THE CANCER REGISTRY by Mr. Abraham Ringel* *Mr. Ringel is a member of the staff of the Operations Research and Systems Analysis Branch of the Regional Medical Programs Service, Health Services and Mental Health Administration, Department of Health, Education and Welfare in Rockville, Maryland 20852 March 1970 INTRODUCTION Regional Medical Programs have evidenced keen interest in establishing cancer registries. Registries have been conceived as mechanisms to improve the primary care of patients; and as a resource in the planning of cancer control programs, in continuing professional education, public education, and the furthering of cancer research. This paper is an attempt to assist Regional Medical Programs and others to more fully understand the requirements and restraints in the proper organization and operation of cancer registries. The opinions and suggested Alternatives for support of cancer registry activities are those of the author, and in no way represent official endorsement. The author wishes to express sincere appreci- ation to Mr. George Linden, Chief of the California Tumor Registry for his valuable advice and information reflecting the experience in California, and to Dr. Sidney J. Cutler and Mr. William I. Lourie, Jr., of the End Results S6ction, National Cancer Institute, for their constructive suggestions. Dr. Frank R. Mark, Chief, and Mr. Francis C. J. Ichniowski Assistant Chief of the Operations Research and Systems Analysis Branch were most helpful with th e organization of this paper, And Mrs. Grace Kelly with its-preparation. TABLE OF CONTENTS Page INTRODUCTION ...................................................... i THE HOSPITAL CANCER REGISTRY ..................................... 1-10 I. Historical Perspective ...................................... 1 II. Definition of a'Hospital Cancer Registry .................... 2 III. Objectives of a Hospital Cancer Registry .................... 2 A. Assist physicians with follow-up .... .... 2 B. Provide reports ..... : .... 2 .. ...... @a'@ .........................3 C. Resource for professional educ ion .............. D. Resource for public education programs ................. 3 E. Resource for clinical and laboratory research ........... 3 !V. Components of a Cancer Registry .......... ... 3 A. A primary site file of abstracts 3 B. A patient name file .................................... 4 C. A follow-up control file ............................... 4 D. An accession register .................................. 5 V. How to Organize a Hospital Cancer Registry .................. 5 A. A Committee on Cancer .................................. 5 B. Developing objectives .................................. 6 C. Supervision of registry operations ........ ii ........... 6 D. Estimating the cancer caseload ......................... 6 E. Cost considerations .................................... 7 VI. Special Problems ............................................ 8' A. Lack of physician interest .............. .... 9 B. Inadequate supervision ..................... 9 C. Inadequate quality control ..............I ...............9 D. Inadequate utilization of data ......................... 9 E. Unavailability and turnover of personnel ............ O.. 9 F. Manual versus automated operation ...................... 9 G. Continuing financing ................................... 10 THE CENTRAL CANCER REGISTRY ..................................... 11-19 1. Definition and Objectives of Central Cancer Registries ..... 11 A. Facilitate follow-up ................................... 11 B. Resource to identify problems and needs ............... 11 C. Resource to measure quality of patient care ........... 11 D. Resource for professional education ........... ......... 11 E. Resource for public education ......................... 11 F. Epidemiological and clinical research ................. 11 TABLE OF CONTENTS (Continued) Page II. Types of Central Cancer Registries ........................... 12 A. The Connecticut Tumor Registry .......................... 12 B. The California Tumor Registry ........................... 12 C. The Rocky Mountain States Cooperative Tumor Registry .... 13 D. The Iowa Central Tumor Registry ......................... 13 III. Advantages and Benefits ..................................... 14 A. Improved quality of data ................................ 14 B. Efficiencies that come with size ........................ 14 C. Increased resources ..................................... 14 D. Improved educational programs ........................... 14 E. Patient benefits ........................................ 14 IV. How to Organize a Central Cancer Registry ................... 15 A. Medical leadership ...................................... 15 B. Medical advisory committee .............................. 15 C. Definition of objectives ................................ 15 D. Consideration of scope of coverage and caseload ......... 16 E. Method of operation ..................................... 16 F. Cost of the central registry ............................ 17 V. Special Problems ............................................ 18 A. Maintain physician interest ............................. 18 B. Quality control ......................................... 19 C. Availability of qualified personnel ..................... 19 D. Availability of long-term financing ..................... 19 REGIONAL MEDICAL PROGRAMS AND CANCER REGISTRY ACTIVITIES .......... 21-26 I. Regional Medical Programs ................................... 21 II. Grants for Cancer Registry Activities ....................... 21 III. Alternatives for Support and Benefits ....................... 21 A. Regional staff technicians .............................. 22 B. Financial assistance to hospitals ....................... 22 C. Financial assistance to local central registry(s) ........ 22 D. Financial assistance to region-wide central registry .... 22 IV. Questions Considered for Support ............................ 24 A. Purpose of registry ..................................... 24 B. Objectives .............................................. 24 C. Cooperative arrangements ................................ 25 D. Medical advisory group ................................... 25 E. Participating hospitals ................................. 25 F. Personnel ............................................... 25 G. Training of personnel ................................... 25 H. Phasing into system ..................................... 25 I. Automatic data processing equipment ..................... 25 iv TABLE OF CONTENTS (Continued) Page J. Software ............................................... 4 25 K. Justification of budget ................................. 25 L. Other sources of support ................................ 25 V. Consultation ................................................. 26 APPENDICES ........................................................ 27-44 I. Basic Standards for Cancer Programs ............. 0 ........... 27 II. Items of Information for Cancer Registries ................. 29 III. Cancer Registry Abstract Form, Patient Index Card, Follow-up Control Card, Atcesgion Register .................. 37 IV. Organizational Plan for Cancer Programs .................... 41 V. Cost Considerations in the Operation of a Hospital Cancer Registry ............................................ 43 REFERENCES ....................................................... 45 v THE HOSPITAL CANCER REGISTRY Registries of various diseases and types have been organized and operated in hospitals and health departments for many years. The purpose of some registries, such as tissue and bone tumor registries, are chiefly educational and reference. These special purpose registries are ' limited to the collection and analysis of data on a single type of cancer. The epidemiological registry principally develops information from a large volume of data from many hospitals about the prevalence, incidence, and survivorship of various sites and types of cancer for research purposes. Such a registry, preferably, should cover the total population in a cir- cumscribed geographical area. Finally, the service registry is designed primarily to evaluate and improve patient care in the local hospital And community. Epidemiological registries of course, may also provide this service to their participating hospitals. 1. Historical Perspective The American College of Surgeons has conducted a program of periodic inspection and evaluation of cancer facilities since 1930 to better the care of cancer patients by early diagnosis, improved treatment, And informed follow-up through a coordinated cancer activities program. The medical profession and hospital governing boards have sought and accepted the desirability of such official recognition by the College. The College's Commission on Cancer, com- posed of fellows of the College, liaison members representing pro- fessional associations of physicians, radiologists, and pathologists, the American Cancer Society, and government agencies, acts in an advisory capacity in approving hospital cancer programs. In order to enable hospitals to evaluate their cancer workload and the quality of medical care provided patients, the College en- couraged the organization of hospital-based cancer registries. In 1956, registries were made one of the conditions for approval of 'a hospital's cancer program (see Appendix I for the Basic Standards for Cancer Programs). (1,2) At present, there are about 850 College ap- proved cancer programs in the United States and Puerto Rico. About 800 of the pro grams are in general hospitals, with about 700 in non- federal hospitals and the remainder in federal hospitals .