Abstract
Background
Despite the initial success of HIV/AIDS policy, an increasing number of patients are failing the first-line antiretroviral therapy (ART) each year and the failure rates are increasing. There is a need for identification of novel strategies to reduce failure rates. The aims of the study are (1) to design a novel strategy to reduce ART failure rates and (2) to create a stochastic model using Monte Carlo (MC) simulation comparing the novel strategy with existing strategy.
Methods
A novel strategy based on annual plasma viral load testing and resistance testing for HIV treatment at baseline and at the time of failure was designed. A cohort of 1000 patients each was created for the existing strategy and a novel strategy. Assumptions were included from Indian studies and own data. The two strategies were compared over 20 years of follow-up using stochastic modeling and MC simulation was done for death rates, failure rates, and cost-effectiveness analysis. SimVoi add-in software for MS Excel was used for simulations. Student's t-tests were performed for comparing continuous variables, and the cumulative rates for various outcomes were plotted using Kaplan–Meier analysis.
Results
The novel strategy resulted in lower mortality over a 20-year period (279.9 + 7.13 deaths vs 130.43 + 6.03 deaths) with incremental cost per life saved at Rs 32,925 per year. Incremental cost-effectiveness ratio cost per quality-adjusted life year was Rs 1.33 lakhs/annum at constant rate of discounting and just under Rs 90,000 per annum using differential discounting.
Conclusion
Armed Forces are likely to benefit by adopting the novel strategy that is cost-effective with a significant mortality benefit.
Keywords: Cost-effectiveness, Mortality reduction, Antiretroviral therapy, Resistance testing, Strategy comparison
Introduction
Provisioning of antiretroviral therapy (ART) has added more than 30–40 years of life to HIV patients.1 Improved survival has led to an increase in disease burden and prevalence of HIV. Although the mortality has reduced dramatically nationwide, the challenge of HIV drug resistance (HIVDR) is progressively increasing.2 The long-term effectiveness of antiretroviral drugs is dependent on strict adherence to the prescribed regimen because HIV resistance to these drugs can develop with subtherapeutic doses. Antiretroviral drugs need to be taken lifelong with more than 95% adherence.3, 4 Consequences of poor adherence include not only diminished outcome for the patient but also the public health threat of widespread transmission of drug-resistant virus.5
The role of various factors related to adherence and resistance including transmitted drug resistance (TDR), non-compliance–related resistance mutations (NCRMs), and spontaneous mutations (SMs) in contributing to drug failure is well documented.6 Despite this knowledge, failure rates are increasing in the country. There is a need for identification of novel strategies to reduce failure rates.
The Armed Forces Medical Services (AFMS) HIV policy was first formulated in May 2003 and is under 4th revision now. Despite the initial success, there have been an increasing number of patients who fail the first-line ART each year. There is a need for evaluation of the current strategy and identification of factors that can be corrected.
Therefore, the aim of this study was to design a novel strategy to reduce mortality rates and assess its cost-effectiveness from the societal perspective. This was carried out by creating a stochastic model using Monte Carlo (MC) simulation in comparison with the existing strategy.
Material and methods
The analytic overview
As per the existing strategy, ART is initiated irrespective of CD4 cell count of the individual (Fig. 1). The regimen comprises a combination of 2 nucleoside reverse transcriptase inhibitors (NRTIs) with one non–nucleoside reverse transcriptase inhibitor (NNRTI). These patients are followed up clinically and by six-monthly CD4 cell count, and second-line ART is initiated if clinical or immunological failure occurs.7 Both clinical and immunological failures in HIV are preceded by virological failure due to development of drug resistance, which is predominantly of four types (Table 1).
Fig. 1.
A pictorial depiction of the existing strategy. ART, antiretroviral therapy; NCRM, non-compliance–related resistance mutation; SM, spontaneous mutation; TDR, transmitted drug resistance.
Table 1.
Types of drug resistance.
|
|
|
ART, antiretroviral therapy; ARV, antiretroviral.
