HIV Risk Report
Version 6.8
Private and anonymous · Built by people who've been in your shoes.
Your Risk, Step by Step
This report explains the encounter you described, the evidence used, the assumptions in the model, and the follow-up tests that can confirm your HIV status. It separates observed clinical evidence from modelling choices.
Step One: What Happened
| Exposure date | 7 October 2026 | Country | Republic of Korea (South Korea, assumed) |
|---|---|---|---|
| Your sex | Male, cisgender | Partner's sex | Female, cisgender |
| HIV status | Unknown | Reported ART | Not sure |
| Sexual activities | Oral and vaginal sex | Sexual role | Insertive |
| Condom | Yes, presumed intact | Ejaculation | No |
| Sex worker | Yes | Multiple recent partners | Yes |
| Injection-drug risk | Unknown, not multiplied | Wounds or inflammation | No |
| Circumcision | No | PrEP | No |
| Additional information | “The girls get tested monthly from what I know.” | ||
The condom is treated as intact because no breakage or slippage was reported. The model applies the condom reduction to vaginal intercourse. Condom coverage during oral sex was not separately stated and is not assumed for numerical oral-sex modelling. “Insertive oral” here refers to receiving oral sex.
Step Two: Analysing the Risk
1. Sexual activities and biological exposure
You reported insertive vaginal sex and insertive oral sex, meaning that you penetrated your partner vaginally and received oral stimulation. These activities are not interchangeable in an HIV risk assessment. For insertive vaginal intercourse without protection, the CDC's pooled estimate is four transmissions per 10,000 exposures, or 0.0004 as a decimal, when the partner has HIV with transmissible virus. This is an average across studies, not the probability for your particular partner. Infection requires infectious material to reach susceptible tissue, such as the urethral mucosa or vulnerable tissue on the penis. An intact condom covers the exposed penile surface and greatly reduces contact with vaginal secretions. Oral sex is very different. The CDC classifies both giving and receiving oral intercourse as low risk for HIV, and UK professional guidance estimates receiving fellatio at less than one transmission in 10,000 encounters with a person who has HIV and is not virally suppressed. Version 6.8 specifies one in 10,000 as a deliberately conservative oral-sex modelling proxy, not a measured oral transmission rate. Because the true rate is poorly quantified and the vaginal exposure has a validated activity-specific condom calculation, this report calculates vaginal exposure and discusses oral exposure separately rather than adding an arbitrary second probability. [1, 4, 5]
2. Partner background and South Korean prevalence
Because your partner's HIV status was unknown, the calculation starts with a population estimate instead of assuming either a positive or negative test result. The selected study projected approximately 20,839 people living with HIV in South Korea in 2020, explicitly including people who had not yet been diagnosed. South Korea's recorded 2020 population was 51,836,239. Dividing those figures gives an estimated total-population prevalence of approximately 0.0402016%, or about four people per 10,000 residents. That number describes the population of both sexes and all ages, not women specifically, and it does not tell us whether your individual partner had HIV. Because a reliable female-specific estimated-total-prevalence value was not established for this calculation, the all-sex national estimate is used as an explicitly labelled population proxy. This substitution is uncertain and may not reflect the prevalence in adult women. Your partner reportedly worked in the sex industry and had multiple recent partners. Version 6.8's Asia-Pacific modelling rule applies one sex-worker multiplier of 3.0 to this starting prevalence; it does not multiply by a second factor for multiple partners, thereby avoiding double counting. This produces an adjusted modelling estimate of approximately 0.1206048% for partner HIV prevalence. The threefold multiplier is an internal heuristic, not a measured South Korean sex-worker prevalence ratio. In Bayesian terms, population prevalence supplies a starting probability, but meaningful updates would require validated evidence about this individual's test results, timing and circumstances. The study itself was published in 2020 and projected future values using claims data primarily from 2009 to 2015, so the fallback is notably old. [2, 3]
