Background levels of micro-organisms in the busy urban environment of transport hubs.

AIMS: We constantly interact with our surrounding microbiome, including the micro-organisms present in highly populated public places. However, data on everyday exposure to background levels of micro-organisms are limited. To address this, bacteria and fungi were collected and enumerated in settled dust from railway stations. METHODS AND RESULTS: Samples were collected weekly for 52 weeks, from up to three pre-determined surfaces in each of 17 railway stations in England and Scotland. Trained staff at each station took surface wipes, sending them to the laboratory for culture-based analysis for total bacteria and fungi. Maximum yields of bacteria at the stations were 107 -108 colony forming units (CFU) per cm2 , and 104 -105 CFU per cm2 for fungi. CONCLUSIONS: There was evidence of seasonal trends, with bacterial numbers rising from spring through to winter, while fungal numbers peaked in autumn. Microbial numbers were similar in samples taken at the same time at a given station. Influences on contamination levels were likely to be a combination of passenger numbers and station layout, with dust generated from construction work also contributing. SIGNIFICANCE AND IMPACT OF THE STUDY: A baseline of typical human exposure to micro-organisms in public transport hubs was established through the generation of a comprehensive database.

The relationship between clinical and standardized tests for hand-arm vibration syndrome.

BACKGROUND: Standardized laboratory tests are undertaken to assist the diagnosis and staging of hand-arm vibration syndrome (HAVS), but the strength of the relationship between the tests and clinical stages of HAVS is unknown. AIMS: To assess the relationship between the results of thermal aesthesiometry (TA), vibrotactile (VT) thresholds and cold provocation (CP) tests with the modified Stockholm scales for HAVS and to determine whether the relationship is affected by finger skin temperature. METHODS: Consecutive records of workers referred to a Tier 5 HAVS assessment centre from 2006 to 2015 were identified. The diagnosis and staging of cases was undertaken from the clinical information contained in the records. Cases with alternative or mixed diagnoses were excluded and staging performed according to the modified Stockholm scale without knowledge of the results of the standardized laboratory tests. RESULTS: A total of 279 cases of HAVS were analysed. Although there was a significant trend for sensorineural (SN) and vascular scores to increase with clinical stage (P < 0.01), there was no significant difference in scores between 2SN early and 2SN late or between 2SN late and 3SN. There was moderate correlation between the TA and VT scores and the clinical SN stages (r = 0.6). This correlation did not change when subjects were divided into those with a finger skin temperature <30 and >30°C. CP scores distributed bimodally and correlated poorly with clinical staging (r = 0.2). CONCLUSIONS: Standardized SN tests distinguish between the lower Stockholm stages, but not above 2SN early. This has implications for health surveillance and UK policy.

Investigation of gastrointestinal effects of organophosphate and carbamate pesticide residues on young children.

This prospective study was designed to investigate whether there is any association between gastrointestinal effects and pesticide residue exposure (as measured by metabolite levels in urine and faecal samples) in young children and to describe background levels of pesticide residues in samples from healthy children in the UK. Children (N=136) between the ages of 1.0 and 4.2 years were recruited. Of these, 107 provided background baseline samples and 26 provided samples when suffering from gastrointestinal symptoms. Urine samples (from all populations) were positive for (non-specific) carbaryl metabolite (urine 19/78, faeces 9/99), organophosphate metabolites (urine 103/135, faeces 47/111) and pirimicarb metabolite (urine 72/175, faeces 45/135). There were no statistically significant differences between samples from children when healthy or unwell. The urinary 95th percentile values for the healthy population of young children in this study were 31 nmol/l (carbaryl metabolite), 2156 nmol/l (total organophosphate metabolites) and 139 nmol/l (pirimicarb metabolite). In this study, samples from children suffering gastrointestinal symptoms were no more associated with anti-cholinergic pesticide metabolite levels or rotaviral infection than samples from healthy children. Background levels of anti-cholinergic pesticide metabolites in healthy UK children were in agreement with previously reported levels from the US and Germany.

Mortality and cancer incidence among British agricultural pesticide users.