There are two fault lines in the existing strategy. First, it does not take into account the existence of TDR and only tackles it as and when the individual exhibits features of failure. Thus, patients harboring TDR get inappropriate therapy until they fail clinically/immunologically. Second, there are a considerable number of individuals (about 10%) with very poor adherence. These patients harbor wild-type virus despite clinical failure and would still respond well to first-line therapy. However, in absence of plasma viral load and HIVDR testing facilities, they get initiated on second-line regimen as per guidelines, the cost of which is significantly higher.
The novel strategy was designed to overcome these two limitations8 (Fig. 2). In the novel strategy, pretreatment HIVDR testing would be performed before ART initiation to exclude TDR. HIVDR testing would also be performed every time a patient presents with failure to correctly structure the regimen. In addition, yearly plasma viral load testing would be performed in all individuals as a preferred mode of monitoring the response to therapy.
Fig. 2.
A pictorial depiction of the novel strategy. ART, antiretroviral therapy; NCRM, non-compliance–related resistance mutation; SM, spontaneous mutation; TDR, transmitted drug resistance.
Defining the simulation cohorts
Two cohorts of 1000 HIV-positive patients were created for the two strategies. The following assumptions were made about the cohorts.
-
1.
HIV being mainly a disease of youth, the mean age at time of diagnosis was 30 years (range, 15–49 years)
-
2.
The rates of baseline TDR and adherence rates were taken from multiple Indian studies (Table 2).2, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 The death rate in India is about 3.5% (3–6%) per annum as per national program.22, 23 Armed Forces have better follow-up and lower death rates (∼1.5%), which was taken as the base rate for the current strategy.24 Data from western countries (United Nations Programme on HIV/AIDS {UNAIDS} 2018 report) that routinely follow the strategy of baseline drug resistance testing have a death rate ranging from 0.05% to 0.5%.25 For the novel strategy, we modestly assumed a death rate of ∼0.70%.
-
3.Defining the regimens
-
a.First-line regimen was defined as per national guidelines, that is, a regimen of two NRTIs and one NNRTI.22
-
b.Second-line regimen for NRTI/NNRTI failure included replacement of failed drug with various combinations of ritonavir-boosted protease inhibitor (atazanavir/lopinavir + ritonavir) with dolutegravir ± alternate NRTI analog, if sensitive.26
-
c.Third-line ART or salvage regimen was defined as use of darunavir/ritonavir with raltegravir.26
-
a.
-
4.The cost estimates were as follows:
-
a.First-line therapy = Rs 1000–2500 per month
-
b.Alternate first-line therapy/second-line therapy = Rs 2500–3000 per month26
-
c.Third-line/salvage regimen = Rs 6000–10000 per month
-
d.CD4 testing = Rs 500 per test
-
e.Plasma HIV viral load = Rs 5000 per test
-
f.Resistance testing = Rs 9000–15000 per test
-
a.
-
5.Protocol of testing
-
a.Current strategy: CD4 count at baseline and every 6 months
-
b.Novel strategy
-
i.CD4 count at baseline and every 6 months
-
ii.Plasma viral load monitoring at baseline and every year
-
iii.Resistance testing at baseline and whenever Plasma Viral Load (PVL) is > 1000 copies/ml and/or signs of immunological/clinical failure occur.
-
i.
-
a.
Table 2.
Assumptions incorporated into the stochastic model.
