Keeping Care Local: A Workflow to Increase Care Retention in the Community Hospital Setting

Abstract

Community hospitals are a key part of the health care system, offering convenience and high-quality care at lower cost to their patients. Nevertheless, they often struggle to remain solvent in the current traditional business model as many patients choose urban academic tertiary medical centers for routine care. It is particularly difficult for a disproportionate share hospital, such as Lawrence General Hospital (LGH), located close to Boston and its renowned medical centers, that serves a higher proportion of medically complex patients with social determinants and financial barriers to care covered by government insurance programs. When patients leave the local network to obtain care elsewhere, the associated hospital and care clinics lose the downstream revenue and the ability to manage the care for these patients.¹ This drop in revenue for the community institutions means that they are less able to invest in programs and infrastructure that would better serve and attract new patients. As a result, both the patients and the hospital are left in a cycle of poor quality of care at higher cost. Despite the potential benefits of improving referrals within a health system, many hospital networks have been slow to develop comprehensive referral tracking methods.¹

Care Retention as a Strategic Initiative

Care management aims to improve care quality, population health, and reduce health care costs by providing coordinated, non-duplicative, cost-effective care.² Locally enhanced services in primary care continue to be critical in reducing health inequalities as primary care and specialty services target specific components of community-based care.^3,4 It is better for the patient, community, clinics, and hospital to keep health care local as local care improves clinical quality, market share, patient and provider satisfaction, overall efficiency, and shared risk contract management.¹ Local care is also particularly beneficial to the most vulnerable patient communities as disproportionate share hospital settings and local providers can offer better community health to meet the specific needs of their patient population. Moreover, tertiary clinics can then focus on providing the level of care they do best with less demand for community health resources.

Care retention is a central component of population health management, as 68% of patients seeking new specialists make their choice based on the referral of their primary care physician.⁵ That interface between primary care and specialty care has been found to be responsible for major inefficiencies in the delivery of health care, decreased specialist access, and suboptimal patient outcomes, particularly within resource-constrained settings.⁶ A central referral management team approach has been shown to be more effective for patient populations with high rates of homelessness and limited spoken English.⁶ More vulnerable patient populations, such as those in this hospital network, have additional considerations (eg, the ability to participate in setting appointment times for specialty care visits) that a team-based approach is most effective at handling. Additionally, providers are more likely to be part of a managed care program that is able to manage the data and work to improve their practices through evidence-based communication. Data sharing creates a partnership between the population health program and providers within the network to address and serve both community and health plans while maintaining individual autonomy to control care and trust.⁷

Tracking patients through a referral system has been shown to have important downstream value.¹ A team-based approach of high-performing caregivers within the community has been shown to not only improve patient access to specialty care, satisfaction with their care, and the effectiveness of their health care, but also to enhance the ability to manage population health. Today's competitive health care market demands that hospital networks project future requirements accurately to meet the needs of their patient population. Accurate referral management programs provide this intelligence to hospital networks. By analyzing referral retention, hospital networks can address a variety of issues, including clinical quality, service quality concerns, lack of awareness among referring providers, and particular needs of providers across the network.

Our Network and Model

In the challenging and continuously changing health care environment, there have been few studies demonstrating the success of a population health approach to improve care retention within risk-sharing networks. This Physician Hospital Organization (PHO) developed the Central Referral Management program (RMP) that targets provider engagement and communication to improve both population health outcomes and financial goals of risk contracts.

LGH is a 189-bed, 41-bassinet acute care hospital located in a very low-income, largely Hispanic community 30 miles north of Boston. The majority of the patients are of Latino descent. The hospital's gross patient service revenue comes largely from public payers (approximately two thirds); 36% of that comes from low-income payers and nearly 7% of that total is caring for the residually uninsured.⁸ The Greater Lawrence primary service area is comprised of several towns, with approximately 300,000 residents. The hospital is also the only not-for-profit hospital in the region and is an independent community provider. There are 2 hospitals in the primary service area and 3 in the total service area, including LGH. LGH also has numerous competing hospitals within a 20-mile radius and highly-regarded tertiary centers within a 40-mile radius.

The RMP required primary care and specialist physicians in the LGH network's PHO to sign provider agreements to join the program. Specialists were asked to track primary care practice affiliation and schedule procedures accordingly within the PHO network. As many network specialists are on multiple medical staffs, including LGH and the nearest competing hospitals, it was important for the success of the program to provide PHO specialists with feedback about in-network care retention and consequential benefits for their patients and practice. Primary care providers were similarly asked to refer to local specialists and talk with patients about the benefits of staying in the local network. In turn, the RMP alleviated the administrative burden on primary care providers to allow them to focus on quality by having easier and more efficient coordination with specialists' offices.

