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(AU)

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.

SHBG levels are associated with bone loss and vertebral fractures in patients with prostate cancer.

SUMMARY: Fractures are increased among prostate cancer patients. No data have been reported in patients with prostate cancer about the relation between serum sex hormone-binding globulin (SHBG) and bone metabolism. We found that SHBG levels were inversely related to bone mass and vertebral fractures in this population. INTRODUCTION: Fractures are increased among prostate cancer patients, especially those on androgen deprivation therapy (ADT), but few data are available on the role of SHBG in their bone status. Our objective was to analyze the relation between serum SHBG and bone metabolism in prostate cancer patients. METHODS: This is a cross-sectional study including 91 subjects with prostate cancer (54 % with ADT). We measured serum levels of SHBG and sex steroids, bone mineral density (BMD) by dual-energy X-ray absorptiometry, and prevalent radiographic vertebral fractures. RESULTS: SHBG levels were inversely related to BMD (femoral neck: r = -0.299, p = 0.00; total hip: r = -0.259, p = 0.019). Subjects with osteoporosis had higher SHBG concentrations than patients without osteoporosis (60.97 ± 39.56 vs 44.45 ± 23.32 nmol/l, p = 0.022). Patients with SHBG levels in the first quartile (>57.6 nmol/l) had an odds ratio (OR) for osteoporosis of 2.59 (95 % CI, 1.30-5.12; p = 0.009) compared with patients with lower SHBG levels. In patients with SHBG >57.6 nmol/l, the OR for vertebral fractures was 2.34 (95 % CI, 1.15-4.78; p = 0.034). The calculated OR was higher after adjustment for age (OR, 5.16; 95 % CI, 1.09-24.49; p = 0.039), estrogens (OR, 6.45; 95 % CI, 1.44-28.95; p = 0.023), and androgens (OR, 5.51; 95 % CI, 1.36-22.37; p = 0.017). CONCLUSIONS: In prostate cancer patients, SHBG levels were inversely related to bone mass and vertebral fractures. Determination of the serum SHBG level may constitute a useful and straightforward marker for predicting the severity of osteoporosis in these patients.

Logistic regression models.

In the health sciences it is quite common to carry out studies designed to determine the influence of one or more variables upon a given response variable. When this response variable is numerical, simple or multiple regression techniques are used, depending on the case. If the response variable is a qualitative variable (dichotomic or polychotomic), as for example the presence or absence of a disease, linear regression methodology is not applicable, and simple or multinomial logistic regression is used, as applicable.

Measures of frequency, magnitude of association and impact in epidemiology.

Epidemiology is concerned with groups of subjects belonging to populations, not with each individual subject, and takes into account both the subjects who contract a disease and those who do not. Comparison, thus, is a basic element of this discipline. Measures of frequency, association and impact are the main statistical resources employed in epidemiology to describe the distribution of healthcare problems, establishing a causal relationship between exposure and disease, enabling users to evaluate the impact of preventive measures in the field of public health.

Association of Vibrio cholerae with plankton in coastal areas of Mexico.

The El Niño event of 1997/1998 provided an opportunity to carry out a field experiment in which the relationship of sea surface temperature and the association of Vibrio cholerae with marine plankton could be assessed in Mexican coastal and estuarine areas. Plankton samples were collected from May 1997 through June 1999. Sites included the Mexican ports of Veracruz, Coatzacoalcos and Frontera in the Gulf of Mexico and Ensenada, Guaymas, Mazatlán, Manzanillo, Acapulco and Oaxaca in the Pacific Ocean. Sampling was also accomplished during two oceanographic cruises in the Yucatan channel of the Caribbean Sea. Bacteriological analyses for V. cholerae serogroups O1 and O139 were carried out. Also, the taxonomic structure of the plankton populations was determined. Vibrio cholerae O1 was detected only in Veracruz samples collected during April, May and June 1999, when La Niña climatic conditions prevailed. It is concluded that V. cholerae O1 in Mexico derives from its marine and estuarine origin and not from sewage contamination. The significant number of Acartia tonsa copepodites and V. cholerae copepodite-positive samples suggests a significant role of this copepod in the occurrence and distribution of V. cholerae in coastal areas of Mexico.

Diagnóstico y conducta en el hídrops fetal temprano / Diagnostic and management of early foetal hydrops

El hídrops fetal temprano es un fenómeno frecuente que forma parte de numerosos procesos patológicos. Sin embargo, la tasa de diagnósticos se encuentra por debajo de su incidencia. Los hallazgos ecográficos indirectos relacionados con fallo cardíaco de diferente origen son más frecuentes en el primer trimestre. Es estos casos, el aumento del pliegue nucal suele ser el primer signo ecográfico de hídrops temprano. En el estudio realizado sobre 30 casos en los que se determinó el cariotipo fetal por un pliegue nucal mayor de 3 mm antes de la semana 14 de gestación como indicación, el índice de cromosomopatías fue del 36,7 frente al 3,1 por ciento de cariotipos anormales obtenidos cuando la indicación fue otra. Además, se observó un 50 por ciento de anomalías asociadas entre los que presentaron algún signo temprano de hídrops y un 47 por ciento de mortalidad fetal asociada. La medición del pliegue nucal ofrece la oportunidad de realizar un cribado temprano de aneuploidías, cardiopatías u otras afecciones fetales (AU)

Plasma levels of beta-endorphin and ACTH during labor and immediate puerperium.

We measured concentrations of beta-endorphin (beta-EP) and adrenocorticotropic hormone (ACTH) in maternal peripheral plasma during dilation, the expulsive period and immediate puerperium, and in the umbilical vein. Plasma levels of both peptides increased markedly and were directly correlated during labor, and decreased 24 h after birth; they were no longer correlated during immediate puerperium. In the umbilical vein, beta-EP and ACTH were also directly correlated. Moreover, the concentration of each of the two peptides in maternal plasma during the expulsive period was correlated with the corresponding concentration in umbilical vein. We discuss the importance of placental corticotropin-releasing factor in the regulation of maternal and fetal stress.

[Evaluation of the pregnancy control programme in the La Chana Health Centre in 1989. A comparison with previous years]. / Evaluación del programa de control de embarazo del Centro de Salud La Chana en 1989. Comparación con años anteriores.

OBJECTIVE: An evaluation of the programme with respect to the objective set for the year 1989 and a comparison with the evaluation results from previous years. DESIGN: It is a retrospective diachronic study, by observation. LOCATION: The study was carried out in the primary health area in the basic health zone of La Chana in the Granada North health district. PATIENTS: All pregnancies attended in the basic health zone during 1989. FIGURES AND PRINCIPAL RESULTS: 223 women reached the end of the pregnancy in 1989 of whom 207 joined the programme. The level of puerperal visits (83.5% of the pregnancies) and the coverage of the programme (59.4%) were below target. Other indicators were above target. The average number of visits per pregnancy has reduced over the years (5.5 in 1987; 4.5 in 1988; 4.4 in 1989), while the percentage for risk levels has been maintained. CONCLUSIONS: The programme seems valid but could be improved, particularly where coverage is concerned. Analysis and comparison of results with previous years highlights the lack of coordination existing between the different levels of attendance. Finally we have to point out that we are unaware of the quality of attention which is received, and this aspect should receive serious study.