ORIGINAL STUDY

Quebec Trophoblastic Disease Registry How to Make an Easy-To-Use Dynamic Database Philippe Sauthier, MD, Magali Breguet, Alexandre Rozenholc, MD, and Michae¨l Sauthier, MD

Objective: To create an easy-to-use dynamic database designed specifically for the Quebec Trophoblastic Disease Registry (RMTQ). Introduction: It is now well established that much of the success in managing trophoblastic diseases comes from the development of national and regional reference centers. Computerized databases allow the optimal use of data stored in these centers. Methods: We have created an electronic data registration system by producing a database using FileMaker Pro 12. It uses 11 external tables associated with a unique identification number for each patient. Each table allows specific data to be recorded, incorporating demographics, diagnosis, automated staging, laboratory values, pathological diagnosis, and imaging parameters. Results: From January 1, 2009, to December 31, 2013, we used our database to register 311 patients with 380 diseases and have seen a 39.2% increase in registrations each year between 2009 and 2012. This database allows the automatic generation of semilogarithmic curves, which take into account A-hCG values as a function of time, complete with graphic markers for applied treatments (chemotherapy, radiotherapy, or surgery). It generates a summary sheet for a synthetic vision in real time. Conclusions: We have created, at a low cost, an easy-to-use database specific to trophoblastic diseases that dynamically integrates staging and monitoring. We propose a 10-step procedure for a successful trophoblastic database. It improves patient care, research, and education on trophoblastic diseases in Quebec and leads to an opportunity for collaboration on a national Canadian registry. Key Words: Trophoblastic diseases, Registry, Database, Electronic health record Received July 22, 2014, and in revised form December 24, 2014. Accepted for publication December 25, 2014. (Int J Gynecol Cancer 2015;25: 729Y733)

Centre Hospitalier de l’Universite´ de Montre´al, Montreal University, Montreal, Quebec, Canada. Address correspondence and reprint requests to Philippe Sauthier, MD, Centre Hospitalier de l’Universite´ de Montre´al, Montreal University, 1560 Sherbrooke E, Montreal, Quebec, Canada H2L 4M1. E-mail: [email protected]. Supplemental digital content is available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal’s Web site (www.ijgc.net). The authors declare no conflicts of interest. Copyright * 2015 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0000000000000401 International Journal of Gynecological Cancer

n 1971, Dr John Brewer of Chicago was already suggesting Itrophoblastic that the morbidity and mortality of patients monitored by a disease center were 9 times lower compared with treatment outside such centers.1 This statement is still valid today, in addition to limiting the costs of minor or advanced neoplasia treatment.2 The International Society for the Study of Trophoblastic Diseases and the European Society of Medical Oncology, through its reference work, confirms that the most successful management of trophoblastic diseases is caused by the development of regional or national centers.3,4 Canada currently has only 2 registries: one in Nova Scotia and one in Quebec. The importance of verified electronic health record registries5Y7 and database for a rare cancer type8 is widely

& Volume 25, Number 4, May 2015

Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

729

Sauthier et al

International Journal of Gynecological Cancer

recognized. Many countries have introduced this type of structure to assess and organize health care in this area.9Y11 In 2003, FIGO (International Federation of Gynaecology and Obstetrics) recommended staging and calculating a risk score for all gestational trophoblastic neoplasia.12Y16 The score and stage change during the patient’s trajectory, with a new score and a new stage in each recurrence of the disease, which is generally different from other cancers.14 The database must be specific and dynamic, which distinguishes it from static databases in use in oncology. Each trophoblastic disease center has developed its more or less complex and expensive database but, to our knowledge, a concordant database in this area does not exist. Because our budget was limited, we had to develop a simple, effective, specific, and inexpensive solution with commercial software usable by all and especially by a noninformatician. In addition, our database had to be centered on the patient and not on the disease, unlike conventional oncology databases. We assessed the value of trophoblastic disease data available at our institution from before 2008. This finding confirmed the importance of creating a practical database available at our institution and for the province as a whole. Based on recommendations of the 1997 Programme Que´be´cois de lutte Contre le Cancer, the project and preparatory work for a trophoblastic disease reference center was undertaken in 2006 at the Centre Hospitalier de l’Universite de Montre´al in the gynecologic oncology division. We have been developing a dedicated software since 2007 and opened our reference center in 2008. This step allowed the Quebec Trophoblastic Disease Registry (RMTQ) to be created and opened on January 1, 2009, recognized by major medical associations involved in this field. A basic research partnership was established with McGill University in 2010, and a link was

& Volume 25, Number 4, May 2015

established in 2013 with the Quebec computerized health record, the Que´bec Health Record. Another link is in the pipeline with the Registre Que´be´cois du Cancer (RQC).

