How patient navigators view the use of financial incentives to influence study involvement, substance use, and HIV treatment

Abstract

Background and Aims:

While patient navigation has been shown to be an effective approach for linking persons to HIV care, and contingency management is effective at improving substance use-related outcomes, Project HOPE combined these two interventions in a novel way to engage HIV-positive patients with HIV and substance use treatment. The aims of this paper are to examine patient navigator views regarding how contingency management interacted with and affected their navigation process.

Design:

Semi-structured qualitative interviews.

Participants:

22 patient navigators from the original 10 Project HOPE study sites.

Measurements:

Individual, semi-structured interviews lasting approximately 60 minutes addressed the patient navigator’s professional background, descriptions of the participant population, substance use disorder versus HIV treatment entry and engagement issues, and the use of contingency management within the navigation service delivery protocol.

Findings:

Patient navigators believed that financial incentives helped motivate participant attendance at navigation sessions, particularly early in study involvement, which helped them to establish rapport and develop relationships with participants. Patient navigators often noted that financial incentives positively influenced targeted HIV health-related behaviors, such as attending medical appointments, which provided a rapid pay-off with an escalating sum. Contingency management was more complex when used by the patient navigators for substance use-related behaviors, particularly when incentives revolved around negative urine screening. Patient navigators noted that not all participants responded the same way to the contingency management and that the incentives were particularly helpful when participants were financially strained with limited resources or when internal motivation was lacking.

Conclusions:

Overall patient navigators found the inclusion of contingency management to be helpful and affective at influencing participant behaviors, particularly concerning navigation session attendance and HIV healthcare-related participation. However, issues and concerns surrounding the inclusion of contingency management for drug-related behaviors as delivered in Project HOPE were noted.

1. Introduction

Patient navigation is a patient-centered approach to care coordination that provides individuals with assistance to overcome barriers to receiving and adhering to health care. Patient navigation has been effective in linking persons to HIV care (Craw et al., 2008; Gardner et al., 2005). Some people with untreated HIV concurrently have substance use disorders and may need additional assistance in accessing substance use treatment as well as HIV treatment (Bell et al., 2010; Gardner et al., 2016; Uhlmann et al., 2010).

Contingency management is a system of delivering financial incentives to increase the frequency of desirable behaviors. It is effective in improving substance use-related outcomes, including promotion of abstinence (Lussier, Heil, Mongeon, Badger, & Higgins, 2006) and attendance at counseling sessions (Fitzsimons, Tuten, Borsuk, Lookatch, & Hanks, 2015; Sigmon & Stitzer, 2005). Among substance-using populations, the use of financial incentives has increased health-related behaviors, such as returning for medical test results (Malotte, Hollingshead, & Rhodes, 1999), completion of hepatitis B vaccine series (M. L. Stitzer, Polk, Bowles, & Kosten, 2010; Weaver et al., 2014), and promoting HIV-testing among disadvantaged populations (Saxena, Hall, & Prendergast, 2016). While incentives have been shown to be effective at increasing behaviors related to specific, short term outcomes, even with challenging populations such as HIV-infected patients with active substance use, they appear to be less successful over the long-term at increasing outcomes such as virologic suppression or adherence to a complex series of health behaviors, as are required for chronic disease management, especially once the incentives are removed (Bassett, Wilson, Taaffe & Freedberg, 2015).

Hospital Visit as Opportunity for Prevention and Engagement for HIV-infected Drug Users (Project HOPE)

Project HOPE compared patient navigation with or without financial incentives to increase viral suppression among 801 HIV-positive patients with any opioid, stimulant, or heavy alcohol use within the past 12 months, recruited from 11 hospitals across the United States. Both patient navigation groups were hypothesized to be superior to a treatment as usual group. Viral suppression (≤200 copies/mL) relative to viral load non-suppression or death at the 12-month follow-up was the primary study outcome. Secondary outcomes included HIV health care engagement (e.g., HIV care provider visits, medication adherence), drug treatment entry, urine drug screens, and patient navigation intervention participation.

