Areas Of Biostatistics Research Articles

On a matrix‐valued autoregressive model

Many data sets in biology, medicine, and other biostatistical areas deal with matrix‐valued time series. The case of a single univariate time series is very well developed in the literature; and single multi‐variate series (i.e., vector time series) though less well studied have also been developed. A class of matrix time series models is introduced for dealing with situations where there are multiple sets of multi‐variate time series data. Explicit expressions for a matrix autoregressive model along with its cross‐autocorrelation functions are derived. Stationarity conditions are also provided. Least squares estimators and maximum likelihood estimators of the model parameters and their asymptotic properties are derived. Results are illustrated through simulation studies and a real data application.

Grouping one-degree trapezoids into biostatistical areas on example of the Okhotsk Sea

A method for grouping one-degree trapezoids into biostatistical areas of any size is proposed. For this purpose, all stations of survey are marked by the number of trapezoid, its area and central coordinates. Then the stations are sorted by the trapezoid numbers; their biological parameters (biomass, abundance, etc.). The mean values are integral parameters with the weights proportional to size of trapezoids — they can be easily grouped into small or large areas of any shape, depending on research tasks.

From clinical data management to clinical data science: Time for a new educational model.

The purpose of this article is to propose and provide a blueprint for a graduate-level curriculum in clinical data science, devoted to the measurement, acquisition, care, treatment, and inferencing of clinical research data. The curriculum presented here contains a series of five required core courses, five required research courses, and a list of potential electives. The coursework draws from but does not duplicate content from the foundational areas of biostatistics, clinical medicine, biomedical informatics, and regulatory affairs, and may be reproduced by any institution interested in and capable of offering such a program. This new curriculum in "clinical" data science will prepare students for work in academic, industry, and government research settings as well as offer a unifying knowledge base for the profession.

Clinical biostatistics in the 2020s

Clinical biostatistics in the 2020s

Integrating computational thinking into a longitudinal data analysis course for public health students

After completing an introductory biostatistics course, public health students often need to take one or more follow-on courses focusing on specialized areas of biostatistics. While there exists decades’ worth of pedagogical research on teaching introductory statistics to non-statistics majors, few systematic attempts have been made to explore innovative ways to teaching follow-on courses to non-statistics majors such as public health students. Extending previous research on teaching categorical data analysis to public health students, this paper explores ways to harness the power of computational thinking in teaching conceptual knowledge in a follow-on course on longitudinal data analysis. The proposed approach aims to keep students in their zone of proximal development by using computational experiments as a tool for developing understanding of conceptual knowledge. Learning activities center on experiments that explore the likelihood function. Illustrative examples of actual student work are used to demonstrate a practical way of integrating computational thinking into biostatistics course content.

Feeding and food supply of salmons in the Okhotsk Sea in autumn of 2018-2019

In the autumn months of 2018-2019, food spectra of salmons were quite diverse in the Okhotsk Sea. Amphipods prevailed in the diet of juvenile pink and chum salmon that corresponded to these zooplankton species abundance in the southern part of the sea; other significant portions of their diet were presented by euphausiids, copepods, arrowworms, pteropods, and oikopleura. Decapods, fish and squids were in the diet of masu, coho and chinook salmon. Portion of zooplankton was about 100 % in the diet of pink salmon fingerlings, but only 98 % in the diet of chum salmon fingerlings with 2 % portion of squids. Grazing by salmons was insignificant: 0.2-0.3 % of the total zooplankton biomass within the periods of observation (18 days in 2018 and 13 days in 2019), or 0.04-2.0 % for certain taxonomic groups, except of pteropods and oikopleura with the consumed portions of 4-7 % and 7-12 %, respectively. However, the daily consumption by salmons relative to daily production of some groups was significant in October, 2018 and reached 97 % for amphipods and > 200 % for euphausiids in the northeastern Okhotsk Sea (biostatistical areas 6-8) and 48 and 27 %, respectively, in the entire surveyed area. This ratio was twice lower in October, 2019, as 24 % for amphipods and 12 % for euphausiids. Daily consumption of copepods and arrowworms never exceeded 2 % of their daily production. The most favorable conditions for nekton feeding in fall of both years were observed in the northwestern Okhotsk Sea, and the worst conditions — in its northeastern part, with moderate conditions without deficit of prey in the central part (biostatistical area 9) and southern part (biostatistical areas 12, 13), where the biomass of forage plankton exceeded the biomass of nekton in 30-97 times in 2018 and in 56-76 times in 2019.

