Tensorial Principal Component Analysis in Detecting Temporal Trajectories of Purchase Patterns in Loyalty Card Data: Retrospective Cohort Study.

BACKGROUND: Loyalty card data automatically collected by retailers provide an excellent source for evaluating health-related purchase behavior of customers. The data comprise information on every grocery purchase, including expenditures on product groups and the time of purchase for each customer. Such data where customers have an expenditure value for every product group for each time can be formulated as 3D tensorial data. OBJECTIVE: This study aimed to use the modern tensorial principal component analysis (PCA) method to uncover the characteristics of health-related purchase patterns from loyalty card data. Another aim was to identify card holders with distinct purchase patterns. We also considered the interpretation, advantages, and challenges of tensorial PCA compared with standard PCA. METHODS: Loyalty card program members from the largest retailer in Finland were invited to participate in this study. Our LoCard data consist of the purchases of 7251 card holders who consented to the use of their data from the year 2016. The purchases were reclassified into 55 product groups and aggregated across 52 weeks. The data were then analyzed using tensorial PCA, allowing us to effectively reduce the time and product group-wise dimensions simultaneously. The augmentation method was used for selecting the suitable number of principal components for the analysis. RESULTS: Using tensorial PCA, we were able to systematically search for typical food purchasing patterns across time and product groups as well as detect different purchasing behaviors across groups of card holders. For example, we identified customers who purchased large amounts of meat products and separated them further into groups based on time profiles, that is, customers whose purchases of meat remained stable, increased, or decreased throughout the year or varied between seasons of the year. CONCLUSIONS: Using tensorial PCA, we can effectively examine customers' purchasing behavior in more detail than with traditional methods because it can handle time and product group dimensions simultaneously. When interpreting the results, both time and product dimensions must be considered. In further analyses, these time and product groups can be directly associated with additional consumer characteristics such as socioeconomic and demographic predictors of dietary patterns. In addition, they can be linked to external factors that impact grocery purchases such as inflation and unexpected pandemics. This enables us to identify what types of people have specific purchasing patterns, which can help in the development of ways in which consumers can be steered toward making healthier food choices.

Data Type Agnostic Visual Sensitivity Analysis.

Modern science and industry rely on computational models for simulation, prediction, and data analysis. Spatial blind source separation (SBSS) is a model used to analyze spatial data. Designed explicitly for spatial data analysis, it is superior to popular non-spatial methods, like PCA. However, a challenge to its practical use is setting two complex tuning parameters, which requires parameter space analysis. In this paper, we focus on sensitivity analysis (SA). SBSS parameters and outputs are spatial data, which makes SA difficult as few SA approaches in the literature assume such complex data on both sides of the model. Based on the requirements in our design study with statistics experts, we developed a visual analytics prototype for data type agnostic visual sensitivity analysis that fits SBSS and other contexts. The main advantage of our approach is that it requires only dissimilarity measures for parameter settings and outputs (Fig. 1). We evaluated the prototype heuristically with visualization experts and through interviews with two SBSS experts. In addition, we show the transferability of our approach by applying it to microclimate simulations. Study participants could confirm suspected and known parameter-output relations, find surprising associations, and identify parameter subspaces to examine in the future. During our design study and evaluation, we identified challenging future research opportunities.

Multivariate Independent Component Analysis Identifies Patients in Newborn Screening Equally to Adjusted Reference Ranges.

Newborn screening (NBS) of inborn errors of metabolism (IEMs) is based on the reference ranges established on a healthy newborn population using quantile statistics of molar concentrations of biomarkers and their ratios. The aim of this paper is to investigate whether multivariate independent component analysis (ICA) is a useful tool for the analysis of NBS data, and also to address the structure of the calculated ICA scores. NBS data were obtained from a routine NBS program performed between 2013 and 2022. ICA was tested on 10,213/150 free-diseased controls and 77/20 patients (9/3 different IEMs) in the discovery/validation phases, respectively. The same model computed during the discovery phase was used in the validation phase to confirm its validity. The plots of ICA scores were constructed, and the results were evaluated based on 5sd levels. Patient samples from 7/3 different diseases were clearly identified as 5sd-outlying from control groups in both phases of the study. Two IEMs containing only one patient each were separated at the 3sd level in the discovery phase. Moreover, in one latent variable, the effect of neonatal birth weight was evident. The results strongly suggest that ICA, together with an interpretation derived from values of the "average member of the score structure", is generally applicable and has the potential to be included in the decision process in the NBS program.

