Systematic review of medication safety assessment methods

Purpose The accuracy, efficiency, and efficacy of four commonly recommended medication safety assessment methodologies were systematically reviewed. Methods Medical literature databases were systematically searched for any comparative study conducted between January 2000 and October 2009 in which at least two of the four methodologies—incident report review, direct observation, chart review, and trigger tool—were compared with one another. Any study that compared two or more methodologies for quantitative accuracy (adequacy of the assessment of medication errors and adverse drug events) efficiency (effort and cost), and efficacy and that provided numerical data was included in the analysis. Results Twenty-eight studies were included in this review. Of these, 22 compared two of the methodologies, and 6 compared three methods. Direct observation identified the greatest number of reports of drug-related problems (DRPs), while incident report review identified the fewest. However, incident report review generally showed a higher specificity compared to the other methods and most effectively captured severe DRPs. In contrast, the sensitivity of incident report review was lower when compared with trigger tool. While trigger tool was the least labor-intensive of the four methodologies, incident report review appeared to be the least expensive, but only when linked with concomitant automated reporting systems and targeted follow-up. Conclusion All four medication safety assessment techniques—incident report review, chart review, direct observation, and trigger tool—have different strengths and weaknesses. Overlap between different methods in identifying DRPs is minimal. While trigger tool appeared to be the most effective and labor-efficient method, incident report review best identified high-severity DRPs.

Mortality from road traffic accidents in Switzerland: longitudinal and spatial analyses

Road traffic accidents (RTA) are an important cause of premature death. We examined socio-demographic and geographical determinants of RTA mortality in Switzerland by linking 2000 census data to RTA mortality records 2000-2005 (ICD-10 codes V00-V99). Data from 5.5 million residents aged 18-94 years, 1744 study areas, and 1620 RTA deaths were analyzed, including 978 deaths (60.4%) in motor vehicle occupants, 254 (15.7%) in motorcyclists, 107 (6.6%) in cyclists, and 259 (16.0%) in pedestrians. Weibull survival models and Bayesian methods were used to calculate hazard ratios (HR), and standardized mortality ratios (SMR) across study areas. Adjusted HR comparing women with men ranged from 0.04 (95% CI 0.02-0.07) in motorcyclists to 0.43 (95% CI 0.32-0.56) in pedestrians. There was a u-shaped relationship with age in motor vehicle occupants and motorcyclists. In cyclists and pedestrians, mortality increased after age 55 years. Mortality was higher in individuals with primary education (HR 1.53; 95% CI 1.29-1.81), and higher in single (HR 1.24; 95% CI 1.05-1.46), widowed (HR 1.31; 95% CI 1.05-1.65) and divorced individuals (HR 1.62; 95% CI 1.33-1.97), compared to persons with tertiary education or married persons. The association with education was particularly strong for pedestrians (HR 1.87; 95% CI 1.20-2.91). RTA mortality increased with decreasing population density of study areas for motor vehicle occupants (test for trend p<0.0001) and motorcyclists (p=0.0021) but not for cyclists (p=0.39) or pedestrians (p=0.29). SMR standardized for socio-demographic and geographical variables ranged from 82 to 190. Prevention efforts should aim to reduce inequities across socio-demographic and educational groups, and across geographical areas, with interventions targeted at high-risk groups and areas, and different traffic users, including pedestrians.

Bayesian statistics in oncology: a guide for the clinical investigator

The rise of evidence-based medicine as well as important progress in statistical methods and computational power have led to a second birth of the >200-year-old Bayesian framework. The use of Bayesian techniques, in particular in the design and interpretation of clinical trials, offers several substantial advantages over the classical statistical approach. First, in contrast to classical statistics, Bayesian analysis allows a direct statement regarding the probability that a treatment was beneficial. Second, Bayesian statistics allow the researcher to incorporate any prior information in the analysis of the experimental results. Third, Bayesian methods can efficiently handle complex statistical models, which are suited for advanced clinical trial designs. Finally, Bayesian statistics encourage a thorough consideration and presentation of the assumptions underlying an analysis, which enables the reader to fully appraise the authors' conclusions. Both Bayesian and classical statistics have their respective strengths and limitations and should be viewed as being complementary to each other; we do not attempt to make a head-to-head comparison, as this is beyond the scope of the present review. Rather, the objective of the present article is to provide a nonmathematical, reader-friendly overview of the current practice of Bayesian statistics coupled with numerous intuitive examples from the field of oncology. It is hoped that this educational review will be a useful resource to the oncologist and result in a better understanding of the scope, strengths, and limitations of the Bayesian approach.

