Utilization of synoptic reporting in Vancouver Island Health Authority (VIHA) and the effect on quality and processes of pathology reporting

(1)

by

Jennifer Furtado

BMLSc, University of British Columbia, 2006

A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of

MASTER OF SCIENCE

in the School of Health Information Science  Jennifer Furtado, 2013 University of Victoria

(2)

Supervisory Committee

Utilization of synoptic reporting in Vancouver Island Health Authority (VIHA) and the effect on quality and processes of pathology reporting

by

Jennifer Furtado

BMLSc, University of British Columbia, 2006 Supervisory Committee

Dr. Francis Lau, School of Health Information Science

Supervisor

Mr. Jeff Barnett, School of Health Information Science

Departmental Member

Dr. Cyril Blake Gilks, School of Health Information Science

(3)

Abstract

Supervisory Committee

Dr. Francis Lau, School of Health Information Science

Supervisor

Mr. Jeff Barnett, School of Health Information Science

Dr. Cyril Blake Gilks, School of Health Information Science

The pathology report is a key document that contains critical information for the proper diagnosis and prognosis of a patient. It is used to communicate key pathologic findings to users of the pathology report: oncologists who will use this document to base treatment options and cancer registries that use key attributes of cancer cases for surveillance and research purposes. The synoptic report can

provide a standard format that contains key information consistency for users of the report. The Vancouver Island Health Authority (VIHA) was the first to adopt this innovative method of pathology reporting within British Columbia.

A mixed methods research study design using both qualitative and quantitative techniques to evaluate a synoptic reporting system’s impact on pathology reporting was used in this study. The impact of synoptic reporting was evaluated based on the utilization of the synoptic summaries and the perceived effect on the quality and processes of pathology reporting from the perspectives of pathology reporting stakeholders. Pathology reports obtained from VIHA and British Columbia Cancer Agency (BCCA) databases were audited to obtain information on the amount of synoptic summaries included in pathology reports and whether this inclusion was associated with any pre-‐defined variables. Pathology reporting stakeholders’ perceptions data was obtained by a self-‐administered web survey distributed to VIHA pathologists, Victoria location British Columbia Cancer Agency (BCCA-‐VICC) oncologists and BCCA Cancer Registry management.

(4)

Out of the reports generated by VIHA pathologists, only 3.5% (593/17430) of them contained a synoptic summary, 5.4% (590/10990) when only considering the pathology reports that were associated to a CAP checklist and activated within the system. According to the VIHA database, it was determined that 16.2%

(2819/17430) cases were malignant suggesting that inclusion of synoptic

summaries is not being done for the vast majority of malignant cases contrary to expectations. This was further confirmed by the BCCA audit, which contained primarily malignant cases but also had additional non-‐malignant (benign) cases requested by the BCCA. The BCCA audit revealed that 18.6% (58/312) reports included a synoptic summary, 19.7% (58/294) when only considering the pathology reports that were associated to a CAP checklist and were activated within the

system. Differences with the synoptic summary inclusion were found in: individual pathologists, procedures performed, sites of report generation, tumor groups, tumor types, and malignancy status of the specimens. Variation was observed in the

perceptions surrounding the quality of the pathology report. Pathologists’ responses were highly variable to the majority of questions in all categories,

demonstrating the polarizing views within the community. Pathologists overall had a more favorable opinion of the current pathology report (including synoptic

summary) generated, where users (oncologists and registry) had a more favorable opinion of the perceived impact of synoptic summaries on pathology reports and a more favorable opinion regarding improvements that could be made in pathology reporting.

While VIHA has taken a significant step forward, being the first health authority in British Columbia to implement synoptic reporting (synoptic summaries), further improvements can be made to better enhance the adoption.