A review of second-order blind identification methods.

Second-order source separation (SOS) is a data analysis tool which can be used for revealing hidden structures in multivariate time series data or as a tool for dimension reduction. Such methods are nowadays increasingly important as more and more high-dimensional multivariate time series data are measured in numerous fields of applied science. Dimension reduction is crucial, as modeling such high-dimensional data with multivariate time series models is often impractical as the number of parameters describing dependencies between the component time series is usually too high. SOS methods have their roots in the signal processing literature, where they were first used to separate source signals from an observed signal mixture. The SOS model assumes that the observed time series (signals) is a linear mixture of latent time series (sources) with uncorrelated components. The methods make use of the second-order statistics-hence the name "second-order source separation." In this review, we discuss the classical SOS methods and their extensions to more complex settings. An example illustrates how SOS can be performed. This article is categorized under:Statistical Models > Time Series ModelsStatistical and Graphical Methods of Data Analysis > Dimension ReductionData: Types and Structure > Time Series, Stochastic Processes, and Functional Data.

Blind Source Separation for Compositional Time Series.

Many geological phenomena are regularly measured over time to follow developments and changes. For many of these phenomena, the absolute values are not of interest, but rather the relative information, which means that the data are compositional time series. Thus, the serial nature and the compositional geometry should be considered when analyzing the data. Multivariate time series are already challenging, especially if they are higher dimensional, and latent variable models are a popular way to deal with this kind of data. Blind source separation techniques are well-established latent factor models for time series, with many variants covering quite different time series models. Here, several such methods and their assumptions are reviewed, and it is shown how they can be applied to high-dimensional compositional time series. Also, a novel blind source separation method is suggested which is quite flexible regarding the assumptions of the latent time series. The methodology is illustrated using simulations and in an application to light absorbance data from water samples taken from a small stream in Lower Austria.

Fast tensorial JADE.

We propose a novel method for tensorial-independent component analysis. Our approach is based on TJADE and k-JADE, two recently proposed generalizations of the classical JADE algorithm. Our novel method achieves the consistency and the limiting distribution of TJADE under mild assumptions and at the same time offers notable improvement in computational speed. Detailed mathematical proofs of the statistical properties of our method are given and, as a special case, a conjecture on the properties of k-JADE is resolved. Simulations and timing comparisons demonstrate remarkable gain in speed. Moreover, the desired efficiency is obtained approximately for finite samples. The method is applied successfully to large-scale video data, for which neither TJADE nor k-JADE is feasible. Finally, an experimental procedure is proposed to select the values of a set of tuning parameters. Supplementary material including the R-code for running the examples and the proofs of the theoretical results is available online.

Extracting Conditionally Heteroskedastic Components using Independent Component Analysis.

In the independent component model, the multivariate data are assumed to be a mixture of mutually independent latent components. The independent component analysis (ICA) then aims at estimating these latent components. In this article, we study an ICA method which combines the use of linear and quadratic autocorrelations to enable efficient estimation of various kinds of stationary time series. Statistical properties of the estimator are studied by finding its limiting distribution under general conditions, and the asymptotic variances are derived in the case of ARMA-GARCH model. We use the asymptotic results and a finite sample simulation study to compare different choices of a weight coefficient. As it is often of interest to identify all those components which exhibit stochastic volatility features we suggest a test statistic for this problem. We also show that a slightly modified version of the principal volatility component analysis can be seen as an ICA method. Finally, we apply the estimators in analysing a data set which consists of time series of exchange rates of seven currencies to US dollar. Supporting information including proofs of the theorems is available online.

