兽研1. 引言
狗和猫日益被视为伴侣和家庭成员，而非仅为实用目的而饲养的动物[1,2]。这一转变提高了人们对所提供的照护质量和动物福利的期望。然而，将动物描述为伴侣本身并不能确保其所有需求得到满足，且主人对责任的理解和履行方式各不相同[3,4]。因此，负责任的所有权应通过具体的照护和管理实践来考察，而非仅从所有权或情感本身推断。
在本研究中，负责任的宠物饲养是指主人在动物整个生命周期中为满足其需求并履行对动物及更广泛社区的责任而采取的实践和决策。这些包括负责任的获取、植入微芯片和登记、预防性医疗保健、繁殖管理、适当的营养和饲养、控制无人看管的游荡以及及时的兽医护理[3,5,6]。这些实践支持个体动物的健康与福利、主人的问责制以及伴侣动物种群管理[6,7]。兽医的参与还提供了早期疾病检测以及营养、繁殖和行为方面的专业指导机会[8,9]。然而，一些主人在动物看似健康或未感知到迫切需求时不寻求兽医护理[10]，导致照护方式更偏向反应性而非预防性。
主人如何履行这些责任可能影响动物福利的各个方面，而这些方面可能比疾病或伤害更不显而易见。当代动物福利方法认识到，福利受营养、身体健康、生活环境、表现正常行为的机会以及动物由此产生的心理状态所影响[11]。主人可能低估狗和猫的体况[12]，而跨伴侣动物物种的研究已发现主人对行为和环境需求的理解以及对应激相关体征的识别存在差异[8,13,14]。因此，动物可能在获得基本照护的同时，仍经历未满足的营养、行为、环境或心理福利需求。
这些责任的履行程度可能因社会和地理环境而异。农村社区在此背景下尤为相关，因为伴侣动物的饲养模式、饲养实践和预防服务的可及性可能与城市地区不同。在智利，农村家庭更可能饲养伴侣动物，且人均动物数量更多，而绝育和微芯片植入率低于城市地区[6]。在美国四个服务不足的社区中，农村社区的整体宠物饲养率和养狗率高于城市社区[15]。在英国，估计的狗和猫密度在大都市区最高，但每户宠物数量通常在农村地区最高[16]。一项全国住户调查同样发现，狗更可能由农村家庭饲养，猫更可能由半城市或农村家庭饲养[17]。丹麦的研究也表明，农村地区猫的微芯片植入率仍低于人口更密集的地区[18]。当每户饲养更多动物时，不一致的微芯片植入、预防性护理和繁殖控制可能影响更多动物。然而，农村居住本身不应被解释为照护较差的证据。城乡差异可能反映的是动物数量、饲养传统、社会经济状况、服务可及性以及主人对动物需求的理解等因素的综合作用，而非仅取决于居住地。
在立陶宛，伴侣动物福利和所有权受1997年通过的《动物福利和保护法》管辖[19]，该法规定了一般福利要求，禁止忽视和虐待，并自2021年修正案起强制要求所有狗和猫植入微芯片并登记。强制性疾病控制和狂犬病疫苗接种由1991年通过的《兽医法》[20]、2007年通过的关于狂犬病疫苗接种要求的政府决议[21]以及疾病控制义务[22]加以规范。在立陶宛，约三分之一的人口居住在农村地区[23]。尽管由于全国性普查尚未完全实施，狗和猫的总数未知，但截至2026年1月1日，国家伴侣动物登记处记录了345,085只狗和188,971只猫，其中约65%的登记伴侣动物位于三大城市市镇以外的地区[24]。考虑到其他研究报告的农村地区微芯片植入率较低[6,18]，居住在农村地区的伴侣动物数量可能远高于官方记录所示，这使得该群体对动物福利和种群管理研究尤为重要。
既往立陶宛研究已探讨了影响兽医服务选择及服务提供特定方面的因素，包括兽医远程医疗[25,26]。然而，关于伴侣动物数量、获取途径、微芯片植入、预防保健、生殖控制、饲养管理实践及兽医诊疗利用方面的城乡差异，目前知之甚少。对国际犬猫福利文献的文献计量分析显示，英国、意大利、西班牙、德国和奥地利是欧洲的主要贡献者，提示波罗的海及东欧地区的研究仍代表性不足[1]。据我们所知，此前尚无立陶宛研究将饲主实践的城乡比较与对农村家庭犬猫的直接、结构化福利评估相结合。
为填补上述空白，本研究将关于饲主、家庭及所养犬猫（以下统称伴侣动物）的问卷数据与直接动物福利评估相结合。本研究旨在比较立陶宛农村与城市家庭在伴侣动物数量、特定饲养与照护实践、饲养管理实践及兽医诊疗利用方面的差异，描述总体样本中伴侣动物的获取途径，并使用改编的动物福利评估网格（AWAG）识别农村家庭中受评估犬猫的主要福利问题。
我们假设农村受访者报告的伴侣动物饲养数量将多于城市受访者，且在若干负责任饲养实践的执行上一致性较低，同时总体福利评分将随饲主社会人口学特征而变化。
2. 材料与方法
2.1. 研究设计与场所
本研究于2025年9月至2026年1月在立陶宛开展，包括匿名在线问卷和面对面饲主问卷结合直接动物福利评估两部分，采用便利抽样策略。
符合条件的参与者为至少饲养一只犬或猫的成年人。本研究主要聚焦农村社区。在线问卷亦纳入了通过公开渠道获取调查链接的城市饲主，而面对面调查部分及所有直接动物福利评估均仅在农村居民中开展。农村地区依据经济合作与发展组织（OECD）标准定义为人口密度低于150人/km²的地区[27]。受访者被要求提供居住地名称，随后依据官方人口密度数据将其归类为农村或城市。
2.2. 伦理批准与知情同意
本研究（包括问卷及对私人所有犬猫的非侵入性评估）经立陶宛健康科学大学生物伦理中心批准（批准号2025-BEC2-0990，2025年9月26日）。参与为自愿性质，所有参与者均被告知研究目的、结果在科学出版物中的预期用途、退出权利及数据保护措施。
未收集可直接识别个人身份的信息或确切居住地址。在访问并填写在线问卷之前，所有受访者均须阅读有关研究目标和个人数据处理的信息。继续进入并完成问卷即视为知情同意的确认。
在进行面对面问卷和动物评估之前，参与者亦同意参与并允许对其动物进行非侵入性检查。评估仅涉及外部观察和检查，不包括疼痛性或侵入性操作。所有与参与者和动物相关的信息在分析前均已匿名化处理。
2.3. 问卷开发与预试验
根据研究目标和相关文献，在具有兽医流行病学、动物福利和伴侣动物行为学专长的研究人员指导下，以立陶宛语开发了研究专用饲主问卷。问卷初稿在主要数据收集前以少量犬猫饲主样本进行了预试验，以评估问题的清晰度和选项的适宜性，并据此修订直至形成最终版本。在线调查和面对面调查使用相同的核心问卷，家庭访视期间增加了额外问题并纳入动物福利评估。作为补充材料S1和S2提供的英文版本仅为出版准备，未用于数据收集。
2.4. 在线问卷
匿名在线问卷使用Google Forms创建，于2025年11月至2026年1月期间通过当地Facebook群组分发。由于邀请通过公开社交媒体帖子传播，无法确定查看或收到邀请的人数，亦无法计算正式应答率。完成问卷约需10–15分钟。
问卷包含封闭式、多项选择和开放式问题，分为四个主要部分。第一部分收集饲主的人口学信息，包括年龄、性别、教育程度、就业状况和居住地。第二部分收集家庭饲养动物的信息，包括犬和猫的数量以及动物的获取途径。第三部分涉及负责任养宠和预防保健实践，包括微芯片植入、狂犬病疫苗接种、寄生虫预防、绝育、繁殖、饲喂和饲养。第四部分涉及兽医诊疗利用、对疾病或损伤的应对、就诊原因、就诊频率、感知到的诊疗障碍以及有关动物健康、营养、护理和福利的信息来源。
问卷的英文版本作为补充材料S1提供。
2.5. 饲主面对面调查
面对面部分采用辅助式问卷。研究者将参与者的回答录入Google Forms，其中包含与在线调查相同的核心问题，并补充了动物福利评估所需的问题和观察内容。
在三次伴侣动物微芯片植入和绝育活动期间，以及随后在饲主请求兽医帮助的个别病例中，接触了潜在参与者。通过这些活动接触到的所有符合条件的成年犬或猫饲主均被邀请参与，不论饲主或动物特征如何，未应用额外的选择参数。参与完全自愿，决定参与、拒绝参与或退出均不影响其获得或接受这些活动或收容所提供的任何服务或援助。仅纳入同时同意填写问卷并接受至少一只动物检查的成年人。
每位参与饲主被分配一个匿名识别号。使用个体动物识别号将饲主问卷与相关福利评估相关联，同时保持匿名。饲主访谈和动物评估合并进行通常需要15–20分钟。
问卷的英文版本作为补充材料S2提供。
2.6. 动物福利评估
动物福利采用改编版动物福利评估网格（AWAG）进行评估，该工具此前已在犬中进行了初步验证[28]。在该工具中，身体、心理、环境和程序性福利参数内的各项因素按1分（代表最有利的福利状态）至10分（代表最不利的福利状态）评分。计算每个参数的平均因素得分，并用于得出基于多边形的累积福利评估分数。
在本研究中，该方法经改编用于在单一时间点对犬和猫进行一次性评估。两种物种均保留基本的1–10评分结构和严重程度阈值，但在解释描述符时使用了物种特异性的行为表达和需求。书面修改包括在攻击性指标中纳入猫的表达（如嘶嘶声和抓挠）以及犬的咆哮和咬击。在结构化访谈期间，评估者还解释了应激的物种适宜表现形式以及物种特异性的环境和行为需求，包括猫的躲藏、抓挠、探索、狩猎相关行为和丰富化。根据指标不同，评分或由饲主在此解释后选择，或由评估者根据直接观察和饲主提供的信息进行评定。这些物种特异性解释基于评估者的正规兽医教育以及在评估和管理犬猫行为与福利需求方面的丰富实践经验。本研究中使用的改编评估表作为补充材料S3提供。
所有评估均由同一名兽医学学生使用改编评估表中的预定义描述符进行。评估了十三项福利指标：活动能力；体况；临床评估；进食和饮水；对饲养者的攻击性；对陌生人的攻击性；恐惧和焦虑频率；对 stressors 的反应；选择、控制和可预测性；丰富化；社会互动；评估期间的行为；以及评估期间的保定。
