Lisa I. Iezzoni, MD, MSc

 Emergency departments (EDs) frequently provide care for nonemergent health conditions outside of usual physician office hours. A nonprofit, fully integrated health insurer/care delivery system that enrolls socioeconomically disadvantaged adults with complex health needs partnered with an ambulance service provider to offer after-hours urgent care by specially trained and equipped paramedics in patients’ residences. The Massachusetts Department of Public Health gave this initiative, the Acute Community Care Program (ACCP), a Special Project Waiver. We report results from its first 2 years of operation.

 We used descriptive methods to analyze administrative claims, financial and enrollment records from the health insurer, information from service logs submitted by ACCP paramedics, and self-reported patient perceptions from telephone surveys of ACCP recipients.

 ACCP averaged only about 1 call per day in its first year, growing to about 2 visits daily in year 2. About 15% to 20% of ACCP patients ultimately were transported to EDs and between 7.2% and 17.1% were hospitalized within 1 day of their ACCP visits. No unexpected deaths occurred within 72 hours of ACCP visits. Paramedics stayed on scene approximately 80 minutes on average. About 70% of patients thought that ACCP spared them an ED visit; 90% or more were willing to receive future ACCP care. Average costs per ACCP visit fell from $844 in year 1 to $537 in year 2 as volumes increased.

 This study using observational data provides preliminary evidence suggesting that ACCP might offer an alternative to EDs for after-hours urgent care. More rigorous evaluation is required to assess ACCP’s effectiveness.

The Acute Community Care Program (ACCP) is a collaboration between a nonprofit, fully integrated health insurer/care delivery system that enrolls socioeconomically disadvantaged adults and an ambulance service provider to offer after-hours urgent care by specially trained and equipped paramedics in patients’ residences. Without ACCP, these patients would typically be sent to emergency departments (EDs). Early results suggest that: 

ACCP appears to reduce ED visits for these urgent care patients. 

At least 90% of patients are willing to receive ACCP care in the future. 

More research is needed to quantify the effects of ACCP on ED use and patients’ experiences.

As managed care programs seek ways to cut costs while improving care quality, reducing emergency department (ED) visits is an obvious target. Most ED visits do not involve medical emergencies, and many patients could safely receive care in less costly and intensive settings.

However, various strategies to decrease ED use among managed care members have yielded mixed results.

 One study of Medicaid managed care found that patients assigned to primary care practices with more than 12 evening hours per week had 20% fewer ED visits than other patients.

 Nonetheless, EDs continue to serve important roles in after-hours care and situations in which primary care clinicians have limited same-day or next-day appointments available.

Here, we describe experiences during the first 2 years of the Acute Community Care Program (ACCP), an innovative program in eastern Massachusetts designed to prevent unnecessary ED visits by providing after-hours nonemergency urgent care to adults with complex care needs.

ACCP represents a collaboration between EasCare, an ambulance service provider, and Commonwealth Care Alliance (CCA), a nonprofit health insurer and integrated healthcare delivery system providing a full range of care, from tertiary care to long-term services and supports. CCA’s care model involves interdisciplinary teams, integrating behavioral and physical health services with extensive community outreach.

All CCA enrollees have Medicaid (MassHealth), the majority also have Medicare, and most care is reimbursed through capitated Medicare and MassHealth payments.

CCA members have complex health and social service needs. Among CCA members 65 years or older and dually eligible for Medicare and MassHealth, approximately 75% are community-dwelling, albeit nursing home—certifiable, according to Medicare criteria; 70% have 4 or more chronic health conditions; and 62% are not primary English speakers. Among enrollees aged 21 to 64 years and dually eligible for Medicare and MassHealth, approximately 70% have behavioral health diagnoses, 15% have current or prior alcohol dependence or substance use disorder, and 7% are homeless, with many others experiencing housing insecurity. CCA enrollees visited EDs 3 times more often than other Massachusetts residents, with average annual ED utilization rates of 810 per 1000 for enrollees 65 years and older and 1564 per 1000 for those aged 21 to 64 years.

 Much of this ED care involved conditions that could have been effectively treated in community settings. Furthermore, ED visits were frequently emotionally uncomfortable for CCA patients, who felt stigmatized by their health conditions or sociodemographic attributes.

