Correcting reported AIDS incidence: a statistical approach

This study is motivated by the time lag between date of diagnosis of AIDS cases and date of reporting, which results in incomplete data about the epidemic. A maximum likelihood procedure has been developed to adjust the actual numbers of diagnosed AIDS cases for reporting delay. If a parametric function for describing past and future incidence is assumed, its parameters and the adjustment for reporting delay can be estimated simultaneously. Data from the WHO Collaborating Centre on AIDS, Paris, are used. Practical problems related to data collection are dealt with.     

Changing pattern of delays in reporting AIDS diagnoses in Australia

To accurately monitor and predict the progress of the HIV/AIDS epidemic, it is important to adjust reported AIDS counts for reporting delays. This requires estimation of the reporting delay distribution. This paper aims to use a statistical model to identify the main factors influencing reporting delays in Australia and to adjust reported incidence data for these delays among cases of AIDS diagnosed from 1993 and reported before 30 June 1997. Reporting delays were found to vary significantly across states/territories. The influence of calendar time of diagnosis was also significant, with an overall trend toward longer delays over time. AIDS cases diagnosed in the fourth quarter of a year were reported significantly more quickly than those diagnosed in the first or third quarters. No significant differences were found due to sex, age and HIV exposure category, except people with haemophilia, in whom AIDS cases appeared to be reported more slowly. After adjusting for under-reporting and reporting delay, we found that the AIDS incidence in Australia was declining from about 1000 cases per year in 1994 to about 760 cases per year in 1996.     

Prognosis concerning HIV-infection and AIDS epidemic in The Netherlands based on mathematical analysis

In the Netherlands by the 1st of January 1990 1074 AIDS patients have been reported to the Department of the Chief Medical Officer. In the last few years the proportion of intravenous drug users increased and the proportion of homo/bisexual men decreased. After adjustment for the effect of delay in reporting the total number of AIDS patients by 1st January 1990 is estimated to be 1173. It appears that the reporting delay outside Amsterdam is longer than in this city. The time required for doubling of the half-yearly incidence of new AIDS patients (doubling time, dt) increased from 9 months in the beginning of the epidemic to 34 months. It is expected on the assumption of constant dt that 1120 new AIDS patients will be diagnosed in 1990 and 1991 together. The present growth among the homo/bisexual men (dt 34 months) is smaller than the one among the intravenous drug users (dt 23 months). The growth in Amsterdam (dt 36 months) is less than that in the rest of the Netherlands (dt 32 months). Based on the course of the AIDS epidemic the number of HIV infected (including the AIDS patients) is estimated as 9,000-12,000 by the 1st of January 1990.     

Impact of antiretroviral therapy on the magnitude of the HIV/AIDS epidemic in Brazil: various scenarios

We applied the back-calculation method to estimate the magnitude of the HIV epidemic in Brazil, using the EM and EMS algorithms. Under certain assumptions regarding the behavior of infected patients towards combined antiretroviral therapy, we discuss five different scenarios applied to the Brazilian epidemic. Our objective was to illustrate the impact of combined antiretroviral treatment on the incubation period and thus on estimates of the size of the HIV-infected population, based on reported AIDS cases.     

Trends in HIV and AIDS based on HIV/AIDS surveillance data in Japan

BACKGROUND: In recent years a decline in the number of new AIDS cases has been observed in several industrialized countries. It is important to know whether these recent trends observed in North America and Europe are also occurring in Japan. METHODS: The number of people reported with HIV and AIDS by nationality, route of infection, and sex was calculated based on the HIV/AIDS surveillance data available in Japan through December 1997. The effect of reporting delay, which was defined as those HIV and AIDS cases reported in the calendar year following diagnosis, on the trends was examined. The coverage rate in reporting HIV cases was estimated as the ratio of the reported AIDS cases with prior report as an HIV-positive to the total number of reported AIDS cases. RESULTS: The cumulative number of reported cases of HIV among Japanese and non-Japanese residents of Japan up to the end of 1997 were 1,300 and 1,190, respectively. The cumulative number of reported cases of AIDS among Japanese and non-Japanese up to the end of 1997 were 758 and 298, respectively. The number of reported cases of HIV among Japanese was found to be still increasing, with the major contribution from male cases. The increasing trend in the number of reported AIDS cases among Japanese began to slow in 1996 and 1997. The number of reported cases of HIV among non-Japanese residents of Japan peaked in 1992, and has decreased since then, and remained constant after 1994. In contrast, the number of reported AIDS cases among these non-Japanese tended to increase gradually. There was a slight reporting delay for people with HIV and AIDS. The estimated coverage rate in reporting HIV cases tended to decrease in 1996 and 1997 (1/7.2, 1/10.2, respectively). We point out several reasons for this recent decline and suggest the possibility of an ostensible decline in the estimates. CONCLUSIONS: We suggest that the number of people with HIV among Japanese has continued to increase, and that the increase in the number of AIDS cases among Japanese is now slowing.     