3. Condom protection and other modifiers
The most important protective feature in your history is condom use during vaginal intercourse, with no reported breakage or slippage. The CDC estimates an 80% reduction in HIV acquisition among heterosexual people who consistently report condom use. Version 6.8 represents this with a multiplier of 0.20. This estimate comes from studies of consistent self-reported use over time, not a trial measuring the exact effectiveness of one intact condom during one encounter. It is deliberately not treated as 100% protection. Your partner's treatment and viral-suppression status were unknown. Treatment that maintains an undetectable viral load prevents sexual transmission, but reported ART use alone would not establish suppression. To represent the chance that a person living with HIV was diagnosed, receiving treatment and had a low measured viral load, this report uses a carefully labelled historical population model. A 2020 Korean study projected that 62.53% of people living with HIV were diagnosed. KDCA reported an ART treatment rate of 94.7% among reported living cases and a 96.9% rate of viral loads below 1,000 copies/mL among people with recorded tests in its counselling programme. Multiplying these measures gives a modelled 57.3802% suppressed fraction, leaving 42.6198% as the risk multiplier (x0.4262). These sources do not form a fully compatible treatment cascade: their denominators differ, and the 1,000-copy threshold is not the below-200 threshold used to establish U=U. Therefore, this is an estimated adjustment, not proof of this partner's treatment or viral suppression. [2, 11] No genital wounds or inflammation were reported, and therefore no wound-related increase was applied. You were uncircumcised, but Version 6.8 applies circumcision adjustments for insertive vaginal sex only when no condom was used. You did not use PrEP, so no additional PrEP reduction was calculated. Finally, not ejaculating is not counted as HIV protection for an insertive participant, because acquisition risk arises from the partner's fluids contacting your own susceptible tissues. [1, 4]
4. Your note about monthly testing
You added that “the girls get tested monthly from what I know”. This information matters because repeated screening can help people detect infection and obtain treatment earlier. However, the wording also makes clear that you do not have a confirmed test result for this specific partner. We do not know which tests were performed, their exact dates, whether the results were negative, or whether a subsequent exposure could have occurred during a test's window period. Therefore, monthly testing is treated as a potentially reassuring contextual detail rather than an independently quantified reduction to HIV prevalence. The calculation does not assume the partner was HIV negative, undetectable or on successful treatment. This is especially important because a negative HIV test from before a recent infection may not identify that infection immediately. Conversely, sex work itself does not mean the partner has HIV, and the model does not make such a claim. The final numerical estimate combines historical country-wide HIV estimates, a population-based treatment and viral-suppression proxy, your reported activities and condom use, and a transparent sex-work modelling assumption. These inputs carry substantial uncertainty even though the arithmetic can be calculated precisely. Your next practical step is a properly timed HIV test, not an attempt to infer individual infection status from occupation or second-hand descriptions of testing practice. [2, 6]
Step Three: Your Final Risk Estimate
Partner HIV estimate
The baseline is estimated total prevalence, including modelled undiagnosed HIV, not diagnosed prevalence. The national all-sex rate is used as a clearly labelled proxy because a compatible female-specific estimated-total-prevalence figure was not established.
Estimated treatment and viral-suppression adjustment
Estimated fraction diagnosed, treated and with low measured viral load: 0.6253 × 0.947 × 0.969 = 0.5738021679, or 57.38021679% of people living with HIV in this hybrid model. The complementary estimated fraction is 1 − 0.5738021679 = 0.4261978321, or 42.61978321%. This remaining fraction is the HIV risk multiplier applied below.
