Introduction
The act of selecting and prescribing a drug for a patient is a complex process that is key for successful treatment. Wherever possible, 3 areas of research should be consulted for informed drug prescription in a pertinent population or individual1–3: pharmacological studies that investigate what concentration drugs reach in the body and the associated biological response, effectiveness studies that confirm that the drug is able to treat or prevent the pertinent condition, and safety studies that ensure that all potential adverse effects associated with the drug are accounted for. This ideal framework is rarely available for drugs in veterinary medicine, and the status is even worse in exotic animal medicine.4
Drug prescription and administration are among the most common interventions for clinical veterinarians. In small animal first-opinion practices, veterinarians prescribe or administer medications in 83% (40/48) of their consultations.5 However, veterinarians may have limited time to perform these tasks, considering that the average time spent with the client is 12 minutes,5 and that most first-opinion practices schedule consults for 10 to 15 minutes.6 It is therefore likely that this type of decision-making, at least in first-opinion practices, is under substantial time pressure.7 Drugs can be prescribed at a standard amount per individual (ie, fixed dosing) or at an individualized dose, often based on individual body weight (ie, weight-based dosing).8 The first approach is commonly used in general human medicine and in specific instances in veterinary medicine (such as vaccination),8 whereas the latter approach is commonly used for prescribing drugs, such as antimicrobials or anti-inflammatories, in small animal practice. In addition, compared with human medical formularies, veterinary medical equivalents may have lower levels of oversight, wider dose ranges (dose ranges of drugs for exotic animals can be extremely wide9), and more varying levels of research-supported evidence,10–12 which further highlights greater freedom for veterinarians in selecting drug dosages for exotic animals. It is therefore important to understand on what basis veterinarians select drug dosages for these animals to address potential shortcomings in this process. The primary objective of the study reported here was to assess what information sources veterinarians use to select drug dosages for treating exotic animals and how veterinarians implement this information.
Materials and Methods
Target population and survey development
We searched 2 leading reference databases13,14 with the following string: “(information) AND (sources) AND (exotics OR zoological) AND (survey OR questionnaire)” on May 25, 2021. No reports of surveys on information sources used by veterinarians treating exotic animals were found doing these searches.
A questionnaire was developed to be administered through an online survey platform.15 The target population of this voluntary survey were veterinarians practicing exotic animal medicine. No eligibility criteria were applied for the type of animal treated, years in practice, geographic, demographics, level of specialization, or the setting in which veterinarians practiced. An open survey design was selected to achieve a larger sample size and include a more diverse population. A first version of the questionnaire was drafted and tested by sending an online invitation to 5 veterinarians. These veterinarians were informed that their results would not be included in the final survey. The preliminary version of the questionnaire was then submitted to the Oklahoma State University Institutional Review Board (IRB) for institutional revision. The questionnaire was modified based on the requests of the IRB by adding the answer option ‘prefer not to respond’ to each question. The final version of the questionnaire was approved as exempt of IRB review by the Oklahoma State University IRB (Application No.: VM-19-1).
Technical details of the questionnaire
The questionnaire was estimated to take about 5 minutes to complete and consisted of 18 questions divided over 9 adaptive sections. Several of these sections depended on the answers in other sections (Supplementary Appendix S1).16 At any point in the survey, participants could review and possibly modify answers to previous questions.
Section 1 consisted of preliminary statements (ie, “You have read the information above,” “You voluntarily agree to participate,” “You are 18 years of age or older,” and “You are a veterinarian”) to which participants were asked to indicate whether they agreed or disagreed. By selecting the agree option participants were directed to Section 2, whereas by selecting the disagree option the survey was ended.
Section 2 collected information on geographic location, years of veterinary practice, degree or degrees held, work setting, type of species treated, and how they heard about the survey. Two questions in this section had the answer option of “I am not a veterinarian,” which if selected ended the survey. Section 3 included 3 questions that focused on the type of information sources used by the participants and the language of these sources.
Sections 4 and 5 focused on printed information sources, and in both sections, the answer options (14 titles and images of covers of textbooks and formularies) were shown in a randomized order per participant to prevent order bias (ie, respondents could be prone to select certain answer options due to the order in which answers were presented17). Section 4 included only 1 question, “Did you obtain drug dosage information in the last month from any of these sources (including previous editions)?” for which participants were allowed to select multiple answer options. Section 5 asked participants “From the following sources, where did you look for drug dosages in the majority of your cases?” for which only 1 answer option could be selected. Selecting the answer option “Exotic Animal Formulary by Carpenter,” which included images of the cover art for 3 editions9,18,19 as examples, directed the participant to Section 6, whereas selecting any other answer option directed the participant to Section 7. Sections 6 and 7 had open questions on how the participants used the formulary or other textbooks, respectively.
Section 8 included 3 multiple choice questions about how often participants would check the source of a dosage in a textbook or a formulary, how often they would search the scientific article on which the dosage was based, and how they would deal with not having access to the full-text article and 1 open question about how to make dosages of drugs for treating exotic animals quicker and easier to retrieve.
On the last page of the questionnaire, participants had the option to submit the questionnaire. The software kept no track of unsubmitted questionnaires.
Outcomes
The primary outcomes for this survey were: (1) the most commonly reported information sources used for selecting drug dosages, (2) the proportion of respondents who reported only sometimes or never checking the sources of a dosage when using a new drug, (3) the proportion of respondents who reported only sometimes or never searching the original publication on the dosage, (4) the proportion of respondents who reported that they would base their dosages on the abstract of an article if they had no access to its full text. The secondary outcomes were: (1) the most commonly reported printed information source used for selecting drug dosages, and (2) the association of respondent characteristics (geographic location, years of practice, qualifications [ie, degrees and certifications held], work setting, species treated) with the primary outcomes.
Survey dissemination and informed consent
The survey was open from September 2, 2019, to December 2, 2019. A link to the survey was disseminated by means of 4 different strategies:
(1) Email: email addresses of the corresponding authors of articles published in the Journal of Avian Medicine and Surgery between March 1996 and August 2019, the Journal of Zoo and Wildlife Medicine between January 2005 and August 2019, and the Journal of Exotic Pet Medicine between January 2016 and August 2019 were automatically extracted through the use of an abstract and citation database.20 In addition, email contacts were exported from the professional veterinary mailing list of the principal investigator (ND). A first email was sent the week of September 2, 2019, to 2,434 recipients, and a reminder email was sent the week of September 16, 2019, to 2,055 recipients (Supplementary Appendix S2). Email recipients had the option to opt out by replying “stop” to the emails. The reminder email was not sent to email addresses of individuals who had opted out of the survey or acknowledged completion of the survey after the first email or to those email addresses that were not functional; (2) Professional forums: posts were published on international and national forums for exotic animal veterinarians (Exotic DVM forum21 and Italian Society of Veterinarians for Exotic Animals [SIVAE] forum22); (3) Social media: posts were published on online veterinary groups and on the professional account of the principal investigator; (4) A professional association: an interim evaluation of the geographic location of the respondents showed that respondents from Australia were a minority. The Unusual Pet and Avian Veterinarians (UPAV) group, a group in the Australian Veterinary Association, was contacted and asked to advertise the survey among their members.
