Introduction

Following the onset of the COVID-19 pandemic, the demand for veterinary services increased, as reflected by an increase in the number of appointments at companion animal practices in 2020 and 2021 versus 2019.¹ Simultaneously, many practices were forced to operate with fewer team members and new workflows, while pet owners waited longer to secure appointments and to be seen for emergencies.¹

On the employment front, hiring struggles are clear. The monthly number of veterinary job openings in the AVMA’s Veterinary Career Center approximately doubled between January 2021 and August 2022.² Further, a survey of employers revealed that these job openings are taking longer to fill, with an average of only one-third (0.6 of 1.8) of veterinarian job openings filled per practice.² These and other factors have helped fuel the perception that veterinary practices are experiencing a labor shortage and that one solution to the resulting “busyness” for existing staff would be to increase the supply of veterinarians and other veterinary professionals.

The tolls of this busyness on practice health are seen in results of recent surveys. Veterinarians worked a median of 5 more h/wk in 2020 through 2022 than they did in previous years.³ Burnout, as measured with the Professional Quality of Life scale, continues to increase among veterinarians, with the average burnout score in 2022 exceeding 30 (of 50) for the first time since measured for relief veterinarians, associate veterinarians, practice owners, and public practice veterinarians alike.³ Indeed, the cost of veterinarian burnout attributed to the US veterinary industry has been estimated to be between $1 and $2 billion annually.⁴ The percentage of veterinarians with serious psychological distress has also increased.⁵

A long-term solution to meeting an increased demand for veterinary services—and alleviating at least some of the observed tolls on practice health—would be to increase the supply of veterinarians. However, such a strategy would be effective only if the demand for veterinarians truly exceeds supply and only for as long as that situation continues. The potential long-term impacts must also be considered. For example, decisions to hire more staff can have long-term financial impacts on practices—impacts that might cause problems when markets change. In addition, more veterinarians could mean less job competition and thus lower or stunted salaries, as well as more graduates with educational debt as veterinary schools strive to increase class sizes to meet anticipated needs.

As we settle into the new, postpandemic “normal,” along with the rest of the world, it would be valuable to consider other ways of meeting demand that may provide more immediate—and more sustainable—gains, no matter how markets are behaving. One such strategy would be to improve practice efficiency. Efficiency essentially means doing things in the most economical way. Various types of efficiency are described in the economic and healthcare literature. We focus here on technical efficiency, which we consider synonymous with operational efficiency for discussion purposes. Technical efficiency captures how resources are used during the provision of services.⁶ In this regard, a highly efficient practice uses the least amount of resources (inputs) to achieve maximal production (outputs). An improvement in technical efficiency could therefore lessen the need for more veterinarians and other veterinary team members, allowing more patients to be seen by staff on hand each day and translating to higher revenues and profits and greater employee engagement and well-being.

Any strategy to improve technical efficiency presupposes that veterinary practices have room for growth in this area. Therefore, it becomes important to evaluate how practices are doing in this regard. Efficiency analysis can be used for this purpose and to provide evidence-based guidance on how efficiency might be improved. The purpose of the present study was to address these aims using data from US companion animal practices captured by the 2022 AVMA Veterinary Practice Owner Survey.

Materials and Methods

Statistical analysis

Background—

Two approaches are commonly used to estimate relative efficiency (RE) in healthcare industries: stochastic frontier analysis and data envelopment analysis (DEA).⁸ The former is a parametric model and requires specification of a functional form that reflects the relationship between the inputs and the output. In addition, and as is often the case with econometric models, stochastic frontier analysis handles only 1 response variable (output) at a time. DEA was designed to address these restrictions. With DEA, no a priori knowledge is needed of the data-generating process underlying the performance data under investigation.⁸ This means it is unnecessary to specify the functional form of the relationship between inputs and outputs. Second, DEA is effective and provides useful performance insights even in situations where data on the prices of inputs are unavailable or difficult to access.^9–11 Third, DEA can appropriately handle multiple response variables (outputs) and multiple predictor variables (inputs) at a time.

