Caregiver burden, treatment complexity, and the veterinarian-client relationship in owners of dog with skin disease.

BACKGROUND: Increasing complexity of treatment plans is associated with higher levels of caregiver burden in owners of dogs with skin disease. It is possible that elevated caregiver burden resulting from treatment complexity could, in turn, affect the veterinarian-client relationship. HYPOTHESES/
OBJECTIVES: We expected that treatment complexity, caregiver burden, and the client's perception of the veterinarian-client relationship would be related to each other. We also expected an indirect effect of caregiver burden on the cross-sectional association between treatment complexity and the veterinarian-client relationship, and that this effect would be robust to adjustment for the dog's skin disease course and severity. PARTICIPANTS: Participants were 349 owners of dogs with skin disease recruited through online consumer panels.
MATERIALS AND METHODS: Cross-sectional online assessments were completed for caregiver burden, treatment plan complexity, veterinarian-client relationship, and skin disease course and severity. Demographic information also was collected.
RESULTS: The indirect effect of caregiver burden on the relationship between treatment complexity and veterinarian-client relationship was statistically significant, accounting for 42.76% of the variance in the model. After controlling for disease severity and course, that effect remained statistically significant, accounting for 37.76% of the variance. CONCLUSIONS AND CLINICAL IMPORTANCE: Findings support the notion that greater treatment complexity is related to the owner's perception of the veterinarian-client relationship via caregiver burden. Efforts to reduce caregiver burden by using the simplest effective treatment may benefit the veterinarian-client relationship.

Introduction

Chronic disease in a companion animal is associated with caregiver burden, or strain from the challenges of providing care, for the owner. , This burden is linked with negative owner outcomes such as depressive symptoms and lower quality of life, , and predicts owner decisions about the animal, including consideration of euthanasia. Beyond impact to the owner and the animal, burden also may affect the relationship between the veterinarian and owner. Caregiver burden in the owner has been posited to underlie stressful encounters with the veterinary provider through a “burden transfer,“ , which could undermine the veterinarian–client relationship. Numerous studies have demonstrated the value of good rapport between the veterinarian and client, such as adherence and satisfaction, , making this a crucial issue to consider in successful veterinary practice.

Owners of dogs with skin disease show elevated caregiver burden compared to owners of healthy control dogs, and that burden is significantly related to the complexity of the treatment plan. Although differences in burden between owners of dogs with skin disease and healthy controls are reduced when disease severity is decreased, controlling for severity does not eliminate the significance of the relationship between burden and treatment complexity. The more complicated the treatment regimen, the greater burden the owner feels, even after accounting for severity of the dog’s problems. Simplicity and effectiveness of treatment thus both appear to be key influencers of caregiver burden. This is of critical concern in treating skin disease, where treatments range in complexity from periodic injection to multimodal treatment involving combinations of oral, injectable and topical medications.

In order to effectively partner with an owner toward sustainable management of skin disease, associations between caregiver burden, treatment complexity and the veterinarian–client relationship are important to understand. Do complicated or difficult treatments lead to feelings of caregiver burden, in turn affecting how the client feels about the veterinarian? The goal of the current study was to explore these relationships in a sample of owners of dogs with skin disease through a cross‐sectional mediation model. We hypothesised that significant relationships would be detected among three primary variables: treatment complexity, caregiver burden and the client’s perception of the veterinarian–client relationship. Furthermore, we expected that the link between treatment complexity and the client’s perception of their relationship with the veterinarian would be cross‐sectionally mediated by caregiver burden.

Materials and methods

Ethics

This study was approved by the local Institutional Review Board (Kent State University IRB #21‐177). Informed consent was obtained electronically from all participants before completion of study activities.

Procedure

This study was undertaken using STROBE criteria for cross‐sectional studies. Data were collected between 23 April and 13 June 2021 through Qualtrics consumer panels targeting pet owners. Following informed consent, the protocol opened. Participants were reimbursed in accordance with an agreed‐upon amount and pay‐out option from Qualtrics (e.g. airline miles, gift card for a retail outlet). No identifying information was obtained from the dog owners, ensuring participant confidentiality.

