Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Statistical Regression”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 739 records · Page 41Linked to original sources
PubMed1991

A comparison of the direct immunobead test and other tests for sperm antibodies detection.

The authors report the results of a correlation study between the direct Immunobead Test (d-IBT) and other techniques for antisperm antibodies detection. The Gelatin Agglutination Test (GAT) and Tray Agglutination Test (TAT) were used to detect antibodies in blood serum and seminal plasma ("indirect methods"). The Direct IgG Mixed Antiglobulin Reaction Test (d-MAR test) was used to detect sperm antibodies bound to the sperm surface ("direct method"). A good concordance between the methods, measured by phi and K tests, was found and satisfactory mathematical models were established by regression analyses.

Agglutination Tests↗
PubMed2004

A regression model using sediment chemistry for the evaluation of marine environmental impacts associated with salmon aquaculture cage wastes.

This study was undertaken to develop an approach for modelling changes of sediment chemistry related to the accumulation of aquaculture waste. Metal composition of sediment Al, Cu, Fe, Li, Mn, and Zn; organic carbon and < 63 microm particles were used to determine the extent of detectable effects around the cage. This study showed marked differences in the sediment chemistry between aquaculture sites and the natural background: (1) negative correlations between sediment Cu and Zn with Al, (2) poor correlations between metals and Li, and (3) concentrations of Fe and Mn decreased with increased accumulation of organic carbon. There is a trend among normalised metals, organic carbon and particles related to normal, hypoxic and anoxic sediment conditions. The trends are useful for detecting and assessing the cumulative effects from aquaculture wastes to the marine environment. Lithium is less interactive with other metals in aquaculture sediments compared with the natural background sediments. Principal components analysis (PCA) was carried out on the metals, organic carbon, and particles to cluster the similarities of the variables so as to establish the predicted or adjusted environmental monitoring program (EMP) ratings. This approach, using the adjusted EMP rating based on sediment chemistry, yields a regression model with R2 = 0.945 compared to R2= 0.653 for the regression model using unadjusted EMP for assessing the environmental conditions.

Animals↗
PubMed1999

Robust regression for clustered data with application to binary responses.

Generalized estimating equations (GEE) can be highly influenced by the presence of unusual data points. A generalization of the GEE procedure, which yields parameter estimates and fitted values that are resistant to influential data, is introduced. Resistant generalized estimating equations (REGEE) include weights in the estimating equations to downweight influential observations or clusters. Influential observations are downweighted according to their leverage or residual in an example of correlated binary regression applied to 137 urinary incontinent elderly patients from 38 medical practices.

Aged↗
PubMed2005

Trick or treat: the effect of placebo on the power of pharmacogenetic association studies.

The genetic mapping of drug-response traits is often characterised by a poor signal-to-noise ratio that is placebo related and which distinguishes pharmacogenetic association studies from classical case-control studies for disease susceptibility. The goal of this study was to evaluate the statistical power of candidate gene association studies under different pharmacogenetic scenarios, with special emphasis on the placebo effect. Genotype/phenotype data were simulated, mimicking samples from clinical trials, and response to the drug was modelled as a binary trait. Association was evaluated by a logistic regression model. Statistical power was estimated as a function of the number of single nucleotide polymorphisms (SNPs) genotyped, the frequency of the placebo 'response', the genotype relative risk (GRR) of the response polymorphism, the strategy for selecting SNPs for genotyping, the number of individuals in the trial and the ratio of placebo-treated to drug-treated patients. We show that: (i) the placebo 'response' strongly affects the statistical power of association studies--even a highly penetrant drug-response allele requires at least a 500-patient trial in order to reach 80 per cent power, several-fold more than the value estimated by standard tools that are not calibrated to pharmacogenetics; (ii) the power of a pharmacogenetic association study depends primarily on the penetrance of the response genotype and, when this penetrance is fixed, power decreases for larger placebo effects; (iii) power is dramatically increased when adding markers; (iv) an optimal study design includes a similar number of placebo- and drug-treated patients; and (v) in this setting, straightforward haplotype analysis does not seem to have an advantage over single marker analysis.

Clinical Trials as Topic↗
PubMedDate not supplied

A community study of gynecological and related morbidities in rural Egypt.

