As of 2009, more than 12.5 million Americans were living with cancer.¹ To help address the incidence of malignant or neoplastic disease, hospital and state-based cancer registries collect and report statistical data to state and federal cancer agencies. As of 2010, there are an estimated 1,500 hospital and 51 central cancer registries in the United States.² These registries employ over 7,500 specially trained cancer registrars and other staff who collect, manage, and analyze data on persons diagnosed with cancer. The work of cancer registries is critical to informing national and state policy on cancer treatment, cancer research, cancer screening, and cancer preventive services. Despite the invaluable work of the registrars in hospital-based and central cancer registries, little is known about their workload, staffing levels, and challenges in staffing.

Cancer registrars are expected to be experts in cancer patient data management. Their primary responsibilities are to “provide timely, accurate, and complete data” on all cancer diagnoses and patients in the United States.³ In 2006, the National Cancer Registrars Association (NCRA) formed a recruitment and retention taskforce that commissioned a study of cancer registrars to seek more information about the profession, characteristics of registrars, educational pathways and certification, and registrar concerns about their work and workload. That study found that recruiting and keeping qualified cancer registry staff was a major concern of registry managers and directors.⁴ Although most cancer registrars expressed a strong commitment to their field, many revealed that they felt overwhelmed by the demands of the job, and undervalued given the amount of work they do and its importance to cancer surveillance. A key theme from the focus groups and key informant interviews was the absence of staffing standards and guidelines across registries. Participants stated that workload standards would help them advocate for adequate staffing as well as assist them with staffing plans.⁵

The purpose of the 2011 Workload and Time Management Survey of Central Cancer Registries was to describe the environment in which registrars work, current issues that may make performing central cancer registry work more challenging, and the relationship between workload and staffing within the registry. The University of California, San Francisco (UCSF) conducted the study with funding from the Centers for Disease Control and Prevention (CDC). Findings from this study provide national data so that central cancer registries may have a benchmark for comparison to their own cancer registry data to make decisions about staff size and configuration. In addition, these findings provide central cancer registry administrators with the data needed to advocate, plan, and budget for their cancer registries.

With input from the funding organizations, researchers formed a 26-member technical advisory committee (TAC) comprised of cancer registry experts with extensive experience in hospital and central cancer registries. The study team developed and submitted the survey tool and data collection instruments to the US Office of Management and Budget (OMB). The survey instrument included 39 multi-item questions covering the following subject areas:

Facility and registry characteristics

OMB approval (control no. 0920-0706) was received on December 5, 2012. An email was sent to each of the 51 CCR managers and included an invitation to the survey, a glossary of words and terms used in the survey, the Work Activities Journal, and instructions for completing the journal. The Work Activities Journal included 20 work activities identified by the researchers and the TAC as the most important components of registrar workload in central cancer registries. Activities in the Work Activities Journal were divided into 3 categories: weekly, monthly, and yearly activities. For frequently performed activities, staff were asked to record the amount of time spent performing those activities each day for 1 week. For less frequently performed activities, staff were asked to estimate the amount of time required to perform those activities on a monthly or annual basis. The manager or director then totaled the times reported for the entire staff and entered the totals for the registry as a whole in the online survey.

After multiple email and telephone reminders, 24 central cancer registries responded to the survey for a response rate of 47% (Appendix 1). One of the central cancer registries was involved in an experimental source record study at the time of this data collection. They received over 1 million source records as part of that study, so their data were excluded from all analyses of this study. Eighteen central cancer registries responded to all questions and 5 responded to some, but not, all questions. Respondents were divided into 4 nearly equivalent group sizes. Data were analyzed for 23 registries using the statistical package Stata. The number of respondents in this survey do not allow for correlation or inferential analyses.