Public health risks of enterobacterial isolates producing extended-spectrum β-lactamases or AmpC β-lactamases in food and food-producing animals: an EU perspective of epidemiology, analytical methods, risk factors, and control options

The blaESBL and blaAmpC genes in Enterobacteriaceae are spread by plasmid-mediated integrons, insertion sequences, and transposons, some of which are homologous in bacteria from food animals, foods, and humans. These genes have been frequently identified in Escherichia coli and Salmonella from food animals, the most common being blaCTX-M-1, blaCTX-M-14, and blaCMY-2. Identification of risk factors for their occurrence in food animals is complex. In addition to generic antimicrobial use, cephalosporin usage is an important risk factor for selection and spread of these genes. Extensive international trade of animals is a further risk factor. There are no data on the effectiveness of individual control options in reducing public health risks. A highly effective option would be to stop or restrict cephalosporin usage in food animals. Decreasing total antimicrobial use is also of high priority. Implementation of measures to limit strain dissemination (increasing farm biosecurity, controls in animal trade, and other general postharvest controls) are also important.

Taking up national safety alerts to improve patient safety in hospitals: The perspective of healthcare quality and risk managers.

BACKGROUND National safety alert systems publish relevant information to improve patient safety in hospitals. However, the information has to be transformed into local action to have an effect on patient safety. We studied three research questions: How do Swiss healthcare quality and risk managers (qm/rm(1)) see their own role in learning from safety alerts issued by the Swiss national voluntary reporting and analysis system? What are their attitudes towards and evaluations of the alerts, and which types of improvement actions were fostered by the safety alerts? METHODS A survey was developed and applied to Swiss healthcare risk and quality managers, with a response rate of 39 % (n=116). Descriptive statistics are presented. RESULTS The qm/rm disseminate and communicate with a broad variety of professional groups about the alerts. While most respondents felt that they should know the alerts and their contents, only a part of them felt responsible for driving organizational change based on the recommendations. However, most respondents used safety alerts to back up their own patient safety goals. The alerts were evaluated positively on various dimensions such as usefulness and were considered as standards of good practice by the majority of the respondents. A range of organizational responses was applied, with disseminating information being the most common. An active role is related to using safety alerts for backing up own patient safety goals. CONCLUSIONS To support an active role of qm/rm in their hospital's learning from safety alerts, appropriate organizational structures should be developed. Furthermore, they could be given special information or training to act as an information hub on the issues discussed in the alerts.

Immune reconstitution inflammatory syndrome in patients starting antiretroviral therapy for HIV infection: a systematic review and meta-analysis

In patients with HIV-1 infection who are starting combination antiretroviral therapy (ART), the incidence of immune reconstitution inflammatory syndrome (IRIS) is not well defined. We did a meta-analysis to establish the incidence and lethality of the syndrome in patients with a range of previously diagnosed opportunistic infections, and examined the relation between occurrence and the degree of immunodeficiency. Systematic review identified 54 cohort studies of 13 103 patients starting ART, of whom 1699 developed IRIS. We calculated pooled cumulative incidences with 95% credibility intervals (CrI) by Bayesian methods and did a random-effects metaregression to analyse the relation between CD4 cell count and incidence of IRIS. In patients with previously diagnosed AIDS-defining illnesses, IRIS developed in 37.7% (95% CrI 26.6-49.4) of those with cytomegalovirus retinitis, 19.5% (6.7-44.8) of those with cryptococcal meningitis, 15.7% (9.7-24.5) of those with tuberculosis, 16.7% (2.3-50.7) of those with progressive multifocal leukoencephalopathy, and 6.4% (1.2-24.7) of those with Kaposi's sarcoma, and 12.2% (6.8-19.6) of those with herpes zoster. 16.1% (11.1-22.9) of unselected patients starting ART developed any type of IRIS. 4.5% (2.1-8.6) of patients with any type of IRIS died, 3.2% (0.7-9.2) of those with tuberculosis-associated IRIS died, and 20.8% (5.0-52.7) of those with cryptococcal meningitis died. Metaregression analyses showed that the risk of IRIS is associated with CD4 cell count at the start of ART, with a high risk in patients with fewer than 50 cells per microL. Occurrence of IRIS might therefore be reduced by initiation of ART before immunodeficiency becomes advanced.