(70 percent Veterans' Administration). The 800 hospitals have a median bed capacity of about 350 beds, and about 75 percent have bed capacities of over 200 beds. The non-federal general hospitals constitute about 12 percent of all short term general hospitals and about 30 percent are affiliated with medical schools. (3) gistry II. Definition of a Hos2ital Cancer Re A hospital cancer registry, sometimes called A tumor registry, has been described as a "mirror" which can reflect to the hospital staff its overall and individual performance in the diagnosis and treatment of cancer patients. To provide valid information, the registry must consist of records of all cancer patients diagnosed or treated at any of the facilities of the hospital as of a given date. Patients diagnosed elsewhere who received any therapy for their malignant condition in the hospital are to be included, as well as those initially diagnosed at the hospital or its clinics. The cancer registry should include outpatients as well as inpatients, and patients with only a clinical diagnosis as well as those with pathological confirmation of their disease. The registry record,, usually referred to as the abstract form, should contain pertinent information on the diagnosis, treatment, and follow-up of each patient, on a continuing basis until death, for each primary malignancy. This information is obtained principally from the patient's hospital medical chart. It is therefore imperative that the medical chart be readily available to registry personnel within Ia reasonable time after the patient's discharge from the hospital, or each outpatient visit. 'Incidentally, since the registry is dependent on complete and accurate medical charts for its information, it may serve as an impetus to up- grade the content and quality of all medical records in the hospital. III. Obiectives of a Hospital Cancer str A well organized and operating registry should: A. Assist physicians in the provision of continuous medical.care of the cancer patients with reminders to them, or to patients with icians's consent, of the need for re-examination annually, the phys or more frequently. Follow-up exapinations should continue during the life-time of the patient regardless of such eventualities as change of residence, or retirement or death of the physician. This is necessary to insure medical care for many patients-who might otherwise not be seen by a physician, and to help in the early diagnosis of local recurrences, metastases, or new primary lesions, and possibly to further their survival and comfort. B. Provide the hospital staff with annual or more frequent sta tistical and analytical reports which evaluate the cancer problem in the institution and community, by site and histologic type, extent of disease (stage), methods of diagnosis, treatment modalities, and survival by age, race and sex. Meaningful reports may lead to the adoption of measures to improve the management of cancer administrators with their scheduling and patients, and assist operational problems. Such reports may also assist in the de- velopment of comprehensive cancer programs. The roportion of p 2 patients successfully followed should also be reported, since thi s is an index of the statistical reliability of 'the survival data, and a measure of the quality of patient care. C. Be a resource for the continuing education of physicians and paramedical personnel at regular clinical conferences, medical society meetings, seminars, and institutes. D. Be a resource in the development of public educational programs in the ed by the registry. geographic area serv E. Be a stimulus and resource for clinical inve stigations and research by hig ig ting areas which require further stu y. IV. Components of a Cancer Registry Although the individual forms and files in cancer registtIies are not standardized,' most registries consist of separate or combination files to facilitate the identification, follow-up, and tabulation of patient information. (4) On August 5, 1968, the Division of Regional Medical Programs brought together an ad hoc committee of representatives from the American College of Surgeons, the End Rd sult8 Section of the National Cancer Institute, the American Cancer Society, several operating registries, and consultants to discuss recommendations for items of information to be collected by cancer registries. These were presented at a cancer registry work- shoo held in Denver, Colorado, on September 17, 1968 (see Appendix II). In general a cancer registry consists of: A. a primary site file of abstracts of significant information about the history, diagnosis, treatment, and end results of each primary cancer, with follow-up notes during the life- time of the patient. (If the patient has multiple primarids there should be separate abstracts for each.) If this file is also to serve as the master control file and. follow-up control file, separate tabs should be attached to the forms to identify the primary site and to remind the registrar when the (living) patients are due for follow-up re-examin- ations. To serve these multiple purposes the abstracts should be filed alphabetically by patient name. If the master control file and follow-up control file are separate (see below), abstracts should be grouped by major primary sites and filed in alphabetical order. This permanent file of abstracts should contain the following minimum information: 1. the name, address, registry and hospital chart numbers, and the age or d ate of birth, race, sex, and marital status of the patient; 3 2. the dates of admission and discharge from the reporting hospitals; and address of a relative or other contact person; 3. the name ing the primary site of the 4. diagnostic informations includ 1 diagnosis f the diagnosis, the histologica cancer, the basis o iagnosis (by any means), and the if made, the date of initial d extent of disease (stage) at initial diagnosis; 5. the history of the cancer, i.e., when and where the cancer was diagnosed and treated before this admission, the type(s) and date(s) of treatment, and a note about any other primary site(s) of cancer for which the patient may have been treated; 6. the condition of the patient at discharge; 7. the name and address of the hospital and/or physician responsible for follow-up; and 8. periodic notations it least annually, of follow-up information concerning additional therapy and the status of the patient. B. a e file of every registered cancer patient, al permanent master control file enables the secreta to avoid duplicate accessions in the registry. The perman file co.uld consist of 3"x5" cards kept in alphabetical order, and should contain the following infor- mation: 1. the patients full name and address, and that of the spouse if married, or parents if a child; 2. the hospital medical chart number; 3. the patient's date of.birth, race, and sex; 4. the cancer diagnosis and primary site,, 5. the date of initial diagnosis; 6. the name and address of the referring physician; and 7. the (eventual) cause and date of death. C. file of living patients to remind the t should be followed. This file 4 could consist of 3"x 5"cards kept in alphabetical order by patient's name within each month of fellow-up. After each follow-up the patient's card is to be re-filed according to the month of next follow-up. After the patient dies his card should be destroyed. The file cards should contain the following information: 1. the patient's full name and address; 2. the patient's registry number; 3. the primary site of cancer; 4. the date of diagnosis; 5. the dates when follow-up information about the patient was obtained; and 6. the name of the attending physician or hospital to whom requests for follow-up information are to be sent. D. an accession resister, or list of all cancer inpatients and outpatients initially admitted to the hospital, preferably grouped by year of initial diagnosis and major sites. This can be useful for the preparation of administrative reports to measure the.cancer workload in the hospital, and for summary repor ts to the medical staff. [See Appendix III for a suggested cancer registry abstract form (prepared by the staff of the Arkansas State Cancer Commission), patient name and follow-up control cards, and a page from an accession register.] V. How to Organize a Hospital Cancer Registry A cancer registry is a self-contained but integral part of a 'hospital's cancer program, under the overran jurisdiction of the hospital's Committee on Cancer. A. The Committee on Cancer The hospital's Committee on Cancer should be a standing Committee, appointed by the medical staff from its membership and confirmed by the governing board of the hospital. It should include re- presentatives of the departments of surgery,- internal medicine, radiology, gynecology, general practice, and pathology; and may include representatives of other medical specialities concerned with the diagnosis and treatment of cancer. it should provide the impetus and over-all direction for the organization of a cancer program, including the conduct of cancer conferences, educational activities, and the operation of the registry. The 5 Committee must determine the objectives and scope of the regis- try, and'concern itself with such major questions as gathering, dissemination, and analysis of data, personnel, facilities, and other matters relating to its effective organization and operation. (See Appendix IV for Organization Plan for Cancer Programs. (5)) B. Developing Objectives The Committee on Cancer must develop the specific objectives of the registry and obtain the agreement and cooperation of the medical staff for their implementation. . In order to obtain approval of the American College of Surgeons the registry will have to meet their requirements outlined in the Basic Standards for Cancer Programs (Appendix I). Depending upon interests and needs, the Committee may also develop educational programs for the medical and paramedical staffs and the public, and en- courage clinical and laboratory research, using registry data as a tool and resource. C. Supervision of Registry Operations The registry secretary should be under the general jurisdiction of the Committee on Cancer, and the direct dailysupervision of a designated person who need not be a physician (the medical record librarian or tumor clinic secretary). If the superv sor is notla physician, then a medical consultant, usually the pathologist or radiologist, should be appointed for regular consultation with her. D. Estimating the Cancer Caseload An important consideration in, organizing . a hospital cancer registry is an estimation of the cancer caseload, preferably for at least five years. There is a rough relationship between the number of new cancer patients and hospital type (whether public and private), and hospital size For example, in Cali- fornia there was an average of about 75 new cancer patients per 100 beds in county hospitals, as compared with an average of about 160 patients per 100 beds in private hospitals in .1963. (6) The average number of cancer cases per 100 beds in 1964 through 1969 inclusive, by size of hospital, in Connecticut was as follows: 6 Average New Cases ,Number of Beds Number of Hospitals per 100 