These forms of disseminating the survey were selected in an attempt to increase the generalizability of the results of the survey. Regardless of the contact method, no incentives were offered to participating veterinarians. Additionally, participation was voluntary.
Statistical analysis
All data recorded in the survey were categorical. Percentages reported were calculated over the total number of respondents, unless otherwise specified. The impact of 5 categorical respondent characteristics (geographic location, years of practice, degree or degrees held, work setting, and species treated) was evaluated on 4 outcomes (primary source of drug dosages [question 12], checking references [question 15], searching for articles [question 16], and using findings in abstracts for clinical decisions [question 17]).
The outcome for primary source of drug dosages had 5 different nominal answer options, and as such, multinomial logistic regression was employed to model the chances that the respondents selected an answer option other than formularies, compared with selecting the formularies option (reference answer option). Five univariable multinomial regression models were built for each individual predictor variable to calculate unadjusted ORs and 95% CIs. Predictor variables were then included in a multivariable model if they had ≥ 1 significant (P < 0.05) univariable association. Goodness-of-fit was assessed with the Pearson χ2 test (P < 0.05 indicated poor model fit) and with the likelihood ratio test (P < 0.05 indicated an improvement upon the model including the intercept alone). Nagelkerke pseudo R2 values were used to compare 2 similar multivariable models after the exclusion of a small category (eg, respondents who indicated their work setting as wildlife).
The outcomes checking references and searching for articles were dichotomized as answer options of always or most of the time versus sometimes or never. The outcome using findings in abstracts for clinical decisions was dichotomized as proceed with the abstract versus wait for the full text or search another full text. For these 3 outcomes, the following 2 answers by respondents were excluded: “prefer not to respond” and “I have never been in that situation.” Binary logistic regression models were built to establish associations between respondent characteristics and outcomes. Initially, univariable models were built including only 1 predictor variable to calculate unadjusted ORs and 95% CIs. If at least one of the categories of the predictor variable was associated (P < 0.05) with the outcome, the predictor variable was included in the final multivariable models. Multivariable logistic regression models were subsequently developed to evaluate whether the association between outcomes and predictor variables persisted after multivariable adjustment. To avoid overfitting of the model, no more than 1 predictor variable was included for 10 events.23 The Hosmer-Lemeshow statistic and the Nagelkerke R2 were used to assess goodness-of-fit of the model.
Analysis was performed with the use of a commercial software (SPSS statistics version 22.0; IBM Corp). A 2-sided P value of < 0.05 was considered to be statistically significant.
Results
There were 951 respondents; however, 15 respondents were excluded: 1 for indicating “disagree” as a response to question number 1, and 14 for indicating “I am not a veterinarian” as a response to question number 3 or 4. Therefore, responses of 936 respondents were included in the study.
Respondent characteristics
When asked about their location, the 936 respondents reported either that they were from Europe and the United Kingdom (n = 438 [46.8%]), North America (295 [31.5%]), Central and South America (78 [8.3%]), Asia (73 [7.8%]), Australia (31 [3.3%]), or Africa (19 [2.0%]) or that they preferred not to respond (2 [0.2%]). Regarding years in practice, 459 (49.0%) respondents reported practicing veterinary medicine for 5 to 20 years, 228 (24.4%) for 1 to 4 years, 179 (19.1%) for > 20 years, and 70 (7.5%) for < 1 year. As their highest level of qualification, respondents reported holding a veterinary degree (DVM or equivalent) alone (475/936 [50.7%]), ≥ 1 other postgraduate degree or certification (321 [34.3%]), or ≥ 1 board certification (American College of Zoological Medicine, European College of Zoological Medicine, or American Board of Veterinary Practitioners) related to exotic, zoo, or wild animal medicine (140 [15.0%]).
When asked about their work setting, respondents reported private practice (n = 631/963 [67.4%]), academia (153 [16.3%]), zoo (96 [10.3%]), wildlife center (32 [3.4%]), government (12 [1.3%]), industry (6 [0.6%]), or that they preferred not to respond (6 [0.6%]). When asked about species treated, respondents reported that they worked with only exotic, zoo, or wildlife species (456/936 [48.7%]); a combination of dogs, cats, and exotic, zoo, or wildlife species (54 [5.8%]); or any combination of the aforementioned species plus horses and production animals (54 [5.8%]); or that or that they preferred not to respond (3 [0.3%]).
Most respondents (574/936 [61.3%]) reported that they received an email about the survey. The remaining respondents reported that they found it on social media (264/936 [28.2%]) or heard about it through a colleague (68 [7.3%]), a professional association (24 [2.6%]), or an exotic veterinary medicine forum (5 [0.5%]); 1 (0.1%) respondent answered “prefer not to respond.”
Sources of drug dosages
Most respondents (877/936 [93.7%]) reported that they consulted information sources written in English.
All information sources—When asked about the single most common source for establishing drug dosages, drug formularies were most commonly reported (682/936 [72.9%]), followed by scientific journals (96 [10.3%]), textbooks other than formularies (68 [7.3%]), suggestions by a colleague(s) (47 [5.0%]), and continuing education notes (38 [4.1%]); and 5 (0.5%) answered “prefer not to respond” (Table 1). When asked what sources they consulted in the past month to decide a drug dosage, respondents reported formularies (876/936 [93.6%]), scientific journals (579 [61.9%]), textbooks (522 [55.8%]), continuing education notes (494 [52.8%]), or colleague suggestions (453 [48.4%]), alone or in combination; 2 (0.2%) respondents answered, “prefer not to respond.”
Numbers and percentages of 936 veterinarians grouped by various characteristics (ie, years of practice, qualifications, species treated, location, and work setting) reported in an anonymous, online survey between September 2, 2019, and December 2, 2019, stratified by respondents’ reported single most common source used for establishing drug dosages when treating exotic, zoo, or wildlife species
| Single most common source used for establishing drug dosages | |||||
|---|---|---|---|---|---|
| Respondent characteristic | Formularies | Textbooks | Scientific journals | CE notes | Colleagues |
| Years of practice | |||||
| < 1 | 51 (72.9) | 9 (12.9) | 3 (4.3) | 1 (1.4) | 6 (8.6) |
| 1–4 | 176 (77.5) | 12 (5.3) | 17 (7.5) | 9 (4.0) | 13 (5.7) |
| 5–20 | 331 (72.3) | 31 (6.8) | 56 (12.2) | 22 (4.8) | 18 (3.9) |
| > 20 | 124 (70.5) | 16 (9.1) | 20 (11.4) | 6 (3.4) | 10 (5.7) |
| Qualifications | |||||
| Board certification related to exotics | 95 (68.8) | 6 (4.3) | 32 (23.2) | 2 (1.4) | 3 (2.2) |
| DVM degree or equivalent alone | 351 (73.9) | 31 (6.5) | 35 (7.4) | 21 (4.4) | 37 (7.8) |
| Other postgraduate degree/certification | 236 (74.2) | 31 (9.7) | 29 (9.1) | 15 (4.7) | 7 (2.2) |
| Species treated | |||||
| EZW alone | 332 (73.0) | 22 (4.8) | 66 (14.5) | 18 (4.0) | 17 (3.7) |
| EZW combined with dogs, cats, or both | 318 (75.7) | 35 (8.3) | 21 (5.0) | 19 (4.5) | 27 (6.4) |
| EZW combined with horses, production animals or both | 29 (54.7) | 11 (20.8) | 9 (17.0) | 1 (1.9) | 3 (5.7) |
| Location | |||||
| North America | 231 (78.6) | 12 (4.1) | 32 (10.9) | 3 (1.0) | 16 (5.4) |
| Europe and the UK | 311 (71.7) | 32 (7.4) | 45 (10.4) | 24 (5.5) | 22 (5.1) |
| Other locations | 140 (69.7) | 24 (11.9) | 17 (8.5) | 11 (5.5) | 9 (4.5) |
| Work setting | |||||
| Academia | 99 (65.6) | 11 (7.3) | 32 (21.2) | 2 (1.3) | 7 (4.6) |
| Private practice | 473 (75.3) | 42 (6.7) | 48 (7.6) | 32 (5.1) | 33 (5.3) |
| Wildlife center | 23 (71.9 | 3 (9.4) | 0 (0.0) | 2 (6.3) | 4 (12.5) |
| Zoo | 71 (74.0) | 6 (6.3) | 15 (15.6) | 1 (1.0) | 3 (3.1) |
CE = Continuing education. EZW = Exotic, zoo, or wildlife species, alone or in combination. UK = United Kingdom.
