Chapter 6. Care Coordination
National Healthcare Disparities Report, 2011
Health care in the United States is often fragmented. Clinical services are frequently organized around small groups of providers who function autonomously and specialize in specific symptoms or organ systems. Therefore, many patients receive attention only for individual health conditions rather than receiving coordinated care for their overall health. For example, the typical Medicare beneficiary sees two primary care providers and five specialists each year (Bodenheimer, 2008). Communication of important information among providers and between providers and patients may entail delays or inaccuracies or fail to occur at all.
Care coordination is a conscious effort to ensure that all key information needed to make clinical decisions is available to patients and providers. It is defined as the deliberate organization of patient care activities between two or more participants involved in a patient's care to facilitate appropriate delivery of health care services (Shojania, et al., 2007). Care coordination is multidimensional and essential to preventing adverse events, ensuring efficiency, and making care patient centered (Powell-Davies, et al., 2008).
Patients in greatest need of care coordination include those with multiple chronic medical conditions, concurrent care from several health professionals, many medications, and extensive diagnostic workups, or transitions from one care setting to another. Effective care coordination requires well-defined multidisciplinary teamwork based on the principle that all who interact with a patient must work together to ensure the delivery of safe, high-quality care.
In early 2011, the Partnership for Patients was created to improve the quality, safety, and affordability of health care for all Americans. One of the two major goals of this public-private partnership is to heal patients without complications arising. This goal specifically ties to care coordination by seeking to decrease preventable complications during transition from one care setting to another. The objective is to decrease all hospital readmissions by 20% overall by the end of 2013 (compared with 2010).
One example of the Federal Government's efforts to support care coordination is the Health Resources and Services Administration's (HRSA) initiative "Enhancement & Evaluation of Existing Health Information Electronic Network Systems for PLWHA (People Living with HIV/AIDS) in Underserved Communities." Begun in 2007, the initiative funded six demonstration sites throughout the Nation for up to 4 years.i
Another more recent funding opportunity also offered by HRSA is "Systems Linkages and Care Initiative to High Risk Populations Evaluation and Technical Assistance Center." This initiative promotes the development of innovative strategies to successfully integrate different components of the public health system into quality HIV care of hard-to-reach populations who have never been in care. AHRQ intends this chapter to be the leading step in the evolving national discussion on measuring care coordination. Furthermore, AHRQ hopes that this chapter will stimulate productive discussions in the area of care coordination, including development and use of valid, reliable, and feasible quality measures.
Morbidity and Mortality
- Care coordination interventions have been shown to:
- Reduce mortality among patients with heart failure.
- Reduce mortality and dependency among patients with stroke.
- Reduce symptoms among patients with depression and at the end of life.
- Improve glycemic control among patients with diabetes (Shojania, et al., 2007).
- Care coordination interventions have been shown to:
- Reduce hospitalizations among patients with heart failure.
- Reduce readmissions among patients with mental health conditions.
- Be cost-effective when applied to treatment of depression (Shojania, et al., 2007).
The National Strategy for Quality Improvement in Health Careii identified care coordination as one of six national priorities for health care. The vision is health care providers working together to "ensure that the patient gets the care and support he needs and wants, when and how he needs and wants it." While measurement of care coordination is at an early stage in development, key goals include: coordinating transitions of care, reducing hospital readmissions, communicating medication information, and reducing preventable emergency department visits. Measures reported in this chapter are organized around these goals:
- Transitions of care:
- Adequate hospital discharge information.
- Patients who reported that they always received test results (Massachusetts only; measure reported separately for adults and children).
- Hospital readmissions:
- Readmissions for congestive heart failure.
- Medication information:
- Provider asking about medications from other doctors.
- Hospital electronic exchange of medication information.
- Preventable emergency department visits:
- Emergency department visits for asthma.
Transitions of Care
As health care conditions and needs change, patients often need to move from one setting to another. These transitions of care place patients at heightened risk of adverse events. Important information may be lost or miscommunicated as responsibility is delivered to new parties.