Participants

Participants were required to be English‐speaking, 18 years of age or older, and reporting on a dog that is routinely monitored for health by a veterinarian and which currently is experiencing “excessive itching (i.e. scratching, licking, chewing, or biting at self).” Participants also were required to provide complete data that passed attention checks including expected completion speed (>231 s) and response consistency. From the initial reach (n = 6537), a total of 526 met these criteria. Due to the focus of the study on treatment complexity and the veterinarian–client relationship, several further exclusion criteria then were applied. Participants who reported a diagnosis that did not fit the study’s chronic dermatological disease focus (e.g. “anxiety,” “ear infection;” n = 4), who reported that their dog was not currently receiving treatment specifically for itch (n = 64), or that, although their dog’s health is overseen by a veterinarian, any treatment for itch in particular was not being overseen by a veterinarian (n = 52) were removed from the sample. To ensure data quality, a veterinarian on the research team (MDC) then reviewed treatment descriptions for participants listing treatment that appeared improbable or inappropriate for the stated condition, removing an additional 57 participants. The final analytic sample was 349.

Measures

Caregiver burden

The 18 item, previously validated Zarit Burden Interview (ZBI), adapted for companion animal use measured caregiver burden. Questions on this measure reflect emotional (e.g. feelings of guilt, anger, frustration) and instrumental (e.g. physical health, social life, financial impact) burden from caregiving. Likert‐type responses (0, never to 4, nearly always) are summed for a total score. Higher values indicate a greater degree of burden, with a total score of ≥18 indicating meaningful caregiver burden. , The sample internal consistency was α = 0.91.

Treatment complexity

Subjective and objective treatment plan complexity were measured as described previously using treatment plan‐related items from the Pet Owner Adherence Scale (POAS) and total number of individual treatments used to manage their dog’s skin condition each week. Specifically, subjective treatment complexity was assessed using the POAS, which is a measure of owner perception of a companion animal’s disease and treatment. Although designed to examine adherence, this scale addresses several issues relevant to treatment needs. Items related to the treatment plan (e.g. how difficult it is to follow, how well it has been explained; n = 7) were utilised. A Likert‐type response format was used (1, strongly disagree to 5, strongly agree), with reverse scoring as indicated. Items then were summed for the subjective treatment complexity score. The internal consistency reliability was α = 0.67 in the current sample. Objective treatment complexity was assessed by asking participants to indicate weekly frequency of individual treatments used to manage their dog’s skin condition. Frequency for various treatment modalities was asked about in a manner facilitating owner response (i.e. oral medications per day, quick topical treatments per day, time‐intensive topical treatments per week, injections per year). Weekly treatment frequency for each variable then was calculated (e.g. oral medications per day multiplied by seven, injections per year divided by 52) and these numbers were summed for a total objective treatment complexity score. To create a single variable reflecting both subjective and objective aspects of treatment complexity, these scores were normalised into t‐scores, a standardized score calculated in order to consider both measures of treatment complexity in the analyses [mean 50, standard deviation (SD) 10] and added together for an “Overall treatment complexity” variable.

Veterinarian–client relationship

The previously validated Context and Relationship subscale of the revised Patient Perception of Patient‐Centeredness (PPPC‐R) questionnaire was used to assess the client’s perception of their relationship with the veterinarian. This scale was chosen as a result of its focus on perception that a provider is caring, trusted, compassionate, and considerate of the rater’s thoughts and feelings. As suggested by measure creators, the specific provider type (in this case “veterinarian”) was used in place of the general term “provider.” Because the PPPC‐R was developed for use in human medicine, an item specifically related to caring about the companion animal was added (“How much would you say this person cares about your dog?”), using the same Likert‐type response format (1, very much/completely to 4, not at all). Item scores were summed for a total subscale score, with a higher number indicating poorer client perception of the veterinarian–client relationship. The sample internal consistency with this added item was α = 0.87 in the current sample.

Skin disease severity

Skin disease severity was rated by owners using a measure validated previously. Responses to the single item disease severity question, “How severe and disturbing has your dog’s skin disease been?” ranged from 0, not at all to 3, very much.

Skin disease course

Skin disease course was rated by owners using a single item question with multiple choice format, “Since your dog’s symptoms started, they have overall: Improved, Declined, Been stable – No improvement, but no decline.” For the purpose of the current study, responses were condensed into either “Improved” or “Not improved” (i.e. declined or stable) in order to create a binary variable.