 Many ED visits occurred during evening hours, when patients’ primary care clinicians were unavailable for direct patient care.

In July 2014, EasCare and CCA jointly approached the Massachusetts Department of Public Health Office of Emergency Medical Services (OEMS) to obtain a Special Project Waiver to implement ACCP. The waiver allowed pilot testing of ACCP in eastern Massachusetts. ACCP paramedics receive 300 hours of didactic training concerning complex clinical assessments in patients’ homes, interpreting laboratory and other test results, and communicating effectively with on-call clinicians. Trainees’ technical skills and communication competencies are assessed through a multifaceted simulation program that incorporates standardized patients, manikins, direct observation by supervisors and ACCP-trainee paramedic peers, and trainees’ self-assessments.

ACCP started operations in October 2014, serving patients between 6 pm and 2 am. Beginning in late afternoon, CCA members who call their primary care clinician are evaluated by nurses by telephone. Emergency needs, defined as requiring immediate medical attention, are referred to 911 and thus EDs. Urgent needs, defined as nonemergent but requiring attention before the next business day, span a range of acuity and are triaged using a color-coded system to the ACCP paramedic. Paramedics visit the patients with the most severe needs each night, traveling in their specially equipped SUV to patients’ residences.

 To oversee clinical decision making, on-call CCA primary care physicians interact with ACCP paramedics in the field through real-time smartphone communication and shared electronic health records. These CCA physicians are either general internists or family practitioners, experienced in the care of complex patients with a heavy burden of chronic conditions, disability, and sociodemographic disadvantages. In their practices, they sometimes make home visits and therefore appreciate firsthand what the paramedics will confront on scene.

During this ongoing pilot period, Massachusetts’s OEMS monitored ACCP for safety (deaths within 72 hours, postvisit ED transports) and patient satisfaction reports obtained by EasCare. Here, we present these observational data compiled during ACCP implementation. We examine ACCP visits and postvisit service use and deaths, ACCP on-scene activities, and self-reported patient satisfaction. These observational findings provide important preliminary insights into patient outcomes during ACCP’s first 24 months.

The Massachusetts General Hospital/Partners HealthCare Institutional Review Board approved this study.

ACCP Visits, Post-ACCP Hospital Use, and Deaths

CCA received encounter records from EasCare for each ACCP visit. These claims records are linked to CCA member identification numbers, which also link to member enrollment information, claims for all services provided to members, and death dates. We obtained these claims files from CCA and identified numbers of ACCP visits for individual members, ED use, hospital observation unit stays and admissions post ACCP visits, and deaths. Less than 1% of claims did not distinguish between ED visits and observation stays, lumping them into a combined category. We excluded these claims and retained only claims that clearly differentiated the 2 services. We counted claims to quantify the numbers of different visit types and used service dates to identify events within specified time intervals.

EasCare tracks all ACCP visits and patients’ dispositions, recording details about each visit, including arrival time and time on scene; patient’s chief complaint treated by the paramedic, as determined by a senior clinician upon reviewing the record; vital signs; tests performed; medications administered and other interventions; and disposition (eg, treated at home, referred to ED). We obtained this information from EasCare.

Patient-Reported Satisfaction With ACCP Visit

From the outset, CCA and EasCare clinical leaders surveyed ACCP patients to identify patient-reported problems and areas for improvement. Because of the novelty of the program, no existing survey precisely fit their needs. They therefore took questions commonly employed in other patient-reported experience surveys as models and created several new questions specific to ACCP. In addition to a final open-ended question, the survey (available upon request) includes 12 questions on topics such as communication, whether ACCP prevented ED visits, comparing ACCP with ED visits, willingness to use ACCP again, and overall satisfaction with care.

A senior EasCare staff member administered the survey by telephone within 30 days of the ACCP visit. Given the survey’s operational purpose (ie, not research), the numbers of calls made, whether a proxy responded, and refusals to participate were not recorded. After 15 months of conducting the survey, staff recognized that frequent ACCP users were weary of being asked to answer the same survey (ie, some persons were surveyed 2 or 3 times). After January 1, 2016, staff administered the survey only once to patients with ACCP visits. Therefore, before year 2, quarter 2, individual patients could have contributed up to 3 survey responses. EasCare provided us with monthly spreadsheets of survey responses, and we compiled these results. Results here represent all surveys conducted.