When results for respondents grouped by years in practice were considered, the odds of reported use were similar for formularies, compared with textbooks, scientific journals, continuing education notes and colleague suggestions (Table 2). Compared with veterinarians who had board certification related to exotic, zoo, or wildlife species, other veterinarians with or without other postgraduate degrees had about one-third the odds (OR, 0.36 [95% CI, 0.21 to 0.64] and 0.30 [95% CI, 0.17 to 0.50], respectively; both P < 0.001) of reporting that they used scientific journals more often than formularies in establishing drug dosages, and veterinarians without other postgraduate degrees had about 3 times the odds (OR, 3.34; 95% CI, 1.01 to 11.06; P = 0.049) of reporting that they consulted colleagues more often than formularies. Compared with veterinarians who treated exotic, zoo, or wildlife species exclusively, veterinarians who also treated cats and dogs had one-third the odds (OR, 0.33; 95% CI, 0.20 to 0.56; P < 0.001) of reporting the use of scientific journals more often than formularies, whereas veterinarians who also treated horses and production animals had over 5 times the odds (OR, 5.72; 95% CI, 2.50 to 12.96; P < 0.001) of reporting the use of textbooks more often than formularies in established dosages. Compared with veterinarians from North America, veterinarians from Central and South America, Asia, Australia, and Africa had 3 times the odds (OR, 3.30; 95% CI, 1.60 to 6.81; P < 0.001) of reporting the use of textbooks more often than formularies. In addition, the odds of reporting the use of continuing education notes more often than formularies were about 6 times higher for veterinarians from Europe (OR, 5.94; 95% CI, 1.77 to 20.00; P = 0.004) and those from other locations (OR, 6.05; 95% CI, 1.66 to 22.06; P = 0.006), compared with those from North America. Compared with veterinarians working in academia, veterinarians in private practice had around one-third the odds (OR, 0.31; 95% CI, 0.19 to 0.52; P < 0.001) of reporting the use of scientific articles compared with using formularies for establishing dosages.
Results of univariable and multivariable multinomial logistic regression describing the odds of formularies versus textbooks, scientific journals, continuing education notes, or colleague suggestions, to have been reported by the respondents in Table 1 as the single most commonly used source in establishing drug dosages for exotic, zoo, or wildlife species.
| Textbooks | Scientific journals | Continuing education notes | Colleagues (including online) | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Characteristic | OR (95% CI) | P value | aOR* (95% CI) | P value | OR (95% CI) | P value | aOR (95% CI)* | P value | OR (95% CI) | P value | aOR (95% CI*) | P value | OR (95% CI) | P value | aOR (95% CI*) | P value |
| Years of practice | ||||||||||||||||
| > 20 | Referent | Referent | Referent | Referent | ||||||||||||
| <1 | 1.37 (0.57–3.29) | 0.48 | — | — | 0.36 (0.10–1.28) | 0.12 | — | — | 0.40 (0.05–3.45) | 0.41 | — | — | 1.46 (0.50–4.22) | 0.49 | — | — |
| 1–4 | 0.53 (0.24–1.16) | 0.11 | — | — | 0.60 (0.30–1.19) | 0.14 | — | — | 1.06 (0.37–3.04) | 0.92 | — | — | 0.92 (0.39–2.15) | 0.84 | — | — |
| 5–20 | 0.73 (0.38–1.37) | 0.32 | — | — | 1.05 (0.60–1.82) | 0.86 | — | — | 1.37 (0.54–3.47) | 0.50 | — | — | 0.67 (0.30–1.50) | 0.33 | — | — |
| Qualifications | ||||||||||||||||
| Board | Referent | Referent | Referent | Referent | ||||||||||||
| certification | ||||||||||||||||
| related to EZW | ||||||||||||||||
| DVM | 1.40 (0.57–3.45) | 0.47 | 1.22 (0.43–3.43) | 0.71 | 0.30 (0.17–0.50) | < 0.01 | 0.43 (0.24–0.79) | 0.006 | 2.84 (0.65–12.33) | 0.16 | 2.19 (0.48–9.98) | 0.31 | 3.34 (1.01–11.06) | 0.049 | 3.08 (0.88–10.76) | 0.08 |
| or equivalent | ||||||||||||||||
| alone | ||||||||||||||||
| Other | 2.08 (0.84–5.15) | 0.11 | 1.56 (0.56–4.36) | 0.40 | 0.36 (0.21–0.64) | < 0.01 | 0.47 (0.25–0.86) | 0.01 | 3.02 (0.68–13.46) | 0.14 | 1.80 (0.39–8.39) | 0.45 | 0.94 (0.24–3.71) | 0.93 | 0.84 (0.20–3.44) | 0.81 |
| postgraduate degree | ||||||||||||||||
| Species treated | ||||||||||||||||
| EZW alone | Referent | Referent | Referent | Referent | ||||||||||||
| EZW, horses, | 5.72 (2.50–12.96) | < 0.01 | 4.14 (1.59–10.77) | 0.004 | 1.56 (0.71–3.45) | 0.27 | 1.93 (0.83–4.49) | 0.13 | 0.64 (0.08–4.94) | 0.66 | — | — | 2.02 (0.56–7.30) | 0.28 | 2.24 (0.60–8.38) | 0.23 |
| production | ||||||||||||||||
| animals | ||||||||||||||||
| EZW, dogs, cats | 1.66 (0.95–2.89) | 0.07 | 1.80 (0.93–3.48) | 0.08 | 0.33 (0.20–0.56) | < 0.01 | 0.43 (0.24–0.79) | 0.007 | 1.10 (0.57–2.14) | 0.77 | 0.69 (0.34–1.43) | 0.32 | 1.66 (0.89–3.10) | 0.11 | 1.57 (0.75–3.28) | 0.23 |
| Location | ||||||||||||||||
| North America | Referent | Referent | Referent | Referent | ||||||||||||
| Europe and the UK | 1.98 (0.99–3.93) | 0.051 | 2.33 (1.08–5.02) | 0.03 | 1.04 (0.64–1.69) | 0.86 | 1.57 (0.92–2.68) | 0.10 | 5.94 (1.77–20.00) | < 0.01 | 5.14 (1.49–17.72) | 0.01 | 1.02 (0.52–1.99) | 0.95 | 1.21 (0.60–2.44) | 0.58 |
