| Part one: Analysis of the characteristics of HIV/AIDS epidemic in1996～2012in GuangxiOBJECTIVETo analyze the characteristics of HIV/AIDS epidemic in1996～2012inGuangxi, and explore regional difference of HIV/AIDS epidemic, so as toprovide policy-makers with scientific evidence.METHODSCollect data of cases who were new-confirmed with Guangxi censusregister and with positive HIV antibody in1996～2012(including those of HIVinfections and AIDS patients); Basing on a variety of statistics tables andcharts,descriptive analysis was applied to the medical consumption. In addiction,divide the total of109counties(cities/districts) of Guangxi into three epidemicareas of high, middle and low HIV cumulative prevalence of infection (CPI),cumulative prevalence (CP), cumulative mortality rates(CMR) and cumulativefatality rates(CFR), totally four epidemic index (hereinafter referred as the fourepidemic index) by natural breakout methods of Geography Information System (GIS), compare epidemic characteristics of the four epidemic index amongdifference epidemic areas.RESULTS1.Overall characteristics of HIV/AIDS epidemic in Guangxi.(1)Thecumulative number of source of infection was very large.The cumulative HIVinfections and AIDS patients who were new-reported between1996～2012withGuangxi census register was83241, among them47094were HIV infections,36147were the AIDS patients, the corresponding proportion was56.58%,43.42%,respectively.(2) The speed of cumulative number of infections appearsfirst quick back slow trend. Before2003, the increase of the number ofinfections was slow, after2003, number is growing rapidly, but there is adownturn in2012.(3) The characteristics of infected people is obvious. Theywere predominantly male (72.34%); age concentrated in15to49years old,accounting for68.94%;the Han nationality were61.11%;57.05%were themarried;79.11%with junior middle school or lower education level;Professional for farmers (including migrant farmers) and workers is givenpriority to, accounting for60.39%.(4)The main routes of transmission washeterosexual transmission. Giving priority to heterosexual transmission,accounting for66.31%.(5) The non-marital sexual behavior became graduallyoutstanding. Non-marital sexual behavior (at least one non-marital heterosexualsex partner) was the main risk behavior, accounting for42.56%; of which80.59%with multiple unmarried partners.(6) Clinical diagnosis is the mainfound way. About65.68%of the patients were found when they visit theirphysicians.2. Division of HIV/AIDS epidemic areas and characteristics in Guangxi.Divided total of109countries (cities/districts) of CPI, CP, CMR, CFR in Guangxi into three groups by natural breakout method of GIS software,corresponding to the high, middle and low epidemic areas. High, middle andlow epidemic areas of CPI included15,45,49countries (cities/districts),respectively, the corresponding interval were2.99‰~6.91‰,1.21‰~2.98‰,0.18‰~1.20. High, middle and low epidemic areas of CP included13,41,65countries (cities/districts), respectively, and the corresponding interval were1.54‰～4.24‰,0.55‰～1.53‰,0.06‰～0.54‰. High, medium and lowepidemic region of CMR included4,20,85countries (cities/districts),respectively, and the corresponding interval were0.58‰～1.26‰,0.17‰～0.57‰,0.00‰～0.16‰. High, middle and low epidemic areas of CP included18,45,46countries (city/district), respectively, the corresponding intervalwere27.25%～50.00%,13.28%～27.24%,0.00%～13.27%.High, middle and low epidemic areas of CPI and CP had a higher level ofconsistency. When base on CPI, the coincidence rate of CP and CPI were86.67%（13/15),84.44%(38/45),93.88%(46/49),respectively for the high,medium and low epidemic areas.Countries (cities/districts) with high CPI and CP had a head start ofHIV/AIDS epidemic, they mainly distributed in the border cities, Liuzhoucenter districts and its surrounding. Countries (cities/districts) with mediumCPI and CP can divided into two big gather bundles, one take high epidemiccountries (cities/districts) of Liuzhou as center then covering the surroundingcountries, the other started with high epidemic countries (cities/districts) of theborder cities and then spread to the inside of Guangxi. The low epidemic areasof CPI and CP were mainly distributed surrounding the high and medium areas.When displayed graphically on a map, it’s easy to find that the high andmedium epidemic areas of CPI and CP mainly distributed on the central part of Guangxi tilt about45°angle from northeast to southwest, which with relativelydeveloped economy and more convenient traffic. Consistency existed amongCPI, CP and CMR. However,77.78%(14/18) of the CFR high epidemic areasshowed low CPI, CP and CMR.CONCLUSIONS1. Phenomenon of HIV infects be found later was so serious that about50%infects was AIDS when confirmed in the first time.The infects weredominated by male, married,Han, age15-49years old, Junior middle school orlower education level, profession of farmers (including migrant farmers) andworkers. Non-marital heterosexual intercourse was the dominant risk behavior,and heterosexual intercourse was the main rout of transmission As a result, thecorresponding prevention and control strategies and measures, such aspromoting HIV detection method,strengthen HIV/AIDS related healtheducation among people with lower education level, further roll out use ofcondoms100%in sex, especially in un-married sex will have majorimplications for prevent HIV/AIDS in