(5)

List of Tables

Table 1: Variable under investigation inclusion of synoptic summaries in pathology

reports ... 31

Table 2: Dermatopathologists reporting on skin specimens ... 32

Table 3: Mode Responses to Equivalent questions asked of multiple groups ... 33

Table 4: ‘Quality’ Index ... 34

Table 5: Multiple Choice questions that allowed multiple selections ... 34

Table 6: Multiple Choice questions that allowed one response ... 34

Table 7: Pathologists' inclusion of synoptic summaries in pathology reports submitted to the BCCA ... 39

Table 8: Pathologists' inclusion of synoptic summaries in pathology reports from VIHA audit ... 40

Table 9: Pathologists' inclusion of synoptic summaries in pathology reports by specialization reporting only specimens within their specialization ... 41

Table 10: Inclusion of synoptic summaries for different procedure types in pathology reports submitted to the BCCA ... 43

Table 11: Inclusion of synoptic summaries for different procedure types in pathology reports from the VIHA audit ... 44

Table 12: Inclusion of synoptic summaries for individual sites of pathology report generation, which were later submitted to the BCCA ... 45

Table 13: Inclusion of synoptic summaries for different sites of generation of pathology reports from VIHA audit ... 46

Table 14: Inclusion of synoptic summaries for different tumor groups, which were later submitted to the BCCA ... 47

Table 15: Inclusion of synoptic summaries for different tumor groups frm VIHA audit ... 47

Table 16: Specimen sites with no synoptic summary inclusion ... 48

Table 17: Inclusion of synoptic summaries for specific specimen sites ... 49

Table 18: Inclusion of synoptic summaries for malignant and benign specimens for pathology reports, which were later submitted to the BCCA ... 50

Table 19: Inclusion of synoptic summaries for specimen donation for pathology reports, which were later submitted to the BCCA ... 50

Table 20: Survey Respondents 'Area of Professional Specialization' ... 52

Table 21: Pathologists' self-‐reported inclusion of synoptic summaries in pathology reports ... 53

Table 22: Self-‐Reported reasons why pathologists are more likely to include a synoptic summary ... 54

Table 23: Factors that may influence the pathologists' decision to not include a synoptic summary ... 54

Table 24: Themes identified by pathologists as a concern in synoptic reporting ... 54

Table 25: Mode Response on Content of Current Reporting ... 57

Table 26: Mode Responses of Timeliness of Current Reporting ... 58

Table 27: Mode Responses of Usability of Current Reporting ... 59

(8)

Table 29: Mode Responses of the Impact of Synoptic Summary Inclusion on Content

... 62

Table 30: Mode Responses of the Impact of Synoptic Summary Inclusion on Timeliness ... 62

Table 31: Mode Responses of the Impact of Synoptic Summary Inclusion on Usability ... 63

Table 32: Mode Responses of the Impact of Synoptic Summary Inclusion on Accuracy ... 64

Table 33: 'Quality' Index for the Perceived Impact of Synoptic Summary Inclusion on pathology reporting ... 65

Table 34: Users perceived helpfulness of synoptic summary inclusion ... 66

Table 35: Mode Responses of Improvements in Content that could be made in pathology reporting ... 67

Table 36: Mode responses of Improvements in Usability that could be made in synoptic reporting ... 68

Table 37: 'Quality' index for the Improvements that could be made in pathology reporting ... 69

Table 38: Words used to initiate the key word search via CoPathPlus ... 101

Table 39: BCCA and VIHA variables under investigation in the audits ... 118

Table 40: Pathology Reports not aligning to CAP checklists based on specimen ... 119

Table 41: Pathology Reports not aligning to CAP checklists based on procedure ... 119

Table 42: Checklists not activated by VIHA ... 120

(9)

List of Figures

Figure 1: Provincial distribution of synoptic reporting levels ... 10

Figure 2: Groups of Data ... 30

Figure 3: Values associated with the BCCA audit ... 37

Figure 4: Values associated with the VIHA audit ... 38

(10)

Acknowledgements

This thesis was possible through the influence, guidance and support of

interdisciplinary and inter-‐organizational personnel interested in the potential impact of synoptic reporting on pathology reporting.

UVIC personnel: Dr. Francis Lau for his invaluable guidance and mindful reviewing as a supervisor. His expert knowledge in the area of evaluation studies helped guide the development of the thesis from a concept to a completed research study.

VIHA personnel: Dr. Barry Power for his interest and support of this thesis topic through all stages of the study; Diana Nagy for her invaluable consultation; Amanda Jones for her extraction of data and Dr. Nick van der Westhuizen for his interest and encouragement.

PHSA personnel: Dr. Blake Gilks for his expertise in pathology reporting and support for the thesis that extended well beyond the role of a committee member; Jeff Barnett for his support in the development of the thesis topic and guidance as a committee member; Cheryl Alexander for her extraction of data; Cathy MacKay for her interest and Dr. Paul Blood for his encouragement.