原始AWAG中包含的两个因素——日常作息改变和操作疼痛——被排除，因为它们旨在评估兽医或管理操作的影响，不适用于本研究非侵入性、单次评估的设计。
13项指标中的每一项均使用改编评估表中相应的言语描述符赋予数值评分。在本分析中，研究特异性总体福利分数计算为13项指标评分的算术平均值。该算术平均值用作研究特异性的描述性总结，不代表原始AWAG中使用的基于多边形的累积分数。分数越低表示福利越有利。
直接福利评估部分有意限定于农村家庭，因为其设计目的是确定研究目标农村人群的福利优先事项，并通过农村社区活动开展；该部分并非旨在提供动物个体层面福利的城乡比较。
2.7. 变量定义与数据处理
在线问卷和面对面问卷的应答结果被导出并合并为一个统一的分析数据库。两份问卷中完全相同的问题经过协调统一后合并分析。
饲主教育程度分为高等教育（包括大学和学院高等教育）或无高等教育（包括小学、初中、高中和职业教育）。就业状况分为就业或未就业。
每户动物总数通过将报告的犬和猫数量相加计算，并分为1只、2至5只、6至10只或11只及以上。在线问卷使用预定义类别（0、1、2–5、6–10或≥11）记录犬和猫的数量，而面对面问卷记录确切数量。在合并分析中，面对面问卷的确切数量被重新编码为相应类别。经协调统一后，家庭动物数量按1只、2至5只、6至10只或11只及以上进行分析。
家庭层面的微芯片植入、狂犬病疫苗接种和绝育应答被协调统一为三个类别：家中无动物实施、部分动物实施或全部动物实施。狂犬病疫苗接种指过去三年内的接种。
对于允许多选的题目，每个选项被转换为单独的二分类变量，未选择时编码为0，选择时编码为1。因此，获取途径、饲喂、饲养、兽医护理行为、障碍和信息来源的应答类别并非互斥。开放文本应答和录入“其他”项下的应答经过审核，并在适当情况下归入现有类别。无法归入预定义类别的应答通过识别反复出现的主题，按其内容进行分组。
使用口头描述词记录的动物福利应答根据改编的AWAG评估表转换为数值评分。饲主问卷数据和动物个体层面福利数据通过匿名识别编号进行关联。
2.8. 统计分析
统计分析使用Microsoft Excel（Microsoft Corporation，美国华盛顿州雷德蒙德）和JMP® 18（JMP Statistical Discovery LLC，美国北卡罗来纳州卡里）进行。分类变量计算频率和百分比，连续福利评分计算均值、中位数和范围。使用Pearson卡方检验评估分类变量之间的关联，包括城乡差异以及饲主人口学特征与家庭层面饲养实践结局之间的关联。考虑到同一家庭内动物福利评分之间的非独立性，福利评分比较使用以家庭为随机截距的线性混合模型（限制性最大似然法，REML）进行。固定效应为动物年龄（分类变量：<1岁、1–4岁、5–9岁和≥10岁；一个模型）和每项饲主人口学特征（教育程度、性别、年龄和就业状况；各一个模型）。统计显著性设定为p < 0.05。
2.9. 生成式人工智能的使用
在本手稿准备过程中，作者使用ChatGPT（OpenAI；GPT-5.5和GPT-5.6 Thinking；https://chatgpt.com/；访问日期为2026年7月30日）和Perplexity AI（Perplexity；https://www.perplexity.ai/；访问日期为2026年7月30日）协助手稿组织及各章节间的连贯性、语言润色、潜在相关文献的初步识别，以及期刊格式和出版要求合规性审查。所有生成输出均由作者进行批判性审阅、核实和编辑，作者对手稿最终内容承担全部责任。
3. 结果
3.1. 研究人群
共有383名猫和犬饲主参与本研究，其中330名受访者完成了在线问卷，53名参与了面对面调查。在面对面调查部分，共邀请57名符合条件的饲主参与；53名（93.0%）同意并完成了问卷，4名拒绝完成，因此未被纳入。总体而言，338名（88.3%）受访者居住在农村地区，45名（11.7%）居住在城镇地区。女性占受访者的89.8%，65.6%完成了高等教育。在面对面调查期间，对85只伴侣动物进行了福利评估，包括34只犬和51只猫。研究人群的人口学特征见表1。
Table 1. Survey and sociodemographic characteristics of respondents (N = 383). Values are presented as number and percentage of respondents, n (%).
| Characteristic | n (%) | |
|---|---|---|
| Survey type | Online survey | 330 (86.2) |
| In-person survey | 53 (13.8) | |
| Place of residence | Rural | 338 (88.3) |
| Urban | 45 (11.7) | |
| Gender | Female | 344 (89.8) |
| Male | 38 (9.9) | |
| Other | 1 (0.3) | |
| Age group | 18–24 | 42 (11.0) |
| 25–34 | 64 (16.7) | |
| 35–44 | 103 (26.9) | |
| 45–54 | 96 (25.1) | |
| 55–64 | 49 (12.8) | |
| ≥65 | 29 (7.6) | |
| Education | Higher education | 251 (65.6) |
| No higher education | 132 (34.4) | |
| Employment status | Employed | 305 (79.6) |
| Unemployed | 78 (20.4) | |
3.2. 宠物饲养特征
总体而言，48.6%的受访者饲养了2至5只伴侣动物，30.0%饲养了1只动物，15.7%饲养了6至10只动物，5.7%饲养了11只或更多动物。农村和城市家庭的伴侣动物数量分布存在显著差异（χ2 = 16.675，df = 3，p = 0.0008；图1）。近四分之三的农村受访者（73.1%）饲养了2只或更多伴侣动物，而城市受访者中这一比例为46.7%。相反，超过半数城市家庭（53.3%）仅饲养1只伴侣动物，而农村家庭中这一比例仅为26.9%。
[图：按居住地（农村与城市）划分的每户伴侣动物报告总数分布。Ba]
45岁及以上的受访者也倾向于比年轻受访者饲养更多的伴侣动物（χ2 = 9.977，df = 3，p = 0.0188）。就业状况也显示出类似趋势，失业受访者比就业受访者更常报告饲养较多伴侣动物，尽管这一关联未达到统计学显著性（χ2 = 7.754，df = 3，p = 0.0514）。
伴侣动物主要通过非正式来源获得。总体而言，69.3%的获取来源回答属于非正式途径，包括作为流浪动物收留的动物、熟人或其他家庭成员赠送、通过广告获得、已饲养动物所生、从未注册繁育者处购买或从私人处获得。相比之下，收容所领养占回答的15.3%，从注册繁育者处购买占14.1%，其他来源占1.4%。获取来源的分布见图2。
[图：报告的伴侣动物获取来源分布。百分比根据所有获取来源r计算]
3.3. 微芯片植入、疫苗接种和绝育实践
对农村和城市家庭的微芯片植入、狂犬病疫苗接种和绝育实践进行了比较（图3）。农村受访者中55.5%报告所有宠物均植入微芯片，城市受访者中为86.7%，而分别有32.8%和8.9%的家庭没有宠物植入微芯片（χ2 = 15.987，df = 2，p = 0.0003）。农村受访者中67.1%报告所有宠物在过去三年内完成狂犬病疫苗接种，城市受访者中为90.9%；分别有23.3%和4.5%的家庭没有宠物接种疫苗（χ2 = 10.742，df = 2，p = 0.0047）。农村受访者中50.6%报告所有宠物均绝育，城市受访者中为68.9%，而分别有31.7%和17.8%的家庭没有宠物绝育；这一差异未达到统计学显著性（χ2 = 5.518，df = 2，p = 0.0633）。两组中常规寄生虫预防均较常见，农村与城市之间无显著差异（χ2 = 0.363，df = 1，p = 0.5471）。
[图：按（A）居住地、（B）就业状况划分的选定预防性饲养实践分布，]
高等教育与完全微芯片植入（67.6% vs. 43.1%；χ2 = 23.179，df = 2，p < 0.0001）、狂犬病疫苗接种（78.2% vs. 53.5%；χ2 = 26.949，df = 2，p < 0.0001）和绝育（58.2% vs. 42.4%；χ2 = 16.617，df = 2，p = 0.0002）相关。就业也与完全微芯片植入（62.2% vs. 47.4%；χ2 = 6.034，df = 2，p = 0.0489）和狂犬病疫苗接种（74.1% vs. 52.7%；χ2 = 13.474，df = 2，p = 0.0012）相关。狂犬病疫苗接种覆盖率还因年龄而异（χ2 = 20.444，df = 10，p = 0.0253）；在65岁及以上的受访者中，48.1%报告所有宠物均已接种疫苗，而相同比例报告没有宠物接种疫苗。