Guangxi.2. There were practical significance in natural breakout method of GIS.The high,medium and low epidemic areas divided by natural breakout methodof GIS had a high coincidence rate, but it is necessary to consider moreinfluence factors of society and behavior.3. Since CPI, CP, CMR,CFR appear mixed in different countries (cities/districts), HIV/AIDS prevention and control should focus on areas thatappeared high and medium CPI and CP; and measures should be taken to reducecase fatality in countries (cities/districts) with lower CPI and CP, but also withhigh CFR. Part two: Study on mathematical discriminant model of HIV/AIDSepidemic areas in GuangxiOBJECTIVETake cumulative prevalence of infection（CPI), cumulative prevalence(CP)and cumulative mortality rate (CMR) of HIV/AIDS as the main effect indicators,then stratified analysis, so as to find important related factors that affect theyhappen; establish mathematical discriminant model of high, middle and lowHIV/AIDS epidemic areas in Guangxi; Provide scientific evidence forHIV/AIDS prevention and control strategies.METHODSCollect eight indexes of economic society development which related withHIV/AIDS epidemic in Guangxi, analysis any one of them the relationship withCPI, CP,and CMR; Compare difference of economic society development indexamong the high-medium-low epidemic zones by one-way analysis of variance,(in this step, epidemic areas used by the standard of the first part of the GISnatural breakout method; Because CMR with high endemic areas include only4countries (cities/districts), we merged the high and medium area of CMR intothe high area); Use the method of stepwise regression analysis to establishmultiple linear regression model across by factors that influenced the CPI;Fisher stepwise discriminant method was used to establish mathematicaldiscriminant model of HIV/AIDS of different epidemic area categories inGuangxi, and evaluated the effect eventually.RESULTS1. Simple Spearman correlation analysis. Spearman correlation analysisbetween eight indexes of economic society development (X1: populationdensity, X2: the proportion of non-agricultural population,X3: the natural population growth, X4: per capita GDP, X5: per capita disposable income ofurban residents, X6: per capita net income of rural residents, X7:completion rateof nine-year compulsory education, X8: accumulation treatment rates of AIDS)and three epidemic index showed that X1, X2, X4, X5, X6, X7were associatedwith CPI, CP (P <0.10); X2, X4, X6, X7were related with CMR (P <0.10).2. Eeconomic sociology development factors that influence differentepidemic regions of CPI, CP, CMR:(1) X2、X4、X5、X6、X7were economicsociology development factors that affected the CPI, X7was factor that affectedmulti-factor model, X4、X7were factors that affected the discriminant models.(2) X2、X4、X6、X7were economic sociology development factors that affectedthe CP, X7was factor that affected multi-factor model, X4、X7were factors thataffected the discriminant models.(3) X2、X4、X7were economic sociologydevelopment factors that affected the CMR, X4was factor that affectedmulti-factor model and the discriminant models. So, X4was the commoninfluence factor of the three epidemic index, playing a positive effect. X7wasthe common influence factor of CPI and CP, it also playing positive effect.3. Establishment of the high,medium and low epidemic areas’sdiscriminant models of CPI, CP and CMR: Economic sociology developmentfactors with statistically significant in the Single factor analysis of variance,andX8, which was considered has closely relationship with HIV/AIDS epidemic,were used as the independent variables of stepwise Fisher discriminant model,so the corresponding discriminant models were as follow:3.1Discriminant function of epidemic areas of high, medium and low ofCPIHigh epidemic area of CPI： YH=-19.36+2.39×10-4X4+0.62X7Medium epidemic area of CPI：YM=-19.97+4.68×10-4X4+0.69X7 Low epidemic area of CPI: YL=-15.72+4.27×10-4X4+0.61X73.2Discriminant function of epidemic zones of high, medium and low ofCPHigh epidemic area of CP： YH=-19.32+1.80×10-4X4+0.61X7Medium epidemic area of CP：YM=-20.31+4.43×10-4X4+0.69X7Low epidemic area of CP： YL=-15.97+4.29×10-4X4+0.62X73.3Discriminant function of epidemic zones of high and low of CMRHigh epidemic area of CMR： YH=-3.78+3.78×10-4X4(Because CMRwith high endemic areas include only4countries (cities/districts), we mergedthe high and medium area of CMR into the high area)Low epidemic area of CMR： YL=-2.54+2.92×10-4X44. The effect evaluation of discriminant models of HIV/AIDS epidemicareasA test sample showed the discriminant model of CPI had an accuracy of51.38%, the high,medium and low endemic area was56.25%,58.18%,53.06%,respectively. Model of CP had an accuracy of54.13%, the high,medium and low endemic area was53.84%,48.78%,52.4%, respectively.Model of CMR rate had an accuracy of64.22%, the high and low endemic areawas48.00%,69.05%, respectively.CONCLUSIONS1. CPI, CP, CMR,CFR can fully reflect the strength of the HIV/AIDSepidemic in guangxi.2. Completion rate of nine-year compulsory education and per capita GDPwere factors that promote HIV/AIDS epidemic in Guangxi.3. Mathematical discriminant models could be considered as method thatapply to divide epidemic areas of HIV/AIDS in Guangxi, but more factors, such as biological factors and behavioral factors should be collected to improve theeffect of discriminant models. |
PDF Full Text Request