|
Cost-effectiveness of HIV counseling and testing in US prisons |
Varghese B, Peterman T A |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology Human immunodeficiency virus (HIV) counselling and testing (CT) for prison inmates, at or near the time of their release from United States (US) prisons, was compared with no CT.
Economic study type Cost-effectiveness analysis.
Study population The study population was described as 10,000 prison inmates at or near their time of release from US prisons.
Setting The setting was the community. The economic study was conducted in prisons in the USA.
Dates to which data relate The effectiveness data were obtained from studies published between 1988 and 1999. The cost data were obtained from studies published between 1993 and 2000. The price year used was 1999.
Source of effectiveness data The effectiveness data were derived from a review of completed studies.
Modelling A decision analytic model was used to estimate the costs and benefits of HIV CT in comparison with no HIV CT.
Outcomes assessed in the review The outcomes used as input parameters for the model were the probabilities for the following:
HIV prevalence among otherwise untested inmates;
the acceptance of voluntary CT;
partners of HIV-infected persons being infected;
the risk for HIV transmission from infected individuals;
the risk of HIV acquisition for uninfected individuals;
the reduction of risk after counselling for those infected and uninfected.
Study designs and other criteria for inclusion in the review The effectiveness data were taken from the published literature, which included randomised controlled trials, longitudinal studies and economic evaluations. The authors did not state the inclusion or exclusion criteria.
Sources searched to identify primary studies Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Number of primary studies included Fourteen primary studies were included in the review.
Methods of combining primary studies The authors did not explicitly state the method used to combine the primary studies.
Investigation of differences between primary studies Results of the review The following input parameters were used in the model.
The probability of HIV prevalence among otherwise untested inmates was 1% (range: 0.1 - 15).
The probability of inmates accepting voluntary CT was 50% (range: 20 - 90).
The probability of partners of HIV-infected persons being infected was 20% (range: 15 - 40).
The risk for HIV transmission from infected individuals was 7% (range: 5 - 30).
The risk of HIV acquisition for uninfected individuals was 0.3%.
The reduction of risk after counselling for those infected was 25% (range: 10 - 50).
The reduction of risk after counselling for those uninfected was 20% (range: 10 - 50).
Measure of benefits used in the economic analysis The measure of benefit used in the economic analysis was the number of future HIV infections prevented. The estimates were calculated using information on the risk of HIV transmission among heterosexual couples, combined with the estimates of the effectiveness of counselling on risk reduction. A value of 1 was assigned for HIV transmission and 0 assigned for no HIV transmission. Therefore, the expected value obtained from the analysis gives the number of HIV infections that would occur with and without CT services. The difference between the two is the number of infections that can be prevented by the CT intervention.
Direct costs The cost and the quantities were not reported separately. The direct costs for providing the HIV CT service in US prisons were reported. The costs were reported as the additional cost required for the provision of HIV CT through an existing programme, which already offered serologic CT in prisons. No fixed costs were reported. The authors stated that the cost estimates for CT services in prison were unavailable in the literature. The estimates were therefore collected from HIV/STD (sexually transmitted disease) clinics at Michigan Department of Community Health and from appropriate literature.
For infected inmates, the costs of CT included the wages and time costs for the staff (administrators, counsellors, phlebotomists and laboratory staff) and the cost of testing (serum collection kits, enzyme immunoassay, Western blot tests and controls). The provider costs of CT were reported as $78.17 (range: 78.17 - 98.17) for HIV-infected inmates and $24.63 (range: 24.63 - 34.63) for uninfected inmates. The lifetime treatment cost of HIV was estimated using previous studies that adopted a 3% discount rate, and was $186,900 (range: 107,000 - 267,000). It was not reported how the lifetime treatment cost of HIV was estimated in these previous studies. All costs were expressed in 1999 US dollars.
Statistical analysis of costs No statistical analysis of the costs was reported.
Indirect Costs The indirect costs of the HIV CT were not included in the analysis. The authors justified the omission of the costs of patient time or productivity losses on the grounds that the study population consisted of prison inmates.
Currency US dollars ($). No currency conversions were reported.
Sensitivity analysis Sensitivity and threshold analyses were conducted in order to test the robustness of the parameters used in the model. The effect of possible additional costs of CT services in prisons, such as training costs for counsellors and guards and travel to the health clinic, was also considered. This was investigated by adding $20 per infected person, and $10 per uninfected person, to the provider cost of CT in the analysis.
Estimated benefits used in the economic analysis The baseline analysis reported that offering CT to 10,000 inmates, compared with not offering CT, would detect 50 new or previously undiagnosed infections and avert 4 future cases.
Offering no CT services to 10,000 inmates would result in 35 future cases of HIV.
Cost results The baseline analysis reported that offering CT to 10,000 inmates cost $125,000 to prison systems. This is the equivalent of $12.50 per inmate offered CT or $25 per inmate tested and counselled. The saving to society of offering CT to 10,000 inmates was over $500,000.
Offering no CT services to 10,000 inmates would cost $6.6 million in medical treatment costs alone.
Synthesis of costs and benefits From a societal perspective, offering CT was found to be more effective and less expensive than not offering CT. Offering CT to 10,000 inmates, compared with not offering CT, would detect 50 new or previously undiagnosed infections and avert 4 future cases of HIV. This would be achieved at a cost of $125,500 to prisons systems, with a societal saving of over $500,000.
