Date of Award

January 2014

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

William Sledge

Second Advisor

Marc Potenza

Subject Area(s)



Sickle cell disease (SCD) accounts for approximately 100,000 hospital admissions annually. Most admissions are for painful vaso-occlusive crises (VOCs). A small minority of patients with extremely high hospital use (EHHU) account for the majority of inpatient medical costs, and exacerbate issues of distrust in the provider-patient relationship surrounding opioid analgesia, the mainstay of treatment for acute VOCs. This study was designed to generate hypotheses surrounding the causes and consequences of EHHU in adults with SCD, and describe the ability of a coordinated, multidisciplinary intervention, based on the Patient-Centered Medical Home (PCMH) model, to address these phenomena and mitigate EHHU.

We carried out a qualitative study of SCD patients with EHHU (defined as greater than 100 hospital days/year averaged over three years), pre and post-intervention, as well as their family members, their medical providers, and a group of SCD patients with low hospital use (LHU). We conducted descriptive data analysis of hospital utilization and cost trends before and after the intervention. Forty in-depth, one-on-one interviews with 29 individuals took place from 2010 to 2014. We analyzed interview transcripts using narrative summaries and qualitative analysis software. Administrative data was abstracted to describe trends in overall and EHHU-specific trends of SCD healthcare resource utilization and inpatient and outpatient direct costs from 2008 to 2013.

Qualitative analysis yielded a common narrative of thwarted educational, social, and vocational development among SCD patients with EHHU due to repeated hospitalizations from a young age as well as accelerating opioid use. SCD patients with EHHU and matched LHU demonstrated similar experiences of provider distrust surrounding opioid use; however, patients with EHHU lacked the interpersonal and symptom-related strategies to ally with providers demonstrated by patients with LHU. Outside the hospital SCD patients with EHHU were socially isolated, and described symptoms of depression and suicidality. The implementation of the medical home model of care coincided with decreases in overall acute care utilization trends among patients with SCD (13% ED visits, 55% inpatient days, 40% average length of stay (ALOS)), as well as reductions among the SCD patients with EHHU (32% ED visits, 78% inpatient days, 62% ALOS). During this same time period there was a 16% and 30% increase in outpatient visits in SCD patients overall and patients with EHHU, respectively. Follow-up qualitative analysis of patients with EHHU demonstrated a difficulty adapting to new policies, as well as a tradeoff between the benefits of reintegrating into family/society and managing higher levels of pain at home.

In addition to reducing reliance on acute care and augmenting outpatient services, the medical home model may address the causes and consequences of EHHU in SCD in four key dimensions: first, psychiatric and social rehabilitation; second, improved therapeutic relationships through continuity of care; third, monitoring of opioid use with coordinated protocols; fourth, overcoming cultural barriers through community outreach. The PCMH has been shown to reduce costs and improve clinical and patient satisfaction outcomes across a wide range of chronic diseases. This study demonstrates the same potential in a clinically and psychosocially complex, sub-specialty population.