报告的不完全微芯片植入常见原因包括认为没有必要、经济或操作困难、动物年龄、担心造成应激、难以捕捉动物、知识有限以及不愿意植入微芯片。不完全狂犬病疫苗接种的原因包括因动物的生活方式或年龄而认为没有必要、经济和操作困难、捕捉或运输动物的问题、知识有限、不信任疫苗以及推迟接种。未绝育的原因包括认为没有必要、动物年龄、繁育意图、经济或操作困难以及对可能的健康或行为影响的担忧。
总体而言，49.9%的受访者报告他们认为没有必要就诊兽医或更频繁就诊。
3.4. 饲养管理实践
饲养方式在不同居住环境之间呈现出描述性差异。由于受访者可以多选，所报告的类别并非互斥。室内饲养占城市饲养回答的93.6%和农村的63.7%。农村地区报告无限制户外活动的频率高于城市地区（20.6% vs. 4.3%），户外犬舍饲养也是如此（11.2% vs. 2.1%）。拴养占农村饲养回答的4.5%，城市受访者未报告此方式。
饲喂类别同样非互斥。城市家庭报告优质或超优质商业日粮的频率高于农村家庭（51.3% vs. 31.3%）。农村受访者更常报告超市宠物食品（19.5% vs. 10.3%）、餐桌残羹（9.9% vs. 3.8%）、骨头（9.4% vs. 6.4%）和粥（6.3% vs. 5.1%）。这些城乡比较为描述性，因为未对这些多响应变量进行推断性统计检验。
3.5. 动物福利评估
采用改编版动物福利评估网格对来自53个农村家庭的85只动物进行了动物福利评估（30户饲养1只、17户饲养2只、4户饲养3只、1户饲养4只、1户饲养5只），包括34只犬（40.0%）和51只猫（60.0%）。犬的总体福利平均评分为2.46 ± 0.88（n = 34），猫为2.18 ± 0.74（n = 51）。在犬组中，平均指标评分最高的是环境丰富度（5.79 ± 3.53）、体况（3.03 ± 2.75）以及选择、控制和可预测性（2.77 ± 2.09）。在猫组中，平均指标评分最高的是环境丰富度（3.29 ± 3.32）、恐惧和焦虑频率（3.04 ± 2.18）、体况（3.00 ± 1.82）以及选择、控制和可预测性（3.00 ± 1.95）。评分越高表示福利越差。
总体而言，总体福利平均评分为2.29，中位数为2.23，评分范围为1至4.38，评分越低表示福利越好。平均指标评分最高的是环境丰富度、体况、评估期间的保定、恐惧和焦虑频率以及选择、控制和可预测性。平均指标评分最低的是对照顾者和陌生人的攻击性、社交互动、活动能力以及饮食（图4）。
[图：使用动物福利评估网格（AWAG）评估的13项动物福利指标在动物中的平均得分]
在考虑家庭聚集性后（线性混合模型，家庭作为随机截距），动物年龄仍与总体福利评分显著相关（F (4, 72.7) = 5.92，p = 0.0004）。一岁以下的动物评分最有利（相对于参照组的估计值为−0.727；p = 0.0004），10岁及以上的动物评分最不利（估计值+0.581；p = 0.003）。中间年龄组（1–4岁和5–9岁）的动物与参照类别无显著差异（分别为p = 0.183和p = 0.273）。各组均值与未调整分析一致：<1岁动物为1.69；1–4岁为2.22；5–9岁为2.17；≥10岁为2.99（图5）。动物年龄模型中家庭的组内相关系数（ICC）为14.2%。相比之下，总体福利评分未因任何主人人口学特征而存在显著差异（教育程度：F (1, 55.5) = 0.45，p = 0.507；性别：F (1, 68.7) = 0.75，p = 0.390；年龄：F (1, 56.9) = 0.36，p = 0.552；就业状况：F (1, 55.2) = 0.44，p = 0.511），这些模型中家庭ICC范围为21.6%至23.9%。
[图：按动物年龄组划分的平均福利得分分布。黑点代表单个动物，绿色菱形]
4. 讨论
本研究发现，农村家庭饲养的伴侣动物更多，且报告微芯片植入和狂犬病疫苗接种的实施一致性低于城市家庭。农村家庭的绝育似乎也较不完善，但居住地差异无统计学显著性。相比之下，总体福利评分会因主人人口学特征而变化的假设未得到支持。微芯片植入、疫苗接种和生殖控制仍是负责任饲养的重要要素，因为它们有助于动物福利、主人问责和伴侣动物种群管理[3,7]。
农村家庭倾向于饲养更多动物，这与智利、美国和爱尔兰的研究一致，这些研究同样报告农村社区的宠物饲养比城市社区更常见[6,15,29]。在立陶宛，仅31.6%的人口居住在农村地区，而约64%的登记宠物位于三个最大城市之外[23,24]。鉴于农村家庭观察到的微芯片植入率较低，官方记录可能大幅低估了这些地区动物的真实数量。类似的模式也可能出现在波罗的海和东欧地区具有相似农村背景的其他地方，但仍需进一步研究。在规划微芯片植入、疫苗接种和生殖控制项目以及更广泛的负责任饲养和福利干预措施时，应考虑这种地理分布。
在本研究中，植入微芯片是显示城乡差异最明显的做法之一。农村受访者比城市受访者更不可能报告其家中所有伴侣动物均已植入微芯片。在立陶宛，2021年5月1日或之后出生的犬、猫和雪貂必须在出生后四个月内、且在转移给另一所有者或饲养者之前植入微芯片，而该日期之前出生的动物必须在2022年5月1日前植入微芯片[30]。只有注册在伴侣动物登记册中的动物才能转移给另一所有者或饲养者[30]。微芯片植入和准确注册有助于证明所有权和法律责任，促进走失动物的归还，并在动物转移时提供可追溯性[7,30,31,32]。因此，较低的普及率可能降低流浪或走失动物被归还给其所有者的可能性，并使动物走失、被遗弃或非正式转移时更难识别前饲养者。智利关于犬和猫以及丹麦关于猫的研究也报告了类似的城乡差异[6,18]。本研究中观察到的与教育程度和就业状况的关联表明，居住地并非与微芯片植入相关的唯一因素。可能的解释包括意识、经济状况和获得注册服务的机会方面的差异，但这些机制未被直接测量。来自其他强制微芯片植入系统的证据也表明，即使存在法律要求，知识有限、对注册数据库的困惑和低参与度也可能降低依从性[33]。
狂犬病疫苗接种在不同居住环境之间也存在差异，农村受访者更不可能报告家中所有动物在过去三年内均已接种疫苗。在立陶宛，犬、猫和雪貂的定期狂犬病疫苗接种是强制性的。爱沙尼亚和邻国波兰也存在类似要求，尽管所涵盖的物种和再接种间隔不同[34,35,36]。因此，较低的报告覆盖率表明部分受访者对法律要求的疫苗接种存在不完全接种的情况。疫苗接种还与教育程度和就业状况相关，表明居住地并非唯一相关因素；然而，意识、经济状况和获得兽医服务的机会的可能作用未被直接评估。维持疫苗接种覆盖率仍然很重要，因为受感染动物可能从流行地区输入，如2026年一例涉及从俄罗斯带到德国的犬的病例所示[37]。因此，狂犬病疫苗接种既是个体动物健康责任，也是群体层面的“同一健康”措施。提高接种率可能需要可及的服务以及对所有者法律和公共卫生责任的有针对性沟通[7]。
绝育仅显示出非显著的城乡趋势。农村受访者更不可能报告家中所有动物均已绝育，更可能报告没有动物绝育，但本研究未证明居住地差异具有统计学显著性。智利此前曾报告城市地区绝育率高于农村地区[6]。在本研究中，教育程度与绝育的关联比居住地更明显。先前的研究已确定兽医建议、感知到的所有者责任以及预期的健康或福利结果作为影响绝育决策的因素[38]。因此，有针对性的、基于证据的沟通可能有助于所有者做出知情的繁殖管理决策，尽管本研究中未直接评估观察到的与教育程度关联背后的原因。
值得注意的是，城市受访者比农村受访者更常饲养单只动物（53.3% 对 26.9%）；因此，“所有动物”标准在城市家庭中更容易满足，因为需要植入微芯片、接种疫苗或绝育的动物更少。因此，观察到的这些做法上的城乡差异可能部分反映的是每户饲养动物数量的差异，而非仅居住地差异。此外，由于所有权和护理做法是在家庭层面而非物种层面报告的，家庭物种构成对观察到的城乡差异的贡献无法确定。
伴侣动物主要通过非正式途径获得。总体而言，69.3%的获取来源 responses 为流浪动物收容、熟人或者家人赠送、通过广告获得、已拥有动物所生、从未注册繁殖者处购买或从其他私人个体处获得。其他国家也报道了类似的依赖赠送、社区网络、私人转让和流浪动物的情况[39,40,41,42]。其中一些途径可能涉及计划外窝产或发生在正式身份识别和登记程序之外的转让。然而，本研究未将个体获取来源与随后的微芯片植入、登记或绝育联系起来，因此无法确定非正式获得的动物是否接受了较不一致的照护。非正式获取占主导地位应被解释为确定了种群管理干预的一个潜在重要切入点，包括转让前微芯片植入、准确记录所有权变更和预防计划外窝产[7,43]。
总体而言，近一半的受访者不认为需要就诊兽医或更频繁地就诊。这一发现可能与更为被动的兽医照护方式一致，即主要在主人认为存在即时健康问题时才寻求专业建议。定期兽医照护不仅可以提供疫苗接种和早期疾病检测的机会，还可以提供关于营养、行为、环境丰富、应激以及其他身体和心理福利方面的循证指导[8,9,10,11]。当不定期寻求兽医照护时，主人教育和早期识别不太明显的福利问题的机会可能会减少。因此，仅增加服务可及性可能不够，除非主人也了解兽医在明显疾病发生之前维护健康和福利方面的作用[8,44]。