On the estimation of the reproduction number based on misreported epidemic data

Epidemic data often suffer from underreporting and delay in reporting. In this paper, we investigated the impact of delays and underreporting on estimates of reproduction number. We used a thinned version of the epidemic renewal equation to describe the epidemic process while accounting for the underlying reporting system. Assuming a constant reporting parameter, we used different delay patterns to represent the delay structure in our model. Instead of assuming a fixed delay distribution, we estimated the delay parameters while assuming a smooth function for the reproduction number over time. In order to estimate the parameters, we used a Bayesian semiparametric approach with penalized splines, allowing both flexibility and exact inference provided by MCMC. To show the performance of our method, we performed different simulation studies. We conducted sensitivity analyses to investigate the impact of misspecification of the delay pattern and the impact of assuming nonconstant reporting parameters on the estimates of the reproduction numbers. We showed that, whenever available, additional information about time‐dependent underreporting can be taken into account. As an application of our method, we analyzed confirmed daily A(H1N1) v2009 cases made publicly available by the World Health Organization for Mexico and the USA. Copyright © 2013 John Wiley & Sons, Ltd.     

Assessing the completeness of reporting of human immunodeficiency virus diagnoses in 2002-2003: capture-recapture methods

To determine the completeness of reporting of human immunodeficiency virus (HIV) diagnoses to state surveillance systems, the authors used capture-recapture methods. The numbers of cases diagnosed in the areas were estimated using HIV diagnoses reported to nine surveillance programs by different sources (e.g., laboratories, health-care providers). To account for dependencies between reporting sources, the authors used log-linear models to estimate the number of cases that had been diagnosed but were not identified by any reporting sources. Completeness of reporting (observed cases/expected cases) was determined for two time frames: cases diagnosed within a 1-year period (from October 1, 2002, to September 30, 2003, for most US states) reported up to 6 months after that diagnosis period and cases diagnosed within a 6-month period reported up to 12 months after that diagnosis period. A total of 11,266 HIV diagnoses were reported for the 1-year period with 21,589 report documents. Completeness of reporting of HIV diagnoses was 76% (95% confidence interval: 66, 83) when allowing 6 months of reporting delay (range: 72-95%) and improved to 81% (95% confidence interval: 72, 88) with 12 months' follow-up. When reporting systems retain all relevant documents, capture-recapture is a feasible approach for assessing completeness of reporting of HIV diagnoses. Completeness should be measured by allowing 12-months' reporting delay.     

HIV prevalence estimates--United States, 2006

Accurate and timely data on the number of persons in the United States living with human immunodeficiency virus (HIV) infection (HIV prevalence) are needed to guide planning for disease prevention, program evaluation, and resource allocation. However, overall HIV prevalence cannot be measured directly because a proportion of persons infected with HIV have neither been diagnosed nor reported to local surveillance programs. In addition, national HIV prevalence data are incomplete because local reporting systems for confidential, name-based HIV reporting have been fully implemented only since April 2008. With the advent of highly active antiretroviral therapies that delay the progression of HIV to acquired immunodeficiency syndrome (AIDS), and of AIDS to death, and changes in the AIDS case definition to include an immunologic diagnosis, earlier back-calculation methods from the 1990s for estimating HIV prevalence based on the number of reported AIDS cases are no longer reliable. With 80% of states reporting name-based HIV diagnoses as of January 2006, an extended back-calculation method now can be used to estimate HIV prevalence more accurately. Based on this method, CDC now estimates that 1.1 million adults and adolescents (prevalence rate: 447.8 per 100,000 population) were living with diagnosed or undiagnosed HIV infection in the United States at the end of 2006. The majority of those living with HIV were nonwhite (65.4%), and nearly half (48.1%) were men who have sex with men (MSM). The HIV prevalence rates for blacks (1,715.1 per 100,000) and Hispanics (585.3 per 100,000) were, respectively, 7.6 and 2.6 times the rate for whites (224.3 per 100,000).     