Management: Complete Written Discharge Instructions
Effective care coordination begins with ensuring that accurate clinical information is available to support medical decisions by patients and providers. A common transition of care is discharge from the hospital. Giving patients and caregivers self-management support after discharge has been shown to reduce readmissions to the hospital and lower costs (Coleman, et al., 2006).
Discharge from a hospital typically indicates improvement in a patient's condition so that the patient no longer requires inpatient care. It also means that the patient and family must resume responsibility for the patient's daily activities, diet, medications, and other treatments. The patient also needs to visit his or her personal doctor and know what to do if his or her condition deteriorates. Written discharge instructions are critical to help ensure that a patient receives the information needed to stay healthy after leaving the hospital.
Figure 6.1. Hospitalized adult patients with heart failure who were given complete written discharge instructions, by race/ethnicity, 2005-2009
Key: AI/AN = American Indian or Alaska Native.
Source: Centers for Medicare & Medicaid Services, Quality Improvement Organization Program,2005-2009.
Denominator: Hospitalized adult patients with a principal discharge diagnosis of heart failure.
Note: Complete written discharge instructions needed to address all of the following: activity level, diet, discharge medications, followup appointment, weight monitoring, and actions to take if symptoms worsen.
- From 2005 to 2009, the overall percentage of hospitalized adult patients with heart failure who were given complete written discharge instructions improved from 58% to 86% (data not shown).
- Improvements were observed among all racial and ethnic groups (Figure 6.1).
- In 2009, American Indians and Alaska Natives (AI/ANs) were less likely to receive complete written discharge instructions compared with Whites.
- The 2008 top 5 State achievable benchmark was 88%.iii By 2009, Hispanics had attained the 2008 benchmark, with 88% having received written discharge instructions. This benchmark could be attained by most of the racial groups in less than a year. The one exception is AI/ANs, who would require almost 2 years to reach the benchmark.
Also, in the NHQR:
- Statistically significant differences in receipt of written instructions by age and gender group were not observed.
Integration of Information
Patients often seek care from many providers. Medical information generated in different settings may not be sent to a patient's primary care provider. Actively gathering and managing all of a patient's medical information is an important part of care coordination. Tasks include ensuring that patients are informed of important findings such as test results, primary care doctors are informed of care from specialists, and providers within a practice have access to needed information.
No national survey currently gathers information from patients about these aspects of care coordination. To help fill this gap, we examined subnational data-gathering activities and identified the Massachusetts Health Quality Partners (MHQP) Patient Experience Survey as a unique source of this information. MHQP is an independent organization established in 1995. It is a broad-based coalition of physicians, hospitals, health plans, purchasers, consumers, academics, and government agencies working together to improve the quality of health care services in Massachusetts. MHQP has conducted the Patient Experience Survey since 2005.
In 2007 and 2009, MHQP conducted a mail and Internet survey of commercially insured adult and pediatric patients' experiences of care. The survey included patients being served in primary care practices with at least three doctors. iv Several questions related directly to coordination of information across providers and patients. In 2007, the survey was completed by 51,000 adult patients and 20,000 parents of pediatric patients receiving care in more than 400 medical practices in Massachusetts. The response rate was 42%. In 2009, the survey was completed by 56,000 adult patients and 22,000 parents of pediatric patients.
Figure 6.2. Patients who reported that they always received test results, commercially insured adults age 18 and over in primary care practices, by race/ethnicity and education, Massachusetts, 2007 and 2009
Source: Massachusetts Health Quality Partners, Patient Experience Survey, 2007 and 2009.
Note: Respondents limited to patients who received a test in the past year
- In 2009, of adult patients who were sent for a blood test, x ray, or other test by their personal primary care doctor, 69% reported that someone from the doctor's office followed up to give them the test results (data not shown).