Patient navigation in Project HOPE consisted of up to 11 sessions over a 6-month intervention period. Under the ideal scenario, more frequent meetings would occur in the early post-hospital discharge period with diminishing frequency over time. However, if participants were doing well, the full 11 sessions were not necessarily required and conversely, for those with whom contact was temporarily lost and re-established during the intervention, more frequent sessions could be scheduled later in the 6-month intervention period. The harm-reduction-grounded patient navigation intervention incorporated motivational interviewing techniques along with strengths-based case management (Metsch et al., 2016). Participants randomized to navigation with incentives received the same 11 navigation session intervention with the addition of a contingency management program that provided financial incentives of escalating value for 7 pre-determined target behaviors, including: 1) attendance at navigation sessions; 2) completion of paperwork prerequisite to care (e.g., health insurance forms); 3) attendance at scheduled HIV care visits; 4) possession of a current HIV medication prescription (as a proxy for medication adherence); 5) attending substance use disorder treatment; 6) providing negative breathalyzer readings and drug negative urine specimens for opiates, oxycodone, methadone, cocaine, amphetamine, and methamphetamine; and 7) meeting viral suppression criteria. A total of $1160 in possible earnings was available over the 6-month intervention (M. Stitzer, Calsyn, et al., 2017), with health-related targets garnering half of the financial incentives, and patient navigation meeting attendance and substance use-related goals each accounting for one-quarter of the possible earnings. Project HOPE found that there were no differences between groups in viral load at 12 months follow-up (6 months after completion of the interventions); however, at 6 months participants in the patient navigation group with incentives were more likely to be virally suppressed and to have engaged in HIV care than participants in the treatment as usual condition (Metsch et al., 2016). Additional secondary analyses that compared behavioral outcomes for patients receiving navigation with and without financial incentives have shown that those with incentives attended substantially more navigation sessions (M. Stitzer, Matheson, et al., 2017), and had more HIV care sessions and more validated medication prescription checks than those without incentives (M. L. Stitzer et al., 2018).

The current sub-study, conducted as part of the NIDA Clinical Trials Network, is an exploratory qualitative investigation of patient navigator views regarding the inclusion of contingency management in a navigation intervention with this high-need, seriously ill population. The advantages and disadvantages of combining these very different interventions were explored from the perspective of the patient navigators, who both provided the strengths-based intervention and implemented the contingency management program with their participants, including tracking of target behaviors and dispersal of incentive payments. The purpose of this paper is to examine navigator views as to how contingency management interacted with and affected the patient navigation process.

2. Methods

2.1. Data Source

The 22 patient navigators from the 10 original Project HOPE study sites completed individual, in-person semi-structured interviews between April and September 2013. An additional site was added mid-way through the parent trial to increase recruitment but was not included in this qualitative subsample. Most interviews were conducted in the second year of Project HOPE recruitment to ensure that patient navigators had ample experiences delivering the contingency management intervention. However, length of time in the role and experiences delivering patient navigation services were diverse across the sample, with some having functioned in that capacity for over a year and others newer to the role having served in a different capacity (e.g., research assistant) in the study prior to becoming a patient navigator. All patient navigators completed an informed consent process. They were not paid directly for their sub-study participation, as their interviews were completed during normal business hours. This project was approved by the Johns Hopkins Medicine’s Institutional Review Board.

2.2. Participants

The sub-study sample included all patient navigators working at the 10 sites at the time the interviews were conducted, with the exception of 4 navigators who were still new to the role (with limited experience or still in-training). Nineteen of the participants were full-time patient navigators with the remaining 3 splitting their time between administrative/research roles and patient navigation. Several participants were former Project HOPE research assistants who transitioned into navigator positions as they opened up. The sample included 16 females and 6 males, and the racial/ethnic composition was 64% African American, 9% Indian/Asian, 18% non-Hispanic white, and 9% Hispanic white. The training backgrounds of the patient navigators varied widely, ranging from some college education combined with prior outreach worker experience, through master’s-level clinical degrees. Among the patient navigators, 10 possessed either some college or bachelor’s degrees (often not in clinical or health-related fields) and 12 held master’s degrees (2 in counseling, 3 in public health, 6 in social work, and 1 with a masters in a non-health field). Three patient navigators specifically mentioned possessing either outreach worker, health education, or addiction certifications.

2.3. Semi-structured interviews

All interviews were conducted by the lead author and lasted approximately 60 minutes, addressing the professional background of the patient navigators themselves, descriptions of their participant population, substance use disorder versus HIV treatment entry and engagement issues observed in the intervention, and aspects of the navigation service delivery protocol, with an emphasis on the use of contingency management during the study. A semi-structured interview guide was developed by the lead author and revised, in an iterative manner, following feedback from site investigators possessing expertise in either contingency management or patient navigation. The guide consisted of pre-determined questions and follow-up probes for gathering additional details, but unexpected topics that emerged during the interviews were also explored and the interview guide expanded over the course of the data collection process. For example, earlier responses to the main question “What are your thoughts about the inclusion of contingency management as part of the patient navigator process?” and the sub-question “How do you talk about it with your patients?” led to more specific sub-questions concerning language and framing of discussions about contingency management. Navigators were informed that the purpose of the interview was to examine the Project HOPE patient navigation intervention from their own perspective. All interviews were audio recorded, transcribed, and checked for accuracy prior to analysis using Atlas.ti qualitative analytic software version 6.2.

