兽研国民幸福感不仅与社会经济因素相关，也与生物和生态因素相关。这项生态相关性研究考察了93个国家国民幸福感、猫密度、刚地弓形虫暴露与社会经济指标之间的关联。我们假设，社会经济指标（如GDP、卫生支出、收入水平）较高的国家会表现出更高的幸福感水平。相反，我们预测较高的刚地弓形虫暴露（包括急性IgM和慢性IgG）与国民幸福感呈负相关。幸福感指标、猫密度、人均GDP、收入分类、卫生支出以及孕妇刚地弓形虫血清阳性率（作为更广泛人群暴露的替代指标）的数据均来自公开可获取的来源。与我们的假设一致，我们发现国民幸福感与猫密度及部分社会经济因素呈正相关，与IgM血清阳性率呈负相关。在多变量分析中，每万名居民中的猫数量与IgM血清阳性率仍是最强的独立预测因素。总体而言，这些探索性分析表明，生物和生态因素与社会经济变量共同影响国家福祉。未来国家层面的幸福感研究应整合社会经济与生物学视角，并纳入纵向和跨学科方法，以阐明其潜在机制和过程。
引言
主观幸福感通常通过生活满意度和幸福感指标进行评估，其作为人类和社会发展核心指标的重要性日益增加[1,2]。虽然一些经济指标（如人均国内生产总值（GDP）、家庭收入或卫生支出）已知与福祉呈正相关[例如3,4]，但它们不足以完全解释各国幸福感水平的差异[5–8]。例如，既往研究表明，孤独感与低幸福感相关，尤其是在单身、离异或无子女的个体中，或在失去家庭成员或其他社会支持来源的老年人中[9–13]。而特定人群、社会或国家中孤独感的患病率与社会经济指标仅呈弱相关[10]。这促使人们日益关注理解社会经济指标如何与社会心理因素相互作用，从而在人群层面影响主观幸福感[例如7,10,12,14,15]。
伴侣动物饲养被认为是心理健康的重要促进因素[16–18]。例如，既往研究提示，宠物（尤其是猫和狗）的存在可以减轻孤独感、促进情感联结、减轻压力并改善情绪[16–21]。此外，在许多社会中，宠物常被视为家庭成员，其照护反映了与共情、休闲和关系福祉相关的文化价值观[22]。因此，一个社会或国家内伴侣动物的数量可能作为生活方式质量及与动物相关的社会规范的社会行为学指标。
然而，饲养宠物，尤其是猫，也带来一些健康风险。例如，家猫（Felis catus）是神经寄生虫刚地弓形虫（T. gondii）的终末宿主，这是一种全球广泛分布的原虫，能够感染几乎所有温血动物，包括人类[23–26]。在全球范围内，估计任何特定人群中有约10–30%的成人感染刚地弓形虫，且往往毫不知情[25,27–29]。人类感染主要通过摄入猫粪便排出的卵囊、食用未煮熟的受污染肉类，以及可能与感染者的性接触而发生[23,30–32，综述见33]。在这方面，越来越多的研究表明，潜伏性弓形虫感染可能与精神病理状况的发生相关，如精神分裂症、强迫症、自闭症谱系障碍、焦虑障碍、抑郁症和人格障碍[25,30,34–37]，其途径可能是感染的直接效应，或与慢性应激、神经递质失调和慢性神经炎症相关的中介通路[37,38]。
已知大多数刚地弓形虫感染在健康个体中无临床症状；然而，血清学标志物有助于区分近期与既往暴露。IgM抗体提示急性或近期感染，与可能影响情绪和认知的炎症相关；而IgG抗体则提示潜伏性或慢性感染，后者与多种精神病理状况相关[30,39–41]。刚地弓形虫可在较大范围内对认知、生理、行为和情绪调节产生不良影响[例如26,31,33,42]，这一事实引发了关于其在大范围聚合指标（如国民幸福感）中潜在作用的重要问题[43,44]。
家猫可能为其主人提供心理和情感上的益处，但同时也作为病原体的传播媒介而带来负面的心理健康结局，这一观点代表了一个引人注目的生物心理学悖论，其对各国的福祉和总体健康的影响取决于卫生基础设施差异、宠物管理和公共卫生系统等因素[45,46]。有理由假设，在较富裕的国家，猫更可能被饲养在室内、接种疫苗并喂食工业宠物食品，从而最大限度地减少环境污染和人畜共患传播的机会[47]；而在低收入和中等收入国家，自由放养的猫可能对社区层面的寄生虫暴露构成更大风险[48]。因此，猫对人类福祉的生态影响可能不仅取决于它们的存在，还取决于人类与非人类动物互动的条件；这些条件由国家财富、政策和文华等指标所塑造[如47,49,50]。
尽管这些变量之间存在直观的联系，但据我们所知，此前尚无研究在大量国家中同时考察相对猫种群数量、弓形虫暴露、经济指标与国家幸福感之间的关联。为填补这一空白，本研究旨在探讨每10,000名居民中的猫数量（猫密度）、弓形虫IgM和IgG血清阳性率、GDP、收入分类和卫生支出与93个国家国家幸福感之间的关系。这一生态学相关性框架提供了一种跨学科方法，整合了心理社会、寄生虫学、经济和公共卫生指标，以探讨生物学和生态学风险如何可能被社会经济因素放大或缓冲。
本研究检验的主要假设是，经济指标较高（即GDP、卫生支出、收入水平）的国家将表现出更高水平的幸福感。相反，我们预测较高的弓形虫血清阳性率，无论是慢性（IgG）还是近期（IgM），将与国家幸福感呈负相关。作为探索性分析，我们进一步假设猫密度可能与国家幸福感呈正相关，但这种关联通过弓形虫血清阳性率的中介效应而被削弱甚至逆转，提示寄生虫感染在塑造猫的存在与国家福祉之间关系方面可能与经济因素同等重要。通过考察各国这一系列变量，本研究旨在提供新的概念和实证见解，以阐明社会经济指标、宠物和寄生虫暴露如何相互作用以影响全球健康与幸福感模式。
方法
数据汇编
使用公开可用的来源汇编了93个国家（图1）的国家幸福感水平、猫密度、社会经济指标以及弓形虫血清阳性率（IgG和IgM）数据。研究变量的数据来源和时间覆盖范围总结于表1。最终国家数量取决于所有感兴趣变量完整数据的可获得性，以确保整个分析过程中的可比性。基于G*Power事后分析，我们的样本量（N = 93，α = 0.05）具有85%的效力检测中等效应量（r = 0.3）。然而，本研究效力不足以可靠检测小效应（如r < 0.20）。因此，非显著关联应谨慎解读，因为缺乏统计学显著性并不一定表明不存在真实关联。
Table 1. Data sources and temporal coverage of the study variables.