This study assesses the prevalence of gynecological and related morbidity conditions in a rural Egyptian community. A medical examination was conducted on a sample of 509 ever-married, nonpregnant women. For gynecological morbidities, genital prolapse was diagnosed in 56 percent, reproductive tract infections in 52 percent, and abnormal cervical cell changes in 11 percent of the women. For related morbidities, anemia was present in 63 percent of the women, followed by obesity (43 percent), hypertension (18 percent), and urinary tract infection (14 percent). Regression analysis of risk factors demonstrated the contribution of social conditions and medical factors to these diseases. Reproductive tract infections were shown to occur more frequently with uterovaginal prolapse, IUD use, presence of husband (regular sexual activity), and unhygienic behavior. Genital prolapse increased with age and number of deliveries. Age, recent pregnancy, education, socioeconomic class, and workload showed significant associations with related morbidity conditions. This evidence challenges national health programs to go beyond safe motherhood, child survival, and family planning in its services to women, and to consider the social context of health as well.

Adult↗
PubMed1989

Test of immunopotentialization in colposcopy--a clinical evaluation.

52 infertile women (age range 19-35 years) treated by Gynatren-immunopotentialization (three vaccinations every second week) were divided into two groups according to appearance of cervical intraepithelial neoplasia (CIN). 30 women had CIN cytologically diagnosed and proved by colposcopy (group II). Colposcopic examinations were performed on the first day of therapy (first vaccination), after two weeks (second vaccination) and after four weeks (the day of the third injection). Four weeks of the therapy resulted in normalization of cervical milieu (e.g., complete regression of infection in 68.2% of I group and 56% of II group) as well as in disappearance of CIN in 66% of cases and in remaining ones there was a 7-times decrease of CIN II. It was colposcopically stated that the process of statistical significant regression of inflammation as well as normalization of cervical pathology is less apparent in women with CIN. These findings confirmed our previous cytological observations.

Adjuvants, Immunologic↗
PubMed2003

[Impact of tumor size on long-term survival of patients with organ-confined renal cell cancer].

PURPOSE: As a result of the observation that the potential biological aggressiveness of even early stage renal cell cancer (RCC) strongly correlates with tumor size, the 5th edition of the TNM system (1997), which classifies all renal tumors up to a maximum diameter of 7 cm as T1, was modified in the current classification to distinguish T1a and T1b tumors based on a cut-off of 4 cm. Only a few larger investigations supporting this cut-off are available in the literature and these are based on univariate statistical analysis. To determine whether this cut-off best reflects the correlation between aggressive potential and tumor size and thus differentiates patients at low or high risk of progression, this investigation was performed using both univariate and multivariate statistical analysis. MATERIALS AND METHODS: Between 1981 and 2000, a total of 652 patients underwent tumor nephrectomy for renal cell carcinoma. Of these, the 243 patients with local RCC not extending beyond the renal capsule were subjected to univariate (log rank test) and multivariate (Cox regression model) statistical analysis. RESULTS: It became evident that, while during univariate analysis the prognostic calue of a cut-off size of 4 cm was confirmed, multivariate analysis identified the highest relative risk of cause-specific death (2.93) in those patients with tumors larger than 5 cm maximum diameter. CONCLUSIONS: The 5 cm cut-off thus appears to best determine the clinical prognosis for patients undergoing tumor nephrectomy for early stage renal call cancer. Taking this into consideration, the present study clearly demonstrates the reed for a multivariate statistical approach when the current modification of the TNM classification system is critically reevaluated.

Adult↗
PubMed1992

[Follow-up of 551 patients with node-negative breast cancer].

To evaluate the prognostic significance of established clinical, histological, and biochemical factors, we examined the survival data of 551 node-negative breast cancer patients. At a median follow-up of 5 years, we found 114 recurrences, 79 of them at distant sites. 41 patients died. 84 patients with less than 8 examined lymph nodes, adjuvant systemic treatment, or treatment differing from standard procedures, had a statistically significant shorter overall survival and were excluded from further analysis. With regard to relapse-free and overall survival univariate and multivariate analyses of the remaining 467 patients revealed only few factors with prognostic significance. In multivariate analysis of overall survival by the Cox regression model, statistically significant prognostic value was limited to three factors: lymphangiosis carcinomatosa (relative risk 4.8, 95%-confidence interval 2.0-11.7), postmenopausal status (0.38, 0.17-0.84), and positive progesterone receptor status (0.37, 0.14-1.0). In addition, there was a trend (p = 0.075) of prolonged survival in patients with Bloom and Richardson grade I cancers. The few prognostic factors found were able to identify patients with very good, as well as very bad prognosis. However, for the majority of node-negative breast cancer patients, estimation of prognosis remains unsatisfactory. Therefore, further independent prognostic factors are necessary.