BACKGROUND: Although the acute effects of pesticides in humans are well known, uncertainty still exists about the health effects of chronic low-level exposure to pesticides. AIMS: To compare mortality and cancer incidence experienced by a cohort of British pesticide users to that of the Great Britain (GB) population. METHODS: The Pesticide Users Health Study (PUHS) comprises users of agricultural pesticides who have Certificates of Competence under the Control of Pesticides Regulations 1986. Participants were followed up between 1987 and 2004 (cancer incidence) or 2005 (mortality). Standardized mortality ratios (SMRs) and Standardized incidence ratios (SIRs) were estimated for outcomes of interest identified from the literature. RESULTS: Altogether, 62,960 pesticide users were followed up for 829,709 person-years (to 31 December 2005). Most participants were male (94%) and based in England (86%). All-cause mortality was lower for both men [SMR 0.58, 95% confidence interval (CI) 0.55-0.60] and women (SMR 0.71, 95% CI 0.52-0.98) compared to the GB population. Mortality and incidence were below those expected for all cancers combined among men (SMR 0.71, 95% CI 0.66-0.77; SIR 0.85, 95% CI 0.81-0.90), particularly for cancers of the lip, oral cavity and pharynx, digestive organs and respiratory system. The incidence of testicular cancer, non-melanoma skin cancer and multiple myeloma were above expected. Mortality from injury by machinery was significantly above expected for men (SMR 4.21, 95% CI 2.11-8.42). CONCLUSIONS: This study suggests that pesticide users in the PUHS are generally healthier than the national population but may have excesses of non-melanoma skin cancer, testicular cancer and multiple myeloma.

Mortality among British asbestos workers undergoing regular medical examinations (1971-2005).

OBJECTIVES: The Great Britain Asbestos Survey was established to monitor mortality among workers covered by regulations to control occupational exposure to asbestos. This study updates the estimated burden of asbestos-related mortality in the cohort, and identifies risk factors associated with mortality. METHODS: From 1971, workers were recruited during initially voluntary and later statutory medical examinations. A brief questionnaire was completed during the medical, and participants were flagged for death registrations. Standardised mortality ratios (SMRs) and proportional mortality ratios (PMRs) were calculated for deaths occurring before 2006. Poisson regression analyses were undertaken for diseases with significant excess mortality. RESULTS: There were 15 496 deaths among 98 117 workers followed-up for 1 779 580 person-years. The SMR for all cause mortality was 141 (95% CI 139 to 143) and for all malignant neoplasms 163 (95% CI 159 to 167). The SMRs for cancers of the stomach (166), lung (187), peritoneum (3730) and pleura (968), mesothelioma (513), cerebrovascular disease (164) and asbestosis (5594) were statistically significantly elevated, as were the corresponding PMRs. In age and sex adjusted analysis, birth cohort, age at first exposure, year of first exposure, duration of exposure, latency and job type were associated with the relative risk of lung, pleural and peritoneal cancers, asbestosis and mesothelioma mortality. CONCLUSIONS: Known associations between asbestos exposure and mortality from lung, peritoneal and pleural cancers, mesothelioma and asbestosis were confirmed, and evidence of associations with stroke and stomach cancer mortality was observed. Limited evidence suggested that asbestos-related disease risk may be lower among those first exposed in more recent times.

Occupational exposure to asbestos and mortality among asbestos removal workers: a Poisson regression analysis.

The asbestos industry has shifted from manufacture to stripping/removal work. The aim of this study was to investigate early indications of mortality among removal workers. The study population consisted of 31 302 stripping/removal workers in the Great Britain Asbestos Survey, followed up to December 2005. Relative risks (RR) for causes of death with elevated standardised mortality ratios (SMR) and sufficient deaths were obtained from Poisson regression. Risk factors considered included dust suppression technique, type of respirator used, hours spent stripping, smoking status and exposure length. Deaths were elevated for all causes (SMR 123, 95% CI 119-127, n=985), all cancers including lung cancer, mesothelioma, and circulatory disease. There were no significant differences between suppression techniques and respirator types. Spending more than 40 h per week stripping rather than less than 10, increased mortality risk from all causes (RR 1.4, 95% CI 1.2-1.7), circulatory disease and ischaemic heart disease. Elevated mesothelioma risks were observed for those first exposed at young ages or exposed for more than 30 years. This study is a first step in assessing long-term mortality of asbestos removal workers in relation to working practices and asbestos exposure. Further follow-up will allow the impact of recent regulations to be assessed.

Do simple questions about diet and physical activity help to identify those at risk of Type 2 diabetes?

