In the food supply veterinary medicine (FSVM) area, there has been growing concern and some controversy over the adequacy of the future supply of veterinarians and the extent that demand for their services will decline or grow. Many conclude that this area, and veterinary medicine in general, is undergoing rapid change and is at a crossroads.1 Understanding the key changes and trends driving the future demand for services, the supply of labor into FSVM, and whether veterinary medicine is headed for growing or declining demand and shortages or surpluses of food supply veterinarians is an important factual grounding that leaders in the veterinary profession need so that they will be able to plot strategic action that will improve the profession. A healthy and prosperous FSVM sector will enable food supply veterinarians to continue to serve society by protecting animal health and human and public health concerns associated with food safety, zoonotic diseases, and biosecurity.
Many agree that the veterinary profession is undergoing profound changes and must adapt to new demands.2–6 Consolidation in the food system and the increasing size of farming or ranching operations have dramatically changed the skill sets required of veterinarians as they increasingly focus on herd health systems rather than interventions on individual animals.7 There are many additional changes, including globalization of the food supply system, changing societal trends (including urbanization and the changing gender composition of the veterinary workforce8), and emerging needs that require the unique abilities of food supply veterinarians. There is increasing demand for countering emerging disease threats that affect health of humans and domestic animals, as well as related concerns about biodiversity and the impact of highly concentrated food animal operations on the environment. Since the events of September 11, the veterinary profession has become more cognizant of the vulnerabilities of agriculture to bioterrorism and the expanding role of veterinarians in protecting the public from such threats.9,10 With these many changes, there are new roles for veterinarians in the public health arena and expanding demands for the ability to conduct research and provide other skills.11–13 These changes alter the available labor supply and the demand for services, and they also have the potential for altering the configuration of forces that will shape the FSVM labor supply and demand in the future.
A number of articles2,5–7,14–16 have raised concerns about the growing shortages of food supply veterinarians and the increasing demand for their services. These articles provide compelling arguments, but the anecdotal evidence contradicts the last major scientific look at the supply and demand for the veterinary profession.17
The KPMG LLP study17 (which has been referred to as the Megastudy) projected the future demand for all veterinarians from 1997 to 2015. That study revealed strong increasing demand for companion animal veterinarians. However, the future for food supply veterinarians forecasted by the KPMG LLP study was quite dismal. It was concluded that during the 18-year forecast period (1997 to 2015), there would be no growth in demand for veterinarians in the academic and government areas and a decrease in demand (−1.7%) for veterinarians in large animal practice.
Economic models were used in the KPMG LLP study to make projections concerning the gap between expected demand and supply of veterinarians. Results for the model predicted a large surplus of veterinarians in large animal practice. It was mentioned that the supply patterns assumed in the model will change when future veterinary students migrate away from areas of large animal practice toward areas in which there are more opportunities, such as companion animal practice. Authors of the KPMG LLP study argued that the pattern of supply (attraction of new graduates to areas of lower demand) is likely to change and anticipated that the predicted surplus of 1,080 veterinarians in large animal practice would not likely be reached.
Reactions to the demand and surplus projections of the KPMG LLP study have varied. Some observers wondered whether food animal veterinarians are becoming an endangered species, and concerns have been expressed that there will be dangerous shortages of food supply veterinarians.14,15 There are also concerns about the shortages of veterinarians at federal agencies, which is made even more problematic because of expected retirements in the near term.7 Changes in the public sector are expected to increase the need for food supply veterinarians with training in population health, food safety, food animal production, zoonotic diseases, biosecurity, and research methods. One author15 suggested that the shortage is becoming so severe that rural, mixed–food animal practitioners are in danger of becoming extinct. That author projected a growing demand in private practice areas as well as strong increases in the need for food supply veterinarians in government roles. Another author2 voiced similar concerns and identified several FSVM areas as having tremendous potential for rapid growth.
Results of the KPMG LLP study are also in direct contrast to global concerns about increasing demand for food supply veterinarians and shortages in FSVM. Two national studies18–20 outside North America were prompted by uncertainty and concern over the increasing demand and inadequacy of the future supply of FVSM veterinarians. In a United Kingdom study18,19 published in 2003, investigators concluded that there were fears about an insufficient number of large animal veterinarians available to fulfill needs. They also concluded that changes in food safety rules and animal health and welfare surveillance requirements will increase the demand for veterinary services.
An Australian study20 on rural veterinarian services was also reported in 2003 and provided similar conclusions. The authors of the Australian study concluded that the increasingly stringent requirements of world trade will necessitate enhanced veterinarian service to provide more surveillance (and the need to establish an Australian Veterinary Reserve Corps to address this concern). It was also concluded that rural veterinarians must contend with several challenges that create local shortages and could lead to a chronic shortage of production animal veterinarians. Finally, it was recommended that all parties involved in veterinary services (governments, producers, veterinary practitioners, and veterinary schools) must change to address these problems.
Questions arise as to which set of data should be believed. The KPMG LLP study used an econometric model that projected a potential surplus of food supply veterinarians and flat or decreasing future demand, whereas more-recent anecdotal observations about growing demand and dangerous shortages and conclusions reached in the studies conducted in the United Kingdom and Australia indicate a growing demand and increased shortages of veterinarians. Which should be used as a guide for strategic planning and action? The obvious answer is that more research on the US and Canadian situations is needed.
Analysis of the preceding information led to the conclusion that FSVM is undergoing change, and there is contradictory evidence about the direction of future demand and whether the veterinary profession will have labor shortages or surpluses during the next few years. This uncertainty provided the basis for the following central questions that guided the research reported here:
•What are the issues and forces that will drive future demand for FSVM services during the next several years?
•Will the demand for FSVM services grow or decrease in the future?
•What factors will drive the future supply of veterinarians to FSVM?
•Will there be a shortage or surplus of veterinarians in FSVM during the next several years?
•What are the actions that leaders in the veterinary profession can implement to improve the future of FSVM and improve its capacity to continue to serve society?
