Life expectancy of cats in Britain: moggies and mollies live longer

Introduction

The domestic cat (Felis catus) is one of the most popular companion animals worldwide. Its domestication began approximately 12,000 years ago in the Fertile Crescent and the Levante, with the process being completed in Egypt 6,000 years ago (Nilson et al., 2022). Cats were primarily kept in households for their hunting abilities and rodent control (Crowley, Cecchetti & McDonald, 2020). They were tolerated in homes due to their pest control skills, and some authors suggest a process of self-domestication in a commensal relationship (Hu et al., 2014). The relationship and tolerance between humans and cats evolved naturally due to the benefit for both humans and cats. While cats benefited from the availability of food in a humanised environment, humans benefited from the presence of cats as these could provide a way to control pests. The role of companion animals developed later in the history of the human-cat relationship (Crowley, Cecchetti & McDonald, 2020).

As household pets, cats provide a range of benefits, including enjoyment, affection, companionship, and even health advantages (Amiot, Bastian & Martens, 2016). However, these roles have not led to the specialization observed in dogs, such as herding, guarding, and game hunting. Consequently, differentiation through selection was not evident until the late 1800s, when Victorian Britain began breeding cats for their appearance, marking the evolution of the cat fancy (The British Newspaper Archive, 2013). The National Cat Club, the first cat club, played a key role in promoting the cat fancy, and together with a second club, it became the Governing Council of the Cat Fancy in Great Britain, established in 1910 and recognized as the oldest fancy cat association.

Breeds evolved based on appearance rather than specialization, leading to the fixation of deleterious genes resulting from mutations, such as those seen in brachycephalic cats (e.g., Persian and Burmese) (Plitman et al., 2019), the furless Sphynx (Åhman & Bergström, 2009), the cartilage disorder causing the Scottish Fold’s distinctive appearance (Gandolfi et al., 2016), and the shortened limbs seen in the Munchkin (Lyons et al., 2019).

Modern animal welfare science has highlighted these genetic issues, and the longevity of pets has gained importance as a concern for both pet owners and veterinary professionals (e.g., Farstad, 2018; Grandin & Deesing, 2022; Morel et al., 2024; Sand et al., 2017). Data on mortality and life expectancy are crucial for understanding the health and welfare of companion cat populations. Predicted life expectancy can reveal shifts in overall health and welfare, and comparing lifespans across different cat subgroups can help identify those at higher risk for health issues or shorter lifespans (O’Neill et al., 2015). Previous research has highlighted various biological (e.g., Kent et al., 2022; Mata & Bhuller, 2022), environmental (e.g., Kent et al., 2022; Morel et al., 2024), and behavioral (Mata, 2015; Montoya et al., 2023; Teng et al., 2024) factors that affect cat longevity, including reproductive status, sex, breed, and genetic factors (Egenvall et al., 2009).

While referring to cats’ breeding, the term domestic cat is frequently used to describe non-pedigreed cats, commonly known as moggies in the UK or domestics elsewhere. Unlike pedigreed breeds, which are selectively bred for specific traits, domestic cats are the result of natural genetic diversity and lack standardized physical or behavioral characteristics.

Several methods can be used to report overall mortality: average (mean or median) age at death, life table analysis, and survival analysis. Life tables allow for estimating life expectancy and the associated probability of death at specific age groups. Survival analysis, using Kaplan-Meier and Cox regression methods, complements life table analysis by enabling comparisons between groups with different characteristics and risk factors in single models (Abd ElHafeez et al., 2021).

Previous studies, including those by O’Neill et al. (2014) and recently Teng et al. (2024) in the UK, have constructed life tables, while Kent et al. (2022) in California, USA, calculated lifespans for different cat groups, and Montoya et al. (2023) generated life tables based on clinical records from Banfield Pet Hospitals in the USA.