| S No | Parameters | Mean | Limits | Remarks |
|---|---|---|---|---|
| 1 | Transmitted drug resistance rates | NNRTI resistance is almost 10%. | ||
| A | TDR NRTI | 3% | 2%, 6% | |
| B | TDR NNRTI | 9% | 7%, 11% | |
| C | Total TDR | 12% | 9%, 15% | |
| 2 | Adherence | |||
| A | Good (>95%) | 65% | 50%, 75% | |
| B | Partial (60–95%) | 25% | 20%, 35% | |
| C | Poor (<60%) | 10% | 7%, 13% | |
| 3 | Annual failure rates | The virus in the subgroup with poor adherence is usually the sensitive strain. | ||
| A | Good adherence | 1.5% | 1%, 2% | |
| B | Partial adherence | 10% | 7%, 13% | |
| C | Poor adherence | 20% | 20%, 35% | |
| D | TDR | 10% | 7%, 13% | |
| 4 | Annual failure rates (in percentage) | |||
| Types of failures | Mean | SD | ||
| A | TDR (1C * 3D) | 1.20 | 0.19 | Obtained after simulation |
| B | Good adherence (2A * 3A) | 0.96 | 0.15 | |
| C | Partial adherence (2B * 3B) | 2.64 | 0.47 | |
| D | Poor adherence (2C * 3D) | 1.99 | 0.39 | |
| E | Total failures existing strategy (sum 4A+4B+4C+4D) | 6.8 | 5.78, 7.83 | |
| F | Total failures novel strategy first year (sum 1C+ 4B+4C+4D) | 17.65 | 15.43, 19.92 | |
| G | Total failures novel strategy subsequent years (sum 4B+4C+4D) | 5.6 | 4.64, 6.67 | |
| 5 | Death rates | |||
| A | Annual death rate (as percentage) among failures in the existing strategy | 24% | 18%, 30% | Function of total failures |
| B | Annual death rate (as percent) among failures in the novel strategy | 12% | 6%, 18% | |
| C | Actual death rate in the existing strategy (5A * 4E) | 1.64% | 1.29%, 2.04% | Obtained after simulation |
| D | Actual death rate in the novel strategy (5B * 4G) | 0.68% | 0.49%, 0.88% | |
NRTI, nucleoside reverse transcriptase inhibitor; NNRTI, non–nucleoside reverse transcriptase inhibitor; SD, standard deviation; TDR, transmitted drug resistance.
Outcome measure
The primary outcome measure was annual mortality rate in both the cohorts.
The secondary outcome measure included treatment failure rates and cost-effectiveness analysis.
-
i)Current strategy: failure was defined as per World Health Organization (WHO) criteria for clinical or immunological failure.7
-
(1)Clinical failure: appearance of new WHO stage IV condition indicating severe immune deficiency.
-
(2)Immunological failure: persistent CD4 count below 100 cell/μl or drop in patient's CD4 count to baseline or below.
-
(1)
-
ii)Novel strategy:26
-
(1)Clinical and immunological failure: as mentioned previously
-
(2)Virological failure: a PVL of 1000 copies/ml and above
-
(1)
Simulation runs
MC simulations were run with 1000 iterations over 20 years of follow-up using a stochastic model, for both cohorts.
Cost-effectiveness analysis
Cost-effectiveness analysis was performed.
-
a.
Direct costs of intervention (4 and 5 above) were included. This was done from societal perspective as medical care costs are completely borne by Defence Services, and there are no costs to the patient. Direct non-medical costs (travel, etc) were not considered. Indirect medical costs are fixed in the Armed Forces and hence are not considered. Intangible costs (pain, etc) were not included too; however, it got captured at the benefit end while estimating quality-adjusted life years (QALYs).
-
b.
Costs were discounted at a constant rate of 3% per annum.
-
c.
Survival benefit was converted to QALY based on Global Burden of Diseases (GBD) 2004 guidelines.27
-
d.
The benefits (QALY gained) too were discounted at 3% per annum.
-
e.
Sensitivity analysis was performed for the primary outcome measure.
-
f.
Δ Costs was calculated as difference in cumulative costs over 20 years after discounting (cost novel strategy - cost existing strategy).
-
g.
Δ QALY was calculated as difference in QALY over 20 years after discounting (QALY novel strategy - QALY existing strategy).
-
h.
Incremental cost-effectiveness ratio (ICER) was analyzed as annual costs per QALY gained (Δ Costs/Δ QALY).
-
i.
Similar analysis was performed with differential discounting for costs at 6% and QALY at 1.5% too.
Statistical analysis
MC simulations were performed in MS Excel using SimVoi add-in software. Comparison of continuous variables was performed using Student's t-test. Kaplan–Meier curves were drawn comparing the primary endpoints. Tornado diagram was plotted to depict sensitivity analysis. Data were analyzed using the SPSS software package version 23.0 (SPSS Inc., Chicago, Illinois, USA) and WinPEPI.
