|
nif:isString
|
-
This study protocol was reviewed and approved by the University of California—Davis Institutional Animal Care and Use Committee (Approval #13000). Permission for participation of shelter cats in this study was obtained from the managing veterinarian at each shelter.
Nine shelters participated in the study. Shelter enrollment in the study was convenience-based and criteria for inclusion in the study were a willingness to participate; presence of a veterinarian to serve as lead contact for the study; existence of a defined set of criteria for URI recognition at the shelter; willingness to comply with study protocol; ability to collect required data daily, and completion of an online survey. The study was conducted from August 1, 2008 through July 31, 2009.
Each shelter that was enrolled had an on-site veterinarian and defined set of criteria for URI recognition. A universal case definition for feline URI across all participating shelters was not implemented. Case definitions ranged from very broad—any symptoms of upper respiratory disease present—to more specific, where several symptoms needed to be present prior to diagnosis. Case definitions included some degree of sneezing, along with ocular and/or nasal discharge. Individual shelter URI case definitions are available in the supporting information S1 Doc. Shelter URI Criteria.
Prior to the start of the study, shelters completed a 47-question online survey (S1 Survey link) to establish shelter baselines for housing, management and environmental factors (summarized in Table 1). Specific questions were asked about housing during the first 7 days of shelter care, referred to as “intake housing”. Monthly updates were received via email from each shelter’s veterinarian to report any major changes from the baseline data provided in the survey for housing, management, or the environment that occurred in the shelter during the 12-month study period.
Table data removed from full text. Table identifier and caption: 10.1371/journal.pone.0190140.t001 Baseline housing and environmental factors. a1 = 3 to <6 ft2, 2 = 6–8 ft2, 3 = >8–10 ft2 (0.28–0.56m2, 0.56–0.74m2, > 0.74–0.93m2)bFeline viral rhinotracheitis (feline herpesvirus-1) and calicivirus All shelters used an online URI database (described below) where shelter staff recorded the daily number of adult cats in the shelter population categorized by health status (with or without URI), and each day’s intake of adult felines (new cats entering the shelter population). Feral cats were not included in this study. Cats were observed daily and recorded as either not sick with URI (and therefore at risk for URI) or sick with URI. For each new case of URI observed, the following data were recorded in the online URI database: Cat ID#, intake date and URI diagnosis date. Shelters also had the option to enter outcome date and outcome type (e.g. adoption versus euthanasia), but this was not required nor consistently used and therefore not included in the analysis. The online URI database was developed by the Computing and Technology Services (CATS), UC Davis School of Veterinary Medicine, hosted by UC Davis Koret Shelter Medicine Program. Prior to the start of the study, a one-hour internet sharing session (using Breeze Meeting by Adobe, San Jose, California) was held for each shelter to train the shelter veterinarian and staff on the use of the online database and to answer questions. Each participating shelter had a unique ID and login password to access the online database. Shelters had access to all their data and could monitor their URI rates throughout the study period. The database provided each shelter with a monthly report that summarized feline URI via a linear graph that displayed the shelter’s URI rate in bimonthly intervals, with the ability to view any time range of interest.
To determine the monthly URI rate, the total number of new cases of URI was divided by the total number of cat-days at risk for URI (the sum of the daily inventory of cats without URI) for that month. URI rates in this study were reported per 1000 at risk cat days. Cats diagnosed with URI at the time of intake, or on day 1 or day 2 of their shelter stay were excluded from the calculation, as these were considered to be pre-existing cases of URI and not shelter-acquired disease.
PCR sampling for URI pathogens at intake: PCR sampling was performed at a subset of five shelters during the study period. Selection was based on having skilled medical staff available for sample collection or shelter location proximal to the research team. Cat handling and PCR sampling were consistently performed. Samples were collected by trained RVT and Veterinary staff via the research team (at 4 shelters) or via the shelter medical staff (1 shelter). Approximately 20 cats observed to be healthy at intake were sampled at each shelter during a one-week sampling period every three months. Cats showing any sign of URI, and those designated as feral by shelter staff or that could not be handled for sampling were excluded. Cats were sampled within 24 hours of intake at each of the five shelter facilities. Because sampling was only performed at one shelter at a time, a five-week continuous period was needed to sample cats at the five shelters. (In a few cases, it was not possible to get 20 samples because of low intake during the sampling period.) Samples were collected onto sterile cotton swabs with plastic sticks. Samples for PCR testing consisted of one oropharyngeal swab and one conjunctival swab from each cat. The two sample swabs were placed into a single red top tube and labeled for the individual cat. Samples were stored in a cooler during each sampling day and then placed in a freezer (-20°F or -29C) until all the samples for that shelter were collected for that sampling period. Samples were submitted (delivered or mailed) to IDEXX Laboratories (Sacramento, California) for real-time quantitative polymerase chain reaction (RT-PCR) testing (test code 2512 RealPCR™ FURD panel) designed to detect the following feline upper respiratory disease agents: Chlamydophila felis, feline calicivirus (FCV), feline herpesvirus-1 (FHV-1), Bordetella bronchiseptica and Mycoplasma felis. All assays were designed and validated according to industry standards (Applied Biosystems, User Bulletin #3). Target genes for each application were: B. bronchiseptica: hemagglutinin fusion protein gene (FhaB), AF140678; C. felis: outer membrane protein A (OmpA), AP006861; feline herpesvirus 1: glycoprotein B (18), feline calicivirus: ORF 1, AF109465; M. felis: ssr RNA—ITS-1, AF443608. The total number of cats sampled at intake in the five shelters over the one-year period was 329 cats.
Variables were originally obtained reflecting the range of practices considered possible risk factors for URI and which could reasonably be assessed via questionnaire. However, for some variables, all shelters followed the same practices, preventing comparison. For example, all shelters vaccinated cats on intake with a modified live parenteral vaccine for feline calicivirus, herpesvirus and panleukopenia; fed a consistent diet daily from day to day within the shelter (as compared to some shelters where a variety of donated food is fed from day to day); and used a disinfection labeled as effective for all pathogens implicated in feline URI. The following variables were examined using Poisson regression with robust variance estimation to account for clustering of cats within shelters (Stata 13.1/IC, StataCorp LP, College Station, Texas) for possible associations with monthly URI rates between shelters: double compartment housing (no/yes), intake housing floor space (3–6 ft2, 6–8 ft2, >8-10ft2 {.28-.56m2, .56-.74m2, >.74-.93m2}), hiding space provided in intake housing (no, sometimes, always), mixed-age housing (no, yes), frequency of cat moves in and out of the cage in the first week (≤ 2 moves, >2 moves), use of intranasal vaccine (no, yes) and monthly shelter intake (natural log transformation). A p value < 0.05 was considered significant.
In a separate analysis involving a subset of five shelters, Poisson Regression was used to examine the association of pathogens at intake with monthly URI rate. A p value < 0.05 was considered significant.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}