Predicting and understanding the future of a complex sector of a profession (eg, FSVM) are not easy tasks, and the changing larger context of this sector of the veterinary profession makes simple linear extrapolation of historical trends with econometric models, which were used in the KPMG LLP study, more problematic. Forecasting techniques based on historical trends, such as econometric models, are informative when there are clear predictors that can be accurately measured and have a continuing stable relationship to the criterion variable. In a changing sector of the veterinary profession, such as FSVM, econometric models become suspect as new causes emerge and the historical relationships between predictors and a criterion change.
A forecasting method based on expert judgments, such as the Delphi forecasting method, is an alternative to forecasting methods based on historical trends and is more appropriate for these changing situations. In contrast to the KPMG LLP study in which a single mathematical function of economic factors was used to predict the future in all sectors of veterinary medicine, the Delphi method can draw on insights from experts in several sectors and provide unique forecasts for the diverse areas within the FSVM sector.
The Delphi method21,22 works well when combined with a strategic planning process. It is useful in identifying trends that can be changed through strategic initiatives as well as changes attributable to larger societal and economic factors that are more fixed. A tenet of the Delphi method is that when the mix of unfolding forces that will create the future is understood, leaders can act and change those driving forces before the future arrives. It is designed to identify leverage points that are important in developing and implementing effective strategic initiatives.
Materials and Methods
Delphi method—In the study reported here, the Delphi forecasting process was used to evaluate 13 areas of the FSVM sector. These included 6 areas of private practice (dairy, swine, poultry, beef, small ruminants, and mixed–food animal), 5 areas of government service (state and provincial government service, 3 areas of US federal government service [public health, animal health, and food safety and security], and Canadian federal government service), the area of food supply veterinarians serving in industrial roles at pharmaceutical companies, and the area of academic roles at university settings. It was believed that these 13 areas capture the major segments of the FSVM sector.
Experts for the Delphi panel associated with each of these areas were identified through a nomination and snowball sampling process. Although the nomination process varied slightly among the 13 panels, it generally followed the pattern whereby known leaders in each area (such as current or past leaders of a related professional veterinary medical association) were contacted, and they recommended veterinarians who had a good understanding of the area and the ability to provide future demand and shortage (or surplus) estimates as well as evaluate factors that influence the future supply and demand for veterinary services. For government panels, those in higher-level veterinary-related positions were also contacted for recommendations. Finally, representatives to the Food Supply Veterinary Medicine Coalition also suggested names of potential panel experts. In addition, authors who published in each area were contacted.
Individuals who received multiple nominations were contacted and asked to participate. Typically, these individuals were contacted via an explanatory e-mail and a follow-up telephone call. The 3-survey process was explained, and potential expert panelists were asked to confirm their ability to provide the demand and shortage (or surplus) estimates and evaluate factors that influence supply and demand. They were also asked to suggest additional experts for panels, including experts who may have differing viewpoints. Once a panel was almost completely filled, members were provided a listing of that panel and asked 1 last time to suggest additional names.
The Delphi method gathers expert opinion and provides a structured feedback process during multiple surveys. In the study reported here, 3 rounds of surveys were used. After the initial data-gathering round, panel experts had an opportunity to consider the views of other experts during 2 successive rounds of surveys. Internet-based surveys were used, which enabled experts to provide their views and learn the collective views of other experts. It set up a learning process whereby experts had an opportunity to reconsider their judgments in the face of conflicting viewpoints. Final judgments were made anonymously without dysfunctional social dynamics that can plague interacting groups.
The feedback process used in the second and third surveys included a summary report that indicated trends and identified data patterns that explained some differing views seen in each panel's survey replies (eg, views of those predicting more-extreme versus less-extreme patterns of demand or shortages). Also, mean values and interquartile ranges (25th to 75th percentiles) for each item were incorporated into the second and third surveys so that data patterns from the preceding survey could be easily reviewed while experts made new ratings and forecasts. Accuracy improves over the multiple-round decision process, and those who drop out tend to be less-expert panelists.21
Each of the 3 surveys had a similar general design and required experts to rate the influence of factors on future demand in their area. These items were gathered from the literature on FSVM and helped panelists consider and think about issues driving demand before they made their demand estimates. Panelists predicted the expected changes in demand in their area for each of 5 time periods between the fall of 2004 and fall of 2016. Experts estimated the percentage change in demand from the start to the end of each time period. The near-term time periods were shorter, and time periods farther in the future were longer. Panelists rated the influence of factors gathered from the literature and discussions with leaders in FSVM regarding the expected supply of veterinarians entering FSVM careers in their area. Expected future shortages or surpluses of veterinarians were predicted for the 5 time periods between the fall of 2004 and fall of 2016 in each area. Experts estimated the percentage by which the available supply of veterinarians would differ from the expected demand for veterinarians during each of the 5 time periods.
Each of the 13 panels was independent, and panel members were repeatedly reminded to consider only their particular FSVM area. It required a minimum of 4 months to complete the 3-survey process.
Supply and demand influences—A review of the literature on FSVM was used to identify factors believed to influence supply and demand in FSVM. Open-ended replies in the first survey were used to develop additional items for each panel in the second survey, and items in the first survey with less agreement in the expert ratings were repeated in the latter surveys. Because each item was not included in all subsequent surveys for each panel, comparisons of results among the 13 panels would have resulted in contrasts between data from the first survey for some panels with data for the second or third surveys for other panels. Given that the results reported here focused on data patterns for all 13 panels, it was determined that the factors influencing supply and demand in the first survey provided the most comparable data among all 13 panels. Information that summarizes common and panel area–unique items on supply and demand from the second and third surveys is available in the fulllength report23 of the research project.