(11)

Problem

The Institute of Medicine’s report on medical errors ”To Err is Human: Building a

Safer Health System” (2000) and the report on quality “Crossing The Quality Chasm: A New Health System For The 21st Century” (2001) have shed light on medical errors

within health care environments and have increased public pressure in quality performance in all areas of medicine. The Institute of Medicine defines ‘Quality of Care’ as “the degree to which health services for individuals and population increase

the likelihood of desired outcomes and are consistent with current professional

knowledge”. Errors that can occur in pathology reporting include: Diagnostic errors

(errors in diagnosis, failure to employ indicated tests, use of outmoded tests, and failure to act on results), Treatment errors (an avoidable delay in treatment or in response to an abnormal test or inappropriate care), or a failure of communication. Effective communication among health professionals is vital to ensure patients receive the best possible care. Powsner, Costa and Homer (2000) have identified the communication between the pathologists and clinician as problematic;

pathologists’ terminology can be considered the furthest from daily medical

discourse and information felt to be implicit in pathology reports by the pathologists were not felt to be present by the clinicians. Synoptic reporting is a format that can support standardized reporting to help ensure all pertinent information is explicitly and accurately reported in a timely and usable format.

(12)

Purpose

This study was an evaluation of a synoptic reporting system implemented at a health authority in an attempt to investigate the level of use and application and the perceptions surroundings its use with respect to the quality of pathology reporting.

Organization of Thesis

This paper will focus on oncology however non-‐cancerous specimens will be discussed where appropriate. In Chapter Two, a literature review will be provided to describe the environment of pathology reporting and associated problems. Additionally, how synoptic reporting can impact pathology reporting will also be discussed. Chapter Three will discuss the background surrounding the research environment. Chapter Four will describe the research approach providing details of the study purpose, research questions, study design, case/participant selection, data collection and data analysis methods. Lastly, Chapter Five will describe the study results in detail followed by Chapter Six, which will provide an interpretation of the results and conclusions.

(13)

Chapter Two – Literature Review

Pathology is the study and diagnosis of disease using macroscopic, microscopic, chemical, immunological and molecular techniques to evaluated cells, tissues and/or organs. Pathologists are physicians with at least three additional years of training to specialize in evaluating these specimens. Pathologists often specialize in certain tissues or organs and are trained in cancerous and non-‐cancerous

conditions. Pathologists are responsible for diagnosing the vast majority of cancers and since most cancer treatments are based on a specific diagnosis, the pathologists’ evaluations are key to the quality of patient care. The communication of pathologic finding, therefore, is vital to ensure clinicians are basing their treatment plans on the best available information.

Pathology Reports communicate pathologic findings

The pathology report is a key document that contains critical information for the proper diagnosis and prognosis of a patient. Numerous medical professionals are involved in treating cancer patients during their cancer treatment and it is vital for these professionals to communicate effectively with each other in order to ensure patients receive the best possible care. Different medical specialists use language that may not be fully understood by practitioners in other areas and as such it is critical that important criteria be explicitly, not implicitly, reported. Users of the cancer pathology report include clinicians, tumor registries, government agencies and health planners, epidemiologists and others involved in quality improvement

(14)

activities or population-‐based research. Oncologists are the major clinical users of the cancer pathology report. Oncologists are physicians who co-‐ordinate cancer patients’ treatment and who have specialized in cancer by completing three

additional years in residency. Oncologists can be categorized into two main groups based on how they treat patients: Medical Oncologists specialize in medication treatments such as chemotherapy and hormones while Radiation Oncologists specialize in radiation therapy to treat patients. Cancer patients may receive either of these treatments alone or in combination with each other.

Problems in Pathology Reporting

Pathology reports are requiring more data elements to be reported to adapt to medical advances and to conform to guidelines such as the College of American Pathologists (CAP). This has put a demand on the pathologists to ensure the latest data requirements are included in the report and all required data elements are included in every pathology report. Verleye, Mosgaard, Soeane, van der Velden, Lotocki, and van der Graaf (2011) evaluated the content of ovarian cancer pathology reports. Verleye found a “substantial number of [Ovarian cancer] reports basic

pathologic data missing with possible adverse consequences for the quality of cancer care”. This missing data may include vital information necessary to properly stage

the disease and to develop treatment plans.