| Other locations | 3.30 (1.60–6.81) | < 0.01 | 3.15 (1.41–7.07) | 0.005 | 0.88 (0.47–1.64) | 0.68 | 1.12 (0.57–2.21) | 0.75 | 6.05 (1.66–22.06) | < 0.01 | 4.90 (1.31–18.39) | 0.02 | 0.93 (0.40–2.16) | 0.86 | 0.99 (0.42–2.35) | 0.99 |
| Work setting | ||||||||||||||||
| Academia | Referent | Referent | Referent | Referent | ||||||||||||
| Private practice | 0.80 (0.40–1.61) | 0.53 | 0.67 (0.30–1.48) | 0.32 | 0.31 (0.19–0.52) | < 0.01 | 0.48 (0.27–0.84) | 0.01 | 3.35 (0.79–14.20) | 0.10 | 2.61 (0.59–11.56) | 0.20 | 0.99 (0.42–2.29) | 0.97 | 0.60 (0.23–1.52) | 0.28 |
| Wildlife center | 1.17 (0.30–4.55) | 0.82 | 1.25 (0.30–5.15) | 0.76 | — | — | — | — | 4.30 (0.58–32.18) | 0.15 | 3.79 (0.49–29.08) | 0.20 | 2.46 (0.66–9.11) | 0.18 | 2.09 (0.55–7.94) | 0.28 |
| Zoo | 0.76 (0.27–2.15) | 0.61 | 1.09 (0.37–3.26) | 0.87 | 0.65 (0.33–1.30) | 0.22 | 0.62 (0.30–1.24) | 0.18 | 0.70 (0.06–7.84) | 0.77 | 0.67 (0.06–7.63) | 0.75 | 0.60 (0.15–2.39) | 0.47 | 0.58 (0.14–2.35) | 0.44 |
The model was well fitted: Pearson χ2 test, P = 0.27; Likelihood ratio test, P < 0.001; Nagelkerke pseudo R2 = 0.15.
— = Not calculated. aOR = Adjusted OR. EZW = Exotic, zoo, or wildlife species, alone or in combination. UK = United Kingdom.
*The adjusted ORs (aORs) were only reported if the given characteristic was included in the final multivariable model.
The final multivariable model included variables for the respondents’ reported geographic location, work setting, and species treated. The multivariable model was well fitted (Pearson χ2 test, P = 0.27; likelihood ratio test, P < 0.001) and confirmed all the associations observed in univariable models, except for that of veterinarians with a DVM degree (or equivalent) alone having had higher odds of consulting colleagues than did veterinarians with board certification related to exotic, zoo, or wildlife species (Table 2).
Printed information sources—When asked about the printed source of information where they looked for drug dosages in the majority of their cases, the most commonly reported source was a formulary for exotic animal medicine9 (645/936 [68.9%]), followed by a formulary from a national professional association24 (87 [9.3%]), none of the sources listed as answer options (46 [4.9%]), a reptile textbook25 (33 [3.5%]), a zoo and wildlife textbook26 (27 [2.9%]), any manual published by the British Small Animal Veterinary Association (26 [2.8%]), a small mammal textbook27 (21 [2.2%]), an avian textbook28 (8 [0.9%]), a companion exotic animal textbook29 (8 (0.9%), or various other textbooks listed as answer options (27 [2.9%]); however, 8 (0.9%) respondents answered “prefer not to respond.”
Details on drug dose decision
Checking the source of a dosage—Respondents indicated that they checked the source of a dosage mentioned in a textbook or a formulary sometimes (318/936 [34.0%]), most of the time (304 [32.5%]), always (221 [23.6%]), or never (89 [9.5%]). Four of the 936 (0.4%) respondents answered, “prefer not to respond.”
Veterinarians who reported practicing > 20 years had over 2 times the odds (OR, 2.58; 95% CI, 1.46 to 4.57; P = 0.001) of indicating that they searched for a source of a dosage always or most of the time, compared to veterinarians who reported practicing < 1 year (Table 3). Respondents with board certification related to exotic, zoo, or wildlife species had greater odds of indicating that they searched for a source of a dosage always or most of the time, compared with respondents with a DVM degree (or equivalent) alone (OR, 3.03; 95% CI, 1.98 to 4.62; P < 0.001) and those with other postgraduate degrees (OR, 2.19; 95% CI, 1.41 to 3.41; P = 0.001). Veterinarians who reported treating exclusively exotic, zoo, or wildlife species had over 2 times the odds (OR, 2.67; 95% CI, 2.03 to 3.51; P < 0.001) of indicating that they searched for a source of a dosage always or most of the time, compared with those who reported also working with dogs and cats. Veterinarians who reported they were from North America had 1.7 times the odds of indicating they searched for a source of a dosage always or most of the time, compared with veterinarians who reported they were from Central and South America, Asia, Australia, or Africa (OR, 1.73; 95% CI, 1.21 to 2.50; P = 0.003) but not compared with respondents from Europe. Veterinarians who reported working in academia had greater odds of indicating that they searched for a source of a dosage always or most of the time, compared with veterinarians who reported working in private practice (OR, 4.06; 95% CI, 2.65 to 6.21; P < 0.001) or from wildlife centers (OR, 3.42; 95% CI, 1.54 to 7.60; P = 0.003). In the final multivariable model (Hosmer-Lemeshow test, P = 0.29; Nagelkerke R2 = 0.14), respondents who reported that they held board certification related to exotic, zoo, or wildlife species (vs a DVM or equivalent degree alone), treated exotic, zoo, and wildlife species exclusively (vs also treating dogs and cats), were from North America (vs Central and South America, Asia, Australia, or Africa), or worked in academia (vs private practice or wildlife centers) were still more likely to have report that they searched for a source of a dosage always or most of the time.
