جستجوی مقالات مرتبط با کلیدواژه "poisson regression" در نشریات گروه "پزشکی"

Brucellosis is a common disease between humans and animals that still exists in most parts of Iran. This study investigated the epidemiology of brucellosis in Mazandaran province, Iran, during a nine-year period from the beginning of 2009 until the end of 2017.

This descriptive-analytical study was conducted based on the guidelines of Iran’s Ministry of Health and Medical Education on the data of 3732 brucellosis patients. The data were retrieved from the private and public laboratories gathered in a Ministry’s database. We used the generalized estimating equation (GEE) for the Poisson regression model (Poisson GEE) to examine the disease incidence based on the studied variables.

The age-standardized incidence rate has been 13.2 per 100,000 people (15.6 in men and 21.3 in rural areas). The highest rate (17.2) was in 2012 and the lowest (10.6) was in 2014. The cities of Galugah (35.0) and Qaemshahr and Fereydunkenar (3.0) had the highest and lowest incidence rates. The patients’ professions had been 30% housewives, 14.9% ranchers, and 14.4% livestock-related professions. The highest standard incidence was in people older than 55 years old. The results of the Poisson GEE regression model showed that the disease incidence has been higher in men compared to women and in rural compared to urban areas (risk ratio = 1.5, 3.4, P = 0.001).

Mazandaran faces the risk of increased brucellosis incidence rate. Taking preventive measures, including educating people, are recommended for the near future.

Type I bipolar disorder is a chronic mental illness with a recurrent nature.

This study was done to identify risk factors for relapse in patients with type I bipolar disorder using Poisson regression.

In this retrospective cohort study, data were collected from 331 bipolar type I patients admitted to the largest and the only psychiatric hospital in West of Iran, Kermanshah. Poisson regression was used to identify risk factors for the relapse numbers in STATA software.

One hundred sixty six of 331 patients (50.2%) were men. Mean and standard deviation of relapse rates in males and females were 2.93 ± 0.16 and 2.93 ± 0.15, respectively. In multivariate analysis, the results showed that with increasing age of disease onset (RR = 0.98, 95% CI :(0.97 -0.99)) and adherence to medication (RR = 0.86, 95% CI :(0.75-0.98)), the relapse rate ratio decreased, and being divorced or widowed (RR = 1.20,95% CI :( 1.01-1.45)), cigarette smoking (RR = 1.28, 95% CI: (1.52-1.09)), and birth season of winter (RR = 1.18, 95% CI: (1.00-1.41)) increased the rate ratio of relapse.

The results of this study showed that the frequency of relapse was lower in bipolar patients younger at the time of disease onset, cigarette smokers, divorced or widowed patients, and patients with no treatment adherent. It is necessary to prevent or reduce the likelihood of relapse, especially in patients who do not have therapeutic adherence through educational and therapeutic measurements.

The aim of this study was to estimate the standardized incidence rate (SIR) and also the relative risk (RR) of colorectal cancer (CRC) in Iran and to determine the distribution of CRC risk in a map after adjusting socioeconomic risk factors.

The growth of CRC incidence rate in Iran is a major public health problem and identifying high-risk regions is essential for further intervention.

For this cross-sectional study, all CRC cases that occurred in 30 Iranian provinces between 2005 and 2008 were collected according to the International Classification of Diseases (ICD-10). In addition, socioeconomic information was extracted from statistical center of Iran. Bayesian and Poison regression models were fitted to identify significant covariates. For RR estimating, the spatial analysis using GIS technique was carried out.

The Bayesian method with increasing precision of the parameter estimates had a better fit. According to spatial model, East Azerbaijan province had a high (11.14) and South Khorasan province had a low (0.22) risk of CRC in the period of study. SIR for the male population was 1.92 ± 3.25, and for the female population it was 1.85 ± 3.37.

There is a non-uniform spatial pattern of CRC risk in Iran. According to the results, North, Northwest and some parts of West and Central provinces of Iran are identified as high-risk areas; thus, it is recommended that health policymakers, especially in these areas, have more intervention measures. Further studies are needed to map the RR adjusted for nutrition factors.

Mammography is a valuable tool for early diagnosis of breast cancer in asymptomatic women. Considering the high prevalence of breast cancer in Iranian women and the low participation in mammography screening program, the purpose of this study was to investigate the factors affecting frequency of mammography screening in women over 40 years of age using zero-. In this study, the required information about number of performing mammography in women’s’ lifetime, demographic characteristics and behavioral risk factors were obtained through interview based on a researcher-made questionnaire. To investigate the factors affecting mammography, zero inflation Poisson regression models were performed using Win Bugs software. The mean (SD) age of women participating in this study was 49.87 (6.76). 77% of the participants have never undergone mammography, 8.9% once, 6.9% twice, 6.7% three times, and 0.5% more than three times. Age had a positive effect on the number of mammograms in the women who have perform mammograms at least once. Having a family history of cancer and breast cancer, middle compared to low economic status, higher compared to low education and menopause were significantly associated with lower probability of never performing mammography. Given the relatively low participation of women in mammography, more facilities are needed for high risk women (aged 40-70) to facilitate the diagnosis process.