Team-based "Get-a-Grip" lifestyle management programme in the treatment of obesity.

This study examined weight loss during an extensive 1-year lifestyle programme in primary care in Finland in overweight subjects (n = 134, age 18-69 years; BMI > 30, or BMI > 25 with a comorbidity that would benefit from weight loss) between 2009 and 2013 in a single arm design. The programme included four medical doctor visits, five sessions by a dietitian (advice on diet and on-location shopping behaviour), cooking classes, exercise supervised by personal trainer, and group discussions. A motivational interview method was applied. Of the 134 participants, 92 (69%) completed the 1-year programme. Among the participants 44% lost ≥ 5%, while 21% lost ≥ 10% of their initial body weight. In intention-to-treat-analyses, the mean weight loss during one year was 4.8 kg (p < 0.001). Mean BMI decreased by 1.7 kg/m2 (p < 0.001) and waist circumference by 5.6 cm (p < 0.001). Mean muscle mass increased by 3.3% (p < 0.001), and body fat decreased by 5.0% (p < 0.001). After the programme mean visceral fat content was reduced by 6.4%, systolic blood pressure by 8 mmHg (p < 0.001), and diastolic blood pressure by 6 mmHg (p < 0.001). In conclusion, retention to the team-based lifestyle management programme resulted in moderate but significant weight loss with beneficial changes in body composition, and the trend to lose weight was maintained throughout the year. Trial registration: Clinicaltrials.gov identifier NCT04003259.

On linear dimension reduction based on diagonalization of scatter matrices for bioinformatics downstream analyses.

Dimension reduction is often a preliminary step in the analysis of data sets with a large number of variables. Most classical, both supervised and unsupervised, dimension reduction methods such as principal component analysis (PCA), independent component analysis (ICA) or sliced inverse regression (SIR) can be formulated using one, two or several different scatter matrix functionals. Scatter matrices can be seen as different measures of multivariate dispersion and might highlight different features of the data and when compared might reveal interesting structures. Such analysis then searches for a projection onto an interesting (signal) part of the data, and it is also important to know the correct dimension of the signal subspace. These approaches usually make either no model assumptions or work in wide classes of semiparametric models. Theoretical results in the literature are however limited to the case where the sample size exceeds the number of variables which is hardly ever true for data sets encountered in bioinformatics. In this paper, we briefly review the relevant literature and explore if the dimension reduction tools can be used to find relevant and interesting subspaces for small-n-large-p data sets. We illustrate the methods with a microarray dataset of prostate cancer patients and healthy controls.

Tensorial blind source separation for improved analysis of multi-omic data.

There is an increased need for integrative analyses of multi-omic data. We present and benchmark a novel tensorial independent component analysis (tICA) algorithm against current state-of-the-art methods. We find that tICA outperforms competing methods in identifying biological sources of data variation at a reduced computational cost. On epigenetic data, tICA can identify methylation quantitative trait loci at high sensitivity. In the cancer context, tICA identifies gene modules whose expression variation across tumours is driven by copy-number or DNA methylation changes, but whose deregulation relative to normal tissue is independent of such alterations, a result we validate by direct analysis of individual data types.

Subgroup detection in genotype data using invariant coordinate selection.

BACKGROUND: The current gold standard in dimension reduction methods for high-throughput genotype data is the Principle Component Analysis (PCA). The presence of PCA is so dominant, that other methods usually cannot be found in the analyst's toolbox and hence are only rarely applied. RESULTS: We present a modern dimension reduction method called 'Invariant Coordinate Selection' (ICS) and its application to high-throughput genotype data. The more commonly known Independent Component Analysis (ICA) is in this framework just a special case of ICS. We use ICS on both, a simulated and a real dataset to demonstrate first some deficiencies of PCA and how ICS is capable to recover the correct subgroups within the simulated data. Second, we apply the ICS method on a chicken dataset and also detect there two subgroups. These subgroups are then further investigated with respect to their genotype to provide further evidence of the biological relevance of the detected subgroup division. Further, we compare the performance of ICS also to five other popular dimension reduction methods. CONCLUSION: The ICS method was able to detect subgroups in data where the PCA fails to detect anything. Hence, we promote the application of ICS to high-throughput genotype data in addition to the established PCA. Especially in statistical programming environments like e.g. R, its application does not add any computational burden to the analysis pipeline.