Factors were rated on a 7-point scale to indicate the perceived influence of each factor on the demand or future shortage for FSVM services in the panel's FSVM area. The 7point scale was as follows: 1, strong decrease; 2, decrease; 3, slight decrease; 4, no influence; 5, slight increase; 6, increase; and 7, strong increase. A 1-way ANOVA was used to detect significant differences among mean values for the various panels. Data for most factors approximated a normal distribution and did not have large skewness or kurtosis. Although a 1-way ANOVA is often considered to be a robust technique that is not overly influenced by violations of the normal curve distribution assumptions,24,25 data for these factors were also tested by use of the Kruskal-Wallis nonparametric test as a precaution. Results of this nonparametric test matched those of the 1-way ANOVA. Values of P < 0.05 were considered significant. The Duncan multiple comparisons post hoc test was used to determine homogeneous subgroups (P < 0.05).
Future demand—Panel experts were asked to forecast expected changes in the future demand for FSVM services for each of 5 time periods between the fall of 2004 and fall of 2016 in each survey. The 5 estimates of expected change in demand from the third survey were summed to form the cumulative demand change scale. These estimates were highly correlated and provided a reliable scale with a Cronbach a reliability coefficient of 0.81. The Cronbach a value is the standard measure of internal consistency for multiple-item scales, and values > 0.60 are considered to be acceptable.26 Estimates that were extreme outliers were removed (0 to 2 outliers/panel). This is a standard procedure in analysis of data generated by Delphi panels.21
A 1-way ANOVA was used to detect significant differences among mean values for the various panels. Results for the ANOVA were confirmed by use of the Kruskal-Wallis nonparametric test. Values of P < 0.05 were considered significant. The Duncan multiple comparisons post hoc test was used to determine homogeneous subgroups (P < 0.05).
Future shortages—Panel experts were asked to forecast the mean match (or gap) between expected demand and available supply of FSVM professionals for each of 5 time periods between the fall of 2004 and fall of 2016. These estimates generally resulted in negative numbers, which indicated shortages, and the mean of these forecasts from the third survey comprised the mean shortage scale. This scale had a Cronbach a reliability coefficient of 0.77. Estimates that were extreme outliers were removed (0 to 2 outliers/panel).
A 1-way ANOVA was used to detect significant differences among mean values for the various panels. Results for the ANOVA were confirmed by use of the Kruskal-Wallis nonparametric test. Values of P < 0.05 were considered significant. The Duncan multiple comparisons post hoc test was used to determine homogeneous subgroups (P < 0.05).
Solutions for shortages—Tactics for eliminating labor shortages were developed and included in the third survey. Panel experts rated the effectiveness of strategies for eliminating a shortage of veterinarians in their particular area of FSVM.
Solutions were rated on a 7-point scale to indicate the perceived effectiveness of each solution to eliminate shortages of veterinarians in the panel's FSVM area. The 7-point scale was as follows: 1, not at all effective; 3, slightly effective; 5, effective; and 7, highly effective. A 1-way ANOVA was used to detect significant differences among mean values for the various panels. Results for the ANOVA were confirmed by use of the Kruskal-Wallis nonparametric test. Values of P < 0.05 were considered significant.
Results
Expert panels—Overall, 303 experts originally agreed to serve on the 13 panels. Of these, 282 (93.1%) completed the first survey and 237 (78.2%) completed all 3 surveys. Eight of the panels had a survival rate (ie, panelists who completed all 3 surveys) of ≥ 80%. The dairy panel had the highest survival rate (22/24 [91.7%]), and the state or provincial government panel, which by far had the most original members, had the lowest survival rate (39/63 [61.9%]).
Factors influencing demand—Ratings of the 13 panels on factors influencing future demand were summarized (Table 1). Values differed significantly (P < 0.001) among panel means, as determined by use of the 1-way ANOVA. Results for the Kruskal-Wallis test were the same as those for the 1-way ANOVA, except 1 factor (slow adoption of new technologies by veterinarians) was significant (P = 0.046) for the Kruskal-Wallis test but not significant for the 1-way ANOVA.
Mean scores* for factors influencing the future demand for veterinarians in FSVM, as determined by panels of experts in 13 areas.
The factor with the highest rating for influence to increase demand was public concerns over food safety (mean, 5.96). There were significant differences among panel means. The mean value for the small ruminants panel (6.81) was significantly higher than the mean value for any of the other 12 panels, which ranged from 5.65 for the mixed–food animal panel to 6.08 for the state or provincial government panel.
Analysis of the data revealed a number of patterns. Five of the 7 highest-rated demand factors were logically related to general public concerns and would be construed as having a strong influence on increasing demand. Concerns about food safety, zoonotic disease, bioterrorism, animal welfare, and animal health illustrate general public concerns. There was a general pattern (with a few exceptions) for panels in private practice areas (except small ruminants) and the industrial area to have lower mean values and for panels in the academic and various government areas to have higher mean values. Higher mean values indicate a stronger influence to increase demand.
Other factors that influenced the increase in demand were logically related to drivers of client demand. These included the need to track animals and certification or verification of standards. An additional factor (access to global markets) that influenced the increase in demand logically focused on the theme of expanding market opportunities. There was a pattern for federal government panels to foresee a stronger influence to increase demand for their services from this factor, compared with the influence on services for the private practice areas (except for small ruminant practice).
Two logical groupings of factors were associated with decreases in demand (Table 1). Government budgetary constraint is an underlying theme for the curtailment of government support and budgetary constraints items. The influence of these factors to decrease demand was rated similarly among all panels.
Although government budgetary constraint is not easily changed, the second grouping of factors that influenced a decrease in demand was related to veterinary skills, which can be changed with education. The theme of skill limitations constraint is evident in the factors of practice management and business skills and slow adoption of new technologies. Although there were not significant differences among panel means for the adoption of new technologies, there were significant differences among panels for practice management and business skills. There was not a demand constraint for those in academia (mean, 4.29), industrial (mean, 4.42), or federal animal health roles (mean, 4.12), but all of the private practice areas had mean values < 4.0, which indicated a decrease in demand.