- In both 2007 and 2009, Black, Hispanic, and Asian patients were less likely to receive a followup on test results than White patients (Figure 6.2).
- In 2009, there were no statistically significant differences between education groups in reporting whether someone from the doctor's office followed up to give them test results.
Figure 6.3. Parents of patients who reported that they always received test results for their child, commercially insured children under age 18 in primary care practices, by race/ethnicity and parent's education, Massachusetts, 2007 and 2009
Source: Massachusetts Health Quality Partners, Patient Experience Survey, 2007 and 2009.
Note: White, Black, and Asian populations are non-Hispanic. Respondents limited to patients who received a test in the past year. Education refers to the parents. Data were insufficient for this analysis for American Indians and Alaska Natives and Native Hawaiians and Other Pacific Islanders.
- In 2009, of children who were sent for a test, 70% of parents reported that someone from the doctor's office followed up to give the test results (Figure 6.3).
- In both 2007 and 2009, Black, Hispanic, and Asian children were less likely to receive followup on test results than White children.
- In both 2007 and 2009, children whose parents had less than a high school education were less likely to receive followup than children whose parents had any college education.
Outcome: Readmissions for Congestive Heart Failure
After discharge from the hospital for a chronic condition such as congestive heart failure (CHF), many patients will be rehospitalized. Rehospitalization signals a worsened state of illness and may reflect care that is not optimally coordinated. Rehospitalization also has significant cost implications since it is much more resource intensive than outpatient treatment.
Although not all rehospitalizations for CHF can be prevented, the risk of rehospitalization may increase when patients do not follow their discharge instructions. After discharge, patients need to take their medications regularly, adhere to recommendations related to diet and activity, monitor their weight, and look for signs and symptoms that their CHF is not under good control. When patients do not receive written discharge instructions that they understand, they may be less able to follow them. In addition, postdischarge care should be coordinated with the patient's primary care physician. Patients will need to arrange followup visits with their primary care physician, who can adjust medications early to help prevent rehospitalization.
The estimates below are derived from data for 12 States participating in the Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases that include data on race and ethnicity. They are based on all CHF admissions from January 1 to November 30, 2008. Rehospitalizations are defined as admissions to any hospital in that State with any principal or secondary diagnosis of CHF within 30 days of the discharge date of an index CHF admission.
Prior to the 2010 reports, we reported on readmissions with only a principal diagnosis of CHF only, so comparisons with those previous reports would not be appropriate. It is also important to note that the figures reported below are not national estimates. The States in the analysis account for about one-third of all adult discharges for CHF in the Nation and may provide an indication of patterns in CHF readmissions.
Costs of CHF rehospitalizations were also examined. Total hospital charges were converted to costs using HCUP cost-to-charge ratios based on hospital accounting reports from the Centers for Medicare & Medicaid Services. Cost estimates refer to hospital costs and do not include costs of physician services. In these 15 States, the average cost per CHF readmission was just over $13,000, and the total cost was nearly $900 million.
Figure 6.4. Rehospitalization for congestive heart failure, by race/ethnicity and State, 12 States, 2008
Key: API = Asian or Pacific Islander.
Source: Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project, State Inpatient Databases, 2008.
Denominator: Patients hospitalized for congestive heart failure.
Note: Some States had insufficient data for this analysis for Blacks, APIs, and Hispanics.
- The percentage of State-level CHF hospitalizations resulting in rehospitalization for CHF ranged from a low of 14% to a high of 24% (data not shown).
- Aggregating over all States, Blacks and Hispanics had higher rehospitalization rates than Whites (States F, H, J, K, and L, and States H, K, and L, respectively; Figure 6.4).
- Aggregating over all States, no statistically significant differences were observed between APIs and Whites, although APIs had a higher rate in States D and K.
Also, in the NHQR:
- Overall, rehospitalization rates did not vary by age. In some individual States, however, rates were higher among patients age 65 and over compared with patients ages 18-64.