AIMS: Diet and physical activity interventions can prevent diabetes in those at high risk due to impaired glucose tolerance. We determined whether simple measures of physical activity and diet predicted incident diabetes and enhanced prediction by known risk factors including age, body mass index and family history. METHODS: This was a population-based prospective cohort study (EPIC-Norfolk). Participants aged 40-79 years (n = 25,633) attended a health check between 1993 and 1998 and completed diet and activity questionnaires. We assessed the association between simple behavioural indices of physical activity and diet derived from the questionnaires as well as known risk score variables with incident diabetes at follow-up (mean 4.6 years). We developed a new diabetes risk score incorporating simple behavioural indices in a randomly selected half of the EPIC dataset using forward step-wise multivariate logistic regression, and tested this score in the remaining half. We compared existing and new scores using receiver-operating characteristic (ROC) curves. RESULTS: There were 417 incident cases of diabetes during 115,137 years of follow-up. A simple physical activity index independently predicted risk of diabetes. Eating one or more daily portion of vegetables, fresh fruit and wholemeal bread was associated with reduced risk; whilst eating meat products was associated with increased risk. The area under the ROC curves for the new and original score was the same (76.3%). CONCLUSIONS: Simple indices of diet and activity are feasible to collect, predict future diabetes risk and might enhance routine data collection in primary care. However, they do not improve the prediction of risk scores based on known risk factors.

Elevated serum ferritin levels predict new-onset type 2 diabetes: results from the EPIC-Norfolk prospective study.

AIMS/HYPOTHESIS: The aim of this study was to examine the association between baseline body iron stores and new-onset diabetes. SUBJECTS AND METHODS: We studied the association between baseline serum ferritin concentration and type 2 diabetes in 360 clinically incident diabetes cases and 758 controls nested within the EPIC (European Prospective Investigation of Cancer)-Norfolk Cohort Study. Serum ferritin levels were categorised into five groups: sex-specific quartiles of the normal range of ferritin and a group with clinically raised ferritin below levels indicative of haemochromatosis. RESULTS: Baseline serum ferritin was higher among cases than control participants (geometric mean: men 96.6 vs 67.8 ng/ml, respectively, p < 0.001; women 45.9 vs 34.8 ng/ml, respectively, p = 0.005). In analyses adjusted for known risk factors (age, BMI, sex, family history, physical activity, smoking habit) and dietary factors measured by 7-day food diary, the risk of diabetes was markedly elevated in participants with clinically raised ferritin compared with the lowest quartile (odds ratio [OR] 7.4, 95% CI 3.5-15.4). Further adjustment for potential confounding by inflammation (C-reactive protein, IL-6 and fibrinogen) had no material impact on the observed association, while adjustment for hepatic enzymes (alanine aminotransferase and gamma glutamyl transferase) and adiponectin attenuated the magnitude of association, but it remained statistically significant (OR 3.2 [1.3-7.6]). CONCLUSIONS/INTERPRETATION: Serum ferritin is an important and independent predictor of the development of diabetes. This finding may have important implications for understanding the aetiology of diabetes.

Polymorphisms in the gene encoding sterol regulatory element-binding factor-1c are associated with type 2 diabetes.

AIMS/HYPOTHESIS: The sterol regulatory element-binding factor (SREBF)-1c is a transcription factor involved in the regulation of lipid and glucose metabolism. We have previously found evidence that a common SREBF1c single-nucleotide polymorphism (SNP), located between exons 18c and 19c, is associated with an increased risk of type 2 diabetes. The present study aimed to replicate our previously reported association in a larger case-control study and to examine an additional five SREBF1c SNPs for their association with diabetes risk and plasma glucose concentrations. METHODS: We genotyped six SREBF1c SNPs in two case-control studies (n=1,938) and in a large cohort study (n=1,721) and tested for association with type 2 diabetes and with plasma glucose concentrations (fasting and 120-min post-glucose load), respectively. RESULTS: In the case-control studies, carriers of the minor allele of the previously reported SNP (rs11868035) had a significantly increased diabetes risk (odds ratio [OR]=1.20 [95% CI 1.04-1.38], p=0.015). Also, three other SNPs (rs2236513, rs6502618 and rs1889018), located in the 5' region, were significantly associated with diabetes risk (OR > or =1.21, p< or =0.006). Furthermore, two SNPs (rs2236513 and rs1889018) in the 5' region were weakly (p<0.09) associated with plasma glucose concentrations in the cohort study. Rare homozygotes had increased (p< or =0.05) 120-min post-load glucose concentrations compared with carriers of the wild-type allele. Haplotype analyses showed significant (p=0.04) association with diabetes risk and confirmed the single SNP analyses. CONCLUSIONS/INTERPRETATION: In summary, we replicated our previous finding and found evidence for SNPs in the 5' region of the SREBF1c gene to be associated with the risk of type 2 diabetes and plasma glucose concentration.