Future demand—The final forecast of future demand for each panel represented the results of the third survey (Table 2). Mean values differed significantly (P < 0.001) among panels, as determined by use of the 1-way ANOVA. Although the distribution for the data varied only slightly from normality, the Kruskal-Wallis test was used as a precaution, and that nonparametric test also detected a significant (P < 0.001) difference among panel means.
Cumulative future demand for veterinarians in FSVM for the 12-year period from the fall of 2004 to fall of 2016, as determined by panels of experts in 13 areas.
| Panel area | Mean ± SD (%) | Median (%) | 25th to 75th percentiles |
|---|---|---|---|
| Poultry (n = 19) | +4.11 ± 4.27a | +4.0 | +2.0 to +8.0 |
| Small ruminants (13) | +7.54 ± 5.04a,b | +7.0 | +4.0 to +11.0 |
| Beef (20) | +7.70 ± 12.00a,b | +9.5 | −2.5 to +14.3 |
| Dairy (21) | +8.29 ± 13.84a,b | +14.0 | −5.0 to +17.5 |
| Swine (22) | +9.96 ± 12.91a,b | +14.0 | +4.3 to +16.5 |
| Mixed food animal (20) | +10.70 ± 17.25a,b,c | +14.5 | −4.0 to +23.0 |
| Academia (19) | +12.58 ± 10.45a,b,c | +13.0 | +9.0 to +18.0 |
| Industrial (12) | +12.83 ± 8.53a,b,c | +11.5 | +6.3 to +20.3 |
| Federal–Canada (9) | +15.44 ± 6.02b,c | +16.0 | +10.5 to +20.0 |
| Federal–animal health (14) | +16.29 ± 8.58b,c | +15.0 | +9.5 to +21.8 |
| Federal–public health (12) | +16.75 ± 14.49b,c | +16.0 | +10.3 to +20.5 |
| Federal–food safety and security (13) | +17.46 ± 12.76b,c | +22.0 | +10.0 to +26.0 |
| State or provincial government (35) | +20.80 ± 15.58c | +19.0 | +13.0 to +27.0 |
| All panels combined (229) | +12.46 ± 13.05 | +13.0 | +6.0 to +20.5 |
Results reported represent cumulative values for 5 time periods between the fall of 2004 and fall of 2016. Mean values differ significantly (P < 0.001) among panels, as determined by use of 1-way ANOVA. The Kruskal-Wallis test was used as a precaution, and also indicated a significant (P < 0.001) difference among panel means.
Values with different superscript letters differ significantly (P < 0.05; Duncan multiple comparisons post hoc test).
The cumulative demand was the sum of demand increases (or decreases) for each of 5 periods from the fall of 2004 to fall of 2016. The forecast for the poultry area had the lowest mean, with the poultry panel projecting a demand increase of only +4.11% during the 12-year period (equivalent to an average annual growth of +0.34%). The 6 private practice areas had the lowest demand increases, with the mixed–food animal area having the highest increase (+10.70% during the 12-year forecast period) among those panels. Values for the academia and industrial panels were slightly higher, whereas mean values for the 5 government panels were the highest. It should be mentioned that these forecasts were conservative in nature. Panel members were explicitly instructed to assume that emerging trends and patterns will continue and that there would be no catastrophic events that could dramatically influence demand.
Some panels reached relatively higher amounts of agreement for demand forecasts. In Delphi panels, the initial estimates often have a broad range that is narrowed in successive forecasts. Relative consensus is evident by smaller SD values and narrower interquartile ranges. Panels with the highest consensus were those focused on the poultry, small ruminants, and federal US or Canadian government areas, whereas other panels had considerable disagreement among experts that did not diminish during the 3 rounds of the Delphi process.
Factors influencing the labor supply—The 13 panels rated factors that influenced the labor supply (Table 3). Lower scores indicated that the factor had a stronger influence on decreasing the labor supply. The factor with the strongest effect on decreasing the supply was less emphasis on food animal practice in veterinary schools (overall mean, 2.34). For this factor, there were significant (P < 0.001) differences among the means for the 13 panels. All panels in the private practice areas had the lowest mean values, which ranged from 1.56 for the small ruminants panel to 2.35 for the poultry panel. The means for the industrial (3.42) and federal–public health (3.25) panels were significantly higher, compared with the means for the 6 private practice panels.
Mean scores* for factors influencing the future supply of veterinarians in FSVM, as determined by panels of experts in 13 areas.
Although factors that influenced the labor supply covered diverse areas, 2 sets of factors were logically related to common themes. The first set comprised 3 factors with a strong influence that were related to negative student experiences in veterinary school. Those factors (less emphasis on food animal practice, little exposure to FSVM career options, and lack of externships related to food animal practice) collectively suggested that veterinary students were being discouraged from pursuing an FSVM career track. Although the first of those factors (less emphasis on food animal practice) is most problematic for the private practice areas and less of a constraint for the nonprivate practice career areas, lack of externships seemed to be a strong supply constraint for all areas, except for the industrial area.
The second set comprised 3 factors logically related to rural economic and cultural constraints that decrease the flow of veterinarians into FSVM careers. The 3 factors related to this theme were lack of spousal career options, limited lifestyle and career opportunities, and lack of cultural and recreational opportunities. Analysis of results for the panels indicated that private practice areas (excluding poultry) and the federal–food safety and security area tended to have stronger supply constraints for these 3 factors. Although constraints associated with rural socioeconomic factors will not be easily changed and will require people to work or manage around such inflexible constraints, the negative messages and experiences that veterinary students perceive are more changeable. Other negative influences on supply can be changed with strategic initiatives. These include poor income opportunities and perceived lack of demand (which are perception-based barriers that can be lessened with factual information) and poor role models (which can be lessened with initiatives to provide better mentoring and career support for veterinary students and new veterinarians who are entering into FSVM careers).