The type of the functional cardiovascular response to upright posture is associated with arterial stiffness: a cross-sectional study in 470 volunteers.

BACKGROUND: In a cross-sectional study we examined whether the haemodynamic response to upright posture could be divided into different functional phenotypes, and whether the observed phenotypes were associated with known determinants of cardiovascular risk. METHODS: Volunteers (n = 470) without medication with cardiovascular effects were examined using radial pulse wave analysis, whole-body impedance cardiography, and heart rate variability analysis. Based on the passive head-up tilt induced changes in systemic vascular resistance and cardiac output, the principal determinants of blood pressure, a cluster analysis was performed. RESULTS: The haemodynamic response could be clustered into 3 categories: upright increase in vascular resistance and decrease in cardiac output were greatest in the first (+45 % and -27 %, respectively), smallest in the second (+2 % and -2 %, respectively), and intermediate (+22 % and -13 %, respectively) in the third group. These groups were named as 'constrictor' (n = 109), 'sustainer' (n = 222), and 'intermediate' (n = 139) phenotypes, respectively. The sustainers were characterized by male predominance, higher body mass index, blood pressure, and also by higher pulse wave velocity, an index of large arterial stiffness, than the other groups (p < 0.01 for all). Heart rate variability analysis showed higher supine and upright low frequency/high frequency (LF/HF) ratio in the sustainers than constrictors, indicating increased sympathovagal balance. Upright LF/HF ratio was also higher in the sustainer than intermediate group. In multivariate analysis, independent explanatory factors for higher pulse wave velocity were the sustainer (p < 0.022) and intermediate phenotypes (p < 0.046), age (p < 0.001), body mass index (p < 0.001), and hypertension (p < 0.001). CONCLUSIONS: The response to upright posture could be clustered to 3 functional phenotypes. The sustainer phenotype, with smallest upright decrease in cardiac output and highest sympathovagal balance, was independently associated with increased large arterial stiffness. These results indicate an association of the functional haemodynamic phenotype with an acknowledged marker of cardiovascular risk. TRIAL REGISTRATION: ClinicalTrials.gov NCT01742702.

Effect of fermented milk product containing lactotripeptides and plant sterol esters on haemodynamics in subjects with the metabolic syndrome--a randomised, double-blind, placebo-controlled study.

We investigated the effects of fermented milk product containing isoleucine-proline-proline, valine-proline-proline and plant sterol esters (Pse) on plasma lipids, blood pressure (BP) and its determinants systemic vascular resistance and cardiac output. In a randomised, double-blind, placebo-controlled study, 104 subjects with the metabolic syndrome (MetS) were allocated to three groups in order to receive fermented milk product containing (1) 5 mg/d lactotripeptides (LTP) and 2 g/d plant sterols; (2) 25 mg/d LTP and 2 g/d plant sterols; (3) placebo for 12 weeks. Plasma lipids and home BP were monitored. Haemodynamics were examined in a laboratory using radial pulse wave analysis and whole-body impedance cardiography in the supine position and during orthostatic challenge. There were no differences between the effects of the two treatments and placebo on the measurements of BP at home or on BP, systemic vascular resistance index and cardiac index in the laboratory, neither in the supine nor in the upright position. The changes in plasma LDL-cholesterol concentration were - 0.1 (95% CI - 0.3, 0.1 and - 0.3, 0.0) mmol/l in the 5 and 25 mg/d LTP groups, respectively, and +0.1 (95% CI - 0.1, 0.3) mmol/l during placebo (P= 0.024). Both at baseline and at week 12, the increase in systemic vascular resistance during head-up tilt was lower in the 25 mg/d LTP group than in the 5 mg/d LTP group (P< 0.01), showing persistent differences in cardiovascular regulation between these groups. In subjects with the MetS, intake of LTP and Pse in fermented milk product showed a lipid-lowering effect of borderline significance, while no antihypertensive effect was observed at home or in the laboratory.