How much might achievement of diabetes prevention behaviour goals reduce the incidence of diabetes if implemented at the population level?

AIMS/HYPOTHESIS: Randomised trials targeting high-risk people with impaired glucose tolerance have halved progression to diabetes using behavioural interventions aimed at achieving five goals related to weight, diet and physical activity. The number of people currently meeting these goals in the general population is unknown. The potential impact on the incidence of diabetes of increasing the proportion of people who meet these goals is also unclear. We quantified the association between the achievement of behavioural goals for the prevention of diabetes and the incidence of diabetes in a population-based cohort study. SUBJECTS AND METHODS: European Prospective Investigation into Cancer (EPIC)-Norfolk is a prospective cohort of 24,155 participants aged 40-79 years who attended a baseline health check and completed validated diet and activity questionnaires. We assessed the association between achievement of five diabetes healthy behaviour prevention goals (BMI <25 kg/m(2), fat intake <30% of energy intake, saturated fat intake <10% of energy intake, fibre intake > or =15 g/4,184 kJ, physical activity >4 h/week) and risk of developing diabetes at follow-up (mean 4.6 years). RESULTS: Only 20% of EPIC participants met three or more diabetes prevention goals. Diabetes incidence was inversely related to the number of goals achieved (p<0.001). None of the participants who met all five goals developed diabetes, whereas diabetes incidence was highest in those who did not meet any goals. If the entire population were able to meet one more goal, the total incidence of diabetes would be predicted to fall by 20%. CONCLUSIONS/INTERPRETATION: In this population-based study, the risk of diabetes was inversely associated with the number of behaviour goals for diabetes prevention that were met. Interventions that promote achievement of these goals in the general population could significantly reduce the growing burden of diabetes-related morbidity and mortality.

Habitual fish consumption and glycated haemoglobin: the EPIC-Norfolk study.

OBJECTIVE: To investigate the association between habitual fish consumption and a continuous measure of glycaemia. DESIGN: Cross-sectional study. SETTING: EPIC-Norfolk, a population-based cohort study of diet and chronic disease. SUBJECTS AND METHODS: In all, 4500 men and 5509 women, aged 40-78 y, without self-reported diabetes. Diet was assessed by a semiquantitative food frequency questionnaire, and glycaemia was measured by glycated haemoglobin. RESULTS: In women only, in analyses adjusted for age, the HbA(1c) level was positively associated with eating fried fish and inversely associated with eating oily fish (b=0.036, 95% confidence interval (CI): 0.0033, 0.069; and b=-0.046, 95% CI:-0.086, -0.0064 respectively). These associations were attenuated by adjustment for family history of diabetes, smoking status and physical activity level, but the association with fried fish remained statistically significant (b=0.033, 95% CI: 0.00056, 0.066). Adjusting for total energy, alcohol, fruit and vegetable intakes resulted in further attenuation and both associations were no longer statistically significant. In men, there was no evidence that HbA(1c) level was associated with fish consumption. CONCLUSIONS: The study found no evidence of an association between fish consumption and HbA(1c) after taking other lifestyle factors into account.

Does peroxisome proliferator-activated receptor gamma genotype (Pro12ala) modify the association of physical activity and dietary fat with fasting insulin level?

