WASHINGTON, DC—As health care moves toward electronic record-keeping and away from paper, the way cancer registries collect and record data is also changing, according to speakers here at a session held as part of the International Union Against Cancer (UICC) World Cancer Congress.
The changes are bringing more speed and timeliness to data collection, but are also raising questions about the quality of the data, said speakers from three nations: the United Kingdom, Italy, and the United States.
In the UK, the National Health Service (NHS) has had full population registration of cancer cases since the 1960s, noted Chris Carrigan, National Coordinator for Cancer Registration in the UK.
In England, eight regional registries collect data from 34 established clinical cancer care networks, he explained. Traditionally, these data have been extracted manually directly from patient case notes in acute-care hospitals and specialist treatment centers, a laborious and time-consuming endeavor.
“Until recently, the regional registries have had to focus most of their resources on the process of data collection, which has mainly relied on extensive manual effort,” he said. But by the end of 2006 all registries in England will be able to receive their data from existing electronic sources, and those data will be able to be processed using fully automated rules-based systems.
“My job was to try to modernize the registries” by taking as much of the manual effort as possible out of registry collection and moving it into the electronic age, Mr. Carrigan said. Citing the black hole syndrome—whereby data went in and nothing came out—he added, “Many of the doctors didn't trust the data that was coming out of the registries.”
Updating the cancer registries is part of a nationwide UK effort to modernize England's health care system, he said. “It's a strong push to get cancer data integrated with the whole NHS system.
“The National Health Service is currently in the midst of the most significant period of investment in new technology ever seen. The program will build a secure national clinical patient record [the NHS Care Record] available to appropriate clinical staff at the point of care, wherever that care is being given.”
Secondary Uses Service
Another NHS project, he noted—the Secondary Uses Service (still some way off)—will make data available for research from the clinical patient records in an anonymous, pseudonymous, or summarized format (to protect patient privacy).
“In readiness for this, cancer registries in England are already moving away from routine manual data collection and entry and are implementing routine electronic data receipt and automated rules-based processing, to build a common cancer patient record—which will include data from referral and screening, diagnosis, and treatment and mortality—to satisfy the range of growing needs,” Mr. Carrigan said.
Already, many of the UK cancer registries are using existing clinical data sources, and data collected for other purposes, to construct the extended cancer registration record. The cancer registries now receive early notification of every newly diagnosed cancer case in England, from data submitted by all hospitals within 25 working days of the end of the month in which the patient was first treated.
“You can look for footprints of care,” Mr. Carrigan noted. Thus, sources of data have increased, and registration has become faster, a process that “goes a huge way to solving issues of rapid case ascertainment.” It is also cheaper, he said, after the initial investment in capital costs for hardware and software.
As for quality, Mr. Carrigan said that it is true that those who enter electronic data may lack the extensive training of experienced registry employees who have worked with manual systems.
“In England we train registry staff, but we also go out and train the hospital-based people who enter the data into the registry,” he said.
He also noted that all registries in England are now subject to a formal process of peer review to ensure quality. “As part of the NHS Cancer Services Peer Review program, registries self-assess against a series of national agreed-upon measures,” he noted.
“The evidence base is checked to confirm the self-assessment, to give a quantifiable and comparable indication of the registries' performance. The most important aim of this registry peer review is to identify and spread good practice across the registry community.”
In Italy: Increasing Capacity of Computer Hardware Has Allowed for Storage of Huge Amounts of Health Data
In Italy, the increasing capacity of computer hardware has allowed for the storage of huge amounts of health-related data including hospital admission and discharge, outpatient records, pathology records, death certificates, drug consumption, and laboratory and instrumental exams, said Lorenzo Simonato, MD, PhD, Professor in the Department of Environmental Medicine and Public Health at the University of Padova, Italy.
“The main objective of the introduction of automation in cancer registration has been the epidemiological utilization of existing population-based electronic archives storing health information.”
Dr. Simonato predicted that automated cancer registries could blaze a trail for archived electronic health information—“Cancer registration could become, in the future, just one specific function of an integrated surveillance system based on multiple information sources fed by electronic health archives.”