Results
Our study has shown that the novel strategy, if implemented in Armed Forces, can save almost 150 deaths over 20 years for every 1000 newly infected HIV patients started on therapy. This translated to almost 54% reduction in mortality, with number needed to treat to save one life being 2. The comparison of two strategies in terms of primary outcome measures of cumulative mortality, cumulative costs, cumulative QALY, and ICER is shown in Table 3. The failure rates are higher in the novel strategy as the virological failure gets picked up both at baseline (TDR) and during annual reviews.
Table 3.
Comparison between the two strategies.
| S No | Characteristics of the cohort after 20 years of modeling | Existing strategy N = 1000 | Novel strategy N = 1000 | Difference in values (Δ) | t-test statistic | P value |
|---|---|---|---|---|---|---|
| 1 | Cumulative costs in crores constant discounting (at 3% per annum) | 47.05 (±4.36) | 56.90 (±4.38) | Δ Costs = 9.84 | −50.4 | 0.000 |
| 2 | Cumulative QALY constant discounting (at 3% per annum) | 9934.0 (±235.1) | 10,672.0 (±234.4) | Δ QALY = 737.97 | −70.3 | 0.000 |
| 3 | Total deaths at 20 years (mean ± SD) | 279.9 (±7.13) | 130.43 (±6.03) | Δ Deaths = 149.54 | 506.08 | 0.000 |
| 4 | Cumulative costs in crores after differential discounting (at 6% per annum) | 35.5 (±3.3) | 43.5 (±3.0) | Δ Costs = 8.00 | −56.7 | 0.000 |
| 5 | Cumulative QALY after differential discounting (at 1.5% per annum) | 11,199.3 (±260.6) | 12,097.3 (±277) | Δ QALY = 898.05 | −75.17 | 0.000 |
| ICER | Formula | Value | ||||
| 6 | ICER with constant discounting | Δ Costs/Δ QALY | Rs 1,33,439 | |||
| 7 | ICER with differential discounting | Δ Costs/Δ QALY | Rs 89,139 | |||
| 8 | ICER per life saved | Δ Costs/Δ Deaths | Rs 32,925 | |||
ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life year; SD, standard deviation.
Sensitivity analysis was performed for the influence of the resistance rates on mortality and ICER (cost/QALY) and costs/death averted. The same has been depicted in Fig. 3(a–d). The Kaplan–Meier curves show the effect over time with respect to cumulative mortality and cumulative costs and the number of patients on first- and second-line therapy over 20 years of simulated follow-up (Fig. 4, a-d). A significantly higher number of patients shifted to second- and third-line therapy sooner in the novel strategy as the resistance was detected earlier. This translates to a higher cost.
Fig. 3.
Tornado plots showing sensitivity analysis comparing (a) existing strategy vs (b) novel strategy and also showing (c) ICER (costs/QALY) in lakhs and (d) ICER (costs/death averted). ICER, incremental cost-effectiveness ratio; NRTI, nucleoside reverse transcriptase inhibitor; NNRTI, non–nucleoside reverse transcriptase inhibitor; QALY, quality-adjusted life year; TDR, transmitted drug resistance.
Fig. 4.
Kaplan–Meier curves over 20-year follow-up between the two cohorts. (a) Cumulative mortality. (b) Cumulative costs in crores. (c) Cumulative number of patients on first-line ART. (d) Cumulative number of ART failures. ART, antiretroviral therapy.
The ICER calculated using the constant discounting method for both costs and QALY at 3% per annum was Rs 1.33 lakhs/annum, whereas the ICER calculated using differential discounting (costs at 6% and QALY at 1.5%) was under Rs 90,000 per annum. The cost to save one life was a mere Rs 32,925 per annum.