The RMP worked to improve data collection methods and increase intervention size. An application of Salesforce (Salesforce.com, Inc., San Francisco, CA) shared by the hospital information technology department was designed to gather referral intelligence that was easy to use. Ultimately, the program provided a method to examine 2 PHO referral rates using billed claims data and Salesforce referral data to track patient transition. The first measure tracked was the rate of PHO primary care referrals to PHO specialists for office visits. Prior to the implementation of the program in June 2015, the monthly PHO referral retention rate was extremely low at 77%. The data were broken down to provide a detailed picture of PHO referral rates at the primary care practice and individual physician level. For out of network referrals, detail such as which out of network providers were used and the reported reasons were known. This breakdown allowed the RMP team to go back to the providers with the numbers and trends and collaborate on improvement. Secondly, it gave an opportunity for the team members to educate specialist and primary care physicians on the resources found within the PHO that have been designed with the local population in mind. Thirdly, these 2-way discussions provided an opportunity for the program to gather qualitative feedback from the network providers. Patients and providers shared that they appreciated how the RMP handled transitioning patients and used data to educate and track changes. Specialists, in particular, reported higher satisfaction rates as the RMP worked to send electronic health record notes and results to specialists, had more consistent referral sources, sent reminder letters to patients to help reduce the no-show rate, and focused on overall quality improvement activity. The RMP team provided the hospital, clinics, and providers in its PHO with multipronged support to improve communication and coordination within risk-sharing contracts to retain patients within the network.

The RMP first began tracking its second quantitative measure, in-network outpatient procedures referrals data, using a pilot model at the end of 2013. Outpatient procedures consistently had one of the highest rates of referrals and therefore was determined to be an important marker of the progress of the care retention strategy. In 2013, in-network outpatient procedures referral rates were low at 35%. The intervention included care coordination, provider communication and education, care transition management, and patient tracking and monitoring.

Referral Outcomes

Within 1 year of the full implementation of the RMP, the PHO network primary care to specialist office visit referral retention rate increased from 77% in June 2015, to 91% in December 2016. The average referral rate to in-network specialists for all practices in 2016 is 89%, a marked 12% increase from the initial data collected before the execution of the RMP, with the referral retention rate reaching the highest percentage it has been to date in December. The high percentage is evidence of strong provider engagement resulting from the RMP approach to care management to improve care retention.

The referral rates for outpatient procedures were tracked by facility type. The rate of outpatient procedure referrals within the PHO network increased 4 percentage points from 36% to 40% in the 2016 calendar year, attaining the highest rate over the past 4 years. The percentage of network patients who received outpatient procedures care at the closest competitor hospital dropped 2% from 26% to 24% and those at other community hospitals dropped from 22% to 21%. The percent of patients receiving that care at tertiary clinics decreased 1% from 16% to 15%. Outpatient procedures are an important outcome for the network given patients' desire to stay local and the significant downstream costs to receiving care in other higher cost centers. Therefore, our program succeeded in shifting patients of primary care providers in our network away from competing hospital networks and more expensive tertiary settings to receive high-quality care in our local community-based network.

The increases in total hospital network and outpatient procedure in-network referral rates are significant considering the amount of competition for this community hospital. The fluctuations in referral rates and obstacles faced by the RMP during the period shown can be attributed to the lack of certain specialties in the network, patients having preexisting relationships with out of network specialists, difficulty reporting data manually for feedback, the prevalence of specialist to specialist referrals, shared specialists with competitor networks, and not having enough in-network physicians within certain specialties to meet the volume of patients needing referrals. These are the areas the PHO network will be addressing next to further improve care retention rates in this community hospital setting.

Conclusions

Over the last few years, the hospital network has sought to improve care retention performance in a challenging environment of continuous health care reform. This brief study found that a population health strategy, such as the RMP, with multiple tactics including referral management for data-driven improvements, qualitative discussions about the program with providers, care management that meets the needs of the PHO specialists, primary care physicians, and patients, and provider engagement, improved rates of keeping patients within network. Analyzing and operationalizing available data for providers and administrators is absolutely necessary, particularly in resource-constrained settings where expensive population health platforms and integrated electronic health records are not readily available. Furthermore, data sharing can help change provider behavior in partnership with care management to better retain patients within network under a centralized referral system.

Footnotes

Author Disclosure Statement

Dr. Eamranond, Ms. Sullivan, and Ms. Sedlacek declared no conflicts of interest with respect to the research, authorship, and/or publication of this article. The authors received no financial support for this article.

References

Finley

. The value of understanding patient referral flow. Healthc Financ Manage, 2013; 67:84–89.

Allen

, Des Jardins

, Heider

, et al. Making it local: beacon communities use health information technology to optimize care management. Popul Health Manag, 2014; 17:149–158.

Cook

, Hollar

, Isaac

, Paul

, Amofah

, Shi

. Patient experience in health care medical homes. J Community Health, 2015; 40:1155–1164.

Kumar

, Quigley

, Singh

, Keeping

, Pitman

, Carroll

. Do local enhanced services in primary care improve outcomes? Results from a literature review. Qual Prim Care, 2014; 22:157–169.

Popple

. Referral management pays–in more patients and a better bottom line. MGMA Connex, 2011; 11:25–26.

Bell

, Straus

, Wu

, Chen

, Kushel

. Use of an electronic referral system to improve the outpatient primary care–specialty interface. final report. AHRQ Publication No. 11(12)-0096-EF. Rockville, MD: Agency for Healthcare Research and Quality, 2012:12.

Mechanic

. Managed care and the imperative for a new professional ethic. Health Aff (Millwood), 2000; 19:100–111.

Lawrence General Hospital, Greater Lawrence Family Health Center. Community Health Needs Assessment. August 2016. www.lawrencegeneral.org/uploads/LGH_GLFHC%20CHNA_2016.pdf Accessed January 17, 2017 .