To assess the resources required to design and implement ACCP and the costs to operate the program, we examined financial accounting records and interviewed personnel at both EasCare and CCA. We used visit volume and scale by monthly net operating costs retrospectively gathered from the general ledger accounts and payroll records from both EasCare and CCA. EasCare incurs all operating costs of running ACCP and receives a monthly fee from CCA. Given dates of financial information, the cost analysis looked at calendar years 2015 and 2016.

Between October 2014 and September 2016, CCA membership in the region covered by ACCP grew from 8026 (3435 ≥65 years and 4591 aged 21-64 years) to 9223 (4048 and 5175, respectively).

 shows ACCP visits by quarter and services used and deaths within 90 days of the ACCP visit, stratifying by whether patients were 65 years or older or were younger than 65 years. Numbers of ACCP visits have risen over time but not monotonically. In each year, quarter 2 (January-March) includes peak influenza seasons. As assessed by persons visiting EDs within a day of their ACCP encounter, the percentage of ACCP visits that ultimately required ED transports ranged from around 15% to just above 20% over time. Across the 2 years, quarterly percentages of observation stays or hospitalizations within 1 calendar day of the ACCP visit ranged from 1.4% to 5.7% and 7.2% to 17.1%, respectively. Only 3.8% to 11.5% visited EDs within 2 to 3 days after their ACCP service. However, more than one-third of patients in each quarter had another ACCP visit within 90 days, and within 90 days, up to 35% to 50% of ACCP recipients had either observation stays or hospitalizations. Death rates within 90 days varied by quarter, with the highest rate (25%) in year 2, quarter 2 (Table 1). As determined by OEMS reviews, no unexpected deaths occurred within 72 hours of an ACCP visit.

Across the 2 years, various nonspecific conditions, including dizziness, fatigue, pain, headache, and weakness, accounted for one-fourth to one-third of ACCP complaints (

). In year 1, urinary or respiratory conditions were the precipitating complaint in about one-third of cases, and in year 2, respiratory or abdominal conditions accounted for almost one-fourth of ACCP visits. A wide variety of other complaints also occurred. On average, ACCP paramedics spent about 80 minutes on scene. Approximately one-fifth of patients had an intravenous line inserted, one-fourth received medications, and just over one-third had real-time blood tests performed and interpreted using a portable device.

Patients generally reported high satisfaction with their ACCP experiences. 

 shows results from questions with fixed response categories. The survey closed with an open-ended statement: “We would like to hear anything else about the recent visit of the paramedic to your home.” The number of persons surveyed decreased substantially in year 2, for the reasons previously described. Importantly, approximately 70% of respondents thought that the ACCP visit saved them a trip to the ED, as suggested by these quotations from responses to the open-ended question (note that respondents all referred to their paramedic by his first name; all current ACCP paramedics are male):

[Paramedic’s first name] was just great, and this visit certainly saved me a trip to the emergency room. The only thing I didn’t like was the needle, but the medication made me feel much better. It’s great knowing you guys have this service available. Thank you!

I was so happy that I didn’t have to go to the emergency department again. [Paramedic’s first name] was able to evaluate and treat me at home, which was just fantastic. I run the risk of getting sicker in the hospital, let alone all of the wasted time. … Thank you!

I was shocked to see exactly how much the paramedic could actually do in my home, from the [electrocardiogram] and vital signs to the amazing blood testing machine. Thank you! Definitely saved me a trip to the ER.

I can’t stand going to the hospital because I get put in the hall and everyone walks past me! Having [paramedic’s first name] come to my home was fantastic, and it certainly saved me a trip to the emergency room. Thank you!

Even when the paramedic care could not resolve the problem on scene and the patient required ED transport, the open-ended responses suggested that the paramedic’s involvement facilitated the hand-off to ED clinicians and made that process easier for the patient:

A million thanks to [paramedic’s first name] who helped get my mom to the hospital quickly. She was very sick and [paramedic] helped my entire family with the transition to the hospital. She is doing much better, thank you!

Unfortunately, [patient’s name] had to go to the hospital where she was admitted for 3 days. She is doing much better, and we are thankful that [paramedic’s first name] was able to coordinate the transfer to the hospital so seamlessly. Thank you!