Results of univariable and multivariable binary logistic regression describing the odds of the respondents in Table 1, grouped by characteristics, answering “always” or “most of the time” versus “sometimes” or “never” for the survey question “You are choosing the dosage of a drug that you have never used before on a particular species. There is a dosage mentioned in a textbook or in a formulary. How often would you check on the source of that dosage?”
| Characteristic | No. (%) of respondents who answered “always” or “most of the time” | No. (%) of respondents who answered “sometimes” or “never” | OR (95% CI) | P value | aOR (95% CI)* | P value |
|---|---|---|---|---|---|---|
| Years of practice | ||||||
| > 20 | 113 (63.8) | 64 (36.2) | Referent | |||
| 1–4 | 130 (57.0) | 98 (43.0) | 1.33 (0.89–1.99) | 0.16 | 0.85 (0.54–1.34) | 0.49 |
| 5–20 | 254 (55.5) | 204 (44.5) | 1.42 (0.99–2.03) | 0.06 | 1.25 (0.85–1.85) | 0.25 |
| < 1 | 28 (40.6) | 41 (59.4) | 2.58 (1.46–4.57) | < 0.01 | 1.61 (0.85–3.04) | 0.14 |
| Qualifications | ||||||
| Board certification | 105 (75.0) | 35 (25.0) | Referent | |||
| related to EZW | ||||||
| Veterinary degree alone | 235 (49.8) | 237 (50.2) | 3.03 (1.98–4.62) | < 0.01 | 1.78 (1.10–2.89) | 0.02 |
| Other postgraduate degree | 185 (57.8) | 135 (42.2) | 2.19 (1.41–3.41) | < 0.01 | 1.53 (0.94–2.50) | 0.09 |
| Species treated | ||||||
| EZW alone | 308 (67.5) | 148 (32.5) | Referent | |||
| EZW, dogs, cats | 184 (43.8) | 236 (56.2) | 2.67 (2.03–3.51) | < 0.01 | 1.93 (1.40–2.66) | < 0.01 |
| EZW, horses, production animals | 31 (58.5) | 22 (41.5) | 1.48 (0.83–2.64) | 0.19 | 1.17 (0.60–2.26) | 0.65 |
| Location | ||||||
| North America | 181 (61.6) | 113 (38.4) | Referent | |||
| Europe and the UK | 247 (56.7) | 189 (43.3) | 1.23 (0.91–1.66) | 0.19 | 0.94 (0.67–1.33) | 0.75 |
| Other locations | 96 (48.0) | 104 (52.0) | 1.73 (1.21–2.50) | < 0.01 | 1.57 (1.04–2.35) | 0.03 |
| Work setting | ||||||
| Academia | 120 (79.5) | 31 (20.5) | Referent | |||
| Private practice | 307 (48.8) | 322 (51.2) | 4.06 (2.65–6.21) | < 0.01 | 2.78 (1.76–4.41) | < 0.01 |
| Wildlife center | 17 (53.1) | 15 (46.9) | 3.42 (1.54–7.60) | < 0.01 | 3.09 (1.36–7.00) | < 0.01 |
| Zoo | 68 (70.8) | 28 (29.2) | 1.59 (0.88– 2.88) | 0.12 | 1.72 (0.94–3.17) | 0.08 |
*Model was well fitted: Hosmer-Lemeshow test, P = 0.29; Nagelkerke R2 = 0.14.
See Table 2 for the key.
Searching for the actual article—Respondents indicated that they sometimes (399/936 [42.6%]), most of the time (282 [30.1%]), always (124 [13.2%]), or never (104 [11.1%]) searched for the article at the origin of the source of a dosage mentioned in a textbook or a formulary. Twenty-four of the 936 (2.6%) respondents answered that they had never been in that situation, and 3 (0.3%) respondents answered, “prefer not to respond.”
Veterinarians who reported that they had board certification related to exotic, zoo, or wildlife species had greater odds of indicating that they searched for the actual article always or most of the time, compared with compared with respondents with a DVM or equivalent degree alone (OR, 2.32; 95% CI, 1.58 to 3.43; P < 0.001) and those with other postgraduate degrees (OR, 1.60; 95% CI, 1.06 to 2.38; P = 0.02; Table 4). Veterinarians who reported treating exclusively exotic, zoo, and wildlife species had over 2 times the odds (OR, 2.68; 95% CI, 2.02 to 3.54; P < 0.001) of indicating that they searched for the actual article always or most of the time, compared with veterinarians who reported also working with dogs and cats. Veterinarians who reported working in academia had greater odds of indicating that they searched for the actual article always or most of the time, compared with veterinarians who reported working in private practice (OR, 3.29; 95% CI, 2.26 to 4.80; P < 0.001) or zoos (OR, 1.69; 95% CI, 1.00 to 2.86; P = 0.05). Veterinarians from different geographic locations and with different years in practice had similar odds of searching for the actual article always or most of the time. In the final multivariable model (Hosmer-Lemeshow test, P = 0.71; Nagelkerke R2 = 0.10), respondents who reported that they treated exotic, zoo, and wildlife species exclusively (vs also treating dogs and cats) or worked in academia (vs private practice or zoos) were still more likely to have reported that they searched for the actual article always or most of the time.
Results of univariable and multivariable binary logistic regression describing the odds of the respondents in Table 1, grouped by characteristics, answering “always” or “most of the time” versus “sometimes” or “never” for the survey question “Let’s assume that you have checked the source of the dosage and it derives from a journal article. How often would you search for the article?”
| Characteristic | No. (%) of respondents who answered “always” or “most of the time” | No. (%) of respondents who answered “sometimes” or “never” | OR (95% CI) | P value | aOR (95% CI)* | P value |
|---|---|---|---|---|---|---|
| Years of practice | ||||||
| > 20 y | 82 (46.1%) | 96 (53.9%) | Referent | |||
| 1–4 y | 95 (43.8%) | 122 (56.2%) | 1.10 (0.74–1.63) | 0.65 | — | — |
| 5–20 y | 208 (46.1%) | 243 (53.9%) | 1.00 (0.70–1.41) | 0.99 | — | — |
| < 1 y | 21 (33.3%) | 42 (66.7%) | 1.71 (0.94–3.12) | 0.08 | — | — |
| Degrees | ||||||
| Board certification related to EZW | 82 (59.0%) | 57 (41.0%) | Referent | |||
| Veterinary degree | 172 (38.2%) | 278 (61.8%) | 2.32 (1.58–3.43) | <0.01 | 1.41 (0.92–2.15) | 0.11 |
| Post-DVM degree | 152 (47.5%) | 168 (52.5%) | 1.60 (1.06–2.38) | 0.02 | 1.11 (0.72–1.71) | 0.64 |
| Species treated | ||||||
| EZW alone | 251 (55.8%) | 199 (44.2%) | Referent | |||
| EZW, dogs, cats | 130 (32.0%) | 276 (68.0%) | 2.68 (2.02–3.54) | <0.01 | 2.05 (1.49–2.82) | <0.01 |
| EZW, horses, production animals | 23 (46.0%) | 27 (54.0%) | 1.48 (0.82–2.66) | 0.19 | 1.33 (0.70–2.52) | 0.38 |
| Location | ||||||
| North America | 131 (45.2%) | 159 (54.8%) | Referent | |||
| Europe and UK | 196 (46.0%) | 230 (54.0%) | 0.97 (0.72–1.30) | 0.82 | — | — |
| Other locations | 78 (40.8%) | 113 (59.2%) | 1.19 (0.82–1.73) | 0.35 | — | — |
| Work setting | ||||||
| Academia | 100 (66.7%) | 50 (33.3%) | Referent | |||
| Private practice | 231 (37.8%) | 380 (62.2%) | 3.29 (2.26–4.80) | <0.01 | 2.32 (1.54–3.48) | <0.01 |
| Wildlife center | 14 (48.3%) | 15 (51.7%) | 2.14 (0.96–4.79) | 0.06 | 2.00 (0.87–4.53) | 0.09 |
| Zoo | 52 (54.2%) | 44 (45.8%) | 1.69 (1.00–2.86) | 0.05 | 1.86 (1.09–3.18) | 0.02 |
The model was well fitted: Hosmer-Lemeshow test, P = 0.71; Nagelkerke R2 = 0.10.
