A target for harm reduction in HIV: reduced illicit drug use is associated with increased viral suppression.

By Lauren Strand

In the midst of a fatal drug epidemic and shifting drug policy in the United States, there is continued interest in the relationship between illicit drug use and negative health outcomes. Because substance use is difficult to characterize in individuals, studies often target sub-populations with more substantial and better-documented substance use profiles. One example is people living with HIV, in whom substance use has been associated with poor engagement in the HIV care continuum, lower likelihood of receiving antiretroviral therapy, reduced adherence to therapy, and increased disease-related mortality. Recently I collaborated on a study finding that reduction in frequency of illicit opioid and methamphetamine use is associated with viral suppression among people living with HIV. Since viral suppression is an important consideration for individual’s health as well as disease transmission, this finding has important policy implications for harm-reduction around substance use frequency.

This work was spearheaded by Robin Nance and Dr. Maria Esther Perez Trejo and advised by Dr. Heidi Crane and Dr. Chris Delaney, colleagues and mentors of mine during my time at the Collaborative Health Studies Coordinating Center (University of Washington). The publication in Clinical Infectious Diseases focuses on the longitudinal relationship between reducing illicit drug use frequency and a key biomarker in HIV, viral load (VL), among people living with HIV. This study used longitudinal data from the Centers for AIDS Research Network of Integrated Clinical Sites (CNICS) cohort. CNICS is an ongoing observational study consisting of more than 35,000 people living with HIV receiving primary care at one of eight sites (Seattle, San Francisco, San Diego, Cleveland, Chapel Hill, Birmingham, Baltimore, and Boston). Importantly, CNICS provides peer-reviewed open access to patient data including clinical outcomes, biological data, and patient-reported outcomes. This study also used individual data in four studies from the Criminal Justice Seek, Test, Treat, and Retain (STTR) collaboration. STTR is an effort to combines data from involved observational studies and trials to improve outcomes along the HIV care continuum for people involved in some part of the criminal justice system. One example is individuals recently released from jail who have struggled in the past with substance use disorders.

Within CNICS, substance use was collected at clinical assessment via tablets approximately every six months with instruments including the modified Alcohol, Smoking, and Substance Involvement Screening Test and the Alcohol Use Disorders Identification Test. Drug use was defined as frequency of use in the last 30 days and was further categorized according to longitudinal trends from baseline: abstinence (no use at baseline or follow-up), reduction in use without abstinence (use at baseline that has declined at follow-up), and non-decreasing (similar or increased use). Drug categories were marijuana, cocaine/crack, methamphetamine, and heroin/other illicit opioids.  Viral suppression was defined as an undetectable VL (<=400 copies/mL). Analytic models for each individual drug were joint longitudinal and survival models with time-varying substance use and adjustment for demographics, follow-up time, cohort entry year, and other concomitant drugs including alcohol and binge alcohol. These longitudinal models account for repeated measures and differential loss to follow-up (unbalanced panels).

Analyses (mean follow-up of 3.9 years) included approximately 12,000 people living with HIV with a mean age if 44 and of whom 47% were white. Marijuana was widely used at baseline, though methamphetamine was also common. Relative to non-decreasing use, abstinence was associated with an increase in odds of viral suppression ranging from 42% for marijuana to 118% for opioids (all four substance groups statistically significant). Reduction in use was associated with an increase of 65% for methamphetamine and 172% for opioids. The directionality and statistical significance of these results were maintained in sensitivity analyses with pooled fixed effects meta-analysis using both CNICS and STTR studies.

Ultimately, findings from this large-sample longitudinal analysis suggest that abstinence of all drug groups increases the likelihood of viral suppression and, more interestingly, reducing frequency without abstinence may also increase the likelihood of viral suppression for illicit opioids and methamphetamine. This finding may support the use of medication-assisted treatments (MAT) to reduce substance use, which could have the potential to improve disease-related outcomes for people living with HIV. However, this study did not evaluate why individuals may have increased or decreased use of illicit substances (e.g. MAT, or other treatment programs). In any case, reduction of illicit substance like opioids and meth (even when abstinence is not achieved) seems like a logical target for harm reduction interventions in people living with HIV and likely, in the broader population, to improve overall health outcomes.