DEA—

In the present study, DEA was used to measure the technical efficiency of included companion animal practices. In this model, the practices (or decision-making units [DMUs]) were assumed to use similar production technologies and produce similar goods and services.¹²

A general model to estimate efficiency, defined as the ratio of weighted sum of outputs to weighted sum of inputs (θ_p), can be expressed mathematically as follows¹²:

View Expanded

such that θ_p ≤ 1, p = 1 through n, α_r ≥ 0, and γ_i ≥ 0. In this equation, Q_rp represents the r^th output from the p^th practice, X_ip represents the i^th input from the p^th practice, and α and γ represent the weights placed on each input and output.

The main assumption for this analysis was that an increase in inputs would be met by a proportional increase in outputs (ie, constant returns to scale). Statistical software (R version 4.2.2; R Foundation for Statistical Computing) was used to determine the set of weights that maximized the ratio for each DMU and to identify high-performing DMUs. Each DMU was thus assigned an RE score between 0 and 1, whereby 1 indicated that the DMU had reached the highest level of output that could be achieved given the combination of inputs used. In the healthcare industry, these practices are considered to be exhibiting best practice in performance or benchmark studies. All other practices were assigned a score relative to this best practice level. On the basis of derived RE scores, practices were classified as having high efficiency (RE = 1 or 100% efficient), moderate efficiency (RE > 0.7 but < 1.0), or low efficiency (RE ≤ 0.7).

Descriptive statistics were also computed to summarize characteristics of the included practices, prior to DEA. Due to nonnormality of the data distribution for many values, these characteristics are reported as median, minimum, and maximum values.

Results

Discussion

Understanding the sources and types of inefficiencies in veterinary practices is important for guiding efforts to optimize operations and allocate resources. Optimal use of resources could help enhance quality of services, improve client satisfaction, boost practice profitability, and increase employee engagement and retention.

In the present study, over half (58% [35/60]) of included companion animal practices were found to be less efficient than their peers, suggesting opportunities for improvement. Three general strategies can be used to optimize efficiency: reduce resource levels (inputs) while maintaining production (outputs), increase production while maintaining resources as is, or reduce resource levels and increase production concurrently. Given the aforementioned busyness of veterinary staff, reducing available resources is likely an untenable solution to improving efficiency unless those resources are unnecessary or drain on efficiency. However, our data suggested that it would be possible to optimize efficiency by boosting productivity with existing staff. Indeed, if the DEA model results were extrapolated to 10,000 practices and productivity increased to achieve high efficiency in all as measured here, this could theoretically obviate the need for 3,633 FTE veterinarians, 7,838 FTE veterinary technicians and assistants, and 7,133 FTE nonmedical staff. Alternatively, with existing staff, those 10,000 practices could see a total of almost 9 million more patients/y. The potential impact of improved efficiency is backed by other research,¹³ in which a “practice productivity index” was estimated. According to that study, a companion animal practice could save up to 2,000 h/y (the equivalent of adding 1 FTE veterinarian), depending on its index value, by improving capacity.

Interestingly, high-efficiency practices in the present study employed more FTE veterinarians than other practices, which countered our expectation that the more efficient practices would be ones that used less resources, including less staff. However, this observation makes sense if one considers that those veterinarians also had more appointment slots available to see patients and saw more patients per day. For practices with staffing challenges or limited space, the question then becomes how to increase available appointment slots with the veterinarians and physical space (exam rooms) the practice already has. Underused areas represent expenses and can tax efficiency when not generating revenue. One answer would be to make maximal use of both available exam rooms and time through high-density scheduling.¹⁴ For example, a veterinarian and well-trained aide could cover 2 rooms together simultaneously, with the CVT performing the initial patient screen in one room, then moving on to the next room when the veterinarian comes in and takes over. Building on this, at the same time another veterinarian could handle surgeries and a third could manage hospitalized patients, treatment area procedures such as dental scaling, walk-ins, or emergencies. This example also speaks to the importance of assessing workflows, processes, and the practice’s physical layout for bottlenecks, dis- or misuse, and other detractors from efficiency. Among other potential outcomes, such an assessment could even lead to downsizing of physical space or services offered to promote maximal use of space and staff.