In most quarters, at least 90% of patients strongly agreed that they were satisfied with the quality of the paramedic care, and at least 90% strongly agreed they would be willing to receive ACCP care in the future. The following quotations exemplify the views of survey respondents:

This is the best thing to happen to [my uncle] in a long, long time. Your service has prevented at least four trips to the hospital in the month of November alone. Thank you very much!

This is a great service. The guy [paramedic] was very professional, and I felt very relaxed after the visit. I felt like I could talk comfortably with the paramedic, and I didn’t have to go to the hospital. Thank you!

They [the paramedics] are fantastic guys! You need to advertise this program to make it available to more folks who would benefit from it. Thank you!

ACCP start-up costs were approximately $95,000 and included labor costs (training and administration), capital costs (vehicle and equipment), and other costs (ie, licensing, medical supply inventory). The annual operating costs for 2015 and 2016 were $350,000 and $344,000, respectively, and costs per patient encounter in each year were $844 and $537. Because volume increased more than 50% between 2015 and 2016 and costs slightly decreased, the cost per encounter decreased by 36%, suggesting that EasCare has achieved some economies of scale. As volume continues to increase, it is likely that the cost per encounter will decrease until there is a need to increase capacity and incur more costs of a start-up/overhead nature.

ACCP aimed to provide a clinically effective and patient-centered alternative to ED care for patients with urgent health needs during evening hours. From these preliminary data, it appears that ACCP largely succeeded, because less than one-fifth of persons with ACCP visits ended up going to EDs within the same calendar day as their ACCP encounter. Anecdotally, most same-day ED visits that occurred after ACCP visits were initiated by the paramedic in collaboration with the on-call physician, who determined that the patient’s condition warranted more in-depth evaluation and treatment than could be safely provided in the home setting. Of the patients who were treated at home, in most calendar quarters, less than one-tenth visited EDs within 3 days, suggesting that the immediate urgent issue had been effectively addressed by the ACCP visit. Recipients seemed pleased with their ACCP services, with most reporting their belief that ACCP had prevented them from needing ED care.

From these observational data, it appears that ACCP recipients have a heavy burden of disease, as suggested by substantial fractions requiring both ACCP visits and hospitalizations within 90 days. Early in this pilot program, recognizing the significant symptom burden faced by many CCA members with multimorbid chronic conditions and approaching the end of life, CCA and EasCare clinical leaders developed special workflows and training for ACCP paramedics to care for patients with end-of-life symptoms. This involved an expanded formulary, rapid response, and direct interactions between paramedics and palliative care—trained physicians. The relatively high death rates within 90 days of the ACCP encounter likely reflect this end-of-life care initiative, as well as the overall medical complexity of CCA enrollees.

Although 8000 to 9000 CCA members lived in the ACCP catchment area, during its first 2 years of operation, the program served few individual patients and provided relatively small numbers of visits: on average, roughly 1 visit per day in year 1 and slightly under 2 visits per day in year 2. ACCP needed to build awareness and trust with CCA primary care practitioners that the program offered a safe and appealing alternative for after-hours urgent care management compared with reflexively sending patients to EDs. Equally importantly, the paramedics and CCA clinicians leading ACCP needed to learn from their early implementation experiences to improve the program and extend its reach. From its start, ACCP leaders have met every month in a morbidity and mortality (M&M) rounds format, reviewing each patient who required ED transport, was admitted to hospital within 72 hours, or raised special interest. These M&M meetings led to improvements in ACCP practices and program enhancements. Since the ACCP’s inception, lessons learned include the following:

Patients referred for ACCP have wide-ranging acuity. We had to refine our triage practices to prioritize by medical need, making the highest-value use of the paramedics’ time. We developed a color-coded triage system, honing it with experience, to identify the most acutely ill patients, whom paramedics visit first.

Demand for ACCP is highest in the early evening, falling off substantially after midnight. Although our Special Project Waiver provisions currently prevent us from changing hours of operation, we shall staff as required to accommodate this timing of patient demand in the future.

