Variability in the frequency of rheumatology consultations in patients with rheumatoid arthritis in Spain.

The aim of this work was to assess the variability in the use of health-care resources, based on the number of visits to rheumatology departments by rheumatoid arthritis patients, in Spain, and its association with patient, physician and center characteristics. The sample consisted of patient records of men and women (aged 16 or older), with a clinical diagnosis of RA, who met the American Rheumatism Association 1987 revised criteria and who had been treated in a rheumatology department at a Spanish hospital with at least one visit to a rheumatologist during the two years preceding the date of the study. To analyze which variables were independently associated with the number of consultations, those with a statistically significant result in the bivariate analysis, or which were clinically relevant or deemed confounders, were used in the construction of a linear regression model. The records of 1188 RA patients were studied. The linear regression model explained the 26.67 % of the variability in the number of visits. The number of csDMARDs prescribed, the administration of biological therapy, corticoid prescription, the presence of nursing consultation, mean time to first visit in the department and attended population showed a positive significant association, while the presence of telephone consultation, distance from the hospital to the patient´s residence (≥20 km) and drug monitoring by rheumatology department + primary care physician or by other specialists were negatively associated with the number of consultations. We observed a high variability in the number of visits, which remains partially unexplained even after taking into account individual, physician and center characteristics.

Bias in clinical epidemiological study designs.

Systematic error, or bias, is error that occurs in each measurement made and which has a direction, i.e., the measured value is always either greater or smaller than the true value. The presence of systematic error directly affects the internal validity of the study, and indirectly affects the external validity of the results obtained. In general, such error can be classified as selection bias, classification bias or confounding bias. It is essential to deal with possible bias in the research design phase, since only confounding bias can be controlled in the phase corresponding to analysis of the results.

Survival analysis and Cox regression.

The data provided by clinical trials are often expressed in terms of survival. The analysis of survival comprises a series of statistical analytical techniques in which the measurements analysed represent the time elapsed between a given exposure and the outcome of a certain event. Despite the name of these techniques, the outcome in question does not necessarily have to be either survival or death, and may be healing versus no healing, relief versus pain, complication versus no complication, relapse versus no relapse, etc. The present article describes the analysis of survival from both a descriptive perspective, based on the Kaplan-Meier estimation method, and in terms of bivariate comparisons using the log-rank statistic. Likewise, a description is provided of the Cox regression models for the study of risk factors or covariables associated to the probability of survival. These models are defined in both simple and multiple forms, and a description is provided of how they are calculated and how the postulates for application are checked - accompanied by illustrating examples with the shareware application R.

Correlation of asymptomatic hyperuricaemia and serum uric acid levels with arterial stiffness in women with systemic lupus erythematosus without clinically evident atherosclerotic cardiovascular disease.

The objective of this article was to evaluate whether serum uric acid (SUA) correlates with arterial stiffness and inflammation markers in a cohort of women with systemic lupus erythematosus (SLE) without overt atherosclerotic cardiovascular diseases, who attended a community hospital. One hundred and two women with SLE were assessed as part of this cross-sectional study. Carotid-femoral pulse wave velocity (PWV) was measured using an automatic device (Complior). C-reactive protein (CRP), fibrinogen and homocysteine levels as well as other metabolic results were recorded. Duration and activity of SLE, damage accrual and treatments were recorded. SLE women were categorized as having or not having hyperuricaemia (HU) according to SUA levels (greater than or up to 6.2 mg/dl, respectively). A multiple linear regression analysis was used to determine the independent link between SUA levels and other variables. Women with SLE and HU (n = 15, 15%) had a worse cardiovascular risk profile that included ageing, hypertension, obesity, higher total cholesterol levels, renal failure and presence of metabolic syndrome. Also, the duration of SLE was increased and damage accrual was greater. In the unadjusted analysis, SUA levels correlated with PWV, CRP, fibrinogen and homocysteine. However, in a multivariate linear regression analysis, SUA levels independently correlated with the duration of SLE, creatinine, total cholesterol and homocysteine levels but did not correlate with PWV. In conclusion, SUA was associated with arterial stiffness, but not independently of age and homocysteine levels. Nevertheless, SUA might be an ancillary indicator of subclinical atherosclerosis in SLE women without clinically evident atherosclerotic cardiovascular disease.

Statistical inference: hypothesis testing.

The aim of statistical inference is to predict the parameters of a population, based on a sample of data. Inferential statistics encompasses the estimation of parameters and model predictions. The present article describes the hypothesis tests or statistical significance tests most commonly used in healthcare research.

[Monitoring of quality indicators of prescriptions after acute myocardial infarction]. / Seguimiento de indicadores de calidad de la prescripción tras infarto agudo de miocardio.

OBJECTIVES: Cardiovascular disease is the primary cause of death in the world. Various drugs have shown their efficiency in the management of patients with acute coronary syndrome, along with the recommendations established by different Scientific Societies. The objective of this survey is to assess the monitoring of the quality standards established in Andalusia for the management of patients with myocardial infarction with evaluation of the ST segment and to analyse the source of variability of the prescription provided by the physician. METHOD: An observational descriptive survey, of indication-prescription, which analysed all the prescriptions included in the hospital discharge reports in patients treated in a tertiary hospital after having suffered acute myocardial infarction with ST segment elevation, between August 2005 and July 2007. RESULTS: A total of 287 patients were included. Antiplatelets were prescribed in 99.3% of the patients, and acetylsalicylic acid in 95.8%. The percentage of beta-blockers received was 89.5%, angiotensin converting enzyme inhibitors (ACEI) in 74.2%, with ARA-II indicated in 2.85%. The use of statins was 84.6%. The indication of conditional sublingual nitroglycerin was 60.3%. The use of ACEI and sublingual nitroglycerin showed statistically significant differences according to the prescribing physician. CONCLUSIONS: The fitness of the prescriptions to the recommendations established in Andalusia is high. The prescription of conditional sublingual nitroglycerin is the indicator which varies most in the quality standards established. Its use in conjunction with the indication of ACEI, are the recommendations which show a higher deviation depending on the prescribing physician.

Descriptive statistics.

Statistics is a science that provides precise techniques for collecting and sorting information made easy by tools and methods for further analysis. The object of descriptive statistics, from sample data, is to describe the most important characteristics, by which we refer to those amounts that provide information on the topic of interest which we are studying.

[Use of the propensity score methodology in health care research]. / Uso de la metodología propensity score en la investigación sanitaria.

Observational studies are frequently used in biomedical research to determine associations between a treatment or exposure and the effects they can produce. Randomized control trials have been developed with the same purpose. Although they provide more precise results, they are more complex and costly. The use of propensity score methodology in observational studies helps to decrease the appearance of bias that they normally present, making them more accurate and with better reliability than randomized control trials. Given the increase in the use of the propensity score methodology in clinical research in recent years, we consider that it is of particular interest to make a description of it, highlighting its application scope and different use techniques.