METHODS Medical databases are generally an accumulation of recorded data, such as that found on a spreadsheet. Because each patient could potentially have several diseases (moles and/or neoplasia), we needed a special database centered on the patient and not on the disease, allowing for individualized calculations. We developed a cross-platform database (Mac OS and Windows) to facilitate the exchange of data among the various stakeholders. Because the data have to be protected, it was encrypted and protected by a password. Its development also led us to choose a multiuser database that can be transferred to a mobile platform (iOS, Android). Because financial resources were limited, the choice was limited to an architecture that is already commercially available or based in a LibreOffice suite to meet these criteria. We created an electronic data registration system by producing a database currently using FileMaker Pro 12. In 2007, we started (version 1.0) using FileMaker Pro 8 to establish the principles of the database and test its use at the Centre Hospitalier de l’Universite´ de Montre´al Reference Centre for Trophoblastic Diseases. Subsequently, we have improved and regularly updated the database by taking into account our needs and the development of software support. The current version (2.2) uses 11 external tables, including a main table (list of patients) associated with a unique identification number for each patient (Fig. 1). Each table allows specific data to be recorded, incorporating demographics, contacts, diagnosis, automated staging, laboratory values, pathological diagnosis, and imaging parameters.

FIGURE 1. The main table is associated with a unique identification number for each patient.

730

* 2015 IGCS and ESGO

Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

International Journal of Gynecological Cancer

& Volume 25, Number 4, May 2015

These data are available as drop-down menus or text fields with alerts for extreme values to reduce data entry errors. We have developed a graphic representation that is characteristic of the development of a marker (A-hCG) for trophoblastic diseases, and we use these data as a semilogarithmic curve using a specific table. The various data collected during patient monitoring required for the implementation of the semilogarithmic curve are encoded in a Uniform Resource Locator address specific to the Google Chart application, which is free and available online. The data, which are completely anonymous and unidentifiable, are transmitted via the HTTP GET method (Supplemental Digital Content: Chart Script: http://links.lww.com/IGC/A275, Database Structure: http://links.lww.com/IGC/A276, Detailed Chart Script: http://links.lww.com/IGC/A277, Detailed Database Structure: http://links.lww.com/IGC/A278). We used the Google Chart application because, to our knowledge, FileMaker software currently does not allow specific customization of curves based on data integrated by the program.

RESULTS From 1982 to 2008, we identified 325 patient records through the archives, but only 135 (42.8%) were actually able to be used because of the lack of available data. Histological diagnoses were not reviewed, as they systematically have been since 2009, and no semilogarithmic curve was available before 2009. Currently, access to our data is immediate and no longer depends on data extraction procedures related to the archives. The data in the database are all verified and entered solely by the registry’s coordinating nurse to reduce the risk of error. The centralization of data and its availability have quickly increased our experience in this area, as we have been directly or indirectly exposed to 8 to 12 times more cases per year since 2009. The database gives us permanent access to demographics during consultation. Similarly, record monitoring, the summary of stage and score development, as well as a blank field for

Quebec Trophoblastic Disease Registry

comments are available. Presentation is in 6 separate tabs, which facilitates navigation within the database: Diseases, Exams, Labs, A-hCG Curve, Treatments, and Contact (Fig. 2). Clinical data are grouped under the Diseases tab and allow automated staging according to the FIGO 2000 score and dynamic staging of moles according to Kohorn 2007. This allows for a visual identification of the risk of neoplasia. Imaging and pathology data are grouped under the same tab. Laboratory data, especially A-hCG, may appear at specific times. The use of A-hCG values and treatment data (Treatment tab) allows for the automatic generation of semilogarithmic curves, which take into account A-hCG values as a function of time, complete with graphic markers for applied treatments (chemotherapy, radiotherapy, or surgery). The visual integration of various therapeutic events provides a better understanding of the development of this disease, and the curve analysis facilitates early identification of recurrence. The database also automatically generates summary tables to export data into Microsoft Excel or Adobe PDF formats. A summary sheet is generated for a synthetic vision of patient monitoring in real time and sent to the treating physician (Fig. 3). To illustrate the practical use of our database, we registered 311 patients with 380 diseases from January 1, 2009, to December 31, 2013 (Fig. 4). The majority of patients (75%) were monitored remotely in collaboration with the treating physician. These data also include the recruitment of 70 patients in our genetic research protocol. On January 1, 2014, the registry included 39 patients being actively monitored or in treatment, 255 completed cases, and 10 patients lost to follow-up. On that date, it also counted 41 babies born after a mole or neoplasia and 8 normal pregnancies in process.