Actuarial Analysis↗
PubMed1991

Clinical findings as outcome predictors in rehabilitation of patients with sciatica.

In rehabilitation of patients with sciatica, factors prognostic of long-term outcome could prove beneficial in interventions and in evaluating progress toward fitness or capacity for work. In this study results of physical examinations, carried out one year after hospital treatment for suspected lumbar intervertebral disc herniation, were related to outcome after five-years, as assessed by the categorization of occupational handicap of the International Classification of Impairments, Disabilities and Handicaps (ICIDH). The sample consisted of 276 patients admitted to hospital in the years 1980-1982. Based on established clinical indications 179 patients underwent lumbar disc surgery and the remaining 97 received further conservative treatment. One year after hospitalization a physician and physiotherapist examined the patients at the Rehabilitation Research Centre (RRC) of the Social Insurance Institution in Finland. In 1986 the patients returned to the RRC for further evaluation of their occupational handicap at that time. Stepwise logistic regression analysis showed that sensory deficits of legs, tenderness in lumbar extension, decreased repetitive trunk flexion performance (sit-up test), decreased lumbar lordosis and tightness of hamstrings were indicative of a poor outcome in the operated group. For the non-operated patients no statistically acceptable regression analysis model could be established.

Adult↗
PubMed1984

Hormonal and personal effects on estrogen receptors in breast cancer.

Since high estrogen receptor protein in breast cancer tissue has a favorable effect on patient prognosis, factors influencing estrogen receptor levels are the focus of this research. From April 1977 to December 1978, 246 histologically confirmed primary breast cancer patients from three North Carolina hospitals were interviewed and provided with photographic displays of hormonal products. Their tumor tissue was analyzed for estrogen receptor protein by the sucrose density gradient assay. Data were analyzed by multiple regression and logistic regression techniques. No statistically significant relationship was found between gonadal hormone usage and estrogen receptor levels, although recent users of either oral contraceptives or replacement estrogens had lower receptor levels. Users of progestogen-predominant and -intermediate types of oral contraceptives also had lower estrogen receptor protein. An effect of menopausal status, independent of age, was not evident. Statistically significant associations involved age and race among postmenopausal patients. With black breast cancer patients aged less than 70 years as the reference group, the relative odds of having high estrogen receptor protein (greater than 10 fmol/mg of cytosol protein) were 2.4 times greater among whites aged less than 70, 1.6 times greater in blacks aged greater than or equal to 70, and 12.8 times greater in whites greater than or equal to 70 years. These differences in estrogen receptor protein may contribute to the black-white differential in breast cancer survival.

Adult↗
PubMed1991

Socioeconomic, demographic and environmental determinants of infant mortality in Nepal.

The Nepal Fertility and Family Planning Survey of 1986 demonstrated that demographic variables, previous birth interval and survival of preceding child, still predominated as determinants of infant mortality, particularly in rural areas of Nepal. However, in urban Nepal, where the level of socioeconomic development is higher, an environmental variable, along with previous birth interval and survival of preceding child emerges as important in determining infant mortality. Separate policy measures for child survival prospects in rural and urban Nepal are suggested.

Adolescent↗
PubMedDate not supplied

On the convergence of Das Gupta's model of fertility projection.

A regression model expressing the age-specific fertility rates at a given point of time as a variable dependent upon the same at the preceding points of time has been developed by Das Gupta for purposes of fertility projection. The model, judged by the values of the multiple correlation coefficients, appears to work very well with the U.S. data set, and it also produces limiting values of the fertility rates when the projection is carried far into the future. The conditions of convergence of such a model have been determined in this paper according to which the model's usefulness for purposes of long-term projection of fertility rates appears to be severely restricted.

Adult↗
PubMed1999

Morphometric analysis of primordial follicle number in pigtailed monkey ovaries: symmetry and relationship with age.