饲养管理方式在居住环境之间也存在差异，并揭示了农村地区若干令人担忧的模式。农村家庭的动物较少被完全饲养在室内，更常被允许不受限制的户外活动、饲养在户外犬舍或拴系。拴系尤其引起福利关注，因为长时间约束可能限制运动、行为选择和社交接触，同时增加挫败感、受伤和暴露于不利环境条件的风险[45]。这些担忧越来越多地反映在欧洲立法中：奥地利禁止将犬只拴在链子上或以其他方式拴系，除了在商店前短期拴系外，而2026年通过的欧盟法规一旦适用，将禁止操作者将犬或猫拴系超过一小时，除非在医疗治疗或参加表演、展览和比赛期间；对于用于军事、警察或海关服务的犬只，也可能适用有限的豁免[46,47]。喂养方式也存在差异，城市受访者更频繁地报告使用优质或超优质商业日粮，而农村受访者更常报告使用超市宠物食品、餐桌残羹、骨头和粥。在其他伴侣动物种群中也报道了依赖家庭自制食物和餐桌残羹的情况[48,49]。适当的营养是负责任饲养的核心义务，因为主人必须提供适合动物物种、生命阶段和健康状况的完整均衡日粮。当餐桌食物或未配制的家庭自制食物构成饮食的很大一部分时，它们可能稀释必需营养素、提供过多能量或使动物面临其他营养风险[48,50,51]。这些发现表明，农村社区的负责任饲养干预不仅应解决预防性医疗保健问题，还应解决日常饲养、活动自由、社交机会和循证喂养实践问题。
直接福利评估显示，在丰富度、体况、评估过程中的操作处理、恐惧与焦虑频率以及选择、控制和可预测性方面，平均指标得分相对较高。由于不存在经过验证的临界值可将研究特定的适应性总体评分分类为良好或不良，因此这些发现应作比较性解读，而非作为总体福利普遍良好的证据。结果识别出可能需要关注的营养、行为和环境方面，但该研究未评估主人对这些需求的认识或理解。体况结果与既往证据相关，即主人可能低估超重体况，而环境和行为方面的发现与已报道的主人对猫行为、环境需求和应激相关信号知识有限的结论一致[12,13,52]。因此，定期兽医指导不仅对疾病预防重要，对体重管理、环境丰富和行为福利也同样重要[9,50,53]。与我们的假设相反，在农村福利评估样本中，总体福利评分与主人年龄、教育程度、性别或就业状况无显著关联。年龄较大的动物总体福利评分较差。慢性疾病、疼痛以及与年龄相关的功能或行为变化可能导致老年动物福利较差；然而，本研究未将这些因素作为解释变量直接评估，其与观察到的关联的贡献无法确定。福利较差并非衰老的必然结果，因为早期发现、预防性护理和适龄管理有助于维持生活质量[54]。由于犬特异性AWAG初步在犬中进行了验证，在解读混合物种结果时应考虑本研究对其进行的猫适应性调整[28]。
综上所述，这些发现表明，主人与动物的关系不仅通过饲养动物的决定来体现，还通过为其健康、可追溯性、繁殖、居住、营养和行为福利承担的日常责任来体现。在农村环境中，家庭动物数量较多、微芯片植入和疫苗接种要求依从性较低、繁殖控制较弱、非正式转让、预防性兽医服务参与有限以及令人担忧的饲养管理实践等因素叠加，可能放大福利和种群管理两方面的后果。这一模式表明，负责任养宠是人与动物关系的一个重要实践维度，不能仅从饲养行为本身推断。因此，有效应对措施应结合法律责任、可及的预防性服务、可追溯的饲养权转让、繁殖控制以及针对生理和心理需求的实用教育。类似挑战可能在其他农村环境中同样存在，尤其是伴侣动物数量众多但正式饲养和预防性护理制度仍未能一致落实的地区。
研究局限性
应承认若干局限性。便利抽样设计不允许得出因果结论，并限制了研究结果的普适性。在线问卷通过公开社交媒体帖子分发，而面对面参与者是在微芯片植入和绝育活动期间以及通过兽医援助请求招募的。两种招募途径均可能导致自我选择。较小的城市对照组降低了城市估计的精确度和检测较小城乡差异的统计功效；因此，非显著比较不应被解读为等效的证据。城乡差异应解读为研究样本内的比较，而非全国代表性估计。此外，使用在线自填问卷和辅助面对面问卷可能引入调查模式效应。居住地使用自我报告的地点信息或粗略的居住描述进行分类，因此可能发生一些误分类。
大多数饲养和饲养管理实践为家庭层面的自我报告，未经独立核实。个体获取途径无法与特定动物或其后续护理相关联，而居住和喂养变量允许多选并进行了描述性分析。直接福利评估包括来自53个农村家庭的85只动物，这些家庭通过社区倡议和援助请求招募。该非随机子样本旨在识别福利优先事项，无法提供种群水平的患病率估计。福利在单一时间点评估，因此无法检查时间变化。评估期间观察到的行为反应，包括对不熟悉评估者的防御性反应，可能反映即时评估环境而非动物的平常行为或长期福利状态。
AWAG此前已在犬中进行了初步验证，并在本研究中针对猫进行了改编。尽管在实施过程中考虑了物种特异性的行为表达和需求，但改编后的评估未经过独立的猫特异性专家审查或正式验证。这限制了对跨物种合并结果的解读以及犬与猫之间的直接比较。此外，福利评估样本量相对较小，包括34只犬和51只猫，不足以支持可靠的物种特异性比较。未来研究应使用更大的按物种分层的样本来考察犬和猫在福利需求和结局方面的差异。
5. 结论
本研究突显了立陶宛农村地区伴侣动物饲养规模与负责任饲养实践的一致实施之间的不匹配。研究结果表明，仅拥有动物并不能确保动物的健康、行为和福利需求得到充分满足，尤其是在饲养数量较多且所有权转让仍不规范的情况下。
改善农村地区伴侣动物福利需要更好的微芯片植入和疫苗接种覆盖率、繁殖控制以及更定期的预防性兽医护理。饲养者还需要关于营养、行为和动物心理需求的实用指导。类似的挑战可能也存在于波罗的海和东欧其他农村地区，但需要比较研究来揭示这些模式的普遍适用程度。
致谢
作者感谢所有参与本研究并允许对其动物进行检查的伴侣动物饲养者；感谢Miško trobelė动物收容所的志愿者协助组织和开展数据收集所在的社区活动；感谢Laurynas Pašakarnis在数据管理和统计分析方面的指导。第2.8节中披露的生成式人工智能工具的产品详情如下：ChatGPT（OpenAI；GPT-5.5和GPT-5.6 Thinking；于2026年访问）和Perplexity AI（Perplexity；于2026年访问）。
缩写
本手稿中使用了以下缩写：
| AWAG | Animal Welfare Assessment Grid |
| OECD | Organisation for Economic Co-operation and Development |
补充材料
以下支持信息可从以下网址下载：https://www.mdpi.com/article/10.3390/ani16192988/s1，补充材料S1. 在线问卷英文翻译；补充材料S2. 面对面饲养者问卷英文翻译；补充材料S3. 基于动物福利评估网格（AWAG）改编的动物福利评估表。
[内容请查看原文]
作者贡献
概念化，E.M.R.、A.M.和T.Š.；方法学，E.M.R.、A.M.和S.M.；形式分析，E.M.R.和A.M.；调查，E.M.R.；数据管理，E.M.R.和A.M.；可视化，E.M.R.和S.M.；写作——初稿准备，E.M.R.；写作——审阅与编辑，E.M.R.、A.M.、S.M.和T.Š.；监督，A.M.。所有作者均已阅读并同意手稿的出版版本。
机构审查委员会声明
本研究按照《赫尔辛基宣言》进行，并获得立陶宛健康科学大学生物伦理中心的批准（批准号2025-BEC2-0990；2025年9月26日）。
知情同意声明
已获得所有参与本研究人员的知情同意。对于匿名在线调查，在展示研究信息后提交问卷即构成知情同意。面对面部分的参与者在数据收集前提供了知情同意，并许可对其动物进行检查和参与。
数据可用性声明
支持本研究结果的匿名化数据可在合理请求下向通讯作者获取。数据未公开提供，因为其包含详细的地点和家庭层面信息，可能增加参与者被重新识别的风险。
利益冲突
E.M.R.是Miško Trobelė动物收容所的创始人和主任，该收容所为招募面对面调查和动物福利评估参与者所在的社区活动提供了便利。参与完全自愿，决定参与、拒绝参与或退出对通过这些活动或收容所提供的任何服务或援助的获取或接受没有影响。此角色出于透明度考虑予以披露。作者声明无商业或财务利益冲突。
资助声明
本研究未获得外部资助。
脚注
[内容请查看原文]
参考文献
[内容请查看原文]
关联数据
补充材料
[内容请查看原文]
数据可用性声明
支持本研究结果的匿名化数据可在合理请求下向通讯作者获取。数据未公开提供，因为其包含详细的地点和家庭层面信息，可能增加参与者被重新识别的风险。
查看英文原文▾
1. Introduction
Dogs and cats are increasingly regarded as companions and family members rather than animals kept solely for practical purposes [1,2]. This shift has raised expectations regarding the quality of care and welfare provided to them. However, describing an animal as a companion does not in itself ensure that all of its needs are met, and responsibility is understood and enacted variably among owners [3,4]. Responsible ownership should therefore be examined through specific care and management practices rather than inferred from ownership or affection alone.