2.4. Analysis

Coding was conducted by a team of researchers with qualitative expertise, including the primary author (SGM, LBM and EL). Transcribed interviews were entered into Atlas.ti for analysis using a grounded theory approach (Strauss & Corbin, 1991). During the open coding phase, the investigators approached the data looking for descriptions of contingency management and how it was perceived to influence participant behaviors throughout the study (e.g., linkage and attendance at HIV treatment, linkage and participation in drug treatment, study participation). Inter-coder reliability was achieved by having all three analysts’ first code the same transcript independently and then coming together to compare identified text segments until agreement was reached. This process was repeated with additional transcripts until coding reliably had been established for all codes, at which point all remaining transcripts were coded by a single analyst. Emergent sub-themes related to contingency management were then identified and categorized during the selective coding phase using the constant comparative method, in which subthemes were repeatedly revisited in the data to detect outliers and exceptions (Glaser & Strauss, 1967).

3. Results

3.1. Contingency Management and Study Engagement

Consistent with the objective data on patient navigation session attendance (M. Stitzer, Matheson, et al., 2017) patient navigators reported that the financial incentives helped motivate people to attend navigation sessions early on. This early engagement was beneficial in that it gave the navigators opportunities to develop better rapport, figure out what was most important to their participants, and employ strength-based and case management strategies to help them.

With contingency management I think it helps motivate the patients [R: Does it?] just what it’s intended to do. And I believe it’s working because my participant, in the beginning he wasn’t very excited about coming to his appointments, more or less waiting in the clinic or any of that, but now that he knows he will get an incentive for meeting with me, for attending his doctor’s appointments, for taking his meds, for getting his lab works done, he’s more eager to do it. And if that works then that’s great because it is and has resulted with suppressed viral load. (Patient Navigator 1)

Other patient navigators agreed that the financial incentives promoted early study engagement, but also noticed that participants were able to use contingency management incentives to purchase goods and services that facilitated their access to improved health and overall care.

It’s so much easier to get participants to come in for money first of all. Most of the time the attendance is much lower for PN-only [patient navigation-only] folks. I think contingency management is great as a behavioral intervention, but I think, honestly, at the end of the day for our folks it’s a resource. Our folks can earn a lot of money. And they buy food and they buy medicine and they buy a place to sleep or clothing or bus tickets or all these things. It’s almost like contingency management was another way of breaking down barriers for folks to access services. So I think it’s great. (Patient Navigator 18)

3.2. Contingency Management and HIV Health Care Behaviors

According to the patient navigators, financial incentives also positively affected their participants’ behaviors for some of the targeted activities, such as completing health care-related tasks (e.g., attending appointments).

Oh yeah, no, I’ve had tremendous success stories because of the incentive; absolutely there’s no question about it. Yes, you’ve had it and they’ve gone to the doctor’s appointment because of it. Oh no, I’m all for the incentive. (Patient Navigator 15)

One navigator mentioned that continued engagement in medical appointments, particularly once participants are starting to feel better, can be a challenge, but that the incentives helped to keep navigation with incentives participants’ adherent with appointments later in the program.

Contingency management, I think it helps motivate the patients… because it does get harder at the end when you’re tired of coming to the doctor’s. This really helps… (Patient Navigator 1)

There were a range of health behaviors that were reinforced as part of the study protocol and financial incentives may have helped motivate participants, particularly when the payoff occurred quickly and was of a considerable sum.

I think for doctor’s visits or like, yeah. Because yeah, I mean it’s quite a bit of money [escalating payments starting at $30] right there. I think for most of them it’s, there are things that they know that they can do… And so it’s the things that seem feasible and the things that seemed doable and realistic and also the big ticket items. (Patient Navigator 17)

3.3. Contingency Management and Substance Use Behaviors

Since the patient navigation intervention was grounded in harm reduction, the navigators did not press for substance use cessation or entry into drug abuse treatment programs, such as methadone treatment or drug counseling, but they did assess on a regular basis their participants’ willingness to enter treatment. Irrespective of their substance use treatment status, patient navigation with incentives participants were offered opportunities to earn other substance use-related incentives by providing negative urine specimens for 6 targeted substances as well as negative alcohol breathalyzer readings. While this feature was meant to motivate substance abstinence independent of treatment participation, it could also engender unwanted behaviors, such as attempts at cheating by falsifying urines, or manipulation of the system, such as timing drug use to avoid positive drug screens at navigation appointments. These behaviors, while perhaps predictable given the participant population, raised concerns among some of the navigators.

