CID 0911 - Adaptation of A Depression Treatment Intervention for HIV Patients in Cameroon

Patient registry only applicable to Specific Aim 3. This aim provided depression care management decision support to all depressed HIV patients who enrolled. There was no comparison arm. This study was an R34 feasibility study adapting measurement-based depression care to a low income sub-Saharan African setting, and Aim 3 assessed the logistics enrolling and monitoring patients.

(Per original protocol, which was followed) Specific Aim 3. Demonstrate the feasibility of recruiting and retaining patients over time and collecting and processing laboratory specimens.

In this aim the investigators seek to quantify objective measures of the feasibility of our planned RCT to assess the effectiveness of the MBC intervention on ART treatment outcomes in HIV patients in Cameroon. Specific measures will include percentage of eligible patients that consent to participate in the study, patient retention over 12 months, percentage of clinic visits and outcome interviews completed at each time point, and percentage of blood samples collected, tested and having valid results. These measures will be important in the quantification of the sample size needed for the definitive RCT. The investigators will also identify procedures that could hamper the successful implementation of the RCT and propose appropriate modifications.

Feasibility study: To achieve this aim the investigators will conduct a pilot prospective study with a sample of 60 participants. This study will begin during year 2 of the project and patients will be accrued over 4 months (15 patients per month) in order to allow for a smooth processing of participants and samples. Standard clinical care will be modified at Bamenda ATC so that the DCM will screen all newly presenting HIV patients for depressive symptoms with the PHQ-9. Patients scoring ≥10 on the PHQ-9 (or above an alternate threshold established in the validation study) will complete further evaluation with the DCM for presence of MDD. These patients will have their MDD confirmed by the HIV clinician. This two-stage screening procedure has been successfully applied by our team at rural medical clinics with low income patients.

The sampling frame for the feasibility study will consist of all patients who are HIV+, 18-55 years old, and have confirmed MDD. Patients will be excluded if they do not speak English well enough to complete the instruments or if they are not mentally competent to provide informed consent, as assessed by the medical team. As noted above, based on Dr. Atashili's clinical experience the investigators expect >95% of clinic attendees to be able to communicate effectively in English. Pregnant and breastfeeding women will not be excluded from the study but will be counseled regarding the particular risks and benefits of participation for them. AMI has the potential to adversely affect the developing fetus, primarily through its anticholinergic properties, although recent observational studies and systematic reviews indicate no increased risk of congenital malformations or developmental delays associated with AMI use. Further, because ADPs such as AMI are potent agents that act on the central nervous system, use by nursing mothers presents theoretical short and long-term risks to the infant's neurodevelopment. Infant behavioral monitoring before and after initiating treatment with the mother is clinically useful to monitor the effects of ADPs during nursing. Infants may exhibit short-term effects such as lethargy, poor weight gain, hypotonia and inconsolability, although the limited data available show that most infants of nursing mothers taking AMI doses consistent with our protocol have had no detectable AMI level. Potential neonatal effects must be weighed against the benefits to both mother and child of successfully treating a MDD. These issues and our related informed consent plans are discussed in detail in the Human Subjects Protection plan in the grant application.

Each day, the DCM will follow a random number algorithm to randomly select 12 patients from the daily register of scheduled patients. The investigators expect this to be sufficient to generate 4 eligible and consenting participants per week (assuming a 25% no show rate, 10% prevalence of MDD and <20% refusal) or ~15 per month and will adjust it if needed in order to meet recruitment goals. The DCM will assess eligibility and will invite eligible patients to participate. Non-consenting patients will be asked their reasons for denial. All patients with MDD (consenting or not) will be proposed a management plan by the DCM (as described under Aim 2 above) which will be discussed with the treating physician. Participants will be encouraged, to the extent they wish, to bring family members to appointments and to involve their social support structure in their depression treatment plan. For consenting patients the DCM will collect a whole blood sample to assess viral load ,CD4 count, and liver function. These samples will be stored at -20o C and transported to the CSCCD laboratory to be tested using standard procedures - VL using the automated Roche Amplicor assay of HIV VL, and CD4 counts using flow cytometry. The DCM will issue a follow-up plan to each participant and collect contact information (home address and phone number when available) to remind patients of follow-up appointments. To assist patient and provider in balancing pregnancy- and breastfeeding-related risks and benefits of AMI use, female participants will take a urine pregnancy test at enrollment and annually thereafter as long as they remain on AMI. Female participants who are not pregnant will be advised to discuss any plans for fertility with their provider, and those not wishing to become pregnant will be counseled to either avoid sexual intercourse or use contraception. Of note, all HIV+ patients at Bamenda are strongly counseled to either practice abstinence or use barrier methods of protection during sexual intercourse in order to avoid HIV transmission, with the secondary effect of avoiding pregnancy.