Shortages in FSVM—In making forecasts regarding future shortages of food supply veterinarians, panel experts were required to consider both emerging demand and supply forces. The final forecast of future shortages or surpluses for each panel were calculated as the mean for each of the 5 time periods from the fall of 2004 to fall of 2016 (Table 4). Negative numbers indicated the percentage of positions that will be in excess of the veterinarians available to fill those positions (ie, the shortage). Combined results for all panels revealed an estimated mean shortage of −4.61% (median, −4.0%; 25th to 75th percentiles, −1.7% to −6.3%).
Forecast of the future shortage of veterinarians in FSVM for the 12-year period from the fall of 2004 to fall of 2016, as determined by panels of experts in 13 areas.
| Panel area | Mean ± SD (%) | Median (%) | 25th to 75th percentiles |
|---|---|---|---|
| Poultry (n = 19) | −0.06 ± 1.03a | 0.0 | 0.0 to −0.8 |
| Small ruminants (13) | −2.23 ± 1.40a,b | −2.2 | −1.0 to −3.6 |
| Industrial (12) | −3.30 ± 3.78b,c | −2.8 | −0.8 to −5.1 |
| Dairy (21) | −3.80 ± 3.24b,c,d | −3.5 | −1.6 to −5.4 |
| Swine (21) | −4.42 ± 2.74b,c,d | −4.0 | −2.7 to −6.3 |
| State or provincial government (36) | −4.85 ± 4.79b,c,d | −4.3 | −1.8 to −6.6 |
| Federal–public health (12) | −5.24 ± 3.78b,c,d | −4.9 | −3.0 to −8.6 |
| Beef (20) | −5.40 ± 5.46b,c,d | −4.6 | −2.2 to −6.5 |
| Federal–Canada (10) | −5.45 ± 3.64b,c,d | −4.6 | −3.3 to −7.1 |
| Academia (19) | −5.46 ± 4.00b,c,d | −5.0 | −3.0 to −7.3 |
| Federal–food safety and security (14) | −6.57 ± 6.01b,c,d | −5.3 | −3.8 to −8.6 |
| Mixed food animal (20) | −6.60 ± 5.00c,d | −5.8 | −2.9 to −9.8 |
| Federal–animal health (14) | −6.86 ± 5.16d | −5.6 | −2.5 to −9.9 |
| All panels combined (231) | −4.61 ± 4.43 | −4.0 | −1.7 to −6.3 |
Results reported represent mean values for 5 time periods between the fall of 2004 and fall of 2016. Negative numbers indicate the percentage of positions that will be in excess of the veterinarians available to fill those positions (ie, a shortage). Mean values differ significantly (P < 0.001) among panels, as determined by use of the 1-way ANOVA. Distribution for the data varied from normality in that there was positive kurtosis (leptokurtic). The Kruskal-Wallis test was used as a precaution, and also indicated a significant (P < 0.001) difference among panel means.
See Table 2 for remainder of key.
Significant (P < 0.001) differences in mean values existed among panels, as determined by use of the 1way ANOVA. Distribution for the data varied from normality in that it had positive kurtosis (leptokurtic). The Kruskal-Wallis test was used as a precaution, and that nonparametric test also revealed a significant (P < 0.001) difference among panel means. The poultry panel predicted the lowest mean shortage (−0.06%), which indicated that experts in the poultry area foresee a close match between supply and demand during the 12-year forecast period. The highest forecasted shortages were in the federal–animal health career area (mean, −6.86%), which was significantly higher than the mean values for 3 panels with the lowest estimates (poultry, small ruminants, and industrial). Similar to the demand estimates, these estimates for shortages of veterinarians in FSVM were conservative because experts were instructed to assume that supply and demand trends would not change and that there would be no catastrophic events that could influence their forecasts.
Solutions to shortages—Tactics for lessening or decreasing shortages were identified and listed in the third survey, and each panel rated the effectiveness of each tactic for eliminating shortages in their area. Panel experts were instructed to use a high standard when evaluating these tactics. Tactics were rated by use of a 7point scale, with a value of 7.0 indicating that the tactic would be highly effective in eliminating shortages. The solutions to shortages were ranked in order of rated effectiveness (Table 5). Ratings for several solutions differed significantly among panels as determined by use of the 1-way ANOVA. Data for most factors approximated a normal distribution and did not have large kurtosis or skewness. Results for the nonparametric Kruskal-Wallis test were the same as those for the 1-way ANOVA, except for 2 factors (marketing campaigns to increase awareness and increased focus on food supply coverage were significant for the 1-way ANOVA but not significant for the Kruskal-Wallis test).
Mean scores* for solutions to the future shortage of veterinarians in FSVM, as determined by panels of experts in 13 areas.
For a number of solutions, there was a general pattern that the 6 private practice panels had lower ratings than for the other panels. The beef and swine panels typically had the highest mean values, and the small ruminants panel frequently had the lowest effectiveness rating for the private practice panels. For example, on the highest-rated tactic (student debt repayment and scholarship programs), the mean values for private practice panels ranged from 4.38 for poultry to 4.95 for swine. Mean values for the academia and US federal government panels (which ranged from 5.63 for academia to 6.07 for federal–animal health) were higher than the means for private practice panels of poultry, small ruminants, mixed food animal, and beef.
Several solutions, including highly rated tactics, have direct implications for colleges or schools of veterinary medicine. These include more involvement of food supply practitioners in training of veterinary students, appointment of more food animal faculty, expanded postgraduate fellowships, paid externship requirements, expanded Centers of Excellence, and increased food supply coverage early in the curriculum. These represent strategies for changing the educational experience and better preparing veterinary students for FSVM careers. Other tactics considered effective were oriented toward supporting careers in FSVM. These included mentoring initiatives and providing expanded job placement services.
Open-ended responses on factors that influence the labor supply revealed that panels often suggested changes in current practices used to select incoming veterinary students. Three tactics were related to doing a better job of attracting students who were more likely to enter FSVM careers (marketing campaigns, focused recruitment of high school and college students, and reserving class slots for qualified FSVM-oriented students; Table 5).
