Improving performance of hurdle models using rare-event weighted logistic regression: an application to maternal mortality data.

In this paper, performance of hurdle models in rare events data is improved by modifying their binary component. The rare-event weighted logistic regression model is adopted in place of logistic regression to deal with class imbalance due to rare events. Poisson Hurdle Rare Event Weighted Logistic Regression (REWLR) and Negative Binomial Hurdle (NBH) REWLR are developed as two-part models which use the REWLR model to estimate the probability of a positive count and a Poisson or NB zero-truncated count model to estimate non-zero counts. This research aimed to develop and assess the performance of the Poisson and Negative Binomial (NB) Hurdle Rare Event Weighted Logistic Regression (REWLR) models, applied to simulated data with various degrees of zero inflation and to Nairobi county's maternal mortality data. The study data on maternal mortality were pulled from JPHES. The data contain the number of maternal deaths, which is the outcome variable, and other obstetric and demographic factors recorded in MNCH facilities in Nairobi between October 2021 and January 2022. The models were also fit and evaluated based on simulated data with varying degrees of zero inflation. The obtained results are numerically validated and then discussed from both the mathematical and the maternal mortality perspective. Numerical simulations are also presented to give a more complete representation of the model dynamics. Results obtained suggest that NB Hurdle REWLR is the best performing model for zero inflated count data due to rare events.

Determinants of COVID-19 Vaccine Acceptability among Healthcare Workers in Kenya-A Mixed Methods Analysis.

Healthcare workers (HCWs) were a priority group for COVID-19 vaccination. Adopting the World Health Organization's 3C and the expanded 5C vaccine hesitancy models, we assessed the factors associated with COVID-19 vaccine acceptability among HCWs in Kenya. In a mixed methods study, respondents were from eight selected counties across the country. An online survey (n = 746), key informant interviews (n = 18) and focus group discussions (n = 3) were conducted. The data were analyzed concurrently. Quantitative data showed that all the 3C antecedents were strong predictors of vaccine acceptability. The association of vaccine acceptability was strongest with convenience (aOR 20.13, 95% CI 9.01-44.96), then complacency (aOR 10.15, 95% CI 4.63-22.21) and confidence (aOR 6.37, 95% CI 2.90-14.02). Marital status was a significant independent factor associated with vaccine acceptability (aOR 2.70, 95% CI 1.20-6.08). Qualitatively, convenience presented as the no-cost availability of vaccines at the health facilities, whereas non-complacency manifested from the first-hand observed experience of COVID cases, and the need to protect oneself and family members. Confidence was mainly attributed to increased knowledge, resulting from multiple training sessions and trust in regulatory authorities. Other social factors including workplace pressure, religion and misinformation had a role in influencing HCW vaccination decisions. In the background of a pandemic, the 3C model is a strong predictor of vaccine acceptability, and making the vaccines easily available and convenient to HCWs significantly impacts their uptake.

The trypanocidal Cape buffalo serum protein is xanthine oxidase.

Plasma and serum from Cape buffalo (Syncerus caffer) kill bloodstream stages of all species of African trypanosomes in vitro. The trypanocidal serum component was isolated by sequential chromatography on hydroxylapatite, protein A-G, Mono Q, and Superose 12. The purified trypanocidal protein had a molecular mass of 150 kDa, and activity correlated with the presence of a 146-kDa polypeptide detected upon reducing sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Amino acid sequences of three peptide fragments of the 146-kDa reduced polypeptide, ligand affinity and immunoaffinity chromatography of the native protein, and sensitivity to pharmacological inhibitors, identified the trypanocidal material as xanthine oxidase (EC 1.1.3.22). Trypanocidal activity resulted in the inhibition of trypanosome glycolysis and was due to H2O2 produced during catabolism of extracellular xanthine and hypoxanthine by the purine catabolic enzyme.

African buffalo serum contains novel trypanocidal protein.

