Research Spotlights: September 2023

Progress made but gaps remain: Availability of mental health services at HIV treatment sites

There is extensive literature indicating that common mental disorders (CMDs) are highly prevalent among people living with HIV (PWH) globally. Supported by NIMH, NIAID, and NICHD, researchers assessed trends in the reported availability of screening for the most common types of CMD, i.e., depression, anxiety, and post-traumatic stress disorder (PTSD), as well as available treatment options in the International epidemiology Databases to Evaluate AIDS (IeDEA) Consortium HIV treatment sites, which are located across a diverse range of countries across levels of income. The IeDEA Consortium was established by the NIH in 2006 to provide a dataset of globally diverse HIV/AIDS data collected and defined in various regions around the world, across Africa, Asia-Pacific, the Caribbean, and Central, South, and North America, allowing researchers to apply methodology to harmonize and pool data cost-effectively, in order to address high-priority research questions impacting PWH. The overall goal of this research is to provide insight into the need for greater integration of mental health services into regular HIV treatment, to better serve the needs of PWH in the U.S. and elsewhere around the world.

Inclusion criteria for the survey data collected in 2020 included IeDEA sites that were actively contributing site-level data. Survey questions assessed the availability of screening and treatment for depression, anxiety, or PTSD prior to the start of the COVID-19 epidemic. A list of validated screening tools were provided to each site, which helped sort responses into whether the practice at the site was to have patients “screened with a validated instrument” or “not screened with a validated instrument.” Sites were also asked to describe treatment interventions provided for those CMDs when screening was available. For the 2016/2017 survey period, responses were collected from a stratified random sample for IeDEA sites in LMICs, with questions limited to depression and PTSD screening and treatments.

A total of 227 out of 238 sites participated in the 2020 site survey; 223 sites responded with screening and treatment data, representing 41 countries. Half of the responding sites (52%) reported using validated instruments to screen for CMDs. Screening for CMDs with a validated instrument and providing counseling or medication as treatment options was most commonly reported in high income countries (HICs) versus LMICs.

Out of all eligible IeDEA sites located in LMICs, only 68 sites participated in both 2016/2017 and 2020 survey periods. A majority of these 68 sites (85%) were located in urban versus rural settings. While there was a higher count of any screening for depression reported in 2020, the reported availability for screening and treatments decreased in 2020, when compared to the 2016/2017 period. No significant difference was found between availability for screening or treatment for PTSD between 2016/2017 and 2020.

The biggest gaps in mental health services according to the researchers were:

  • the combined availability of screening and group counseling for depression, where 54% of sites surveyed reported that these services were unavailable during both survey time points,
  • PTSD screening, which 59% reported unavailable at both time points, and
  • combined availability of screening and individual counseling for PTSD, which 65% of sites reported unavailable at both time points.

Forty-four percent of sites reported availability of depression screening alone at both time points. Of the sites that did not offer depression screening in 2016/2017, 31% offered it in 2020. For PTSD screening, 24% of sites that did not offer it at the first time point were offering it in 2020. Limitations to this study include a small sample size for comparison data points and that survey responses were self-reported and not independently verified. Nevertheless, this study is a useful guide regarding the need for improvements in mental health services for PWH globally in resource-constrained areas. In addition, researchers offer recommendations for further research into sustainable practices to address care gaps experienced by resource-constrained sites.

Parcesepe AM, Stockton M, Remch M, Wester CW, Bernard C, Ross J, Haas AD, Ajeh R, Althoff KN, Enane L, Pape W, Minga A, Kwobah E, Tlali M, Tanuma J, Nsonde D, Freeman A, Duda SN, Nash D, Lancaster K; IeDEA Consortium. Availability of screening and treatment for common mental disorders in HIV clinic settings: data from the global International epidemiology Databases to Evaluate AIDS (IeDEA) Consortium, 2016-2017 and 2020. J Int AIDS Soc. 2023 Aug;26(8):e26147. doi: 10.1002/jia2.26147. PMID: 37535703; PMCID: PMC10399924.

Preventing revictimization among recent survivors of sexual violence

Previous research has shown that over half of girls, young women, and adult women who have experienced rape will suffer another incident of sexual assault and that 50% of individuals with histories of child sexual abuse experience sexual or physical violence in adulthood. Thus, it is important that trauma-informed approaches to addressing sexual violence emphasize the impact of repeated exposure. In a recent study, supported by NIDA, researchers undertook a secondary analysis to assess whether certain ecological risk factors, including poverty, structural racism, prior history of sexual or physical violence, and unemployment, could be assessed by emergency department personnel at intake to provide tailored risk reduction programs to those survivors of sexual or physical violence who are at increased risk for further violence within a short period of time.

In this study, researchers analyzed data from their completed randomized controlled trial (RCT) that measured the impact of a video intervention on post-traumatic stress disorder (PTSD) levels among recent survivors of sexual violence who presented for the sexual assault medical forensic exam (SAMFE) at an emergency department (ED). The goal of this secondary analysis was to isolate characteristics of women who experienced additional instances of sexual or physical violence at follow-up during the RCT versus those who did not. Demographic information included racial/ethnic identification, marital status, income, and education. The RCT’s inclusion criteria were: 15 years or older, female survivors of rape or attempted rape who completed a SAMFE within 7 days of the crime at either of two emergency departments in a midwestern urban area. The secondary analysis sample consisted of a subset of the original RCT recruitment.

Researchers analyzed unadjusted data points to assess risk factors for experiencing a secondary sexual or physical crime during the 6-month period following the initial SAMFE. Low-income status was significantly associated with the occurrence of subsequent sexual or physical attacks. Additionally, respondents who recalled more details of the initial experience of perpetrated violence, experienced life-threatening conditions during the crime, had a prior history of sexual or physical violence, and those presenting with higher distress at the time of the initial SAMFE had higher associations with secondary attacks at follow-up. Using a model with adjusted data points, researchers found that those with a prior history of assault and income lower than $10,000 annually were at an increased risk of physical or sexual violence at the six month follow up time point.