A new approach to estimating AIDS incubation times: results in homosexual infected men.

STUDY OBJECTIVE--The aim was to develop a new approach for estimating the incubation period of acquired immunodeficiency syndrome (AIDS), based on age distributions. DESIGN--Incubation period was expressed as the difference between age at time of diagnosis and age at time of contamination. Assuming independence between age at time of infection and incubation period, the age distribution of newly diagnosed AIDS cases is thus the convolution product between the distributions of the age of freshly infected patients and of the incubation times. AIDS incubation time can therefore be estimated from the age distribution of newly HIV infected subjects and newly diagnosed AIDS cases. SUBJECTS--Subjects were 2220 AIDS cases diagnosed until 1987, reported to the Ministry of Health, France, and 172 subjects discovered to be HIV-1 seropositive during a blood donation in Paris between August 1985 and July 1988. In both groups, the only known risk factor was homosexuality. MAIN RESULTS--The estimated median incubation time was 9.9 years (90% CI 9.0-10.9 years). Confidence intervals were narrow, even when taking into account the uncertainty in serodetection delay (90% CI 6.7-13.5 years). CONCLUSIONS--The incubation estimate is as accurate as previous estimates based on other models. This technique could therefore be applied to other risk groups.     

Reporting of newly diagnosed HIV infections in Catalonia, Barcelona, Spain. Implementation and results

OBJECTIVES: To describe newly diagnosed HIV infections from the HIV Reporting System in Catalonia (2001-2003), and to compare the characteristics of the epidemic based on the use of the HIV Reporting System and the Catalonian AIDS Registry versus those based on the Catalonian AIDS Registry alone. METHODS: Data were collected from newly diagnosed HIV infections and AIDS cases between 2001 and 2003 in Catalonia. RESULTS: Among the newly diagnosed HIV infections (1,765) the most frequent route of HIV infection was heterosexual transmission (46.8%), followed by men who had sex with men (26.7%), and injecting drug use (19.9%). Out of the 1,210 AIDS cases, the most common route of HIV transmission was injecting drug used (42.2%), followed by (heterosexual transmission 34.5%) and MSM (18.0%). Comparison of routes of HIV transmission in the two reporting systems (HIV/AIDS versus AIDS) revealed statistically significant differences. CONCLUSIONS: The HIV/AIDS Reporting System based on reporting of newly diagnosed HIV infections is feasible, since it has been useful in achieving the objectives of epidemiological HIV infection surveillance. It also provides more accurate information than does the AIDS Registry, which can be used to describe recent patterns of HIV transmission. The completeness of the new reporting system may be enhanced by including the diagnosis of HIV infection among the diseases of mandatory notification.     

Delays and continuation of hospital visits among HIV-infected persons and AIDS cases in Japan

This study attempts to clarify the distribution patterns of delay between HIV transmission and the first hospital visit among HIV-infected persons and AIDS cases in Japan except those infected through blood products. Such hospital visit patterns were analyzed, and the rates of reporting for HIV/AIDS surveillance among diagnosed HIV-infected persons and AIDS cases in hospitals were shown. From 1991 to 1997, a survey and subsequent follow-up were conducted among HIV-infected persons and AIDS cases diagnosed at 74 hospitals in Tokyo. The numbers of HIV-infected persons and AIDS cases were 590 and 208, respectively. The percentage of patients whose estimated date of HIV transmission was obtained ranged 23-41% among Japanese and non-Japanese HIV-infected persons and AIDS cases. Among these patients, 28% to 86% showed a 3-year delay between HIV transmission and their first hospital visit. The rate of HIV-infected persons who continued to visit hospitals within 1 year after their first visit was 77% for Japanese and 45% for non-Japanese; among those after 1 year or more following their first hospital visit the rate was more than 80% among Japanese and over 70% among non-Japanese. The rate of reporting to HIV/AIDS surveillance among diagnosed HIV-infected persons and AIDS cases was 90% or more after 1994 in Japan. The delay between HIV transmission and the first hospital visit was suggested to be very long. Not a few patients stopped visiting hospitals after only a short time. Most diagnosed HIV-infected persons and AIDS cases were reported to the surveillance system of Japan.     