Central cancer registries were asked a number of questions about where they are housed, what organizations they report to, and the reference year for the registry. These factors may be related to workload in that some states may require additional data reporting. Reference year is important in that older registries are likely to follow more cases than newer registries. Nearly 80% of the survey respondents reported that their central cancer registry was housed in a state health department. The remaining 20% reported their central cancer registries were located either in universities or through a consultant relationship with an entity in the state. The average age of the registries, based on reference year, was 15.2 years, with a minimum of 10 years and a maximum of 17 years. The 3 responding registries that report to the National Cancer Institute’s (NCI), Surveillance Epidemiology and End Results (SEER) program had an older mean age; 1 SEER registry reported a reference year dating back 38 years. Of the 24 respondents, 23 respondents reported to the CDC’s National Program of Cancer Registries (NPCR), 2 reported to both CDC-NPCR and NCI-SEER, and 1 reported exclusively to NCI-SEER.

This survey of central cancer registries was an important first step at the national level to describe and collect information on registry staffing and workload. These results provide some important new information that can be used by registries to compare their staffing and workload with other registries. The survey also provides a framework for assessing workload and staffing needs that could be replicated by states over time and may be useful for staff training or in the distribution of workload within a registry.

A key finding from this study is that workload standards are in place in some registries, although we do not have information on when those workload standards were established or how they compare or differ across registries. About 20% of respondents reported that they have workload standards for all positions. Central cancer registries may want to develop a resource to share best practices in workload staffing across registries.

This study also highlighted important changes in budgeted staffing, filled positions, and vacancy rates. In recent years, there has been a decrease in filled positions and higher rate of vacancies in central cancer registries. This may be due to state budget cuts as well as difficulties in recruiting that have been highlighted in other studies.^2,4 Registries reported an average of 0.6 fewer filled positions between 2008 and 2009 and an average vacancy of 1.5 FTEs in 2009. This may impact productivity and the ability to meet state and national reporting requirements. Registries may use these workload data as evidence to support the need for specific positions, such as the need for more certified tumor registrars. Using similar data collection tools, registries could study workload over time as operational changes occur, such as the introduction of additional reporting requirements, and use that data to identify changes needed in staffing.

This study highlighted that both the number of source records reviewed and the number of cases are measures that can indicate workload. The relationship between the number of source records and reportable and non-reportable cases varied greatly among registries. The number of cases ranged from about 3,400 to 160,500 with a mean of nearly 37,000 cases. The vast majority of these cases were reportable. The number of source records had even more variation, ranging from about 56,000 to over 290,000 with a mean of 72,000 source records. If a registry reviews a greater than average number of case records to form a single case, such as in the case of the registry that reviewed over 1 million source records, the need for staff would be greater than indicated by merely looking at the number of cases reported.

The survey has several limitations that were noted as well by some of the respondents. In the open-ended comments at the end of the survey, several respondents noted that the list of work activities on the data collection tool was not comprehensive enough to capture all the work that they do, or sufficiently detailed to allow them to determine which practices fell under which questions. Therefore, reports in this section may be an underestimate or a lower bound of the actual hours spent by registrars performing their work.

Another limitation with these data is that not every registry reported hours for all activities listed. In addition, the total FTEs reported may be an overestimate if senior level staff were counted in the FTE total but do not perform any or most of the activities listed in the Work Activities Journal. Central registry staff who perform non-cancer registration functions may also have been erroneously included in the FTE count. However, their hours and type of work are not captured in the activities journal. This limitation may have the impact of causing an underestimate of the workload and staffing estimates compared to caseload. Future workload studies should further delineate and track the time it takes for cancer registration and non-cancer registration activities.

This first time survey of staffing and workload in central cancer registries provides descriptive information about these registries as well as baseline information about workload and staffing. Respondents expressed a concern about adequate staffing and the need to find qualified cancer registry staff to hire. These concerns are consistent with those expressed in the 2011 hospital registry workload study and the 2006 NCRA workforce study of cancer registrars.^2,4,5 These workload findings and staffing guidelines provide an opportunity for registries to compare to their own registries and assess how their registries may differ. These findings may help central cancer registries to begin to build the evidence for the staffing needed to meet the cancer data reporting objectives and requirements.