Survival analysis approach to insect life table analysis and hypothesis testing: with particular reference to Russian wheat aphid (Diuraphis noxia (Mordvilko)) populations

A unified survival-analysis approach to insect population development and survival times

There are two major categories of observation data in studying time-dependent processes: one is the time-series data, and the other is the perhaps lesser-recognized but similarly prevalent time-to-event data (also known as survival or failure time). Examples in entomology include molting times and death times of insects, waiting times of predators before the next attack or the hiding times of preys. A particular challenge in analyzing time-to-event data is the observation censoring, or the incomplete observation of survival times, dealing which is a unique advantage of survival analysis statistics. Even with a perfectly designed experiment being conducted perfectly, such ‘naturally’ censoring may still be unavoidable due to the natural processes, including the premature death in the observation of insect development, the variability in instarship, or simply the continuous nature of time process and the discrete nature of sampling intervals. Here we propose to apply the classic Cox proportional hazards model for modeling both insect development and survival rates (probabilities) with a unified survival analysis approach. We demonstrated the advantages of the proposed approach with the development and survival datasets of 1800 Russian wheat aphids from their births to deaths, observed under 25 laboratory treatments of temperatures and plant growth stages.

A randomized, blinded, controlled trial to assess sand fly mortality of fluralaner administered orally in dogs

Background

Leishmania infantum is the parasite responsible for the disease in humans known as zoonotic visceral leishmaniasis (ZVL). Dogs are considered the main domestic reservoir of ZVL and sand flies are the proven vectors. The use of systemic insecticides in dogs has been studied as an alternative strategy to control ZVL in endemic areas. One systemic insecticide in dogs, fluralaner, has a proven anti-sand fly effect in membrane-fed studies. However, the efficacy and duration on sand flies directly feeding from dogs treated with fluralaner remains unknown.

Methods

Direct feeding bioassays were performed on 10 beagle dogs that had been randomly assigned to two groups: one with five dogs orally treated with Bravecto® (fluralaner) and other five as a control. About 30 females of Phlebotomus papatasi were allowed to directly feed from dogs at seven days before the administration of the treatment and Days 3, 17, 31, 45 and 73 post-treatment. Sand fly mortality after feeding was observed every 24 h for 5 days. The Kaplan-Meyer method, Henderson-Tilton formula and a negative binomial mixed model were used to respectively calculate: (i) mortality and its 95% confidence interval (CI); (ii) efficacy of the insecticide at killing sand flies in 24 h; and (iii) differences in the risk of sand fly death at 24 h after feeding.

Results

Control sand fly mortality 24 h after feeding was always ≤ 20% and mortality in the fluralaner group ranged from 2% (95% CI: 0–4%) 7 days before treatment to 100% at 3 days post-treatment. Fluralaner efficacy was 100, 93, 94 and 75% at Days 3, 17, 31 and 45, respectively (P < 0.0001). The increase in the risk of sand fly death was 32.9 (95% CI: 4–263), 76 (95% CI: 8–705), 95.8 (95% CI: 9–1029) and 10.6 times (95% CI: 1.43–79) on Days 3, 17, 31 and 45, respectively

Conclusions

The efficacy of fluralaner, orally administered to dogs, against sand-flies was above 90% for 31 days. Fluralaner administered to dogs should be further evaluated as a control strategy in ZVL endemic areas.

Electronic supplementary material

The online version of this article (10.1186/s13071-018-3231-8) contains supplementary material, which is available to authorized users.

Control of subterranean termites (Isoptera: Rhinotermitidae) infesting power poles

A trial was conducted to determine the efficacy of termiticidal dusts (arsenic trioxide, triflumuron, and Metarhizium anisopliae), a timber fumigant (dazomet) and liquid termiticides (bifenthrin, chlorfenapyr, chlorpyrifos, fipronil, and imidacloprid) for controlling subterranean termites (Isoptera: Rhinotermitidae) infesting in-service power poles in New South Wales, Australia. Dusts were applied to parts of the pole where termites were present. Fumigant was inserted into holes drilled into the base of the pole. Liquid termiticides were mixed with soil around the base of the pole and injected into internal voids if present. Poles were inspected for up to 5 yr, and the time taken for reinfestation to occur was recorded. Before the start of the trial, the major Australian pole owners were surveyed to obtain an estimate of the annual national cost of termite infestation to the power supply industry. The annual costs of termite treatment and replacing damaged poles were estimated at AU$2 million and AU$13 million, respectively. Infestation rates were lower for all treatments compared with controls within the first 12 mo of the study. Dazomet, arsenic trioxide, fipronil, and chlorpyrifos were the most efficacious treatments. Efficacy was positively related to the amount of termiticide applied and negatively related to the infestation severity but was unaffected by geographical location. Survival curves were calculated of the time elapsed before the recurrence of termite infestations (survival absence of reinfestation). Survival was highest for poles treated with liquid termiticides.