| Variable | Source | Time period |
|---|---|---|
| Happiness scores | World Population Review (based on World Happiness Report / Gallup World Poll) | 3-year average: 2022–2024 |
| Cat density (cats per 10,000 inhabitants) | Cat Population by Country 2025 dataset (worldstats.com/World Population Review) | 2025 estimates (projections based on compilation data) |
| GDP per capita (current US$) | World Bank database | Most recent available year (typically 2023–2024, depending on country reporting) |
| Health expenditure (% of GDP) | WHO Global Health Expenditure Database | Primarily 2021–2023 (latest update includes 2023 data) |
| Country income classification | World Bank 2025 classifications | Based on GNI per capita for 2023 or 2024 (FY2025/FY2026 classifications) |
| T. gondii IgM and IgG seroprevalence (in pregnant women) | Bigna et al. (2020) | Individual studies: 1976–2017 (published 1979–2018); meta-analysis published 2020 |
[图：紫色所示国家为纳入本研究的国家。注：使用mapchart.net在CC BY 4.0许可下创建]
幸福感指数
幸福感得分来自世界人口综述网站（https://worldpopulationreview.com/country-rankings/happiest-countries-in-the-world），基于其2025年《世界最幸福国家》报告。这些排名反映三年平均值（即2022–2024年），通过分析盖洛普在149个国家开展的全面民意调查数据得出，涵盖六个特定类别：人均GDP、社会支持、健康预期寿命、自主选择生活的自由、普通民众的慷慨程度以及对内外腐败程度的感知（详见世界人口综述网站）。本研究使用了93个国家的数据，这些国家在所有其他变量（见下文）上均有完整信息。
猫密度
各国家猫种群估计数据来自《2025年各国家猫种群》数据集（https://worldstats.com/country-stats/cat-population-by-country）。为便于跨国比较，各国家猫数量按人口规模进行标准化。我们通过将各国猫总数除以总人口数（来源：https://data.worldbank.org/indicators/SP.POP.TOTL），再将结果乘以10,000，计算出每万名居民中的猫数量。这种标准化使得针对各国人口规模差异的生态学比较更具可解释性[51]。
国内生产总值
人均国内生产总值（以美元计）取自世界银行2025年数据库（https://data.worldbank.org/indicator/NY.GDP.PCAP.CD）。该指标表示人均经济产出，常被用作国家财富的标志。
卫生支出
当前卫生支出数据（占GDP的百分比）取自世界卫生组织全球卫生支出数据库（https://apps.who.int/gho/data/view.main.GHEDCHEGDPSHA2011v?lang=en），该数据库于2023年更新。该指标表示与卫生相关的总支出（包括公共和私人支出），不包括水和环境卫生。以占GDP的百分比报告支出，便于比较各国卫生优先级相对于经济产出的情况，常用于全球卫生研究[52]。
国家收入分类
国家收入组别依据世界银行2025年收入分类标准（https://datahelpdesk.worldbank.org/knowlegebase/articles/906519），该标准基于按阿特拉斯法计算的人均国民总收入。据此，各国被分类为：1=低收入，2=中低收入，3=中高收入，4=高收入。
IgM和IgG血清阳性率
由于本研究纳入的93个国家中，普通人群的弓形虫血清阳性率综合数据稀缺，我们采用Bigna等[28]报告的孕妇血清阳性率数据作为社区暴露水平的替代指标。使用这些特定类别可确保方法学与国际基准一致，便于与最新的同行验证流行病学模型进行直接比较。该方法得到既往研究的支持，这些研究表明孕妇血清阳性率密切反映更广泛人群的暴露水平[如24,27]，Bigna等[28]也指出了这一相关性。尽管该方法仅关注孕妇可能引入人口学偏倚，但由于先天性传播而对这一人群进行弓形虫常规筛查，提供了稳健且标准化的数据集，适用于跨国比较[28]。关于使用该替代指标的进一步讨论和意义，见局限性部分。
如Bigna等[28]所报告，国家IgG血清阳性率分为五个水平：1=<10.0%，2=10.0–19.99%，3=20.0–29.9%，4=30.0–49.99%，5=>50.0%。同样，IgM血清阳性率分类如下：1=<0.50%，2=0.5–0.99%，3=1.0–1.99%，4=2.0–4.99%，5=>5.0%。
统计分析
在主要分析之前，我们对非正态分布的数据进行对数转换以改善正态性。计算Pearson相关系数以评估变量间的双变量关联。在探索性分析中，使用线性回归模型检验猫密度、卫生支出、GDP、收入和弓形虫血清阳性率作为自变量对国家幸福感得分（因变量）的影响。基于回归结果，随后进行中介分析（5000次Bootstrap抽样），引入IgM血清阳性率作为猫密度与国家幸福感之间关联的中介变量。由于本研究为探索性研究，我们报告未校正的p值，并侧重于效应量而非统计显著性，以尽量减少忽略潜在有意义关联的风险[53,54]。所有分析使用Jamovi（版本2.3.28，[55]）和JASP（版本0.95.3.0，[56]）完成。所有检验均为双尾，统计显著性设定为p≤0.05。
需要强调的是，由于本研究提供的数据集由有限数量的国家组成，这些国家是基于数据可获得性而非随机抽样选出的，因此我们在此将推断统计量（如p值）解释为基于模型和基于设计的不确定性指标，而非基于抽样对更大总体的推断[57,58]。该方法与近期利用假设检验评估跨国和生态学分析中观察到的关联的稳健性和内部一致性的研究相一致[53,57]。
结果
描述性分析
所有变量的描述性数据见表2。研究变量不满足正态性假设，因此进行了对数转换。
Table 2. Descriptive statistics of the study variables.
