Schlitz and Braud’s 1997 review¹ catalogued the terminological diversity in the field: transpersonal imagery, intercessory prayer, spiritual healing, non-directed prayer, intentionality, energy healing, shamanic healing, non-local healing, non-contact Therapeutic Touch, and Reiki are all treated as DHI variants. Each involves a distinct theoretical, theological, cultural, or pragmatic approach, but all share the assumption that healing can occur under conditions that transcend conventional sensory communication, a claim that challenges standard causal ontology. Schlitz’s 2014 review⁴ noted that a US National Center for Health Statistics survey ranked prayer for self as the most popular complementary and alternative medicine practice among American adults, and prayer for others as the second most popular, making DHI a phenomenon of substantial public-health relevance regardless of its ultimate mechanistic status.

The Schlitz and Braud 1997 meta-analysis¹ synthesized 30 formal experiments in which self-reported healers, “psychics,” and other self-selected volunteers attempted to influence autonomic nervous system activity in a distant person. The primary outcome measure was electrodermal activity (skin conductance level) recorded during intention periods versus randomly interspersed control periods. Results across the 30 experiments showed a significant and characteristic variation during distant intentionality periods. The review explicitly considered possible alternative explanations including suggestion and self-regulation, arguing that the experimental design, physical isolation of sender and receiver, random scheduling of intention epochs, and physiological rather than self-report outcome measures, mitigated but did not fully eliminate these alternatives. The review noted that the effect was not uniform across studies and that healer selection, motivation, and training were potential moderators. No independent replication of the full 30-study meta-analysis by a skeptical research team is reported in the pool.

In her 1995 paper,⁵ Schlitz outlined three research domains for intentionality in healing: (1) self-healing, where intentions interact with one’s own body via psychoneuroimmunological and biofeedback mechanisms; (2) indirect influence on others through placebo and “noxebo” effects, where communication is the operative pathway; and (3) nonlocal influence on others, where healing occurs under conditions that preclude conventional mind-body interactions. She identified methodological challenges specific to each tier, for self-healing, the difficulty of controlling for unconscious self-regulatory processes; for placebo effects, the challenge of constructing appropriate blinds; and for nonlocal healing, the fundamental epistemological challenge of ruling out all known causal pathways. The paper argued that remissions research, well-documented cases of recovery against medical odds, represented an underexplored natural-history resource for understanding self-healing mechanisms. No specific effect sizes are reported in this conceptual paper.

The Schlitz et al. 2012 study³ was conducted at California Pacific Medical Center and examined whether distant healing intention directed toward post-surgical patients would accelerate wound healing compared with a control condition. The study was explicitly described as exploratory. Patients and wound assessors were blinded to healing assignment, addressing the artifact of demand characteristics and expectancy effects on outcome ratings. The null result on the primary wound-healing outcome is notable because it came from a team that had previously found positive effects in laboratory paradigms, suggesting that the translation from tightly controlled physiological measures (skin conductance) to complex clinical outcomes (tissue repair) introduces confounds that the laboratory design does not. The paper is cited in the broader DHI literature as an example of the methodological challenges facing clinical translation of laboratory psi findings.

Schlitz et al. 2003⁷ addressed the methodological complexity of differentiating among psychological, psychosocial, and psychic influences in intentional healing research. The paper discussed experimenter effects (the finding that the experimenter’s own beliefs and expectations can influence outcomes, a phenomenon Schlitz herself demonstrated in the Wiseman collaboration), patient expectation effects, baseline psychological factors as predictors of medical outcomes, and experimental design questions including healer selection criteria, structure and duration of the intervention, and the construction of appropriate blinds. The paper argued that these design variables must be standardized before cross-study comparisons are meaningful. It also noted that the distinction between self-healing (psychoneuroimmunological) and nonlocal healing (putatively transpersonal) cannot be made on the basis of outcome data alone, the causal pathway must be independently assessed.

Radin, Stone, Levine, Eskandarnejad, Schlitz et al. 2008² enrolled 36 couples in 38 test sessions. In 22 couples, one partner was a cancer patient; in 12 of those couples, the healthy partner had been trained to direct intention toward the patient and had practiced daily for three months prior to the experiment (trained group). In 10 couples, the pair was tested before the partner received training (wait group). Fourteen healthy couples received no training (control group). Skin conductance level was measured simultaneously in both members of each couple. The sending person directed intention toward the receiver during repeated 10-second epochs separated by random interepoch periods; the receiver relaxed in a distant shielded room for 30 minutes per session. The primary outcome, normalized skin conductance means during intention epochs versus randomly selected interepoch control periods, was assessed using nonparametric bootstrap procedures. The overall result was z = 3.9, p = .00009 (two-tailed), indicating significantly elevated skin conductance during intention epochs across all groups. The preplanned group comparison (trained vs. wait vs. control) was not significant. A post-hoc analysis showed that peak deviations were largest and most sustained in the trained group, followed by the wait group, then the control group. The post-hoc nature of this moderator finding means it carries exploratory rather than confirmatory weight and requires independent replication with a preregistered group-comparison hypothesis. The artifact of shared physiological rhythms between intimately bonded couples, which could produce correlated skin conductance fluctuations independent of intention, was partially addressed by the shielded room design (eliminating direct sensory contact) but not fully eliminated, as low-frequency physiological coupling mechanisms are not fully understood.

The 2008 study’s post-hoc finding that trained, motivated cancer-patient couples showed the largest autonomic effects² aligns with a broader theoretical claim in Schlitz’s work: that the quality of the sender’s intention, its depth, training, and motivational grounding, is a meaningful variable, not merely a binary on/off switch. This claim has methodological implications: if motivation and training are genuine moderators, then studies using untrained volunteers as senders may systematically underestimate effect sizes relative to studies using experienced healers or intimately bonded partners. Schlitz’s 2003 guidelines paper⁷ identified healer selection and training as among the most important unresolved design variables in the field. The 2015 overview by Radin and Schlitz⁶ noted that heterogeneity in healer selection across studies is a primary source of between-study variance in DHI meta-analyses, making pooled effect-size estimates difficult to interpret.