Disagreements among experts—One unexpected finding was the sometimes-high amount of disagreement in demand and shortage forecasts within several panels. Through the process of analyzing complete data for each panel, it became apparent that although there were a number of factors on which panel members disagreed, their views on the influence of consolidation and larger-sized operations systematically explained the reason that 1 panelist had a pessimistic view and another panelist had an optimistic view of future demand. For factors that influenced the future demand for food supply veterinarians, movement toward larger-sized operations had an overall mean of 4.0, which was the point of no influence (Table 1). More than half of the panel members rated this factor between 4 and 7, which indicated that they believed this factor would have no influence or lead to an increase in demand, whereas the other panel members scored it between 1 and 3 because they believed it would lead to a decrease in demand. Consolidation and its influence have been dominant issues and key factors affecting change in the FSVM sector of the veterinary profession.
To better understand how this issue influenced demand forecasts of panel experts, 2 groups were created. Results for panel experts who rated the influence of consolidation (ie, movement toward larger-sized operations) between 5 and 7 (slight increase to a strong increase on demand) were contrasted with results for panel experts who rated this factor between 1 and 3 (slight decrease to a strong decrease on demand); analyses were conducted by use of t tests. There was a significant (P < 0.001) difference in the forecasted cumulative demand between these 2 groups. Panel experts who believed that the influence of larger operations would decrease the demand for food supply veterinarians forecasted a mean increase of +8.4%, whereas panel experts who believed that the influence of larger operations would increase the demand for food supply veterinarians projected a mean increase in demand of +15.7% for the 12-year forecast period.
Other influences on demand also helped clarify differences between panel experts who predicted higher values for demand increases and those who predicted lower values for demand increases. Another set of 2 groups was created by use of the median value for the forecasted cumulative demand (ie, panel experts with values greater than the median [high-demand group] and panel experts with values less than the median [low-demand group]). The t test analyses of all demand factors revealed several significant differences between these 2 groups of panel experts. Factors related to general public concerns (food safety, zoonotic disease, and bioterrorism) were considered to have a stronger influence on demand by panel members in the high-demand group. Similarly, expanding market opportunities (as indicated by the response to more access to global markets for food exports) was a stronger driver of the increase in demand for the high-demand group. Use of nonveteriarnians, such as veterinary technicians, had only a slightly positive influence on demand for the highdemand group, whereas it had a slightly negative influence on demand (ie, decrease in demand) for the low-demand group.
To better understand variance within panels on shortage estimates, we created 2 groups by use of the median value for the forecasted shortage (ie, panel experts with values greater than the median [highshortage group] and panel experts with values less than the median [low-shortage group]). Comparisons between the 2 groups were made by use of t tests for responses on all demand and supply factors. Several significant differences were identified.
Of the supply factors, only 2 were rated significantly differently by the high-shortage and low-shortage groups. Panel experts in the high-shortage group rated more women veterinarians entering the workforce and physical demand of large animal veterinary work as having a significantly stronger effect on decreasing the labor supply and contributing to a shortage, compared with responses for the low-shortage group. The high-shortage group also projected higher increases for cumulative demand (+17.9%), compared with the projection for the low-shortage group (+7.7%).
Several demand influences were also rated significantly differently between the groups. These differences included a stronger influence for an increase in demand as a result of the use of nonveterinarians, such as veterinary technicians; more access to global markets for food exports; and public concerns over bioterrorism for the high-shortage group. Compared with responses for the low-shortage group, the high-shortage group did not believe that client concerns about veterinary service costs will decrease future demand.
Although the results indicated several areas that can be used to partially explain the range of projections provided in some panels, the range of forecasts also indicated that panel experts recognized that the future is not fully determined but will be a function of strategic actions undertaken in the near future by those involved with the veterinary profession. In that sense, the range of forecasts also reflected the degree of optimism or pessimism about the effect of strategic actions undertaken in the near term to influence the outcome for FSVM.
Discussion
It was clear in the study reported here that there are opportunities and challenges facing FSVM among the 13 areas of the FSVM sector evaluated. On the important issue of whether the future demand for food supply veterinarians is increasing or decreasing, analysis of the data confirmed many anecdotal views and conflicted with the forecast reported in the KPMG LLP study.17 The 13-panel Delphi study reported here predicted that demand will increase in the future.
Although there are important differences in the rates of increase forecast for each area, it was clear that the net effect of various trends is that food supply veterinarians will have expanding career opportunities. The percentage increases forecast for each area were conservative in that they assumed that trends will continue and there will be no major events that will change FSVM demand. Realistically, there eventually will be an outbreak of an important disease. Such events have the strong potential to increase demand well beyond that projected by the mean and median point estimates, which would make the higher values in the 25th to 75th percentile range more probable.
A number of factors are currently influencing and will continue to influence future demand. Some demand factors are not highly changeable and require that people work or manage around them. The decrease in demand attributable to the constraint of overburdened government budgets is a prime example. People developing strategic initiatives to improve the veterinary profession must be cognizant of this factor. In contrast, many demand influences are actionable and can be changed to increase demand. Those include drivers of client demand, such as animal tracking, certification or verification, herd management services, and expanding market opportunities related to the globalization of the food supply system. Constraints attributable to skill limitations, such as lack of business and practice management skills and resistance to new technologies, are also highly actionable.
Future demand trends should be viewed as providing many opportunities, not threats. Even the trend of consolidation within the food system and the increasing size and scale of operations are opportunities. There are many experts in the private practice areas who believed that consolidation was an opportunity to expand services offered and deliver value-added benefits to these operations. If those who believed that consolidation was an opportunity to increase demand are correct, then veterinarians who are able to provide added value to these clients will see strong increases in demand that are well in excess of the mean demand forecasts for the private practice areas. Changes that improve the capacity to deliver value to large farming and ranching operations will make the higher estimates for increases in demand (ie, those closer to the 75th percentile estimates) more probable. With the exception of the federal–food safety and security panel, nonprivate practice panels (government, academia, and industrial) believe that the increasing size of operations would have no effect or be a positive driver to increase future demand.