The findings from this study indicate the value of screening for certain demographic risk factors at ED intake and highlights the potential impact of offering low-income survivors material resources to support their recovery. The limitations of this study include a smaller overall sample size, and lower rates of T1 and T3 assessment completion compared to initial enrollment post-SAMFE, and the lack of controlling for factors relating to intimate partner violence. In conclusion, the researchers note that more research is needed to prevent revictimization among recent rape victims. Additionally, policies to provide financial support to recent rape victims and/or targeted prevention for those with pre-rape victimization at the SAMFE may be useful for reducing revictimization risk in this population.

Walsh K, Gilmore AK, Jaffe AE, Frazier P, Ledray L, Acierno R, Ruggiero KJ, Kilpatrick DG, Resnick HS. A preliminary examination of sexual and physical victimization 6 months after recent rape. Arch Womens Ment Health. 2023 Aug;26(4):495-501. doi: 10.1007/s00737-023-01335-y. Epub 2023 Jun 8. PMID: 37286883.

Positive impacts of an adaptive digital health intervention for patients prescribed opioid analgesics

Overdose deaths involving opioid analgesic (OA) misuse comprised 80% of overdose deaths in the United States in 2021. Patients at risk of OA misuse can be screened for risk reduction interventions during emergency department (ED) clinical encounters. However, behavior change motivation decreases after the ED discharge is completed. Thus, providers need a systematic tool that could address the lack of health services after discharge that may increase OA misuse risk. Providing reinforcement and assistance with decision-making beyond the clinical setting could help address this urgent public health challenge.

Researchers, funded by NIDA, sought to address the challenge by developing a mobile intervention called “Prescription Opioid Wellness and Engagement Research in the ED” or “PowerED” that utilizes reinforcement learning (RL), an artificial intelligence (AI) process that incorporates user feedback to provide treatment choices. In this study, the researchers assessed if patient-reported OA risk indicators (the RL system’s “reward” function) would improve as PowerED “learned” how to suggest treatment options that were best-suited to individual patient needs. They tracked longitudinal changes in patient-reported OA risk behaviors after baseline data collected at the time of clinical trial enrollment.

Researchers took a patient-centered approach, ensuring they pilot-tested PowerED first to confirm its positive effects on OA misuse behaviors, and also evaluated patient burden by varying frequencies of OA risk assessments delivered via interactive voice response (IVR). Participant inclusion criteria at time of ED intake was any reported OA misuse (utilizing the Current Opioid Misuse Measure or “COMM” questionnaire) in the prior 3 months, an age range of 18 to 70 years, and presenting with a pain-related complaint. After initial assessments, patients were randomized to the PowerED system or enhanced usual care.

The intervention sample consisted of 228 patients. The average patient age was 40 years, with a majority of participants (approximately 60%) identifying as White women who were employed, and whose highest level of education was high school. A majority (76.8%) reported experiencing high levels of chronic pain. PowerED patients engaged in 12 weeks of behavioral intervention via the app, which in turn addressed their OA risk and provided non-OA pain management options. Response rates were similar across all baseline COMM score subgroups, approximately 51%.

Therapeutic interventions were structured on enhancing motivation and CBT, focusing on improving overall health, patient safety, and improving communication with providers. The PowerED intervention consisted of IVR calls utilizing a female human voice lasting no more than 5 minutes, 5 times a week, during which patients were prompted to report their use of prescribed OAs, pain levels, and respond to questions regarding OA-related risks. The PowerED program then chose each week from among 3 types of treatment options and delivered that selection to the patient:

  1. A less than 5-minute long IVR call matching self-reported levels of pain severity that provided motivational messaging focused on safe use and healthy pain management (selected 43% of the time across all participants)
  2. A 5 to 10-minute IVR call communicating OA risks and behavior change strategies in keeping with self-reported pain severity and OA use (selected 36% of the time)
  3. A 20-minute telephone conversation with a trained human counselor (selected 22% of the time), who was supervised biweekly and who had to participate in 8 hours of structured training prior to participating in this trial. Completion rates were higher for patients assigned a live call, and higher still if they had recorded higher pain scores and OA misuse scores at baseline.

Researchers found that PowerED was able to successfully identify when live counseling was appropriate. The program was also able to identify when brief or extended IVR could substitute for live counseling, and ultimately led to improvements in patient reported OA risk over time. The program showed the greatest potential for effectiveness in patients with higher baseline risk scores. There were no significant improvements over time for patients with lower baseline scores.

In conclusion, these data indicate the potential usefulness of PowerED in reducing OA misuse. One limitation of this study was that IVR response rates were low, and that the program had no access to data that was not self-reported regarding patient status and actual OA risk. Researchers advise proceeding with caution when developing AI-supported behavioral interventions given that limited diversity in AI training samples can lead to biased decisions when utilized in groups that reflect real-world heterogeneity. Researchers also advise including more user-friendly language, as well as electronic health records (EHR) integration. Evaluating this program with a larger and more diverse sample of patients with high OA risk is recommended.

Piette JD, Thomas L, Newman S, Marinec N, Krauss J, Chen J, Wu Z, Bohnert ASB. An Automatically Adaptive Digital Health Intervention to Decrease Opioid-Related Risk While Conserving Counselor Time: Quantitative Analysis of Treatment Decisions Based on Artificial Intelligence and Patient-Reported Risk Measures. J Med Internet Res. 2023 Jul 11;25:e44165. doi: 10.2196/44165. PMID: 37432726; PMCID: PMC10369305.