Key limitation: The three figures come from related but non-identical populations and denominators. KDCA's 96.9% endpoint is based on patients with registered RNA results and uses a 1,000-copy threshold, which is not the below-200 threshold for established U=U. The result is a transparent modelling proxy, not a verified national cascade or evidence of the individual partner's status. [2, 11]
Per-act transmission basis
For insertive vaginal sex: 0.0004 as a decimal, or 0.04%, approximately 1 in 2,500, conditional on a partner who has HIV with transmissible virus and before the modelled prevention adjustment. Receiving oral sex has a much lower and poorly quantified HIV transmission rate; it is not added as an invented independent risk figure. [1, 5]
Calculation Breakdown
| Factor | Value applied | Direction | Reason |
|---|---|---|---|
| National estimated total HIV prevalence | 0.000402016049 | Starting probability | 20,839 divided by 51,836,239, 2020 |
| Sex-worker adjustment | × 3.0 | Upward | Asia-Pacific heuristic, multiple partners not added again |
| Adjusted partner prevalence | 0.001206048147 | Intermediate probability | Approximately 0.1206048% |
| Insertive vaginal base per act | 0.0004 | Per-act baseline | CDC, four per 10,000 |
| Condom protection | × 0.20 | Downward | 80% reduction, consistent heterosexual condom-use evidence |
| No ejaculation, insertive role | × 1.00 | Neutral | No unsupported adjustment for an insertive participant |
| Partner injection-drug risk | × 1.00 | Neutral | Unknown, no affirmative history |
| Viral suppression, hybrid Korean estimate | × 0.4261978321 | Downward, modelled | 1 − (0.6253 × 0.947 × 0.969), not a matched official cascade |
| Wounds or inflammation | × 1.00 | Neutral | None reported |
| Circumcision | × 1.00 | Neutral | Adjustment excluded when a condom was used |
| PrEP | × 1.00 | Neutral | None taken |
| Final combined risk | 0.0000000411212 decimal | Model output | 0.00000411212% after matched display rounding |
Approximately 1 in 24 million for the condom-protected vaginal component, rounded to avoid false precision. The oral component is discussed separately. This is a scenario calculation, not a diagnosis or a clinically validated prediction.
Mathematical consistency
Unrounded computation: (20,839 ÷ 51,836,239) × 3.0 × 0.0004 × 0.20 × [1 − (0.6253 × 0.947 × 0.969)] = 0.00000004112120846. As a percentage this is 0.00000411212085%, with a reciprocal of approximately 24.32 million. The displayed decimal 0.0000000411212 and percentage 0.00000411212% agree at their displayed precision. The reader-facing ratio is rounded to approximately 1 in 24 million. All these figures are model outputs, not measured personal probabilities.
How Can I Put This in Perspective?
What the number means
This model places the selected condom-protected vaginal encounter in a very-low-probability category. It does not mean the encounter had zero risk or prove that your partner was HIV negative. Because the calculations combine an older projected national prevalence, an unvalidated partner-background multiplier and an estimated hybrid treatment cascade with mismatched populations, the many decimal places reflect arithmetic, not true individual-level precision.
Even a small modelled probability cannot replace a test. An HIV laboratory test taken after the relevant window period offers a much more reliable answer about your personal HIV status than further refinement of a population-level formula.
Comparison to Other STIs
Other sexually transmitted infections can spread during vaginal or oral sex and may be present without symptoms. A single numerical “times more likely than HIV” factor is not evidence-based without a matching pathogen, infected-partner prevalence, activity, protection and population. The per-infection proxies below therefore describe plausible exposure pathways, not fabricated comparative probabilities. [7, 8, 9]
| Infection | Relative likelihood versus HIV | Encounter-specific proxy | Appropriate follow-up |
|---|---|---|---|
| HIV | 1× reference | Very low modelled risk for condom-protected insertive vaginal exposure, oral HIV risk extremely low | Use the HIV testing schedule below |
| Syphilis | Not reliably quantifiable | Contact with an infectious sore, which may be outside the condom-covered area | Discuss a blood test with a sexual-health clinician |
| Gonorrhoea | Not reliably quantifiable | Can spread to the penis during vaginal sex or when receiving oral sex from an infected throat | Urine NAAT for genital infection; site-specific testing if indicated |
| Trichomoniasis | Not reliably quantifiable | Primarily a vaginal-to-penile exposure concern, less associated with oral sex | Discuss NAAT if symptoms or known exposure |
| Chlamydia | Not reliably quantifiable | Potential genital exposure during vaginal sex; oral transmission patterns are less well quantified | Urine NAAT for genital infection |
NAAT means nucleic acid amplification test. Condom use reduces exposure, but it does not eliminate all STI risk, particularly skin-contact infections. No comparative multiplier in this table should be interpreted as a measured single-encounter probability.