(15)

correct information and has been attributed to deaths in Canada (Hede, 2008). In Newfoundland, the first of a series of Canadian scandals was revealed in the media. Specifically, a laboratory failed to accurately report the results of a key test between 1997 and 2005, during which nearly 400 out of 1000 patients received incomplete or inaccurate pathology reports and therefore lead to an inaccurate diagnosis. Over one hundred of these patients have since died, leaving family members to wonder if the proper diagnosis was made at the time, would their family member still be alive? The inquiry into this scandal, lead to the conclusion that overworked pathologists and a lack of standards at the Newfoundland laboratory contributed to these critical medical errors.

Even when the essential content is included in the pathology report some critical information can be lost in translation. Powsner, Costa and Homer (2000) conducted a study entitled “Clinicians Are From Mars and Pathologist Are From Venus”

evaluating the clinician’s ability to comprehend crucial information contained in the pathology report. This study noted a discrepancy with what the pathologists

determined necessary to report and what the users, such as oncologists, of the pathology report required to perform their role. Powsner et al. cited the Diagnostic Anatomic Pathologists’ terminology as potentially the furthest from daily medical discourse and hence the motivation for selecting this medical profession to study. Powsner et el. administered pathology reports to general surgeons and conducted questionnaires to assess their comprehension of these reports. It was unclear in this study how much practical exposure the surgeons had with pathology reports as

(16)

surgeons typically are not the primary users of the report. Powsner et el. found a crude 30% discrepancy rate between the pathologists’ intended meaning and the interpretation by the clinicians. Some of these discrepancies have great clinical ramifications. For example, this study found in the interpretation of a kidney biopsy report, a staggering 38% of clinicians had mistaken impressions of ‘no acute

rejection’. Additionally, in reference to bladder biopsy reports, Powsner et al. cited a “widespread failure to recognize deeply invasive bladder cancer and carcinoma in

situ, both of which are key features in determining therapy”. The greatest amount of

error occurred in implicit values, values which the pathologist felt was implicit in the data elements reported, but were not understood to be present by the clinician. Powsner et al. rightfully acknowledged that in a real clinical environment, clinicians would have the opportunity to resolve ambiguous findings by contacting the

reporting pathologist. The clinician however may not realize they need clarification or busy clinical work environments may hinder this interaction.

To prevent errors like the above mentioned and to increase the comprehension of the information contained in the pathology report, the issue of quality pathology report generation and reporting processes needs to remain a high priority.

Standardization in Pathology Reporting

An initial step for ensuring the comprehension of the pathology report is to ensure content is explicitly reported for users of the pathology report and presented in an

(17)

guidelines and checklists to ensure consistency in reporting requirements and these checklists have been internationally recognized as well recognized by the Canadian Partnership Against Cancer (CPAC), an independent organization funded by the federal government, as a pan-‐Canadian standard. CAP is an organization that was formed in 1947 and is the leading organization of board-‐certified pathologists. It is a leader in laboratory quality assurance and has been an advocate for high-‐quality and cost-‐effective patient care. CAP was the original creator of the medical

vocabulary, Systematized Nomenclature of Medicine-‐Clinical Terms (SNOMED CT) fostering its development for over 40 years and it continues to develop and

maintain the medical vocabulary on behalf of the International Health Terminology Standards Development Organization (IHTSDO). CAP guidelines and checklists assist by ensuring consistency in reporting requirements including site

specific/cancer specific staging and prognostic information. There has been an observed increase in CAP guidelines and this is largely due to advancements in medicine for more targeted cancer treatments, thereby resulting in an increase in the complexity of the pathology report (Amin, 2010). Information is required on various fields such as tumor grade, specimen size, tumor size, local extent, vessel involvement, marginal status, other morphologic findings and tumor markers, and other reporting requirements vary for each individual cancer type. This is a

challenge for the pathologist who is obliged to keep up with the reporting guidelines for each cancer type and to remember to report each of these findings every time evaluating a specimen and this was identified as one of the problems contributing to the Newfoundland incident.

