Gastroenterology Practitioner and Trainee Numbers in Canada 2018: Annual Report From the Canadian Association of Gastroenterology

Abstract Background It is necessary for health planners, educators and physician and patient organizations to be aware of trends in gastroenterologist (GI) numbers in order to ensure that patients have timely access to care. Methods The number of GIs in practice and the number of trainees in the specialty was determined for 2018 using three national databases compared with previous years. Results In 2018, there were 787 GIs in Canada, which equated to 2.1 GIs per 100,000 population. There are marked differences between provinces with numbers ranging from 1.1 to 2.9 per 100,000. There are 53 GIs specializing in pediatric GI care. Forty-six per cent of practitioners under the age of 35 years are female. Seventy-two residents are training in adult GI and six in pediatrics. Approximately 75% of fellows in adult and pediatric GI are training on temporary visas. The number of adult GIs is decreasing despite increasing national population growth and service demand. The numbers of trainees in both adult and pediatric GI are lower than in 2010. If these trends continue, wait times for GI care, which are already poor, will likely increase further. Conclusions Continued monitoring of human resource numbers, patient access to care and validation of current data is required. The purpose of this report is to present the number of gastroenterologists (GIs), both in practice and in training, in Canada for 2018. We also wished to examine the 2018 numbers by province and gender, compare the 2018 numbers with those of previous years and describe the practice settings and organization.


Background and rationale
In order for patients to receive timely access to gastrointestinal health care, gastroenterology consultation and investigation are required. GIs also need to contribute to education, research and administration. It is important that the Canadian Association of Gastroenterology (CAG) tracks the number of GIs in training, and in practice, in order to contribute to health care planning.
Information-Scott's Medical Database (CIHI-SMDB) (2). The Canadian Post-MD Education Registry (CAPER) (3) was used to generate trainee and fellow counts. The methods that these databases use to record physician numbers have previously been reported (4) The CMA has wider inclusion criteria and narrower exclusion criteria. However, despite this, the CMA database has consistently reported lower numbers than the CIHI-SMDB database until 2018 (Table 1), when both databases recorded almost identical numbers (CIHI-SMDB 787 adult GI + 53 pediatric GI, Total = 840, CMA 845). Analysis of the numbers of physicians by province between the two databases indicates that they are highly correlated (r = 0.99). For clarity, we have chosen to use the CIHI-SMDB database in this report for physician numbers and the CMA database and workforce survey (2017) (1) for practice characteristics and gender balance.

Adult GI: National and Provincial Data
The total number of Adult GIs in 2018 was 787, a decrease of 14 from the previous year. Nationally, this gives a number of 2.1 GI per 100,000 population (Table 2). There is significant variation between provinces with the lowest number in Saskatchewan and the highest in Quebec.

Pediatric GI: National and Provincial Data
Pediatric GI numbers are very small with 1.4 physicians per million population on a national basis. There are approximately 6 million children under the age of 15 (Table 3), which gives a number of 8.8 pediatric GI per million children in this age group. There is variation between provinces; three provinces do not have any practitioners.
Practice Setting and Time in Clinical Practice (CMA 2017 Workforce Summary) (1) The main, but not necessarily only, practice settings include: academic health science centres (AHSC; 36%), private offices/ clinic (28%), community hospitals (23%), non-AHSC teaching hospitals (7%), walk-in clinics (2%) and community clinics (2%). It is likely that hospital-based practitioners are not exclusively engaged in GI practice. We do not know their percentage devoted to GI clinical practice. The majority are in solo practice (39%), while 20% are in group practice, 34% are hospital based and 5% are in interprofessional practice.

Age of the GI Physician Population
There are smaller numbers in the under 35 years of age and over 65 years of age cohorts. The largest number of GI physicians is in the 35-44 age group (Table 4). Approximately 110 are over the age of 65 and we can expect that some will retire each year. An average of 12 gave up their licenses per year between 2014 and 2017 (CMA). Others may have retired from clinical practice but kept their license or reduced workload. We do not know whether the number of trainees and fellows entering practice is appropriate for the number retiring and decreasing their workload.