| Variables | Mean (SD) | 95%CI | Minimum | Maximum | Shapiro-Wilk | |
|---|---|---|---|---|---|---|
| W | p | |||||
| IgM | 2.88 (1.284) | 2.62-3.15 | 1 | 5 | 0.904 | < 0.001 |
| IgG | 3.33 (1.210) | 3.08-3.58 | 1 | 5 | 0.882 | < 0.001 |
| GDP | 24,047.68 (27,022.096) | 18,482.55-29,612.82 | 415.70 | 128,678.20 | 0.803 | < 0.001 |
| Income group | 3.14 (0.904) | 2.95-3.33 | 1 | 4 | 0.802 | < 0.001 |
| Health expenditure | 7.82 (3.306) | 7.14-8.50 | 2.36 | 21.83 | 0.937 | < 0.001 |
| Cat density | 989.28 (769.516) | 830.80-1,147.76 | 67.35 | 3,245.80 | 0.915 | < 0.001 |
| Happiness | 5.72 (1.118) | 5.48-5.97 | 1.72 | 7.74 | 0.956 | 0.003 |
相关性
使用Spearman's rho和Pearson's r进行的相关分析得出一致的结果，并揭示了变量之间的多重关联。在此，我们报告参数检验结果（见图2）。两种方法的完整结果汇总于补充材料的S1表中。简言之，国家幸福感得分与猫密度（r = 0.620，p < 0.001，95%CI = 0.480，0.733）、卫生支出（r = 0.350，p < 0.001，95%CI = 0.162，0.521）、收入水平（r = 0.780，p < 0.001，95%CI = 0.686，0.849）和GDP（r = 0.800，p < 0.001，95%CI = 0.718，0.866）呈中至高度正相关。相反，幸福感得分与刚地弓形虫IgM血清阳性率（r = −0.540，p < 0.001，95%CI = −0.672，−0.380）和IgG血清阳性率（r = −0.330，p = 0.001，95%CI = −0.504，−0.141）呈中至高度负相关。猫密度与收入（r = 0.770，p < 0.001，95%CI = 0.676，0.844）、GDP（r = 0.790，p < 0.001，95%CI = 0.699，0.856）和卫生支出（r = 0.600，p < 0.001，95%CI = 0.448，0.714）呈高度正相关。还观察到猫密度与IgG血清阳性率之间存在小幅度负相关（r = −0.200，p = 0.058，95%CI = −0.385，0.007）。收入水平与GDP呈高度正相关（r = 0.930，p < 0.001，95%CI = 0.895，0.953）。收入水平与IgM血清阳性率呈高度负相关（r = −0.530，p < 0.001，95%CI = −0.661，−0.364），与IgG血清阳性率呈中度负相关（r = −0.320，p = 0.002，95%CI = −0.493，−0.126）。GDP与IgM血清阳性率呈高度负相关（r = −0.590，p < 0.001，95%CI = −0.710，−0.442），与IgG血清阳性率呈中度负相关（r = −0.330，p = 0.001，95%CI = −0.502，−0.138）。收入水平和GDP均与幸福感、猫密度和卫生支出呈高度正相关（所有情况下r > 0.570，p < 0.001）。IgM和IgG血清阳性率呈中度正相关（r = 0.460，p < 0.001，95%CI = 0.282，0.606），且两者均与社会经济指标和幸福感水平呈中至高度负相关（所有情况下r < −0.182，p < 0.076）。这些发现支持了我们的假设，即较高水平的财富、公共投资和猫密度与较低的刚地弓形虫暴露和较高的国家幸福感相关。探索性收入组分层分析见补充材料S1和S2图。
[图：显示研究变量之间相关性的热图。颜色越深表示相关性越强，紫色]
探索性预测与中介分析
1. 由于GDP与收入之间存在高度多重共线性（VIF分别为7.61和7.41），这些变量被排除在探索性回归分析之外。虽然残差显示非正态分布（Shapiro-Wilk p < 0.05），但鉴于样本量（N = 93）以及线性和方差齐性的维持，模型对该偏差具有稳健性。卫生支出被保留作为主要社会经济指标，因为它代表了国家财富更直接的代理变量（例如，因为它与公共卫生基础设施相关）。因此，仅将猫密度、卫生支出以及T. gondii IgM和IgG血清阳性率纳入作为预测变量（所有情况下VIF < 1.71）。我们的模型解释了国家幸福感47.0%的方差（R2 = 0.470，F(4,88) = 19.500，p < 0.001），并显示猫密度正向（β = 0.520，p < 0.001）且IgM负向（β = −0.240，p = 0.021）显著贡献于幸福感评分的预测。卫生支出和IgG血清阳性率与幸福感的关系发现为小且不显著的关联（表3）。我们还进行了回归分析，将IgM和IgG血清阳性率作为分类预测变量使用虚拟编码而非连续预测变量。结果模式在定性上保持一致（补充材料中的S2表），表明所报告的关联并非由血清阳性率类别之间等间距的假设所驱动。探索性中介分析（图3）显示，猫密度对国家幸福感有正向直接效应，且该效应被IgM血清阳性率负向中介（表4）。总体而言，我们的发现一致地显示了国家幸福感、猫密度、T. gondii暴露和社会经济因素之间关联的强度和内部一致性。此处的中介分析旨在对变量间关联进行描述性统计分解，并不意味着因果顺序或机制通路。
Table 3. Results from the regression analysis.
| Variables | b | SE | t | p | 95%CI |
|---|---|---|---|---|---|
| Constant | 0.559 | 0.074 | 7.506 | < 0.001 | 0.411, 0.707 |
| Health Expenditure | −0.042 | 0.054 | −0.784 | 0.435 | −0.150, 0.065 |
| Cat density | 0.114 | 0.023 | 4.905 | < 0.001** | 0.068, 0.161 |
| IgM | −0.019 | 0.008 | −2.358 | 0.021* | −0.037, −0.003 |
| IgG | −0.012 | 0.007 | −1.567 | 0.121 | −0.027, 0.003 |
Table 4. Path estimates of the mediation analysis.