This study aims to further investigate the lifespan of cats in the UK, extending the research conducted by Teng et al. (2024) with a survival analysis focusing on how cat genetic type, reproductive status, and sex interact to influence lifespan. By analyzing the interactions between these variables, this study seeks to provide a more comprehensive understanding of their contributions to feline longevity. Particularly concerning breeding effects, we aim to demonstrate that pure breeding may have negative implications for the health and welfare of cats, as previously shown in dogs by Mata & Mata (2023).

Materials and Methods

The data used in this study were publicly available under a Creative Commons Attribution 4.0 license (CC BY 4.0) and were obtained from the Royal Veterinary College, University of London’s repository on FigShare (O’Neill & Teng, 2024). The sample consists of all cats receiving primary veterinary care in clinics affiliated with the VetCompass™ Program in the UK during 2019, meaning they had at least one electronic patient record in that year (Teng et al., 2024). The dataset included variables for ‘reproductive status’ (entire and neutered/spayed), breeding status or ‘cat type’ (identifying pure breeds and cross-breeds as cats with only one pure breed ancestor, and moggies as all types of domestic cats without any type of purebred ancestry), ‘age at death’ in years, and ‘sex’ (male and female). Spayed females will, from now on, be referred to as mollies.

Initially, the dataset contained N = 7,935 entries. However, after cleaning the data by removing unknown breed entries, unknown reproductive statuses, and outliers, the final sample included N = 7,708 cats. Detailed figures of observations in the different categories can be found in Table 1. Outliers were detected following the transformation of the data to a standard normal distribution, characterized by a mean of zero and a standard deviation of one, by identifying values with z-scores exceeding 3 or falling below −3 standard deviations.

An ANOVA with a least significant difference (LSD) post-hoc test was used as a full factorial general linear model to compare mean differences between the ‘cat types,’ ‘reproductive status,’ and ‘sex,’ along with their respective interactions. The LSD test was chosen as the homogeneity of variances was verified, and no more than three levels were compared for each factor analysed.

A one-way ANOVA was also performed using the factor joining together sex and reproductive status, with the levels ‘entire female’, spayed female’, ‘neutered male’, ‘entire male’. Tukey’s honest significant difference (HSD) test was used as the post-hoc as there were four levels in the factor analysed.

The prerequisites of the ANOVA—namely, homogeneity of variances and normal distribution of residuals—were assessed and confirmed via Levene’s test and via Kolmogory-Smirnov test, respectively.

The data were analyzed using a Cox proportional hazards model to assess survival, with ‘age at death’ as the time-to-event variable and ‘cat type’ (moggy, crossbred, or purebred), ‘reproductive status’ (entire or neutered/spayed), and ‘sex’ as the factors of interest. No data were censored, and the event was defined as the ‘age at death’ (in years). The model was evaluated using the −2 Log likelihood test, while the parameters were assessed with the Wald test. Cumulative survival plots were generated. The prerequisites of linearity and proportional hazards assumptions are met.

To investigate the interaction between ‘reproductive status’ and ‘sex,’ a new variable combining both factors was created, resulting in four levels: ‘mollies,’ ‘female entire,’ ‘male neutered,’ and ‘male entire.’ The data were entered into a Kaplan-Meier model tested with a log-rank (Mantel-Cox) test to calculate differences between lifespan means and medians. A survival plot was also produced.

Data were first imported into a spreadsheet (Microsoft Excel for Microsoft 365 MSO, version 2204 Build 16.0.15128.20240, 64-bit) for cleaning. Descriptive statistics were also computed using this software. The Cox proportional hazards snd the Kaplan-Meier models were fit using the software package IBM SPSS Statistics (Version 29.0.2.0 (20); IBM Corp., Armonk, NY, USA). The level of significance was set to p < 0.05 in all tests.

Results

The overall mean lifespan for the cats in the data set analysed was 10.61 years with a 95% confidence interval of [10.30–10.91].

The ANOVA test yielded a significant result, indicating significant differences in all main effects and the interaction between the ‘reproductive status’ and the ‘cat type’. Details of the ANOVA results are displayed in Table 2. The results of the post hoc tests showing the mean values found significant, are shown in Table 3.