Demographic information

Participants reported their age (continuous, with free text response format), gender (multiple choice response format: Male, Female, Other), education level [multiple choice response format: Less than high school, High school graduate, Associate’s degree or equivalent, Bachelor’s degree or equivalent, Advanced degree (e.g. Master’s degree, Doctorate)], race/ethnicity (multiple choice response format: African‐American or Black, Asian American or Asian, Caucasian or White, Latin American or Hispanic, Native American or Indigenous, Other), annual income (multiple choice response format: $125,000 per year). A “Prefer not to say” option was provided for all multiple choice response formats.

Participants also provided information regarding their dog and its disease, including age (continuous, with free text response format), sex (multiple choice response format: Male/neutered/ Male/intact, Female/spayed, Female/intact), diagnosis (multiple choice response format with the option to endorse more than one diagnosis: Allergic dermatitis – undetermined/unknown, Atopic dermatitis – environmental allergies, Flea allergy, Food allergy, Contact allergy, No formal diagnosis, Other), duration of disease (continuous, with free text response format), and type of provider overseeing the dog’s treatment for itch [General (practitioner) or Dermatologist (referral)].

Statistical analyses

Descriptive statistics first were conducted for demographic and primary variables; primary variables were deemed acceptable for parametric analyses. To examine the hypothesis that significant relationships would exist across the three primary variables (caregiver burden, treatment complexity and the veterinarian–client relationship), Pearson’s correlations were conducted. Disease course and severity were considered to be potential confounding factors that might influence primary variables of interest; correlations of these with primary variables also were thus examined so that further analyses could be adjusted in the event of significant relationships.

In order to examine the hypothesis that owner caregiver burden would mediate the cross‐sectional relationship between treatment plan complexity and owner perception of the veterinarian–client relationship, the Hayes PROCESS macro was employed. Two models were run: first, the three primary variables were included in the model with treatment plan complexity entered as the independent variable, the veterinarian–client relationship entered as the dependent variable, and caregiver burden entered as the mediating variable. Given the significance of the associations among disease course and severity with primary variables, a second model adjusting for these potential confounders also was run. All analyses were conducted in Spss 27.

Results

See Table 1 for sample characteristics and Table 2 for descriptive statistics of primary variables. Average ZBI scores (M = 11.96) did not reach the threshold for clinically meaningful caregiver burden in this sample. Mean subjective treatment complexity (total POAS raw score) was 13.08 (SD = 3.59), while average objective treatment complexity (total weekly skin disease treatments) was 9.06 (SD = 5.85). Most dogs were described as having skin disease that was “not at all” (42.1%) or “a little” (47.3%) severe or disturbing, while smaller proportions were described as having skin disease that was “quite a bit” (9.2%) or “very much” (1.4%) severe or disturbing.

Table 1

Sample demographic characteristics

	Total sample n = 349
Owner variables
Gender (n, %female)	192 (55.0)
Age (M/SD)	47.25/16.59
Education (n, %)
<High school	5 (1.4)
High school	119 (34.1)
College	180 (51.6)
Advanced degree	44 (12.6)
Declined to answer	1 (0.3)
Race/ethnicity (n, %)
Caucasian or White	278 (79.7)
African American or Black	23 (6.6)
Asian American or Asian	13 (3.7)
Latin American or Hispanic	21 (6.0)
Native American or Indigenous	7 (2.0)
Other	3 (0.9)
Declined to answer	4 (1.1)
Annual Income (n, %)
<US$25,000	40 (11.5)
US$25–74,999	190 (54.4)
>US$75,000	110 (31.5)
Declined to answer	9 (2.6)
Animal and disease variables
Sex (n, %)
Female, spayed	145 (41.5)
Female, intact	40 (11.5)
Male, neutered	150 (43.0)
Male, intact	12 (3.5)
Age (M/SD)	6.83/3.73
Diagnosis (n, %)
Allergic dermatitis (undetermined/unknown)	117 (33.5)
Atopic dermatitis (environmental allergies)	19 (22.1)
Flea allergy	87 (24.9)
Food allergy	44 (12.6)
Contact allergy	23 (6.6)
Other	6 (1.7)
No formal diagnosis	67 (19.2)
Types of treatment used (n, %)
Topical	208 (59.6)
Oral	197 (56.4)
Injection	30 (8.6)
Ear cleaner	66 (18.9)
Other	8 (2.3)
Duration of disease – years (M/SD)	2.35/2.48
Veterinarian type overseeing treatment (n, %)
General practitioner	318 (91.1)
Dermatological specialist	31 (8.9)

Diagnosis category adds to >100% owing to the option to endorse more than one diagnosis as appropriate.