Schemes of multi-parameter stations in the Bering and Chukchi Seas and North Pacific made in research cruises of TINRO (1986–2019)

Spatial distribution of stations made in research expeditions conducted by Pacific Res. Inst. of Fisheries and Oceanography (TINRO) in the Bering and Chukchi Seas and in the North Pacific in 1986–2019 is discussed. Schemes of the stations for every cruise are presented and sorted by decades, years, seasons, and months. The data sufficiency is estimated at regional and temporal levels by calculation of the stations number in the daytime and nighttime, per years, per seasons, and per biostatistical areas.

Schemes of multi-parameter stations in the Okhotsk Sea made in research cruises of TINRO (1984–2019)

Spatial distribution of stations made by research expeditions of TINRO in the Okhotsk Sea in 1984–2019 (82 surveys) is considered. The schemes of stations are sorted by decades, years, seasons, and months. Information the number of daylight and nighttime stations and the stations distribution by time of day, years, seasons, and by 33 biostatistical areas is presented in table form, that allows to estimate the data coverage at regional and temporal levels.

OPPORTUNITIES AND TECHNIQUES OF USING THE DATABASES OF TINRO «ZOOPLANKTON OF THE NORTH PACIFIC, OKHOTSK, BERING, AND CHUKCHI SEAS», «NEKTON TROPHOLOGY», AND «MARINE BIOLOGY»

Structure of the databases «Zooplankton…» and «Nekton trophology» is described and some techniques are proposed for nekton studies using these bases in conjunction with the database «Marine Biology». All three databases are regularly updated. The bases «Zooplankton…» and «Nekton trophology» contain raw data on plankton and feeding of nekton collected in the North Pacific and the Okhotsk, Bering and Chukchi Seas in 1984–2018. The «Nekton trofology» database contains information for 97 species of nekton, mostly for mass species (72–78 % of samples belong to 5 most numerous species), and 156 species of prey, including 27 species of Copepoda, 7 species of Euphausiacea, 9 species of Amphipoda, 14 species of Decapoda, 6 species of Coelenterata, 15 species of Cephalopoda, and 60 species of Pisces, other groups of prey are represented by 1–2 species. The data are spatially sorted by biostatistical areas and their sub-areas, in total 64 sub-areas in the Okhotsk Sea, 32 areas in the Bering Sea, 30 areas in the North Pacific, and 5 areas in the Chukchi Sea. Mean depth is determined for each sub-area. Method of spatial distribution mapping is demonstrated with using the sub-areas as integral stations or the 1-degree grid for Surfer software. Technique of regional inventory is explained with summarizing and averaging the data and calculation of various indicators as plankton–nekton ratio, etc. The 1-degree trapeziums are numbered for easier usage. Some useful examples are presented with the author’s comments (showing his personal opinion).

Shared random parameter models: A legacy of the biostatistics program at the National Heart, Lung, and Blood Institute

Shared random parameter models (SRPMs) were first introduced by researchers at the National Heart Lung and Blood Institute (NHLBI) Biostatistics Branch for analyzing longitudinal data with informative dropout (Wu and Carroll, 1987; Wu and Bailey, 1988; Follmann and Wu, 1995; Albert and Follmann, 2000; Albert et al, 2002). This work was all focused on characterizing the longitudinal data process in the presence of an informative missing data mechanism that is treated as a nuisance. Shared random parameter modeling approaches have also been developed from the perspective of characterizing the relationship between longitudinal data and a subsequent outcome that may be an event time, a dichotomous measurement, or another longitudinal outcome. This article will review the early contributions of the NHLBI biostatisticians on SRPMs for analyzing longitudinal data with dropout and demonstrate how these ideas have, more recently, been applied in these other areas of biostatistics. Rather than focus on technical details or specific analyses, this article presents a conceptual framework for SRPMs within a historical context.

LONG-TERM AVERAGE BIOMASS AND DOMINANT FISH SPECIES IN THE BOTTOM BIOTOPES OF THE OKHOTSK SEA. PART 1. COMPOSITION AND QUANTITATIVE RATIO OF SPECIES ON SHELVES IN DIFFERENT AREAS OF THE SEA

Composition of the demersal fish community in the bottom biotopes of 14 biostatistical areas of the Okhotsk Sea is considered on the data of 9,189 standard trawl catches obtained in 95 expeditions conducted by Pacific Fish. Res. Center (TINRO) in 1977–2010. The most abundant fish species in each area are ranked by biomass. The total demersal fish biomass is estimated as 9583.0 . 103 t (2124.9 . 103 t without pollock and herring). The most numerous demersal species are: pacific cod Gadus macrocephalus, yellowfin sole Limanda aspera, pacific sand lance Ammodytes hexapterus, great sculpin Mycocephalus polyacanthocephalus, and saffron cod Eleginus gracilis.

