Modelo predictivo para uso del condón y consumo de alcohol como conductas de riesgo relacionadas con el contagio de VIH/SIDA en trabajadoras sexuales de Bogotá- 2015
Chavarriaga Rios, Marcia
Objective: To determine a predictive model for condom use and alcohol-related risk behaviors as the spread of HIV / AIDS in women sex workers in the city of Bogota in 2015. Methods Cross-sectional study with observational design, 255 female sex workers in the city of Bogotá were taken; The information analyzed was taken from the study in five cities in Colombia in 2015, the hypotheses were supported in the association between sociodemographic conditions, knowledge, practices, habits, social support and self-employment of women sex workers they could explain and predict the adoption of risk behaviors for HIV / AIDS as they are condom use and alcohol use in the exercise of their occupation. Results The average age of onset in sex work was 22.1 ± 7.1 years, three-quarters are single and living in stratum two and three; 96.5% reported using a condom with the last client and 27.8% of them consumed alcohol during their last service. In the conduct of risk condom use, were found associated with each other, age [OR = 1.10 (1.03 to 1.17)], living in stratum two [OR = 7.7 (1.5 to 39 5)], the income from sex work [OR = 1.0 (1.0 to 1.0)], the availability of condoms for service [OR = 0.03 (0.008 to 0.16)] and count with another planning method (tubal ligation) [OR = 4.47 (1.0 to 18.3)]. Risk behavior in alcohol , was found associated entity other : socioeconomic two [OR = 5.8 ( 1.54 to 22.3 ) ] , secondary school level [OR = 0.12 ( 0,16- 0.96) ] , living with other relatives [OR = 3.45 ( 1.7 to 7.02 ) ] , income from sex work [OR = 1.0 ( 1.0 to 1.0 ) ] and the site where the service [OR = 0.07 ( 0.04-0.15 ) ] is offered . After adjusting, it was found that the variables that best explain condom use were age [OR = 1.1 (1.02 to 1.17)] and condom availability [OR = 0.04 (0.008 to 0.024)] the model had little sensitivity 33.3% and good predictive value ( 84.6 %). The variables that best explained the consumption of alcohol during service were age [ OR = 0.95 ( 0.91 to 0.98 ) ] , number of clients per week [OR = 0.9 ( 0.90 to 0.98 ) ] , where the service site offers [OR = 7.1 ( 3.45 to 14.8 ) ] , and socioeconomic status [OR = 1.8 ( 0.90 to 3.83 ) ] , resulting in a model with good sensitivity ( 71.8 % ) and good predictive value (86.4 %). Conclusions Aspects such as age, socioeconomic status, education, marital status, income from sex work, age of onset in sex work, many old customers in the last week, condom availability to provide the service and tubal ligation as a method different planning, were associated statistically with condom use. However when adjusting the variables only age and condom availability remained as explanatory variables. It should be noted that although the model showed good predictive value (84.6%), the accuracy of their estimates was low due to the infrequency of no condom use with the last (3.5%) client, and the sensitivity of just model was 33.3%. On the other hand, factors such as age, socioeconomic status, education level, income, site offering the service, family composition, number of children, number of clients served in the last week and a number of old customers showed statistical association with alcohol consume. However, by adjusting the variables only age, socioeconomic status, place where the customer service number is offered weekly statistical association maintained; also observed that the socioeconomic status (one and two) and place where the service (establishment) is offered, are risk factors for alcohol consumption in exercise of the occupation and the young age and a small number of clients per week behave as protective factors for alcohol consumption. The predictive model developed for risk behavior of drinking, with a sensitivity of 71.8% and 86.4% predictive power.