Table 2

Descriptive statistics for primary measures

	Total sample n = 349
Caregiver burden (M/SD; min–max)
Zarit Burden Interview, adapted	11.96/9.87; 0–53
Veterinarian–client relationship (M/SD; min–max)
Patient Perception of Patient‐Centeredness, revised (PPPC‐r):	15.77/4.68; 9–32
Context and relationship subscale
Treatment complexity (M/SD; min–max)
Subjective plus objective treatment complexity	101.04/14.58; 70.89–153.20
Disease severity (M/SD; min–max)
Noli skin disease severity	0.70/0.69; 0–3
Disease course (n, %)
Improved	183 (52.4)
Not improved (i.e. stable or declined)	166 (47.6)

M, mean; SD, standard deviation; min–max, minimum to maximum.

Correlations among the three primary variables (treatment complexity, caregiver burden, and the client’s perception of the veterinarian–client relationship) and two potential confounding variables (disease course and disease severity) are shown in Table 3. Significant relationships were observed among all three primary variables (P < 0.001 for all). Several significant relationships also emerged with potential confounding factors of disease course and severity, suggesting that adjusting for these variables would be of benefit in understanding results of the subsequent mediation analysis.

Table 3

Correlations among primary and potentially confounding variables

	r
	1	2	3	4
1. Caregiver burden	__
2. Veterinarian–client relationship	0.29**	__
3. Treatment complexity	0.44**	0.24**	__
4. Disease severity	0.40**	0.09	0.23**	__
5. Disease course	0.23**	0.18**	0.16*	0.15*

*, p < 0.01, ** indicates p < 0.001.

Cross‐sectional mediation analysis demonstrated that the indirect effect of caregiver burden on the relationship between treatment complexity and the client‐veterinarian relationship was statistically significant [B = 0.03, BC 95%CI(0.01, 0.05)], accounting for 42.76% of the variance in the model (see Table 4 and Figure 1). After controlling for disease severity and course, the indirect effect remained statistically significant [B = 0.03, BC 95%CI(0.01, 0.04)], accounting for 37.76% of the variance in the model (see Table 4 and Figure 2).

Table 4

Analysis of the indirect effect of caregiver burden on the relationship between treatment complexity and client perception of the veterinarian–client relationship

	Model 1 Coefficient (95%CI)	Model 2 Coefficient (95%CI)
Association:	0.30	0.24
treatment complexity × caregiver burden	(0.23, 0.36)	(0.17, 0.16)
Association: caregiver burden × veterinarian–client relationship	0.11 (0.06, 0.16)	0.11 (0.05, 0.16)
Association:	0.04	0.04
treatment complexity × veterinarian–client relationship	(0.01, 0.08)	(0.01, 0.08)
Indirect effect	0.03 (0.01, 0.05)	0.03 (0.01, 0.04)
Variance explained by indirect effect (%)	42.76%	37.76%

Model 1, unadjusted; Model 2, adjusted for disease severity and course. CI, confidence interval.

Figure 1

Coefficients for the indirect effect of caregiver burden on the association between treatment complexity and client perception of the veterinarian–client relationship (Model 1).

Figure 2

Coefficients for the indirect effect of caregiver burden on the association between treatment complexity and client perception of the veterinarian–client relationship, adjusted for disease course and severity (Model 2).

Discussion

This study explored relationships among caregiver burden, treatment complexity, and the veterinarian–client relationship in a sample of owners of dogs with skin disease. Highly significant relationships among all three variables were demonstrated. Moreover, the association between complexity of the treatment plan and veterinarian–client relationship was mediated by caregiver burden in the owner, even after adjusting for disease course and severity in this sample. Several aspects of the findings warrant further discussion.

