| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
According to the World Health Organization and Red Cross and Red Crescent societies, as many as two-thirds of developing countries have not mandated screening of donated blood products for the presence of HCV, making receiving transfusions risky in these areas.8,16 Furthermore, in many developing countries, HCV transmission also may be linked to blood exposure through use of improperly sterilized needles and syringes for administration of therapeutic injections.17–28
In India, where most health-care visits include one or more therapeutic injections,29,30 there is great potential for exposure to infected blood. Individuals seeking health care from providers with little or no medical training are at particularly high risk when injections are administered with reused, unsterilized needles and syringes.31,32
The city of Chennai (formerly Madras), the site of this study, is located in the southeastern state of Tamil Nadu and has a population of over 6 million. In a recent survey, over two-thirds of the general population reported receiving at least one injection in the past six months, averaging four injections per year.29 Individuals receiving health care in the context of poor adherence to universal blood precautions are at increased risk of not only HCV infection but also infection with HIV, hepatitis B, and blood-borne parasites.
This study aimed to determine whether use of different types of formal allopathic, informal allopathic, and traditional non-allopathic health-care providers was associated with increased HCV infection in adults living in slum communities in Chennai.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Each health camp was held in a community for 6–7 hours on a single day. Those who attended received physical examinations, laboratory testing, and prescription drugs, as needed. Those selected were informed of the risks and benefits of participation, advised of their rights as study participants, and given copies of the informed consent form to read and sign (or mark an "X"). All study methods, procedures, and consent forms were approved by the Johns Hopkins University Bloomberg School of Public Health and YR Gaitonde Centre for AIDS Research and Education (YRG-CARE) institutional review boards.
Those who agreed to participate were taken to a private setting in the health camp where same-sex, study-trained staff members interviewed them in the Tamil language regarding health-care use, blood exposures, and sexual risk practices. Participants were then given HIV pre-test counseling and were asked to provide blood and urine specimens for testing.
Biological specimens were tested in a dedicated laboratory in Chennai, and samples were tested for quality control at the Johns Hopkins School of Medicine central laboratory. Serum was tested for antibody to HCV using HCV Enzyme-Linked Immunosorbent Assay (ELISA) 4.0 (Abbott Murex Biotech S.A. Pty Ltd., Kyalami, South Africa). Samples that tested repeatedly positive on ELISA were deemed positive for analysis. Serum was tested for several STDs including HIV, using HIV ELISA 1.2.0 (Abbott Murex Biotech Limited, Kent, England), Genscreen HIV 
Version 2 ELISA (BioRad, Marnes LaCoquette, France) and Western blot (BioRad, Marnes LA Coquette); HSV-2, using Herpeselect 2 EIA (MRL, Focus Technologies, Los Angeles); and syphilis using the Treponema pallidum Passive Particle Agglutination Test (Serodia TP-PA Fujirebio Inc., Tokyo, Japan).
Data were transferred electronically and analyzed using SAS (SAS Institute, Cary, NC) and STATA (Stata Corp., College Station, TX) software. For the purpose of analysis, health posts/dispensaries, hospitals, and pharmacies were designated "formal allopathic health-care providers (formal)," traditional healers were designated "traditional non-allopathic health-care providers (traditional)," and other providers were designated "informal allopathic health-care providers (informal)." Parameters for all unadjusted, and adjusted odds ratios (ORs) were estimated using generalized estimating equations to adjust for the inherent correlation between individuals within the same slum communities. Unadjusted ORs were calculated using simple logistic regression. Variables were considered for inclusion into the multiple logistic regression model to estimate the adjusted odds ratio (AOR) of the main association based on the following criteria: variables at least marginally associated (P < 0.25) with the outcome, affecting the magnitude or significance of the association of the main exposure of interest with the outcome by more than 10%, measuring an effect not already measured by another variable in the model, or important purely by content or interest. For example, whether participants had tattoos and at least one measure of health-care use were included in final models regardless of statistical significance because these were exposures of interest. All possible effect modification was explored. Associations were deemed statistically significant if 95% CIs did not include unity.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Injection drug users. Participants reporting injection drug use in the prior three months (N = 7) were almost 28 times as likely to be HCV infected as those denying injection drug use (OR = 27.8; 95% CI: 7.25, 106.4). To focus on health-care use exposures, these participants were excluded from further analyses.