From a societal perspective, offering no CT services would result in 35 future cases of HIV and would cost society $6.6 million in medical treatment costs alone.
The mean provider cost to prevent a future case of HIV was $34,000.
The one-way sensitivity analysis showed that an increase in HIV prevalence, risk transmission, or the effectiveness of counselling increased the magnitude of societal savings. HIV CT in US prisons remains less expensive than no HIV CT for society provided:
HIV CT prevents 1.45 cases of HIV;
the lifetime treatment costs of HIV infection exceed $40,000; and
the risk of HIV transmission from infected to uninfected individuals is greater than 1% per year, and the risk of acquiring HIV for the uninfected is greater than 0.05% per year.
Authors' conclusions Voluntary human immunodeficiency virus (HIV) counselling and testing (CT) of prison inmates could be a cost-saving HIV prevention programme, which might prevent future cases of HIV and save societal dollars.
CRD COMMENTARY - Selection of comparators No explicit justification was given for the comparator used, although it would appear to represent current practice in the authors' setting. The authors reported that correctional facilities in 17 states have mandatory testing and that the rest offer some form of voluntary or "on request" HIV testing, which was the intervention considered in this evaluation. You should decide if the intervention and the comparator represent current practice in your own setting.
Validity of estimate of measure of effectiveness The authors did not state explicitly that a systematic review of the literature had been undertaken. It was unclear if the review had been conducted in a systematic way to identify relevant research and minimise bias. Further, it was unclear whether the authors used the data from the available studies selectively. The impact of differences between the primary studies was not considered when estimating the effectiveness. Some of the effectiveness estimates taken from the literature and used in the analysis related to the general population, rather than the prison population. This may have affected the estimation of effectiveness, particularly as ethnic minorities and intravenous drug users are over represented in US correctional facilities.
Validity of estimate of measure of benefit The estimation of benefits was modelled using the effectiveness estimates taken from the literature. The instrument used to derive a measure of health benefit, the decision analytic model, was appropriate.
Validity of estimate of costs The quantities and the costs were not reported separately. The indirect costs were not included, even though the authors reported that the costs were estimated from a societal perspective. The authors stated explicitly that the costs of patient time and productivity were omitted from the analysis given that the study population comprised prison inmates. You should decide if such omissions are likely to affect the authors' conclusion when applied to your own practice setting. The study population was actually defined as "soon-to-be-released" prison inmates and, therefore, loss of time and productivity may have an impact on the future lifetime costs associated with HIV infection and treatment. These need to be evaluated in a formal study.
A sensitivity analysis of the costs was conducted using ranges that appear to have been appropriate. Appropriate currency conversions were not performed. Discounting at a rate of 3% was used when quantifying the lifetime treatment costs for HIV infection. The authors did not explore the impact of the discount rate used on the study's findings in the sensitivity analysis. This may limit the generalisability to other health care settings that use a different discount rate, such as the UK. The authors reported that the cost estimates for the prison setting were unavailable in the literature. The model used the costs for providing CT services in a facility that already offers serologic testing and counselling. It is unclear if the costs from this facility represented unit prices or charges. If charges were used to estimate the total cost then this would further limit the generalisability of the model's results to other health care settings.
Other issues The authors did not make appropriate comparisons of their findings with those from other studies. The sensitivity analysis only partly addressed the issue of generalisability, since it omitted to explore the impact of the discount rate and the possible use of charges rather than unit prices. The authors do not appear to have presented their results selectively. The study enrolled soon-to-be-released prison inmates and this was reflected in the authors' conclusions. The authors reported further limitations to their study. For instance, as with all studies, the results should be considered within the context of the probabilities and information used in the analysis. In particular, the lack of information on the effectiveness of counselling and the cost estimates for prison populations may underestimate the benefits. Further, the use of risk of HIV infection for 1 to 2 years may miss second- or third-generation transmission of HIV and thus underestimate the societal cost-saving. Finally, the authors reported that the model used was a prevention model, which did not estimate the benefits and costs associated with treating HIV-infected persons who were identified by prison CT.
Implications of the study The authors suggest that the average provider cost of HIV CT to prevent a future case of HIV ($34,000) should be regarded as a reasonable price to pay given the high treatment costs of HIV. The authors add that it may be difficult for providers to accept the cost of the HIV CT intervention in prisons and suggest alternative means of funding.
Bibliographic details Varghese B, Peterman T A. Cost-effectiveness of HIV counseling and testing in US prisons. Journal of Urban Health 2001; 78(2): 304-312 Indexing Status Subject indexing assigned by NLM MeSH AIDS Serodiagnosis /economics; Cost Savings /statistics & Cost of Illness; Cost-Benefit Analysis /statistics & Counseling /economics; Decision Trees; HIV Infections /economics /epidemiology /prevention & HIV Seroprevalence; Health Care Costs; Humans; Michigan /epidemiology; Preventive Health Services /economics; Prisoners; Prisons /economics; Risk Assessment /economics; United States; control; numerical data; numerical data AccessionNumber 22001001297 Date bibliographic record published 30/11/2002 Date abstract record published 30/11/2002 |
|
|
|