beds Total 28 139 Under 100 beds 3 114 100-199 beds 8 115 200-299 beds 6 133 300-399 beds 6 132 400 and more beds 5 159 .Source: Derived from information provided by the Connecticut Tumor Registry. There is also a rough relationship between the number of new cancer patients, total inpatient admissions, and size of hospital. The ratio of new cancer cases to total inpatient admissions by size of hospitals in California and Connecticut was as follows: California Connecticut New Cases New Cases Number of as Percent Number of as Percent Number of Beds Hospitals of Admissions Hospitals of Admissions Total 57 3.0% 26 3.2% Under 100 beds 8 2.1 3 2.4 100-199 12 2.1 8 2.5 200-299 14 3.5 6 3.0 300-399 5 4.6 5 3.6 400 and over 18 3.0 4 3.8 Source: Communication from California Tumor Registry, and de- rived from information provided by the Connecticut Tumor Registry. In the absence of more specific information these data may be used to develop crude estimates of the number of expected new cancer cases in general hospitals.' E. Cost Considerations The cost of a registry is directly related to the size of the cancer caseload, and the programs for which it was organized (patient follow-up, professional education, program planning, research, etc.). Some of the components of the total cost of a hospital cancer registry are: 1. the time of the supervisor of the registry personnel (the physician, medical record librarian, or tumor clinic secretary); 7 2. the salary of a registry secretary and other clerical person- nel needed to operate the registry; 3. overhead costs, including the use of a separate room(s) or part of a room, furniture, file cabinets, office equipment and supplies, telephone, electricity, postage, etc.; 4. the time of the medical consultants to the registry; 5. the, time of personnel in other departments in the hospital to make available the inpatient and outpatient medical charts, pathology and radiology reports, and any other information needed to complete the registry abstracts; 6. the time of public health or social service personnel to assist with patient follow-up; 7. the cost of special purpose registry forms; 8. the cost of special purpose books (medical dictionary, Manual of Tumor Nomenclature and Coding, etc.); travel costs to Attend training courses and workshops; 10. the cost of computer hardware or computer time, if the registry is computerized; 11. personnel to code and keypunch the data, develop software, and operate the computer; 12. the cost of preparing statistical and analytical reports, and their duplication and distribution. A survey in Ohio in 1967 found that the cost of maintaining a hospital cancer registry ranged from an average of about $8.50 to $11.50 per new case; in California this cost averaged about $10.00 per case. A survey of College approved registries in 1968 found that about two of every three able to respond had annual operating budgets of $10,000 or less, with a median of $6,285. (7) The average cost of accessioning a new patient was $7.60, and of entering follow-up information,'$3.44.(For a discussion of the factors to be considered in estimating these costs see Appendix V.) VI. Special Problems If the registry is to be other than a sterile repository of information and achieve its objectives and full potential, the following problems must be overcome: 8 A. the lack of interest, cooperation and involvement of the hospital medical staff, This usually reflects the failure of the Committee on Cancer to convey to the staff an appreciation of the purpose and value of the registry, and to see to it that the registry fulfills its service and educational potentials. The survey of College approved registries found that a majority did not provide physicians with regular reports of their manage- ment of cancer patients. Perhaps this is why most registries reported fewer than ten requests for data from physicians. (7) B. inadequate guidance and assistance to registry personnel in their daily tasks, Continuous guidance and assistance should be avail- able to abstract cases (often from inadequate medical charts), to deal with uncooperative physicians, and to develop routine reports for the medical staff. C. inadequate quality control. It is imperative that registry per- sonnel be checked routinely on their accuracy in abstracting and coding. The hospital Records Committee should develop and en- force criteria and regulations to ensure the availability of complete, accurate, and uniform medical charts from attending physicians within a reasonable time after discharge of the patient. D. inadequate utilization of registry data to audit the performance of the medical staff in their management of cancer cases, and as a resource in professional and public educational programs. E. the unavailability and high turnover of secretarial and techni- qal personnel due to competitive factors. Training of secre- taries is, for the most part, on-the-job or at infrequent and short-term workshops. (The University of California Medical Center in San Francisco offers up to two months of training which includes lectures in registry methodology, the medical and sociological aspects of cancer, elementary statistical methods and epidemiology, films on the diagnosis and treatment of cancer, extensive in-service training in abstracting, follow- up, indexing, and the preparation of reports, attendance at consultative tumor board meetings, and field trips. However, they can only accomodate a limited number Of trainees at any one time.) With respect to the availability of statistical personnel, programmers, and systems analyst for large registries, it should be noted that these persons are also in short supply and this is not likely to improve in the foreseeable future. F. the desirability and need to use mechanical or electronic equip" ment to process and retrieve data, The use of such equipment will depend upon the volume of cases, financial resources of the hospital, and the uses to be made of the information. In 9 general, a hospital cancer registry with fewer than 400 new cases per year can be operated manually for at least five years. After the caseload exceeds 2,000 active (living) cases it may be more economical and efficient to use automatic data@@ro- dessing equipment, particularly if it can also be used in other departments of the hospital. However, it must be emphasized that the use of such equipment by itself does not improve the quality of the abstracted information, but rather increases the possibility of errors. The effective utilization of such equip- ment will depend upon the availability of qualified programmers and systems analysts, and will significantly increase the cost of operation of the registry. G. the question of continuing financing, The registry must be assured of continuing financial support if it is to provide the services for which it was organized. Costs increase not only because of comp8titive'and inflationary factors in our economy, but also because of the increasing follow-up load even when the number of new cancer cases remain stable. Also, as physicians develop an appreciation of the use and value of the data in the registry, more staff time will be needed to provide additional services, tabulations, and analyses. 10 THE CENTRAL CANCER REGISTRY The American College of Surgeons does not have an official position regarding local, state, or regional central cancer registries. However, hospitals wishing to participate in such registries.are expected to 11 maintain or have available their own data on cancer cases so as to meet the requirement for approval by the College.'? (1) This is in keeping with the policy and fundamental purpose of the College's individual hospital cancer program "to provide the hospital staff members with a continuing record of what is being accomplished in cancer patient care in their hospital and to insure proper follow-up." (1) The service and educational benefits of individual hospital cancer registries can be enhanced when they are organized into a central cancer registry system. Properly organized, a central facility can promote and expedite patient follow-up, And provide participating hospitals with separate and comparative reports of their activities. If the central registry is population-based, it can also develop information on the incidence of cancer in the area, and engage in epidemiological research and special studies. I. Definition and Objectives of Central Cancer Registries . A central cancer registry is a coordinating facility of coopera- ting hospital registries in a geographic Area to collect, combine,, compare, and evaluate uniformly defined information on cancer patients which can: A. facilitate and improve patient follow-up; B. be a resource in the identification of community problems in cancer control,and the development of progra.ms to cope with these needs; C. measure and compare the quality of diagnosis and the effective- ness of various treatment modalities in the participating hospi- tals, separately and as a group; D. stimulate inter-hospital and area-wide educational programs for physicians, nurses, and technicians; E. provide a resource for community-wide public educational programs; and F. engage in epidemiological and cooperative clinical research efforts. II. Types of Central Cancer Registries Central registries vary in purpose, scope of coverage, and method of operation. Following are brief descriptions of four registries which reflect these differences: A. The Connecticut Tumor Registry organized more than 35 years ago, requires the reporting and annual follow-up of essentially all diagnosed cases in the state from 38 hospitals. The registry receives-abstracts on about 9,000 new cases, and follow-up reports on about 50,000 active cases each year. A fee based on credit points for completeness of each abstract submitted is paid to 31 community hospitals to h-elp defray the cost of opera- tion of the individual,.hospital registries, and as an incentive for the maintenance of quality reporting. Registry staff code and process the abstracts and follow-up reports, train hospital secretaries in the operation of their registries, assist them with follow-up, and prepare follow-up, letters for the hospitals for transmission to the physicians. The registry publishes in- formation about the extent and nature of the cancer problem in the state, and makes available indices to evaluate progress in bringing the disease under control. Data is developed on the number, characteristics, and geographic distribution of cancer patients to plan public health programs, and to formulate and test hypotheses concerning the etiology of the disease. The registry also provides information on trends on the extent of disease at diagnosis, survival experience, and cancer incidence to evaluate progress made over a period of years. B. The California Tumor Registry was established in 1947 and has grown, on a voluntary basis, to 57 participating hospitals. Many more hospitals wish to join this system but the number of parti- cipants has been limited in order to maintain the registry's excellent record of collecting uniform data-of high quality with the available staff. The Registry includes all 24 hospitals in Alameda County to make a population-based sys.tem in that County. The 57 hospitals report about 20,000 new cases, (about one-third of the total cancer caseload in the State), and follow about 75,000 cancer cases each year. The registry pays $3.10 to each hospital for each completed abstract. It is estimated that this fee covers approximately one-fourth to one-half of the hospital's total cost. The registry staff codes and processes the abstracts and follow-up reports, and prepares annual reports, diagnostic indices, and reports on the survival experience (with comparisons to the total registry), for each participating hospital. Data is also prep@red to answer physicians' requests,, for planning public, health programs, and for special studies and publications. related to cancer diagnosis, treatment, and survival. 