Just as an example I had a participant who used two days a week she said by the time she got to us… when she saw the cash incentives that she was losing she began to challenge herself to not use before the visits. And then when she had her first clean UA she was like, “Awesome! Okay this is good.” When she would go out with her friends they were like, “Do you want?” “No, no,” basically, I don’t want to miss my money. Unfortunately, as our visits became more and more spaced apart she had time in between to use. So as opposed to using before the visits, she would use the day of the visit, after the visit or the next day. And so it was hard for me to confront her about that but I did. (Patient Navigator 4)

The example above illustrates the potential power of the incentives to affect drug use while also suggesting that parameters of the intervention necessitated by the overall study protocol, (e.g. low frequency of urine testing) may not have been ideal.

A portion of the incentive money could be earned for entering substance use disorder treatment and attending the first few treatment sessions, although incentives for negative urines could be earned independent of whether the participant was in treatment. While acknowledging the way incentives affected drug use patterns, patient navigators did not think that the incentives were sufficient to motivate participants to enter drug treatment if they did not already want to do so, as illustrated in the comment below about two particular cocaine users in the navigator’s caseload who managed to quit on their own without treatment:

Both of them had used cocaine like regularly…. They didn’t [go to treatment], they just stopped. Neither of them were interested in treatment. Yeah, so I mean, that was huge. I mean, that they quit cocaine is a big, big deal! (Patient Navigator 9)

A final feature of the substance use intervention that was troublesome to navigators was that not all drugs tested in a 10-drug panel were included in the incentive program. In particular, marijuana and benzodiazepines were not included, as the intervention developers wanted to avoid possible complications with verifying prescribed medications. This led to uncertainties about how to justify this detail of the incentive program to participants. One example is given below.

There’s people that show up positive for certain drugs that the study doesn’t care about. It’s hard; it’s hard. I say this is, I just say the drugs that the study is really looking at are…, and those [that the participant tested positive for] aren’t the drugs that the study is really looking at, so it’s strange. If they’re not interested in treatment then I’m not going to explore it and do what they want, the motivational interviewing stuff, but I don’t push it at all…. I mean and I don’t even mind from a harm reduction perspective. It’s great. Like, “Great, you’re using pot and you’re not smoking crack. That’s awesome. Like, that’s cool. Okay here’s twenty bucks. That’s great.” And that, I truly honestly believe that, but it’s confusing. It can be a really confusing situation. (Patient Navigator 7)

3.4. General Issues

3.4.1. Perceived Value of Financial Incentives

This distinction regarding the “value” of the incentives for patients of differing socio-economic status was mentioned frequently by patient navigators. They served a range of patients, spanning from fully-employed and well-paid stimulant-using MSM to participants with opioid use from poorer communities of color. In general, patient navigators perceived that the financial incentives were particularly helpful for those patients who were financially strained with limited resources. When asked about which patient population was especially motivated by incentives, one navigator responded:

Heroin users. Absolutely, because they need it and they come in early because they need to get high, so they’re very dependent, very dependent it, on the incentives. (Patient Navigator 15)

3.4.2. Participant response to failure

One disadvantage cited for the incentives is that they could exacerbate the sense of failure when people who had been doing well relapsed; this is highlighted in the following description:

Well and what was also really difficult is he was in the CM [contingency management] arm and really benefited a lot from those incentives then all of the sudden he was so excited because he knew in the final meeting he was going to get an incentive for being in a treatment program. Then he started using again so he didn’t get that incentive for the negative urine, [and] he didn’t get the treatment. So then we had this final meeting where he was also just really depressed because he just felt like he had failed himself. (Patient Navigator 20)

As this case illustrates, having a lot to gain could also mean having a lot to lose when incentives were not earned.

3.4.3. Internal versus external motivation

Some patient navigators struggled with the distinction between internal and external motivation and with what was more important to patients: their intrinsic health care motives or earning the incentives. In the beginning many participants joined the research to achieve health goals, but once the incentives started they appeared, for some, to become the primary force driving behaviors. In these cases, the patient navigators expressed concern that the participants’ internal motivation, which they considered key to success, was being masked by the financial incentives. Ultimately, however, the navigators felt that the financial incentives could be extremely helpful in motivating some patients to participate in the Project HOPE intervention when other factors, such as motivation to improve their health, was insufficient.