Vasopeptidase Inhibition Corrects the Structure and Function of the Small Arteries in Experimental Renal Insufficiency.

BACKGROUND: We studied whether vasopeptidase inhibition corrects the structure and function of the small arteries in experimental chronic renal insufficiency (CRI). METHODS: After 5/6 nephrectomy (NX) surgery was performed on rats, there was a 14-week follow-up, allowing CRI to become established. Omapatrilat (40 mg/kg/day in chow) was then given for 8 weeks, and the small mesenteric arterial rings were investigated in vitro using wire and pressure myographs. RESULTS: Plasma and ventricular B-type natriuretic peptide (BNP) concentrations were increased 2- to 2.7-fold, while systolic blood pressure (BP) increased by 32 mm Hg after NX. Omapatrilat treatment normalized the BNP and reduced the BP by 45 mm Hg in the NX rats. Endothelium-dependent vasorelaxation was impaired but the response to acetylcholine was normalized after omapatrilat treatment. Vasorelaxations induced by nitroprusside, isoprenaline and levcromakalim were enhanced after omapatrilat, and the responses were even more pronounced than in untreated sham-operated rats. Arterial wall thickness and wall-to-lumen ratio were increased after NX, whereas omapatrilat normalized these structural features and improved the strain-stress relationship in the small arteries; this suggests improved arterial elastic properties. CONCLUSION: Omapatrilat treatment reduced BP, normalized volume overload, improved vasorelaxation and corrected the dimensions and passive elastic properties of the small arteries in the NX rats. Therefore, we consider vasopeptidase inhibition to be an effective treatment for CRI-induced changes in the small arteries.

Supine and upright haemodynamic effects of sublingual nitroglycerin and inhaled salbutamol: a double-blind, placebo-controlled, randomized study.

OBJECTIVES: Pulse wave analysis is widely applied to measure the haemodynamic effects of nitroglycerin and salbutamol as an endothelium-independent and endothelium-dependent vasodilator, respectively. The recordings are usually performed in supine position from 10 to 20 successive heartbeats without simultaneous measurement of vascular resistance and cardiac function. Our objective was to examine the effects of nitroglycerin and salbutamol on central haemodynamics, arterial stiffness, cardiac function, and vascular resistance in supine and upright positions. METHODS: A placebo-controlled, randomized and double-blinded passive head-up tilt protocol was performed after sublingual nitroglycerin (0.25 mg) or inhaled salbutamol (400 µg) in 35 healthy volunteers. Continuous tonometric pulse wave analysis, whole-body impedance cardiography, and plethysmographic finger blood pressure recordings were applied. RESULTS: Nitroglycerin decreased aortic and finger blood pressure, radial DBP, vascular resistance, augmentation index and pulse wave velocity, and increased heart rate, cardiac index, stroke index and aortic reflection time (P < 0.030 for all). Salbutamol moderately decreased radial and aortic blood pressure and finger DBP, augmentation index and vascular resistance, but increased heart rate and cardiac index (P < 0.030 for all). Almost all of the strong haemodynamic effects of nitroglycerin were emphasized during the head-up tilt, whereas the effects of salbutamol on heart rate and cardiac index were more pronounced in the supine position. CONCLUSION: The haemodynamic changes induced by nitroglycerin and salbutamol were dependent on body position: the effects of nitroglycerin were accentuated during the head-up tilt, whereas those of salbutamol were more evident in the supine position.

Reduced systemic vascular resistance in healthy volunteers with presyncopal symptoms during a nitrate-stimulated tilt-table test.