Reporting efficiency during a measles outbreak in New York City, 1991.

OBJECTIVES. During an epidemic of measles among preschool children in New York City, an investigation was conducted in 12 city hospitals to estimate reporting efficiency of measles to the New York City Department of Health. METHODS. Measles cases were identified by review of hospital emergency room and infection control logs and health department surveillance records. The Chandra Sekar Deming method was used (1) to estimate the total number of measles cases in persons less than 19 years old who presented to the 12 hospitals from January through March 1991 and (2) to estimate reporting efficiency. Information on mechanisms for reporting measles cases was collected from hospital infection control coordinators. RESULTS. The Chandra Sekar Deming method estimated that 1487 persons with measles presented to the 12 hospitals during the study period. The overall reporting efficiency was 45% (range = 19% to 83%). All 12 hospitals had passive surveillance for measles; 2 also had an active component. These 2 hospitals had the first and third highest measles reporting efficiencies. CONCLUSIONS. The reporting efficiency of measles cases by New York City hospitals to the health department was low, indicating that the magnitude of the outbreak was substantially greater than suggested by the number of reported cases.     

Estimating AIDS incidence and jack-knife variance from a continuous delay distribution and incomplete data

Usual methods for estimating AIDS incidences are based on the inflation of a discrete reporting delay distribution, which often results in very imprecise estimates of the incidence in the most recent past. In this paper, we propose an alternative approach to estimate the AIDS incidence by inflating a continuous reporting delay distribution for each reported case. Covariate effects on reporting delays are evaluated by a proportional hazards model for the reverse time hazard function. A jack-knife variance for the estimated AIDS incidence is given. Study results showed that precision of estimates is improved by using the continuous time model as compared with those estimates given by its discrete counterpart. This feature is useful in assessing current trends in AIDS incidence.     

Epidemiological environment and response to the HIV epidemic in Bolivia]

With AIDS/HIV, early detection is of key importance to public health, as well as disseminating prevention information and providing timely and appropriate treatment. In Bolivia, at the end of 2006 approximately 50% had AIDS at the time of diagnosis, detection having occurred late in the illness. The HIV/AIDS epidemic in Bolivia is concentrated, with prevalence rates over 5% among the at-risk population, primarily men who have sex with men. From January 1984 through October 2006, the total number of HIV/AIDS cases reported in Bolivia rose to 2 190, with 1 239 HIV and 951 AIDS cases, and underreporting estimated to be over 70% country-wide. The United National Joint Program on AIDS (UNAIDS) estimated that by the end of 2006 there would be 6 700 people living with HIV/AIDS in Bolivia. In the context of this scenario, the article describes the challenges facing the HIV/AIDS program and the strategies developed to address the epidemic in Bolivia. In addition, the UNAIDS/PAHO strategies are stressed and must get underway for HIV/AIDS prevention and control activities in the country.     

江苏省AIDS疫情估计和预测分析研究

目的 利用估计和预测软件包(estimation and projection package,EPP)模型估计和预测江苏省AIDS疫情,为制定防治规划和进行干预活动提供准确的信息.方法 分析江苏全省各类重点人群AIDS、性病综合监测哨点(共80个,其中国家级监测点28个,省级监测点52个,2003-2009年共监测人群10 730 000名)、各类专题调查资料和实验室检测数据,运用EPP疫情模型分析疫情流行规律并预测未来的形势.结果 2003年,共监测人群830 000名,2004年共监测人群1 020 000名,2005-2009年累计监测8 880 000名.截至2009年底累计报告HIV感染者4103例,AIDS患者918例,累计死亡432例.注射吸毒(injection drug users,IDU)占的比例由2003年的66.84%(截至2003年底累计报告感染者392例)(262/392)下降至16 40%(142/868),性途径感染的比例由2003年的21.68%(85/392)上升至77.40%(672/868)[其中同性性传播占39.10%(339/868)、异性性传播占38.30%(333/868)].模型拟合显示2011年存活的感染者将达到14 290例,比2009年增加18.10%(2191/12 099),2011年总人群的感染率为0.02%.结论 EPP预测显示江苏省AIDS疫情呈上升趋势,AIDS疫情已进入快速增长期,应采取积极有效的防控措施,以控制进一步蔓延.     

