RPD106 - Identifying “Difference Maker” Factors in Patient-Reported Pain Interference: A Coincidence Analysis

Key Message : Coincidence analysis (CNA) provides an innovative method to analyze complex and heterogenous symptom data. CNA can help illuminate what combinations of contextual factors link directly to clinically relevant outcomes, such as degree of interference from cancer pain, and can help inform tailored interventions.

Abstract:

Background: The remote health monitoring system, Behavioral and Environmental Sensing and Intervention for Cancer (BESI-C), collects data from patient-caregiver dyads via environmental and wearable sensors to holistically understand the experience of cancer pain at home (1,2).

Objective: Utilize coincidence analysis (CNA) (3,4) to understand which factors are key “difference makers” related to the patient experience of pain interference.

Methods: Data were obtained from 2-week, in-home BESI-C deployments with patient-caregiver dyads recruited from a palliative care clinic and hospice program. Using patient-reported moderate or severe pain interference as the primary outcome of interest, we applied CNA (R v.4.5.1) to 30 patients (242 cases, each case representing one deployment day) to explore potential “difference making” causal factors. Potential causal factors included: baseline clinical and socio-demographic variables and patient-reported: 1) confidence in managing pain and functional status; 2) frequency of pain reports and corresponding severity scores; and 3) daily reported quality-of-life measures (e.g., mood, fatigue). We also explored caregiver perceptions of patient pain interference.

Results: Of the 40+ potential explanatory factors analyzed, two factors emerged as crucial “difference makers” in our strongest model: low mood and persistent pain. Mood was defined as “low” if the patient reported their mood as “poor” or “fair” (versus “good” or “very good”) on a given day. Pain was considered “persistent” if median pain severity was ≥5 on a 0-10 scale 40-minutes after initial pain reports on a given day. The presence of low mood or the persistence of severe pain consistently distinguished cases with reported high pain interference from those with lower pain interference.

Conclusion: CNA analysis can identify configurations of factors that explain clinically relevant outcomes for patients experiencing pain. Our results suggest tailored interventions that address low mood and persistent pain are especially critical to reduce high levels of pain interference for patients with cancer.

References: 1. LeBaron, V., Bennett, R., Blackhall, L., et al. (2020). Understanding the experience of cancer pain management at home from the perspective of patients and family caregivers to inform design of an in-home smart health sensing system. JMIR (Journal of Medical Internet Research) Formative Research, 4(8): e20836

2. LeBaron, V., Alam, R., Bennett, et al. (2022). Deploying the Behavioral and Environmental Sensing and Intervention for Cancer (BESI-C) smart health system to support patients and family caregivers in managing pain: Feasibility and acceptability study. JMIR (Journal of Medical Internet Research) Cancer, 8(3): e36879

3. Whitaker R.G., Sperber N., Baumgartner M., et al. (2020). Coincidence analysis: a new method for causal inference in implementation science. Implementation Science, 11;15(1):108.

4. Yakovchenko, V., Miech, E.J., Chinman, M.J., et al. (2020). Strategy configurations directly linked to higher hepatitis C virus treatment starts: an applied use of configurational comparative methods. Medical Care, 58(5), e31-e38.