Patterns of morphological changes and hybridization between sympatric whitefish morphs (Coregonus spp.) in a Swiss lake: a role for eutrophication?

Whitefish, genus Coregonus, show exceptional levels of phenotypic diversity with sympatric morphs occurring in numerous postglacial lakes in the northern hemisphere. Here, we studied the effects of human-induced eutrophication on sympatric whitefish morphs in the Swiss lake, Lake Thun. In particular, we addressed the questions whether eutrophication (i) induced hybridization between two ecologically divergent summer-spawning morphs through a loss of environmental heterogeneity, and (ii) induced rapid adaptive morphological changes through changes in the food web structure. Genetic analysis based on 11 microsatellite loci of 282 spawners revealed that the pelagic and the benthic morph represent highly distinct gene pools occurring at different relative proportions on all seven known spawning sites. Gill raker counts, a highly heritable trait, showed nearly discrete distributions for the two morphs. Multilocus genotypes characteristic of the pelagic morph had more gill rakers than genotypes characteristic of benthic morph. Using Bayesian methods, we found indications of recent but limited introgressive hybridization. Comparisons with historical gill raker data yielded median evolutionary rates of 0.24 haldanes and median selection intensities of 0.27 for this trait in both morphs for 1948-2004 suggesting rapid evolution through directional selection at this trait. However, phenotypic plasticity as an alternative explanation for this phenotypic change cannot be discarded. We hypothesize that both the temporal shifts in mean gill raker counts and the recent hybridization reflect responses to changes in the trophic state of the lake induced by pollution in the 1960s, which created novel selection pressures with respect to feeding niches and spawning site preferences.

General and specific components of depression and anxiety in an adolescent population

Background Depressive and anxiety symptoms often co-occur resulting in a debate about common and distinct features of depression and anxiety. Methods An exploratory factor analysis (EFA) and a bifactor modelling approach were used to separate a general distress continuum from more specific sub-domains of depression and anxiety in an adolescent community sample (n = 1159, age 14). The Mood and Feelings Questionnaire and the Revised Children's Manifest Anxiety Scale were used. Results A three-factor confirmatory factor analysis is reported which identified a) mood and social-cognitive symptoms of depression, b) worrying symptoms, and c) somatic and information-processing symptoms as distinct yet closely related constructs. Subsequent bifactor modelling supported a general distress factor which accounted for the communality of the depression and anxiety items. Specific factors for hopelessness-suicidal thoughts and restlessness-fatigue indicated distinct psychopathological constructs which account for unique information over and above the general distress factor. The general distress factor and the hopelessness-suicidal factor were more severe in females but the restlessness-fatigue factor worse in males. Measurement precision of the general distress factor was higher and spanned a wider range of the population than any of the three first-order factors. Conclusions The general distress factor provides the most reliable target for epidemiological analysis but specific factors may help to refine valid phenotype dimensions for aetiological research and assist in prognostic modelling of future psychiatric episodes.

Toward an early diagnosis of lung cancer: an autoantibody signature for squamous cell lung carcinoma

Serum-based diagnosis offers the prospect of early lung carcinoma detection and of differentiation between benign and malignant nodules identified by CT. One major challenge toward a future blood-based diagnostic consists in showing that seroreactivity patterns allow for discriminating lung cancer patients not only from normal controls but also from patients with non-tumor lung pathologies. We addressed this question for squamous cell lung cancer, one of the most common lung tumor types. Using a panel of 82 phage-peptide clones, which express potential autoantigens, we performed serological spot assay. We screened 108 sera, including 39 sera from squamous cell lung cancer patients, 29 sera from patients with other non-tumor lung pathologies, and 40 sera from volunteers without known disease. To classify the serum groups, we employed the standard Naïve Bayesian method combined with a subset selection approach. We were able to separate squamous cell lung carcinoma and normal sera with an accuracy of 93%. Low-grade squamous cell lung carcinoma were separated from normal sera with an accuracy of 92.9%. We were able to distinguish squamous cell lung carcinoma from non-tumor lung pathologies with an accuracy of 83%. Three phage-peptide clones with sequence homology to ROCK1, PRKCB1 and KIAA0376 reacted with more than 15% of the cancer sera, but neither with normal nor with non-tumor lung pathology sera. Our study demonstrates that seroreactivity profiles combined with statistical classification methods have great potential for discriminating patients with squamous cell lung carcinoma not only from normal controls but also from patients with non-tumor lung pathologies.