The associations observed in this study support the idea that more complicated treatment is related to poorer perception of the veterinarian–client relationship by way of caregiver burden. Although the study design does not definitively address causality, it is likely that caregiver burden is amplified by greater treatment complexity, which in turn reduces the client’s perception of their rapport with the veterinarian. The importance of current findings is underscored by a recent focus on client‐centred approaches in veterinary medicine, which include efforts to develop rapport and build an authentic partnership with the client. ,

Of note, sustained statistical significance after controlling for the course and severity of skin disease in these relationships shows that these links are present independent of the level of the dog’s problems or any potential client dissatisfaction with outcomes. Both course and severity of skin disease did contribute to the relationships observed; however, their contributions were relatively small. This is useful knowledge for the veterinarian: it suggests that for most clients, a complicated treatment plan could do more to damage trust in the veterinarian than having an imperfect outcome. When treatment is difficult and caregiver burden ensues, rapport with the veterinarian is affected.

Findings thus highlight the importance of treatment that is simple. Dermatological presentations will sometimes necessitate complicated treatment planning, and starting with the simplest possible treatment and increasing complexity only as necessary may help keep owner burden to a minimum. Implications extend beyond the burden experienced by the client. The indirect relationship observed between treatment complexity and the veterinarian–client relationship suggests that by decreasing the burden of complex treatment when possible, clients may view the veterinarian as more caring, compassionate, considerate of their thoughts and feelings, and trustworthy. Efforts to reduce burden for the client through use of the simplest effective treatment may thus facilitate the working relationship between veterinarian and client.

Some differences from past work emerged in the present study, including average burden that was below the clinically meaningful threshold (samples from prior work showed average burden above this threshold). , This difference is probably a reflection of recruitment methods. Whereas past work was conducted using samples recruited from a veterinary dermatological specialism clinic and a dog allergy social media group, the current study broadly targeted pet owners, then screened for owners of dogs experiencing itch, removing those who were not seeking current treatment from a veterinarian. The difference in recruitment methods is reflected in lower frequency of weekly treatments compared to past work and lower disease severity: approximately 90% of the current sample described their dog’s skin disease as “a little” or “not at all” disturbing, compared to 20–60% of the sample in the other studies. ,

Although recruitment methods led to a sample with lower overall caregiver burden, fewer treatments, and less severe disease relative to past work, this methodological difference represents a strength of the current study. The replication of previously demonstrated relationships (e.g. the link between caregiver burden and disease severity, as well as between caregiver burden and treatment plan complexity) in this sample of lower severity cases speaks to the robustness of these associations. Additionally, sampling methods allowed for greater representation of gender, income and race/ethnicity in the current sample relative to past studies, addressing a shortfall in previous work , and increasing the generalisability of study findings to the general population of dog owners.

Limitations of the current work include the use of caregiver report in determining presence of skin disease severity and course in their pet. The measure used in the current study was validated against clinician pruritus assessment, yet a clinician rating would be more objective. As prior work has suggested, however, adjusting for owner ratings of disease severity and course in this type of work may actually be helpful in removing variance attributable to owner reactivity. Additionally, because a majority of the sample reported that a general veterinarian was overseeing their dog’s skin disease treatment, findings may be more reflective of the general practice setting than dermatological specialist practice. Future work might benefit from side‐by‐side comparison of owner and clinician ratings, both general and specialist practice veterinary samples, as well as longitudinal mediation design to firmly establish directionality of the current findings.

Conclusions

Overall, results suggest that more complicated skin disease treatment appears to be related to poorer client perception of rapport with the veterinarian, by way of caregiver burden. Findings emphasise the importance of utilising the simplest effective treatment possible in an effort to minimise the client’s experience of burden, in turn enhancing the veterinarian–client relationship.

Author Contributions

Mary Beth Spitznagel: Study conceptualization and design, acquisition, analysis, and interpretation of data, drafting and critically revising the manuscript. Karlee Patrick: Analysis and interpretation of data, drafting and critically revising the manuscript. Andrew Hillier: Study conceptualization and design, critically revising the manuscript. Margaret Gober: Study conceptualization and design, critically revising the manuscript. Mark D. Carlson: Data acquisition and interpretation, critically revising the manuscript.