**Variable not included in the multivariable model.
See Table 2 for the rest of the key.
Using the abstract—Most respondents (486/936 [51.9%]) reported that if they found the abstract of a journal article but could not download the full text immediately, they would read the abstract and base their dosages on it, whereas 208 (22.2%) respondents indicated that they would base the dosage on another journal article that had the full text available. One hundred forty-six of the 936 (15.6%) respondents reported that they had never been in that situation, 81 (8.7%) respondents reported that they would wait and keep trying to retrieve the article, and 15 (1.6%) respondents answered “prefer not to respond.”
Veterinarians who reported practicing < 1 year had less than half the odds (OR, 0.39; 95% CI, 0.21 to 0.76; P = 0.005) of reporting that they used the abstract as the basis for their dosage decisions, compared with veterinarians who reported practicing > 20 years (Table 5). Veterinarians who reported having other postgraduate degrees had about two-thirds the odds (OR, 0.62; 95% CI, 0.39 to 0.98; P = 0.04) of indicating that they based their dosages on abstracts, compared with veterinarians who reported board certification related to exotic, zoo, or wild animal medicine. Veterinarians who reported that they treated exclusively exotic, zoo, and wildlife species had lower odds of indicating that they based their dosage decisions on abstracts than did veterinarians who reported that they also treated horse and production animals (OR, 0.41; 95% CI, 0.22 to 0.76; P = 0.005) or dogs and cats (OR, 0.69; 95% CI, 0.51 to 0.94; P = 0.02). Veterinarians who reported they were from North America had less than half the odds of indicating that they based their dosages on abstracts than did veterinarians who reported they were from Central and South America, Asia, Australia, or Africa (OR, 0.37; 95% CI, 0.24 to 0.57; P < 0.001) or those who reported they were from Europe (OR, 0.38; 95% CI, 0.26 to 0.54; P < 0.001). Veterinarians who reported working in academia had 3 times the odds (OR, 3.41; 95% CI, 1.73 to 6.71; P < 0.001) of indicating that they based their dosages on abstracts than did veterinarians who reported working in zoos. In the final multivariable model (Hosmer-Lemeshow test, P = 0.31; Nagelkerke R2 = 0.11), respondents who reported that they treated exotic, zoo, and wildlife species exclusively (vs also treating horses and production animals), were from North America (vs other locations), or worked in zoos (vs academia) were still more likely to have reported that they based their dosages on abstracts.
Results of univariable and multivariable binary logistic regression describing the odds of the respondents in Table 1, grouped by characteristics, responding “I read the abstract and base my clinical decision on it” versus “I wait and try to retrieve the article” or “I decide the dosage based on another journal article that has full text available” for the survey question “Let’s assume that you have found the abstract of the journal article but you can not download the full text immediately.”
| No. (%) of respondents who reported to use the abstract if full-text of an article is unavailable | No. (%) of respondents who reported to wait or use a different article | OR (95%CI) | P value | aOR (95% CI)* | P value | |
|---|---|---|---|---|---|---|
| Years of practice | ||||||
| > 20 y | 101 (64.7%) | 55 (35.3%) | Referent | |||
| 1–4 y | 101 (55.8%) | 80 (44.2%) | 0.69 (0.44–1.07) | 0.09 | 0.90 (0.55–1.48) | 0.68 |
| 5–20 y | 263 (67.8%) | 125 (32.2%) | 1.15 (0.77–1.69) | 0.50 | 1.33 (0.87–2.04) | 0.18 |
| < 1 y | 21 (42.0%) | 29 (58.0%) | 0.39 (0.21–0.76) | < 0.01 | 0.49 (0.24–1.00) | 0.05 |
| Degrees | ||||||
| Board certification related to EZW | 85 (70.8%) | 35 (29.2%) | Referent | |||
| Veterinary degree | 234 (62.1%) | 143 (37.9%) | 0.67 (0.43–1.05) | 0.08 | 0.98 (0.58–1.66) | 0.95 |
| Post-DVM degree | 167 (60.1%) | 111 (39.9%) | 0.62 (0.39–0.98) | 0.04 | 0.98 (0.85–1.66) | 0.95 |
| Species treated | ||||||
| EZW alone | 261 (68.0%) | 123 (32.0%) | Referent | |||
| EZW, dogs, cats | 203 (59.5%) | 138 (40.5%) | 0.69 (0.51–0.94) | 0.02 | 0.94 (0.66–1.35) | 0.74 |
| EZW, horses, production animals | 22 (46.8%) | 25 (53.2%) | 0.41 (0.22–0.76) | < 0.01 | 0.41 (0.21–0.84) | 0.01 |
| Location | ||||||
| North America | 194 (77.0%) | 58 (23.0%) | Referent | |||
| Europe and UK | 201 (56.0%) | 159 (44.0%) | 0.38 (0.26–0.54) | <0.01 | 0.37 (0.25–0.55) | <0.01 |
| Other locations | 90 (55.6%) | 72 (44.4%) | 0.37 (0.24–0.57) | <0.01 | 0.43 (0.27–0.68) | <0.01 |
| Work setting | ||||||
| Academia | 73 (58.4%) | 52 (41.6%) | Referent | |||
| Private practice | 320 (61.4%) | 201 (38.6%) | 1.13 (0.76–1.69) | 0.53 | 1.28 (0.81–2.02) | 0.28 |
| Wildlife center | 17 (63.0%) | 10 (37.0%) | 1.21 (0.51–2.86) | 0.66 | 1.08 (0.44–2.67) | 0.86 |
| Zoo | 67 (82.7%) | 14 (17.3%) | 3.41(1.73–6.71) | <0.01 | 3.14 (1.55–6.39) | < 0.01 |
Model was well fitted: Hosmer-Lemeshow test, P = 0.31; Nagelkerke R2 = 0.11.
See Table 2 for the key.
Discussion
A key finding of the present study was that veterinarians prescribing medications to exotic animals have a strong predilection for formularies. Although the use of information sources varied on the basis of postgraduate degrees held, geographic location, work setting, and species treated, formularies were the most commonly consulted information source for drug dosages. Veterinarians who reported working in academia, compared with private practice, were relatively more likely to use scientific articles than formularies. Also, veterinarians without (vs with) board certification related to exotic, zoo, or wild animal medicine were more likely to use formularies than scientific journals. Compared with respondents who reported treating exclusively exotic, zoo, or wildlife species, respondents who reported also treating dogs and cats were more likely to use formularies than scientific journals, and respondents who reported also treating horses and production animals were more likely to use textbooks than formularies. Compared with veterinarians from North America, veterinarians from Europe were more likely to use continuing education notes than formularies, and veterinarians from Central and South America, Asia, Australia, and Africa were more likely to use textbooks. Although we did not have a plausible explanation for these geographic differences, veterinarians who reported working in academia, treating exclusively exotic, zoo, or wildlife species, or holding board certification related to exotic, zoo, or wildlife species were more likely to make use of scientific articles.
