Search

The objective of this study is to simultaneously assess the associations between suboptimal oral health (SOH) and cardiovascular disease (CVD) and competing death (CD).Ontario residents aged 40 years and over who participated in the Canadian Community Health Survey 2003 and 2007-2008 were followed until December 31, 2016 for the incidence of CVD or CD. SOH was assessed based on self-rated oral health and inability to chew. Multivariable competing risk analysis was adjusted for socioeconomic characteristics, behavioural factors and intermediate health outcomes.The study sample included 36,176 participants. Over a median follow-up of 9.61 years, there were 2077 CVD events and 3180 CD events. The fully adjusted models indicate 35% (HR = 1.35, 95% CI: 1.12-1.64) increase in the risk of CVD and 57% (HR = 1.57, 95% CI: 1.33-1.85) increase in the risk of CD among those who reported poor oral health as compared to those who reported excellent oral health. The fully adjusted models also indicate 11% (HR = 1.11, 95% CI: 0.97-1.27) increase in the hazard of CVD and 37% (HR = 1.37, 95% CI: 1.24-1.52) increase in the hazard of CD among those who reported inability to chew.This study provides important information to contextualize CVD risk among those with SOH. The competing risk analysis indicates that those with SOH may benefit from additional interventions to prevent CVD and CD. Accordingly, managing the risk of CVD among those with SOH should fall under a more comprehensive approach that aims at improving their overall health and well-being.RéSUMé: OBJECTIF: L’objectif de la présente étude est d’évaluer simultanément les liens entre la santé bucco-dentaire sous-optimale et les maladies cardiovasculaires (MCV) et le décès concurrent (DC). MéTHODES: Les résidents de l’Ontario de 40 ans et plus ayant participé à l’Enquête sur la santé dans les collectivités canadiennes 2003 et 2007-2008 ont fait l’objet d’un suivi évaluant les risques de MCV ou de DC. La santé bucco-dentaire sous-optimale a fait l’objet d’une évaluation axée sur l’autoévaluation de l’état de santé bucco-dentaire et l’incapacité à mastiquer. L’analyse multivariable des risques concurrents a été adaptée aux caractéristiques socioéconomiques, aux facteurs comportementaux et aux résultats intermédiaires en matière de santé. RéSULTATS: L’échantillon à l’étude comprenait 36 176 participants. Après un suivi médian de 9,61 ans, 2 077 incidents de MCV et 3 180 DC ont été recensés. La modélisation complète indique une hausse des risques de MCV de 35 % (indice de risque (IR) = 1,35, intervalle de confiance (IC) à 95 % : 1,12-1,64) et une hausse des risques de DC de 57 % (IR = 1,57, IC à 95 % : 1,33-1,85) parmi les participants ayant signalé une piètre santé bucco-dentaire comparativement aux personnes ayant déclaré une excellente santé bucco-dentaire. La modélisation complète indique également une hausse des risques de MCV de 11 % (IR = 1,11, IC à 95 % : 0,97-1,27) et une hausse des risques de DC de 37 % (IR = 1,37, IC à 95 % : 1,24-1,52) parmi les participants ayant déclaré une incapacité à mastiquer. CONCLUSION: La présente étude fournit d’importants renseignements nous permettant de contextualiser les risques de MCV chez les personnes ayant une piètre santé bucco-dentaire. L’analyse des risques concurrents indique que les personnes ayant une santé bucco-dentaire sous-optimale pourraient bénéficier d’interventions supplémentaires afin de prévenir les MCV et les DC. Par conséquent, la gestion des risques de MCV chez les personnes ayant une santé bucco-dentaire sous-optimale devrait relever d’une approche plus globale visant à améliorer la santé et le bien-être en général.

Disease-specific quality indicators (QIs) are used to compare institutions and health care providers in terms processes or outcomes relevant to treatment of a particular condition. In the context of surgical cancer treatments, the performance variations can be due to hospital and/or surgeon level differences, creating a hierarchical clustering. We consider how the observed variation in care received at patient level can be decomposed into that causally explained by the hospital performance, surgeon performance within hospital, patient case-mix, and unexplained (residual) variation. For this purpose, we derive a four-way variance decomposition, with particular attention to the causal interpretation of the components. For estimation, we use inputs from a mixed-effect model with nested random hospital/surgeon-specific effects, and a multinomial logistic model for the hospital/surgeon-specific patient populations. We investigate the performance of our methods in a simulation study.