The mother's role in health of family members is very important, and her death has uncompensable losses for the family and society. With regard to the importance of maternal mortal ratio in development indicators of the United Nations (UN) this study is conducted to investigate the impact of socioeconomic factors on maternal mortality ratio
In the current study the role of socioeconomic factors on maternal mortal ratio was investigated by panel regression. Data include number of maternal mortality, total fertility, number of hospital beds, number of midwifes, number of physicians and urbanization in the period between 2008 and 2012. Since the dependent variable was in count form, Poisson estimator, Hausman test, and Breusch-pagan test were used.
Based on the findings, household’s income and fertility rate had direct and inverse association with maternal mortality, respectively. Investigating the association between midwifes and maternal mortality showed that increasing the number of midwifes decrease it, but no significant association was found between the number of physicians and maternal mortality. Number of hospital beds also showed a direct association (significant at 10%).
All of the variables to somehow are related with the level of development. In developed regions, per capita income, access to health resource and urbanization is higher and there is more equity in distribution of health resources. It could be concluded from the findings that by increasing the level of development, maternal mortality would decrease.

Poverty and low socioeconomic status are the most important reasons of increasing the global burden of tuberculosis, not only in developing countries but also in developed countries for particular groups. The purpose of this study was to assess the association between socioeconomic factors and the number of tuberculosis patients using quantile regression for count data.
This cross-sectional study was conducted on 11,320 tuberculosis patients from March 2010 to March 201 in Iran. Data was gathered from the 345 sections of Iran by Ministry of Health and Medical Education and Statistical Center of Iran. The jittering method was applied for smoothing, and then, the quantile regression for count data was fitted. The AIC was used to compare the fitness of quantile regression for count data model and Poisson log-linear model. The R (3.0.1) software and Quantreg and AER packages were used for all analysis and modeling of the data.
The results of fitting the quantile regression for count data showed that in all percentiles, the more increase in immigration rate, illiteracy rate, unemployment and urbanization rates, the more tuberculosis morbidity rate was increased. The maximum increase of tuberculosis due to immigration rate, urbanization rate, unemployment rate, and illiteracy rate was in 95th percentile (β ̂=0.315), 85'Th percentile (β ̂=0.162), 75'Th percentile (β ̂=0.114), and 95'Th percentile (β ̂=0.304), respectively. For 50th percentiles and higher percentiles, with increasing the sum of physicians to the number of population, the tuberculosis morbidity rate was decreased, and the maximum decrease was in 95'Th percentile (β ̂=-0.1). For all percentiles, the AIC showed that quantile regression for count data had been a better fit to data.
With respect to the relationship between socioeconomic factors and TB rate, health care observers should pay close attention to improving these factors in Iran to reduce the TB mortality and morbidity.

Iraq fought three wars in three consecutive decades, Iran-Iraq war (1980-1988), Persian Gulf War in 1991, and the Iraq’s war in 2003. In the nineties of the last century and up to the present time, there have been anecdotal reports of increase in cancer in Ninawa as in all provinces of Iraq, possibly as a result of exposure to depleted uranium used by American troops in the last two wars.This paper deals with cancer incidence in Ninawa, the most importance province in Iraq, where many of her sons were soldiers in the Iraqi army, and they have participated in the wars. The data was derived from the Directorate of Health in Ninawa. The data was divided into three sub periods: 1980-1990, 1991-2000, and 2001-2010. The analyses are performed using Poisson regressions. The response variable is the cancer incidence number. Cancer cases, age, sex, and years were considered as the explanatory variables. The logarithm of the population of Ninawa is used as an offset. The aim of this paper is to model the cancer incidence data and estimate the cancer incidence rate ratio (IRR) to illustrate the changes that have occurred of incidence cancer in Ninawa in these three periods. There is evidence of a reduction in the cancer IRR in Ninawa in the third period as well as in the second period. Our analyses found that breast cancer remained the first common cancer; while the lung, trachea, and bronchus the second in spite of decreasing as dramatically. Modest increases in incidence of prostate, penis, and other male genitals for the duration of the study period and stability in incidence of colon in the second and third periods. Modest increases in incidence of placenta and metastatic tumors, while the highest increase was in leukemia in the third period relates to the second period but not to the first period. The cancer IRR in men was decreased from more than 33% than those of females in the first period, more than 39% in the second period, and regressed to 9.56% in the third period. Our paper confirms the media reports that there are increases in the number of cancer cases, but when it analyzed statistically with population growth in the Ninawa province, there are decreases in incidence rates in most cancer types.