Gender
Based on CMA data, females make up 31% of the GI workforce (Table 5). When the numbers of female and male GIs are analyzed by age group, the proportion of female practitioners shows a progressive increase with decreasing age, with the profession moving closer to parity in the number of females and males if this trend continues.
In adult gastroenterology practice, the majority (60%) of practitioners are male, a proportion unchanged since 2015. In contrast, numbers for pediatric GI practice are near opposite balance, where an average of 64% of practitioners are female.

Core trainees
Core trainees are residents who have completed 3 years of internal medicine and are completing core training in GI. Canadian citizens (CC) and permanent residents (PR) are most likely to add to the Canadian workforce and they are presented separately from visa trainees who are in Canada on educational visas and are likely to return to their home countries.

Adult GI
Trainee numbers are CC-PR just over three quarters of the peak in 2012. Visa trainees are close to their 9-year average ( Table 6).

Pediatric GI
Pediatric trainee numbers peaked in 2011 and are currently about 40% of that number. Visa trainee numbers are very small and no more than one or two per year (Table 7).

Fellows
Fellows are GIs who have completed core training and are doing additional training without exam certification on completion.

Adult GI
There have always been more visa fellows than CC-PR. CC-PR fellows peaked in 2013 and are currently 67% of peak (Table 8).
Fellowship positions are being taken up by non-Canadian trainees who now outnumber CC-PR by nearly 3:1.

Pediatric GI
The pattern is similar to the adult fellows with a ratio of visa to CC-PR of over 3:1. The numbers appear to be stable or increasing slightly (Table 9).

Entering Practice
In 2018, there were 72 CC-PR adult GI trainees in the system. The average over the 5 years from 2014 to 2018 was 75. Assuming that half graduate each year, since GI is a 2-year program, this gives an average of approximately 38 trainees available to enter practice either postcore training or postfellowship.