DISCUSSION The RMTQ provides a support and a source of information for professionals and patients with trophoblastic diseases. This collaboration aims to optimize their treatment.17

FIGURE 2. Presentation in 6 separate tabs facilitates navigation within the database. * 2015 IGCS and ESGO

Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

731

Sauthier et al

International Journal of Gynecological Cancer

& Volume 25, Number 4, May 2015

FIGURE 3. A summary sheet is generated for a synthetic vision of patient monitoring. Patient registration by the RMTQ team allows epidemiological information to be gathered that is currently not available in Quebec and is essential to better understand and better treat these diseases. The main objective of the RMTQ is to be able to offer this service to all patients in the province under the program of the Direction Que´be´coise de Cance´rologie.18 We have created an easy-to-use database specifically for trophoblastic diseases that dynamically integrates staging and monitoring (Table 1). It certainly improves patient care, research, and education.19,20 Unlike other cancers, which are attributed a final and unique initial stage, the score and stage change during the trajectory of the patient, with a new score and a new stage in each recurrence of the disease. The Registry is a patient database and not a cancer database, unlike most of the other tumor registries. The database was developed using free, available, or inexpensive software compared with the development of software created specifically by professionals. It is used and constantly adapted by and for caregivers, allowing constant adjustment caused by new developments in this area. It is currently being evaluated by the Nova Scotia registry, which could possibly allow data exchange between these 2 provinces. This software is also the basis for the development of the Web-based software that the Society of Gynecologic Oncologists of Canada (GOC) is looking to implement. In addition, it may be a model to enable the integration of add-on registries (eg, rare tumors) to the RQC. The strengths of this software are stability and practical use for 5 years. Its adaptability, drop-down menus, automated calculations, and ease of use allow for a better user acceptance and, therefore, a probable better reliability. However, the

732

dependence on external Web sites and, consequently, the Internet makes the system’s reliability more fragile. A multiple table structure allows easy extraction of data sets because they are stored independently of each other and are only dynamically linked when the program consults the database. In 2012 and 2013, the GOC meetings in Toronto expressed the desire for a national vision to be developed to manage these patients. Several options were considered for the management of these disorders in Canada by experts, which are in line with international recommendations. The first step was to create a Web-based community of practice using software (Go-CLIC) developed by the GOC. This community of practice, built as a Village Metaphor, serves as an online medium to facilitate creativity, learning, innovation, and collaboration among GOC members to improve the care of our patients and enhance our individual and group

FIGURE 4. The registrations from January 1, 2009, to December 31, 2013. * 2015 IGCS and ESGO

Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

International Journal of Gynecological Cancer

& Volume 25, Number 4, May 2015

TABLE 1. 10 Steps for a successful trophoblastic database 1. Define your needs. a. Multicenter or multiple users (95) will probably benefit from a serverYclient software (beyond the scope of this article). 2. Choose an appropriate software that will suit your needs. a. Some are free (eg, Base in the LibreOffice suite). 3. Design the database structure with the appropriate tables and links. At least 5 tables are recommended. i Main (name, date of birth, file number, etc) ii Labs iii Radiology iv Treatments v Diseases 4. Design the user interface. 5. Use scripts (easier, faster, and safer for quality data). a. Automatic staging b. Age calculator c. Drop-down menu 6. Set up a graphic to chart and follow the patient’s A-hCG levels. 7. Test and adjust the database with users of the database. 8. Protect and save your data. 9. Routinely review your goals and take action. 10. Remember that the ultimate goal of the database is to meet the needs of the patient. professional development. The second step is to create a Webbased database for a national registry.

CONCLUSIONS The quality of the management of trophoblastic diseases comes from the development of national or regional reference centers. We have created an easy-to-use and inexpensive database specific to trophoblastic diseases that dynamically integrates staging and monitoring. This database facilitates access to data relevant to monitoring, treatment, and research purposes. We plan to migrate our database into a Web database first to allow the integration of the RQC database and, subsequently, as part of a future Canadian trophoblastic disease registry.