In this study, responsible pet ownership refers to the owner’s practices and decisions throughout an animal’s life that are intended to meet its needs and fulfill responsibilities to the animal and the wider community. These include responsible acquisition, microchipping and registration, preventive healthcare, reproductive management, appropriate nutrition and housing, control of unsupervised roaming, and timely veterinary care [3,5,6]. These practices support individual animal health and welfare, owner accountability and companion animal population management [6,7]. Veterinary involvement additionally provides opportunities for early disease detection and professional guidance on nutrition, reproduction and behaviour [8,9]. Nevertheless, some owners do not seek veterinary care when an animal appears healthy or when no immediate need is perceived [10], resulting in a more reactive than preventive approach to care.
How owners fulfil these responsibilities can influence aspects of animal welfare that may be less immediately visible than illness or injury. Contemporary animal-welfare approaches recognise that welfare is shaped by nutrition, physical health, the living environment, opportunities to perform normal behaviour and the animal’s resulting mental state [11]. Owners may underestimate body condition in both dogs and cats [12], while studies across companion animal species have identified variation in owners’ understanding of behavioural and environmental needs and recognition of stress-related signs [8,13,14]. Thus, animals may receive basic care while still experiencing unmet nutritional, behavioural, environmental or psychological welfare needs.
The extent to which these responsibilities are implemented may vary across social and geographical settings. Rural communities are particularly relevant in this context because companion animal ownership patterns, husbandry practices and access to preventive services may differ from those in urban areas. In Chile, rural households were more likely to keep companion animals and had more animals per person, while neutering and microchipping rates were lower than in urban areas [6]. In four underserved communities in the United States, rural communities had higher overall pet-keeping and dog-ownership rates than urban communities [15]. In Great Britain, estimated dog and cat densities were highest in major metropolitan areas, but the number of pets per household was generally highest in rural areas [16]. A national household survey similarly found that dogs were more likely to be owned by rural households and cats by semi-urban or rural households [17]. Research in Denmark has also shown that microchipping cats remained less prevalent in rural areas than in more densely populated areas [18]. Where more animals are kept per household, inconsistent microchipping, preventive care and reproductive control may affect a greater number of animals. However, rural residence should not itself be interpreted as evidence of poorer care. Rural–urban differences are likely to reflect a combination of animal numbers, husbandry traditions, socioeconomic circumstances, service availability and owners’ understanding of animal needs rather than place of residence alone.
In Lithuania, companion animal welfare and ownership are governed by the Animal Welfare and Protection Law adopted in 1997 [19], which sets general welfare requirements, prohibits neglect and cruelty, and since its 2021 amendment mandates the microchipping and registration of all dogs and cats. Mandatory disease control and anti-rabies vaccination is regulated under the Veterinary Law adopted in 1991 [20], the associated Government resolutions on requirements for rabies vaccination adopted in 2007 [21], and disease control obligations [22]. In Lithuania, approximately one-third of the population lives in rural areas [23]. Although the total dog and cat populations are unknown because a nationwide census has not been fully implemented, the national Companion Animal Register recorded 345,085 dogs and 188,971 cats as of 1 January 2026, with approximately 65% of registered companion animals located outside the municipalities of the three largest cities [24]. Taking into account the lower microchipping rates that have been reported in rural areas by other studies [6,18], it is possible that the number of companion animals living in rural areas may be considerably higher than official records suggest, making this population particularly important for animal-welfare and population-management research.
Previous Lithuanian studies have examined factors influencing the choice of veterinary services and selected aspects of service provision, including veterinary telemedicine [25,26]. However, little is known about rural–urban differences in companion animal numbers, acquisition, microchipping, preventive care, reproductive control, husbandry practices and veterinary-care use. A bibliometric analysis of the international dog and cat welfare literature identified the United Kingdom, Italy, Spain, Germany and Austria as its principal European contributors, suggesting that Baltic and Eastern European settings remain underrepresented [1]. To our knowledge, no previous Lithuanian study has combined a rural–urban comparison of owner practices with a direct, structured welfare assessment of dogs and cats living in rural households.
To address these gaps, the present study combined questionnaire data on owners, households and owned dogs and cats—hereafter collectively referred to as companion animals—with direct animal welfare assessment. The study aimed to compare companion animal numbers, selected ownership and care practices, husbandry practices and veterinary-care use between rural and urban households in Lithuania, to describe companion animal acquisition routes in the overall sample, and to identify the principal welfare concerns among dogs and cats assessed in rural households using an adapted Animal Welfare Assessment Grid (AWAG).
We hypothesised that rural respondents would report keeping more companion animals and less consistent implementation of selected responsible ownership practices than urban respondents, and that overall welfare scores would vary according to owner sociodemographic characteristics.
2. Materials and Methods
2.1. Study Design and Setting
The study was conducted in Lithuania between September 2025 and January 2026. It consisted of an anonymous online questionnaire and an in-person owner questionnaire combined with direct animal welfare assessment. A convenience sampling strategy was used.
Eligible participants were adults who owned at least one dog or cat. The study focused primarily on rural communities. The online questionnaire also included urban owners who accessed the openly distributed survey, whereas the in-person component and all direct animal welfare assessments were conducted exclusively among rural residents. Rural areas were defined according to the Organisation for Economic Co-operation and Development (OECD) criterion as localities with a population density below 150 inhabitants/km2 [27]. Respondents were asked to provide the name of their place of residence, which was classified as rural or urban using official population-density data.
2.2. Ethical Approval and Informed Consent
The study, including the questionnaire and the non-invasive assessment of privately owned dogs and cats, was conducted with the approval of the Bioethics Centre of the Lithuanian University of Health Sciences (approval no. 2025-BEC2-0990, 26 September 2025). Participation was voluntary, and all participants were informed about the purpose of the study, the intended use of the results in scientific publications, their right to withdraw, and the protection of their data.
No directly identifying personal information or exact residential addresses were collected. Before accessing and completing the online questionnaire, all respondents were required to read information about the study objectives and the processing of personal data. Proceeding to and completing the questionnaire was considered confirmation of informed consent.
Before the in-person questionnaire and animal assessment, participants also agreed to take part and permitted the non-invasive examination of their animals. The assessments involved external observation and examination only and did not include painful or invasive procedures. All information relating to participants and animals was anonymised before analysis.
2.3. Questionnaire Development and Pilot Testing
A study-specific owner questionnaire was developed in Lithuanian based on the study objectives and relevant literature, under the guidance of researchers with expertise in veterinary epidemiology, animal welfare and companion animal behaviour. The draft questionnaire underwent pilot testing with a small sample of dog and cat owners before the main data collection to assess the clarity of the questions and the suitability of the response options and was revised accordingly until the final version was produced. The same core questionnaire was used for the online and in-person surveys, with additional questions and the animal welfare assessment included during the household visits. The English version provided as Supplementary Materials S1 and S2 was prepared for publication and was not used for data collection.
2.4. Online Questionnaire
The anonymous online questionnaire was created using Google Forms and distributed through local Facebook groups between November 2025 and January 2026. Because the invitation was circulated through open social-media posts, the number of individuals who viewed or received the invitation could not be determined and a formal response rate could not be calculated. The questionnaire required approximately 10–15 min to complete.
The questionnaire contained closed, multiple-response and open-ended questions organised into four principal sections. The first section collected owner demographic information, including age, gender, education, employment status and place of residence. The second collected information about the animals kept in the household, including the numbers of dogs and cats and the routes through which animals had been acquired. The third addressed responsible ownership and preventive-care practices, including microchipping, rabies vaccination, parasite prevention, neutering, breeding, feeding and housing. The fourth addressed veterinary-care use, responses to illness or injury, reasons for veterinary visits, frequency of attendance, perceived barriers to care and sources of information regarding animal health, nutrition, care and welfare.
The English version of the questionnaire is provided as Supplementary Material S1.
2.5. In-Person Owner Survey
The in-person component used an assisted questionnaire. The researcher entered participants’ answers into Google Forms containing the same core questions as the online survey, supplemented with questions and observations required for the animal welfare assessment.
Potential participants were approached during three companion-animal microchipping and neutering campaigns and subsequently in individual cases in which owners requested veterinary assistance. All eligible adult dog or cat owners encountered through these activities were invited to participate, irrespective of owner or animal characteristics, and no additional selection parameters were applied. Participation was entirely voluntary, and the decision to participate, decline participation, or withdraw had no effect on access to or receipt of any services or assistance provided through these initiatives or by the shelter. Only adults who consented to both the questionnaire and the examination of at least one animal were included.
Each participating owner was assigned an anonymised identification number. Individual animal identification numbers were used to link the owner questionnaire with the relevant welfare assessment while maintaining anonymity. The combined owner interview and animal assessment generally required 15–20 min.
The English version of the questionnaire is provided as Supplementary Material S2.
2.6. Animal Welfare Assessment
Animal welfare was assessed using an adapted version of the Animal Welfare Assessment Grid (AWAG), which was previously preliminarily validated for dogs [28]. Within this tool, individual factors within the physical, psychological, environmental and procedural welfare parameters are scored from 1, representing the most favourable welfare state, to 10, representing the least favourable welfare state. Mean factor scores are calculated for each parameter and used to derive a polygon-based cumulative welfare assessment score.
