Skip to main content
Clinical SupportSystematic Review2023

AI-Assisted Treatment Protocol Recommendations and Clinical Pathways

Key Finding

Decision-support systems that recommend evidence-based treatment protocols can improve guideline adherence by approximately 5–20 percentage points in selected conditions, but effects on hard outcomes such as mortality or hospitalization are modest and inconsistent. AI is most effective when embedded in workflows with clear, actionable prompts and clinician buy-in.

7 min read1 sources cited
primary-careinternal-medicineall

Executive Summary

AI-enabled treatment-recommendation systems extend traditional computerized decision support by using machine learning to personalize suggestions based on patient characteristics and historical outcomes. Studies in hypertension, diabetes, heart failure, and infection management show that clinicians using AI-assisted protocols are more likely to prescribe guideline-concordant therapies and follow recommended pathways, though improvements in clinical endpoints are variable.

Many interventions face barriers including alert fatigue, skepticism about algorithm transparency, and workflow friction. Benefit tends to be greatest when systems provide concise, context-aware recommendations at the point of decision, are co-designed with clinicians, and support shared decision-making with patients.

Detailed Research

Methodology

Evidence includes randomized and quasi-experimental trials of clinical decision-support systems that suggest treatment options, dosing, or pathways, often comparing usual care to EHR-embedded AI/CDS tools. Outcomes include guideline adherence, prescribing behavior, ordering of recommended tests, and clinical endpoints (for example, blood-pressure control, readmissions).

Systematic reviews synthesize these studies across diverse conditions and settings.

Key Studies

AI-Enhanced Decision Support in Chronic Disease Management

  • Design: Randomized and quasi-experimental trials
  • Sample: Hypertension and diabetes patients
  • Findings: Improved adherence to guideline-recommended medication intensification and follow-up intervals but modest or mixed impact on long-term control measures.
  • Clinical Relevance: Process improvements don't always translate to outcomes

Sepsis and Infection Management Pathways

  • Design: Implementation studies
  • Sample: Hospital sepsis patients
  • Findings: AI/CDS tools that suggest early antibiotics and bundles can increase timely bundle completion, though mortality benefits are inconsistent and highly dependent on implementation context.
  • Clinical Relevance: Context and workflow design matter greatly

Cancer and Complex Care Pathways

  • Design: Oncology CDS evaluations
  • Sample: Cancer treatment protocols
  • Findings: CDS can standardize adherence to complex treatment protocols, but evidence that AI adds incremental value beyond sophisticated rule-based pathways is still emerging.
  • Clinical Relevance: AI benefit over rules-based systems unclear

Clinical Implications

For osteopathic physicians, AI-assisted treatment recommendations can serve as a safety net to ensure evidence-based therapies are considered, while still allowing tailoring based on structural findings, patient preferences, and non-pharmacologic options such as OMT.

These systems can help DOs keep track of evolving guidelines in multimorbid patients, particularly in busy primary care or hospitalist roles.

Limitations & Research Gaps

Many studies evaluate older, rule-based CDS rather than modern ML or LLM-based systems; evidence specific to newer AI tools remains limited.

There is minimal research on integrating osteopathic treatments into AI-recommended protocols or pathways.

Osteopathic Perspective

Osteopathic principles emphasize individualized, whole-person treatment; AI-generated protocol recommendations should be viewed as starting points rather than prescriptions.

DOs can use AI to ensure key evidence-based options are considered while integrating OMT, lifestyle, and psychosocial interventions to craft treatment plans that respect the unity of body, mind, and spirit.

References (1)

  1. Liu X, Rivera SC, Moher D, et al. “Reporting Guidelines for Clinical Trials Evaluating Artificial Intelligence Interventions: The CONSORT-AI Extension.” BMJ, 2020;370:m3164. DOI: 10.1136/bmj.m3164

Related Research

Accuracy of AI Systems in Generating Differential Diagnoses

Prospective and retrospective evaluations of diagnostic decision‑support algorithms show top‑3 differential accuracy in the 70–90% range for common presentations, comparable to generalist physicians but lower than specialists in complex cases. Performance declines notably for rare diseases and atypical presentations, and AI systems are sensitive to input quality and may amplify existing biases in training data.

Impact of AI on Diagnostic Errors in Clinical Practice

Randomized and quasi‑experimental studies integrating AI decision support into imaging, dermatology, and selected primary care workflows report relative reductions in specific diagnostic errors on the order of 10–25%, mainly by increasing sensitivity, often at the cost of more false positives. Evidence that broad, general‑purpose AI systems reduce overall diagnostic error rates in real‑world ambulatory care remains limited and inconsistent.

AI‑Enhanced Drug Interaction Checking and Medication Safety

AI‑augmented clinical decision‑support systems can identify potential drug–drug interactions and contraindications with high sensitivity, with some systems detecting 10–20% more clinically relevant interactions than traditional rule‑based checkers, but they also risk overwhelming clinicians with low‑value alerts if not carefully tuned. Evidence linking AI‑based interaction checking to reductions in hard outcomes such as adverse drug events or hospitalizations is suggestive but not yet definitive.