Elementary trophic ecology of pacific salmons in the Bering Sea. Species and regional differences. Provision with food in different environments

The final issue in the series of studies on trophic ecology of pacific salmons in the Bering Sea conducted by planktonologists of Pacific Fish. Res. Center (TINRO) for NPAFC. Salmons dwell in the upper epipelagic layer and feed mainly in the daytime by large-sized zooplankton that is more abundant in the upper pelagic layer as compared with the lower pelagic layer in summer-fall season. The zooplankton and nekton species dominant in the salmon diet are determined on the base of 8,000 plankton samples and 45,000 samples of salmon stomachs. Two periods with different environments are revealed in the Bering Sea in the early 21 century: the relatively warm period of 2000-2006 and the relatively cold period of 2007-2012. The regime shift between them affected the plankton community and trophic ecology of salmon, though not by means of water cooling directly but presumably because of the water circulation change. Food composition is determined for each size group of salmons in both periods, by biostatistical areas and bathymetric ranges. Zooplankton grazing by salmons increased significantly in all areas during the cold period, whereas the nekton grazing decreased, but their total consumption did not change much. Transition to the next warm period since 2013 does not affect negatively to the salmon food base; their diet changes significantly but within the list of preferable species. Feeding intensity of pink, chum, and sockeye salmons was similar in any period, as well as in the 2014-2015, food deficit was never observed. The salmons show a great plasticity in feeding: in general, their diet corresponds to the structure of plankton community, though the portion of nekton increases with the age. Large-sized salmons consume a large-sized prey and need more time for its complete digestion, so there is necessary to apply reduction coefficients in calculation of their daily food rations. N​utritional needs of all salmon stocks are estimated for summer-fall season and offered as a guideline for evaluation their load on the food base and their food security for certain spatial and temporal circumstances.

First experience of using a new database on net zooplankton in the Far-Eastern Seas and adjacent Pacific waters

Unique database on zooplankton collected by 0.1 m 2 Juday planktonic net (mesh size 0.168 mm) in the Chukchi, Bering, Okhotsk, and Japan Seas and the North Pacific in 1984-2013 is presented. History of its creation, sources and volume of the data are described and the first operating experience and prospects of its using for applied and fundamental researches of marine biological resources are explained. Some new results based on this database are discussed. In particular, 5 tabular reference-books of species composition, occurrence and abundance of zooplankton for the main fishing grounds in the Russian part of this region were prepared and published in 2014-2015, where the information is summarized by species, stages of development, size fractions, biostatistical areas, water layers, light and dark time of day, four seasons, and quasi-decadal periods. Charts of spatio-temporal distribution of certain taxonomic and dimensional groups are drawn. Assumption is offered on negative dependence of zooplankton body size on water temperature, that needs additional verification. Vertical zooplankton migrations are considered and there is concluded that its biomass in the upper layer increases usually in the nighttime in the Okhotsk and Japan Seas but in the daytime in the Bering Sea and North Pacific, therefore attempts to recalculate the daytime biomass to the nighttime one using constant coefficients give opposite results for these pairs of regions. Prospects for developing of GIS and atlases of zooplankton for the North Pacific are discussed with examples of charts for the data generalized by abovementioned criteria, but this application is postponed until the best times because of funding reduction.

A bivariate measurement error model for semicontinuous and continuous variables: Application to nutritional epidemiology.

Semicontinuous data in the form of a mixture of a large portion of zero values and continuously distributed positive values frequently arise in many areas of biostatistics. This article is motivated by the analysis of relationships between disease outcomes and intakes of episodically consumed dietary components. An important aspect of studies in nutritional epidemiology is that true diet is unobservable and commonly evaluated by food frequency questionnaires with substantial measurement error. Following the regression calibration approach for measurement error correction, unknown individual intakes in the risk model are replaced by their conditional expectations given mismeasured intakes and other model covariates. Those regression calibration predictors are estimated using short-term unbiased reference measurements in a calibration substudy. Since dietary intakes are often "energy-adjusted," e.g., by using ratios of the intake of interest to total energy intake, the correct estimation of the regression calibration predictor for each energy-adjusted episodically consumed dietary component requires modeling short-term reference measurements of the component (a semicontinuous variable), and energy (a continuous variable) simultaneously in a bivariate model. In this article, we develop such a bivariate model, together with its application to regression calibration. We illustrate the new methodology using data from the NIH-AARP Diet and Health Study (Schatzkin et al., 2001, American Journal of Epidemiology 154, 1119-1125), and also evaluate its performance in a simulation study.