For the present study, the method was adapted for a one-time assessment of both dogs and cats at a single time point. The underlying 1–10 scoring structure and severity thresholds were retained for both species, while species-specific behavioural expressions and needs were used when interpreting the descriptors. Written modifications included feline expressions, such as hissing and scratching, alongside canine growling and biting in the aggression indicators. During the structured interview, the assessor also explained species-appropriate manifestations of stress and species-specific environmental and behavioural needs, including hiding, scratching, exploration, hunting-related behaviour and enrichment in cats. Depending on the indicator, the score was either selected by the owner following this explanation or assigned by the assessor based on direct observation and owner-provided information. These species-specific interpretations were grounded in the assessor’s formal veterinary education and extensive hands-on experience in evaluating and managing the behavioural and welfare needs of dogs and cats. The adapted assessment form used in the study is provided as Supplementary Material S3.
All assessments were performed by the same veterinary medicine student using the predefined descriptors in the adapted assessment form. Thirteen welfare indicators were evaluated: mobility; body condition; clinical assessment; eating and drinking; aggression towards the caregiver; aggression towards unfamiliar people; fears and anxieties frequency; reaction to stressors; choice, control and predictability; enrichment; social interactions; behaviour during assessment; and handling during assessment.
Two factors included in the original AWAG—change in daily routine and procedure pain—were excluded because they were intended to evaluate the effects of veterinary or management procedures and were not applicable to the non-invasive, single-assessment design of the present study.
Each of the 13 indicators was assigned a numerical score using the corresponding verbal descriptor in the adapted assessment form. For the present analysis, a study-specific overall welfare score was calculated as the arithmetic mean of the 13 indicator scores. This arithmetic mean was used as a study-specific descriptive summary and did not represent the polygon-based cumulative score used in the original AWAG. Lower scores indicated more favourable welfare.
The direct welfare-assessment component was intentionally restricted to rural households because it was designed to identify welfare priorities in the study’s target rural population and was conducted through rural community initiatives; it was not intended to provide a rural–urban comparison of animal-level welfare.
2.7. Variable Definition and Data Processing
Responses from the online and in-person questionnaires were exported and combined into a single analytical database. Questions that were identical in both questionnaires were harmonised and analysed together.
Owner education was grouped as higher education, comprising university and college higher education, or no higher education, comprising primary, lower-secondary, secondary and vocational education. Employment status was grouped as employed or unemployed.
The total number of animals in each household was calculated by combining the reported numbers of dogs and cats and was grouped as one, two to five, six to ten, or eleven or more animals. The online questionnaire recorded the numbers of dogs and cats using predefined categories (0, 1, 2–5, 6–10, or ≥11), whereas the in-person questionnaire recorded exact counts. For the combined analysis, the exact in-person counts were recoded into the corresponding categories. After harmonisation, household animal numbers were analysed as one, two to five, six to ten, or eleven or more animals.
Household-level microchipping, rabies vaccination and neutering responses were harmonised into three categories: none of the animals, some of the animals, or all animals in the household. Rabies vaccination referred to vaccination within the previous three years.
For questions allowing multiple selections, each option was converted into a separate binary variable, coded as 0 when not selected and 1 when selected. Consequently, response categories for acquisition routes, feeding, housing, veterinary-care behaviour, barriers and information sources were not mutually exclusive. Open-text responses and responses entered under “other” were reviewed and, where appropriate, assigned to existing categories. Responses that could not be assigned to a predefined category were grouped according to their content by identifying recurrent themes.
Animal-welfare responses recorded using verbal descriptors were converted into numerical scores according to the adapted AWAG assessment form. Owner questionnaire data and animal-level welfare data were linked through anonymised identification numbers.
2.8. Statistical Analysis
Statistical analyses were performed using Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) and JMP® 18 (JMP Statistical Discovery LLC, Cary, NC, USA). Frequencies and percentages were calculated for categorical variables, while means, medians and ranges were calculated for continuous welfare scores. Pearson’s chi-square test was used to assess associations between categorical variables, including rural–urban differences and associations between owner demographic characteristics and household-level ownership-practice outcomes. To account for the non-independence of welfare scores from animals within the same household, welfare score comparisons were performed using linear mixed models (restricted maximum likelihood, REML) with household as a random intercept. Fixed effects were animal age (categorical: <1, 1–4, 5–9 and ≥10 years; one model) and each owner demographic characteristic (education, gender, age and employment status; one model each). Statistical significance was set at p < 0.05.
2.9. Use of Generative Artificial Intelligence
During the preparation of this manuscript, the authors used ChatGPT (OpenAI; GPT-5.5 and GPT-5.6 Thinking; https://chatgpt.com/; accessed on 30 July 2026) and Perplexity AI (Perplexity; https://www.perplexity.ai/; accessed on 30 July 2026) to assist with manuscript organisation and coherence across sections, linguistic refinement, preliminary identification of potentially relevant literature, and review of compliance with journal formatting and publication requirements. All generated outputs were critically reviewed, verified and edited by the authors, who take full responsibility for the final content of the manuscript.
3. Results
3.1. Study Population
A total of 383 cat and dog owners participated in the study, including 330 respondents who completed the online questionnaire and 53 who participated in the in-person survey. For the in-person component, 57 eligible owners were invited to participate; 53 (93.0%) agreed and completed the questionnaire, while four declined to complete it and were therefore not included. Overall, 338 (88.3%) respondents resided in rural areas and 45 (11.7%) in urban areas. Women represented 89.8% of respondents, while 65.6% had completed higher education. During the in-person survey, welfare assessments were performed on 85 companion animals, including 34 dogs and 51 cats. The demographic characteristics of the study population are presented in Table 1.
Table 1. Survey and sociodemographic characteristics of respondents (N = 383). Values are presented as number and percentage of respondents, n (%).
| Characteristic | n (%) | |
|---|---|---|
| Survey type | Online survey | 330 (86.2) |
| In-person survey | 53 (13.8) | |
| Place of residence | Rural | 338 (88.3) |
| Urban | 45 (11.7) | |
| Gender | Female | 344 (89.8) |
| Male | 38 (9.9) | |
| Other | 1 (0.3) | |
| Age group | 18–24 | 42 (11.0) |
| 25–34 | 64 (16.7) | |
| 35–44 | 103 (26.9) | |
| 45–54 | 96 (25.1) | |
| 55–64 | 49 (12.8) | |
| ≥65 | 29 (7.6) | |
| Education | Higher education | 251 (65.6) |
| No higher education | 132 (34.4) | |
| Employment status | Employed | 305 (79.6) |
| Unemployed | 78 (20.4) | |
3.2. Pet Ownership Characteristics
Overall, 48.6% of respondents owned two to five companion animals, 30.0% owned a single animal, 15.7% owned six to ten animals, and 5.7% owned eleven or more animals. The distribution of companion animal numbers differed significantly between rural and urban households (χ2 = 16.675, df = 3, p = 0.0008; Figure 1). Nearly three-quarters of rural respondents (73.1%) owned two or more companion animals, compared with 46.7% of urban respondents. Conversely, more than half of urban households (53.3%) kept only one companion animal, whereas this was the case for only 26.9% of rural households.
[图：Distribution of the reported total number of companion animals per household by place of residence (rural vs. urban). Ba]
Respondents aged 45 years and older also tended to own larger numbers of companion animals than younger respondents (χ2 = 9.977, df = 3, p = 0.0188). Employment status showed a similar tendency, with unemployed respondents more frequently reporting larger numbers of companion animals than employed respondents, although this association did not reach statistical significance (χ2 = 7.754, df = 3, p = 0.0514).
Companion animals were acquired predominantly through informal sources. Overall, 69.3% of acquisition-source responses represented informal routes, including animals taken in as strays, gifted by acquaintances or family, obtained through advertisements, born to already-owned animals, purchased from unregistered breeders, or acquired from private individuals. In comparison, shelter adoptions accounted for 15.3% of responses, purchases from registered breeders for 14.1%, and other sources for 1.4%. The distribution of acquisition sources is presented in Figure 2.
[图：Distribution of reported companion animal acquisition sources. Percentages were calculated from all acquisition-source r]
3.3. Microchipping, Vaccination and Neutering Practices
Microchipping, rabies vaccination and neutering practices were compared between rural and urban households (Figure 3). Complete microchipping was reported by 55.5% of rural respondents and 86.7% of urban respondents, while no pets were microchipped in 32.8% and 8.9% of households, respectively (χ2 = 15.987, df = 2, p = 0.0003). Complete rabies vaccination within the previous three years was reported by 67.1% of rural and 90.9% of urban respondents; no pets were vaccinated in 23.3% and 4.5% of households, respectively (χ2 = 10.742, df = 2, p = 0.0047). Complete neutering was reported by 50.6% of rural and 68.9% of urban respondents, whereas no pets were neutered in 31.7% and 17.8% of households, respectively; this difference did not reach statistical significance (χ2 = 5.518, df = 2, p = 0.0633). Routine parasite prevention was commonly reported in both groups, with no significant rural–urban difference (χ2 = 0.363, df = 1, p = 0.5471).
[图：Distribution of selected preventive ownership practices according to ( A ) place of residence, ( B ) employment status, ]
Higher education was associated with complete microchipping (67.6% vs. 43.1%; χ2 = 23.179, df = 2, p < 0.0001), rabies vaccination (78.2% vs. 53.5%; χ2 = 26.949, df = 2, p < 0.0001) and neutering (58.2% vs. 42.4%; χ2 = 16.617, df = 2, p = 0.0002). Employment was also associated with complete microchipping (62.2% vs. 47.4%; χ2 = 6.034, df = 2, p = 0.0489) and rabies vaccination (74.1% vs. 52.7%; χ2 = 13.474, df = 2, p = 0.0012). Rabies vaccination coverage additionally differed according to age (χ2 = 20.444, df = 10, p = 0.0253); among respondents aged 65 years and older, 48.1% reported that all pets were vaccinated, while the same proportion reported that none were vaccinated.