Clinical information collected in electronic health records (EHRs) is becoming an essential source to emulate randomized experiments. Since patients do not interact with the healthcare system at random, the longitudinal information in large observational databases must account for irregular visits. Moreover, we need to also account for subject-specific unmeasured confounders which may act as a common cause for treatment assignment mechanism (e.g. glucose-lowering medications) while also influencing the outcome (e.g. Hemoglobin A1c). We used the calibration of longitudinal weights to improve the finite sample properties and to account for subject-specific unmeasured confounders. A Monte Carlo simulation study is conducted to evaluate the performance of calibrated inverse probability estimators using time-dependent treatment assignment and irregular visits with subject-specific unmeasured confounders. The simulation study showed that the longitudinal weights with calibrated restrictions improved the finite sample bias when compared to the stabilized weights. The application of the calibrated weights is demonstrated using the exposure of glucose lowering medications and the longitudinal outcome of Hemoglobin A1c. Our results support the effectiveness of glucose lowering medications in reducing Hemoglobin A1c among type II diabetes patients with elevated glycemic index ($$\ge 8.5 \%$$ ≥ 8.5 % ) using stabilized and calibrated weights.

Purpose We evaluated the level of oral hygiene and its association with oral health status and need for oral treatment among older residents in long-term care facilities. In addition, the association between oral hygiene level and health-related quality of life (HRQoL) was explored. Methods This cross-sectional study assessed 231 dentate residents in long-term care facilities (71% female, mean age 81 years, 70% had dementia). Nurses assessed residents and completed questionnaires on participants' background information, diagnoses, oral healthcare habits, and HRQoL with the 15D instrument. Two qualified dentists performed clinical oral examinations (number of teeth, plaque index, periodontal condition, open caries lesions, and dry mouth). We used a modified plaque index (PI) to measure the level of oral hygiene (good, moderate, and poor) and calculated the clinical Asymptotic Dental Score (ADS) to determine the oral inflammation burden. Results Of the residents, 21% had good, 35% moderate, and 44% poor oral hygiene according to PI. Poor oral hygiene was associated with poorer cognitive status (P = 0.010) and higher oral inflammation burden (P < 0.001). Moreover, poor oral hygiene was associated with poorer HRQoL in a correlation analysis adjusted for age and gender. Conclusions Oral hygiene of older individuals in long-term care is insufficient. Poor oral hygiene is a marker for poor HRQoL. Residents also have a high burden of oral inflammatory diseases and a need for dental care. Older residents' oral hygiene and HRQoL may be improved with oral care education of caregivers and regular dental check-ups. Key summary pointsAim We evaluated oral hygiene level and its association with oral health and general health-related quality of life (HRQoL) among older residents in long-term care facilities. Findings Only one-fifth of residents had good oral hygiene. Poor oral hygiene was associated with poor oral health and diminished HRQoL. Message Oral hygiene, oral health, and HRQoL may be improved with oral care education of caregivers, professional cooperation, and regular oral healthcare of older residents in long-term care facilities.

Causal variance decompositions for a given disease-specific quality indicator can be used to quantify differences in performance between hospitals or health care providers. While variance decompositions can demonstrate variation in quality of care, causal mediation analysis can be used to study care pathways leading to the differences in performance between the institutions. This raises the question of whether the two approaches can be combined to decompose between-hospital variation in an outcome type indicator to that mediated through a given process (indirect effect) and remaining variation due to all other pathways (direct effect). For this purpose, we derive a causal mediation analysis decomposition of between-hospital variance, discuss its interpretation, and propose an estimation approach based on generalized linear mixed models for the outcome and the mediator. We study the performance of the estimators in a simulation study and demonstrate its use in administrative data on kidney cancer care in Ontario.

The Swedish-speaking ethnolinguistic minority in Finland has a health advantage over the Finnish-speaking majority in terms of lower levels of sickness allowance and disability pension receipt. We examine whether this differential has consequences within the social security system by analysing the long-term risk for disability pension after sickness allowance receipt. With longitudinal population-register data covering the period 1989–2010, we observed 110,390 individuals aged 16–40 years at baseline. Using discrete-time hazard models, we estimated how first observed receipt of sickness allowance relates to disability pension receipt in the Finnish- and Swedish-speaking groups. Receipt of sickness allowance was found to be an important predictor for future disability pension receipt, but there were no systematic differences between the two ethnolinguistic groups in this respect, and the results were highly similar for men and women. These findings suggest that the social security system in Finland is non-discriminatory and provides equal treatment of both ethnolinguistic groups despite their overall health differential.