(18)

Additionally, the British Columbia Association of Laboratory Physicians (BCALP) has developed pathology reporting guidelines. The BCALP is a section of the BC Medical Association and is composed of medical doctors with specialty certifications in Laboratory Medicine and includes experts from all aspects of laboratory testing and its application to patient care. The BCALP pathology reporting guidelines of practice has been well received by the pathologist community in BC. Other provinces in Canada do not have an equivalent to the BCALP reporting guidelines. This has created an environment in British Columbia where pathologists are left to determine which elements from which guidelines (CAP or BCALP) to report.

Synoptic Reporting in Pathology

The term ‘Synoptic Report’ can refer to a spectrum of reporting formats and can simply be considered a summary of pertinent findings. The term however is most often extended to mean the inclusion of structured data elements in discrete data fields using headers to prompt required information. Srigley, a leader in Canadian synoptic pathology reporting, categorized this spectrum of reporting formats. Narrative reports were divided into Level One: narrative containing no CAP content and only single text data fields and Level Two: narrative containing CAP content and only single test field data. Synoptic reports were divided into Level Three: Level 2 with the addition of ‘synoptic-‐like’ structured format, Level Four: Level 3 with the addition of electronic reporting tools using drop down menus, Level Five: Level 4

(19)

discrete fields and Level Six: Level Five with the addition standardized coding such as the International Classification of Disease for Oncology (ICD-‐O) and SNOMED CT. (Srigley, McGowan, Maclean, Raby, Ross, Kramer, 2009)

The electronically captured synoptic report can be coded using standard

terminology such as SNOMED CT, Logical Observation Identifiers Names and Codes (LOINC) or ICD-‐O and messages can be transmitted using the Health Level 7 (HL-‐7) standard. Synoptic reporting software can include features such as drop down menus, radial buttons, peer-‐review checklists, calendars, built in logic, branching and automatic staging, automatic generation of final diagnosis and risk factor calculation. Additionally, the tool can provide restrictions that can prevent the user from moving onto the next data element until the mandatory elements are

completed.

‘Synoptic-‐like format’ was defined by Srigley et al. (2009) as providing a

standardized ‘look’ with headings but the text was largely narrative. ‘Synoptic-‐like’ and synoptic reports alone can clarify findings for clinicians and if reporting is extended to include structured data elements it clarifies findings for computers. For the purposes of this study the term ‘synoptic’ will refer to Level 5 or higher:

synoptic-‐like structured format, using electronic reporting tools with drop down menus and incorporation of standardized reporting language into discrete data fields.

(20)

Currently most Canadian pathology reports are written as narrative reports though there has been a movement in recent years to make these reports more synoptic in format. The distribution of the provinces by their synoptic reporting level can be seen in Figure 1.

Figure 1: Provincial distribution of synoptic reporting levels

Source: Canadian Partnership Against Cancer (CPAC) (2012)

The Canadian Association of Medical Oncologist (CAMO) the Canadian Association of Radiation Oncology (CARO) and the Canadian Association of General Surgeons (CAGS) are professional organizations that have supported the change in the report format from narrative to synoptic. In Ontario, Srigley et al. (2009) found that the

(21)

synoptic report adoption rate varied by cancer site, by region (Local Health

Integration Networks-‐ LHIN) and also by hospital within a given region. Srigley et al. found that most pathology reports were a hybrid of narrative and ‘synoptic-‐like’ format.

Advantages of synoptic pathology reporting

The advantages of using synoptic pathology reports have been documented in recent years. Advantages can be seen with the improved quality of the report

generated and with the reporting processes. Srigley et al. (2009) defined the quality in terms of the completeness (content), timeliness, usability and accuracy of the pathology report and the reporting process.

Synoptic reports provide a consistent format to assist the reporting of key

information and they also help to alleviate differences in format and terminology among pathologists, which have sometimes lead to problems in understanding the reports (Kang, Devine, Piccoli, Seethala, Amin, and Parwani, 2009). The synoptic report can assist in guiding the pathologist to report all of the recommended requirements and can be considered a learning tool for both pathology residents and well-‐established pathologists. The content of the pathology report can be improved with synoptic reporting by increasing the ability to adhere to the most recent version of guidelines and reporting more required content (Srigley et al., 2009). Srigley et al. found a higher percent of reporting requirements with synoptic reporting whether using an electronic tool or paper document, compared to

(22)

narrative reports. Hemmings, Jeffery and Frizelle (2003) found that GI and GU biopsy specimen pathology reports had significantly more content when a synoptic report was used. Ensuring minimum data requirements are always met has obvious clinical, research and cancer registry benefits and also is necessary for accreditation purposes. This could have also prevented the breakdown in the Newfoundland laboratory that failed to accurately reporting findings.