Patients often seek care from many providers, and different providers may prescribe medications for the same patient. Patients are responsible for keeping track of all their medications, but medication information can be confusing, especially for patients on multiple medications. When care is not well coordinated and some providers do not know about all of a patient's medications, patients are at greater risk for adverse events related to drug interactions, overdosing, or underdosing. In addition, providers need to periodically review all of a patient's medications to ensure that they are taking what is needed and only what is needed. Medication reconciliation has been shown to reduce both medication errors and adverse drug events (Whittington & Cohen, 2004).
Management: Provider Asking About Medications From Other Doctors
Medication information generated in different settings may not be sent to a patient's primary care provider. In the absence of communication from other providers, the patient is the primary source of medication information. Actively gathering and managing all of a patient's medical information is an important part of care coordination.
Figure 6.5. People with a usual source of care whose health provider usually asks about prescription medications and treatments from other doctors, by race and activity limitations, United States, 2002-2008
Key: AI/AN = American Indian or Alaska Native, NHOPI = Native Hawaiian or Other Pacific Islander.
Source: Agency for Healthcare Research and Quality, Center for Financing, Access, and Cost Trends, Medical Expenditure Panel Survey, 2002-2008.
Denominator: Civilian noninstitutionalized population who report a usual source of care.
Note: Data for NHOPIs were insufficient for analysis except in 2004 and 2005.
- From 2002 to 2008, the percentage of people with a usual source of care whose health provider usually asked about prescription medications and treatments from other doctors improved from 75% to 81% (data not shown).
- Improvements were observed among all racial and disability groups (Figure 6.5).
- In 2008, the percentage of patients with a usual source of care whose health provider usually asked about prescription medications and treatments from other doctors was significantly lower for Asians than for Whites (74% compared with 81%).
Also, in the NHQR:
- From 2002 to 2008, children's health providers were less likely to ask about medications from other doctors. In all years except 2003, the health providers of older adults were also less likely to ask about medications from other doctors.
Structure: Electronic Exchange of Medication Information
Ideally, information about medications prescribed for a patient by one provider would be available to all providers taking care of that patient. One way to exchange this information efficiently is to build this function into health information technologies. The American Hospital Association recently surveyed hospitals about their use of health information technologies. Questions about whether a hospital electronically exchanged patient information on medication history with other providers were included, and 2,112 hospitals responded.
Data are shown by region and geographic location (urban or rural) of the hospitals. Urban hospitals provide a disproportionate share of care to low-income and minority patients.
Figure 6.6. Hospitals with electronic exchange of patient information on medication history, by region and geographic location, 2009
Key: MSA = metropolitan statistical area.
Source: American Hospital Association Annual Survey Information Technology Supplement, 2009.
- Overall, in 2009, 73% of hospitals electronically exchanged patient information on medication history with other hospitals in their system, 13% exchanged information with hospitals outside their system, and 28% exchanged information with ambulatory providers outside their system (data not shown).
- Hospitals in the West were the most likely to exchange information with ambulatory providers outside their system, followed by hospitals in the Midwest, Northeast, and South (74%, 73%, 73%, and 71%, respectively; Figure 6.6).
- Urban (metropolitan statistical area, or MSA) hospitals were more likely to exchange information with hospitals in their system and with ambulatory providers outside their system than rural (non-MSA) hospitals.
Also, in the NHQR:
- Federal hospitals were most likely to have electronic exchange with hospitals in their system, followed by nonprofit, for-profit (investor-owned), and non-Federal hospitals, respectively.
Preventable Emergency Department Visits
Potentially preventable, high-cost encounters with the medical system occur not only in hospitals, but also in emergency departments (EDs). There were more than 125 million ED encounters in 2008 (AHRQ, 2008). ED crowding, boarding (i.e., holding patients until an inpatient bed is available), and ambulance diversion have become more prevalent and have given rise to increasing concerns about the quality of care delivered in EDs.