Commonly reported reasons for incomplete microchipping included a perceived lack of need, financial or logistical difficulties, the animal’s age, concerns about causing stress, difficulty catching the animal, limited knowledge and unwillingness to microchip. Reasons for incomplete rabies vaccination included a perceived lack of need because of the animal’s lifestyle or age, financial and logistical difficulties, problems catching or transporting the animal, limited knowledge, distrust of vaccines and postponement. Reasons for not neutering included a perceived lack of need, the animal’s age, breeding intentions, financial or logistical difficulties and concerns about possible health or behavioural effects.
Overall, 49.9% of respondents reported that they did not perceive a need to visit a veterinarian or to do so more frequently.
3.4. Husbandry Practices
Housing practices varied descriptively between residential settings. Because respondents could select more than one option, the reported categories were not mutually exclusive. Keeping animals indoors accounted for 93.6% of urban and 63.7% of rural housing responses. Unrestricted outdoor access was reported more frequently in rural than urban settings (20.6% vs. 4.3%), as was housing in outdoor kennels (11.2% vs. 2.1%). Tethering accounted for 4.5% of rural housing responses and was not reported by urban respondents.
Feeding categories were also non-mutually exclusive. Premium or super-premium commercial diets were reported more frequently in urban than rural households (51.3% vs. 31.3%). Rural respondents more frequently reported supermarket pet food (19.5% vs. 10.3%), table scraps (9.9% vs. 3.8%), bones (9.4% vs. 6.4%) and porridge (6.3% vs. 5.1%). These rural–urban comparisons are descriptive, as inferential statistical tests were not performed for these multi-response variables.
3.5. Animal Welfare Assessment
Animal welfare was assessed in 85 animals from 53 rural households (30 households with 1 animal, 17 with 2, 4 with 3, 1 with 4, 1 with 5), including 34 dogs (40.0%) and 51 cats (60.0%), using the adapted Animal Welfare Assessment Grid. The mean overall welfare score was 2.46 ± 0.88 in dogs (n = 34) and 2.18 ± 0.74 in cats (n = 51). Within the dog group, the highest mean indicator scores were recorded for enrichment (5.79 ± 3.53), body condition (3.03 ± 2.75), and choice, control and predictability (2.77 ± 2.09). Within the cat group, the highest mean indicator scores were recorded for enrichment (3.29 ± 3.32), fears and anxieties frequency (3.04 ± 2.18), body condition (3.00 ± 1.82), and choice, control and predictability (3.00 ± 1.95). Higher scores indicate poorer welfare.
Overall, the mean overall welfare score was 2.29, the median was 2.23, and scores ranged from 1 to 4.38, with lower scores indicating better welfare. The highest mean indicator scores were recorded for enrichment, body condition, handling during assessment, fears and anxieties frequency, and choice, control and predictability. The lowest mean indicator scores were recorded for aggression towards the caregiver and unfamiliar people, social interactions, mobility, and eating and drinking (Figure 4).
[图：Mean scores for the 13 animal welfare indicators assessed using the Animal Welfare Assessment Grid (AWAG) among animals ]
After accounting for household clustering (linear mixed model, household as random intercept), animal age remained significantly associated with the overall welfare score (F (4, 72.7) = 5.92, p = 0.0004). Animals younger than one year had the most favourable scores (estimate −0.727 relative to the reference group; p = 0.0004), and animals aged 10 years and older had the least favourable scores (estimate +0.581; p = 0.003). Animals in the intermediate age groups (1–4 and 5–9 years) did not differ significantly from the reference category (p = 0.183 and p = 0.273, respectively). Group means were consistent with the unadjusted analysis: animals aged <1 year, 1.69; 1–4 years, 2.22; 5–9 years, 2.17; ≥10 years, 2.99 (Figure 5). The intraclass correlation coefficient (ICC) for household was 14.2% in the animal age model. In contrast, the overall welfare score did not differ significantly by any owner demographic characteristic (education: F (1, 55.5) = 0.45, p = 0.507; gender: F (1, 68.7) = 0.75, p = 0.390; age: F (1, 56.9) = 0.36, p = 0.552; employment: F (1, 55.2) = 0.44, p = 0.511), with household ICC ranging from 21.6% to 23.9% across these models.
[图：Distribution of mean welfare scores according to animal age group. Black dots represent individual animals, green diamon]
4. Discussion
The present study found that rural households kept more companion animals and reported less consistent implementation of microchipping and rabies vaccination than urban households. Neutering also appeared less complete in rural households, but the residential difference was not statistically significant. By contrast, the hypothesis that overall welfare scores would vary according to owner demographic characteristics was not supported. Microchipping, vaccination and reproductive control remain important elements of responsible ownership because they contribute to animal welfare, owner accountability and companion animal population management [3,7].
The tendency for rural households to keep more animals is consistent with studies from Chile, the United States and Ireland, where pet ownership has likewise been reported to be more common in rural than urban communities [6,15,29]. In Lithuania, only 31.6% of the population lives in rural areas, while approximately 64% of registered pets are located outside the three largest cities [23,24]. Given the lower microchipping rates observed in rural households, official records may substantially underestimate the true number of animals in these areas. Comparable patterns may also occur elsewhere in the Baltic and Eastern European regions with similar rural contexts, although further research is needed. This geographical distribution should be considered when planning microchipping, vaccination and reproductive control programmes, as well as broader responsible ownership and welfare interventions.
Microchipping was one of the practices showing the clearest rural–urban differences in the present study. Rural respondents were less likely than urban respondents to report that all companion animals in their household were microchipped. In Lithuania, dogs, cats and ferrets born on or after 1 May 2021 must be microchipped within four months of birth and before transfer to another owner or keeper, whereas animals born before that date had to be microchipped by 1 May 2022 [30]. Only animals registered in the Companion Animal Register may be transferred to another owner or keeper [30]. Microchipping and accurate registration support proof of ownership and legal accountability, facilitate the return of lost animals and provide traceability when animals are transferred [7,30,31,32]. Lower uptake may therefore reduce the likelihood that roaming or lost animals are returned to their owners and make it more difficult to identify the previous keeper when an animal is lost, abandoned or transferred informally. Similar rural–urban differences have been reported in Chile for dogs and cats and in Denmark for cats [6,18]. The associations with education and employment observed in the present study indicate that residence was not the only factor associated with microchipping. Possible explanations include differences in awareness, financial circumstances and access to registration services, but these mechanisms were not directly measured. Evidence from other compulsory microchipping systems also indicates that limited knowledge, confusion about registration databases and low engagement may reduce compliance even when a legal requirement exists [33].
Rabies vaccination also differed between residential settings, with rural respondents less likely to report that all animals in the household had been vaccinated within the previous three years. Regular rabies vaccination of dogs, cats and ferrets is mandatory in Lithuania. Comparable requirements also exist in Estonia and neighbouring Poland, although the species covered and revaccination intervals differ [34,35,36]. The lower reported coverage therefore indicates incomplete uptake of legally required vaccination among some respondents. Vaccination was also associated with education and employment, suggesting that residence was not the only relevant factor; however, the possible roles of awareness, financial circumstances and access to veterinary services were not directly assessed. Maintaining vaccination coverage remains important because infected animals may be imported from endemic areas, as illustrated by a 2026 case involving a dog brought from Russia to Germany [37]. Rabies vaccination therefore represents both an individual animal-health responsibility and a population-level One Health measure. Improving uptake may require accessible services together with targeted communication about owners’ legal and public-health responsibilities [7].
Neutering showed only a non-significant rural–urban tendency. Rural respondents were less likely to report that all animals in the household had been neutered and more likely to report that none had been neutered, but the present study did not demonstrate a statistically significant residential difference. Higher neutering rates in urban areas than in rural areas have previously been reported in Chile [6]. In the present study, education was more clearly associated with neutering than place of residence. Previous research has identified veterinary advice, perceived owner responsibility and expected health or welfare outcomes as factors influencing decisions about neutering [38]. Targeted, evidence-based communication may therefore help owners make informed reproductive-management decisions, although the reasons underlying the observed association with education were not directly assessed in the present study.
It is worth noting that urban respondents more frequently kept a single animal than rural respondents (53.3% vs. 26.9%); therefore, the “all animals” criterion was easier to meet in urban households because fewer animals had to be microchipped, vaccinated or neutered. The observed rural–urban differences in these practices may thus partly reflect differences in the number of animals kept per household rather than residence alone. Furthermore, because ownership and care practices were reported at the household rather than species level, the contribution of household species composition to the observed rural–urban differences could not be determined.
Companion animals were acquired predominantly through informal routes. Overall, 69.3% of acquisition-source responses represented animals taken in as strays, gifted by acquaintances or family, obtained through advertisements, born to already-owned animals, purchased from unregistered breeders or acquired from other private individuals. Similar reliance on gifts, community networks, private transfers and strays has been reported in other countries [39,40,41,42]. Some of these routes may involve unplanned litters or transfers occurring outside formal identification and registration procedures. However, the present study did not link individual acquisition sources with subsequent microchipping, registration or neutering, and therefore cannot establish that informally acquired animals received less consistent care. The predominance of informal acquisition should instead be interpreted as identifying a potentially important point for population-management interventions, including microchipping before transfer, accurate recording of ownership changes and prevention of unplanned litters [7,43].