Characteristics of the study population. A slight majority of study participants were female (53%), most were married (71%), about a third (34%) had completed high school, and the average age was 28 (standard deviation <obrSD] = 6.6). More than half the study participants rated their health status as moderate to very bad (60%), and a third reported having a tattoo. Forty-two percent of men (but no women) reported drinking alcohol more than weekly, and eight percent of men reported having been circumcised. Most participants had used hospitals (57%), about a third had used a health post (34%), and a few had used traditional healers (6%) and other informal health-care providers (1%) in the past three months. (Data not shown.)
HCV antibodies were detected in 1.9% of men and 2.8% of women. Approximately 16% of women and 10% percent of men were HSV-2 infected; 5% of women and 17% of men reported having ever had a genital ulcer, and less than 10% of women and men had evidence of HIV infection or infection with any other STI. (Data not shown.) Because of effect modification noted between gender and multiple health-care use exposure variables, subsequent analyses of the associations of demographic and health-care use and HCV infection were stratified and displayed by gender.
Women.
Women reported low levels of parenteral risk behavior (Table 1
). Neither HCV infected (N = 24) nor HCV uninfected (N = 830) women reported using informal health-care providers. However, compared with HCV-uninfected women, HCV-infected women were somewhat less likely to have gone to a hospital for medical care in the prior six months (OR = 0.50, P = 0.16), although this association could have been seen purely chance.
|
When assessing the association of hospital use and HCV infection, after controlling for having had genital ulcers and tattoos, those who used a hospital were half as likely to be HCV-infected as those who did not use a hospital (AOR = 0.45, 95% CI: 0.16, 1.29). This association was not statistically significant.
Men.
As shown in Table 1
, although there was little difference in hospital use by HCV status, HCV-infected men (N = 15) visited informal health-care providers almost nine times as frequently in the prior six months as HCV-uninfected men (N = 3 752; OR = 8.76, P < 0.01).
HCV-infected and -uninfected men did not differ with respect to marital status or age; however, HCV-infected men were more likely than HCV-uninfected men to report moderate to very bad health status (compared with good and very good health status; OR = 4.10, P = 0.03). Furthermore, HCV-infected men were more likely to drink alcohol at least once a week (OR = 3.76, P < 0.01) than HCV-uninfected men. More HCV-infected men reported having had genital ulcers (OR = 1.92, P = 0.24, not statistically significant) and almost three times as many had HSV-2 (OR = 3.31, P = 0.04) relative to HCV-uninfected men. Moreover, HCV-infected men were over three times as likely as HCV-uninfected men to report ever having had sex with a man (OR = 3.66, P = 0.06).
Finally, when compared with HCV-uninfected men, HCV-infected men were more likely to have had a tattoo (OR = 2.01, P = 0.22), and more likely to have been circumcised (OR = 2.17, P = 0.27), although these associations were not statistically significant.
As shown in Table 2, after controlling for health status, frequency of alcohol use, HSV-2 infection, having had sex with a man, and having had a tattoo, men who used informal health-care providers were more than five times as likely to be HCV-infected compared with men not reporting using informal providers (AOR = 5.83, CI: 1.57, 21.6).
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Local staff members suggest that informal health-care providers were interpreted to be practitioners working from storefronts and market stands, also non-disparagingly known as "quacks." The term used on the questionnaire for these providers explicitly excluded trained allopathic providers (specifically: health posts/dispensaries, hospitals, and pharmacies) and non-allopathic providers (specifically: traditional healers). Anecdotal reports suggest that quacks are visited primarily to obtain inexpensive, expedient remedies for common ailments, and that these providers typically offer therapeutic and vitamin injections usually administered with reused or unsterilized equipment. Quacks are ubiquitous in southern India, as in many developing countries. If respondents interpreted informal providers to be quacks, it would not be surprising to find an association between using a quack and being HCV-infected.