The high ability of African buffalo, as compared to domestic cattle, to control infections with Trypanosoma brucei brucei ILTat 1.4 organisms did not correlate with the timing or magnitude of parasite surface coat-specific antibody responses and may have resulted from the constitutive presence in buffalo blood of a novel trypanocidal factor. Buffalo plasma and serum contained material that killed bloodstream stage T. b. brucei, T. b. rhodesiense, T. b. gambiense, T. evansi, T. congolense, and T. vivax organisms during four h of incubation at 37 degrees C in vitro. Serum from eland was also trypanocidal whereas serum from oryx, waterbuck, yellow-back duiker, cattle, horse, sheep, goat, mouse, rat, and rabbit was not trypanocidal. The buffalo serum trypanocidal material was not lipoprotein, or IgG, and had the following properties: 1) a density of > 1.24 g/ml determined by flotation ultracentrifugation; 2) insolubility in 50% saturated ammonium sulphate; 3) non-reactivity with anti-bovine IgM, and anti-bovine IgG; 4) non-reactivity with protein G, and protein A; 5) a relative molecular mass of 152 kDa determined by chromatography on Sephacryl S 300, and of 133 kDa determined by chromatography of the 50% SAS cut of IgG-depleted buffalo serum on Superose 12; 6) no associated cholesterol; and 7) inactivation by digestion with proteinase K that was immobilized on agarose.

Control of G1 to S cell cycle progression of Trypanosoma brucei S427cl1 organisms under axenic conditions.

Trypanosoma brucei S427cl1 organisms made 6 divisions in modified minimal essential medium (BMEM) supplemented with fetal bovine serum (FBS)-low or high density lipoprotein (LDL, HDL) and fatty acid-free bovine serum albumin (FAF-BSA). Omission of lipoproteins or FAF-BSA from the medium caused the parasites to accumulate in G1 of the cell cycle and to lose the ability to replicate at 37 degrees C. Proteinase K-treated LDL or HDL, which did not have detectable apolipoprotein, supported the G1 to S cell cycle transition of T. brucei S427cl1 organisms in BMEM supplemented with FAF-BSA. Addition of C6:0, C7:0 or fatty C8:0 fatty acid (1 mol fatty acid mol-1 FAF-BSA in the incubation mixture) to serum-free medium supplemented with LDL or HDL and FAF-BSA prevented T. brucei S427cl1 organisms from progressing through G1 into S of the cell cycle. T. brucei S427cl1 organisms became stumpy-like forms during plateau phase growth under axenic conditions. Stumpy-like T. brucei S427cl1 organisms were mainly in G1 of the cell cycle, expressed raised levels of NAD diaphorase activity, were unable to replicate at 37 degrees C, but were able to differentiate to replicating procyclic organisms. Medium collected from plateau phase cultures of T. brucei S427cl1 did not support the G1 to S cell cycle transition of exponentially growing T. brucei organisms. The capacity of plateau phase medium to support G1 to S transition of T. brucei S427cl1 organisms was restored by addition of FAF-BSA and its capacity to support 4 cycles of replication of the parasites was restored by addition of FAF-BSA and LDL or HDL.

Analysis of Propionibacterium acnes-induced non-specific immunity to Trypanosoma brucei in mice.

Mice treated with dead Propionibacterium acnes (previously called Corynebacterium parvum), up to 30 days before infection with any of three strains of Trypanosoma brucei, were more able to limit the level of first-wave parasitaemia than untreated controls. Reduced parasitaemia was not due to enhanced phagocytosis of input parasites and was associated with a dramatic reduction in the proportion of multiplying T. brucei in the blood of treated as compared to control mice. For 4 days after P. acnes treatment, T. brucei growth-inhibitory molecules, assayed by their effect on T. brucei multiplication under axenic culture conditions, were detected in the serum of recipient mice. The molecules were released by macrophages collected from the peritoneal cavity of P. acnes-treated mice, and similar molecules were produced in vitro by macrophages from normal mice after incubation with P. acnes. Accessory studies suggested that the molecules were breakdown products of P. acnes and were unlikely to be responsible for the long-term in-vivo effects of the P. acnes treatment. It was also shown that monokines and lymphokines which are likely to be induced by in-vivo P. acnes treatment, i.e. IL-1, IL-2, TNF alpha, INF alpha, INF beta, INF gamma, PGE1, PGE2, PGF2 alpha and biological mediators present in Con-A and LPS-induced spleen cell supernatants (collected 20, 40, 60 or 80 h after mitogen stimulation) had no influence on T. brucei growth under axenic culture conditions over a wide range of concentrations. The studies suggest that the P. acnes effect was not due to a direct interaction of these biological mediators with the T. brucei. We suggest that the reduction in T. brucei parasitaemia in P. acnes-treated mice reflects secondary physiological effects of one or more unidentified biological mediators.