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT: Nitrates may facilitate syncope through various pathways, but the precise mechanism of nitrate-induced syncope is still under debate. The purpose of the present study was to compare the underlying haemodynamic mechanisms in subjects without and with presyncopal symptoms during a nitroglycerin-stimulated tilt-table test. WHAT THIS STUDY ADDS: A major decrease in systemic vascular resistance was documented in subjects with presyncope during 0.25 mg nitroglycerin-stimulated tilt-table test, in the absence of changes in cardiac output. These findings indicated that even a small dose of nitroglycerin significantly decreased arterial resistance and cardiac afterload. AIMS The mechanism of nitrate-induced syncope remains controversial. We examined the haemodynamic changes in healthy volunteers during nitroglycerin-stimulated tilt-table test. METHODS: Continuous radial pulse wave analysis, whole-body impedance cardiography and plethysmographic finger blood pressure were recorded in a supine position and during head-up tilt in 21 subjects with presyncopal symptoms (6 male/15 female, age 43 ± 3 years) after 0.25 mg sublingual nitroglycerin and 21 control subjects (6 male/15 female, age 43 ± 2 years). The drug was administered in the supine position and a passive head-up tilt followed 5 min later. Additionally, nitroglycerin was only administered during head-up tilt in 19 subjects and the haemodynamics were recorded. RESULTS: Supine and upright haemodynamics were similar before nitroglycerin administration in the two groups. During the nitroglycerin-stimulated tilt test, aortic and radial mean blood pressure decreased significantly more in the presyncope group when compared with the controls (P= 0.0006 and P= 0.0004, respectively). The decreases in systemic vascular resistance (P= 0.0008) and heart rate (P= 0.002), and increase in aortic reflection time (P= 0.0002) were greater in the presyncope group, while the change in cardiac index was not different between the groups (P= 0.14). If nitroglycerin was administered during the upright tilt and not in supine position, the haemodynamic changes were quite corresponding. CONCLUSIONS: Presyncopal symptoms during nitrate-stimulated tilt test were explained by decreased systemic vascular resistance and increased aortic reflection time, while cardiac output remained unchanged. These findings indicated reduced arterial resistance in nitroglycerin-induced presyncope.

Persistent pain following knee arthroplasty.

BACKGROUND AND OBJECTIVE: The prevalence of persistent pain after orthopaedic surgery has been the subject of only few studies and the risk factors for persistent pain have been evaluated even more rarely. The purpose of the present study was to evaluate the degree and the risk factors of persistent pain after knee arthroplasty. METHODS: The prevalence of persistent postoperative pain after knee replacement was evaluated with a questionnaire in a large, register-based cross-sectional prevalence study. The main hypothesis was that the type of operation (primary, bilateral, revision) would influence the prevalence of persistent postoperative pain. Logistic regression analysis was performed to test the hypothesis and to find other possible risk factors for the development of persistent pain. RESULTS: The total number of patients was 855. The operation was a primary arthroplasty in 648 patients (75.7%), a bilateral arthroplasty in 137 patients (21.1%) and a revision arthroplasty in 70 patients (8.2%). The response rate was 65.7%. The type of operation was not associated with the prevalence of persistent pain, but the degree of early postoperative pain was the strongest risk factor. If the degree of pain during the first postoperative week was from moderate to intolerable, the risk for the development of persistent pain was three to 10 times higher compared with patients complaining of mild pain during the same period. Other risk factors were the long duration of preoperative pain and female sex. CONCLUSION: Intensity of early postoperative pain and delayed surgery increase the risk of the persistent pain after knee arthroplasty.

Ageing and cardiovascular responses to head-up tilt in healthy subjects.