Intra-articular corticosteroid for knee osteoarthritis.

BACKGROUND Knee osteoarthritis is a leading cause of chronic pain, disability, and decreased quality of life. Despite the long-standing use of intra-articular corticosteroids, there is an ongoing debate about their benefits and safety. This is an update of a Cochrane review first published in 2005. OBJECTIVES To determine the benefits and harms of intra-articular corticosteroids compared with sham or no intervention in people with knee osteoarthritis in terms of pain, physical function, quality of life, and safety. SEARCH METHODS We searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, and EMBASE (from inception to 3 February 2015), checked trial registers, conference proceedings, reference lists, and contacted authors. SELECTION CRITERIA We included randomised or quasi-randomised controlled trials that compared intra-articular corticosteroids with sham injection or no treatment in people with knee osteoarthritis. We applied no language restrictions. DATA COLLECTION AND ANALYSIS We calculated standardised mean differences (SMDs) and 95% confidence intervals (CI) for pain, function, quality of life, joint space narrowing, and risk ratios (RRs) for safety outcomes. We combined trials using an inverse-variance random-effects meta-analysis. MAIN RESULTS We identified 27 trials (13 new studies) with 1767 participants in this update. We graded the quality of the evidence as 'low' for all outcomes because treatment effect estimates were inconsistent with great variation across trials, pooled estimates were imprecise and did not rule out relevant or irrelevant clinical effects, and because most trials had a high or unclear risk of bias. Intra-articular corticosteroids appeared to be more beneficial in pain reduction than control interventions (SMD -0.40, 95% CI -0.58 to -0.22), which corresponds to a difference in pain scores of 1.0 cm on a 10-cm visual analogue scale between corticosteroids and sham injection and translates into a number needed to treat for an additional beneficial outcome (NNTB) of 8 (95% CI 6 to 13). An I(2) statistic of 68% indicated considerable between-trial heterogeneity. A visual inspection of the funnel plot suggested some asymmetry (asymmetry coefficient -1.21, 95%CI -3.58 to 1.17). When stratifying results according to length of follow-up, benefits were moderate at 1 to 2 weeks after end of treatment (SMD -0.48, 95% CI -0.70 to -0.27), small to moderate at 4 to 6 weeks (SMD -0.41, 95% CI -0.61 to -0.21), small at 13 weeks (SMD -0.22, 95% CI -0.44 to 0.00), and no evidence of an effect at 26 weeks (SMD -0.07, 95% CI -0.25 to 0.11). An I(2) statistic of ≥ 63% indicated a moderate to large degree of between-trial heterogeneity up to 13 weeks after end of treatment (P for heterogeneity≤0.001), and an I(2) of 0% indicated low heterogeneity at 26 weeks (P=0.43). There was evidence of lower treatment effects in trials that randomised on average at least 50 participants per group (P=0.05) or at least 100 participants per group (P=0.013), in trials that used concomittant viscosupplementation (P=0.08), and in trials that used concomitant joint lavage (P≤0.001).Corticosteroids appeared to be more effective in function improvement than control interventions (SMD -0.33, 95% CI -0.56 to -0.09), which corresponds to a difference in functions scores of -0.7 units on standardised Western Ontario and McMaster Universities Arthritis Index (WOMAC) disability scale ranging from 0 to 10 and translates into a NNTB of 10 (95% CI 7 to 33). An I(2) statistic of 69% indicated a moderate to large degree of between-trial heterogeneity. A visual inspection of the funnel plot suggested asymmetry (asymmetry coefficient -4.07, 95% CI -8.08 to -0.05). When stratifying results according to length of follow-up, benefits were small to moderate at 1 to 2 weeks after end of treatment (SMD -0.43, 95% CI -0.72 to -0.14), small to moderate at 4 to 6 weeks (SMD -0.36, 95% CI -0.63 to -0.09), and no evidence of an effect at 13 weeks (SMD -0.13, 95% CI -0.37 to 0.10) or at 26 weeks (SMD 0.06, 95% CI -0.16 to 0.28). An I(2) statistic of ≥ 62% indicated a moderate to large degree of between-trial heterogeneity up to 13 weeks after end of treatment (P for heterogeneity≤0.004), and an I(2) of 0% indicated low heterogeneity at 26 weeks (P=0.52). We found evidence of lower treatment effects in trials that randomised on average at least 50 participants per group (P=0.023), in unpublished trials (P=0.023), in trials that used non-intervention controls (P=0.031), and in trials that used concomitant viscosupplementation (P=0.06).Participants on corticosteroids were 11% less likely to experience adverse events, but confidence intervals included the null effect (RR 0.89, 95% CI 0.64 to 1.23, I(2)=0%). Participants on corticosteroids were 67% less likely to withdraw because of adverse events, but confidence intervals were wide and included the null effect (RR 0.33, 95% CI 0.05 to 2.07, I(2)=0%). Participants on corticosteroids were 27% less likely to experience any serious adverse event, but confidence intervals were wide and included the null effect (RR 0.63, 95% CI 0.15 to 2.67, I(2)=0%).We found no evidence of an effect of corticosteroids on quality of life compared to control (SMD -0.01, 95% CI -0.30 to 0.28, I(2)=0%). There was also no evidence of an effect of corticosteroids on joint space narrowing compared to control interventions (SMD -0.02, 95% CI -0.49 to 0.46). AUTHORS' CONCLUSIONS Whether there are clinically important benefits of intra-articular corticosteroids after one to six weeks remains unclear in view of the overall quality of the evidence, considerable heterogeneity between trials, and evidence of small-study effects. A single trial included in this review described adequate measures to minimise biases and did not find any benefit of intra-articular corticosteroids.In this update of the systematic review and meta-analysis, we found most of the identified trials that compared intra-articular corticosteroids with sham or non-intervention control small and hampered by low methodological quality. An analysis of multiple time points suggested that effects decrease over time, and our analysis provided no evidence that an effect remains six months after a corticosteroid injection.