Significant differences were found between all the levels of the main factors analysed, except for cat types, where differences were not found significant between cross-bred and pure breed cats. Moggies live longer on average than ‘pure breeds’ and ‘cross-bred’, without significant differences being found between the last two. Relatively to ‘sex’, females have a longer average lifespan. Concerning ‘reproductive status’, ‘neutered’ tom cats, or ‘spayed’ queens have also significantly different lifespans, in comparison to ‘entire’ cats. The interaction between ‘cat type’ and ‘reproductive status’ showed that neutered/spayed cats have a longer average lifespan, while entire purebreds have the lowest average lifespan.

The one-way ANOVA showed that spayed females have the longer average lifespan, followed by neutered males, entire females, and entire males. The results are presented in Table 4.

In both ANOVA tests, both the Levine’s tests and the Kolmogorov-Smirnov tests showed homogeneity of variances (p > 0.05) and normal distribution of the residuals (p > 0.05).

The cox regression model was found significant (−2 Log Likelihood 122368, ${\chi }^2$ = 222.69, 4 df, p < 0.001). The full parameterization of the model, the respective statistical tests of parameters, and the hazard ratio (HR) for the parameters are shown in Table 5. The baseline function as covariate means values of: ‘sex’ 0.512, ‘reproductive status’ 0.273, ‘cat type’ pure breed 0, cross-bred 0.33, moggy 0.863. The plot of the baseline cumulative hazard is presented in Fig. 1. Figures 2–4 represent respectively the plots of the cumulative survivals for the three ‘cat type’, the two ‘sex’, and the two ‘reproductive status’.

The interpretation of the hazard ratio tells that ‘males’ had a lower probability of survival with an HR 28.9% higher than ‘females’; ‘entire’ cats had a lower probability of survival with an HR 22.7% higher than ‘neutered/spayed’ cats; ‘moggies’ had a higher probability of survival with an HR 22.2% lower than both ‘cross-bred’ and ‘pure-breed’ cats, with no significant differences between these last two.

The Kaplan-Meier model was found significant (Mantel-Cox log-rank 187335, ${\chi }^2$ = 187.34, 3 df, p < 0.001). The medians of the combined ‘reproductive status’ and ‘sex’, together with the respective confidence intervals, are shown in Table 6. The survival plot for the four combinations (‘mollies’, ‘female entire’, ‘male neutered’, and ‘male entire’) is presented in Fig. 5.

The means confidence intervals don’t intersect each other and therefore were significantly different (p < 0.05). Nevertheless, the survival curves of neutered males and entire females intersect each other at some point in time, with entire females showing a lower lifespan before reaching ≈ 12–13 years, but in later ages, lifespan increases comparatively. Mollies showed the longer lifespans, and entire males the lower.

Discussion

The overall mean lifespan, along with the respective 95% confidence interval, for the cats in the analyzed dataset was 11.83 years (11.70; 11.96). The median lifespan and the respective interquartile range were 12.92 years (12.73; 13.11). Other authors have reported median lifespans of 14.0 years (IQR 9.0–17.0; range 0.0–26.7) using UK data (O’Neill et al., 2015); 9.07 years (IQR 4.20–12.92; range 0.01–21.85) in a California, USA population (Kent et al., 2022); 11.18 years (mean 11.16–11.20) in a USA population (Montoya et al., 2023); 11.01 years (mean 10.55–11.46) in a population from Buenos Aires, Argentina (Gennaro, Isturiz & Pucheta, 2023); and over 12.5 years (median) in Sweden (Egenvall et al., 2009). The results of Egenvall et al. (2009) are close to those of the present study; however, the results of O’Neill et al. (2015) are higher, which can be explained by the fact that the authors excluded cats under five years of age from their analysis. The means obtained by Kent et al. (2022) are lower than those in the present study, likely due to local conditions affecting the cat population in California, as their results are also below those reported by Montoya and colleagues for the entire USA. The results obtained by Montoya et al. (2023) and Gennaro, Isturiz & Pucheta (2023) are lower than those in the present study, indicating higher survival rates in the UK. Nevertheless, it is important to note that different cat populations, influenced by varying environmental factors, exhibit different lifespans, as highlighted by Teng et al. (2024).