Peroxisome proliferator-activated receptor gamma (PPARgamma) has a role in controlling adipogenesis and insulin sensitivity. Previous studies have suggested that a common polymorphism (Pro12Ala) in the PPARgamma-2 isoform of this gene may be associated with markers of insulin resistance. We have previously shown that in combination, the relationships with fasting insulin of dietary polyunsaturated to saturated fatty acid ratio (P:S ratio) and physical activity are additive. We have also demonstrated that the association between P:S ratio and fasting insulin level is modified by the Pro12Ala genotype. The purpose of the present study was to investigate whether the Pro12Ala genotype modified the combined relationships of P:S ratio and physical activity level (PAL) on fasting insulin concentration. A population-based cohort of 506 Caucasian men and women aged 31 to 71 years was genotyped for the Pro12Ala polymorphism. P:S ratio was assessed by food-frequency questionnaire (FFQ) and PAL was estimated from 4 days of free-living heart rate monitoring following individual calibration of heart rate against energy expenditure during an exercise stress test. The combined associations of PAL and P:S ratio on fasting insulin level were examined stratified by Pro12Ala genotypes in a dominant model for the Ala allele. Among Pro allele homozygotes, there was no interaction between PAL and P:S ratio on fasting insulin (P =.929). However, in carriers of the Ala allele the association of P:S ratio with fasting insulin was modified by activity level (interaction P = 0.038). In those who were inactive and carried the Ala allele, the age-, sex-, and body mass-adjusted relationship between P:S ratio and log insulin was not significant (beta = -0.03, P =.93). In contrast, in physically active Ala carriers, the association of P:S ratio with log fasting insulin was highly significant (beta = -0.93, P =.004). In conclusion, this study examined the modification by PPARgamma genotype of the association between energy expenditure, P:S ratio, and fasting insulin level, a measure of insulin resistance. These data show that in Pro allele homozygotes the combined associations of P:S ratio and PAL are additive. In contrast, in Ala allele carriers, PAL modifies the association between P:S ratio and fasting insulin level in a multiplicative manner.

Cross-sectional association between total level and type of alcohol consumption and glycosylated haemoglobin level: the EPIC-Norfolk Study.

OBJECTIVE: To investigate the association between total level and type of alcohol consumed and glycaemia. DESIGN: Cross-sectional study. SETTING: The EPIC-Norfolk Study, a population-based cohort study of diet and chronic disease. SUBJECTS AND METHODS: Non-diabetic men (n=2842) and women (n=3572), aged 40-78 y. Alcohol intake was assessed by self-reported questionnaire, and glycaemia measured by glycosylated haemoglobin (HbA(1c)). RESULTS: Ten percent of men and 18% of women reported drinking no alcohol. Among drinkers, median alcohol intake was 8 units/week for men and 3 units/week for women. In analyses stratified by sex and adjusted for age, total energy intake, education, fruit and vegetable intake, smoking, family history of diabetes, physical activity, body mass index and waist:hip ratio, alcohol intake was inversely associated with HbA(1c) in men and women, although the association was stronger in women. A 1 unit/week increase in alcohol intake was associated with 0.0049% (s.e.=0.00223; P-value=0.028) and 0.017% (s.e.=0.00343; P-value <0.001) reduction in HbA(1c) in men and women respectively. In similar multivariate analyses, wine intake was inversely associated with HbA(1c) in men, and wine, spirits and beer intake were inversely associated with HbA(1c) in women. When also adjusted for total alcohol intake, only the association between wine intake and HbA(1c) in men remained significant. CONCLUSION: Alcohol intake was associated with lower HbA(1c) level, an association not explained by confounding. The distinction between type of alcohol consumed was particularly important in men.

Is the association between dietary fat intake and insulin resistance modified by physical activity?