Bifidobacterium spp. are abundant gut commensals, especially in breast-fed infants. Bifidobacteria are associated with many health-promoting effects including maintenance of epithelial barrier and integrity as well as immunomodulation. However, the protective mechanisms of bifidobacteria on intestinal epithelium at molecular level are poorly understood. In this study, we developed a high-throughput in vitro screening assay to explore binding receptors of intestinal epithelial cells for Bifidobacterium bifidum. Short interfering RNAs (siRNA) were used to silence expression of each gene in the Caco-2 cell line one by one. The screen yielded four cell surface proteins, SERPINB3, LGICZ1, PKD1 and PAQR6, which were identified as potential receptors as the siRNA knock-down of their expression decreased adhesion of B. bifidum to the cell line repeatedly during the three rounds of siRNA screening. Furthermore, blocking of these host cell proteins by specific antibodies decreased the binding of B. bifidum significantly to Caco-2 and HT29 cell lines. All these molecules are located on the surface of epithelial cells and three out of four, SERPINB3, PKD1 and PAQR6, are involved in the regulation of cellular processes related to proliferation, differentiation and apoptosis as well as inflammation and immunity. Our results provide leads to the first steps in the mechanistic cascade of B. bifidum-host interactions leading to regulatory effects in the epithelium and may partly explain how this commensal bacterium is able to promote intestinal homeostasis.

Proton pump inhibitors (PPIs) are commonly prescribed medications that have been shown to have contradicting effects on cancer. We aimed to investigate the effect of pantoprazole and other PPIs on prostate cancer (PCa) specific mortality (PCSM), use of androgen deprivation therapy (ADT), and PCa diagnosis using a large Canadian population-based cohort. We identified 21,512 men aged ≥ 66, with a history of a single negative prostate biopsy and no previous use of any of the analyzed medications between 1994 and 2016. Multivariable Cox regression models with time-dependent covariates were used to assess the associations of PPIs with PCa outcomes. All models included other medications with a putative chemopreventative effect on PCa-outcomes, and were adjusted for age, rurality, comorbidity, and study inclusion year. Over a mean follow-up of 8.06 years (SD 5.44 years), 10,999 patients (51.1%) used a PPI, 5187 patients (24.1%) had PCa, 2043 patients (9.5%) were treated with ADT, and 805 patients (3.7%) died from PCa. For every 6 months of cumulative use, pantoprazole was associated with a 3.0% (95% CI 0.3–6.0%) increased rate of ADT use, while any use of other PPIs was associated with a 39.0% (95% CI 18.0–64.0%) increased risk of PCSM. No association was found with PCa diagnosis. Upon validation of the potentially negative association of PPIs with PCa, PPI use may need to be reassessed in PCa patients.

Forest management practices in boreal peatlands increase nutrient export and suspended solids to watercourses calling for development of new water protection methods. One potential solution could be adsorption-based purification of runoff water using biochar. The aim of this study was to determine the adsorption rate and capacity for Norway spruce and silver birch biochars to design a biochar-filled reactor for a ditch drain. In a 10-day laboratory experiment, biochar was stirred with runoff water from a clear-cut peatland forest, and changes in water pH, total nitrogen, nitrate nitrogen, ammonium nitrogen, phosphorus, and total organic carbon concentrations were measured. Based on the concentration changes, adsorption was quantified and adsorption model containing the adsorption rate and capacity was fitted to the data. Our results indicate that biochar effectively adsorbs both inorganic and organic nitrogen from runoff water. Birch biochar had higher adsorption capacity of nitrogen than spruce biochar. This study demonstrates that the adsorption of nitrogen compounds onto biochar surfaces increases with increasing initial concentrations. Thus, aquatic ecosystems exposed to high nutrient loads from fertile peatlands would particularly benefit from biochar-based water purification.

Forest regeneration operations increase the concentration of nitrogen (N) in watercourses especially outside the growing season when traditional biological water protection methods are inefficient. Biochar adsorption-based water treatment could be a solution for nutrient retention. We studied the total nitrogen (TN) and nitrate–nitrogen (NO3−–N) adsorption–desorption properties of spruce and birch biochar. The adsorption test was performed under four different initial concentrations of TN (1, 2, 3, and 4 mg L−1) using forest runoff water collected from ditch drains of boreal harvested peatland. The results showed that the TN adsorption amount increased linearly from the lowest to the highest concentration. The maximum adsorption capacity was 2.4 and 3.2 times greater in the highest concentration (4 mg L−1) compared to the lowest concentration (1 mg L−1) in spruce and birch biochar, respectively. The NO3−–N adsorption amount of birch biochar increased linearly from 0 to 0.15 mg NO3−–N g biochar−1 when the initial concentration of NO3−–N increased from 0.2 to 1.4 mg L−1. However, in spruce biochar, the initial concentration did not affect NO3−–N adsorption amount. The results indicate that concentration significantly affects the biochar’s capacity to adsorb N from water. The desorption test was performed by adding biochar extracted from the adsorption test into the forest runoff water with low TN concentration (0.2 or 0.35 mg L−1). The desorption results showed that desorption was negligibly small, and it was dependent on the TN concentration for birch biochar. Therefore, biochar can be a complementary method supporting water purification in peatland areas.

