Continuity of Primary Care and Preventable Hospitalization for Acute Conditions: A Machine Learning-Based Record...

Area of Science:

• Health Services Research
• Public Health
• Epidemiology

Background:

• Potentially preventable hospitalizations (PPH), also known as ambulatory care-sensitive conditions, represent a significant global health concern.
• Continuity of care is hypothesized to reduce PPH, but establishing a causal link requires robust methodologies.

Purpose of the Study:

• To establish a causal relationship between continuity of care and acute PPH.
• To quantify the impact of continuity of care on reducing acute PPH in older adults.

Main Methods:

• Linked data from the 45 and Up Study (New South Wales, Australia) and Medicare claims (2007-2017).
• Employed a double machine learning technique with four algorithms (LASSO, Random Forest, XGBoost, ANN) for robust estimation.
• Controlled for key patient characteristics including age, multimorbidity, cultural diversity, sex, education, psychological status, physical limitation, smoking, and socioeconomic status.

Main Results:

• Among 54,376 participants, 50.8% experienced at least one acute PPH episode.
• Increased continuity of care was associated with a reduction in acute PPH incidence.
• A modest increase in continuity of care (from 45th to 50th percentile) reduced the probability of acute PPH by 9.8% to 23.5% across models.

Conclusions:

• Continuity of care at the primary level is crucial for reducing acute PPH.
• Policies promoting person-centered and integrated care should prioritize initiatives to enhance continuity of care.
• Supporting general practitioners in improving continuity of care is essential for PPH reduction strategies.