As in some other developing countries, there is widespread use of injections in Chennai. In a recent rapid assessment of injection related-health risk there, 63% of respondents reported having had at least one injection, and a median of three, in the prior six months. In that study, vitamin, therapeutic, and antibiotic injections were given for a variety of ailments including asthma, fever, and "old age." During that assessment, approximately 20% of injections observed were administered with unsterilized needles and syringes, and only 8% of injection prescribers, 5% of injection providers, and 0.02% of the general population were aware that the hepatitis C virus could be transmitted through injections.29
Why are injections in such high demand? Anthropologic studies have given some insight into the attraction of injections. Reeler suggested that worldwide demand stems from perceptions that injections are fast-acting, highly effective, and more "western" than other therapies, although the author notes that rationales vary somewhat according to culture.17 According to Reeler, practitioners are increasingly compelled to provide injections to maintain credibility in their communities,18 and even those practicing traditional and ayurvedic medicine typically offer injectable medications.17 Furthermore, reporting on a study in nearby Vellore, India, Lakshman and Nichter offered insights into injection practices of registered medical practitioners (RMPs), providers who are registered by the government as having some health-care training. According to the authors, the training experience of these providers is minimal and, like informal health- care providers (as designated in this analysis), injections are typically administered by RMPs for non-specific ailments such as "weakness" and body pain, as well as for fever, headache, and cough.30
We could not detect an association between use of informal health-care providers and HCV infection in women because neither HCV-infected nor HCV-uninfected women reported using informal health-care providers. It was surprising to find no women reporting such use, and ethnographic research is needed to determine whether these women failed to report use, had reduced access to use, or did not choose to use informal providers.
Although the association was not statistically significant, we were surprised that HCV-infected women sought health care from a local hospital less often in the prior six months than HCV-uninfected women. This finding was contrary to results from a 1995 study in Sweden, where significantly more HCV-infected blood donors reported having been hospitalized compared with HCV-uninfected blood donors.33 It is possible that in a culture that holds women to very high standards of morality, riskier, HCV-infected women were reluctant to access conventional health care, including hospitals. However, more research is needed to assess the reliability of these results and describe these associations in social and cultural contexts.
Results should be interpreted while acknowledging study limitations. First, we examined the association of HCV and exposure to informal health-care providers or "quacks" instead of exploring the association of HCV infection and injection exposures directly. Although we asked about therapeutic injections in the interviews, we believe that the validity of responses to those questions was adversely affected by imperfect translations unwittingly linking licit to illicit injection, and by undesirable placement between questions on sex behavior and illicit injection drug use. In our study, less than one percent (0.06%) reported receiving any injection in the prior three months. These results differ starkly from those found in other studies, and particularly an assessment of injection risk in the general population of Chennai in which over half of the population reported having received a therapeutic injection in the prior six months.29 Though exposure to quacks may have been an imperfect surrogate marker for exposure to therapeutic injections, because of the absence of stigma attached to use of health care, we felt that the validity of responses to these questions was superior to that of questions about injection exposure. Moreover, the link between use of informal health-care providers and unsafe injections in the developing world has been established previously.17–29
Because all behavioral risk data relied on self-reported risk, associations of high-risk behavior and disease may be underestimated across risk behaviors. When answering questions about sensitive behaviors, participants may have given what they perceived to be socially desirable responses. The traditional structure of Indian society, which lends itself to close monitoring and scrutiny of the behavior of women, may lead women to have more of a perceived need to give socially desirable responses than men. While socially desirable responding may explain in part why we could not detect substantial high-risk behavior in women, we see no reason that this bias should be more pronounced in HCV-infected compared with HCV-uninfected male or female participants. As such, this bias should not have adversely affected estimates of associations.
Secondly, most studies confirm HCV test results with the recombinant immunoblot assay, but in this study, we used repeated positive results on EIA to confirm results. Though some misclassification of HCV test results may have resulted from adoption of this protocol, again, it is expected that this bias would occur without regard to exposure categories, and thus not affect measures of association of exposure and disease.