Discussion
The key finding of our study was a 54% reduction in mortality over 20 years of follow-up at a modest cost of under Rs 33,000 per life saved per year with the novel strategy. We have analyzed ICER using both constant discounting and differential discounting methods. Controversy exists between the best way to apply discounting, with the opinion veering toward differential discounting.28 As per WHO - Choosing Interventions that are Cost Effective; Interventions costing less than three times the national annual Gross Domestic Product (GDP) per capita for each QALY gained represent a cost-effective intervention, whereas one that costs less than one national annual GDP per capita is considered highly cost-effective.29 As per Census and Economic Information Center (CEIC) data, India's GDP per capita in 2018 was USD 1977.2 (Rs 1,38,000).30 Thus, any intervention costing up to Rs 4.2 lakhs per annum can be considered cost-effective, whereas that costing less than 1.4 lakhs per annum is highly cost-effective. The sensitivity analysis performed showed that even at the extreme values, the intervention remained cost-effective. The novel strategy suggested by us for the Armed Forces is thus highly cost-effective with either way of discounting (constant or differential).
The true cost incurred per annum is likely to be even lower as the cost assumptions have been taken at existing market rates. The true rates of all the tests and medications are lower by 10–20% (post-negotiations), whereas the costs of in-house resistance testing (as available at Armed Forces Medical College (AFMC), Pune) are almost 30% lower than the market rates.
Baseline viral load and resistance testing is a part of National Institute of heath (NIH) guidelines and Department of Health and Human Services (DHHS) guidelines, whereas the same is not yet recommended by NACO.22 HIV care in Armed Forces differs from the national program because of limited clientele, efficient referral and logistic services, disciplined patient population, good record-keeping/availability of data, and negligible dropouts. In many ways, it is akin to Western standards. Thus, there was a need felt for evaluation whether the novel strategy might work in the context of Armed Forces.
In this study, we not only addressed the gaps in the current policy but also assessed the effectiveness of the new strategy to overcome those gaps by stochastic modeling using MC simulations. The newer strategy involves two important decisions. First, HIVDR testing before ART initiation in newly diagnosed patients to rule out presence of TDR. In a large multicohort study in sub-Saharan Africa, it was found that compared with individuals without pre-treatment drug resistant (DR), the odds ratio for virological failure and acquired drug resistance was increased in participants with pretreatment DR to at least one prescribed drug.8 Second, as regards monitoring the response to therapy by yearly PVL, a prior study from South Africa has shown that delayed ART following virological failure was associated with elevated mortality.31 Immunological monitoring has lower sensitivity and positive predictive value, relying on which may result in delayed switch to second-line regimen.32 The public health benefit of maintaining adequate virological suppression also includes breaking the chain of transmission of HIV.
The current priority of the national program is to increase the coverage of ART to all eligible patients. This involves strengthening of services related to early HIV diagnosis, initiating treatment, and maintaining supply chain of antiretrovirals and retention in care. A question arises as to the efficacy of the novel strategy at the national level. The challenge is that the failure rate and non-adherence rates are almost 2–3 times higher among the civilian populations. Studies from various parts of India show a loss to follow-up (LFU) rates of almost 12–13% in the NACO-led program.33 With such a high rate of LFU, the novel strategy may not be feasible or cost-effective in the national scenario.
Unlike a conventional study that looks at a single dimension, the strength of the present study was that it looked at the implementation of the HIV program in Armed Forces in totality. The study identified program constraints, through a combination of different types of studies, and carried out mathematical modeling to identify key areas where changes done would be most beneficial.
Our study results should be interpreted in light of few limitations. First, the outcome of any mathematical modeling study depends on the assumptions made. All our assumptions have been drawn from our published and unpublished data and extensive review of Indian literature. Similarly, the death rates in the current and novel strategy was modeled stochastically based on failure rates and found to be around 1.64% and 0.68%, respectively, which related closely to existing data.24, 25 Assumptions, especially for the novel strategy, are open to challenge. We have consciously erred toward assumptions that make our analysis stringent. For example, our cost estimates are at market rates, almost 20–30% higher than actual costs. Similarly, we have assumed that despite following the best strategy, we would still have mortality rates 10–12 times higher than the USA (0.05% vs 0.68%) and almost twice that of other western nations.25 However, by modifying HIV policy comprehensively, we can aspire for even better mortality rate than 0.68% as estimated by us.