Another critical finding was that most respondents selected the same specific formulary,9 when given several printed information sources to choose from. This implies that this source has the control of the information for the selection of drug dosages for exotic animals worldwide. This specific formulary has references indicated for each dosage. However, when these references have been systematically assessed, they often do not cite primary resources.10–12 This is worrisome particularly in the context that only 221 of the 936 (23.6%) respondents in our survey reported that they always checked the source of a dosage given in a textbook or a formulary. Education of veterinary professionals on how to consult formularies seems indicated.
Our findings were difficult to compare with previous surveys30–32 of sources used by veterinary professionals because of differences in questions or source types. The results of the present study indicated that veterinarians use published, peer-reviewed literature, when making decisions on drug dosages for exotic animals. We found that 62% (579/936) of the respondents used scientific journals in the last 30 days for establishing drug dosages at least once, and that 10% (96/936) used scientific journals as their most common source. In a previous international survey33 of veterinarians, 90% (1,448/1,609) of the respondents declared that they had read peer-reviewed scientific articles, and for 66% (1,207/1,835) of clinicians that was the most common source of information.33 These differences could have been results of spurious or real factors or both. Methodological limitations could explain these spurious differences such as various degree of selection bias (ie, the way the participants were selected in 1 or both the surveys was in some way associated with higher or lower reported use of scientific articles), various degree of reporting bias (ie, the way the questions were formulated in 1 or both surveys could have led the surveyees to give answers that would please the researchers who sent the survey), or both.34,35 If the difference observed is reflecting a real difference in behaviors, this could be due to a different way of consulting information sources for drug-related reasons vs for other medical reasons (eg, when searching for drugs veterinarians often do not consult scientific articles, contrary to when they search for evidence on diagnosis, prognosis or treatment in general), or it could be due to the differences in the sampled populations, such as veterinarians that prescribe treatment to exotic animals do not use scientific articles compared with veterinarians that prescribe treatments for other animals.
Less than half of the respondents reported that they would always or most of the time double-check a dosage given in a formulary or a textbook, by searching for the original source. Results of the present study indicated that veterinarians with (vs without) board certification related to exotic, zoo, or wildlife species, veterinarians treating exotic, zoo, and wildlife species exclusively (vs also treating dogs and cats), and veterinarians from North America (vs Central and South America, Asia, Australia, or Africa), and veterinarians from academia (vs private practice and wildlife centers) were more likely to search for a source of a dosage always or most of the time. Possible explanations for this finding included (1) veterinarians with board certification related to exotic, zoo, or wildlife species and veterinarians treating only those species have more time available per consultation, (2) possibly such board-certified veterinarians also are more likely to examine referral cases than are other veterinarians, and (3) veterinarians working in an academic setting may need to justify their choices of drug dosages, potentially incentivizing them to search for the actual source of a dosage. Searching and critically appraising original articles that are listed as references for a drug, are critical steps for veterinarians when deciding on drug dosages. Reasons for not implementing these steps could include the trust in the published literature and in the authors of textbooks and formularies, paywalls to access published articles, and a lack of time. However, the veterinary literature has critical deficiencies36–39 and the number of articles that are currently freely available online is increasing, especially in interdisciplinary journals.40 In this context it is critical that either clinicians learn to search and critically appraise the scientific literature, or authors of formularies and textbooks focus on evidence-based information and state clearly when information reported is rather anecdotal.
Finally, most (486/936 [51.9%]) respondents indicated that if they found the abstract of a journal article but were unable to download the full text immediately, they would just base their clinical decisions on the abstract. This outcome is congruent with the findings of previous surveys32,33 that showed that the abstract was the single most read section of an article. A recent study41 on reporting in abstracts of veterinary randomized controlled trials demonstrated that only a limited number of recommended reporting items were actually reported. The frequent consultation by veterinarians of just abstracts and the poor reporting in abstracts imply 2 key issues: (1) clinicians should retrieve and critically appraise full texts of articles when selecting drug dosages and (2) authors, peer-reviewers, and editors need to improve the reporting quality of abstracts in veterinary studies.
The present study had limitations. Options to study drug prescription habits include direct monitoring of veterinarians during clinical activity (eg, video-monitoring or direct observation)5,42 or interviewing or surveying these clinicians.43 Video-monitoring and direct observation are likely to be most accurate of the 3 methods, but are also expensive, logistically complicated, and probably would have been limited to a specific sample of veterinarians, reducing the generalizability of outcomes. In open interviews, veterinarians could intentionally or involuntarily conceal habits that could be considered negative and highlight habits considered positive. Based on these reasons, we opted for an anonymous survey. We chose the open survey design format because we aimed to obtain a large sample of diverse respondents (n = 936), compared with the generally poor response rates reported in closed surveys on similar topics.44,45 However, open surveys do not permit the calculation of the response rate. The present survey did not use monetary or other forms of incentives. Incentives may have both positive and negative effects on the results of a survey; however, prepaid incentives are considered an effective way to increase cooperation as compared to postpaid incentives.46 The software kept no track of unsubmitted or incomplete answers; therefore, it was impossible to calculate the completion rate.16 No measures were implemented to prevent the potential risk of multiple entries from the same surveyee.16 Finally, no specific algorithms were implemented to screen for careless responses.47 However, the fact that the survey was voluntary and short should have limited the impact of careless responses.47 Further studies on the prescribing habits of veterinarians treating exotic animals are warranted, and they should overcome some of the limitations of the present study.
Formularies (ie, books containing specific information on drug prescription and dosages) have been developed to help veterinarians in the difficult process of providing the best medical treatment for their patients. The results of our survey showed that formularies, and especially one specific formulary, are the information sources most commonly consulted for deciding drug dosages in exotic animals. We also found that there are limitations in the way these formularies are consulted, in that many veterinarians do not routinely verify the sources on which the dosage in the formularies was based. Tailored strategies to educate veterinarians on the importance of primary sources, in light of potential limitations of currently available formularies and textbooks, are recommended.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org
Acknowledgments
No external funding was used in this study. The authors declare that there were no conflicts of interest.
References
- 1. ↑
Bhatti SF, De Risio L, Muñana K, et al. International Veterinary Epilepsy Task Force consensus proposal: medical treatment of canine epilepsy in Europe. BMC Vet Res. 2015;11:176.doi: 10.1186/s12917-015-0464-z
- 2.
Shaughnessy AF. STEPS drug updates. Am Fam Physician. 2003;68:2342–2348.