Integral values of biomass and stock of zooplankton in the epipelagic layer of the area 71 in the North Pacific, including the Bering and Okhotsk Seas, and patterns of distribution for mass species

The Okhotsk and Bering Seas and the subarctic zone of the North Pacific are surveyed well by the expeditions conducted by Pacific Fish. Res. Center (TINRO) in the last three decades (115 surveys in 1984-2013), with sampling of zooplankton for monitoring of feeding resources of commercial species. Mean total zooplankton biomass and biomasses of size fractions of zooplankton, its taxonomic groups and mass species are calculated, by biostatistical areas. General patterns of the biomass distribution between the biostatistical areas are determined and presented as the charts of the biomass averaged by the areas. The most abundant species of zooplankton with similar systematic and trophic status are spatially separated that allows them to reduce competition for food, for example the pairs of small-sized, medium-sized and large-sized Copepoda species ( Pseudocalanus minutus - Oithona similis, Metridia pacifica - M. okhotensis , Calanus glacialis - Neocalanus plumchrus ), the pair of Hyperiida species ( Themisto pacifica - T. libellula ) and four Euphausia species ( Thysanoessa raschii - Th. inermis - Th. longipes - Euphausia pacifica ) in the Okhotsk Sea. Detailed tables on biomass and stock of the main components of zooplankton are presented, by regions and biostatistical areas.

Spatial-temporal patterns of euphausiids ecology in the Okhotsk Sea

The most common regularities in spatial distribution of mass Euphausia species and their size classes and ontogenetic stages are determined for the Okhotsk Sea using the data averaging by biostatistical areas and applying the average results to the centers of the areas that allows to smooth small-scale patterns as patchiness. Thysanoessa raschii and Th. inermis occupy mostly shelf areas and Th. longipes and Euphausia pacifica - deep-water areas; among them Th. raschii and Th. longipes are more abundant and Th. inermis and E. pacifica are supposedly allochtonous species which penetrate to the Okhotsk Sea through the Kuril Straits, so they are distributed mainly in the southern part of the Sea. Mean biomass and abundance of all species in the upper (0-50 m) and deeper (50-200 m) layers are calculated for each stage and size class in the daytime and nighttime, separately: most of them, including both adults and furcilias, concentrate mainly in the deeper layer and their night biomass is higher in both layers though about a half of animals raise to the sea surface at night. Only eggs, nauplii and calyptopes of euphausiids don’t change their abundance considerably between day and night, and most of them occupy the upper layer round the clock. Obviously, the increasing of late furcilias and adults in the deeper layer at night is caused by their diurnal migration from the layers below 200 m. Seasons of the highest abundance for early stages come in natural order that is a sign of their short duration.

Marginal Models for Censored Longitudinal Cost Data: Appropriate Working Variance Matrices in Inverse-Probability-Weighted GEEs Can Improve Precision

When cost data are collected in a clinical study, interest centers on the between-treatment difference in mean cost. When censoring is present, the resulting loss of information can be limited by collecting cost data for several pre-specified time intervals, leading to censored longitudinal cost data. Most models for marginal costs stratify by time interval. However, in few other areas of biostatistics would we stratify by default. We argue that there are benefits to considering more general models: for example, in some settings, pooling regression coefficients across intervals can improve the precision of the estimated between-treatment difference in mean cost. Previous work has used inverse-probability-weighted GEEs coupled with an independent working variance to estimate parameters from these more general models. We show that the greatest precision benefits of non-stratified models are achieved by using more sophisticated working variance matrices.