The timeliness can be improved by decreasing the time to generate a pathology report (Mack, Bathe, Herbert, Tamann, Ruie, and Fields, 2009) and decreasing the time for the user of the pathology report to interpret the report as the information is laid out in an intuitive structure (Srigley et al. 2009, Hassell, Parwani, Weiss, Jones and Ye, 2010, Kang et al. 2009). Mack et al. (2009) found a significant five-‐minute decrease in generation of the report by the pathologist compared to the narrative.

Srigley et al (2009), Hassell et al. (2009) and Kang et al. (2009) all found that the synoptic report was easier to decipher by alleviating differences in terminology among pathologists, therefore minimizing the risk of misinterpretation. Synoptic reports are also cited as reducing the amount of clinician follow-‐up for clarification of findings therefore reducing amount of pathologists re-‐reviewing slides (Mohanty, Piccoli, Devine, Patel, Williams and Parwani, 2007). Insertion of ‘canned comments’ can be inserted based on data in a specific field. For example, a bladder biopsy specimen’s diagnosis of dysplasia could trigger an automatic message detailing the clinical significance of this finding and could recommend clinical follow up (Parwani,

(23)

Mohanty and Becich, 2008). Additionally, synoptic reporting can be easily

completed at different points in the work flow allowing special procedures i.e. flow cytometry to be added and easily linked to the pathology report. Some vendors allow direct access to website information for easy resource information for the clinician. Furthermore, synoptic reporting can allow structured data to

automatically populate Laboratory Information System (LIS) fields, facilitating quick access to data thus enhancing patient care (Mack et al. 2009). Built in logic can additionally allow for staging to be done automatically (Parwani et al. 2008).

The elimination of transcription services with synoptic reporting introduction has resulted in error reduction in typographical and transcription errors (Amin, 2010). Additionally, many synoptic reporting vendors have built-‐in logic and derived values incorporated into the checklists to reduce the possibility of errors at the point of data entry, which may have prevented the Newfoundland reporting errors. An additional benefit associated with the reduction of transcription services is the significant cost savings to a hospital and when including other environments such as the cancer registry and research transcription and abstraction needs, the costs are even more exorbitant (Amin, 2010).

Synoptic reporting facilitates secondary use of the data. If the data has been coded to a standard terminology, such as SNOMED CT, and messaging standards, such as HL-‐7, are also used, the data can be reused in a plethora of ways. Secondary usage can include cancer registries, tumor banking, quality improvement reporting, stage

(24)

capture, quality management and evaluation, patterns of care and outcome analysis, system planning, population research and other research applications. Data can also be incorporated into relational databases for more efficient searches (Amin, 2010). Since searches can be streamlined, synoptic databases will enhance basic science, clinical and translational research.

Limitations of Synoptic pathology reporting

Though introduction of synoptic reporting can offer many benefits, some research has cited limitations. With respect to the content, cited limitations of synoptic reporting include the checklists are often incomplete, such as the absence of certain staging-‐modifiers terms and limitations on histology choices and the checklists would include clinically meaning-‐less elements (Hassell et al, 2010). Standardized reporting by its very design can inhibit the extreme biological variability of

specimens. The tick box style of many synoptic reporting systems can give a

misleading sense of certainty in the findings and standardized terminology may not accurately describe cancer of mixed, indeterminate or unusual histology (Balleine, Bilous and Morey, 2008). To minimize these constraints for atypical specimens, free text commentary options can be incorporated. An additional limitation is the

significant work that goes into designing templates (Parwani et al, 2008). Smaller organizations may simply not have the resources to dedicate.

(25)

synoptic reports some other studies found an increase in time for the pathologist to complete the pathology report (Mohanty et al, 2007, and Hassell et al, 2010).

