Sexual Behavior Changes And Medical Costs With Markov Analysis Of Rectal Swab Self-collection For Testing Chlamydia Trachomatis And Neisseria Gonorrhoeae In HIV-positive Men Who Have Sex With Men

Objective: Men who have sex with men(MSM)are disproportionately affected by human immunodeficiency virus(HIV),genital tract Chlamydia trachomatis(CT)and Neisseria gonorrhoeae(NG)infections.Coinfection with CT and NG can prolong the course of disease.CT/NG infections in HIV-positive MSM are predominantly asymptomatic in the rectal region,and failure to screen promptly can lead to further transmission of CT and NG.Regular CT and NG testing of HIV-positive MSM to maximize the detection of CT and NG infection and timely link-to-care for treatment after diagnosis is an important tool to control the disease burden in HIV-positive MSM.In this study,we propose to conduct a study on the status of previous CT and NG testing and rectal infection in the HIV-positive MSM population,with the aim of capturing the rates of previous CT and NG testing and rectal infection in the HIV-positive MSM population,and obtaining information on changes in condom and sexual behavior characteristics after knowledge of infection.The cost-utility evaluation by Markov model was conducted to further explore the feasibility of the rectal swab self-collection testing strategy in the HIV-positive MSM.Methods: In the first chapter of this study,HIV-positive MSM were recruited jointly based on recommendations from the First Affiliated Hospital of China Medical University and community work organizations,and completed an online self-administered epidemiological questionnaire to obtain information on their socio-demographics,sexual behavior,STIs testing behavior,other HIV-related behaviors,and STIs testing perceptions.This section was analyzed by 2test/Fisher exact /Logistic regression model for factors associated with previous CT/NG testing,and multivariable regression model corrected for education and monthly income.In the second chapter of this study,HIV-positive MSM were recruited,epidemiological surveys and laboratory tests were conducted at baseline,and study subjects with positive laboratory tests for CT and NG completed follow-up epidemiological questionnaires as required 6 months after enrollment,and information on socio-demographics and sexual behavior of study subjects was collected online during baseline and follow-up to obtain CT and NG infection rates among HIV-positive MSM,and to longitudinally analyze Differences in behavioral changes among different subgroups of HIV-positive MSM after knowledge of infection status and assessment of acceptance of rectal swab self-collection.Logistic regression models were used to analyze the association factors of CT/NG rectal infection and generalized estimating equation was used to compare the differences in frequency of sexual behavior,type of sexual partners and other relevant information before and after the knowledge of CT and NG infection in different subgroups.The third chapter of this study was to assess the concordance between rectal swab selfcollection testing for CT and NG and medical clinic clinician-collection for CT and NG and construct a Markov model for cost effectiveness evaluation.HIV-positive MSM were recruited offline and enrolled after signing informed consent for dual testing of CT and NG by medical staff swabbing at medical institutions and home-based self-collection,to assess the consistency of CT and NG testing between the two groups.Kappa concordance test was used to assess the concordance between home self-sampling for detection of CT and NG and sampling by medical personnel in health care institutions.A cost-effectiveness Markov model was constructed to analyze the medical costs of self-collection this test for CT and NG,with parameters corrected for half-cycle and sensitivity analysis by cyclone diagram.The third part of this study was to assess the concordance between rectal swab self-testing for CT and NG and medical facility medical staff sampling for CT and NG and construct a Markov model for cost-utility evaluation.HIV-positive MSM were recruited offline and enrolled after signing informed consent for dual testing of CT and NG by medical staff sampling at a medical institution and home self-sampling.Kappa concordance test was used to assess the concordance between home self-sampling for CT and NG and medical staff sampling at the medical institution.A Markov model was constructed to analyze the medical costs of self-collection for CT and NG using discounting and half-cycle correction for cost and utility parameters,and sensitivity analysis was performed using tornado diagrams and cost-effect quadrant plots.Results: 1.(1)From May 2021 to January 2022,430 HIV-positive MSM were included.334 were asymptomatic and 96 were symptomatic.The median age was 35 years(IQR: 30-45 years),10.7% psychologically identified gender as female,2.1% were not adherent to antiretroviral therapy,and in terms of sexual orientation,69.8% self-identified as homosexual.60.5% of the study subjects had a regular same-sex sexual partner and 52.3% adhered to condom use when having sex with a regular same-sex sexual partner.(2)The testing rate for previous testing for either CT/NG was 19.5%,the rates of previous CT and NG testing were 5.1% and 18.8%,respectively,and 4.4% had dual testing for CT and NG.The rate of testing for previous sexually transmitted infections(STIs)in HIV-positive MSM was 72.6%,with 7 cases having been tested for rectal site STIs and only 1 case for rectal NG testing.In the past 12 months,the detection rate for either CT/NG testing was 5.1%,CT and NG testing rates were 1.6% and 4.7%,respectively,and 1.2% had dual testing for CT and NG.In the last 12 months,the detection rate of HIV-positive MSM with previous STIs was 44.7%.(3)Factors associated with testing for STIs among HIV-positive MSM: Barriers included distance from the nearest hospital with Tertiary care/dermatological specialties(2 km-10 km/>10 km vs.