The effects of a regional telepathology project: a study protocol

Pathology is a scientific field concerned with the study of the nature and causes of diseases. Anatomic pathology, which is at the heart of this paper, is a medical specialty concerned with the diagnosis of diseases based on the gross, microscopic, chemical, immunologic, and molecular examination of organs, tissues, and whole bodies (autopsies). Because pathology examinations serve as the basis for a large percentage of decisions about diagnosis, treatment, hospital admission, and discharge, their annual number have increased in recent years in step with health care demand and, ultimately, with population growth and demographic aging. While an increase in the supply of pathologists would appear to be essential to maintaining patient care, the number of pathologists has declined in many developed countries over the last decade. This has been the case in Japan [1, 2], Australia [3], the UK and New Zealand [4], as well as Canada [5], to name a few. In the province of Quebec - the site of our study - the supply of pathologists fell 10% from 1998 to 2008 [5]. Moreover, pathology is known to be unpopular among medical students [6], making it even more difficult to reverse this trend.

A shortage of pathologists has several consequences. The most direct is a longer turn-around time before a diagnosis can be made on a specimen [7]. Knowing that in the case of cancer - an area where pathology diagnoses are crucial - prompt diagnosis/treatment is particularly important, the repercussions of this labor shortage can be dramatic. The shortage also influences workload and quality assurance. When there are fewer people to do a job, workers become overloaded and it is difficult to guarantee good quality [6, 8]. Finally, because of the shortage many pathologists are compelled to work past retirement age, and this may add to the overall quality problem. Once purely theoretical, such consequences were confirmed in recent years when news accounts about errors in anatomic pathology testing began to make national headlines. In addition to this shortage, pathology is gaining in complexity and pathologists working alone increasingly require support from colleagues. It is estimated that 10 to 20% of oncology cares require a second opinion [9].

One way to counteract the scarcity of pathologists and help them obtain a second opinion is to adopt telepathology, which is defined as the practice of pathology at a distance. Telepathology uses telecommunications technologies to facilitate the transfer of image-rich pathology data between distant locations for the purposes of diagnosis, education, and research [10]. Telepathology is performed in three modes: static, dynamic, and hybrid. When a pathologist is reviewing one or several preselected still images, he is working in static mode (also known as store-and-forward). In contrast, when a pathologist examines images while communicating live with the remote site where the specimen is located, he is working in dynamic mode. Finally, whenever a pathologist uses both methods to provide a diagnosis, he is working in hybrid mode [11].

Microscopic static examinations work well when there is no ambiguity as to the images the laboratory technologist should submit to the pathologist [12]. A microscopic dynamic examination involves the use of a robotized remote-controlled microscope by the pathologist to explore an entire slide surface, change the magnification level, or focus on and digitize a portion of interest. The glass slide is usually prepared by a well-trained laboratory technologist who must also position it within the microscope [13]. A macroscopic (or gross) dynamic examination is performed when the pathologist can first look at the whole specimen (e.g., tumour, organ, etc.) and talk to the technologist to orient the cuts. The histological section is then fixed on a glass slide and completely digitized by a slide scanner to provide a virtual slide of outstanding quality that can be remotely accessed and examined [14]. Virtual slides, as they are called, therefore constitute a newer form of microscopic static examination.

Once installed, a telepathology system can be used for distant primary diagnoses, expert referrals, quality assurance, and education [12]. For distant diagnoses in the absence of a local pathologist or a telepathology system, slides must be physically sent to another facility, hence delaying the diagnoses. With telepathology, the slides are examined remotely, either statically or dynamically, and diagnoses are swiftly provided. In some particular instances, however, pathology examinations are performed while the patient is still under anaesthetics and undergoing surgery, and the surgeon needs a pathology diagnosis to properly resume his procedure. For these specific examinations, called intraoperative frozen sections, delays are undesirable. When there is no local pathologist in a hospital and a surgical procedure requires an intraoperative frozen section examination, either the patient is transferred to another hospital, a visiting pathologist is called on site, or a procedure is performed in two steps (creating time for the slides to be sent and read elsewhere).

With telepathology, intraoperative frozen section analyses can be performed whenever a pathologist is available at a remote location. The procedure usually unfolds as follows: a laboratory technologist manipulates the specimen locally while it is filmed; a pathologist performs a gross examination of the specimen from another location examining the specimen and interacting with the technologist via a monitor and tells her where to cut; the technologist prepares one or more slides from the section then digitizes and stores them in a repository; and the pathologist accesses the images, performs his examination and provides his diagnosis directly to the surgeon by phone. The present study focuses specifically on this hybrid telepathology service: intraoperative frozen section examinations of virtual slides.

Examinations of intraoperative frozen sections have been performed via telepathology for over 20 years. In 1989, a Norwegian hospital decided to cover another facility located 400 km away using a motorized video-microscope that could be remotely controlled. The project was then expanded to four other hospitals [15]. In Japan, where experiments with telepathology began in 1982, intraoperative rapid diagnosis has been intensively used since the early 1990's to alleviate the effect of a severe shortage of pathologists - only 0.7% of the total number of physicians in Japan are pathologists [1]. In the United States, a fully implemented telepathology service was introduced between Veterans Affairs Medical Centers in Milwaukee (Wisconsin) and Iron Mountain (Michigan) in 1996. This service has been used, among other things, to examine intraoperative frozen section cases at Iron Mountain, which has had no on-site pathologist since that time [16]. In Canada, pathologists from the University Health Network (UHN) started using telepathology to examine frozen sections in 2004. More precisely, pathologists located at the Toronto General Hospital cover for Toronto Western Hospital and its affiliated Neuroscience Center.