As per the UNAIDS 90-90-90 initiative, by 2020, 90% of all people living with HIV will know their HIV status, 90% of all people with diagnosed HIV infection will receive sustained ART, and 90% of all people receiving ART will have viral suppression.34 The current thrust in Armed Forces should be to achieve the third objective of the ambitious 90-90-90 target, that is, achieving viral suppression in 90% of individuals on ART. Our novel strategy can help to achieve this goal. To implement this new strategy, there is a need to strengthen laboratory capabilities in terms of PVL testing and HIV drug resistance testing. The novel study of structuring individual regimens for each HIV patient based on baseline drug resistance testing and plasma HIV viral load provides a scientific rationale for incorporation of the novel strategy of HIV treatment in the Armed Forces while being eminently cost-effective.
Conflicts of interest
The authors have none to declare.
Acknowledgment
This paper is based on Armed Forces Medical Research Committee Project No 4272/12 granted and funded by office of the Directorate General Armed Forces Medical Services and Defence Research Development Organization, Government of India.
References
- 1.Miller V., Hodder S. Beneficial impact of antiretroviral therapy on non-AIDS mortality. Aids. 2014;28(2):273–274. doi: 10.1097/QAD.0000000000000079. [DOI] [PubMed] [Google Scholar]
- 2.Braithwaite R.S., Shechter S., Chang C.C.H., Schaefer A., Roberts M.S. Estimating the rate of accumulating drug resistance mutations in the HIV genome. Value Health. 2007;10(3):204–213. doi: 10.1111/j.1524-4733.2007.00170.x. [DOI] [PubMed] [Google Scholar]
- 3.Cinatl J., Cinatl J., Rabenau H., Doerr H.W., Weber B. Failure of antiretroviral therapy: role of viral and cellular factors. Intervirology. 1994;37(6):307–314. doi: 10.1159/000150394. http://www.ncbi.nlm.nih.gov/pubmed/8586528 [DOI] [PubMed] [Google Scholar]
- 4.Wilkin T.J., Gulick R.M. When to start antiretroviral therapy? Clin Infect Dis. 2008;47(12):1580–1586. doi: 10.1086/593311. [DOI] [PubMed] [Google Scholar]
- 5.Ekstrand M.L., Shet A., Chandy S. Suboptimal adherence associated with virological failure and resistance mutations to first-line highly active antiretroviral therapy (HAART) in Bangalore, India. Int Health. 2011;3(1):27–34. doi: 10.1016/j.inhe.2010.11.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Shafer R.W. Genotypic testing for human immunodeficiency virus type 1 drug resistance. Clin Microbiol Rev. 2002;15(2):247–277. doi: 10.1128/CMR.15.2.247-277.2002. http://www.ncbi.nlm.nih.gov/pubmed/11932232 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Consolidated guidelines onthe use of antiretroviral drugs for treating and preventing hiv infection 2016, recommendations for a public health approach. https://apps.who.int/iris/bitstream/handle/10665/208825/9789241549684_eng.pdf;jsessionid=4D443FF54B41E0E981B7E749C9F20B3C?sequence=1. Accessed February 16, 2019.