OBJECTIVE: Ageing is associated with increased postural blood pressure changes and attenuated cardiovascular reactivity. As the background of these changes remains uncertain, we examined arterial stiffness, cardiac function and vascular resistance in healthy subjects in supine position and during orthostatic challenge. METHODS: Haemodynamics of 179 normotensive subjects (109 female and 70 male, 21-59 years) were examined using continuous radial pulse wave analysis, whole-body impedance cardiography, and plethysmographic finger blood pressures. RESULTS: Both in supine position and during head-up tilt central and peripheral blood pressure and augmentation index (amplitude of the reflected pressure wave divided by pulse pressure) increased, and time of pulse wave reflection decreased with age. Supine pulse wave velocity progressively increased with age. There were only minor differences in supine and upright systemic vascular resistance, cardiac index, stroke index, and heart rate that were not systematically related to age. In regression analysis, the explanatory factors for a more pronounced decrease in central systolic blood pressure during the head-up tilt were in the age of 50-59 years, higher baseline pulse wave velocity and central systolic BP. None of the changes in other haemodynamic variables during tilt were related to age. CONCLUSION: As no systematic age-related differences were observed in cardiac function or vascular resistance, these results support the view that progressive reduction of large arterial compliance contributes to the exaggerated age-related decrease in central systolic blood pressure in response to head-up tilt.

Reanalysis of systematic reviews: the case of invasive strategies for acute coronary syndromes.

OBJECTIVES: The objective of this study was to collect all systematic reviews on invasive strategies for acute coronary syndromes (ACS) and reanalyze the data in these reviews to reach combined estimates, as well as to make predictions on the effectiveness and risk of harm so as to facilitate relevant decision making in health care. METHODS: The data sources used were the following electronic databases, searched from 1994 to September 2004: Cochrane Database of Systematic Reviews; Cochrane Central Register of Controlled Trials; DARE, HTA, EED (NHS CRD); MEDLINE In-Process, Other Non-Indexed Citations, MEDLINE, and PubMed (2000 to 2004). References to the identified systematic reviews were checked. An ancillary search to identify recent randomized controlled trials (RCTs) covering the period from January 2003 to January 2006 was done in MEDLINE(R). We included systematic reviews of RCTs on patients with ACS. In unstable angina and non-ST-elevation myocardial infarction (UA/NSTEMI), eligible reviews had to compare early routine invasive strategy with early selective invasive strategy. In ST-elevation myocardial infarction (STEMI), a comparison between primary percutaneous coronary intervention (PCI) and thrombolytic therapy was required. The methodological quality of the reviews was assessed, and a standardized data extraction form was used. Results for the main outcomes of the RCTs in the reviews were reanalyzed. An additional search of those RCTs not included in the meta-analyses was performed for UA/NSTEMI and short-term morality data on STEMI. Bayesian models were constructed to estimate the uncertainty about a possible treatment effect and to make predictions and probability statements. Main results are based on these analyses. Mortality was considered as the primary outcome measure. RESULTS: One systematic review on invasive strategies was identified for UA/NSTEMI and nine on invasive strategies for STEMI. Five reviews of the latter that were published after the year 2000 were included for the final analysis. The median quality score was 10.5 (range, 7-13; n = 6) on a scale from 0 to 18 points. An updated literature search identified one further RCT on UA/NSTEMI. Regarding NSTEMI and mortality, the average risk difference favoring an early invasive treatment strategy compared with early conservative strategy was .6 percent (95 percent credible interval [CrI], -2.1 to 1.0). Predicted risk (relative risk/risk difference scales) of doing harm was 26.7/26.6 percent. Regarding STEMI and mortality, the absolute risk reduction in favor of primary PCI over thrombolysis was 4.1 percent (95 percent CrI, -7.1 to -1.1) when PCI was compared with streptokinase and 1.2 percent (95 percent CrI, -2.7 to .2) when compared with fibrin-specific thrombolytics. Predicted risk of harm was 8.9/5.3 percent and 8.0/13.3 percent, respectively. CONCLUSIONS: There seems to be at present no solid evidence for survival benefit on early invasive strategy for UA/NSTEMI as a broad diagnostic group, and the risk of doing harm should be considered. Also, the evidence for PCI to decrease early mortality after STEMI is scanty. Estimations of predicted harm may further aid decisions on whether to implement the new treatment over the old one. It may also give an additional dimension for interpreting the results of any meta-analysis.