What Helped and What Hurt
Factors that reduced or limited risk
✓ Intact condom assumed. An 80% reduction estimate was applied to vaginal intercourse.
✓ No wounds or inflammation. No added tissue-injury multiplier.
✓ Oral HIV exposure is inherently low. Receiving oral sex was discussed separately, not inflated into a second numeric result.
✓ Population treatment estimate. A modelled 57.38% low-viral-load fraction was included, with important source and threshold limitations.
✓ Reported monthly screening. Useful context, but no numerical credit without documented results.
Factors that increased risk or uncertainty
✕ Unknown partner HIV status. A population starting probability was necessary.
✕ Sex work and multiple partners. Captured by one unvalidated threefold heuristic, not two multipliers.
✕ No PrEP. No additional protection credited.
✕ Country-data limitations. National rather than female-specific prevalence, and an imperfect 2020 treatment proxy rather than a matched population cascade.
FAQ Based on Your Risk Tier
Am I overreacting?
Feeling uncertain after an unfamiliar sexual encounter is understandable. The available evidence points to a very-low-risk protected vaginal exposure, but the statistical model cannot determine the partner's individual HIV status. Testing is the practical way to resolve the uncertainty.
Does this mean I have HIV?
No. A risk model is not an HIV diagnosis. Only appropriately timed diagnostic testing can establish your status.
What if my partner had HIV and was not virally suppressed?
Removing population prevalence and applying the CDC insertive-vaginal baseline with the modelled condom reduction gives 0.0004 × 0.20 = 0.00008 as a decimal, or 0.008%, approximately 1 in 12,500. This is a hypothetical average modelled risk conditional on a partner with transmissible HIV, not the estimated risk of your actual encounter. Confirmed sustained viral suppression prevents sexual HIV transmission. [1]
How is the viral-suppression risk multiplier calculated?
For a partner with unknown HIV status, this report first estimates the probability of HIV infection and then asks, among people living with HIV in South Korea, what fraction may be diagnosed, receiving HIV treatment and have a low measured viral load? Effective HIV treatment can reduce transmission to zero when viral suppression below 200 copies/mL is maintained, commonly known as U=U. A partner being diagnosed, taking ART, or participating in a monitoring programme does not on its own establish this. The historic Korean data below are combined to make a transparent estimated adjustment, not to claim a measured individual probability. [2, 11, 1]
| Hybrid cascade input or derived result | Value | Definition and scope |
|---|---|---|
| Estimated diagnosed among all people living with HIV | 0.6253, 62.53% | 100% minus 37.47% modelled undiagnosed, South Korea projected 2020, all sexes |
| ART among reported living HIV cases | 0.947, 94.7% | KDCA 2020 national treatment rate, reported cases, not all estimated infections |
| Low viral load among monitored people with an RNA result | 0.969, 96.9% | KDCA 2020 counselling programme, HIV RNA below 1,000 copies/mL, not necessarily the same denominator as all ART recipients |
| Derived diagnosed and on ART | 0.6253 × 0.947 = 0.5921591 | 59.21591% of all estimated people living with HIV, modelled |
| Derived low viral-load fraction | 0.6253 × 0.947 × 0.969 = 0.5738021679 | 57.38021679% of all estimated people living with HIV, modelled |
| Complement and final model multiplier | 1 − 0.5738021679 = 0.4261978321 | 42.61978321% modelled complement, applied to the per-act risk |
| Method and population scope | Historical hybrid cascade, national all-sex proxy | Not an observed 2026 national or adult-female suppression percentage |
The model's 1,000-copy outcome is a surveillance proxy. It is not equivalent to confirmed sustained viral suppression below 200 copies/mL, and the multiplication implicitly assumes the programme's suppression percentage extends to treated cases outside its monitored sample. Those assumptions may understate or overstate actual transmissible infection prevalence. The partner's individual diagnosis, ART use and viral load remain unknown. [2, 11]
What We Recommend
Based on your reported exposure date, 7 October 2026, testing can be timed as follows. These dates refer to tests performed without antiretroviral medications that might alter detection. A clinician may choose another schedule based on symptoms or other exposures. [6]
HIV RNA / nucleic acid test
Date: 17 October 2026, 10 days after exposure, the earliest point in the CDC's usual 10 to 33-day NAT detection window. An early negative result does not conclusively exclude infection. NAT is typically a venous-blood test used when very recent infection is suspected. A clinician can advise whether it is indicated.