Effects of subthalamic nucleus stimulation on cognition and mood in Parkinson’s disease (PD)

Objective: To assess the neuropsychological outcome as a safety measure and quality control in patients with subthalamic nucleus (STN) stimulation for PD. Background: Deep brain stimulation (DBS) is considered a relatively safe treatment used in patients with movement disorders. However, neuropsychological alterations have been reported in patients with STN DBS for PD. Cognition and mood are important determinants of quality of life in PD patients and must be assessed for safety control. Methods: Seventeen consecutive patients (8 women) who underwent STN DBS for PD have been assessed before and 4 months after surgery. Besides motor symptoms (UPDRS-III), mood (Beck Depression Inventory, Hamilton Depression Rating Scale) and neuropsychological aspects, mainly executive functions, have been assessed (mini mental state examination, semantic and phonematic verbal fluency, go-no go test, stroop test, trail making test, tests of alertness and attention, digit span, wordlist learning, praxia, Boston naming test, figure drawing, visual perception). Paired t-tests were used for comparisons before and after surgery. Results: Patients were 61.6±7.8 years old at baseline assessment. All surgeries were performed without major adverse events. Motor symptoms ‘‘on’’ medication remained stable whereas they improved in the ‘‘off’’ condition (p<0.001). Mood was not depressed before surgery and remained unchanged at follow-up. All neuropsychological assessment outcome measures remained stable at follow-up with the exception of semantic verbal fluency and wordlist learning. Semantic verbal fluency decreased by 21±16% (p<0.001) and there was a trend to worse phonematic verbal fluency after surgery (p=0.06). Recall of a list of 10 words was worse after surgery only for the third attempt of recall (13%, p<0.005). Conclusions: Verbal fluency decreased in our patients after STN DBS, as previously reported. The procedure was otherwise safe and did not lead to deterioration of mood.

Strengthening the Reporting of Observational Studies in Epidemiology for respondent-driven sampling studies: "STROBE-RDS" statement.