- 3. ↑
de Vries TP, Henning RH, Hogerzeil HV, Fresle DA. Guide to good prescribing. A practical manual. World Health Organization Action Programme on Essential Drugs. Accessed January 12, 2021. whqlibdoc.who.int/hq/1994/WHO_DAP_94.11.pdf
- 4. ↑
Visser M, Oster SC. The educated guess: determining drug doses in exotic animals using evidence-based medicine. Vet Clin North Am Exot Anim Pract. 2018;21(2):183–194. doi:10.1016/j.cvex.2018.01.002
- 5. ↑
Everitt S, Pilnick A, Waring J, Cobb M. The structure of the small animal consultation. J Small Anim Pract. 2013;54(9):453–458.doi: 10.1111/jsap.12115
- 6. ↑
Corah L, Lambert A, Cobb K, Mossop L. Appointment scheduling and cost in first opinion small animal practice. Heliyon. 2019;5:e02567.doi: 10.1016/j.heliyon.2019.e02567
- 7. ↑
Belshaw Z, Robinson NJ, Dean RS, Brennan ML. “I always feel like I have to rush…” Pet owner and small animal veterinary surgeons’ reflections on time during preventative healthcare consultations in the United Kingdom. Vet Sci. 2018;5:20.doi: 10.3390/vetsci5010020
- 8. ↑
Pan SD, Zhu LL, Chen M, Xia P, Zhou Q. Weight-based dosing in medication use: what should we know? Patient Prefer Adherence. 2016;10:549–560. doi:10.2147/PPA.S103156
- 10. ↑
Greenwell K, Hanna D, Womble W, Di Girolamo N. A twenty-year trend in references used for avian drug dosages in an exotic animal formulary. Presented at: Exoticscon 2019; St Louis; 340.
- 11.
Caron M, Smead C, Brandão J, Di Girolamo N. A twenty-year trend in type of references used for reptile drug dosages in an exotic animal formulary. Presented at: Exoticscon 2019; St Louis; 564.
- 12. ↑
Kline S, Tarpalechee L, Brandão J, Di Girolamo N. Twenty-year trend in type of references used for small mammal drug dosages in an exotic animal formulary. Presented at: Exoticscon 2019; St Louis; 412.
- 14. ↑
CAB direct. Centre for Agriculture and Bioscience International. Accessed May 25, 2021. https://www.cabdirect.org
- 16. ↑
Eysenbach G. Improving the quality of web surveys: the Checklist for Reporting Results of Internet E-Surveys (CHERRIES). J Med Internet Res. 2004;6(3):e34.doi: 10.2196/jmir.6.3.e34
- 20. ↑
Scopus. Elsevier. Accessed August 8, 2019. https://www.scopus.com/search/form.uri?display=basic#basic
- 21. ↑
Exotic DVM. Veterinary Forum for Companion Exotic Animal Medicine. HBD International, Inc. Accessed September 27, 2019. https://groups.io/g/ExoticDVM
- 22. ↑
Italian Society of Veterinarians for Exotic Animals [SIVAE] forum. Società Italiana Veterinaria per Animali Esotici. Accessed September 24, 2019. http://groups.google.com/a/anmvi.vet/group/sivae-forum?hl=it
- 23. ↑
Peduzzi P, Concato J, Kemper E, et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49:1373–1379.
- 25. ↑
Divers SJ, Stahl SJ, eds. Mader’s Reptile and Amphibian Medicine and Surgery. Elsevier Health Sciences; 2018.
- 26. ↑
Miller ER, Fowler ME, eds. Fowler’s Zoo and Wild Animal Medicine, Volume 8. Elsevier Health Sciences; 2014.
- 27. ↑
Quesenberry K, Mans C, Orcutt C. Ferrets, Rabbits and Rodents: Clinical Medicine and Surgery. Elsevier Health Sciences; 2020.
- 30. ↑
Pelzer NL, Leysen JM. Use of information resources by veterinary practitioners. Bull Med Libr Assoc. 1991;79:10–16.
- 31.
Wales T. Practice makes perfect? Vets’ information seeking behaviour and information use explored. Aslib Proc. 2000;52:235–246.
- 32. ↑
Nielsen TD, Dean RS, Massey A, Brennan ML. Survey of the UK veterinary profession 2: sources of information used by veterinarians. Vet Rec. 2015;177(7):172.doi: 10.1136/vr.103068
- 33. ↑
Huntley SJ, Dean RS, Massey A, Brennan ML. International evidence-based medicine survey of the veterinary profession: information sources used by veterinarians. PLoS One. 2016;11:e0159732.doi: 10.1371/journal.pone.0159732
- 34. ↑
Delgado-Rodríguez M, Llorca J. Bias. J Epidemiol Community Health. 2004;58(8):635–641.doi: 10.1136/jech.2003.008466
- 35. ↑
Sackett DL. Bias in analytic research. J Chronic Dis. 1979;32(1-2):51–63.doi: 10.1016/0021-9681(79)90012-2
- 36. ↑
Giuffrida MA. Type II error and statistical power in reports of small animal clinical trials. J Am Vet Med Assoc. 2014;244(9):1075–1080.doi: 10.2460/javma.244.9.1075
- 37.
Beaufrère H, Kearney MT, Tully TN Jr. Can we trust the avian medical literature: survey and critical appraisal of the use of statistics in avian medicine from 2007 to 2011. J Exot Pet Med. 2015;24(4):415–426.doi: 10.1053/j.jepm.2015.08.005
- 38.
Sargeant JM, Elgie R, Valcour J, et al. Methodological quality and completeness of reporting in clinical trials conducted in livestock species. Prev Vet Med. 2009;91(2-4):107–115. doi:10.1016/j.prevetmed.2009.06.002
- 39. ↑
Di Girolamo N, Meursinge Reynders R. Deficiencies of effectiveness of intervention studies in veterinary medicine: a cross-sectional survey of ten leading veterinary and medical journals. PeerJ. 2016;4:e1649. doi:10.7717/peerj.1649
- 40. ↑
Page JR. Accessibility of published research to practicing veterinarians. J Med Libr Assoc. 2018;106(3):330–339. doi:10.5195/jmla.2018.196
- 41. ↑
Maranville RE, Popken A, Meursinge Reynders R, Brandão J, Di Girolamo N. Reporting quality of abstracts of veterinary randomized controlled trials. J Am Vet Med Assoc. 2021;258(3):303–309.doi: 10.2460/javma.258.3.303
- 42. ↑
Robinson NJ, Brennan ML, Cobb M, Dean RS. Capturing the complexity of first opinion small animal consultations using direct observation. Vet Rec. 2015;176(2):48.doi: 10.1136/vr.102548
- 43. ↑
Belshaw Z, Asher L, Dean RS. The attitudes of owners and veterinary professionals in the United Kingdom to the risk of adverse events associated with using non-steroidal anti-inflammatory drugs (NSAIDs) to treat dogs with osteoarthritis. Prev Vet Med. 2016;131:121–126.doi: 10.1016/j.prevetmed.2016.07.017
- 44. ↑
Gochenauer AE, Holmes ER, Barber KE, Forsythe LR. The current landscape of veterinary compounding in the pharmacy setting. Int J Pharm Compd. 2019;23(5):422–427.
- 45. ↑
Adrian DE, Rishniw M, Scherk M, Lascelles BDX. Prescribing practices of veterinarians in the treatment of chronic musculoskeletal pain in cats. J Feline Med Surg. 2019;21(6):495–506.doi: 10.1177/1098612X18787910
- 46. ↑
Wiant K, Geisen E, Creel D, et al. Risks and rewards of using prepaid vs. postpaid incentive checks on a survey of physicians. BMC Med Res Methodol. 2018;18(1):104.doi: 10.1186/s12874-018-0565-z
- 47. ↑
Meade AW, Craig SB. Identifying careless responses in survey data. Psychol Methods. 2012;17(3):437–455.doi: 10.1037/a0028085