A key finding of the study reported here was the predicted shortages of food supply veterinarians. Although the mean overall forecast for demand was approximately an increase of 1%/y (mean of +12.46% for the 12-year period; Table 2), the mean overall forecast for shortage is approximately 4%/y. This means that for every 100 veterinarians needed in FSVM, there will be only approximately 96 veterinarians available. Ensuring that we have an adequate supply of food supply veterinarians is a pressing challenge. Similar to the situation with demand, there are important differences in forecasted shortages among areas. Although the small ruminants area and, especially, the poultry area will have a close match between supply and demand, several areas are facing more extreme shortages. These include 2 US federal areas (animal health and food safety and security), mixed–food animal practice, and academia.
Although the academia area did not have the highest projected shortage, shortages in this area may pose the most problems. Solving shortages in all other areas requires veterinary medical schools to graduate more veterinarians who are focused on FSVM careers. Shortages in the academia area make it less likely that the veterinary schools will graduate more food supply veterinarians. Negative messages to students about careers in FSVM will be confirmed when they see inadequate faculty resources. Also, veterinary students who pursue an FSVM focus are less likely to develop needed skills when faculty positions are vacant or veterinary schools have an inadequate complement of faculty.
Factors influencing the supply of veterinarians into FSVM careers are a mixture that ranges from relatively fixed factors that are not likely to change and must be addressed by working or managing around them to factors with more actionable supply constraints that can be directly decreased. Negative experiences that discourage students from a focus on a career in FSVM are high on the list of supply constraints (eg, less emphasis on food animal practice, little exposure to FSVM careers, and lack of food animal externships) that can be mitigated. Similarly, constraints stemming from a lack of positive role models can also be changed. Better mentoring and career support while students are in veterinary school and as they enter into FSVM careers as recent graduates will help to attract and develop veterinary students as well as to retain competent veterinarians in FSVM careers. Other supply constraints stem from rural socioeconomic factors, such as lack of spousal career options, and are less actionable; thus, they will require creative solutions to enable people to work or manage around such inflexible constraints.
Tactics that will be effective in lessening shortages varied slightly among areas. Overall, strong levers of action directly involve colleges or schools of veterinary medicine. Those ideas require changes in resource allocations (including more FSVM faculty and the use of practitioners in training of veterinary students) and curricular initiatives (eg, postgraduate fellowships, paid externships, and more coverage of FSVM early during the curriculum). Expanding the concept of Centers of Excellence is a broader-scale solution that will require external resources and cooperation among veterinary schools. Mentoring is another tactic that was mentioned, and the tactic with the highest overall rank was student debt reduction or scholarships for service in areas of need.
Conclusions
The FSVM sector of the veterinary profession is at a crossroads.1 There are many changes in the larger context of the veterinary profession that provide both challenges and opportunities. These include new and unique demands and opportunities that require different skills and more veterinarians to serve in these changing roles. At the same time, there should be serious concerns about the adequacy of the supply of veterinarians who will be available to enter into these changing roles. Although results of the study reported here are in direct contrast to conclusions of the KPMG LLP study17 with regard to FSVM, we believe that conclusions for the KPMG LLP study offered in another report27 accurately characterize the current situation: “…if our [the veterinary] profession doesn't make a series of strategic and substantive changes to create a different future, it is likely that we will wind up in a state of lost opportunity. Most important, the profession will fail to meet societal needs and demands.”
Changing the FSVM sector is a demanding task with many alternatives and multiple competing constituencies, but there are good opportunities for leaders in the veterinary profession to move forward with strategic initiatives that will create a better future—for those in the veterinary profession and for the clients and society that FSVM serves. Every profession has strengths and weaknesses. Leaders need to capitalize on the many strengths of FSVM and move forward to address challenges and take advantage of the numerous opportunities.
Resolving the continuing problems of labor shortages in FSVM is the central challenge for the veterinary profession. Even the conservative numbers reported here indicate unacceptable risks, and the actual labor shortages will probably be worse than predicted. Continuing shortages have the potential to lead to catastrophic economic and human health problems for the United States and Canada. There are too many historical examples to reach any other conclusion.
Many opportunities exist that can be used to address the challenge of potential future shortages. In providing solutions for such shortages, the following should be key areas of focus:
•The veterinary profession should reconsider how colleges or schools of veterinary medicine select, retain, and educate students for careers in FSVM. Veterinary schools need adequate resources to accomplish this task and must make difficult choices among competing demands for limited resources.
•Private practitioners and the larger population of veterinarians involved in other aspects of FSVM need to be part of any solution. Initiatives are needed that will provide career support and mentoring for veterinary students and recently graduated veterinarians. These efforts should continue the revitalized professional development process started in veterinary schools and ensure that there will be an adequate cadre of veterinarians involved in FSVM.
•Student debt reduction and scholarship programs for graduates who enter into career areas in which there is a need (ie, shortage) should be expanded. Government and privately sponsored programs will probably be needed to eliminate financial barriers for pursuit of an FSVM career.
•The shortage of veterinarians needed for FSVM roles in government is a particularly striking result. Protecting the public and the integrity of the global food supply system is an important priority, and it is critical to have an adequate number of veterinarians in this role. Developing effective partnerships with private practice veterinarians, particularly those in rural areas, may help lessen the burdens currently placed on government food supply veterinarians and create a more effective system.
•Given the complexity of changes involved with the process, it is particularly important to have a systematic method of evaluation that will enable interested parties to track progress toward solving shortage problems. An effective strategy must have multiple tactics focused on many outcomes. It is unlikely that all efforts will be perfectly implemented. Evaluating the impact of these strategies will provide opportunities to learn and fine-tune initial efforts.
A decrease in the future demand for FSVM services is not the looming problem some have feared. Decreased demand for veterinarians is likely to be evident only in selected areas of FSVM, such as poultry. Other areas will see increases in demand, and this will contribute to the shortage problem. If labor shortages can be resolved, then opportunities to increase the demand for FSVM should be pursued.