Fourth-generation lab blood test
Date: 21 November 2026, 45 days after exposure, the end of the CDC's usual 18 to 45-day window for a laboratory antigen/antibody test using venous blood. Earlier testing can sometimes detect infection, but a negative result before the window closes may require repeat testing.
OraQuick or antibody-only test
Date: 5 January 2027, 90 days after exposure. The CDC's usual antibody-test window is 23 to 90 days. Oral self-tests and finger-prick antibody tests can have longer windows than fourth-generation laboratory blood tests.
Other STI testing
Consider a sexual-health appointment for urine gonorrhoea and chlamydia NAAT, a syphilis blood test and additional investigations according to local guidance. Receiving oral sex may create exposure to gonorrhoea from a partner's throat. The clinician should choose which anatomical sites need testing. [7, 8, 9]
PEP timing information
PEP, or post-exposure prophylaxis, is a 28-day course of antiretroviral medication that may be prescribed by a healthcare professional after a potential HIV exposure. Official guidance is to start as soon as possible and within 72 hours. The reported encounter on 7 October 2026 falls outside the standard 72-hour starting window as of this report's generation on 23 October 2026. A clinician, not this report, decides whether PEP is appropriate for an individual exposure. Speak with a sexual-health professional about your questions and use the testing schedule to confirm your status. [10]
Real Talk, A Note from the Creator
I know what it is like to be in an HIV panic spiral. I have experienced a couple throughout my life and each time it derailed me for months. I used to spend hours on Reddit and Google searching for anything to reassure me, but the information was vague, conflicting, and never truly calmed me down. That is why I built this tool, to cut through the noise, show you the mathematical risk, and give you facts grounded in real world data. My goal is simple: help you avoid the hell I went through so you can feel less anxious about your situation. You should not have to dig through one hundred sites to feel sane again. This report does that work for you. If guilt is the thing that drove you to buy this report, do not punish yourself. People make mistakes. You are human. Thank you for supporting this product. I wish you peace, good health, and clarity. No matter what happens, you are going to be okay. It is 2026, not 1980.
Your Path Forward: From Anxiety to Agency
1. Clarity
You know which encounter details and assumptions affected the estimate, and why the result is not a diagnosis.
2. A plan
Choose the most appropriate HIV test with a clinician. The 45-day fourth-generation laboratory date is 21 November 2026.
3. Knowledge
Effective condoms, properly timed testing, PrEP when appropriate and sustained viral suppression are established prevention and care tools.
The aim is not to spend more time calculating decimals. It is to take a sensible next step and let reliable testing resolve uncertainty.
Data Provenance and Model Limitations
The table lists every country or effectiveness input that influenced the numeric result. “Estimated total prevalence” includes modelled undiagnosed infections; it is not the number of reported diagnoses. Access date for the listed sources: 23 October 2026.