Studies have found that sibling loss is associated with an increased risk of death from external causes (i.e. suicides, accidents and homicides). Increased psychiatric health problems following bereavement could underlie such an association. We studied the influence of sibling loss during childhood on psychiatric care in young adulthood, adjusting for psychosocial covariates shared by siblings in childhood. A national cohort born in Sweden in 1973–1982 (N = 701,270) was followed prospectively until 2013. Cox proportional hazards models were used to analyse the association between sibling loss during childhood and psychiatric inpatient and outpatient care identified by the Hospital Discharge Register. After adjustment for confounders, the HRs of psychiatric care in men who experienced sibling loss were 1.17 (95% CI 1.07–1.27) while the associations turned non-significant in women after adjustment for family-related psychosocial covariates, HR 1.07 (95% CI 0.99–1.16). An increased risk was found in men bereaved in early childhood (1.22 95% CI 1.07–1.38) and adolescence (1.27 95% CI 1.08–1.48). Among women, loss of a sibling during adolescence was significantly associated with psychiatric care (1.19 95% CI 1.03–1.36). Increased psychiatric health problems following bereavement could underlie the previously found association between sibling loss and mortality from external causes. Family-related psychosocial conditions shared by siblings in childhood may account for the association between sibling death and psychiatric care in adulthood. Electronic supplementary material The online version of this article (10.1007/s00787-019-01324-6) contains supplementary material, which is available to authorized users.

Explainable artificial intelligence is an emerging research direction helping the user or developer of machine learning models understand why models behave the way they do. The most popular explanation technique is feature importance. However, there are several different approaches how feature importances are being measured, most notably global and local. In this study we compare different feature importance measures using both linear (logistic regression with L1 penalization) and non-linear (random forest) methods and local interpretable model-agnostic explanations on top of them. These methods are applied to two datasets from the medical domain, the openly available breast cancer data from the UCI Archive and a recently collected running injury data. Our results show that the most important features differ depending on the technique. We argue that a combination of several explanation techniques could provide more reliable and trustworthy results. In particular, local explanations should be used in the most critical cases such as false negatives.

Digitalisation has revolutionised health service delivery, which has provided global business opportunities for start-ups that specialise in digital innovations. Such start-ups challenge the traditional healthcare service industry by introducing radical and sustainable innovations in the agile product development cycle and the creative acquisition of resources within their networks. Starting up is the most critical period in establishing a new digital health service company. However, little is known about the critical early growth processes of newly established digital healthcare service businesses. The aim of this study is to clarify the experience-based priorities of managers of digital health service businesses in California during the critical start-up stage. Based on this multiple case study, the qualitative and contextual characteristics of growth in California-based digital health service start-ups were clarified, and a framework of management priorities was formed. Network management is a high priority in digital service start-ups that are focused on bringing radical innovations to the complex and hard-to-access market where fundraising is an integral part of success.

Diabetes and cancer are common diseases both with enormous impact on health burden globally. The increased risk of several types of cancer among people with type 2 diabetes mellitus has been indicated repeatedly. This study aimed at exploring and describing the association between type 2 diabetes and cancer incidence. A cohort of 428,326 people with type 2 diabetes was identified from the Finnish National Diabetes Register and followed up through a register linkage with the Finnish Cancer Registry for cancer incidence during 1988–2014. A total of 74,063 cases of cancer occurred in this cohort in 4.48 million person-years. This accounted for 16% more than the expected cancer incidence in the Finnish general population; the standardized incidence ratio (SIR) was 1.16 (95% confidence interval [CI] 1.15–1.16). There was a statistically significant excess of cancers of lip (SIR = 1.40, CI = 1.28–1.53), liver (SIR = 2.44, CI = 2.35–2.53), pancreas (SIR = 1.75, CI = 1.70–1.79), stomach (SIR = 1.22, CI = 1.18–1.26), colon (SIR = 1.22, CI = 1.19–1.25), gallbladder and bile ducts (SIR = 1.29, CI = 1.21–1.36), non-melanoma skin (SIR = 1.18, CI = 1.15–1.22), kidney (SIR = 1.42, CI = 1.37–1.47), bladder (SIR = 1.17, CI = 1.13–1.21), and thyroid (SIR = 1.22, CI = 1.12–1.31). There was a small statistically significant decrease in prostate cancer incidence (SIR = 0.95, CI = 0.93–0.96). This study showed an association between type 2 diabetes mellitus and the incidence of cancer at numerous sites in the Finnish population.