One extension of this work would be to evaluate the relative value of programs targeting abstinence and substance use reduction among individuals with HIV compared with other programs. This, of course, requires a true causal relationship between substance use and viral load, which is likely mediated through ART adherence. A simple Markov model could include states for suppressed and not suppressed; however, because suppression reduces the risk of transmission, we might also incorporate shifting dynamics of the population of people living with HIV. Both transmission and individual outcomes were considered in a recent cost-effectiveness analysis of financial incentives for viral suppression authored by CHOICE Alumna Dr. Blythe Adamson and Professors, Dr. Josh Carlson and Dr. Lou Garrison. The main study finding was that paying individuals to take HIV medications was associated with health improvement, reduced transmission, and reduced healthcare costs. While this finding is fascinating, substance use may be an important contextual consideration. One previous study found that financial incentives did not improve viral suppression among substance users and it is unclear how financial incentives may impact drug use and addiction. This is an active area of research and debate. Our study did not look at increases in substance use and viral suppression because we wanted to address the question around reduction and abstinence. Regardless, additional research on strategies to improve viral suppression are needed as well as a better understanding of the interplay between substance use behavior, other risk behaviors, adherence, and viral suppression among people living with HIV.

Ecological Studies of Marijuana

If you live in one of the states which has legalized recreational marijuana or a state that is considering it, you have possibly seen one of the following billboards:

Available via http://www.wltx.com/article/news/local/verify/verify-do-states-that-legalize-marijuana-have-25-fewer-opioid-deaths/482535530
Credit Steven Lemons, available via https://frontpageconfidential.com/weedsmaps-billboards-marijuana-arizona/

The simple black background and white lettering makes them pop, but the statements themselves are even more captivating. The content covers contentious, hot-ticket topics: the opioid epidemic, health spending, and marijuana legalization. But what gets left out is the context:  for most readers, these statements imply causality, despite there being limited evidence for these relationships to date.

To the casual observer, these are impressive, exciting statements. A beneficial effect of a historically outlawed and much maligned substance is indeed fascinating! A more cautious observer might be wondering about the source of these claims, and, indeed the fine print appears to contain references! The more cautious observer might now be appeased.

But really, we should all pause here, for two reasons:

Firstly, these billboards are advertising. I will not get into further discussion of advertising, political or otherwise, however I will note that “Weedmaps,” the billboard producers, is poised to be your go-to search engine and rating site for marijuana strains and producers.

Secondly, causality is complex and elusive. The two studies cited on these particular billboards (Bachhuber et al. 2014; Bradford & Bradford 2017) are ecological in design, meaning they have aggregated data (in this case, states) as the unit of analysis. The variables in these analyses are features of the states, including the main variable of interest, implementation of medical cannabis laws (note that medical is missing from both billboards).  The research design is appropriate for questions about the average effects of medical cannabis laws on an outcome of interest (more on this later). But, these findings are subject to residual confounding on the state level. In addition, they are subject to the ecological fallacy in their interpretation, and as we all know, interpretation is what matters most.

Both studies include other state-level variables in the analyses that might explain the change in their outcomes over time, such as the implementation of state-wide Prescription Drug Monitoring Programs (PDMPs). PDMPs occurred in many states over the period studied, and based on similar analytic designs, may be largely responsible for improvements in opioid outcomes. Both studies account for PDMPs, and the first also considers several other opioid laws and policies that effectively restrict availability. These authors also performed several nice checks of robustness. For example, a secondary model was used to adjust for state-level linear time trends in outcome (i.e. including a random slope for state).  Authors note this technique may account for changes in concepts that are difficult to measure, such as attitudes, and other time-varying confounders. The study employed analysis of negative controls: death rates from other conditions supposedly not associated with cannabis (e.g. heart disease and septicemia), which authors would expect to remain unaffected by legalization.

Despite these nice checks, it is unlikely that these analyses accounted for all potential confounding variables, especially those that change over time.  And this is almost always the case, as it’s virtually impossible to observe, let alone control for, all sources of confounding. Adjusting for linear trends produced results that were only marginally statistically significant. Especially in dealing with states, the inclusion of a large number of explanatory variables quickly becomes a high-dimensional problem, where there are only 50 states with a few years of data, but many more variables than that. The question then becomes whether this residual confounding is enough to change our interpretation of these studies.

Interpretation of these studies (especially in the media) may suffer from the ecological fallacy, a logical fallacy when inference made on a group does not necessarily translate into inference on an individual’s behavior or risk. From these findings, we cannot make any inference about individuals’ patterns of opioids and cannabis use (i.e. substitution) and individuals’ underlying risk of negative opioid outcomes (i.e. substitution effects). In other words, we cannot link marijuana legality to the use patterns of individuals.

So where do we go from here?