Usability of the synoptic report can also be brought into question. The checklist can be inflexible and not allow for nuanced diagnosis or microscopic findings (Mohanty et al, 2007) and cannot accommodate for more rare conditions. Additionally, pathologists can have difficulty adapting to negative reporting, that is reporting when certain pathological findings are not present (Branston, Williams, Greening Newcombe, Daoud and Abraham, 2002) and the synoptic reporting can be

cumbersome (Mohanty, 2007). Mohanty (2007) noted some pathologists were not willing to change their method of reporting. This may be related to discomfort with using information technology. Hassell (2010) cited that the willingness to adopt the new method of reporting negatively correlated with the age of the pathologist and positively correlated with the facilities adoption of computer technologies. Urquhart, Porter, Grunfeld and Sargeant (2012) also noted the prospect of being monitored as potentially an inhibiting factor to the pathologists’ adoption. A further limitation is that some vendor solutions do not allow for more than one checklist to be selected for a specimen thus limiting the usability.

The only cited limitation with respect to accuracy of the synoptic report was that long drop down menus potentially contributed to errors when attempting to select the right value (Karim, van den Berg, Colman, McCarthy, Thompson and Scolyer 2008).

(26)

Implementing Synoptic Reporting into clinical practice

Though the benefits of synoptic reporting likely outweigh the disadvantages, synoptic reporting is a complex innovation to implement. The complex nature stems from requirements that significant change must be made in physician behavior and culture, significant support is required from the organization in the form of policies and processes and from local government in addition to the technical aspects of the integration with existent health information technology infrastructure. Implementation of synoptic reporting must occur at the individual, organizational and system levels. The individual level being supported by clinical leaders and clinical practice guidelines; organizational level being supported by a culture that will embrace the change, clinical champions, management support and evaluation/feedback mechanisms. The system level being the broader socio-‐

political context and is supported by policies, financial incentives/disincentives and inter-‐organizational networks. Synoptic pathology reporting implementation processes have a high degree of interdependency amongst organizational members thus while interventions designed at the individual level are important to clinical change the complex nature of the health care environment means that the individual level alone cannot change the widespread clinical practice.

(27)

Chapter Three -‐ Background

British Columbia (B.C.) Current State

Provincially British Columbia is moving towards a plan to implement synoptic pathology reporting across the entire province. Ontario has been the national leader in pathology synoptic reporting while Alberta has been the national leader for surgical synoptic reporting. British Columbia is a complex environment, where current reporting practice is variable. Pathologists often pick certain requirements from numerous pathology-‐reporting standards, predominately BCALP and CAP, and reporting some requirements from the Memorial Sloan-‐Kettering Cancer Center. Much of the disarray arises out of current literature lacking conclusive evidence to which standard or standards to choose and why. Resistance to standardized reporting arises from complaints that they can be too restrictive, not allowing for unique observations; checklist items have limited therapeutic value and are often a result of outdated information; information needs are promulgated by oncologists; and reporting is taking longer with extra workload induced by oncologists ‘wants’ and is not considered by pathology laboratory management when assessing pathologists manpower needs.

Vancouver Island Health Authority (VIHA)

The Vancouver Island Health Authority (VIHA) is one of six health authorities in B.C. and it provides health care to over 750,000 residents on Vancouver Island, the Gulf Islands and the mainland communities north of Powell River and South of Rivers

(28)

Inlet. VIHA was chosen for this study, as currently it is the only health authority in B.C. using a synoptic pathology reporting system. Implementation began with the South Island region in September 2009 followed by North Island and Central Island over the next few months. All tumor site checklists were implemented at the same time.

VIHA uses the term ‘synoptic summary’ when referring to the synoptic report portion of the pathology report. Including synoptic summaries into the pathology report is an option available to VIHA pathologists. Pathologists will dictate their findings using Dragon Speech Recognition by Nuance and elect to complete a

synoptic summary before signing out the case. Templates for the synoptic reporting checklists were provided by the Anatomical Pathology Laboratory Information System vendor, Cerner’s CoPath PLUS, and modified by a VIHA pathologist lead in the specific tumor group. These checklists are based on the CAP standardized checklists. Currently, VIHA has 79 checklists activated with the 2009 CAP checklists being the latest version installed.