| Metric used | Value | Classification or cascade stage | Population and geography | Numerator and denominator | Data year | Publication year | Publisher and title | Direct URL | Limitations and fallbacks |
|---|---|---|---|---|---|---|---|---|---|
| Estimated PLHIV | 20,839 | Estimated total HIV prevalence count, including undiagnosed | All sexes and ages, South Korea | Modelled people living with HIV, no population denominator until combined with the next row | 2020 projected | 2020 | Lee et al., Journal of Korean Medical Science, Estimation of the Number of HIV Infections and Time to Diagnosis in the Korea | doi.org/10.3346/jkms.2020.35.e41 | Projection from claims data mainly through 2015; sex-specific prevalence not available from selected count; patient population not identical to residents |
| Population denominator | 51,836,239 | Total population | All residents, South Korea | 2020 population count, Koreans and foreign nationals; denominator for national estimated total prevalence | 2020 | 2023 cited data series | National statistical indicators, Total Population and Population Growth Rate and Natural Growth Rate | index.go.kr, 2020 population series | All-age and all-sex denominator, not adult female denominator |
| Partner-background adjustment | 3.0× | Version 6.8 heuristic, not empirical country statistic | South Korean female sex worker, proxy from Asia-Pacific region | Not a measured numerator/denominator ratio | Not applicable | Not applicable | HIV Risk Report Version 6.8 internal model rule | Internal model rule, not an external source | Unvalidated for South Korean sex workers; multiple partners not multiplied again |
| Undiagnosed among estimated total PLHIV | 37.47% undiagnosed; 62.53% diagnosed | Estimated modelled diagnosis fraction, cascade stage 1 | All sexes, South Korea | Modelled 7,809.11 undiagnosed / 20,838.90 total HIV infections; 13,029.79 diagnosed / 20,838.90 | 2020 projected | 2020 | Lee et al., Journal of Korean Medical Science, Estimation of the Number of HIV Infections and Time to Diagnosis in the Korea | PMC7025911, table 3 | Projection from historical inputs, differs from contemporaneous reported diagnosed case counts; not a direct 2026 observation or a woman-specific rate |
| ART treatment rate among reported living cases | 94.7%, 0.947 | Hybrid cascade stage 2, KDCA national treatment coverage proxy | Living diagnosed HIV cases, Korean nationals in KDCA reports | 13,766 individuals with recorded ART prescription / 14,538 reported living HIV cases, 2020 | 2020 | Not stated; displayed figures dated as of 20 Dec 2022 | Korea Disease Control and Prevention Agency, Current status of AIDS in Korea, HIV/AIDS Management Indicators | kdca.go.kr, HIV/AIDS management indicators | Denominator excludes some anonymous reports, while numerator may include them; Korean nationals focus; not a conditionally compatible follow-on from modelled diagnosis estimate |
| Viral suppression in monitored programme | 96.9%, 0.969 | Hybrid cascade stage 3, low viral-load proxy | People registered in KDCA Medical Institution Infected Person Counseling Project with a recorded RNA result | Last registered HIV RNA under 1,000 copies/mL / programme participants with at least one RNA result during 2020; underlying counts not published in displayed table | 2020 | Not stated; displayed figures dated as of 20 Dec 2022 | Korea Disease Control and Prevention Agency, Current status of AIDS in Korea, HIV/AIDS Management Indicators | kdca.go.kr, HIV/AIDS management indicators | Subset with viral-load tests, not demonstrably all ART recipients; below-1,000 threshold differs from U=U below-200; not a compatible conditional cascade stage |
| Derived estimated suppression and complement | 57.38021679% low viral-load fraction, 42.61978321% complementary fraction, x0.4261978321 | Historical hybrid population proxy, not measured national suppression | National all-sex proxy applied to female partner with unknown HIV/ART status | 0.6253 × 0.947 × 0.969 = 0.5738021679; 1 − 0.5738021679 = 0.4261978321 | 2020 inputs | Derived for report, 2026 | Calculated from Lee et al. (2020) and KDCA HIV/AIDS Management Indicators | Model study; KDCA data | Stages do not have matched denominators; using monitored viral-load rate for all treated cases is an unverified extrapolation; not established U=U coverage; wide unquantified model uncertainty |
