This is the first comprehensive study of a wide range of welfare risk factors concerning an unowned unsocialised domestic cat’s environment or health status that can be linked to the animal’s being in a problematic nutritional state in terms of being thin or emaciated (i.e., the complete absence of fat deposits). Two (0.3%) of the 598 necropsied cats were emaciated, and no specific cause of this was identified. Irrespective of cause, emaciation is considered a severe welfare issue for the individual animal, and it seems to be uncommon in unowned unsocialised domestic cats in Denmark. The BCS ‘thin’ was more common (68 cats; 11.0%), but this body condition is much less challenging for the animal than emaciation: the main potential welfare issue it presents is the risk that it might progress to emaciation.
The model pseudo-R2 was 0.26, which suggests a reasonably fit, but with additional information not included in the model. Cats trapped in the season spring/summer had the highest odds of low BCS. Also, only one juvenile cat was categorised as thin, and no neutered cats were thin or emaciated. It was intact mature/grown-up males and pregnant females that were primarily affected, which is not surprising considering the nutritional needs of pregnancy and potential roaming of males. The only internal health-related factor associated with low BCS when age-group was controlled for was GI findings, and the only parasites found to be associated with low BCS were lice. However, skin lesions, tooth lesions and the presence of ear mites, FIV and H. taeniaeformis worms, A. abstrusus L1, and T. cati eggs were all associated with low BCS when age-group and reproductive status were not controlled for. Such a finding could be common for age-related conditions, but it is not possible to verify in this study. We have previously determined the prevalence of a number of conditions that unowned unsocialised cats in Denmark suffer from, and have estimated that 17% have major health issues . These include mostly damaged teeth, which has also been reported as a major finding among TNR cats in Japan .
Season, age-group and reproductive status were all significant factors, and age-group and reproductive status appeared to confound the effect of some of the other variables. This complicated the interpretation, particularly because the small sample size did not permit stratification into subgroups. It should be noted that lactating cats were excluded from sampling, and inference could not be made on these. One way to overcome confounding is through stratification; however, this would require a larger sample size.
The finding that the adult age-group was associated with low BCS conflicts with an Israeli study of urban TNR cats  describing that kittens were more likely to have lower BCS than adults, and that neutered adult cats were more likely to be obese than intact adults. However, because age-group is known to be associated with a number of infections, diseases and conditions, our result makes good sense. The older cats become, the more likely they are to have been exposed to, and acquire, for example, FIV and some parasites, or to have become involved in situations resulting in injuries. Although exposure to some pathogens may lead to immunity, this means there are many conditions from which age-group cannot be separated, which must be considered in the analyses and their interpretation.
Season was found to be associated with low BCS. Cats that had been trapped during spring/summer had the highest odds of low BCS, whereas those trapped during summer/autumn had lower odds, as has also been observed in Australia, although in stray cats . These findings are not surprising given that the availability of feed increases over summer, when the number of prey animals is higher and people spend more time outside and may be more likely to provide feed for cats. It is surprising, however, how little BCS of cats trapped in winter/spring seemed to differ from those trapped in summer/autumn (Table 6), but the peak in low BCS coincides with the breeding season, which may be physiologically demanding for these cats. Reproductive status also displayed associations with BCS: neutered cats had low odds of having low BCS, while intact males and females, as well as pregnant cats, had higher odds (Table 6). This finding is consistent with one study on TNR cats , where neutering has been identified as a protective factor against poor body condition. Another study not assessing the potential confounding effects reported that intact male TNR cats are more likely to be injured than intact female TNR cats , and this may in turn cause loss of body condition.