Nearly half of respondents overall did not perceive a need to visit a veterinarian or to do so more frequently. This finding may be consistent with a more reactive approach to veterinary care, in which professional advice is sought mainly when owners perceive an immediate health problem. Regular veterinary care can provide opportunities not only for vaccination and early disease detection, but also for evidence-based guidance on nutrition, behaviour, environmental enrichment, stress and other aspects of physical and psychological welfare [8,9,10,11]. When veterinary care is not sought regularly, opportunities for owner education and early recognition of less visible welfare concerns may be reduced. Increasing service availability alone may therefore be insufficient unless owners are also informed about the veterinarian’s role in maintaining health and welfare before overt disease develops [8,44].
Husbandry practices also differed between residential settings and revealed several concerning patterns in rural areas. Animals in rural households were less often kept exclusively indoors and were more commonly allowed unrestricted outdoor access, housed in outdoor kennels or tethered. Tethering is of particular welfare concern because prolonged restraint may restrict exercise, behavioural choice and social contact while increasing the risk of frustration, injury and exposure to adverse environmental conditions [45]. These concerns are increasingly reflected in European legislation: Austria prohibits keeping dogs on a chain or otherwise tethered, apart from short-term tethering in front of a shop, while the EU regulation adopted in 2026 will, once applicable, prohibit operators from tethering dogs or cats for more than one hour, except during medical treatment or participation in shows, exhibitions and competitions; limited exemptions may also apply to dogs intended for military, police or customs services [46,47]. Feeding practices also differed, with urban respondents more frequently reporting premium or super-premium commercial diets, whereas rural respondents more often reported supermarket pet food, table scraps, bones and porridge. Reliance on home-prepared food and table scraps has also been reported in other companion animal populations [48,49]. Appropriate nutrition is a core responsible ownership obligation, as owners must provide a complete and balanced diet suited to the animal’s species, life stage and health status. When table foods or unformulated home-prepared foods constitute a substantial part of the diet, they may dilute essential nutrients, provide excessive energy or expose animals to other nutritional risks [48,50,51]. These findings indicate that responsible ownership interventions in rural communities should address not only preventive healthcare but also everyday housing, freedom of movement, social opportunities and evidence-based feeding practices.
The direct welfare assessment identified comparatively higher mean indicator scores for enrichment, body condition, handling during assessment, fears and anxieties frequency, and choice, control and predictability. Because no validated cut-off exists for classifying the study-specific adapted overall score as favourable or unfavourable, these findings should be interpreted comparatively rather than as evidence that overall welfare was generally favourable. The results identify nutritional, behavioural and environmental aspects that may require attention, but the study did not assess owners’ recognition or understanding of these needs. The body-condition result is relevant to previous evidence that owners may underestimate excess body condition, while the environmental and behavioural findings are consistent with reported limitations in owners’ knowledge of feline behaviour, environmental requirements and stress-related signals [12,13,52]. Regular veterinary guidance is therefore important not only for disease prevention, but also for weight management, environmental enrichment and behavioural welfare [9,50,53]. Contrary to our hypothesis, within the rural welfare-assessment sample, the overall welfare score was not significantly associated with owner age, education, gender or employment status. Older animals received poorer overall welfare scores. Chronic disease, pain and age-related functional or behavioural changes may contribute to poorer welfare in senior animals; however, these factors were not directly assessed as explanatory variables in the present study, and their contribution to the observed association cannot be determined. Poorer welfare is not an inevitable consequence of aging, as early detection, preventive care and age-appropriate management can help preserve quality of life [54]. As the dog-specific AWAG was preliminarily validated for dogs, its adaptation for cats in the present study should be considered when interpreting the combined-species results [28].
Taken together, these findings show that the owner–animal relationship is expressed not only through the decision to keep an animal, but also through the everyday responsibilities assumed for its health, traceability, reproduction, housing, nutrition and behavioural welfare. In rural settings, the combination of larger household animal populations, lower compliance with microchipping and vaccination requirements, weaker reproductive control, informal transfers, limited engagement with preventive veterinary care and concerning husbandry practices may amplify both welfare and population-management consequences. This pattern indicates that responsible ownership represents an important practical dimension of the human–animal relationship and cannot be inferred from ownership alone. Effective responses should therefore combine legal accountability, accessible preventive services, traceable ownership transfer, reproductive control and practical education addressing both physical and psychological needs. Similar challenges may be relevant across other rural contexts, particularly where companion animals are numerous but formal ownership and preventive-care systems remain inconsistently applied.
Study Limitations
Several limitations should be acknowledged. The convenience-sampling design does not permit causal conclusions and restricts the generalisability of the findings. The online questionnaire was distributed through open social-media posts, whereas in-person participants were recruited during microchipping and neutering campaigns and through requests for veterinary assistance. Both recruitment routes may have resulted in self-selection. The smaller urban comparison group reduced the precision of urban estimates and the statistical power to detect smaller rural–urban differences; therefore, non-significant comparisons should not be interpreted as evidence of equivalence. Rural–urban differences should be interpreted as comparisons within the study sample rather than as nationally representative estimates. In addition, the use of an online self-administered questionnaire and an assisted in-person questionnaire may have introduced survey-mode effects. Residence was classified using self-reported locality information or broad residential descriptions, and some misclassification may therefore have occurred.
Most ownership and husbandry practices were self-reported at household level and were not independently verified. Individual acquisition routes could not be linked to specific animals or their subsequent care, while the housing and feeding variables permitted multiple responses and were analysed descriptively. The direct welfare assessment included 85 animals from 53 rural households recruited through community initiatives and requests for assistance. This non-random subsample was intended to identify welfare priorities and cannot provide population-level prevalence estimates. Welfare was assessed at a single time point, so temporal variation could not be examined. Behavioural responses observed during the assessment, including defensive responses towards the unfamiliar assessor, may reflect the immediate assessment context rather than the animal’s usual behaviour or longer-term welfare state.
The AWAG was previously preliminarily validated for dogs and was adapted for cats in the present study. Although species-specific behavioural expressions and needs were considered during its administration, the adapted assessment did not undergo independent feline-specific expert review or formal validation. This limits interpretation of the combined-species findings and direct comparisons between dogs and cats. Moreover, the relatively small welfare-assessment sample, comprising 34 dogs and 51 cats, was not sufficient to support reliable species-specific comparisons. Future studies should use larger species-stratified samples to examine differences in the welfare needs and outcomes of dogs and cats.
5. Conclusions
This study highlights a mismatch between the scale of companion animal ownership and the consistent implementation of responsible ownership practices in rural Lithuania. The findings suggest that ownership alone does not ensure that animals’ health, behavioural and welfare needs are adequately met, particularly where larger numbers of animals are kept and ownership transfers remain informal.
Improving companion animal welfare in rural areas will require better microchipping and vaccination coverage, reproductive control and more regular preventive veterinary care. Owners also need practical guidance on nutrition, behaviour and animals’ psychological needs. Similar challenges may be relevant in other rural Baltic and Eastern European settings, but comparative research is needed to reveal how widely these patterns apply.
Acknowledgments
The authors thank all companion animal owners who participated in the study and permitted the examination of their animals; the volunteers of the Miško trobelė animal shelter for their assistance in organising and carrying out the community activities during which data collection took place; and Laurynas Pašakarnis for his guidance on data curation and statistical analysis. Product details for the generative artificial intelligence tools disclosed in Section 2.8 are as follows: ChatGPT (OpenAI; GPT-5.5 and GPT-5.6 Thinking; accessed in 2026) and Perplexity AI (Perplexity; accessed in 2026).
Abbreviations
The following abbreviations are used in this manuscript:
| AWAG | Animal Welfare Assessment Grid |
| OECD | Organisation for Economic Co-operation and Development |
Supplementary Materials
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ani16192988/s1, Supplementary Material S1. English Translation of the Online Questionnaire; Supplementary Material S2. English Translation of the In-Person Owner Questionnaire. Supplementary Material S3. Adapted Animal Welfare Assessment Form Based on the Animal Welfare Assessment Grid (AWAG).
[内容请查看原文]
Author Contributions
Conceptualization, E.M.R., A.M. and T.Š.; methodology, E.M.R., A.M. and S.M.; formal analysis, E.M.R. and A.M.; investigation, E.M.R.; data curation, E.M.R. and A.M.; visualization, E.M.R. and S.M.; writing—original draft preparation, E.M.R.; writing—review and editing, E.M.R., A.M., S.M. and T.Š.; supervision, A.M. All authors have read and agreed to the published version of the manuscript.
Institutional Review Board Statement
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Bioethics Centre of the Lithuanian University of Health Sciences (approval no. 2025-BEC2-0990; 26 September 2025).
Informed Consent Statement
Informed consent was obtained from all participants involved in the study. For the anonymous online survey, submission of the questionnaire after presentation of the study information constituted informed consent. Participants in the in-person component provided informed consent and permission for the examination and participation of their animals before data collection.
Data Availability Statement
The anonymized data supporting the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available because they contain detailed locality- and household-level information that may increase the risk of participant re-identification.
Conflicts of Interest
E.M.R. is the founder and director of the Miško Trobelė Animal Shelter, which facilitated access to the community initiatives during which in-person survey and animal welfare assessment participants were recruited. Participation was entirely voluntary, and the decision to participate, decline participation, or withdraw had no effect on access to or receipt of any services or assistance provided through these initiatives or by the shelter. This role is disclosed for transparency. The authors declare no commercial or financial conflicts of interest.
Funding Statement
This research received no external funding.
Footnotes
[内容请查看原文]
References
[内容请查看原文]
Associated Data
Supplementary Materials
[内容请查看原文]
Data Availability Statement
The anonymized data supporting the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available because they contain detailed locality- and household-level information that may increase the risk of participant re-identification.