We conducted a scoping review [17] to identify the nature and extent of telepathology research evidence. First, we searched two databases (PubMed and ABI Inform) with the keyword "telepathology" to obtain a general sense of the existing research on this topic. We chose these databases because we felt that they were offering a good sample of knowledge dissemination within the medical field (PubMed) as well as the business field (ABI Inform). Our queries, performed on September 9, 2011, resulted in 763 citations (730 found in PubMed, 25 found in ABI Inform and 8 found in both databases.) The titles and abstracts of these articles were skimmed and they appeared to be predominantly validations of the diagnostic accuracy of telepathology systems. At that point, we added the keyword "frozen section" to our search algorithm, as our study focuses on this type of examination. Our revised queries resulted in 72 citations (68 found in PubMed, three found in ABI Inform and one found in both databases). Four of these articles were excluded: three were not in English or French and one could not be found through our interlibrary system. Of the remaining 68 articles, 38 were concerned with diagnostic accuracy (19 of these articles also examined the time spent to arrive at these diagnoses), 10 were general papers introducing the topic/technology, nine were qualitative research on matters such as organizational consequences, learning, and cooperation in setups where telepathology was used (seven of them were subsections of a single research project by a single author), four were technical papers related to specific systems/features, and two were descriptive case studies presenting projects without evaluating them. Finally, only two articles were looking at the outcomes of telepathology, which is at the heart of our project. This literature dates back to 1991 when the first papers on the Norwegian installations were published.

According to Drummond et al. [18], evaluations must first be carried out on efficacy (can this theoretically work?), effectiveness (does this work in practice?), and availability (is this accessible to those who need it?) before an in-depth economic evaluation can be performed. The results of our scoping review confirm this evolution and suggest that the field is mature enough for more benefit assessments, economic or not. In this line of thought, the primary objective of this study is to evaluate the benefits of a major regional telepathology project implemented in 17 health care organizations located in eastern Quebec, Canada. The main goal of the project is to allow remote analysis of intraoperative frozen sections, although the telepathology system in place can also be used for other purposes, including secondary opinions and education sessions. As explained below, expected benefits encompass quality, access and productivity outcomes and will be measured at both the local and regional levels.

Our evaluative approach is based on DeLone and McLean's information systems success model, illustrated in Figure 1. According to this framework, the individual benefits of an information system are derived from expressed satisfaction with the system and its use, which are in turn influenced by three factors: the perceived quality of the system, the perceived quality of the information contained in the system, and the perceived quality of the technical and organizational support offered. Finally, the greater the benefits at the individual level, the more the positive effects associated with use of the new system will be felt at the organizational and regional levels [19‐21].

In the last 30 years, the number of telemedicine applications has multiplied due to advancements in telecommunications. One of these applications, telepathology, has fostered considerable academic research since the mid-1980s. The extant literature has mainly followed the path described by Drummond et al. [18], and evaluations have so far been carried out on reliability, efficacy, effectiveness, and availability [e.g., [12, 28]]. For frozen section examinations, the accuracy of the technique has been demonstrated [e.g., [29, 30]] as has its speed [e.g., [14, 31]]. However, very few empirical studies have evaluated the benefits of telepathology at the patient, provider, organizational and regional levels.

Using a mixed-method design that combines qualitative and quantitative data collection methods, the present study attempts to fill this gap. More specifically, our research will add to the extant body of knowledge by offering a thorough evaluation of a large telepathology project, assessing outcomes at multiple levels of analysis. To our knowledge, our study will be the first to examine a regional telepathology project by focusing on intraoperative frozen section analyses. Our study will also be the first to perform an exhaustive pre-post evaluation using secondary data from operating room and other health information systems. By drawing a clearer picture of the benefits to be obtained by deploying a large regional telepathology project, our study should help decision makers and regional authorities redistribute regional workforces around optimal workflows that take advantage of telepathology. Ultimately, positive results should influence health policy decisions targeted at broader societal and environmental issues.

Notwithstanding these anticipated contributions, our study protocol has limitations and, hence, more research will be needed to further explore the implications of our work. For one thing, we must acknowledge the usual limitations on generalizability associated with collecting data through mail questionnaires as well as the cross-sectional nature of Phase 1. Significantly, even though we will be able to compare contrasting situations by investigating a wide variety of settings in Phase 2, the fact remains that this research is limited to the study of a single telepathology project and our results will be context-dependent. Indeed, the province of Quebec has its own specificities in terms of the healthcare system (public), density of population (medium to very low), physicians' remuneration mode (fee-for-service) and pathologist shortage (moderately severe). Therefore, more evaluative studies will need to be conducted in other regions or countries to enhance the generalizability of our findings.

To our knowledge the ongoing telepathology project at the Laval UIHN is unique in Canada and is expected to yield significant results that are relevant both nationally and internationally. Results will provide valuable information to healthcare managers and decision makers who must find more effective and efficient ways to provide accessible care services to patients and, more specifically, patients living in rural regions. A demonstration of both perceived and actual impacts will likely contribute to the promotion of telepathology at the individual (patient and provider), organizational and regional levels. We hope that the study protocol provided here will guide other researchers interested in evaluating such projects and contribute to the overall development of a solid knowledge base regarding the range of benefits associated with telepathology in health care organizations.