<2 km,AOR=0.386/AOR=0.221,P<0.05),perceived severity of infection STIs dimension(AOR=0.453,95% CI: 0.293-0.702).The promoting factors were age(26 to 45 years / >45 years vs.18 to 25 years,AOR=3.170/AOR=5.788,P<0.05),disclosure of sexual orientation to others(vs.no disclosure)(AOR=3.611;95% CI: 2.000-6.520),condom use with regular sexual partners(vs.no use)(AOR=2.071,95% CI: 1.151-3.726)and self-efficacy of STIs testing(AOR=2.393,95% CI: 1.575-3.636).2.A total of 389 HIV-positive MSM were laboratory tested for CT and NG at baseline,of whom 189 tested positive and completed 6-month follow-up,and 1 was lost to followup.(1)Baseline sexual behavior and other characteristics: low frequency of sexual behavior,mainly at less than 1 time per month(58.9%),60.9% had regular same-sex sexual partners,and 51.9% used condoms for each sexual act,38.0% had same-sex casual sexual partners,19.0% had HIV-positive same-sex sexual partners,22.4% had symptoms of anal bleeding in the last 3 months,and 5.7% had oral/ 5.7% had experience of oral/oral inhalation of recreational special drugs.(2)Rectal CT/NG infection status: The prevalence of any CT/NG infection in the rectal area of HIV-positive MSM population was 19.0%,with asymptomatic percentage of those infected was 73.0%.The rates of rectal CT and NG infection were 15.9% and 6.4%,respectively,and the rate of mixed infection was 3.3%.The association factor for any CT/NG rectal infection was age(18 to 25 years vs.older than 45 years,AOR=3.862,95% CI: 1.457-10.236).(3)Changes in sexual behavior before and after knowledge of infection: Compared with the period before knowledge of rectal CT/NG infection,the proportion of sexual roles“both”decreased in the last 3 months after knowledge of infection(6.5% vs.25.8%/20.0% vs.40.0%,P<0.05)and sex with casual partners(17.7% vs.35.5%,P<0.05),a decrease in the proportion of sex with HIV-positive homosexual partners(16.1% vs.41.9%/16.0% vs.44.0%,P<0.05),and a decrease in the proportion of enema before and after sex(29.0% vs.59.7% /28.0% vs.60.0%,P<0.05),an increase in the proportion of sex with a regular homosexual partner,and no more than 20% with multiple regular homosexual partners.(4)Link-to-care behavior(link-to-care): the proportion of link-to-care after positive test was 56.1%,and the time interval between linkto-care visits was mostly within 1 month 67.9%,of which the most chose to visit medical institutions,accounting for 85.8%.19.0% of the timely link-to-care(within 7 days),and the most influential factor was "fear that someone I know might find out"(very/very influential was 40.2%),followed by the belief that the disease would heal on its own(37.6%)and concern about available treatments(33.3%).(5)Acceptance of self-collected rectal specimens for STIs testing was 75.8%,mainly due to time saving(75.61%)and better privacy(70.73%).371(86.3%)respondents were willing to recommend their sexual partners to self-collect rectal specimens for testing.3.A total of 296 investigation subjects were included in the home self-collected rectal sample consistency evaluation study.(1)The results of Kappa test showed that the positive rate of CT in the home collected rectal swab was 11.8%,and the positive rate of CT in the medical institution was 12.8%,with a concordance rate of 94.26% and a Kappa value of 0.73.The positive rate of NG in the home collected rectal swab was 5.1%,and the positive rate of NG in the medical institution was 4.7%,with a concordance rate of 97.64% and a Kappa value of 0.75.The above analyses were statistically significant(P<0.001).(3)Costutility analysis was performed by constructing a Markov model,and the results showed that the cost-utility ratio(CUR)of home self-sampling for CT/NG testing was RMB 35.93/QALY,and the CUR of physician-collected samples for CT/NG testing in medical institutions was RMB 104.34/QALY.The incremental cost-utility ratio(ICUR)for physician sampling was RMB-204,702.27/QALY,which is below the willingness-to-pay(WTP)threshold criterion,and self-collection for CT/NG testing is the recommended option.Sensitivity analysis of the effect of changes in each parameter on the simulation results showed that the scatter points were located below the WTP threshold line and the model analysis results were more stable.Conclusion: 1.The low prevalence of previous CT and NG testing in our HIV-positive MSM population is mainly related to the high proportion of asymptomatic STIs,inaccessible healthcare services,subjective lack of awareness of STIs,stigma,and sexual behavior characteristics of this population.Therefore,screening for CT and NG in the HIV-positive MSM population should be strengthened,and knowledge of STIs and the efficacy of STIs testing should be disseminated to this population to improve the detection rate of CT and NG in the HIV-positive MSM population in China and find more infections.2.HIV-positive MSM had a higher rate of rectal CT/NG infection,and a lower rate of unsafe sex after knowing the infection,suggesting that screening for CT/NG in this population should be improved to reduce further transmission of CT/NG;the proportion of link-to-care behavior after HIV-positive MSM infection was higher,among which the rate of timely link-to-care(within 7 days)is low,the main barriers for delaying medical treatment were exposure of privacy and asymptomatic.However,there was a high willingness to accept self-collection testing,mainly because of "time saving" and "good privacy",which can be tried as a screening instrument.3.The consistency of CT/NG testing by home self-collected and CT/NG testing by physician sampling in medical institutions is good,which can be used as an alternative to daily CT/NG testing;the strategy of CT/NG testing by self-collected is more cost-effective than the standard group,and it is worth promoting as it provides more health benefits at a lower cost.