12 C. The Rocky Mountain Stat es Cooperative Tumor Registry (organized in 1967 and supported by the Intermountain, Mountain States, and Colorado@Wyoming Regional Medical Programs) includes hospi- tals in Utah, Idaho, Montana, Colorado, and Wyoming. LargO. institutions submit abstracts of all diagnosed cancer cases to the Registry, and medical students abstract cases in small hospitals for the Registry four times a year. The Registry sends follow- up inquiries directly to the patients' physicians annually, and copies of their res.ponses to the hospital from which the abstract originated. Twice a year the Registry sends listings of the characteristics and status of the patients at last follow-up to the hospitals and the attending physicians. Various medical specialty groups are requested to review the literature and choose references to be included in the listings to the physicians, on a voluntary basis. Each group is also given an opportunity to use the registry data to analyze the malignancies with which they are most intimately concerned, and publish the results in the Rocky Mountain Medical Journal. The articles review the diagnosis, treatment, and survival of cancer patients in the area, and sug- gest methods for improvement. D. The Iowa Central Tumor Registry founded in 1965, was organized to provide continuing cancer education for physicians who take part in the cancer programs of the 54 participating hospitals. These hospitals admit more than 75 percent of the cancer patients in Iowa, and accessioned about 11,000 cases during lq69. More than 18,000 cases of cancer are being followed by the 54 liospi- tals. The central registry provides the following services: (1) storage, retrieval, and analysis of the cancer data collected by participating hospitals; (2) a semi-autothated follow-up system; (3) a field program which assists hospitals with the organization and evaluation of their cancer programs; and (4) financial support. The principal educational impact of the Iowa Central Tumor Regis- try is provided by the annual reports sent to each participating hospital in October. This report includes a tablulation of the cancer data collected during the preceding year by the hospital, their current survival results computed on the basis of all cases obtained from the hospital, and a cancer patient listing. Each participating hospital is also given a report of the combined experience of all hospitals participating in the registry to pro- vide them with a basis for evaluating their own experience. The registry also assists physicians in cancer clinical research and in providing cancer data for use in hospital cancer educational programs. The central registry biostatistician is available to assist physicians with the interpretation of the data. 13 III. Advantages and Benefits A properly organized central cancer registry has several dis- .tinct advantages over individually operated hospital registries. Among these are: A. improved uniformity and quality of data abstracted and coded in accordance with mutually agreed upon definitions. This can be ensured by continuous monitoring and training of secretarial and coding personnel in the participating hospitals and central facility. An added dividend is a general up- grading of the medical charts in the participating hospitals due to the requirements of the central registry for complete reporting. B. efficiencies that come with size. Centralization of coding, keypunching, programming, and computer usage can result in substantial savings when compared to the aggregate costs of these activities in separate hospitals. C. availability of technology and statistical personnel (which the participating hospitals could not afford individually) to assist physicians and hospitals with the follow-up of cancer patients, and the preparation of comprehensive and com- parative analysis of the management of cancer in each of the hospitals by pooling of financial resources or centralized funding. D. improved public and professional educational programs, and more useful analytical reports of the cancer control problem in the community due to the larger volume and diversity of cancer cases. E. the more immediate benefits that can accrue to patients and physicians. Routine requests for follow-up information by a registry promotes medical care of the patient. For example, following the suggestion of their Advisory Committee, the California Tumor Registry ranked the hospitals on their follow- up efforts (using only codes to identify each hospital). The proportion of patients on whom current follow-up information was received increased 19 percentage points from 71 percent to 90 percent within a two and a half year period. In a sample survey over a six-month period, the California Tumor Registry found that a little more than 60 percent of the patients had received at least one medical examination during the last year. If it is assumed instead, that 75 percent (or three-quarters) of the patients on whom follow-up information was obtained were seen by their physicians without stimulus from the registry, an 14 additional one-quarter were brought under medical care as a result of the registry's intensified follow-up program. Thus, in California an additional five percent (1/4 times 19 percent) of the total file of living patients were brought under medical supervision. With about 75,000 patients currently in the active file in the California Tumor Registry, about 3,500 addi- tional patients each vear are benefitting from medical care they otherwise might not have. Also, an evaluation and comparison of central registry data re- lating to the extent of disease at diagnosis in the participating hospitals may improve early case finding and treatment. In California the percent of patients diagnosed with early cancers (excluding skin) increased by six percentage points over a fifteen year period. Thus, with about 20,000 new cancer patients added to the registry each year, about 1,200 additional patients may be benefitting from earlier care. IV. How to Organize a Central Cancer Registry The organization of a central cancer registry is not too unlike the organization of an individual hospital registry. The proper organization of such a combined effort requires: A. medical leadership. A central registry is a cooperative arrange- ment among the medical staffs of partici ating hospitals. It .p should have the endorsement of local and state medical societies, other professional organizations, the state health department, hospital administrators, and voluntary agencies. The leadership for such a combined effort may come from physicians from one or (preferably) more hospitals, local medical societies, and the state health department. B. a medical advisory committee. The central registry itself re- quires a representative professional advisory committee to con- sider policy and operating questions which relate to the interests of;the participating-hospitals as well as the objectives of the centralized facility. C. a definition of objectives. The advisory committee will have to define the objectives of the combined effort in the light of the specific needs and objectives of the participating hospitals. They will have to consider whether the data in the registry will be used to evaluate questions such as, the referral patterns of the different kinds of hospitals (community, specialty, and in medical centers), the diagnostic and therapeutic resources in the participating hospitals, and other questions concerning the utilization of health care facilities in the area. The advisory 15 committee will also have to consider whether they can or even want to operate a registry which will assist in the follow-up of patients, engage in comparative studies and the evaluation of diagnosis and therapy, and epidemiological research, versus one with more limited objectives. In this connection it should be noted that the potential for epidemiological research is not, in itself, sufficient justification for the establishment of a central registry. While it is desirable that several states in different parts of the country develop such registries for purposes of comparison, epidemiologists have long noted that research needs do not require such registries in all states in the country. (8) D. consideration of the scope of coverage And total caseload. A central cancer registry can: 1. include all hospitals in the state (or region), 2. consist of a selected number of hospitals in the state, 3. consist of hospitals in A local and limited geographical area (county or city), or 4. consist of a tonferation of several local central cancer registries. Whatever the scop e and caseload a central registry mu t e nutured patiently and carefully over a period of years (at least five) before it can prove its effectiveness in the pro- vision of services and useful information. This can be accom- plished only if the participating,hospitals are incorporated and