Some hospitalizations and ED encounters cannot be avoided, but appropriate ambulatory care can help keep some patients from having to visit an ED or from being hospitalized. Reducing potentially avoidable ED encounters, in particular, holds promise for reducing cost, improving quality, and enhancing efficiency.
Outcome: Emergency Department Visits for Asthma
Asthma is an ambulatory care-sensitive condition, but is one that typically requires those with asthma to avoid environmental conditions that exacerbate their disease, take their medications regularly, and monitor their symptoms. Good primary care can help patients with self-management and treatment adjustments before exacerbations of asthma become severe and require emergent attention. Population-based rates of visits to the ED with and without hospitalization could be lowered with better patient education and outpatient management. For the purposes of added insight, the following analysis distinguishes between visits that resulted in hospitalization and those that did not.
For this analysis, the adult and pediatric asthma measures from the AHRQ Pediatric Quality Indicators software were applied to the 2007 HCUP Nationwide Emergency Department Sample (NEDS). Results related to area income are presented. While other studies have demonstrated higher ED visits for asthma among Blacks and Hispanics, the NEDS does not support analyses of race/ethnicity at this time.
Figure 6.7. Potentially avoidable emergency department encounters for asthma, among adults, by income and geographic location, and among children, by income and geographic location, 2008
Key: ED = emergency department.
Source: Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project, Nationwide Emergency Department Sample, 2008.
Denominator: U.S. population.
Note: Annual rates are adjusted for age and gender.
- Overall, the rate of ED visits for asthma was 513 per 100,000 among adults and 852 per 100,000 for children (Figure 6.7). About 21% of ED visits for asthma among adults led to hospitalization (107 per 100,000), and 79% had other dispositions (406 per 100,000). About 11% of ED visits for asthma among children led to hospitalization (93 per 100,000) and 89% had other dispositions (759 per 100,000).
- Compared with residents of high-income neighborhoods, residents of all other income quartiles had higher rates of both ED visits for asthma that led to hospitalization and ED visits that did not end in hospitalization (this was true for adults as well as children).
- Compared with residents of other geographic locations, noncore residents had the lowest rates of both ED visits for asthma that led to hospitalization and ED visits that did not end in hospitalization (this was true for adults as well as children).
Also, in the NHQR:
- Compared with adults ages 18-44, all older age groups had lower rates of ED visits for asthma, but an increasing percentage that led to hospitalization.
- Children ages 0-4 had the highest overall rates of ED visits for asthma compared with any group presented.
Bodenheimer T. Coordinating care—a perilous journey through the health care system. New Engl J Med 2008;358(10):1064-71.
Coleman EA, Parry C, Chalmers S, et al. The care transitions intervention: results of a randomized controlled trial. Arch Intern Med 166(17):1822-8.
Powell-Davies G, Williams AM, Larsen K, et al. Coordinating primary health care: an analysis of the outcomes of a systematic review. Med J Aust 2008;188(8 Suppl):S65-68.
Shojania K, McDonald K, Wachter R, et al. Closing the quality gap: a critical analysis of quality improvement strategies—Volume 7: Care Coordination. Rockville, MD: Agency for Healthcare Research and Quality; 2007. Available at: http://www.ahrq.gov/clinic/tp/caregaptp.htm.
Weighted national estimates from HCUP Nationwide Emergency Department Sample. Rockville, MD: Agency for Healthcare Research and Quality; 2008. Available at: http://hcupnet.ahrq.gov.
Whittington J, Cohen H. OSF Healthcare's journey in patient safety. Qual Manag Health Care 2004;13(1): 53-59
i For more information, go to: http://hab.hrsa.gov/abouthab/special/underservedcommunities.html.
ii Available at www.healthcare.gov/center/reports/quality03212011a.html.
iii The top 5 States that contributed to the achievable benchmark are Colorado, Delaware, New Hampshire, New Jersey, and Utah.
iv The survey and results are available at http://www.mhqp.org/quality/pes/pesMASumm.asp?nav=031600 .