Strategies to expand the demand and help veterinarians in FSVM careers thrive should include the following:
•Analysis of factors that influence the demand for FSVM services has identified several opportunities. The unique skill sets that veterinarians can offer to clients can add value beyond the cost of their services. Use of technology, improving practice management skills, and focusing on emerging areas with high-demand needs will make an adequate demand situation even better.
•The challenge of consolidation in the food supply system must be pursued as an opportunity. It will be a threat only if veterinarians do not adapt to this widespread reality. There are veterinarians who have established themselves as good role models, and they should be studied. Learning points obtained from these veterinarians need to be understood and communicated so that all can benefit. Offering highvalue service for client operations will provide an economic platform to improve incomes, and this will contribute to resolving future labor supply problems before they arrive.
•Effectively enhancing the demand for food supply veterinarians is a complex process, and systematically tracking the efforts and learning from successes and failures will make the process more effective for achieving long-term strategic goals. The systematic research processes used in the study reported here can also be used to evaluate efforts to improve demand.
Results of the study reported here provide a substantial empirical basis for moving forward. The supply and demand factors provide points of reference and leverage points for development of strategic initiatives.
In the full-length report,23 information on factors unique to specific panels provides additional information for each area. Future demand opportunities and problematic shortages should motivate all who are involved to move forward with haste to change the projected pattern of shortages forecasted for many areas. The list of tactics that can be used to resolve shortages provides a good starting point for leaders who will develop strategic initiatives to strengthen FSVM and ensure that it will continue to serve society by protecting animal health and addressing related public health concerns. The future collective health and economic well-being of the United States and Canada are at stake. That future is not yet determined but will be created by the collective strategic actions for FSVM that the veterinary profession will pursue in the near future.
References
- 1↑
US National Research Council. Animal health at the crossroads: preventing, detecting, and diagnosing animal diseases. Washington, DC: National Academy Press, 2005.
- 2↑
Larson RL. Food animal veterinary medicine: leading a changing profession. J Vet Med Educ 2004;31:341–345.
- 3
Marshak RR. Veterinary schools and the profession: a search for bearings in the new century. J Am Vet Med Assoc 2005;227:1234–1238.
- 4
Radostits O. Engineering veterinary education: a clarion call for reform in veterinary education—let's do it! J Vet Med Educ 2003;30:176–190.
- 5
Nielsen NO. Will the veterinary profession flourish in the future? J Vet Med Educ 2003;30:301–306.
- 7↑
Hird D, King L, Salman M, et al. A crisis of lost opportunity—conclusions from a symposium on challenges for animal population health education. J Vet Med Educ 2002;29:205–209.
- 9
Moon HW, Kirk-Baer C, Ascher M, et al. US agriculture is vulnerable to bioterrorism. J Vet Med Educ 2003;30:96–104.
- 10
King L. Veterinary medicine's role in biodefense and public health. J Vet Med Educ 2003;30:161–163.
- 11
Becker KM. An epiphany: recent events highlight the responsibilities, roles, and challenges that veterinarians must embrace in public health. J Vet Med Educ 2003;30:115–120.
- 12
Buntain BJ. Emerging challenges in public health protection, food safety, and security: veterinary needs in the USDA's Food Safety and Inspection Service. J Vet Med Educ 2004;31:334–340.
- 13
Kelly A, Heider LE, Prasse KW. Renewed directions in veterinary medical educations. J Am Vet Med Assoc 2003;222:1511–1512.
- 14
Radostits O. Are too few veterinary graduates choosing food animal practice? What is the problem? American Association of Bovine Practitioners Web site. Available at: www.aabp.org/members/resources/Endangered.pdf. Accessed May 19, 2006.
- 16
Elmore RG. Recruitment and retention of veterinary students for food animal practices. J Am Vet Med Assoc 2003;222:1697–1699.
- 17↑
Brown JP, Silverman JD. The current and future market for veterinarians and the future market for veterinarians and veterinary medical services in the United States. J Am Vet Med Assoc 1999;215:161–183.
- 18
Macleod C. Department for Environment, Food and Rural Affairs Web site. House of Commons Environment, Food and Rural Affairs Committee report—vets and veterinary services. Available at: www.defra.gov.uk/corporate/consult/vetservices/letter.htm. Accessed May 18, 2004.
- 19
The United Kingdom Parliament Web site. Select Committee on Environment, Food and Rural Affairs: sixteenth report. Available at: www.publications.parliament.uk/pa/cm200203/cmselect/cmenvfru/703/70304.htm. Accessed May 18, 2004.
- 20↑
Australian Government: Department of Agriculture, Fisheries and Forestry Web site. Review of rural veterinary services report. January 2003. Available at: www.affa.gov.au/corporate_docs/publications/pdf/animalplanthealth/chief_vet/vetreport.pdf. Accessed Dec 11, 2004.
- 21↑
Rowe G, Wright G. Expert opinions in forecasting: the role of the Delphi technique. In: Armstrong J, ed. Principles of forecasting. Boston: Kluwer Academic Publishers, 2001;125–144.
- 22
Rowe G, Wright G. The Delphi technique as a forecasting tool: issues and analysis. Int J Forecast 1999;15:353–375.
- 23↑
Andrus DM, Gwinner KP, Prince JP. Estimating FSVM demand and maintaining the availability of veterinarians of careers in food supply related disciplines in the United States and Canada, 2006. Available at: www.avma.org/public_health/fsvmc/fsvmc_toc.asp.
- 24
Stevens J. Applied multivariate statistics for the social sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Publishers, 1992.
- 25
Bray J, Maxwell S. Multivariate analysis of variance. Newbury Park, Calif: Sage Publications, 1985.
- 26↑
Cronbach L. Coefficient alpha and the internal structure of tests. Psychometrika 1951;16:297–334.
- 27↑
King L. “It was the best of times, it was the worst of times”: a perspective on the KPMG study. J Am Vet Med Assoc 2000;217:996–998.