Acquired immunodeficiency syndrome in Brazil and the State of Ceará

Since February 1998, a total of 129,000 cases of acquired immune deficiency syndrome have been reported in Brazil. The cumulative frequency of the disease is 82 per 100,000 which makes Brazil one of the countries moderately affected. There are considerable differences between regions in the frequency of cases, from 25 per 100,000 in the north and north east to 152 per 100,000 in the south east. Sexual intercourse is still the predominant means of transmission. Transmission in the early years of the epidemic was mostly between homosexuals and bisexuals, but transmission via heterosexual intercourse is increasing. The contribution made by intravenous drug use differs between the regions, and is particularly large in the mid-south region. A pilot project in the city of Fortaleza has shown that it is possible to successfully integrate the diagnosis of STD and AIDS in health care units at an intermediate level. This appears to be an appropriate strategy for the integration of STD treatment into primary health care in Ceará State. The non-uniform pattern of development of this epidemic must be taken into account in epidemiological analyses of AIDS in Brazil.     

Routine surveillance data on AIDS and HIV infections in the UK: a description of the data available and their use for short-term planning

In the UK surveillance of AIDS and HIV infection is based on routine reporting systems. Whilst attempts are made to ensure that AIDS data are as complete as possible, numbers of reports fluctuate from month to month for reasons which are described. In 1986 there was an increase in death certificates naming AIDS as a cause of death in patients who were not identifiable in the surveillance data. More active surveillance is now undertaken in order to minimize this and other possible discrepancies. It is probable that most cases of AIDS are reported and therefore these data can be used to describe trends in the epidemic by 'risk group'. Laboratory reports of HIV antibody-positive tests could give an earlier indication of trends because of the long incubation period of AIDS. But these laboratory data are difficult to interpret because they represent an incomplete and biased sample of all positive persons. AIDS cases are still being reported at a rate which is increasing approximately exponentially. Short-term predictions are presented showing a growth in the epidemic which is consistent with previously published predictions. Most cases are in the homosexual risk group. New asymptomatic homosexual patients with HIV antibody are still being identified. The epidemic of AIDS in haemophilia patients should be of finite size although new cases of AIDS are likely to continue to be diagnosed for several years. AIDS due to blood transfusion given in the UK before donor screening appears to be a much smaller epidemic. The epidemic in drug abusers is increasing. Heterosexually acquired AIDS and HIV infections are being reported in small but increasing numbers.     

Short-term predictions of HIV prevalence and AIDS incidence.

Reports of AIDS cases in Amsterdam up to February 1990 were used to make predictions of future cases up to 1993. Two published methods were applied, which make extrapolations from current cases and simultaneously estimate the extent of delay in reporting. The choice of the exact model greatly influenced the predictions, as did predictions for distinct transmission groups. We present results for the homo/bisexual male group, and the total population of Amsterdam. The AIDS case predictions are used to predict the HIV prevalence using the ratio of HIV prevalence to AIDS incidence and through 'back calculation'. We suggest that the ratio is a simple technique that may be used to estimate HIV prevalence. The estimated number of cumulative HIV infected homo/bisexual males in Amsterdam in January 1990 was between 2100 and 4100 in a total of 2200-4600 infected people.     

A change-point model for reporting delays under change of AIDS case definition

Accurate monitoring of disease incidence is of major public health concern. The time delay between diagnosis and the date of reporting creates bias in estimating disease incidence. Changes in case definition are expected to have an impact on the time lag of case reporting. We propose a change-point model for reporting delays in AIDS that takes into account recent changes in the AIDS definition in US and European countries. The model was applied to California AIDS surveillance data and the distribution of reporting delays before and after the recent change of definition in 1993 were analyzed in terms of contributing factors. The overall significance of the model with change-point as compared to the model without change-point indicates that the effect of the 1993 change in definition on the distribution of reporting delays was highly significant (p < 0.0001). Overall, reporting delay of cases initially diagnosed with AIDS-defining diseases before 1993 was shorter compared to after 1993; reporting delay of cases initially diagnosed meeting the 1993 immunologic case definition was shorter than of those initially diagnosed with AIDS-defining diseases. Region of residence, mode of exposure, race/ethnicity and time of diagnosis emerged as the main covariates in the models. The method introduced here applies to current and possible future changes of the AIDS case definition as well as changes in diagnostic criteria or case definition in diseases other than AIDS. We demonstrate that such changes may be accompanied by sizeable changes in the distribution of reporting delays, and thus adjustment for reporting delays must be recalibrated after a change in definition.