| 95% Confidence Interval | |||||||
|---|---|---|---|---|---|---|---|
| Estimate | Lower | Upper | Z | p | |||
| Cat density | → | IgM | −1.4060 | −1.8466 | −0.9765 | −6.29 | <.001 |
| IgM | → | Happiness | −0.0247 | −0.0389 | −0.0112 | −3.52 | <.001 |
| Cat density | → | Happiness | 0.1040 | 0.0549 | 0.1616 | 3.87 | <.001 |
[图：中介分析的效应，代表猫密度与IgM血清阳性率之间关联的统计分解]
讨论
与我们的假设一致，我们的发现显示猫密度和一些社会经济指标与幸福感呈中等到较大的正相关，而T. gondii IgM和IgG血清阳性率与幸福感呈中等到较大的负相关。这些结果强调了人-动物互动、传染病生态学和国家福祉之间复杂但尚未充分研究的关系。
观察到的猫密度与幸福感之间的正相关与先前显示宠物饲养心理和生理益处的研究一致[例如，18–21]，包括减少孤独感和改善情绪健康[16,17,59]。猫密度是一个聚合指标，因此是宠物饲养的不完美代理变量，不能捕捉人-猫互动的质量或背景。尽管如此，这些关联表明，即使在粗略的生态尺度上检查，人-动物关系也可能解释主观幸福感的人群水平差异。因此，猫饲养、减少孤独感和缓解压力之间的联系可能解释猫密度与更高国家幸福感之间的关联，而这反过来可能反映对宠物更有利的态度、更高的可支配收入和更好的生活条件，所有这些都与更高的主观幸福感密切相关[1,3,59]。
本研究一个有趣且反直觉的发现是猫密度与T. gondii血清阳性率之间中等到较大的负相关，挑战了猫数量越多意味着更多寄生虫暴露的常见假设[例如，46]。对这一发现的一种解释是，在较富裕的国家，猫更可能被养在室内、喂食加工食品并接受常规医疗护理，显著降低了T. gondii感染的可能性。相反，在低收入和中等收入国家，猫通常自由放养，增加了T. gondii感染的几率[27,44,47]。
值得注意的是，虽然无法从这些生态相关性中推断因果关系，但中等至较大的关联进一步支持了越来越多的证据表明弓形虫暴露与负面神经心理结局相关[33,60]。例如，慢性弓形虫感染与人格的细微变化以及抑郁、焦虑、精神分裂症和其他精神障碍的发生相关[23–25,29,30,35–37,40–42]，而急性感染可能导致免疫学改变，进而影响情绪和认知功能[25,39]。弓形虫感染对心理过程的影响可能是通过这种神经寄生虫作用于大脑神经递质以及参与情绪调节和认知过程的特定回路所介导的[61,62]。
需要强调的是，本文报告的所有关联，包括来自中介模型的关联，都应解释为统计关系而非因果通路的证据。虽然生态相关性有助于识别宏观模式和生成假设，但无法厘清方向性或排除未测量的混杂过程[61,62]。因此，中介效应是谨慎讨论的，并被界定为间接统计关联而非因果效应。同样重要的是要强调，我们的发现描述的是国家层面的关联，不应被用来推断与养猫、幸福感和弓形虫暴露相关的个体层面效应或个人风险。
从更广泛和更一般的角度来看，GDP、收入水平、卫生支出与幸福感之间的大幅关联支持了健康的社会决定因素框架[63]，根据该框架，医疗保健、教育和环境安全已知可减少传染性病原体的暴露并增强心理韧性[64,65]。我们的发现扩展了这一框架，表明养猫和寄生虫暴露——这些在幸福感研究中通常被忽视的因素——可以在国家层面影响幸福感。我们的探索性回归和中介分析结果进一步阐明了相关性分析中出现的关联。具体而言，这些分析显示猫密度和IgM血清阳性率与国家幸福感具有最大的关联，表明生物学和生态因素可能与既有的社会经济驱动因素共同促进国家福祉。此外，我们的发现有可能为旨在减轻弓形虫暴露负面影响、促进负责任的宠物饲养以及提高国家层面福祉的公共卫生政策提供参考。
局限性与优势
重要的是要认识到我们的研究存在若干局限性。首先，从孕妇获得的弓形虫感染血清阳性率数据虽然在流行病学和生态学研究中被广泛采用[例如，27,28]，但可能不能完全代表更广泛的人群，从而在我们的研究中引入潜在的人口学偏倚。孕妇在各国之间的医疗保健可及性、检测方案和暴露风险方面可能存在差异。然而，在缺乏全面国家监测规划的情况下，这些数据代表了全球比较中最稳健和标准化的来源。同样，猫密度测量将宠物猫、流浪猫和野猫种群混为一谈，它们可能对人类心理益处和寄生虫传播过程产生不同的影响。因此，未来研究应区分这些猫种群并考虑文化差异，以更精确地分离它们对弓形虫暴露和国家幸福感的具体影响。其次，我们未考虑若干混杂变量，如城市化水平、气候指标或卫生基础设施，这些也可能影响养猫、寄生虫暴露和主观幸福感。例如，在卫生基础设施较好的城市化地区，流浪猫可能较少，从而限制了弓形虫向其他动物的传播[66]。此外，湿度和温度等气候指标可能影响卵囊的孢子化[67]，这也可能影响与弓形虫感染率、人群健康、养猫以及最终国家福祉相关的过程。
第三，本研究的横断面生态数据和相关性设计使我们无法推断因果关系。因此，在探索性预测和中介分析中观察到的模式应被解释为关联的统计分解。替代解释，包括反向因果和未测量的混杂变量，不能被排除，并可能解释部分观察到的关系。未来研究可能受益于使用预测性方法而非相关性方法来检验猫密度、弓形虫暴露和财富的社会经济指标对国家幸福感的因果效应。此外，由于世界幸福指数部分纳入了GDP等社会经济指标，结局测量与某些预测变量之间存在一定程度的概念重叠。
最后，数据质量和可获取性因国家而异，这在依赖二手数据的跨国生态学研究中很常见，可能引入研究变量之间的时间错位，并限制研究结果的普适性。尽管这种不匹配可能引入一些噪声或偏倚，但在缺乏重大公共卫生干预的情况下，刚地弓形虫的血清阳性率往往随时间缓慢变化。因此，任何潜在偏倚很可能是保守的，因为富裕国家暴露水平随时间下降可能会低估刚地弓形虫与国家幸福感之间的负相关。未来研究应使时间数据紧密对齐，以确认这些生态学模式的稳健性。
尽管存在这些局限性，我们的研究也有若干优势。首先，虽然既往研究已探讨过刚地弓形虫等寄生虫对经济和文化模式的潜在作用[如43,68]，但本研究首次在涵盖大量国家的全球幸福感分析中整合了生态学、生物学和社会经济指标。通过纳入养猫拥有率、寄生虫暴露和经济发展数据，本研究为幸福感的决定因素提供了新颖的跨学科视角。此外，同时纳入IgM和IgG血清阳性率使我们能够区分急性感染和慢性感染，提高了分析的生物学特异性。总体而言，我们的发现凸显了在研究和理解全球主观幸福感模式时考虑生物学和社会生态学因素的重要性。
结论
本生态学研究为93个国家中幸福感、猫密度、刚地弓形虫暴露与社会经济因素之间关联的幅度和稳健性提供了新见解。我们的主要发现表明，除了主观幸福感与某些社会经济因素之间的典型关联外，国家幸福感与猫密度在很大程度上呈正相关，与IgM刚地弓形虫暴露在很大程度上呈负相关。这些结果强调，未来研究需要整合幸福感的社经和生物学决定因素，并采用纵向和跨学科方法以阐明潜在机制和过程。
支持信息
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]
查看英文原文▾
National happiness is related not only to socioeconomic factors but also to biological and ecological factors. This ecological correlational study examined associations between national happiness, cat density, Toxoplasma gondii exposure, and socioeconomic indicators across 93 countries. We hypothesized that countries with higher socioeconomic indicators (e.g., GDP, health expenditure, income level) would exhibit higher happiness levels. In contrast, we predicted that higher T. gondii exposure, both acute (IgM) and chronic (IgG), would be negatively associated with national happiness. Data on happiness measures, cat density, GDP per capita, income classification, health expenditure, and T. gondii seroprevalence in pregnant women (as a proxy for broader population exposure) were obtained from publicly available sources. In line with our hypothesis, we found that national happiness was positively associated with cat density and some socioeconomic factors and negatively associated with IgM seroprevalence. In multivariate analyses, cats per 10,000 inhabitants and IgM seroprevalence remained the strongest independent predictors. Overall, these exploratory analyses suggest that biological and ecological factors are associated with socioeconomic variables in determining national well-being. Future studies of happiness at the national level should integrate both socioeconomic and biological perspectives and include longitudinal and interdisciplinary approaches to elucidate the underlying mechanisms and processes.