- 8.Hamers R.L., Schuurman R., Sigaloff K.C. Effect of pretreatment HIV-1 drug resistance on immunological, virological, and drug-resistance outcomes of first-line antiretroviral treatment in sub-Saharan Africa: a multicentre cohort study. Lancet Infect Dis. 2012;12(4):307–317. doi: 10.1016/S1473-3099(11)70255-9. [DOI] [PubMed] [Google Scholar]
- 9.Sen S., Tripathy S.P., Chimanpure V.M., Patil A a, Bagul R.D., Paranjape R.S. Human immunodeficiency virus type 1 drug resistance mutations in peripheral blood mononuclear cell proviral DNA among antiretroviral treatment-naive and treatment-experienced patients from Pune, India. AIDS Res Hum Retrovir. 2007;23(4):489–497. doi: 10.1089/aid.2006.0221. [DOI] [PubMed] [Google Scholar]
- 10.Karade S.K., Ghate M.V., Chaturbhuj D.N. Cross-sectional study of virological failure and multinucleoside reverse transcriptase inhibitor resistance at 12 months of antiretroviral therapy in Western India. Medicine (Baltimore) 2016;95:37–4886. doi: 10.1097/MD.0000000000004886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bezabhe W.M., Chalmers L., Bereznicki L.R., Peterson G.M. Adherence to antiretroviral therapy and virologic failure: a meta-analysis. Medicine (Baltimore) 2016;95(15):e3361. doi: 10.1097/MD.0000000000003361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Deeks S.G. Treatment of antiretroviral-drug-resistant HIV-1 infection. Lancet. 2003;362(9400):2002–2011. doi: 10.1016/S0140-6736(03)15022-2. [DOI] [PubMed] [Google Scholar]
- 13.Allam R.R., Murhekar M.V., Bhatnagar T. Survival probability and predictors of mortality and retention in care among patients enrolled for first-line antiretroviral therapy, Andhra Pradesh, India, 2008-2011. Trans R Soc Trop Med Hyg. 2014;108(4):198–205. doi: 10.1093/trstmh/tru025. [DOI] [PubMed] [Google Scholar]
- 14.Balakrishnan P., Kumarasamy N., Kantor R. HIV type 1 genotypic variation in an antiretroviral treatment-naive population in southern India. AIDS Res Hum Retrovir. 2005;21(4):301–305. doi: 10.1089/aid.2005.21.301. [DOI] [PubMed] [Google Scholar]
- 15.Kannangai R., David S., Sundaresan V.C. Frequency of transmitted drug resistance mutations among treatment-naïve HIV-1-infected individuals at a tertiary care centre in South India. Mol Diagn Ther. 2015;19(5):273–275. doi: 10.1007/s40291-015-0160-5. [DOI] [PubMed] [Google Scholar]
- 16.Kumar M., Sen S., Gupta R.K., Chaturbhuj D., Tripathy S.P., Paranjape R.S. Prevalence of drug resistance in human immunodeficiency virus type 1-infected treatment-naive Children in Pune, India. AIDS Res Hum Retrovir. 2014;30(3):294–298. doi: 10.1089/aid.2013.0222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lall M., Gupta R.M., Sen S., Kapila K., Tripathy S.P., Paranjape R.S. Profile of primary resistance in HIV-1-Infected treatment-naive individuals from western India. AIDS Res Hum Retrovir. 2008;24(7):987–990. doi: 10.1089/aid.2008.0079. [DOI] [PubMed] [Google Scholar]
- 18.Karade S., Patil A.A., Ghate M. Short Communication: limited HIV pretreatment drug resistance among adults attending free antiretroviral therapy clinic of Pune, India. AIDS Res Hum Retrovir. 2016;32(4):377–380. doi: 10.1089/aid.2015.0277. [DOI] [PubMed] [Google Scholar]
- 19.Claxton A.J., Cramer J., Pierce C. A systematic review of the associations between dose regimens and medication compliance. Clin Ther. 2001;23(8):1296–1310. doi: 10.1016/S0149-2918(01)80109-0. [DOI] [PubMed] [Google Scholar]
- 20.Huang Y. Long-term HIV dynamic models incorporating drug adherence and resistance to treatment for prediction of virological responses. Comput Stat Data Anal. 2008;52(7):3765–3778. doi: 10.1016/j.csda.2007.12.016. [DOI] [Google Scholar]