The effect of sex

The effect of sex in the present analysis determined that male cats have a shorter life than female cats. Table 7 summarizes the results also in comparison with studies of other authors. All these authors confirm a higher lifespan for females.

Table 7:

Cats’ lifespan estimation by sex.

Study	Location	Male cat	Female cat
Present study (θ)	UK	12.00 (6.51; 15.699)	13.88 (8.60; 17.04)
Present study (μ)	UK	11.17 (10.99; 11.35)	12.52 (12.33; 12.70)
O’Neill et al. (2015) (θ)	UK	13.0 (7.6; 16.0)	15.0 (11.0; 17.4)
Montoya et al. (2023) (μ)	USA	10.72 (10.68; 10.75)	11.68 (11.65; 11.71)
Gennaro, Isturiz & Pucheta (2023) (μ)	Buenos aires	10.92 (10.50; 11.35)	12.26 (11.78; 12.73)

DOI: 10.7717/peerj.18869/table-7

Note:

Comparison of the results obtained in the present study with those obtained by other authors. Medians (θ) with interquartile range and means (μ) with 95% confidence intervals.

Higher mortality rates for male cats may be attributed to traumatic, health-related, and genetic (epigenetic) factors. Traumatic causes include accidents and fights, which are more common among males than females. Agonistic behaviour between males is frequent for the definition of sexual dominance, which may result in serious injuries (Albright, Calder & Learn, 2022; Martin, 2023; Stella, 2021; Vitale, 2022) (Fig. 6). Tom cats, driven by their sexual instincts, roam over large areas in search of receptive females, while females typically do not venture out in search of males (Hall et al., 2016). O’Neill et al. (2015) identify trauma as the most common cause of mortality in England, accounting for 12.2%, with falls, traffic accidents, and dog attacks being notable contributors. Road traffic accidents are the leading cause of traumatic death for cats in both Sweden (Egenvall et al., 2009) and the United States (Loyd et al., 2013), responsible for 60% to as much as 87% of traumatic deaths in cats (Egenvall et al., 2009; Olsen & Allen, 2001; O’Neill et al., 2015).

Several health conditions that are more prevalent in male cats can also shorten their lifespan. Urinary (urethral) tract obstruction has been identified as a condition predominantly affecting male cats (Montoya et al., 2023; Segev et al., 2011). Infections with feline immunodeficiency virus (FIV) and feline leukemia virus (FeLV) are also more common in males, as their tendency to roam outdoors increases their risk of exposure and infection (Kent et al., 2022). A study by O’Neill et al. (2023a) found that male cats have a higher predisposition to periodontal disease, renal tubular acidosis, heart murmurs, lameness, obesity, abscesses, wounds, and cat bite injuries (often related to fights between males). Periodontal disease in cats has been associated with comorbidities such as hepatic, cardiorespiratory, and renal disorders (Watson, 1994), as well as systemic bacteremia (Perry & Tutt, 2015).

The epigenetic likelihood of shorter lifespans in tom cats is highlighted by Kent et al. (2022). This phenomenon is observed not only in cats but across various mammal species, as noted by Clutton-Brock & Isvaran (2007). In polygynous species, more intense intrasexual male competition can limit opportunities for successful individual male breeding, resulting in weaker selection pressures favoring longevity in males compared to females.

Conclusions

This study emphasizes the importance of considering multiple factors when assessing cat longevity, suggesting that owners should be mindful of ‘sex’, ‘reproductive status’, and breeding status or ‘cat type’ when caring for their pets. The observed variations in lifespans across different populations and environmental contexts further indicate the necessity for tailored approaches to feline health and welfare, such as giving preference to outbreeding practices. Future research should explore the underlying mechanisms influencing these trends to develop targeted interventions aimed at enhancing the quality of the health and welfare, and duration of ‘cats’ lives.