Health Services Research – Access Measurement, Care Delivery Patterns

Definition and Core Concept

This article defines Health Services Research (HSR) as a multidisciplinary field that examines how people access healthcare, how much care costs, what happens to patients as a result of care, and how healthcare systems and policies influence these outcomes. HSR integrates methods from epidemiology, economics, sociology, statistics, and management science. Core questions: (1) access – do individuals receive needed care? (2) quality – what is the appropriateness and effectiveness of care? (3) costs and efficiency – are resources used wisely? (4) equity – are outcomes similar across population groups? (5) organisation and delivery – how do system structures affect outcomes? The article addresses: objectives of HSR; key concepts including access frameworks, practice variation, and risk adjustment; core mechanisms such as administrative data analysis, patient surveys, and qualitative interviews; international comparisons and debated issues (data privacy, comparative effectiveness research, implementation science); summary and emerging trends (learning health systems, artificial intelligence for health services, patient-reported outcomes integration); and a Q&A section.

1. Specific Aims of This Article

This article describes health services research without endorsing specific findings. Objectives commonly cited: improving healthcare quality and safety, reducing unwarranted variation, controlling costs, informing policy decisions, and reducing disparities. The article notes that HSR findings have influenced payment reforms (e.g., value-based purchasing), guidelines, and accreditation standards.

2. Foundational Conceptual Explanations

Key terminology:

• Access (Andersen model, 1995): Potential access (have insurance, usual source of care), realised access (actually received care), and equitable access (care not determined by income, race, location).
• Practice variation (Wennberg, 1970s): Differences in medical care rates not explained by patient illness or preferences. High variation suggests unwarranted factors (supply of resources, physician practice style).
• Risk adjustment: Statistical methods to account for differences in patient severity or case mix when comparing outcomes across providers. Prevents penalising providers caring for sicker patients.
• Comparative effectiveness research (CER): Direct comparison of two or more interventions to determine which works better for which populations under which circumstances.

Data sources in HSR:

• Administrative claims data (insurance billing records – demographics, diagnoses, procedures, costs).
• Electronic health record (EHR) data (clinical details, lab results, medications, notes).
• Surveys (patient experience, health status, access).
• Registries (disease-specific, device registries).
• Vital statistics and census data.

3. Core Mechanisms and In-Depth Elaboration

Access measurement (access frameworks):

• Financial: insurance coverage, out-of-pocket costs, medical debt.
• Structural: provider availability, wait times, after-hours care.
• Cultural: language concordance, trust, discrimination experiences.

Practice variation methods:

• Small area analysis (rates of procedures per capita across regions).
• Inflation-adjusted variation indices (coefficient of variation, systematic component of variation).
• Dartmouth Atlas of Health Care (US) documents 2-5 fold variation in many procedures (knee replacement, cardiac catheterisation, back surgery).

Risk adjustment models (examples):

• Hierarchical Condition Categories (HCC) for Medicare payment.
• Charlson comorbidity index, Elixhauser comorbidity measure.
• Expected vs observed outcomes ratios for provider profiling.

Effectiveness evidence (examples):

• RAND Health Insurance Experiment (1974-1982) – cost-sharing reduced healthcare use by 20-30% with minimal health effects (except low-income individuals).
• Dartmouth Atlas (1990s-present) – regions with higher spending did not have better outcomes; higher spending due to more specialist visits, hospital days, and procedures.
• Patient-Centered Outcomes Research Institute (PCORI) – funded over 1,000 CER studies; findings have changed practice (e.g., spinal fusion vs non-surgical treatment for back pain, medication comparisons for hypertension).

4. International Comparisons and Debated Issues

HSR infrastructures:

Debated issues:

1. Data privacy vs research access: De-identified administrative data and EHRs are highly valuable for HSR, but privacy regulations (HIPAA, GDPR) create barriers. Secure data enclaves, data use agreements, and privacy-preserving methods (federated analysis) balance needs.
2. Implementation science (translating research into practice): HSR identifies effective interventions; implementation science studies how to adopt and sustain them in real-world settings. Gap between evidence and practice remains (10-20 years).
3. Comparative effectiveness research and industry influence: CER may threatens profits for interventions that are more costly but not more effective. Industry funded research may favour their product; independent, publicly funded CER (e.g., PCORI) aims to reduce bias.

5. Summary and Future Trajectories

Summary: HSR examines access, costs, quality, equity, and delivery of care. Key methods include analysis of variation, risk adjustment, and comparative effectiveness. Landmark studies (RAND HIE, Dartmouth Atlas) shaped understanding of cost-sharing and regional variation. Implementation science bridges research-practice gap. Data access and privacy remain challenges.

Emerging trends:

• Learning health systems (continuous feedback from routine care data to improve practice): HSR embedded in healthcare delivery (e.g., VA, Kaiser Permanente).
• Artificial intelligence for health services (predictive models for readmission risk, length of stay, patient complexity).
• Patient-reported outcome measures (PROMs) integrated with administrative data: Enable assessment of outcomes that matter to patients (function, pain, quality of life).
• Rapid cycle evaluation for policy changes (e.g., telemedicine expansion, hospital-at-home programmes).

6. Question-and-Answer Session

Q1: How does health services research differ from clinical research?
A: Clinical research focuses on individual patients and biological mechanisms (e.g., drug efficacy in a trial). HSR focuses on populations, systems, and policies (e.g., does a new payment model reduce readmissions across all hospitals in a state?). Methods and questions differ.

Q2: What is the Dartmouth Atlas?
A: A US research project (1996-present) documenting regional variation in healthcare use, spending, and outcomes. Key finding: regions with more intensive care (more visits, procedures, hospital days) do not have better patient outcomes; supply-sensitive care (hospitalisation for chronic conditions) varies with available beds.

Q3: How is risk adjustment used in provider payment?
A: Medicare Advantage, ACO shared savings, and hospital readmission penalties use risk adjustment to account for patient mix (e.g., HCC scores). More accurate risk adjustment reduces incentives to avoid sick patients and reduces payment errors.

Q4: What is the evidence on telemedicine from HSR?
A: Before 2020, evidence limited. Post-2020, studies show telemedicine reduces no-show rates (10-20% improvement), improves access for mental health, but may worsen disparities for elderly, low-income, and those with limited digital literacy. Effects on total spending mixed (substitution vs add-on visits).

https://www.ahrq.gov/
https://www.pcori.org/
https://www.dartmouthatlas.org/
https://www.rand.org/health-care.html