Search bioRxiv⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/084541

Power calculator for instrumental variable analysis in pharmacoepidemiology

Walker, V. M.·Davies, N. M.·Windmeijer, F.·Burgess, S.·Martin, R. M.
View original source ↗Export metadata ↓

Abstract

BackgroundInstrumental variable analysis, for example with physicians prescribing preferences as an instrument for medications issued in primary care, is an increasingly popular method in the field of pharmacoepidemiology. Existing power calculators for studies using instrumental variable analysis, such as Mendelian randomisation power calculators, do not allow for the structure of research questions in this field. This is because the analysis in pharmacoepidemiology will typically have stronger instruments and detect larger causal effects than in other fields. Consequently, there is a need for dedicated power calculators for pharmacoepidemiological research.\n\nMethods and resultsThe formula for calculating the power of a study using instrumental variable analysis in the context of pharmacoepidemiology is derived before being validated by a simulation study. The formula is applicable for studies using a single binary instrument to analyse the causal effect of a binary exposure on a continuous outcome. A web application is provided for the implementation of the formula by others.\n\nConclusionsThe statistical power of instrumental variable analysis in pharmacoepidemiological studies to detect a clinically meaningful treatment effect is an important consideration. Research questions in this field have distinct structures that must be accounted for when calculating power.\n\nFUNDING STATEMENTThis work was supported by the Perros Trust and the Integrative Epidemiology Unit. The Integrative Epidemiology Unit is supported by the Medical Research Council and the University of Bristol [grant number MC_UU_12013/9]. Stephen Burgess is supported by a post-doctoral fellowship from the Wellcome Trust [100114].\n\nKey MessagesO_LIResearch questions using instrumental variable analysis in pharmacoepidemiology have distinct structures that have previously not been catered for by instrumental variable analysis power calculators.\nC_LIO_LIPower can be calculated for studies using a single binary instrument to analyse the causal effect of a binary exposure on a continuous outcome in the context of pharmacoepidemiology using the presented formula and online power calculator.\nC_LIO_LIThe use of this power calculator will allow investigators to determine whether a pharmacoepidemiology study is likely to detect clinically meaningful treatment effects prior to the studys commencement.\nC_LI

Source connections

Explore related subjects

epidemiologyPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Walker, V. M., Davies, N. M., Windmeijer, F., Burgess, S., Martin, R. M.. 2016-10-31. Power calculator for instrumental variable analysis in pharmacoepidemiology. https://doi.org/10.1101/084541

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2021

A recombinant protein containing influenza viral conserved epitopes and superantigen induces broad-spectrum protection

Influenza pandemic poses public health threats annually for lacking vaccine which provides cross-protection against novel and emerging influenza viruses. Combining conserved antigens inducing cross-protective antibody response with epitopes activating cross-protective cytotoxic T-cells would offer an attractive strategy for developing universal vaccine. In this study, we constructed a recombinant protein NMHC consisting of influenza viral conserved epitopes and superantigen fragment. NMHC promoted the mature of bone marrow-derived dendritic cells and induced CD4+ T cells to differentiate into Th1, Th2 and Th17 subtypes. Mice vaccinated with NMHC produced high level of immunoglobulins which cross-bound to HA fragments from six influenza virus subtypes with high antibody titers. Anti-NMHC serum showed potent hemagglutinin inhibition effects to highly divergent group 1 (H1 subtypes) and group 2 (H3 subtype) influenza virus strains. And purified anti-NMHC antibodies could bind to multiple HAs with high affinities. NMHC vaccination effectively protected the mice from infection and lung damage challenged by two subtypes of H1N1 influenza virus. Moreover, NMHC vaccination elicited CD4+ and CD8+ T-cell responses to clear the virus from infected tissue and prevent virus spreading. In conclusion, this study provided proof of concept for triggering both B cells and T cells immune responses against multiple influenza virus infection, and NMHC may be a potential candidate of universal broad-spectrum vaccine for various influenza virus prevention and therapy.

epidemiology↗
bioRxiv2019

Factors associated with cholera outbreaks, Nairobi County, July 2017: a case control study

