.
[9] Plebani M. Errors in clinical laboratories or errors in labo-ratory medicine? Clin Chem Lab Med 2006;44(6):750—9.
10] Vecellio E, Maley MW, Toouli G, Georgiou A, WestbrookJ. Data quality associated with handwritten laboratorytest requests: classification and frequency of data-entryerrors for outpatient serology tests. Health Inf Manage J2015;44(3):7.
11] Wiwanitkit V. Errors in Laboratory Requests in Chula-longkorn Hospital. Maharat Nakhon Ratchasima Hosp Med
Bull 2013;24(2):83—90.
12] Hanan AA, Roland M. Quality of primary health care in SaudiArabia: a comprehensive review. Int J Qual Health Care2005;17(4):331—46.