OBJECTIVES Respondent-driven sampling (RDS) is a new data collection methodology used to estimate characteristics of hard-to-reach groups, such as the HIV prevalence in drug users. Many national public health systems and international organizations rely on RDS data. However, RDS reporting quality and available reporting guidelines are inadequate. We carried out a systematic review of RDS studies and present Strengthening the Reporting of Observational Studies in Epidemiology for RDS Studies (STROBE-RDS), a checklist of essential items to present in RDS publications, justified by an explanation and elaboration document. STUDY DESIGN AND SETTING We searched the MEDLINE (1970-2013), EMBASE (1974-2013), and Global Health (1910-2013) databases to assess the number and geographical distribution of published RDS studies. STROBE-RDS was developed based on STROBE guidelines, following Guidance for Developers of Health Research Reporting Guidelines. RESULTS RDS has been used in over 460 studies from 69 countries, including the USA (151 studies), China (70), and India (32). STROBE-RDS includes modifications to 12 of the 22 items on the STROBE checklist. The two key areas that required modification concerned the selection of participants and statistical analysis of the sample. CONCLUSION STROBE-RDS seeks to enhance the transparency and utility of research using RDS. If widely adopted, STROBE-RDS should improve global infectious diseases public health decision making.

Simvastatin and ezetimibe in addition to nonpharmacological risk factor modification for achieving new low-density lipoprotein cholesterol targets

BACKGROUND: Though guidelines emphasize low-density lipoprotein cholesterol (LDL-C) lowering as an essential strategy for cardiovascular risk reduction, achieving target levels may be difficult. PATIENTS AND METHODS: The authors conducted a prospective, controlled, open-label trial examining the effectiveness and safety of high-dose fluvastatin or a standard dosage of simvastatin plus ezetimibe, both with an intensive guideline-oriented cardiac rehabilitation program, in achieving the new ATP III LDL-C targets in patients with proven coronary artery disease. 305 consecutive patients were enrolled in the study. Patients were divided into two groups: the simvastatin (40 mg/d) plus ezetimibe (10 mg/d) and the fluvastatin-only group (80 mg/d). Patients in both study groups received the treatment for 21 days in addition to nonpharmacological measures, including advanced physical, dietary, psychosocial, and educational activities. RESULTS: After 21 days of treatment, a significant reduction in LDL-C was found in both study groups as compared to the initial values, however, the reduction in LDL-C was significantly stronger in the simvastatin plus ezetimibe group: simvastatin plus ezetimibe treatment decreased LDL-C to a mean level of 57.7 +/- 1.7 mg/ml, while fluvastatin achieved a reduction to 84.1 +/- 2.4 mg/ml (p < 0.001). In the simvastatin plus ezetimibe group, 95% of the patients reached the target level of LDL-C < 100 mg/dl. This percentage was significantly higher than in patients treated with fluvastatin alone (75%; p < 0.001). The greater effectiveness of simvastatin plus ezetimibe was more impressive when considering the optional goal of LDL-C < 70 mg/dl (75% vs. 32%, respectively; p < 0.001). There was no difference in occurrence of adverse events between both groups. CONCLUSION: Simvastatin 40 mg/d plus ezetimibe 10 mg/d, on the background of a guideline-oriented standardized intensive cardiac rehabilitation program, can reach 95% effectiveness in achieving challenging goals (LDL < 100 mg/dl) using lipid-lowering medication in patients at high cardiovascular risk.

A multifactorial analysis of obesity as CVD risk factor: use of neural network based methods in a nutrigenetics context

Obesity is a multifactorial trait, which comprises an independent risk factor for cardiovascular disease (CVD). The aim of the current work is to study the complex etiology beneath obesity and identify genetic variations and/or factors related to nutrition that contribute to its variability. To this end, a set of more than 2300 white subjects who participated in a nutrigenetics study was used. For each subject a total of 63 factors describing genetic variants related to CVD (24 in total), gender, and nutrition (38 in total), e.g. average daily intake in calories and cholesterol, were measured. Each subject was categorized according to body mass index (BMI) as normal (BMI ≤ 25) or overweight (BMI > 25). Two artificial neural network (ANN) based methods were designed and used towards the analysis of the available data. These corresponded to i) a multi-layer feed-forward ANN combined with a parameter decreasing method (PDM-ANN), and ii) a multi-layer feed-forward ANN trained by a hybrid method (GA-ANN) which combines genetic algorithms and the popular back-propagation training algorithm.