BackgroundCholera affects 1.3-4 million people globally and causes 21000-143,000 deaths annually. Nairobi County in Kenya reported cholera cases since April 2017. We investigated to identify associated factors and institute control measures.\n\nMethodsWe reviewed the line-list of patients admitted at the Kenyatta National referral Hospital, Nairobi and performed descriptive epidemiology. We carried out a frequency-matched case control study, using facility-based cases and community controls. We defined a case as acute onset of watery diarrhoea of at least >3 stools/24hours with or without vomiting in person of any age, admitted in Kenyatta National Hospital as from July 1st, 2017. We calculated odds ratios and their respective 95% confidence intervals. We also took water samples at water reservoirs, distribution and consumer points, and made observation on hygiene and sanitation conditions in the community.\n\nResultsWe reviewed 71 line-listed cases; median age 30 years (range 2-86 years); 45 (63%) were male. First case was admitted on 14th April 2017. Culture was performed on 44 cases, 30 (68%) was positive for Vibrio cholerae, biotype El-Tor, serotype Ogawa. There were 2 deaths (case fatality ratio 2.8%). Age-group [≥]25 years was most affected. Drinking unchlorinated water (aOR 14.57, 95% CI 4.44-47.83), eating in public places (aOR 9.45, 95% CI 3.07-29.12) sourcing water from non-Nairobi city water company source (aOR 4.92, 95% CI 1.56-15.52) and having drank untreated water in the previous week before the outbreak (aOR 3.21, 95% CI 1.12-9.24) were independently associated with being a case in the outbreak. Out of 28 water samples, 4 (14%) had >180 coliforms/100mls; all were at consumer points.\n\nConclusionPoor water quality and sanitation were responsible for this outbreak. We recommended adequate, clean water supply to unplanned settlements in Nairobi County, as well education of residents on water treatment at the household level.\n\nAuthor summaryCholera, a disease causing outbreaks in areas with low standards of hygiene and sanitation has afflicted humans for millennia. It is caused by a bacterium, Vibrio Cholerae, transmitted mainly through water contaminated by faecal matter. The resultant disease is acute watery diarrhoea, which causes death rapidly due to dehydration and shock. Virtually brought under control in the developed world due to improvements in hygiene, the disease still ravages many communities in low and middle income countries, as well as regions affected by conflict or natural disasters. In outbreak situations, rapid response in water treatment, sanitation improvement and setting up of cholera treatment centres for rehydration therapy reduces impact and saves lives. Long-term control can only be achieved through sustainable improvements in sanitation and standards of living. Case control studies in outbreak situations provide quick, actionable information to public health specialists during outbreak response. This study provides a cholera outbreak investigation in an urban informal settlement setting; the approach reported here can guide in outbreak investigations and response in similar settings globally.

epidemiology↗
bioRxiv2019

Infection of Borrelia burgdorferi sensu lato of small mammals in Yunnan Province, China

BackgroundLyme disease is caused by Borrelia burgdorferisensulato (BBSL) which is usually found in wild and domestic mammals worldwide. Human cases of B. burgdorferi infections have been identified in China, but little direct surveillance of potential rodent reservoirs has been performed in Yunnan Province, Southwestern China. Yunnan Province is a tropical area with a diverse topographic range and sustains a high biodiversity of small mammals that could potentially play an important role in the transmission of a variety of B. burgdorferigenospecies.\n\nMethods3659 small mammals were captured in 159 sample siteslocated 23 countries inYunnan Province and screened for BBSL infection by nested PCR based on 5S-23S rRNA intergenic spacer gene of BBSL.Univariate and multivariate forward stepwise logistic regression analysis was used to access the association between infections and related risk factors.\n\nResultsInfection with BBSL was confirmed in 3.99%(146/3659) of small mammals. Significant differences in prevalence rates of BBSL were observed at varying landscape types and altitudes.Small mammals in forested areas had higher prevalence rates than other landscape types as did small mammals found at altitudes greater than 2500 meters. The 5S-23S rRNA intergenic spacergene revealed that there were 5 genotype of BBSL, including B. afzelii, B. burgdorferisensustricto, B.japonica, B.gariniiand B.valaisiana, which demonstrate the genetic diversity and regional distribution.\n\nConclusionsThere exists a wide distribution and genetic diversity of endemic BBSL in Southwestern China, warranting further investigations and monitoring of clinical disease in individuals presenting with symptoms of Lyme disease in these areas.\n\nAuthor summaryLyme disease is caused by Borrelia burgdorferi sensu lato (BBSL) which is usually found in wild and domestic mammals worldwide. Human cases of Borrelia burgdorferi sensu lato infections have been identified in China, but little direct surveillance of potential rodent reservoirs has been performed in Southwestern China. This study documents potential small mammal reservoir hosts collected from a large of sample sites from different landscape types and altitudes, with PCR and sequencing identifying the wide distribution and genetic diversity of endemic Borrelia burgdorferi sensu lato in Southwestern China. This was the first report that B. japonica was detected in Apodemus draco and Niviventer excelsior in China. This study adds to body of literature on Borrelia burgdorferi sensu lato in China. This work will provide insight regarding small mammals to target for surveillance and we access the association between gender, developmental stage of rodents, environmental landscape and altitude to better prevent human exposure.

epidemiology↗