phased into the system methodically over a period of time. The number and bad capacities of the participating hospitals will determine the total caseload in the registry. The methods sug- gested on page 6 can be used to estimate the caseload in the central registry. An estimate of the increasing number of cases to be followed each successive year will also be necessary. If we assume that the number of new cases will remain constant from year to year, the total active (living) caseload to be followed, based on the survival experience of the National Cancer Institute .End Results Group, will be about three times as large by the fifth year of operation of the registry, about four and a half times by the tenth year,' and about five and a half times as large by the fifteenth year of operation of the registry. E. consideration of the method of operation'. A central registry can acquire data by having the participating hospitals send copies of their abstracts, or the registry can send out circuit-riding 16 abstractors to the hospitals, or use a combination of both methods. Of paramount,importance, however, is the development of mechanisms to obtain total reporting from the hospitals with uniform interpretation of the reported information. The regis- try may ensure consistency of information with periodic reviews of the procedures of the participating hospitals, re-abstracting of a sample number of medical charts, and training sessions and workshops of registry personnel and medical advisors. (9) Without such quality control measures the accumulated data could be meaningless. Another aspect of the method of operation to be considered is the question of manual operation of the registry versus opera- ti.on with mechanical or electronic equipment. @A registry with a large number of cases will have difficulty handling the volume of reported data, and servicing the hospitals without the assistance of automatic data processing equipment. Auto- mation of registry files can assist hospitals with the follow- up of patients, and greatly enhance the usefulness of the data in the form of frequent reports to physicians and hospitals. It should be noted, however, that the need to accurately transform (code) and keypunch data for computer processing can be one of the most time-consuming, difficult, and expensive procedures. F. consideration of the cost of a central registry. The objectives of the registry and their effective implementation, the 'number of physicians and hospitals participating in-t@,e program, and the size of the caseload bear on the cost of op .&ration of the registry. Some of the components of the cost O'f.a central r6gis- try are: 1. the employment of a qualified supervisor (who may be a statistician); 2. personnel needed for the-routine operation of the registry (s6cretarial.- clerical, coders, keypunch operators); 3. technical personnel (statisticians, systems analysts, programmers); 4. computer hardware or computer time; 5. overhead costs.which includes the use of one or more large rooms for the supervisor, the other registry staffi and the computer;'the necessary desks, chairs, ordinary and special purpose file cabinets; office equipment,-such as calculators, adding machines and supplies, telephones, postage, electricity, etc.; 6. the time of the medical advisory committee; 17 7 special purpose registry stationery, forms, binders, etc.; 8. special purpose books and manuals; '9. travel costs for visits to participating hospitals to abstract and/or review data, to attend training workshops, professional meetings, etc. The California Tumor Registry estimates that the current cost associated with the routine operation of their registry (in- cluding follow-up) is about $10 per new case. This includes salaries and fringe benefits (in a relatively high labor cost area), computer and other data processing costs field visits of registry personnel to ensure the quality of the abstracts from the hospitals, the cost of obtaining copies of.death certificates, and overhead costs (rent, furniture, office equipment and supplies, telephone, electricity, postage, etc.) A new registry with a small follow-up load will cost considerably less per new case in the early years of operation. The Cali- fornia Tumor Registry also estimates that it costs another $4 per new case to report to each hospital on its own experience; to answer requests for data from physicians, hospitals, and other agencies; to develop research studies; and to prepare repor ts and publications for distribution. It must be emphasized that these estimates do not include.fees or subsidies to hospitals to cover their costs of reporting to the central registry, or the cost to the participating hospitals of maintaining their own registry. V. special Problems The organization and operation of a central cancer registry pre- sents several special problems. Among these are: A. the need to stimulate and maintain the interest and involvement of the physicians in the participating hospitals. This may be accomplished by the representatives of the hospitals by communi- cating to the registry the interests and needs. of the physicians. Physicians must see t'he results'of their efforts, and made to feel members of a team effort. Periodic reports about their specialty and the activities of the hospitals in which they practice, and the organization of a professional educational program which shows how they relate to the total effort of cancer control in the area can stimulate their involvement. Rapport between physicians in the participating hospitals and the central registry'can be promoted if their reciprocal responsibilities and obligations are explicitly stated and agreed upon at the outset. B. the problem of effective quality control. The maintenance of quality control is most important in the operation of a central cancer registry (see section'IV E on page 16 above for suggested techniques). Without such control the data and any resultant analysis is of dubious validity. C. the availability of qualified personnel. A central registry that has more than minimum objectives requires an epidemiolog- ist, statisticians, systems analysts, and programmers. These persons are in great demand in health and other areas, and there is little likelihood that this situation will improve in the-near future. Thus, the shortage of technical personnel may short-circuit any projected benefits envisioned by the organi- zation of a central registry. Consideration must also be given to the availability and retention of other trained registry personnel and clerical staff since they play a vital part in the daily operations of the registry and maintaining the quality of the data. D. the availability of long-term financing. Proponents of a central cancer registry must keep in mind that costs will con- tinue to rise especially because of the increasing volume, and increasing demands which will be made on the registry as its usefulness is recognized. Long-range financing arrangements must therefore be made for continuation of the facility if it is to bevorth the initial efforts. Also, without assurance of long-range support there is likely to be abnormal turnover of personnel. This may require operating short-cuts to the detri- ment of the quality of the data, and the service and educational benefits. 19 REGIONAL MEDICAL PROGRAMS AND CANCER REGISTRY ACTIVITIES I. Regional'Medical Programs Public Law 8.9-239, enacted on October 6, 1965, authorizes the establishment and maintenance of Regional Medical Programs to assist the nation's health resources in making available the best possible patient care for heart disease, cancer, stroke, and related dis- eases. Through a system of grants the law attempts to provide the means for conveying to medical i'nstitutions and the professions the latest advandes in medical science for the prevention, diagnosis, treatment, and rehabilitation of patients afflicted with t ese diseases. The grants assist in the establishment of regional co- operative arrangements among medical schools, research institutions, hospitals, and other medical institutions and agencies to achieve these ends by research, education, and demonstrations of patient care. Since the enactment of the law representative groups have organized themselves to conduct Regional Medical Programs in 55 regions using functional as well as geographic criteria. Regions include combinations of entire states, portions of several states, single states, and portions of states around a metropolitan center. (10) II. Grants for Cancer Registry Activities Between June 1966 and the end of February 1970, about 180 million dollars were awarded (not spent) to the 55 Regions for all'planning and operational activities. Of this amount, about 128 million dollars were awarded to 53 Regions for project grants since April 1967; about two million of which has been made available in the current program period to 18 Regions for twenty operational projects with cancer registry components, in whole or part... Addi- tional funds have been spent by some Regions for cancer registries for purposes of program planning. Except for the Rocky Mountain Co- @operative Tumor Registry (page 13) registry projects are in the begin- ning stages of organization and operation, hence it is too.early to evaluate their progress. Additional cancer registry proposals have been submitted for funding by Regional Medical Programs and others are considering doing so. III. Alternatives for Support and Benefits The decision to engage in cancer registry activities should be made only after the Regional Medical Program has carefully consid- ered the purpose and use of dancer registries in a compr ehensiveiand cohesive cancer program. A Region may wish to develop a limited registry, whereas another may wish to develop on& which can provide extensive services. The choice should only be made after a careful assessment of needs, and a realistic evaluation of available re- 21 sources. Regions thinking about promoting cancer registry activities, may wish to consider one or more of the following options: A. To make available technicians to assist hospitals in the organi- zation and operation of their individual cancer registries. Such technicians must have training in registry operations, and have some background in the preparation of statistical tabulations and reports. To service large hospital registries technicians also require competence in the development and analysis of special studies, and the use