Another option for increasing production, without changing available resources, would be to use the best person for the task, based on their qualifications, competencies, and comfort level and as allowed per relevant state veterinary practice acts. Logic suggests that when a veterinarian performs tasks another team member could do, they see fewer patients, limiting their production and impacting efficiency. This appears to happen commonly, as 40% of veterinary technicians report not feeling fully utilized.¹⁵ And yet, in a previous study,¹⁶ higher staff-to-veterinarian labor ratios were found to be associated with higher gross revenue and productivity.¹⁶ Optimal engagement of staff could also improve job satisfaction and morale and reduce turnover^15,17,18—all things that also can further tax efficiency and the practice’s bottom line.

A third option to boost production would be to adopt and fully utilize technologies designed to decrease the onus on people and make routine tasks or processes more efficient. In a 2022 survey³ of veterinary practice owners, only 58% of practices were reported to use electronic medical record software or practice information management system (PIMS), 45% communications software that integrates with PIMS, 33% online appointment scheduling systems, 34% digital inventory management systems, and 21% telehealth services. Among respondents in another study,¹³ 78% reported not using PIMS to its fullest. The veterinarian owners of high-efficiency practices in the present study spent less time than those in less efficient practices on practice management–related activities, freeing the rest of their time for seeing patients and clients. Whether those owners embraced technology remains unknown but would be interesting to explore in future studies.

Finally, leadership can also influence efficiency. In a 2022 survey of veterinary technicians,¹⁵ 22% chose “under-managed veterinary practice” as one of the most challenging aspects of their job. Leadership efforts that support efficiency include developing a mission statement for the practice and making efficiency a strategic pillar, hiring a practice manager or training an interested teammate to fill the role, clearly defining staff roles and tying them into the mission, promoting teamwork and culture of respect and support, and providing or encouraging training or continuing education opportunities for interested staff. It is also useful to measure productivity before and after any efficiency-related changes are made to see if those changes are working.

The present study had some limitations that need to be considered when interpreting the results. These include the small sample size and limitations inherent to DEA. The DEA approach allowed estimation of the efficiency of companion animal practices relative to those displaying best practice but not the absolute efficiency of each. High-efficiency practices were assumed to be 100% efficient, yet in reality even high-efficiency practices have room for improvement. Further, the fact that 42% of included practices were classified as having high efficiency suggested the developed DEA model may have lacked some discriminatory power.

Another limitation was that participating practices differed somewhat from the sample from which they were drawn,³ including more veterinarians and CVTs on staff, suggesting a certain amount of selection bias. Consequently, the generalizability of the DEA results to the general US companion animal practice population is unclear. However, it should be noted that these practices were comparable to the sample in other aspects (eg, number of examination rooms, gross annual revenue, and hours of operation),³ so some of the general patterns observed may extend to other practices. Additional research into routines or workflows associated with high or low efficiency could yield more actionable insights.

In conclusion, the present study revealed opportunities to improve efficiency in companion animal practices and the potential benefits of doing so. These findings could be considered hopeful, as they highlight a possible solution to current staffing challenges through improving productivity with available staff, rather than through hiring and onboarding new staff. Employers may do well to ride out the short-term pain of an uncertain labor supply and take the opportunity now to ensure their practice is running at optimal efficiency before making financial decisions that could have long-term impacts. Improved efficiency could put the practice in a better position and ease some of the frustrations arising from the challenge of finding new team members, without compromising the quality of, access to, or cost of care. Solutions could be as simple as engaging existing staff to their full potential, identifying and addressing workflow issues, leveraging technology, and providing strong leadership.