Environmental conditions and compositions of plankton and nekton in the epipelagic zone of the northwestern Pacific Ocean in summer 2009, the year of the largest spawning migration of pink and Chum Salmon

The distribution of temperature, salinity, phytoplankton, zooplankton, and nekton, including Pink Salmon and Chum Salmon, is analyzed based on the data from a complex survey conducted in the upper epipelagic zone of the northwestern Pacific Ocean in June–July 2009. An estimate of water transfer, plankton biomass (by size fractions and taxonomic groups), and biomass and abundance of nekton is made. The water circulation pattern was notable for the “compressed” state of the Western Subarctic Gyre. Copepods predominated in zooplankton (60.8% biomass), and Pacific Salmon dominated in the nekton. The ratio of the plankton biomass to the nekton biomass in various biostatistical areas varied from 23.9 to 195.1 (mean 46.7). It was found that the nekton biomass in the area of the spreading subarctic waters reaches its maximum in the summer months and that in the area of subtropical waters reaches its maximum in the fall. In the upper 30-meter layer, the biomass of fish and squid rose during the night to 2.92 million tons, or 1.6 times as much, due to mesopelagic migrants. The nekton communities of the upper epipelagic of the northwestern Pacific Ocean functioned normally, providing a high level of biological and fish productivity.

Letters

We are pleased that our initial letter to the editor, “The PharmD Degree in Developing Countries,” received attention and resulted in several letters of response being published.1 We would like to summarize some of the issues raised by other colleagues. In the letter “Controversy of PharmD Degree,” published in AJPE, the principal author opinionated the similarity of pharmacy degrees issued throughout the world. We found it difficult to agree with this perspective--that the “PharmD tag” is the only solution-because the curriculum is neither fully industry-oriented nor clinically aligned. In contrast, the letter penned by Ghayur not only deserves merit but energizes us to share the experiences we had during 3 months in pharmacy institutes and pharmacies in Pakistan.2 In fact, the doctor of pharmacy (PharmD) degree in Pakistan was started without any planning. Thus, it is not incorrect to say that the PharmD degree is just a tool to help students qualify for job opportunities in Gulf countries1 as well as to make the so-called “doctors” eligible to sit for the license examination in the United States or elsewhere. While working on my PhD project in Pakistan, I (S.J.) had an opportunity to interview final-year PharmD students at 3 public and 3 private universities. The students were apprehensive about their future because “this new curriculum, without any proper clinical attachment, with just a couple of visits to hospitals, would serve the purpose is a doubtful issue.” They expressed a strong sense of discontentment over the forced conversion from the 4-year bachelor's degree program for which they were admitted to a 5-year PharmD program without their consent. Not even proper clinical affiliation with inadequately trained staff (in one instance clinical pharmacy was taught by a zoologist) they perceived the PharmD degree program as a mere “hoax” with no light at the end of tunnel. Likewise, the co-author of this letter (Z.B.), while delivering a lecture as a Malaysian pharmacy expert to pharmacy institutes in Pakistan, was speechless to answer one pharmacy student's question: “Sir, why in Pakistan, clinical pharmacy comes under the umbrella of Pharmaceutics”? Just to endorse the views penned down by Ghayur2 regarding non-availability of pharmacists on pharmacies as well as selling of prescription drugs without prescription, we would like to describe our experience here. A visit to a pharmacy in Karachi, Pakistan, being run by non-professionally trained drug sellers, exhibited not only brand substitutions without the consent of a doctor but also antibiotic dispensing without a prescription. The most nerve-racking observation was the supply of benzodiazepine available on consumer demand. Interestingly 25 consumers mentioned different names of the benzodiazepine class of drugs and bought a whole month's supply without a prescription.3 Interestingly, despite the absence of pharmacists, every pharmacy has a license to sell medicines thanks to many of our fellow pharmacists who lease their degrees annually to any layman who wants to open a chemist shop or pharmacy. With gaps even in the bachelor of pharmacy curriculum, accompanied with sparse recognition of pharmacists as healthcare providers in the healthcare system of Pakistan, we are unable to comprehend to what extent switching to the PharmD degree would be justified. The PharmD could be required in developing countries; however, a systematic plan of action would be needed to give due recognition to pharmacists as they are still an underrated group of health professionals. Regarding India's recent scheme for adding the PharmD, the Health Ministry has paved the way for Indian pharmacy practice training and education by giving formal approval for PharmD regulations.4 The PharmD course is comprised of 6 academic years, with 5 years of study and 1 year of internship and residency. Six months of the internship and residency is spent in a general medicine department and 2 months each in other speciality departments. The clerkship, coupled with a research project covering areas of biostatistics and research methodology, as well as concepts of pharmacoeconomics and clinical research, is also in place.4 However the effectiveness of any such scheme largely depends on its delivery and execution, as well as the availability of clinical pharmacy staff members. Shazia Jamshed Universiti Sains Malaysia Zaheer-ud-din Babar University of Aukland, New Zealand Mohamed Izham Mohamed Ibrahim Universiti Sains Malaysia