of automatic data processing machines. B. To provide financial assistance to hospitals to enable them to organize and operate their registries. C. To provide financial assistance to one or more local (count@) central registries (preferably population-based) which would receive or obtain reports on all cancer cases in the participat- ing hospitals. Central registry personnel should be available to assist hospitals with the organization and operation of their registries, the analysis of statistical data,, and the development of special studies. D. To provide financial assistance for a Region-wide central cancer registry (preferably populdtion-based) which would receive or obtain reports on all dancer cases in the participating hospitals. Personnel resources should be the same as that for local central registries noted in item C above. Some benefits of the separate alternatives are, Support of Support of individual Support of Region-wide Support of hospital local central central Benefits('X) technicians) registries_ registrv(s) registry 1. Ready ac- x x Where hosp. Where hospital ces to data registries registries by hospital organized organized staff 2. Encourage x x More likely More likely medical with hospi- with hospital staff par@ tal registry registry ticipation & involve- ment (Continued) 22 individual f Region-wide Support o ,lsupport.of hospital local central central registry(s) rep Berxifits (x) technicians) istr x x 3. Training x resource x x 4. Provide additional assistance in follow- ing pa- tients @imited Limited x x 5. Develop experience reports, experience analysis, and and studies resources resources I x x 6. Develop- x ment of uniform data among hospitals 7. Preparation x of compara- tive reports of hospitals x x 8. Preparation of combined reports of hospitals Reflects x gt Resource to develop local area-wide needs cancer pro- gram . e Limited Limited Reflects x 10. Resourc to local for pro- to needs fessional hospital hospital educational programs (Continued) 23 Su individual Support of e upport of hospital .!local centra i St ie registry(s-) Benefits(x) IReflects x 11. Resource f, !,local public ineeds educations programs 12. Engage in epidemiologica x research It appear therefore that maximum benefits may be obtained by s way of local and/or region-wiae population-based central regis- tries. In the last analysis.the decision to support cancer registry activities, and the extent of such supports will depend upon their relative importance as compared to other health needs in the region, the leadership and interest of the medical communi- ty, and the availability of trained personnel and financial re- sources at the time of initiation of such activities, and in the future. IV. uestions Consi4ered for Su Ort As has been indicated, the proper organization of a cancer registry is no simple matter and must be approached with much thought. Vollow- ing are some of the question which staff of the Division of Regional Medical Programs and reviewers consider when evaluatin .g proposals for support of a registry: A. How does this new registry activity or expansion of an existing registry fit into the overall cancer program iii the Region? B. Are the objectives clear with reference to: 1. patient services, 2. follow-up services-for physicians and hospitals, 3. the number of physicians that mi'ht benefit from professional 9 educational programs utilizing registry data, 4. whether the registry activity will attempt any unique services to patients, physicians,, hospitals, the community, (with examples of such possible services), 5. how the registry will fulfill a regionAl and/or national need, 24 registry data in public educational programs 6. the use of and possibly, ticipated? 7. the kinds of research studies an C. Does the proposal include documentation or other evidence of cooperative arrangements with: 1. medical societies (county, state), 2. the administrators and staffs of participating hospitals, 3. other professional organizations (pathologists, radiologists, surgeons dental society, etc.), and 4. paramedical groups and voluntary organizations? D. Will the medical advisory group of the proposed registry (which will consider registry policies and operating questions) be representative of the participating hospitals and professional groups? E. How many hospitals are to be included in the.central registry, .how many hospitals have,cancer registries presently,.what is the estimate of the cancer load in each of the participating hospitals, and the anticipated combined cancer load over a five year period? F. What will be the composition of the personnel, both technical and auxiliary, available to the central registry? G. What mechanism is to be used or developed to train personnel in participating hospitals, and to review the completeness and accuracy of the abstracts they will submit? R. How will additional hospitals be phased into'the system, and at what rate? I. What kinds of automatic data processing equipment will be used, and what is the basis for the selection-of the equipment? J. Will competence in the development of software be required, what personnel or time will be needed for this, and the cost'? K. What are the justifications for the budget data-for personnel, space, furniture, equipment, supplies, travel, etc.? L. What other sources of support will be available during and after funding by the Regional Medical Program? 25 v. consultation Bef re A Regional Medical Program submits a'regist y proposal to the 0 r Division of Regional Medical Programs for funding it is suggested that they try to profit from the experience of previously funded and existing registries. Technical staff is also available from the Division to assist in the development of hospital and central cancer registry projects, and the training of registry personnel. Requests for such assistance may be addressed to: Chief, Operations Research and Systems Analysis Branch Attention: Mr. Abraham Ringel Division of Regional Medical Programs Health Services and Mental Health Administration Parklawn Building, Room 10-49 5600 Fishers Lane Rockville, Maryland 20852 Telephone Number (301) 443-1800 26 Appendix I tand ards for Cancer Programs (2) Basic S A. Accreditation of hospital There should be a clear understanding that the 1. Only those hospitals which are accredited ultimate responsibility remains with the pa- by the Joint Commission on Accreditation of tient's physician. Hospitals will be considered for approval. 5. Clinical conferences will be held regu- Should a hospital forfeit it Y. Consultative sessions s accreditation after larly, preferably weekl receiving approval of its cancer program, ap- should be held between regularly scheduled proval will be withdrawn. The hospital cancer conferences when necessary, to avoid delay in program will be re-evaluated upon request instituting treatment. The conferences will when accreditation is restored. provide consultation service only, or consulta- 2. In the evaluation of nonhospital medical tion and treatment service, according to the institutions with cancer programs, only those policy established by the medical staff and con- which are certified by the county or state medi- firmed by the governing board of the hospital. cal societ3r will be considered for approval. a. All members of the hospital staff, includ- ing interns and residents, and physicians B. Hospital committee. The clinical cancer in the community, should be encouraged activities program is to be under the guidance to attend the conferences. of a committee on cancer composed of repre- b. Minutes of the sessions should contain a sentatives from the several-medical specialties record of the attendance, the cases con- concerned with the diagnosis and treatme nt of sidered, and other pertinent information. cancer. It is suggested that representatives of 6. The cancer clinical activities committee the departments of surgery, radiotherapy, should initiate professional educational a pathology, internal medicine, gynecology and ties and encourage clinical research in cancer. general practice be assigned to the committee, It should render an annual report including an with additional representatives as desired. evaluation of the data contained in the cancer registry, a summary of the minutes of the clin- C. Clinical program. The American College of ical conferences, and recommendations lead- Surgeons has established the following minimal ing to improvement in cancer control. standards for hospital and other recognized in- 7. When therapy is part of the total pro- stitutional cancer programs. The College reeog- gram, adequate facilities and equipment shall nizes that local situations may call for modi- be available for diagnosis and treatment. fications which facilitate operation of the pro- S. Responsibility for local fiscal support of gram, and makes provision for indicated vari- the cancer program rests with the individual ations which do not alter the established basic hospital. There is no general formula to fit all principles and policies. institutions. Financing of the program should 1. The clinical program is to be conducted ultimately be a part of the hospital budget. under rules formulated and approved by the medical staff and confirmed by the governing D. Registry. An institution-wide cancer registry body of the hospital or clinic involved. shall be in operation: 2. Clinical cancer conferences and cancer 1. To provide service to the patient by assur- educational activities are conducted by an ap- ing lifetime Interval follow-through examinations, pointed group which may be the same as the regardless of eventualities such as change of hospital committee on cancer. A member of residence, retirement or death of the patient's the group shall be appointed as director. Aux- physician, as well as change of residence of the fliary professional and secretarial personnel patient. A registry has the administrative capa- are to be assigned to the group according to bility of reminding both the physician and the the needs for efficient operation. patient that it is time for a re-examination. S. Nonprivate patients with cancer or sus- Meaningful follow-through can be accomplished pected cancer should, as a matter of policy, be only by a thorough examination by a physician. referred to the cancer clinical conference for This facilitates early recognition of local recur- consultation. rences and metastww and early diagnosis of a 1 4. Physicians should be encouraged th pre- new primary cancer. sent their private patients to the conferences. (Continued on next pagIe) 2 7 Appendix I (Continued) 2. To provide the hospital staff with statistical reports on site, stage, method of diagnosis, treatment and results for all patients with cancer treated in that hospital. Only if the follow-through examinations are maintained at a level approxi- mating 100 per cent, will the accrued statistical data be meaningful. E. Reports on survival and end results. 