Data on exposures and outcome were collected concurrently, so there is no way to know if exposures preceded HCV infection. Longitudinal studies are needed to establish temporality and would help researchers draw causal associations. The associations found in this study should be further explored in cohort studies.
Finally, the results of this study may have limited generalizability to other cities. Because the participants of this study were randomly selected from the general population living in selected slums, conclusions and recommendations may be applicable only to other slum communities in Chennai, or other similar cities.
According to these data, the risk of HCV infection in southern India is associated in men with use of informal health-care providers. This excess risk may be associated with exposure to reused needles in these settings.
Public officials could use the results of this study to initiate development of clear: 1) regulations for screening of all blood products for HCV infection; 2) strategies for oversight of practices of health-care providers; and 3) plans for systematic disposal of contaminated medical waste.
Furthermore, there is a great need for health educators to disseminate prevention messages regarding blood safety. It is worthwhile noting that no practitioners giving injections who were queried in a study in Karachi, Pakistan, knew that HCV could be transmitted through therapeutic injection,20 and in a separate study in Chennai, less than a tenth of injection prescribers, injection providers, and the general population knew that HCV could be transmitted through injections.29 Data from these studies underscore the need to promote awareness of the risks of disease acquisition through contaminated needles among both health care providers and the public. Billboards and signs have been used widely to raise awareness about unprotected sex and HIV in Indian cities. Although there has been some effort to promote safe needle use to prevent HIV infection in India, advertisement campaigns should be expanded and should specifically encourage the purchase of a disposable needle for each health-care visit.
More research is needed to determine the extent to which contaminated needles contribute to HCV infection. Studies have repeatedly shown that most Indians with minimal education prefer injections to oral administration of medications. With proper long-term investment, social marketing may be used to encourage the prescription and use of topical and ingestible therapeutics rather than injectables. Reducing the incidence of blood-borne infections due to health-care-related exposures would be an important step in controlling the spread of HCV infection in developing countries such as India.
Received June 12, 2002. Accepted for publication November 4, 2002.
Acknowledgments: We thank the YRG-CARE staff, especially A.K. Ganesh, P. Balakrishnan, the interviewers who collected the data, and the participants in the study. We are also indebted to Dr. Vivian Go, Dr. Peggy Bentley, Clevetta Chandler, and Dr. Kenneth Mayer of the Johns Hopkins-based NIMH Chennai site for project support; Lisa Strader of Research Triangle Institute for logistical support; Paul Stamper of the Johns Hopkins School of Medicine, NIMH central laboratory for laboratory support; Dr. Steven Reynolds, Johns Hopkins School of Medicine, for STI data from Pune, India; Judith Bass, Dr. Patrick Tarwater, and Dr. Charles Rohde of the Johns Hopkins School of Public Health for statistical advice; and Dr. Susan Sherman for feedback on the manuscript.
Financial support: This study was funded with NIMH collaborative grant U10-MH61543 and a Fogarty New and Minority Investigator’s Award.
Authors’ addresses: Melissa A. Marx, Bureau of Communicable Disease, New York City Department of Health and Mental Hygiene, 125 Worth St., Room 318, New York, NY 10013. K.G. Murugavel, P. Balakrishnan, S. Anand, Suniti Solomon, YR Gaitonde Centre for AIDS Research and Education, 1 Raman St., T. Nagar Chennai, Tamilnadu, India. Sudha Sivaram, Mark Steinhoff, and David Celentano, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe St., Rm E6136, Baltimore, MD 21205. David L. Thomas, Johns Hopkins University School of Medicine, 424 N. Bond St., Baltimore, MD 21231.
Reprint requests: Melissa A. Marx, Bureau of Communicable Disease, New York City Department of Health and Mental Hygiene, 125 Worth St., Room 318, New York, NY 10013, Telephone: 212–788–4220, Fax: 212–788–5470, E-mail: mmarx{at}health.nyc.gov
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