We previously described a modern, three-dimensional counting method for determining primordial follicle (PF) numbers in primate ovaries using a combination of fractionator and physical dissector techniques. The purposes of our current study were 1) to apply our method to describe intraindividual differences in PF numbers between ovaries and 2) perform a linear regression analysis of age versus mean PF number per ovary. Ovaries from 16 pigtailed monkeys (Macaca nemestrina) age 0.85-12.5 yr were examined. Both ovaries were available from 11 subjects. The difference between ovaries ranged from 2% to 22% (mean +/- SD, 10 +/- 7%) and was not statistically significant. Regression analysis of data from all 16 subjects displayed a log-linear relationship according to the equation log N(a) = 4.8542 - 0.0714(age) where N(a) is the number of PF at a given chronological age. The fit for this model was highly significant (r(2) = 0.73, p </= 0.0001). Extrapolation of the model suggests that there are 71 483 PF in each ovary at the time of birth. We conclude that right and left ovaries differ little and that PF numbers follow a log-linear rate of decline during the reproductive years in this species.

Aging↗
PubMed1998

A randomized clinical trial of cryotherapy, laser vaporization, and loop electrosurgical excision for treatment of squamous intraepithelial lesions of the cervix.

OBJECTIVE: To compare cryotherapy, laser vaporization, and loop electrical excision for treatment of squamous intraepithelial lesions (SILs). METHODS: Women at least 18 years old with biopsy-proven SIL, negative pregnancy tests, negative findings on endocervical curettage, satisfactory colposcopy examinations, and congruent Papanicolaou smear and biopsy results were assigned randomly to treatment after stratification by SIL grade, endocervical gland involvement, and lesion size; they were evaluated 1, 4, 8, 12, 16, 20, and 24 months after treatment. Data were analyzed using chi2 statistics, logistic regression analysis, and the Cox proportional hazards model. RESULTS: Of 498 patients assigned, 108 were excluded (most because of inadequate follow-up), leaving 390 (139 cryotherapy, 121 laser vaporization, 130 loop excision) for analysis. All were followed 6-37 months (mean 16). There were no statistically significant differences in complications, persistence (disease present less than 6 months after treatment), or recurrence (disease present more than 6 months after treatment). Risk of persistent disease was higher among women with large lesions (risk ratio [RR], 18.9; 95% confidence interval [CI], 3.2, 110.6). Recurrence risk was higher among women aged 30 years and older (RR, 2.1; 95% CI, 1.2, 4.3), those with human papillomavirus type 16 or 18 (RR, 2.1; 95% CI, 1.1, 4.0), and those who had had prior treatment (RR, 2.1; 95% CI, 1.1, 3.9). CONCLUSION: The data support a high success rate with all three modalities. No significant difference in success rates was observed between the three treatments in our population. Additional attention and research should be directed toward the higher risk patients identified above.

Adolescent↗
PubMed1981

Contraceptive embarrassment and contraceptive behavior among young single women.

The purpose of this paper was to determine factors predictive of contraceptive embarrassment and to determine the relationship of contraceptive embarrassment to contraceptive use among young unmarried females. Contraceptive embarrassment was measured by an 8 item scale. Data were obtained from 265 nonvirgin high school and university females who were currently dating. Using simple multipe regression 30% of the variance of the embarrassment scale was explained, with the most important predictors being parental attitude toward premarital intercouse and sexual guilt. Other significant predictors were attitude to planning ahead to use contraception, perceived difficulty in obtaining contraceptive devices, and peer attitudes toward premarital intercouse. The embarrassment scale had significant correlations with contraceptive use.

Attitude↗
PubMed1990

Oral contraceptives in women with diabetes.

We evaluated the association of oral contraceptive use with the presence and severity of diabetic retinopathy, hypertension, and glycosylated hemoglobin in women of childbearing age who have diabetes. Neither current or past use nor number of years of use of oral contraceptives was associated with severity of retinopathy, hypertension, or current glycosylated hemoglobin. In conclusion, further study of various birth control methods in young women of childbearing age should be considered.

Adolescent↗
PubMed2000

Simultaneously model-unbiased, design-unbiased estimation.

This paper proposes a class of inferential procedures (incorporating both design and estimation elements) that yield estimates of means that are simultaneously model unbiased and design unbiased. Classical regression procedures yield conditionally unbiased estimators for the mean (conditioning on the model and choice of observation points). In contrast, design-based methods yield estimators that are unconditionally unbiased on matter what the form of the underlying model. Variance properties of the proposed class are examined, and applications to bioavailability, water quality from mine run-off, and finite population regression estimation are considered. The proposed procedures perform well, especially in the typical case where a model is only approximately correct.

Animals↗