Follow-up visits will be aimed at assessing the effects of both ARV and ADP treatment. Assessment of depressive symptoms and ADP side effects will occur every 2 weeks until remission and every 4 weeks thereafter. Assessments at weeks 12, 24, etc. will occur during regularly scheduled clinic visits for ART prescription refills. Patients will return to the clinic for additional in-person assessments at weeks 2, 4, and 8; assessments at weeks 6 and 10 will be conducted by phone when feasible or else in person. Based on Dr. Atashili's experience with this patient population, the investigators expect approximately 50% of participants to be reliably reachable by phone. Project funds will pay for additional clinic visits required by the protocol. Critical decision points for potential modification of the depression treatment plan (e.g., maintain dose, increase dose, or switch ADP) will occur at weeks 4, 8, and 12 and will be informed by depressive symptom response and side effects as described under Aim 2 above. Clinical assessment of HIV disease and blood samples for CD4 counts and viral load will be conducted at the visits at 4, 8 and 12 months of follow-up; follow-up liver function tests will be performed based on clinical indication. Pregnancy tests will be repeated annually. Patients will be counseled to avoid alcohol consumption, and the DCM will clinically monitor alcohol use by asking patients at follow-up.

Separate from their ongoing contact with the DCM, participants will complete in-person interviews at baseline and at months 4, 8, and 12 to assess self-reported general health, ADP and ARV adherence, and sexual risk behavior. These in-person interviews will be conducted by a CSCCD interviewer who is not a part of the participants' clinical care to minimize the potential for social desirability bias. Interviews will be conducted in a private location of the participant's choosing and will last ~30 minutes. General health will be measured by the Medical Outcomes Study Short Form-8 (SF-8). Adherence for each medication will be measured with four visual analog scale (VAS) items asking patients what proportion of the time in the past month they have taken all of their medication doses; taken more medication than prescribed; missed or skipped doses; and taken doses on time. VAS adherence measures correlate well with other self-report measures of adherence and predict virologic failure. Sexual risk behavior will be measured with questions asking participants about the frequency of oral, anal, and vaginal sexual intercourse in the preceding 4 months, separately by sex of the partner, by the partner's HIV status, and by whether a condom was used. The interview will additionally include questions on age, sex, religious and ethnic identification, educational attainment, and household income. Finally, the interview will include questions about the acceptability of the intervention, including patients' perceptions of the additional time cost imposed by the intervention as well as their willingness and ability to purchase ADPs (current cost of <$1 per month) if the intervention were part of standard clinical care rather than a research study.

As a second measure of adherence, the investigators will collect pharmacy records of the quantity and date of all HIV and psychiatric medications dispensed to study participants from a period starting 4 months before study enrollment to the end of follow-up. All patients receiving HIV care at Bamenda ATC obtain their HIV medications at the hospital pharmacy, which would also be the source for any ADP medications prescribed as part of this study. Adherence will be calculated as the ratio of the number of days' supply of medication divided by the days elapsed before the next refill. Adherence as measured by pharmacy refills has been shown in other settings to be associated with virologic outcomes.

The DCM will attempt to contact patients who miss any visit by either calling the participants when phone numbers are available or visiting the patient's home. The investigators recognize that not all patients have phone numbers and that poor identification (naming and numbering) of streets/homes will constitute a challenge for identifying participants' homes. The investigators expect 50% of participants to have phone numbers.