1. Reports based on data obtained from the cancer registry are to be presented at least annually to the hospital staff. The reports should include analyses of data on survival and end results for various types of cancer. 2. The periodic reports based upon register data will serve as a guide for the care of cancer patients within the hospital and will be useful in developing the ov6r-all.hospital cancer program. These reports are often a stimulus,for clinical investigations and research by pointing out the areas in which studies are especially indicated. 3. The hospital will have available, at the time of survey of the cancer program, evidence that periodic reports ate being submitted to the hospital staff. A copy of the report must be submitted to the field representative of the College at the time of survey, together with a description of the method of distribu@ tion and presentation to the staff members. These reports based on statistical data from the cancer registry, will be given great weight by the Commission on Cancer in evaluating programs. 28 Appendix II ITEMS OF INFORMATION FOR,CANCER REGISTRIES On August 5, 1968, the Division of Regional Medical Programs brought together an ad-hoc commi.ttee of representatives from the American College of Surgeons, the End Results Section of the National Cancer Institute, the American Cancer Society, several. operating registries, and consultants to discuss recommendations for items of information to be collected by cancer registries. a represents the general consensus of The information that follow s a working paper for consideration this committee and is intended a by on-going registries, organizations, and institutions planning regis- try programs, and by various national and regional bodies. It is not intended as a set of criteria for approval of a registry by any recognized national body. The items listed have been grouped into two categories: Coke items and Optional Items. Whereas the Core Items are generally considered desirable, they do not represent minimum requirements one member of the ad-hoc committee suggests shifting of several items from Core to Optional and vice versa. 'For example, he suggests that the history of diagnosis and treatment of each prior cancer be listed as Opt ional, but that performance status be listed as Core. Another member feels that performance status be listed as Core. Another member feels that the list is a compendium of information that should be in the hospital record rather than a guide for developing a cancer registry abstract. Obviously, the amount of information to be abstracted routinel@ must be geared to the purposes of each individual registry. dditional informa- tion can be abstracted on selected series of cases on a special study basis, in order to answer specific questions. In order to implement the collection of the information listed, a registry program will have to develop specific operational procedures, definitions, And codes. The development of basically uniform,definitions and.codes to facilitate: pooling and comparison of data is a desirable goal. Experience in a number of well-established registry programs provides a basis for developing.appropriate guidelines to promote uniform- ity. 29 Appendix II (Continued) CER REGISTRIES ITEMS OF INFORMATION FOR CAN CORE ITEMS Not required at Central Registry Not required at Local Registry A. Id6ntification: 1. Hospital name or code number 2. Patient a. Name surname, first, middle husband's first name (changes in name) b. Chart number(s) C. Hospital registry accession number d. Central registry accession number e. Address - street, city, (county), state, ZIP code @(changes in address) f. Phone number (and changes) g. R61ative(s) or other contact(s) 1) Relationship 2) Name, address, phone number h. Race i. Sex J. Date of birth month, day, year k. Age at admission for present cancer 1. Marital status single, married, widowed, divorced, or separated 30 Appendix II (Continued) B. History: 1. Prior other cancer (e*cept for non-melanotic skin cancer) a. No, yes b. If yes 1) Number of prior primaries 2) Diagnosis, date, treatment, and place of treatment for each 2. Prior diagnosis of present cancer a. No, yes b. If yes 1) Name of hospital or physician 2) Diagnosis (site and type) and date 3) Method of diagnosis -- histology, hematology, cytology, x-ray, clinical only, other (specify), not reported 4) Was treatment given? a) No, yes, not reported b) Type of treatment and date(s) C. Diagnosis (present cancer): 1. Sequence number (excluding prior non-melanotic skin cancer)-- One rimary only p First of two or more primaries second or later primary Unspecified sequence number 2. Primary site minimum detail as per ICD (8th revision), or as per Manual of Tumor Nomenclature and Coding, 1968 revision 31 Appendix II (Continued) 3. Date of initial diagnosis (may be clinical) --month) year 4. Confirmation of diagnosis -- histology, hematology, cytology, x-ray, clinical only, other (specify), autopsy, not reported 5. Histopathologic diagnosis a. @ologic type __ detail as per Manual of Tumor Nomenclature and Coding, 1968 revision b. Date of histopathology 6. Extent of disease --assessment of extent of disease at initial treatment based on all information avail- able during first course of treatment a. Sumary classification In-situ Lotalized, i.e.@ has not extended beyond primary site Regional Regional node involvement Direct extension to adjacent tissues Regional nodes plus direct extension 'Not.otherwise specified Distant or diffuse spread b. nt of extent of disease histopathology, Basis of assessme surgical explora- tion, x-ray, clinical only9 other (specify) D. Treatment: 1. Fi.rst course include all tum6r-diredte4 treatments that 'Were part.of the initial attack on the cancer. Exclude any Itreatment given because the first prescribed course of therapy failed. a.- Date of initiation of tumor-directed treatment b.' Identify each type of treatment given and date initiate The major types.of treatment are: surgery, beam radiation, other radiation, chemotherapy, hormone therapy, endocrine surgery, and endocrine radiation. 32 Appendix II (Continued) 2. Subsequent tumor-directed therapy a. Record as per first course b. For coding purposes it is sufficient 'to combine all subsequent treatment to identify types given E. Follow-up: 1. Date of contact (or death) 2. Type of contact, e.g., medical examination, letter, phone call 3. Vital status alive or dead Disease status at last contact or death (including autopsy findings) -lio evidence of any cancer In remission Evidence of cancer Residual (never free of this cancer) Reappearance of this cancer Other cancer present, but no evidence of this cancer Cancer present, but origin not known Unknown 5. Cause of death a. Per death certificate b . Per best available information, including autopsy findings; indicate source 6. Survival time -- years and months from date of first diagnosis 7. Physician (or clinic) responsible for patient follow-up 33 Appendix II (Continued) OPTIONAL IT EMS* A. Identification: 1. Location of hospital -- city or county 2. Patient a. Maiden, name b. Social security number -- may be used in lieu of accession numbers c. Employer -- name, address, phone d. Insurance company e. V. A. claim number f. Items o'f epidemiologic interest,.e.g., occupation, county of birth of patient and patents, menopausal status, etc. B. History: Prior skin cancers other than melanoma Same information as for other prior cancers 2. Delay (months elapsed) -- various intervals may be computed by recording: a. Date of first symptoms b. Date first sought medical advice c. Date of first diagnosis d. Date of initiation of greatment C. Diagnosis: 2. Detailed description of location of primary tumor 3. Multiple tumors within primary site information on multip e tumors should include location and histology of each *Outline letters and numbers relate to the letters and numbers under Core Items. 34 Appendix II (Continued) a. At initial diagnosis b. Over time (dates) 5. Histopathologic diagnosis a. Size of tumor in cm. b. Descriptive summary (including type of specimen) C. Identification of laboratory or pathologist d. Slide numbers 6. Extent of disease a.- A more detailed descriptive scheme may be used provided it is compatible with the summary classification. b. Detailed description in text form or via A check list (including bases of assessment of spread to different parts of the body) 7. Clinical assessment of extent of disease a. Summary classification per American Joint Committee b. Detailed description in text form or via a check list D. Treatment 1. First course a. Description of each type of treatment, including extent of surgery; radiation fields and dosage; specific chemo- therapeutic agents, route, and dose; and date of completion of each course. b. If no tumor-diretted treatment was given, state reason C. if other than optimal type of treatment was given, or if treatment plan was modified, give reason 2. Subsequent tumor -- directed therapy Description of each type of treatment (as per la above) .35 Appendix II (Continued) 3. Supportive therapy -- description and dates of non-tumor- directed treatments, such as by-pass surgery and blood transfusion, par- ticularly when this was the only treatment given or when it preceded the first tumor-directed treatment E. Follow-up: 1. Date of first reappearance of disease, or statement that patient was never free of disease 2. Time elapsed (years and months) from date of initiation of treatment to first reappearance of disease 3. Performance status -- at each hospital discharge, or At each contact a. Classification Normal activity Asymptomatic Sympotomatic Unable to work Capable of selfcare Not capable of .4elfcare Severely disabled Not terminal Terminal Dead Not reported b. If disabled, is disability primarily due to other di sease yes, no, not reported 5. Cause of death -- summary of autopsy findings 6. Survival time -- years and months from date of initiation of treatment 