Introduction
Subjective well-being, commonly assessed through life satisfaction and happiness indicators, has increased its relevance as a central indicator of human and social development [1,2]. While some economic indicators such as gross domestic product (GDP) per capita, family income, or health expenditure are known to be positively associated with well-being [e.g., 3,4], they are not sufficient to completely explain the variability of happiness levels across countries [5–8]. For example, previous research has shown that loneliness is linked to low happiness, especially among individuals who are single, divorced, or without children, or among aging individuals who have lost family members or other sources of social support [9–13]. The prevalence of loneliness in a particular population, society, or country, in turn, has been weakly associated with socioeconomic indicators [10]. This has led to a growing interest in understanding how socioeconomic indicators interact with psychosocial factors to influence subjective well-being on a population scale [e.g., 7,10,12,14,15].
Companion animal ownership has been proposed to be an important contributor to psychological health [16–18]. For example, previous studies have suggested that the presence of pets, particularly cats and dogs, can reduce loneliness, foster emotional bonding, diminish stress, and improve mood [16–21]. Moreover, pets are often regarded as family members in many societies, and their care reflects cultural values related to empathy, leisure, and relational well-being [22]. Thus, the number of companion animals within a society or country might serve as a sociobehavioral indicator of lifestyle quality and social norms related to animals.
Pet ownership, however, especially cats, also entails some health risks. For instance, domestic cats (Felis catus) are the definitive hosts of the neuroparasite Toxoplasma gondii (T. gondii), a globally widespread protozoan capable of infecting nearly all warm-blooded animals, including humans [23–26]. Globally, approximately 10–30% of adults in any given human population are estimated to be infected with T. gondii, often without even knowing it [25,27–29]. Human infection occurs mainly through ingestion of oocysts excreted in cat feces, consumption of undercooked contaminated meat, and potentially, sexual intercourse with an infected person [23,30–32, see 33 for a review]. In this regard, a growing body of research has suggested that latent Toxoplasma infection may be associated with the development of psychopathological conditions such as schizophrenia, obsessive-compulsive disorder, autism spectrum disorders, anxiety disorders, depression, and personality disorders [25,30,34–37], either through direct effects of infection or through mediational pathways associated with chronic stress, neurotransmitter dysregulations, and chronic neuroinflammation [37,38].
Most T. gondii infections are known to be clinically asymptomatic in healthy individuals; however, serological markers may help distinguish between recent and past exposures. Whereas IgM antibodies indicate acute or recent infections associated with inflammation that could affect mood and cognition, IgG antibodies signal latent or chronic infections that have been linked to various psychopathologies [30,39–41]. The fact that T. gondii can have adverse effects on cognition, physiology, behavior and emotional regulation on a large scale [e.g., 26,31,33,42] raises important questions about its potential role in modulating aggregate measures such as national happiness [43,44].
The idea that domestic cats may provide psychological and emotional benefits to their owners but also serve as a vector for pathogens with negative mental health outcomes, represents a compelling biopsychological paradox, with implications for well-being and general health in different countries depending on factors such as disparities in sanitation infrastructure, pet management, and public health systems [45,46]. It is reasonable to assume that in wealthier countries, cats are more likely to be kept indoors, vaccinated, and fed industrial pet food, minimizing opportunities for environmental contamination and zoonotic transmission [47], whereas in low- and middle-income countries, free-roaming cats might represent a greater risk for community-level parasite exposure [48]. Thus, the ecological impact of cats on human well-being may depend not only on their presence, but also on the conditions under which humans and non-human animals interact; conditions that are shaped by national indicators such as wealth, policy, and culture [e.g., 47,49,50].
Despite the intuitive connections among these variables, to our knowledge, no prior study has simultaneously examined the associations between relative cat population, T. gondii exposure, economic indicators, and national happiness across a large set of countries. To address this gap, the present study aimed to explore how cats per 10,000 inhabitants (cat density), T. gondii IgM and IgG seroprevalence, GDP, income classification, and health expenditure are related to national happiness across 93 countries. This ecological correlational framework offers a transdisciplinary approach that integrates psychosocial, parasitological, economic, and public health indicators to explore how biological and ecological risks may be amplified or buffered by socioeconomic factors.
The primary hypothesis tested in this study is that countries with higher economic indicators (i.e., GDP, health expenditure, income level) will exhibit higher levels of happiness. In contrast, we predict that greater T. gondii seroprevalence, both chronic (IgG) and recent (IgM), will be negatively associated with national happiness. As an exploratory analysis, we further hypothesize that cat density may be positively associated with national happiness, but that this association is weakened or even reversed through the mediating effect of T. gondii seroprevalence, suggesting that parasite infection may be as important as economic factors in shaping how the presence of cats may relate to national well-being. By addressing this constellation of variables across countries, this study aims to provide new conceptual and empirical insights into how socioeconomic indicators, pets, and parasitic exposure interact to influence global patterns of health and happiness.
Methods
Data compilation
Data on national happiness levels, cat density, socioeconomic indicators, and T. gondii seroprevalence, both IgG and IgM, were compiled for 93 countries (Fig 1) using publicly available sources. Data sources and temporal coverage of the study variables are summarized in Table 1. The final number of countries was determined based on the availability of complete data across all variables of interest, ensuring comparability throughout the analysis. Based on a G*Power post-hoc analysis, our sample size (N = 93 and α = 0.05) had 85% power to detect moderate effect sizes (r = 0.3). However, the present study was underpowered to reliably detect small effects (e.g., r < 0.20). Therefore, the non-significant associations should be interpreted cautiously, as the absence of statistical significance does not necessarily indicate the absence of a true association.