| Insertive vaginal HIV risk | 0.0004 | Per-act conditional transmission baseline | Insertive penile-vaginal sex, HIV-serodifferent partnerships in pooled research | Four transmissions per 10,000 exposures to HIV with transmissible virus | Research pooled through 2014 | 2024 CDC page | CDC, HIV Risk and Prevention Estimates | cdc.gov/hivpartners/php/riskandprevention/index.html | Across-population summary, not individual outcome or South Korea-specific |
| Condom protection | 80% reduction, 0.20× remaining | Consistent condom-use estimate, heterosexual | Heterosexual men and women, not South Korea-specific | Risk among those reporting always versus never using condoms, observational person-year measures | Studies reviewed in 2002 | 2024 CDC page | CDC, HIV Risk and Prevention Estimates; Weller 2002 review | cdc.gov/hivpartners/php/riskandprevention/index.html | Not a direct single-act efficacy estimate, not separately stratified by insertive role; relies on self-report; occasional condom failure is possible |
Interpretation limits
The selected prevalence numerator and population denominator both refer to 2020, but the former is a projection and the latter a population statistic; they are not independent observed same-day counts. National all-sex prevalence is substituted for female prevalence, and neither country-wide prevalence nor the threefold sex-work heuristic tells us the specific partner's HIV status. The treatment adjustment is a hybrid estimate from a projected diagnosis proportion, reported ART treatment coverage and a viral-load measure from a monitored subset; it is not an observed complete national cascade. The treatment sources have different denominators despite all referring to 2020, and the less-than-1,000-copy measure is not an established U=U endpoint. More recent diagnosed case counts cannot replace estimated total HIV prevalence because they omit undiagnosed infection. The model has no separately validated combined oral-plus-vaginal transmission probability and does not compute one. These limitations mean the final point estimate should not be interpreted as a precise personal probability.
The condom percentage derives from observational evidence of consistent self-reported use, not proof of 80% protection for one act. A condom may slip, break or be used incorrectly even though the form provided no such report. PrEP, ejaculation and circumcision adjustments are applied only as allowed by the model rules. Testing, not a calculated probability, determines infection status.
Sources and References
- CDC. HIV Risk and Prevention Estimates. 5 November 2024. https://www.cdc.gov/hivpartners/php/riskandprevention/index.html.
- Lee E, Kim J, Lee JY, Bang JH. Estimation of the Number of HIV Infections and Time to Diagnosis in the Korea. Journal of Korean Medical Science, 2020, data/model through 2020. https://doi.org/10.3346/jkms.2020.35.e41.
- Republic of Korea national statistical indicators. Total Population and Population Growth Rate and Natural Growth Rate. 2020 data, 2023 projection series cited. Official table.
- Patel P et al. Estimating per-act HIV transmission risk: a systematic review. AIDS, 2014, underlying source described in CDC's 2024 estimates. PubMed search.
- Cresswell F et al. UK guideline for the use of HIV post-exposure prophylaxis 2021. HIV Medicine, 2022. https://doi.org/10.1111/hiv.13208.
- CDC. Getting Tested for HIV. Updated 15 September 2026. https://www.cdc.gov/hiv/testing/index.html.
- CDC. About STI Risk and Oral Sex. 6 February 2024. https://www.cdc.gov/sti/about/about-sti-risk-and-oral-sex.html.
- CDC. Getting Tested for STIs. https://www.cdc.gov/sti/testing/.
- CDC. About Trichomoniasis. 2024. https://www.cdc.gov/trichomoniasis/about/index.html.
- CDC. Clinical Guidance for PEP. 7 October 2026. https://www.cdc.gov/hivnexus/hcp/pep/index.html.
- Korea Disease Control and Prevention Agency. Current status of AIDS in Korea: HIV/AIDS Management Indicators. 2020 surveillance data displayed on a page with a 20 December 2022 reference date; webpage publication year not stated. https://www.kdca.go.kr/eng/4459/subview.do.
Sources accessed 23 October 2026. The internal sex-worker multiplier and the derived hybrid treatment cascade are labelled model assumptions, not established personal clinical probabilities.