It helps them, it does help of course. Does that mean that only my CM [patients] are going to do well? No. I have PN’s [patients in the patient navigation without CM arm] that have done better than my CM’s and vice versa. I just think it really depends on the person. I’ve had tremendous success stories because of the incentive; absolutely there’s no question about it. But it doesn’t necessarily mean that the outcome, I think it depends on the person just like stopping drugs. If they want to stop they’ll do it. If they want to go on their meds they’re going to do it. You can guide them and be their guardian angel and make them feel good, which is great, and help them through the system, but the bottom line is they have to get up every morning and put it in their mouth. That’s the hard part. (Patient Navigator 15)

In the quotation below, a navigator describes a contingency management participant that she worked with, expressing her belief that the incentives were the driving force for the patients’ participation

Yeah, the question would just be whether they would continue to find that valuable even without the incentives. Yeah, so I’m not sure. I think one of my participants sometimes just feels like she’s coming in to see me and can’t wait for like the session to be over so she can walk away with the gift cards. (Patient Navigator 21)

A different navigator was more optimistic because the participant appeared to be intrinsically motivated and put his contingency management incentives second to the desire for improved health outcomes.

But for some people on my case load, their motivation is that they want to do this and this is kind of like a bonus, meaning the incentive. So, for them the first it’s primarily it’s their motivation for change and then the secondary is the incentive. So, again for that gentlemen you know he made the decision, he was like, you know, I’m not in this for the money. I want to do this because I need to do this. (Patient Navigator 6)

Navigators also distinguished among different subgroups of participants and felt that the incentives were more important for some than others. As one navigator described, intrinsic motivating factors may have been even more of a motivator for subgroups of participants:

For the guys, I don’t think it’s about the money because some of our guys are MSM [men who have sex with men] so you know it’s always a lot of image. I think they just want to look better, feel better. I feel like they want to be able to walk down the street and feel like their self again. So most of them are not motivated by the [financial] incentive that they receive. Most of them just want the care. And with our older guys, most of them, I think they’re motivated by the incentives due to the fact that some of them, most of them, are homeless and active drug users. (Patient Navigator 13)

4. Discussion

4.1. Findings

Financial incentives were included in Project HOPE as an added component of patient navigation with the premise that they would further boost rates of viral suppression beyond that engendered by patient navigation alone among patients with uncontrolled HIV and substance use. Importantly, the contingency management intervention was delivered by the navigators and thus became an integral component of their navigation activities. Beyond testing efficacy of the combined intervention, this study provided the opportunity to learn how the patient navigators viewed the utility of including contingency management, a perspective that has not previously been explored. Findings from this qualitative sub-study shed new light on the blending of these interventions. They indicate that, from the navigator’s point of view, the incentives were helpful for increasing positive behaviors, including attendance at navigation sessions and HIV care visits, particularly for participants who were financially disadvantaged or lacking in other motivators, but were less helpful for addressing the unhealthy behavior of drug use. The viewpoints expressed by navigators are generally consistent with objective findings from secondary data reported in the main study outcomes paper (Metsch et al., 2016) and post-hoc analyses (M. Stitzer, Matheson, et al., 2017; M. L. Stitzer et al., 2018). While there were no group differences in outcomes at 12 months after effects of the intervention had dissipated, participants in the navigation with incentives arm attended significantly more navigation sessions (Metsch et al., 2016; M. Stitzer, Matheson, et al., 2017), were more likely to have engaged in HIV care, and had more HIV care visits (M. L. Stitzer et al., 2018) than participants in the navigation-only arm. Further, only those in navigation with incentives had a significantly lower viral load by the end of the 6-month intervention compared with usual care (Metsch et al., 2016).