Table 1. Data sources and temporal coverage of the study variables.
| Variable | Source | Time period |
|---|---|---|
| Happiness scores | World Population Review (based on World Happiness Report / Gallup World Poll) | 3-year average: 2022–2024 |
| Cat density (cats per 10,000 inhabitants) | Cat Population by Country 2025 dataset (worldstats.com/World Population Review) | 2025 estimates (projections based on compilation data) |
| GDP per capita (current US$) | World Bank database | Most recent available year (typically 2023–2024, depending on country reporting) |
| Health expenditure (% of GDP) | WHO Global Health Expenditure Database | Primarily 2021–2023 (latest update includes 2023 data) |
| Country income classification | World Bank 2025 classifications | Based on GNI per capita for 2023 or 2024 (FY2025/FY2026 classifications) |
| T. gondii IgM and IgG seroprevalence (in pregnant women) | Bigna et al. (2020) | Individual studies: 1976–2017 (published 1979–2018); meta-analysis published 2020 |
[图：Countries shown in purple represent those included in the study. Note: Created with mapchart.net under a CC BY 4.0 Licen]
Happiness index
Happiness scores were obtained from the World Population Review website (https://worldpopulationreview.com/country-rankings/happiest-countries-in-the-world), based on their 2025 report of the Happiest Countries in the World. These rankings reflect three-year averages (i.e., 2022–2024) and are determined by analyzing comprehensive Gallup polling data from 149 countries in six particular categories: GDP per capita, social support, healthy life expectancy, freedom to make your own life choices, generosity of the general population, and perceptions of internal and external corruption levels (see the World Population Review website for more details). For the purposes of this study, we used data from the 93 countries for which complete information on all other variables (see below) was available.
Cat density
Estimates of domestic cat populations per country were obtained from the Cat Population By Country 2025 dataset (https://worldstats.com/country-stats/cat-population-by-country). To allow cross-national comparisons, the number of domestic cats per country was standardized by population size. Thus, we calculated the number of cats per 10,000 inhabitants by dividing the total number of cats in each country by the total human population (sourced from: https://data.worldbank.org/indicators/SP.POP.TOTL), and then multiplying the result by 10,000. This standardization allows for more interpretable ecological comparisons for differences in population size across countries [51].
Gross Domestic Product
Gross Domestic Product per capita (in US dollars) was retrieved from the World Bank 2025 database (https://data.worldbank.org/indicator/NY.GDP.PCAP.CD). This measure indicates average economic productivity per person and is often used as a marker of national wealth.
Health expenditure
Data on current health expenditure (as percentage of GDP) were obtained from the World Health Organization’s Global Health Expenditure Database (https://apps.who.int/gho/data/view.main.GHEDCHEGDPSHA2011v?lang=en), updated in 2023. This measure indicates the total health-related spending, both public and private, excluding water and sanitation. Reporting expenditure as a percentage of GDP allows for comparisons of national health priorities relative to economic output and is frequently used in global health research [52].
Country income classification
Country income groups were classified according to the World Bank’s 2025 income classifications (https://datahelpdesk.worldbank.org/knowlegebase/articles/906519), which are based on gross national income per capita using the Atlas method. Accordingly, countries were categorized as 1 = low, 2 = lower-middle, 3 = upper-middle, or 4 = high income.
IgM and IgG seroprevalence
Due to the scarcity of comprehensive T. gondii seroprevalence data for the general population across the 93 countries included in this study, we employed the seroprevalence data from pregnant women reported in Bigna et al. [28] as a proxy for community-level exposure. The use of these specific categories ensures methodological alignment with international benchmarks, facilitating a direct comparison with the most recent peer-validated epidemiological models. This approach is supported by prior research showing that seroprevalence rates in pregnant women closely reflect broader population exposure levels [e.g., 24,27], a correlation also noted by Bigna et al. [28]. While this approach might introduce a demographic bias by focusing only on pregnant women, the routine screening of this population for T. gondii due to congenital transmission provides a robust and standardized dataset for cross-national comparisons [28]. See the Limitations section for further discussion and implications of the use of this proxy.
As reported in Bigna et al. [28], national IgG seroprevalence was categorized into five levels as follows: 1 = <10.0%, 2 = 10.0–19.99%, 3 = 20.0–29.9%, 4 = 30.0–49.99%, and 5 = >50.0%. Similarly, IgM seroprevalence was categorized as follows: 1 = <0.50%, 2 = 0.5–0.99%, 3 = 1.0–1.99%, 4 = 2.0–4.99%, 5 = >5.0%.
Statistical analysis
Prior to the main analyses, we log-transformed data that were not normally distributed to improve normality. Pearson’s correlation coefficients were calculated to evaluate bivariate associations among the variables. For an exploratory analysis, a linear regression model was used to examine the influence of cat density, health expenditure, GDP, income, and T. gondii seroprevalence, included as independent variables, on national happiness scores, included as the dependent variable. Based on the regression results, a mediation analysis (5000 bootstrap samples) was then performed introducing IgM seroprevalence as the mediator of the association between cat density and national happiness. As this is an exploratory study, we reported unadjusted p-values and focused on effect sizes over statistical significance to minimize the risk of overlooking potentially meaningful associations [53,54]. All analyses were performed using Jamovi (version 2.3.28, [55]) and JASP (version 0.95.3.0, [56]). All tests were two-tailed and statistical significance was set at p ≤ 0.05.
It is important to emphasize that since the dataset presented in this study comprises a finite number of countries chosen based on data availability rather than random sampling, our inferential statistics (e.g., p values) are interpreted here as indicators of model-based and design-based uncertainty, rather than as sampling-based inferences to a larger population [57,58]. This approach aligns with recent research using hypothesis-testing to assess the robustness and internal consistency of observed associations in cross-national and ecological analyses [53,57].
Results
Descriptive analysis
Descriptive data for all variables are shown in Table 2. The study variables did not satisfy the assumption of normality and therefore were log-transformed.