REFERENCES 1. Brewer JI, Eckman TR, Dolkart RE, et al. Gestational trophoblastic disease. A comparative study of the results of therapy in patients with invasive mole and with choriocarcinoma. Am J Obstet Gynecol. 1971;109:335Y340. 2. Golfier F, Raudrant D, Frappart L, et al. First epidemiological data from the French Trophoblastic Disease Reference Center. Am J Obstet Gynecol. 2007;196:172 e1-5.

Quebec Trophoblastic Disease Registry

3. Seckl MJ, Sebire NJ, Fisher RA, et al. Gestational trophoblastic disease: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2013;24:vi39Yvi50. 4. Kohorn E. Regional centers for trophoblastic disease. Am J Obstet Gynecol. 2007;196:95Y96. 5. ACOG. Committee opinion no. 472: patient safety and the electronic health record. Obstet Gynecol. 2010;116:1245Y1247. 6. DeFrancesco MS. Practice smarterI not harder! Obstet Gynecol. 2008;112:10Y13. 7. Bradley CJ, Penberthy L, Devers KJ, et al. Health services research and data linkages: issues, methods, and directions for the future. Health Serv Res. 2010;45:1468Y1488. 8. Bulusu VR, Fullarton J, Leahy M, et al. Rationale and design of a UK database for a rare cancer type: the GEM Registry for gastrointestinal stromal tumours. Br J Cancer. 2013;109:1403Y1407. 9. Higashi T, Nakamura F, Shibata A, et al. The national database of hospital-based cancer registries: a nationwide infrastructure to support evidence-based cancer care and cancer control policy in Japan. Jpn J Clin Oncol. 2014;44:2Y8. 10. Houser SH, Colquitt S, Clements K, et al. The impact of electronic health record usage on cancer registry systems in Alabama. Perspect Health Inf Manag. 2012;9:1f. 11. Menachemi N, Lee SC, Shepherd JE, et al. Proliferation of electronic health records among obstetrician-gynecologists. Qual Manag Health Care. 2006;15:150Y156. 12. Ngan HY, Bender H, Benedet JL, et al. Gestational trophoblastic neoplasia, FIGO 2000 staging and classification. Int J Gynaecol Obstet. 2003;83:175Y177. 13. Ngan HY. The practicability of FIGO 2000 staging for gestational trophoblastic neoplasia. Int J Gynecol Cancer. 2004;14:202Y205. 14. Kohorn EI. Dynamic staging and risk factor scoring for gestational trophoblastic disease. Int J Gynecol Cancer. 2007;17:1124Y1130. 15. El-Helw LM, Coleman RE, Everard JE, et al. Impact of the revised FIGO/WHO system on the management of patients with gestational trophoblastic neoplasia. Gynecol Oncol. 2009;113:306Y311. 16. Ngan HY, Odicino F, Maisonneuve P, et al. Gestational trophoblastic neoplasia. FIGO 26th Annual Report on the Results of Treatment in Gynecological Cancer. Int J Gynaecol Obstet. 2006;95:S193YS203. 17. Murphy DR, Laxmisan A, Reis BA, et al. Electronic health record-based triggers to detect potential delays in cancer diagnosis. BMJ Qual Saf. 2014;23:8Y16. 18. Hernandez MN, Voti L, Feldman JD, et al. Cancer registry enrichment via linkage with hospital-based electronic medical records: a pilot investigation. J Registry Manag. 2013;40:40Y47. 19. White P, Kenton K. Use of electronic medical recordYbased tools to improve compliance with cervical cancer screening guidelines: effect of an educational intervention on physicians’ practice patterns. J Low Genit Tract Dis. 2013;17:175Y181. 20. Kanas G, Morimoto L, Mowat F, et al. Use of electronic medical records in oncology outcomes research. Clinicoecon Outcomes Res. 2010;2:1Y14.

* 2015 IGCS and ESGO

Copyright © 2015 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.

733

Quebec Trophoblastic Disease Registry: how to make an easy-to-use dynamic database.

To create an easy-to-use dynamic database designed specifically for the Quebec Trophoblastic Disease Registry (RMTQ)...
2MB Sizes 0 Downloads 5 Views