Navigators also had concerns about the strength of internal motivation among those who received incentives. Concerns regarding the potential for external rewards to overshadow or diminish internal motivation have been noted for decades, with the mixed findings attributed to mediating variables pertaining to how rewards are interpreted by recipients (Deci, Koestner, & Ryan, 1999; Wiersma, 1992). In particular, external rewards may be less effective when they are seen as a way to control behaviors or to provide feedback regarding competency. In the case of the present study, it is likely that the tangible-incentives were effective in part because they were combined with positive feedback (i.e., verbal rewards) from navigators while also providing more opportunities for engagement and rapport to be built between patients and navigators. The fact that outcomes for the two patient navigation groups were no longer different at 12 months, 6 months beyond the end of the active intervention phase, suggests that the combined intervention, while successful at improving engagement during the intervention, was not successful at engendering the conditions necessary to sustain health care behaviors after the intervention ended. More research is needed to understand how internal motivation may be related to long-run success and the general conditions under which positive health-related behaviors can be sustained throughout and beyond the end of an active intervention (interpreted as internal motivation), whether or not contingent financial incentives are incorporated.

The main outcome paper reported that participants in both patient navigator groups compared to treatment-as-usual were more likely to engage in professional substance use disorder treatment at 6, but not 12 months, although rates of engagement were low (30% in contingency management and 18% in treatment as usual) (Metsch et al., 2016). It may be that the navigators were able to assist participants in overcoming motivational and concrete barriers to SUD treatment entry and that the financial incentives to enter treatment were not sufficient to improve entry further. Indeed, navigators were less sanguine about the ability of incentives to impact substance-related behaviors, including abstinence and/or SUD treatment entry. This may, in part, reflect inadequate parameters of intervention for the substance-related target behaviors. For example, out of the possible $1160 that a participant could earn, only $310 was allocated for substance use related goals. It was also noted that frequency of testing may have been a barrier to an effective intervention, especially later in the program when the spacing of navigation visits increased for most participants, with protocol-specified visits only scheduled at monthly intervals during the last few months. Nevertheless, the inclusion of urine testing as part of the navigation intervention was a significant aspect of the program that served, at minimum, to focus attention on substance use issues.

Navigators naturally focused on and derived their impressions from their own individual cases, so that they were sensitive to individual differences in response to the incentives. Thus, they described the incentives being a greater motivator for low-income participants than more financially secure participants. They were also sensitive to the potential downside of incentives for those who experienced relapse or failure to carry out other targeted behaviors. Incentive amounts were available on an escalating pay schedule for all target behaviors, with larger amounts available later in the program as the behavior was repeated over time. This is a strategy designed to enhance external motivation later in the program to sustain performance at a time when behaviors may have a natural tendency to degrade. But it could be challenging for participants to reach goals later in the program, leading potentially to disappointment and, in the case of substance use relapse, to attempts at cheating or manipulating the urine testing system. Overall, the intervention appeared to be less successful at changing substance use than health-care behaviors and further refinement in this area may be warranted.

4.2. Limitations

There are a number of limitations to consider when interpreting our findings. First, interviews occurred while the project was underway and may not reflect the views regarding contingency management held by the navigators at the conclusion of the project. Second, the project utilized a complex and novel contingency management intervention that addressed multiple behaviors simultaneously, so that perspectives expressed by navigators may not generalize to other contingency management interventions. Third, not all Project HOPE patient navigators were interviewed, and additional perspectives may have been missed by failing to include them in the sample. Finally, the findings cannot speak to the merits or limitations of combining contingency management with patient navigation for other health issues or with other populations. Strengths include a relative large sample of interviewed navigators working with a large, demographically heterogeneous group of HIV positive substance users during a lengthy (6-month) health-focused protocol conducted in real world settings.

5. Conclusion

In conclusion, navigators noted the particular benefits of adding incentives to help bring people to navigation sessions early in the intervention, giving them a chance to build rapport and promote early engagement. Navigators felt that the addition of financial incentives to the intervention was also helpful for improving HIV health-related behaviors during the active intervention phase of the study, at least for some participants, above and beyond the benefits of the navigation services alone. They also noted that incentives were helpful later in the intervention to give participants additional motivation to attend HIV treatment when they were starting to feel fatigued with the treatment process. On the negative side, they noted deficits in the ability of incentives, as currently structured, to impact substance use behavior and struggled with uncertainty about how financial incentives may have impacted participants’ “internal motivation.” These perspectives on the dynamics of the combined intervention provide insights that will be useful for potential improvements in future. This includes improvements in specific parameters of the combined intervention as well as improved training of navigators to help them better anticipate and deal with individual client reactions relative to the allure and reality of financial incentives.

Highlights.

• Financial incentives were viewed as enhancing targeted HIV health-related behaviors when combined with the patient navigation service.

• The inclusion of financial incentives was more complex when used by the patient navigators for substance use-related behaviors, particularly when incentives revolved around negative urine screening.

• This differing “value” of the incentives for patients of differing socio-economic status was noted by patient navigators.