Table 2. Descriptive statistics of the study variables.
| Variables | Mean (SD) | 95%CI | Minimum | Maximum | Shapiro-Wilk | |
|---|---|---|---|---|---|---|
| W | p | |||||
| IgM | 2.88 (1.284) | 2.62-3.15 | 1 | 5 | 0.904 | < 0.001 |
| IgG | 3.33 (1.210) | 3.08-3.58 | 1 | 5 | 0.882 | < 0.001 |
| GDP | 24,047.68 (27,022.096) | 18,482.55-29,612.82 | 415.70 | 128,678.20 | 0.803 | < 0.001 |
| Income group | 3.14 (0.904) | 2.95-3.33 | 1 | 4 | 0.802 | < 0.001 |
| Health expenditure | 7.82 (3.306) | 7.14-8.50 | 2.36 | 21.83 | 0.937 | < 0.001 |
| Cat density | 989.28 (769.516) | 830.80-1,147.76 | 67.35 | 3,245.80 | 0.915 | < 0.001 |
| Happiness | 5.72 (1.118) | 5.48-5.97 | 1.72 | 7.74 | 0.956 | 0.003 |
Correlations
Correlation analyses using both Spearman’s rho and Pearson’s r yielded consistent outcomes and revealed multiple associations between the variables. Here, we report the parametric results (see Fig 2). Full results for both methods are summarized in S1 Table in Supplementary Materials. In brief, national happiness scores showed moderate-to-large and positive correlations with cat density (r = 0.620, p < 0.001, 95%CI = 0.480, 0.733), health expenditure (r = 0.350, p < 0.001, 95%CI = 0.162, 0.521), income level (r = 0.780, p < 0.001, 95%CI = 0.686, 0.849), and GDP (r = 0.800, p < 0.001, 95%CI = 0.718, 0.866). In contrast, happiness scores showed moderate-to-large and negative correlations with T. gondii IgM seroprevalence (r = −0.540, p < 0.001, 95%CI = −0.672, −0.380), and IgG seroprevalence (r = −0.330, p = 0.001, 95%CI = −0.504, −0.141). Cat density was largely and positively correlated with income (r = 0.770, p < 0.001, 95%CI = 0.676, 0.844), GDP (r = 0.790, p < 0.001, 95%CI = 0.699, 0.856), and health expenditure (r = 0.600, p < 0.001, 95%CI = 0.448, 0.714). A small and negative correlation between cat density and IgG seroprevalence was also observed (r = −0.200, p = 0.058, 95%CI = −0.385, 0.007). Income level and GDP showed a large and positive correlation (r = 0.930, p < 0.001, 95%CI = 0.895, 0.953). Income level was largely and negatively correlated with IgM seroprevalence (r = −0.530, p < 0.001, 95%CI = −0.661, −0.364) and moderately and negatively with IgG seroprevalence (r = −0.320, p = 0.002, 95%CI = −0.493, −0.126). GDP showed a large and negative correlation with IgM seroprevalence (r = −0.590, p < 0.001, 95% CI = −0.710, −0.442) and a moderate and negative correlation with IgG seroprevalence (r = −0.330, p = 0.001, 95%CI = −0.502, −0.138). Both income level and GDP were largely and positively correlated with happiness, cat density, and health expenditure (r > 0.570, p < 0.001, in all cases). IgM and IgG seroprevalence were moderately and positively correlated (r = 0.460, p < 0.001, 95%CI = 0.282, 0.606), and both were moderately-to-largely and negatively correlated with socioeconomic indicators and happiness levels (r < −0.182, p < 0.076, in all cases). These findings support our hypothesis that higher levels of wealth, public investment, and cat density are associated with lower T. gondii exposure and greater national happiness. See S1 and S2 Figs in Supplementary Materials for an exploratory income group-stratified analysis.
[图：Heatmap showing the correlation between the study variables. Darker shades represent stronger correlations, with purple ]
Exploratory prediction and mediation analyses
1. Due to high multicollinearity between GDP and income (VIFs = 7.61 and 7.41, respectively) these variables were excluded from the exploratory regression analysis. While residuals showed a non-normality distribution (Shapiro-Wilk p < 0.05), the model was robust to this deviation given the sample size (N = 93) and the maintenance of linearity and homoscedasticity. Health expenditure was retained as the primary socioeconomic indicator, as it represents a more direct proxy for national wealth (for example, because it relates to public health infrastructure). Therefore, only cat density, health expenditure, and T. gondii IgM and IgG seroprevalence were included as predictors (VIF < 1.71 in all cases). Our model accounted for 47.0% of the variance in national happiness (R2 = 0.470, F(4,88) = 19.500, p < 0.001) and showed that cat density positively (β = 0.520, p < 0.001) and IgM negatively (β = −0.240, p = 0.021) significantly contributed to the prediction of happiness scores. Small and non-significant associations were found for health expenditure and IgG seroprevalence and their relationship with happiness (Table 3). We also conducted the regression analyses treating IgM and IgG seroprevalence as categorical predictors using dummy coding instead of continuous predictors. The pattern of results remained qualitatively identical (S2 Table in Supplementary Materials), indicating that the reported associations were not driven by the assumption of equal spacing between seroprevalence categories. The exploratory mediation analysis (Fig 3) showed that cat density had a positive direct effect on national happiness, and that this effect was negatively mediated by IgM seroprevalence (Table 4). Overall, our findings consistently showed the strength and internal consistency of the associations between national happiness, cat density, T. gondii exposure and socioeconomic factors. Here the mediation analyses are intended as a descriptive statistical decomposition of associations among variables and do not imply causal ordering or mechanistic pathways.
Table 3. Results from the regression analysis.
| Variables | b | SE | t | p | 95%CI |
|---|---|---|---|---|---|
| Constant | 0.559 | 0.074 | 7.506 | < 0.001 | 0.411, 0.707 |
| Health Expenditure | −0.042 | 0.054 | −0.784 | 0.435 | −0.150, 0.065 |
| Cat density | 0.114 | 0.023 | 4.905 | < 0.001** | 0.068, 0.161 |
| IgM | −0.019 | 0.008 | −2.358 | 0.021* | −0.037, −0.003 |
| IgG | −0.012 | 0.007 | −1.567 | 0.121 | −0.027, 0.003 |
Table 4. Path estimates of the mediation analysis.
| 95% Confidence Interval | |||||||
|---|---|---|---|---|---|---|---|
| Estimate | Lower | Upper | Z | p | |||
| Cat density | → | IgM | −1.4060 | −1.8466 | −0.9765 | −6.29 | <.001 |
| IgM | → | Happiness | −0.0247 | −0.0389 | −0.0112 | −3.52 | <.001 |
| Cat density | → | Happiness | 0.1040 | 0.0549 | 0.1616 | 3.87 | <.001 |
[图：Effects of the mediation analysis representing a statistical decomposition of association between cat density, IgM serop]
Discussion
Consistent with our hypotheses, our findings showed that cat density and some socioeconomic indicators were moderately-to-largely and positively associated with happiness, whereas T. gondii IgM and IgG seroprevalence were moderately-to-largely and negatively associated with happiness. These results underscore a complex and yet understudied relationship between human-animal interactions, infectious disease ecology, and national well-being.