- 21.Ghate M., Deshpande S., Tripathy S. Mortality in HIV infected individuals in Pune, India. Indian J Med Res. 2011;133(4):414–420. http://www.ncbi.nlm.nih.gov/pubmed/21537095 [PMC free article] [PubMed] [Google Scholar]
- 22.NACO Annual Report 2016-17. 2017. [Google Scholar]
- 23.Bajpai R., Chaturvedi H., Jayaseelan L. Effects of antiretroviral therapy on the survival of human immunodeficiency virus-positive adult patients in Andhra Pradesh, India: a retrospective cohort study, 2007-2013. J Prev Med Public Heal. 2016;49(6):394–405. doi: 10.3961/jpmph.16.073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Patrikar S., Shankar S., Kotwal A. Predictors of first line antiretroviral therapy failure and burden of second line antiretroviral therapy. Med J Armed Forces India. 2017;73(1):5–11. doi: 10.1016/j.mjafi.2016.09.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Unaids . 2018. UNAIDS DATA 2018.http://www.unaids.org/sites/default/files/media_asset/unaids-data-2018_en.pdf [Google Scholar]
- 26.Updated Recommendations on First-Line and Second-Line Antiretroviral Regimens and Post-Exposure Prophylaxis and Recommendations on Early Infant Diagnosis of HIV: Interim Guidelines. Supplement to the 2016 Consolidated Guidelines on the Use of Antiretrovir. 2018. [Google Scholar]
- 27.WHO | the Global Burden of Disease: 2004 Update. World Health Organization; 2014. https://www.who.int/healthinfo/global_burden_disease/2004_report_update/en/ [Google Scholar]
- 28.Attema A.E., Brouwer W.B.F., Claxton K. Discounting in economic evaluations. Pharmacoeconomics. 2018;36(7):745–758. doi: 10.1007/s40273-018-0672-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.WHO-CHOICE . 2003. Choosing Interventions That are Cost Effective. [DOI] [Google Scholar]
- 30.Up-to-date Chart & Data] India GDP per Capita. https://www.ceicdata.com/en/indicator/india/gdp-per-capita. Accessed February 21, 2019.
- 31.Petersen M.L., Tran L., Geng E.H. Delayed switch of antiretroviral therapy after virologic failure associated with elevated mortality among HIV-infected adults in Africa. AIDS. 2014;28(14):2097–2107. doi: 10.1097/QAD.0000000000000349. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Gunda D.W., Kidenya B.R., Mshana S.E., Kilonzo S.B., Mpondo B.C.T. Accuracy of WHO immunological criteria in identifying virological failure among HIV-infected adults on first line antiretroviral therapy in Mwanza, North-western Tanzania. BMC Res Notes. 2017;10(1):45. doi: 10.1186/s13104-016-2334-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Chakravarty J., Tiwary N.K., Prasad S.R. Determinants of survival in adult HIV patients on antiretroviral therapy in Eastern Uttar Pradesh: a prospective study. Indian J Med Res. 2014;140(4):491–500. http://www.ncbi.nlm.nih.gov/pubmed/25488442 [PMC free article] [PubMed] [Google Scholar]
- 34.Unaids . 2014. 90-90-90: An Ambitious Treatment Target to Help End the AIDS Epidemic.http://www.unaids.org/sites/default/files/media_asset/90-90-90_en.pdf [Google Scholar]
- 35.Chaturbhuj D.N., Hingankar N.K., Srikantiah P. Transmitted HIV drug resistance among HIV-infected voluntary counseling and testing centers (VCTC) clients in Mumbai, India. AIDS Res Hum Retrovir. 2010;26(8):927–932. doi: 10.1089/aid.2010.0032. [DOI] [PubMed] [Google Scholar]
- 36.Deshpande A., Karki S., Recordon-Pinson P., Fleury H.J. Drug resistance mutations in HIV type 1 isolates from naive patients eligible for first line antiretroviral therapy in JJ Hospital, Mumbai, India. AIDS Res Hum Retrovir. 2011;27(12):1345–1347. doi: 10.1089/AID.2011.0066. [DOI] [PubMed] [Google Scholar]
- 37.Braithwaite R.S., Shechter S., Roberts M.S. Explaining variability in the relationship between antiretroviral adherence and HIV mutation accumulation. J Antimicrob Chemother. 2006;58(5):1036–1043. doi: 10.1093/jac/dkl386. [DOI] [PubMed] [Google Scholar]