British Columbia Cancer Agency (BCCA)

The British Columbia Cancer Agency (BCCA) is responsible for providing cancer care to the residents of B.C. (population 4,500,00) and the Yukon (population 30,000). The BCCA includes six cancer centers: Abbotsford Centre, Centre for the North, Sindi Ahluwalia Hawkins Centre for the Southern Interior, Fraser Valley Center,

(29)

All pathology reports that contain a cancer diagnosis or specified non-‐malignant condition are to conform to the B.C. Cancer Agency (BCCA) Research Information Regulation Health Act by reporting the cases to the BCCA. A list of reportable cases is found in Appendix A. All health authorities in B.C. including VIHA are required to send the BCCA the appropriate reports. The reportable pathology reports that were generated by VIHA are batched every weekday and faxed captured to the BCCA Health Records where they are scanned into the BCCA Cancer Agency Information System (CAIS) Document Manager. Approximately 60% of the total cancer cases reported are referred to the BCCA for further treatment. With these cases, the BCCA Health Records Staff will index the pathology report image file that was inputted into the CAIS Document Manager to a patient record. VIHA surgical cancer patients will most often be referred to the Vancouver Island Cancer Centre (BCCA-‐VICC) for potential follow-‐up treatment of radiation and or systemic therapy. When a patient is referred to the BCCA for follow-‐up treatment a Medical and/or Radiation

Oncologist will view the patient’s pathology report in the CAIS application, CAIS Clinic. Oncologists use the scanned pathology report image file as a key document to understand the final pathologic diagnosis, which is used to base treatment plans.

British Columbia Cancer Registry (BCCR)

The British Columbia Cancer Registry (BCCR) has been a population-‐based registry since 1969 and it is operated by the BCCA. The BCCR currently captures data on approximately 25,000 cancer cases annually. Pathology reports are fax captured (or

(30)

mailed) and held in CAIS Document Manager, a document repository. Twenty-‐five full-‐time transcriptionists are employed to view these un-‐structured image files or paper document from external providers, to manually pull out relevant data fields and enter data into the BCCR information system, OncoLog. This highly manual and time-‐consuming process is an essential process for cancer surveillance and

research. It is the BCCR legal obligation that all data captured is reported to the Canadian Cancer Registry (CCR), The National Association of Central Cancer Registries (NAACCR) and the World Health Organization: The International Association for Research in Cancer (IARC). The transcription process is entirely manual and Cancer Registry personnel have dual monitors to facilitate using CAIS and OncoLog concurrently to complete the coding process. Disease data is often not entered into the Oncolog database until a year after diagnosis/treatment for

referred patients but once disease data fields are populated in the Oncolog database the corresponding CAIS disease data fields will automatically be updated. This one-‐ year delay therefore does not permit transcribed data elements to be used to guide cancer treatment for BCCA referred cases.

The pathology report is considered the source of truth for diagnostic information and is the primary document used, though any supporting information entered in CAIS during the course of the patient’s treatment will complement this report. The BCCR relies significantly on external sources (i.e. health authorities) to send

pathology reports for notification of a cancer case and therefore there is recognition that some cancers will go un-‐reported. A significant deviation of the observed-‐to-‐

(31)

expected cancer cases negatively impacts the registry’s certification level. Some un-‐ reported cancer cases are found when the cancer patient enrolls in a research study whereby the research study reports the participants to the registry. Additionally, the BCCR is able to cross reference data from the Vital Statistics Agency (VSA), an agency that records deaths due to cancer, to identify previously un-‐reported cases. There is no cross-‐reference for ‘alive’ cancer patients who have not participated in a research study. As of September 2010, an interface was built with VSA to acquire VSA data but previously this process was manual. As a requirement for

accreditation, the un-‐reported cancer cases found by comparison to Vital Statistics data is required to be less than 5% of the total cancer cases and the BCCR has always just met this. With the combined data from research studies and VSA, the BCCR often discovers approximately 1500 more cases per year. The cancer registry estimates that they are under-‐reporting between 5-‐10% of cancer cases, some cancers being reported less effectively than others. This value is derived by

comparison to other regions, which are considered to do a better job collecting data.

Utilization of synoptic reporting in Vancouver Island Health Authority (VIHA) and the effect on quality and processes of pathology reporting

Supervisory Committee

Abstract

Table of Contents

List of Tables

List of Figures

Acknowledgements

Chapter Two – Literature Review

Chapter Three -­‐ Background

Chapter Three -‐ Background