The observed positive association between cat density and happiness is consistent with previous studies showing the psychological and physiological benefits of pet ownership [e.g., 18–21], including reduced loneliness and improved emotional well-being [16,17,59]. Cat density is an aggregate measure and, therefore, an imperfect proxy for pet ownership and does not capture the quality or context of human-cat interactions. Nevertheless, these associations suggest that human-animal relationships may explain population-level differences in subjective well-being, even when examined at a coarse ecological scale. Thus, it is possible that the links between cat ownership, reduced loneliness, and alleviated stress may explain the association between cat density and greater national happiness, which, in turn, may reflect more favorable attitudes toward pets, higher disposable income, and better living conditions, all of which are closely related to higher subjective well-being [1,3,59].
An interesting and counterintuitive finding of this study was the moderate-to-large negative association between cat density and T. gondii seroprevalence, challenging the common assumption that a higher number of cats would imply more parasite exposure [e.g., 46]. One explanation for this finding is that in wealthier countries, cats are more likely to be kept indoors, fed processed food, and receive routine medical care, significantly reducing the potential of T. gondii infection. In contrast, in lower- and middle-income countries, cats are often free-roaming, increasing the odds of T. gondii infection [27,44,47].
It is noteworthy that while causality cannot be inferred from these ecological correlations, the moderate-to-large associations further support the growing evidence linking T. gondii exposure to negative neuropsychological outcomes [33,60]. For instance, chronic T. gondii infection has been associated with subtle changes in personality and the development of depression, anxiety, schizophrenia, and other mental disorders [23–25,29,30,35–37,40–42], whereas acute infections may result in immunological alterations that could affect mood and cognitive functioning [25,39]. It is possible that the effects of T. gondii infection on psychological processes are mediated by the action of this neuroparasite on brain neurotransmitters and specific circuits involved in emotion regulation and cognitive processes [61,62].
It is important to emphasize that all associations reported here, including those derived from mediation models, should be interpreted as statistical relationships rather than evidence for causal pathways. While ecological correlations can be informative for identifying broad-scale patterns and generating hypotheses, they cannot disentangle directionality or rule out unmeasured confounding processes [61,62]. Accordingly, mediation effects are discussed cautiously and are framed as indirect statistical associations rather than causal effects. It also is critical to highlight that our findings describe associations at the national level and should not be used to infer individual-level effects or personal risks related to cat ownership, happiness, and T. gondii exposure.
From a broader and more general perspective, the large associations between GDP, income level, health expenditure, and happiness support the social determinants of health framework [63], according to which healthcare, education, and environmental safety are known to reduce the exposure to infectious agents and enhance psychological resilience [64,65]. Our findings extend this framework by suggesting that cat ownership and parasitic exposure, which is typically overlooked in well-being research, can influence well-being at the national level. The results of our exploratory regression and mediation analyses further clarify the associations that emerged from the correlational analyses. Specifically, these analyses showed that cat density and IgM seroprevalence have the largest associations with national happiness, suggesting that biological and ecological factors may contribute to national well-being alongside established socioeconomic drivers. Additionally, our findings have the potential of informing public health policies aimed to mitigate the negative effects of T. gondii exposure, to promote responsible pet ownership, and to increase the well-being at a national level.
Limitations and strengths
It is important to recognize that our study has several limitations. First, the seroprevalence data on T. gondii infection obtained from pregnant women, although widely adopted in epidemiological and ecological research [e.g., 27,28], may not fully represent the broader population introducing a potential demographic bias in our study. It is possible that pregnant women may differ in their healthcare access, testing protocols, and exposure risks across countries. Nevertheless, in the absence of comprehensive national surveillance programs, these data represent the most robust and standardized source for global comparisons. Likewise, the cat density measurement conflates pet cats, stray cats, and feral cat populations which could differently impact on human psychological benefits and parasite transmission processes. Thus, future studies should differentiate between these cat populations and account for cultural variations to more precisely isolate their specific impacts on T. gondii exposure and national happiness. Second, we did not consider several confounding variables such as urbanization levels, climate indicators, or sanitation infrastructure, which may also influence cat ownership, parasite exposure, and subjective well-being. For example, it is possible that stray cats are less common in urbanized territories with better sanitation infrastructure, thus limiting the transmission of T. gondii to other animals [66]. Moreover, climate indicators such as humidity and temperature may affect the sporulation of oocysts [67] which may also influence processes linked to T. gondii prevalence, population health, cat ownership, and ultimately, national well-being.
Third, the cross-sectional ecological data and the correlational design of this study prevented us from inferring causality. Thus, the observed patterns in our exploratory prediction and mediation analyses should be interpreted as a statistical decomposition of associations. Alternative explanations, including reverse causation and unmeasured confounding variables cannot be ruled out and may account for some of the observed relationships. Future studies may benefit from using a predictive approach, rather than a correlational one, to test the causal effects of cat density, T. gondii exposure, and socioeconomic indicators of wealth on national happiness. Moreover, because the World Happiness Index partly incorporates socioeconomic indicators such as GDP, some degree of conceptual overlap exists between the outcome measure and certain predictors.
Finally, data quality and availability varied across countries, which is common in cross-national ecological studies relying on secondary data, potentially introducing a temporal misalignment among the study variables and limiting the generalizability of the findings. Although this mismatch may introduce some noise or bias, T. gondii seroprevalence tends to change slowly over time in the absence of major public health interventions. Thus, any potential bias is likely conservative, as declining exposure in wealthier countries over time would probably underestimate the negative association between T. gondii and national happiness. Future studies should closely align temporal data to confirm the robustness of these ecological patterns.
Despite these limitations, our study also has several strengths. First, while previous research has explored the potential role of parasites such as T. gondii on economic and cultural patterns [e.g., 43,68], this is the first study that integrates ecological, biological, and socioeconomic indicators in a global analysis of happiness across a large set of countries. By including data on cat ownership, parasitic exposure, and economic development, this research provides a novel and interdisciplinary perspective on the determinants of well-being. Moreover, the inclusion of both IgM and IgG seroprevalence allowed us to differentiate between acute and chronic infections, increasing the biological specificity of our analysis. Overall, our findings highlight the importance of considering both biological and socio-ecological factors in the study and understanding of global patterns of subjective well-being.
Conclusion
This ecological study provides novel insights into the magnitude and robustness of the associations between happiness, cat density, T. gondii exposure, and socioeconomic factors across 93 countries. Our main findings showed that, besides the typical associations between subjective well-being and some socioeconomic factors, national happiness was largely and positively associated with cat density and largely and negatively associated with IgM T. gondii exposure. These results underscore the need for future studies integrating both socioeconomic and biological determinants of happiness and including longitudinal and interdisciplinary approaches to elucidate the underlying mechanisms and processes.
Supporting information
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]