7. Other interested physician 36 Appendix III Cancer Registry A.bstract NAME (Lost) (First) (Middle) (Spouse)' State Registry No. (leave blank) ADDRESS (No. and Street or R.F.D.) (City) (County) (State Zip) (Phone No.) Social Security No. MARITAL STATUS Hospital Registry a. DATE Of BIRTH SEX (chock on*) RACE (chock one) Ole -white -Never married Omole -Negro arried *me of Hospital an @n. Indian Hospital Chati No. e Other, w-@ I- - NAME, ADDRESS, AND RELATIONSHIP OF TWO CONTACTS Date of Admission (Month, day,-and year) 1. Date of Discharge (Month, day, and year) 12. PRIMARY SITE OF THIS TUMOR HISTOLOGICAL DIAGNOSIS (summary of pathology report) BASIS OF DIAGNOSIS Autopsy -ray -Originally clinical or X-ray, microscopic confirmation ater -Histology -Clinical Only -Unknown STAGE OF DISEASE (chock one) I- -in SiN agional with regional nodes plus C3 alized direct extension egional with direct extension to agional, not otherwise specified co, adjacent tissues Distant or diffuse spread Regional with regional nods -Not recorded involvement C', (Name and FIRST DIAGNOSED 21 PLACE FIRST DIAGNOSED or Hospital) (Month, day, and year) 93 PLACE(S), TYPE(S), AND DATE(S) OF TREATMENT PRIOR TO THIS ADMISSION C3, 2c TYPE(S) OF TREATMENT AT THIS ADMISSION: rgery -Ucm radiation Record actual treatment: -Other radiation -Chemotherapy ormone therapy Endocrine surgery ndocrine radiation -Othor-dirod*d -other treatment, none of abo" a treatment, s@ roaso NAME AND ADDRESS OF ATTENDING PHYSICIAN INTERVAL TO FOLLOW-UP (chock *no) - 3 months NAME AND ADDRESS OF REFERRING PHYSICIAN - 6 months IC," 12 months Name of Person Submitting Report Dot. R*Vtewed A.D. FOLLOW-UP INFORMATION PATIENT STATUS AND STAGE OF DISEASE LATEST CONTACT REMARKS REASON FOR NO FOLLOW-UP c Enter summary of subsequent fro@ent, giving -00 0 date, place, and type of treatment. c 0 4 x IL -0. 0. > z 0 co C. a 0 -E .2 CL IL to 0 E 10 Co 0 0 -0 -4 0 0 v .0 t -0 0 c -C . i g 0 -c Ot0 .0 0 .2E DATE 0 Coe z E fD :3 ci 0 m C6 *SOURCE OF CONTACT- **QUALITY OF SURVIVAL H-hospital readmission A-capable of normal adivity.- Asyynptomatic C-dinic visit S-capable of normal activity. Symptomatic D-d@r's office with ph@ical examination C-incapablo of normal adivity: Capable of self-wre NP-dodor's office Without p@ical examination NC-incapable of normal @vity; Not capable of self-mrs N-public health nor$* @verely disabled: Not terminal P-dired patient contact or, @fy in Remarks column T@erely disabled.- Terminal CAUSE OF D UTOPSY: Ye, No Not ft ATTENDING PHYSICI REFERRING PKYSICI PHYSICIAN OR HOSPITAL RESPONSIBLE FOR FOLLOW-UP Appendix III (Continued) PATIENT INDEX CARD 'NAME HOSP. NO. ADDRESS REG. NO. - RACE BIRTH DATE - SEX D IAGNOS I S DATE OF DIAGNOSIS REFERRED BY CAUSE OF DEATH AND DATE FOLLOW-UP CONTROL CARD NAME: REG. NO. ADDRESS: SITE: DATE DIAGNOSED- DATE OF LAST DOCTOR TYPE OF REMARKS. FOLLOW-UP FOLLOW-UP 39 PRIMARY SITE(S): ISC NO(S). CASES DIAGNOSED FROM 19 TO 19 REGI ISTOLOGICAL DI SUR- CA PATIENTIS NAME OR VIVAL HUM BASIS 0 TIME MOS. T 0 T L S 1. THERAPY THAT AFFECTED THE CANCER: S-SURGERY. R-RADIATION, C-CHEMOTHERAPY. H-HORMONES 2. W-WITH CANCER@ WO-WITHOUT CANCER. U-CANCER STATUS UNKNOWN. DOC-DEATH FROM OTHER CAUSES (AnD CANCCR STATUS) Appendix IV Organizational Plan for Cancer Programs (5) Governing Board Administrator and/or Joint Conference Committee Medical Staff (Executive Committee) Committee Committee Committee Committee Committee on on on on on Tissue Joint Conference Cancer Credentials Medical Records Re Ri Cancer Cancer Education Cancer Clinical Clinical Program Activities Records Cancer Cancer Consultation Consultation (or) and Treatment Service Service Consultati Staff nc;llary Personnel From various specialty groups can. Minimum 1. Secretary 4corned with management of cancer 1. Executive Officer (Dirodar) 2. Medical Social Worker 2.Surgeon 3. Nurse 3. Pathologist 4. Radiologist 5i lnternbt Appendix V Cos t Considerations in the Operation of a Hospital Cancer Registry The major cost in the routine operation of a hospital cancer registry is for secretarial help. Ydeally, a cancer registry secretary should have training and knowledge comparable to that of a medical secretary, with some aptitude for elementary statistical tabulations. In general, the pay scale for a medically knowledgeable registry secretary is comparable to that of a medical secretary. The higher salary for such a person, as compared to that of a general secretary, will be more than compensated for by lower turnover, greater efficiency, and accuracy in abstracting medical records. However, in the absence of such a person, an alert general secretary, with no more than a high school education can operate a registry of moderate size successfully, provided she receives adequate training, and close supervision and guidance by the medical consultant. Respondents to a survey of registries approved by the American College of Surgeons report that 64 percent of the registry personnel had either A high school or secretarial school education and that such personnel could be trained to perform adequately within twelve weeks.7 The personnel time required to operate a hospital cancer registry is not only dependent upon the knowledge and ability of the secretary, but also the following variables- (1) the size of the hospital and the volume of the cancer load; (2) the amount of detail to be recorded on the abstract form; (3) the availability of medical charts (inpatient and outpatient);, (4) the completeness and legibility.,of the medical charts; (5) the availability of information from the pathology laboratory Ana the department of radiology; and (6) the response from physicians and others for follow-up information. Two of every three registries reported that their secretaries worked 40 or fewer hours per week.7 It is estimated that a secretary will require a ma ximum of one hour to completely register a new cancer patient, and an average of one-half hour per case, to obtain and record follow-up information. This is predicated on the assumption that variables three to six mentioned above are favorable. For example, if the secretary must hunt for missing medical charts, or for information missing from the charts, or if she has to send several letters to physicians and follow-up contacts in order to obtain adequate follow-up information, more time will have to be allowed. Estimates of the optimum and maximum number of actual work hours of secretarial help required per week to carry out the routine work of maintaining a hospital cancer registry for 15 years, per 100 new 43 Appendix V (Continued) cancer patients annually, are presented Estimates of the Optimum and Maximum in the table on the right. These Number of Hours Per Week of Secretarial Halt) Required for the Routine Operation estimates include the preparation of of'a H6spital Cancer Registry-for 15 Yeait' Per 100 New Cancer routine manual tabulations of the Patients Annually' registry data:, at least annually. How- Optimum Maximum ever, these estimates do not include Ist Year 1.9 1.9 time to answer numerous inquiries and 2nd Year 2.5 3.0 requests for information, or to code 3rd Year 2.9 3.9 data for computer proce .ssing. Addi- 4th Year 3.3 4.7 5th Year 3.7 5.4 tional provision must also be made for 6th Year 4.0 6.1 work-breaks, sickness, and vacation 7th Year 4.3 6.6 8th Year 4.6 7.2 time. 9th Year 4.8 7.7 1 Oth Year 5.0 8.1 To this should be added the cost 1 Ith Year 5.3 8.6 12th Year 5.4 8.9 of the time of physicians and statis- 13th Year 5.6 9.3 tidal personnel to supervise the 14th Year 5.8 9.6 registry and prepare analytical and 15th Year 5.9 9.9 special study reports to evaluate NOTE: A h ute its personnel rec ng either col- um of Its bed the management.of dancer in the has- 4.8 hospital the fundamental reason dur- for the registry. These additional ini and 9.3 fifth costs will vary according to the ye; 'Based on survival @erience, reported in the interests and needs of the me ica California Tumor Registry Monograph "Can- cer Registration and Survival in California." staff. 2There is a great deal of variation in the rela- ttonship between the number of beds andne.-W The overhead costs of a registry cancer patients. in California, the ca8elbed ratio ranged from 0.73 in county hospitals to must also be considered. It is 1.61 in private hospitals. These estimates are desirable that the registry be located ba8ed on a one-to-one relationship of new pa- tients annually and the average daily total in a separate room so that the secre- bed, quyacity. tary may work with a minimum of dis- traction. Personnel will require the usual office furniture and equip- ment'including letter file,cabinet(s) to house cancer registry abstract forms, small cabinet(s) for the patient name and follow-lip control files, and secretarial and clerical supplies.. The hospital must also provide basic reference books and manuals, follow-up aids, a telephone, and.postage. Finally, registry forms and form letters may be purchased commercially, or be prepared and reproduced by photo-offset for about $50 per thousand.' 44 REFERENCES 1. American College of Surgeons: Manual for Cancer Programs. Chicago, 1966. 2. American College of Surgeons: Bulletin. Vol. 54, No. 1, Chicago, January-February 1969. 3. American College of Surgeons: Bulletin. Vol. 55, No. 4, Chicago, April 1970 (for complete list of cancer programs approved by the College). 4. American Cancer Society: The Hospital Cancer Registry Definition, Purpose, Value, Operation and Cost. New York, 1966. 5. American College of Surgeons: Manual for Registries and Cancer Clinical Activities. Chicago, 1955. 6. California Tumor Registry: Cancer Registration and Survival in California, Berkeley, 1963. 7. Cook, G. B., Watson, F. R., and Thomas, E. C.: A Synergized Cancer Information System. Cancer Research Center, Association of Missouri Tumor Registry Directors, Inc., Columbia, 1969. 8. Haenszel, W. and Hon, and N.B.: Statistical Approaches to the Study of Cancer with Particular Reference to Case Registers Journal of Chronic Diseases, St. Louis, Vol. 4, No. 6, pp. 589- 599. December, 1956. 9. Haenszel W. and Lourie, Jr., W. I.: Quality Control of Data in a Large-Scale Cancer Re@ister Program. Methods of Information in Med icine, Vol. 5, No. 2, pp. 67-74, April 1966. 10. Guidelines, Regional Medical Programs. Division of Regional Medical Programs, Health Services and Mental Health Administration, U.S. Department of Health, Education and Welfare, Bethesda, Revised May 1968. 45