The past decade has been something of an ecological study renaissance. This is not a bad thing. Such studies are useful for hypothesis generation, and population-level risk factors are very relevant in public health and medicine.  Population-level risk factors may be important effect modifiers or cause exposure to individual risk factor. Differences in state laws can make for great “natural experiments” where groups of people are “randomized” to an exposure by a natural process, and a pre-post assessment can be made.

But mostly importantly, it comes down to inference. Inference from these studies might inform marijuana policy but should not inform interventions on individuals. Lots of discussion has been generated by these studies, and there is a great deal of room for misinterpretation (sample headline: “How marijuana is saving lives in Colorado”).

On the bright side, the scientific community recognizes this problem, and it is likely that additional studies of the individual- and population-level effects will be undertaken.  A recent well-designed study from RAND (Powell et al. 2018) replicated Bachhuber et al., finding that adding more state-level variables and additional years of data to the model nullifies the effect of medical marijuana laws on opioid overdose mortality. Moreover, the authors identified that a more meaningful effect on opioid outcomes is achieved through protected and operational dispensaries, i.e. access, where the largest effect was seen during a time period of relatively lax regulation of dispensaries in California, Washington, and Colorado.

How to ensure that other new investigations will be high quality and unbiased is another question. Regardless, the tide for marijuana research appears to be turning. As more and more studies are published, it is imperative that researchers are clear about their analysis limitations, especially when their results might end up on a billboard.

Reminders About Propensity Scores

Propensity score (PS)-based models are everywhere these days.  While these methods are useful for controlling for unobserved confounders in observational data and for reducing dimensionality in big datasets, it is imperative that analysts should use good judgement when applying and interpreting PS analyses. This is the topic of my recent methods article in ISPOR’s Value and Outcomes Spotlight.

I became interested in PS methods during my Master’s thesis work on statin drug use and heart structure and function, which has just been published in Pharmacoepidemiology and Drug Safety. To estimate long-term associations between these two variables, I used the Multi-Ethnic Study of Atherosclerosis (MESA), an observational cohort of approximately 6000 individuals with rich covariates, subclinical measures of cardiovascular disease, and clinical outcomes over 10+ years of follow-up. We initially used traditional multivariable linear regression to estimate the association between statin initiation and progression of left ventricular mass over time but found that using PS methods allowed for better control for unobserved confounding. After we generated PS for the probability of starting a statin, we used matching procedures to match initiators and non-initiators, and estimated an average treatment effect in the treated. Estimates from both traditional regressions and PS-matching procedures found a small, dose-dependent protective effect of statins against left ventricular structural dysfunction. This finding of very modest association contrasts with findings from much smaller, short-term studies.

I did my original analyses using Stata, where there are a few packages for PS including psmatch2 and teffects. My analysis used psmatch2, which is generally considered inferior to teffects because it does not provide proper standard errors. I got around this limitation, however, by bootstrapping confidence intervals, which were all conservative compared with teffects confidence intervals.

Figure 1: Propensity score overlap among 835 statin initiators and 1559 non-initiators in the Multi-Ethnic Study of Atherosclerosis (MESA)

Recently, I gathered the gumption to redo some of the aforementioned analysis in R. Coding in R is a newly acquired skill of mine, and I wanted to harness some of R’s functionality to build nicer figures. I found this R tutorial from Simon Ejdemyr on propensity score methods in R to be particularly useful. Rebuilding my propensity scores with a logistic model that included approximately 30 covariates and 2389 participant observations, I first wanted to check the region of common support. The region of common support is the overlap between the distributions of PS for the exposed versus unexposed, which indicates the comparability of the two groups. Sometimes, despite fitting the model with every variable you can, PS overlap can be quite bad and matching can’t be done. But I was able to get acceptable overlap on values of PS for statin initiators and non-initiators (see Figure 1). Using the R package MatchIt to do nearest neighbor matching with replacement, my matched dataset was reduced to 1670, where all statin initiators matched. I also checked covariate balance conditional on PS in statin initiator and non-initiator groups. Examples are in Figure 2.  In these plots, the LOWESS smoother is effectively calculating a mean of the covariate level at the propensity score. I expect the means for statin initiators and non-initiators to be similar, so the smooths should be close. In the ends of the age distribution, I see some separation, which is likely to be normal tail behavior. Formal statistical tests can also be used to test covariates balance in the newly matched groups.

Figure 2: LOWESS smooth of covariate balance for systolic blood pressure (left) and age (right) across statin initiators and non-initiator groups (matched data)

Please see my website for additional info about my work.