Discussion
The ultimate aims of GI human resource planning are to provide adequate access to care and to allow the specialty to contribute to education, research and administration. Physician head counts are a crude measure of clinical and other service availability. The percentage of time that physicians spend in clinical practice, as opposed to administrative duties or working part-time, clearly has an impact on the availability of clinical care. This may vary by province. We do not have data on this number. Other factors affecting the availability of clinical care include the efficiency of the system, rates of referral and disease burden of the population. The number of physicians in practice is determined by the number of trainees and fellows entering practice, the number of physicians retiring or reducing workload and the number of physicians immigrating to, or emigrating from, the country. We are able to track some, but not all, of these parameters. We do not know what percentage of time these physicians spend in the delivery of liver care or internal medicine.
Descriptions of the main human resource databases that generate gastroenterology numbers have previously been given (4). The two main databases that provide estimates of gastroenterology practitioner numbers are now almost identical in their estimates. Historically, the CMA database has reported a lower number of GIs, but it is now almost identical to CIHI. We can be reasonably confident that 787 GIs are engaged primarily in adult GI practice. There is less certainty with regard to the number engaged in pediatric GI but it is between 53 and 70. We cannot be certain whether a total of 840 represents a small increase or decrease from 2017. It depends on whether CMA or CIHI-SMDB numbers for 2017 are used. As we continue to track the databases, this may become clearer in future years. The need for validation studies of the databases is shown by our results of a review of pediatric GI, which shows poor correlation with the CIHI-SMDB database.
The number of GIs in adult practice was lower in 2018 than in 2017. Canada's population had record growth at 1.4% in 2018-2019 (8); the population is aging and service demand is growing in volume and complexity, all of which would suggest that an increase in numbers, rather than a decrease of 1.8%, is necessary. Trainee numbers are also at close to the lowest they have been since 2010. These trends need to be monitored and action taken to increase numbers if necessary.
Some of the gastroenterology practice, and certainly much of endoscopic practice, can be carried out by surgical colleagues. Therefore, as we have previously shown in the practice audit Table 6. Canadian citizens (CC) and permanent residents (PR) and visa trainees in adult GI by year (7) Year 2010  2011  2012  2013  2014  2015  2016  2017  2018   CC-PR  79  91  94  80  76  78  79  71  72  Visa  15  11  10  9  7  8  13  11  9   Table 7. Canadian citizens (CC) and permanent residents (PR) and visa trainees in pediatric GI by year (7) Year 2010  2011  2012  2013  2014  2015  2016  2017  2018   CC-PR  13  16  7  8  9  14  11  5  6  Visa  1  0  0  2  2  1  1  1  2   Table 9. Canadian citizens (CC) and permanent residents (PR) and visa fellows in pediatric GI by year (7) Year  of access to care, it is not a matter of a "correct" or "incorrect" number of GIs but whether patients have appropriate access to timely care (9). No studies have been done in recent years.
To get a better idea of access trends, there is a need to repeat the practice audits of access to care, or to establish monitoring practices, throughout the country. There is considerable variation between provinces with regard to the number of GIs per 100,000 population. These data are difficult to interpret. Each province regulates its own health care system. The only valid measure of whether the provincial variation in physician numbers is significant would be measures of wait times and access to specialty consultation and investigation.
The number of pediatric GIs is low, and several provinces do not have any access to the specialty. The impact of this on the care of children with gastrointestinal disease is unknown. It is likely that, in the provinces without pediatric GIs, some of that work is performed by physicians who are trained in adult gastroenterology. Studies assessing the impact of low pediatric GI numbers on patient care need to be conducted. The lack of agreement between the CIHI-SMDB numbers and our review is a concern and highlights the need to validate the databases by direct measurement.
In the younger GI age groups, the numbers of males and females are nearly equal. The changing gender balance of the specialty and trends in paternity leave may, or may not, bring about differences in workload, leave or practice settings. We need to conduct age-matched studies of GIs in order to understand and predict any new possible trends.
An appropriate number of trainees could be defined as the number needed to enter practice in order to replace practitioners who are transitioning to part-time clinical practice or retiring to ensure continuing clinical, administrative and academic practice. We cannot say with certainty whether or not the current number of trainees is adequate. However, we do know that over 100 members of the GI workforce are over the age of 65, an age when retirement or a reduction in practice is likely and we do know that an average of 12 retire their licence each year up to 2017.
CC and PR are accessing fellowships in decreasing numbers, but the number of fellows on an educational visa is increasing. Fellows with educational visas are likely to return to their home countries and their expertise will not be available to the Canadian public, nor will they be available to fill replacement positions as they become available. We do not know what the implications of this are for GI care without further detail on patient need for the care provided by fellow graduates and the positions they might replace.
In February 2020, the Royal College of Physicians and Surgeons launched its Medical Workforce Knowledge base (10). This is an interactive, well-designed site that promises to be invaluable GI resource for the study of human resource numbers. The first set of data cover the years up to 2017. Both adult and pediatric data are included. Data is included for residency quota, new trainees, new certificants and licensed physicians, and the site allows comparison with other specialties. The data sources are CARMS, CAPER, and CIHI. Our data is broadly similar to that of the College and we will work with the College so that agreement can be reached on numbers and trends.
This data for 2018 adds to our understanding of trends in gastroenterology human resources. There are significant gaps in our knowledge of GI human resources. We need much more information on the implications of the major trends observed in 2018-a decrease in the number of practitioners, low trainee numbers compared with historical data, increasing female representation and a decreasing number of Canadians in fellowship positions. Continued monitoring of trends is required.

Funding
This study was funded by Canadian Association of Gastroenterology (CAG). The authors declare no conflict of interest.