The rising prevalence of type 2 diabetes, a condition associated with insulin resistance, is commonly attributed to changes in dietary patterns and physical activity levels in susceptible populations. However, few studies have described the independent effects of dietary intake and physical activity on the degree of insulin sensitivity within populations or examined the possibility of interactions between dietary factors and physical activity. This study was undertaken to describe the relationship between the quantity and pattern of dietary fat intake on fasting insulin levels (a marker of insulin sensitivity) and to investigate whether the association was modified by physical activity. A cross-sectional study of 815 nondiabetic men and women (30 to 71 years) recruited from a population-based sampling frame was undertaken. Diet was characterized using a semiquantitative food frequency questionnaire. Physical activity level (PAL), the ratio of total energy expenditure to basal metabolic rate, was estimated using individually calibrated heart rate monitoring, a method previously shown to be an objective and valid method for assessing total energy expenditure. In a linear regression model adjusted for total energy intake, total fat intake bordered on a significant association with fasting insulin (b = 0.000081; P =.058), and the polyunsaturated to saturated fat ratio (P:S ratio) of the diet was negatively associated with fasting insulin concentration (b = -0.37, P <.001). A negative association was observed between the PAL and fasting insulin (b = -0.12, P =.025). The association of the P:S ratio and PAL with fasting insulin remained significant when adjusted for each other and for total fat, total energy intake, body mass index (BMI), waist-to-hip ratio (WHR), age, sex, family history of diabetes, smoking status, and alcohol intake (P:S ratio, b = -0.24, P =.003; PAL, b = -0.13; P =.007). The association with total fat intake was no longer significant in this multivariate model (b = 6.7 x 10(-6); P =.858). There was no evidence for an interaction between total dietary fat intake and PAL (b = -0.000048; P =.243) or between the P:S ratio and PAL (b = -0.013; P =.949). These data demonstrate an independent association between the P:S ratio of the diet, the overall level of physical activity, and the fasting insulin concentration, a marker of insulin sensitivity. There was no evidence that the association between dietary fat intake and insulin resistance was modified by physical activity. The findings provide further support for efforts to promote increases in overall physical activity and modifications in the pattern of dietary fat intake in the whole population.

Fat consumption and HbA(1c) levels: the EPIC-Norfolk study.

OBJECTIVE: To describe the relationship between total dietary fat and the pattern of fat intake and HbA(1c). RESEARCH DESIGN AND METHODS: In this cross-sectional study, 2,759 men and 3,464 women (40-78 years of age) without a previous diagnosis of type 2 diabetes were recruited from a population-based sampling frame. Diet was assessed using a self-reported semiquantitative food frequency questionnaire. RESULTS: The HbA(1c) level was negatively associated with the polyunsaturated fat-to-saturated fat ratio (P:S ratio) of the diet (beta = -0.0338 HbA(1c)% per SD change in P:S ratio; P < 0.001) and positively associated with the total level of fat intake (beta = 0.0620 HbA(1c)% per SD change in total fat intake; P < 0.001), adjusted for age and total energy intake. The associations remained significant when adjusted for each other and for total energy, protein, age, sex, family history of diabetes, BMI, waist-to-hip ratio, physical activity, and smoking (for P:S ratio, beta = -0.0200 HbA(1c)% per SD change in P:S ratio, P = 0.013; for total fat, beta = 0.420% HbA(1c)% per SD change in total fat intake, P < 0.001). The benefits from a high P:S ratio were attributed to a lower saturated fat intake. CONCLUSIONS: These findings demonstrate independent associations between HbA(1c) concentration across the normal range of HbA(1c) and both total fat intake and the pattern of dietary fat intake. They provide further support to efforts promoting modifications in the intake of dietary fat.

Evidence for gene-nutrient interaction at the PPARgamma locus.

The importance of the nuclear receptor peroxisome proliferator-activated receptor-gamma (PPARgamma) in regulating insulin resistance and blood pressure has been demonstrated in families with loss of function mutations. Gain of function mutations has been associated with severe obesity. However, previous population studies of the common variant Pro12Ala have produced conflicting results. As it is likely that the natural ligands for this receptor may include fatty acids, we hypothesized that the effect of this common variant may be altered by the character of the diet, particularly the ratio of dietary polyunsaturated fat to saturated fat (P:S ratio). We studied 592 nondiabetic participants in an ongoing population-based cohort study who were genotyped for the Pro12Ala polymorphism in the PPAR gamma2 isoform. As the Ala homozygotes were uncommon (2.0%), all analyses were conducted comparing Pro homozygotes (79.1%) to Ala allele carriers. There was no difference in fasting insulin concentration or BMI between Ala allele carriers and Pro homozygotes. The fasting insulin concentration was negatively associated with the P:S ratio (P = 0.0119) after adjustment for age and sex, and a strong interaction was evident between the P:S ratio and the Pro12Ala polymorphism for both BMI (P = 0.0038) and fasting insulin (P = 0.0097). The data suggest that when the dietary P:S ratio is low, the BMI in Ala carriers is greater than that in Pro homozygotes, but when the dietary ratio is high, the opposite is seen. This gene-nutrient interaction emphasizes the difficulty of examining the effect of common polymorphisms in the absence of data on nongenetic exposures, and may explain the heterogeneity of findings in previous studies.