Dr. Simonato said that in the 1990s the Venetian Tumour Registry (VTR) “explored for the first time the possibility of using, as primary sources of data for the registration process, electronic routine data from hospitals, pathology departments, and other health institutions in that region of Northeast Italy.”
Today, he noted, Italy has the North East of Italy Cancer Surveillance Network (NEICSN) as well as the VTR and a Milan cancer registry.
SITE Software Conditions
The VTR uses software called SITE, which accepts the following conditions, said Dr. Simonato:
- ▪ Cancer cases with full concordance between two or more sources.
- ▪ Histologically confirmed cases with at least one concordant or compatible hospital discharge or death certificate.
- ▪ Histologically confirmed skin cancer (ICD 173) unless in combination with skin melanoma (ICD 172).
- ▪ Histologically confirmed benign, in situ, or tumors of uncertain behavior.
Like Mr. Carrigan, Dr. Simonato agreed that quality control has to be a major component of a cancer data registry, since the data are only as good as their source and the accuracy of data entry.
He said that data quality can be lower in electronic systems than in manual tumor registries, but that this should not be a deterrent in moving toward automated systems. In the case of the NEICSN surveillance network, for example, cases rejected by the software program “are solved manually by the registry personnel,” he said.
In US, Cancer Registration Has Been Ongoing for 50+ Years
In the United States, cancer registration has been in effect for more than 50 years, said Kenneth A. Gerlach, MPH, CTR, Health Scientist at the CDC's National Program of Cancer Registries. He said that in the US, cancer registration is primarily a state function, and the registry infrastructure has evolved over the years to include all the American states and the District of Columbia.
Challenge: No National US Patient Health Record
The CDC's National Program of Cancer Registries (NPCR) collects data on the occurrence of cancer; the type, extent and location of the cancer; and the initial treatment. The NPCR and the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) program collaborate to publish annual US cancer incidence and cancer mortality data. A specific challenge is that there is no national US patient health record, as there is in England.
Mr. Gerlach noted that today “the methods of cancer registration and surveillance, one of the most complex areas of disease surveillance, have evolved to rely more upon information technology, although the remnants of paper-based systems remain.”
He said one goal is to reduce the clerical functions of tumor registrars, so that they can spend more time on more challenging projects, such as data analyses.
In the US, the complexities of cancer registries are due, said Mr. Gerlach, to the breadth and complexity of the data that are captured; the multiple varying sources from which the data come; the multiple providers from which cancer patients receive care; the longitudinal nature of the data; and the numerous levels of users of cancer surveillance information.
Mr. Gerlach noted that the National Framework for Cancer Surveillance recommends integrating new information technology into current systems for keeping registries, with the ultimate goal of facilitating the transition from paper to electronic records.
“With the implementation of HIPAA, the Health Insurance Portability and Accountability Act, and the national vision to establish interoperable electronic health records, there is a renewed emphasis on moving away from paper-based health records systems to electronic-based systems,” Mr. Gerlach said.
“Cancer registries have a history of taking advantage of receiving electronic files from health care facilities and incorporating those files into associated software systems.”
This practice is especially well-honed for pathology reports, he noted. Currently, “it is not uncommon for information from a health care system's software system to be printed and re-entered by a tumor registrar into the cancer registry database.”
But, he said, “We're starting to see electronic submissions from the laboratory directly into the cancer registry.” In short, he said, “Activities are underway to re-think current methods of cancer registration and model new methods and tools to take advantage of emerging electronic health records.”
Asked by OT about the goal of streamlining and updating US cancer registries given the sheer size and pluralism of the country, Mr. Gerlach said, “I think the challenge is that we have such a diverse country.”
He noted that there is no unique patient identifying number, and that some Americans don't want one because they mistrust the government. Therefore, he said, “I think we have to be a lot more creative in collecting the data.”
‘We'll Get There’
Despite the many current obstacles, Kenneth A. Gerlach, MPH, CTR, of the CDC's National Program of Cancer Registries, predicted that some form of US electronic patient health record is certainly coming.
In an interview following his talk, he told OT that he was highly encouraged to see Senator Hillary Rodham Clinton (D-NY) and staunch Republican politician Newt Gingrich standing together in bipartisan support of a US electronic health record.
“We're going to get there,” Mr. Gerlach said.