We found the presence of lice and the presence of ear mites to be associated with low BCS. Both ectoparasites are likely to influence the health of the cat . Fleas and ticks were present in low numbers  and were not associated with low BCS. It is commonly assumed that high ectoparasite loads on cats that are ill or in poor condition may partly be the result of reduced or insufficient grooming. It is also possible that the higher ectoparasitic loads develop as a result of impaired immunity in animals with poor nutritional status, or that a high ectoparasitic burden is irritating and thus affects the animal’s feeding behaviour, thereby reducing its feed uptake. While no associations between body condition and ectoparasites in cats were found reported in the literature, flea infestation has been associated with season in stray cats .
Univariable associations of endoparasites with low BCS were observed, but we could not accurately determine whether age-group, reproductive status or endoparasitic status explained the effect. Infection with H. taeniaeformis worms was associated with low BCS. This parasite has generally been considered to have a relatively low impact on cats. However, it is known that A. abstrusus can be pathogenic , and this supports our finding of an association between the infection and body condition. A similar association was found for T. cati egg counts, but not for adult T. cati. The eggs are not themselves pathogenic, but they may reflect adult female worm burden. Counts of A. abstrusus L1 or H. taeniaeformis worms were higher in cats with low BCS, as were juvenile T. cati counts.
Although we employed a reasonably large sample, it was still relatively limited for a risk factor study. This can be seen most clearly in connection with FeLV, where only seven cats in total were positive but the prevalence of low BCS among them was 43%. A larger sample size, for which confounding effects could be addressed, is needed to investigate a possible association between this infection and low BCS. Similarly, we would draw attention to the high counts of juvenile T. cati worms, which may be associated with low BCS, but this could not be convincingly demonstrated with our sample. An additional limitation with regards to the sample size was the dichotomisation of many findings; for example, we combined loss of a tooth with that of damage to a tooth, irrespective that the effect on BCS could be different, and damage would likely be more severe in terms of its effect on the welfare of the cat. A larger sample size could have prevented this dichotomisation, which was used for many variables.
We cannot be certain that our sample population was representative of the unowned unsocialised domestic cat population. However, since no special criteria were used in constructing the sample, we believe it can be considered representative of the Danish population of unowned unsocialised domestic cats as a whole. It could, however, be the case that healthy cats are reported more often because they might cause more problems as a result of their higher activity level. On the other hand, it is also possible that sick cats are reported more often because people might feel pity for them. With the data we have, we cannot verify which mechanism is more likely, and thus which way the bias could affect the results. Also, neutered cats may be less of an annoyance with less fighting and less mating behaviour, and therefore may be underrepresented. However, we would not expect the proportion of neutered cats to be high in the population because unsocialised cats are usually not released again once caught, owing to Danish legislation. Furthermore, due to practical and logistical constraints, cats were not sampled for an entire season. This precludes interpretation outside the seasons collected.
With the confounding observed with the above-mentioned variables, we can still observe that: Only one juvenile and no neutered cats were observed with a low BCS, and low BCS was primarily observed in spring/summer (April to July), and mostly in intact males or pregnant females, which accounted for 79% of cats with low BCS (Table 2). Consequently, focus could be on intact adults. Season is a given, and we would expect that pregnant cats have a lower BCS due to higher energy expenditure, and neutered cats would experience lower energy expenditure . Also, 79 of the 80 pregnant cats were from the spring/summer–summer/autumn periods, with 8/12 of those with low BCS observed in the spring/summer period. These cats should probably not be identified with health-related problems just because they are pregnant. However, among the 12 pregnant cats with low BCS, 3 (25%) had problems with their teeth, 4 (33%) had lice, 4 (33%) had ear mites, 5 (out of 10, 50%) were positive for T. gondii. Among 42 intact male cats with low BCS, 12 (29%) had teeth problems, 10 (24%) had FIV, 13 (out of 40, 33%) had lice and 12 (out of 33, 36%) were positive for T. gondii. While it is important to emphasise that causality cannot be inferred with this type of observational data, low BCS in an adult cat could be an indicator of poor health and therefore compromised welfare.
Strengths of the study include its use of a relatively large sample and its investigation of many different variables. The results highlight associations that can be used to define a risk profile.