Ensuring seamless communication among all participants is critical. This includes tracking the ACCP SUV’s location and facilitating calls between paramedics in the field and CCA clinicians. Having online access to CCA’s electronic health records allows paramedics to learn about patients before the visit and gives CCA clinicians immediate access to paramedics’ clinical observations, notes, and test results.

Many CCA members with physical disabilities and/or mental illness have experienced discomfort and stigma at EDs and are therefore reluctant to visit them, even with severe symptoms. These individuals frequently wait at home until their illness is so advanced that they risk worse outcomes. Although we originally conceived ACCP to reduce unnecessary ED use, for these members we believe ACCP allows us to intervene earlier, thereby avoiding complications and more severe illness or debility.

Validating these lessons will require further study.

Articles

The Acute Community Care Program uses paramedics to provide in-home urgent care after regular business hours, aiming to prevent unnecessary emergency department visits.
Check out our website’s new table/figure pop-up feature! Click on the name of a table or figure in the text to see it in your browser.

Early Experiences With the Acute Community Care Program in Eastern Massachusetts

Hybrid approaches to case finding show promise as a method to increase the success of care management programs (CMPs). A large healthcare system implemented a hybrid approach in which clinicians review algorithm-generated lists of potential high-risk patients within their practice and select the patients most appropriate for the CMP.  We sought to understand the criteria clinicians used when selecting patients.

 We conducted 20 semi-structured interviews with a convenience sample of primary care clinicians and their care managers from a diverse set of practices.

 Two independent reviewers thematically coded interview responses through an iterative process.

 In addition to clinical need (eg, disease severity or multiple comorbidities), interviewees considered a number of nonclinical patient characteristics that they felt placed patients at risk for suboptimal healthcare utilization and poor adherence. These include patients’ predisposing (eg, health literacy or navigation challenges, physical vulnerabilities, insight regarding their health) and enabling characteristics (eg, social and home environment, coping skills, financial resources). Interviewees also considered patients’ existing linkages with the health system and whether other clinicians were already meeting a patient’s care needs.

In selecting patients for a CMP, clinicians considered patient characteristics that are not easily captured in standard clinical and administrative data. A hybrid approach to case finding concentrates clinician review on patients who meet standard clinical and healthcare utilization criteria, and allows clinicians to incorporate knowledge of patients’ predisposing and enabling characteristics that are not readily maintained in clinical data.

	This study identified the factors that primary care clinicians and care managers considered as they reviewed algorithm-generated patient lists to identify a final set of patients appropriate for care management. 

		In addition to the complexity of patients’ medical conditions, clinicians considered patients’ health literacy, physical vulnerabilities, social and home environment, and coping skills when selecting patients for care management. 

		Clinicians also considered the extent to which patients’ care needs were met by strong existing linkages to care providers outside of the primary care practice.

Healthcare delivery systems are increasingly entering into risk contracts which require healthcare providers to manage the utilization of defined patient populations. Care management programs (CMPs), particularly those based in primary care practices, have become a common strategy to meet the cost containment and quality targets in these contracts. Due to the concentration of costs in persons with multiple chronic conditions, these persons are often selected for care management interventions that aim to enhance the coordination of health services and improve patients’ ability to navigate the health system.

 Appropriate case finding, or the identification of patients most likely to benefit from care management, is key to maximizing the cost-effectiveness and sustainability of CMPs.

Many CMPs rely on quantitative approaches to high-risk patient identification, such as risk prediction models based on prior healthcare utilization and clinical conditions. These approaches typically rely on administrative data, such as billing claims, to predict future healthcare utilization, but are often unable to include more nuanced patient characteristics, including socioeconomic and psychosocial circumstances.

 Claims-based approaches often fall short of identifying the patients who are most likely to engage in and benefit from care management because they are unable to detect factors such as patient willingness to participate or barriers to participation.

 Moreover, use of data-driven algorithms often introduces a time lag between when patients need help and when they are identified for the program.

An alternative strategy is to rely on direct referrals from primary care physicians (PCPs) to identify appropriate patients. This strategy is based on the premise that PCPs know their patients best and can consider patient characteristics not routinely available in administrative data or medical records.

 However, PCPs vary widely in their referral patterns for such programs,

 and concerns have been raised about underselection of patients with poor existing linkages with the healthcare system.

 Finally, physicians with a poor understanding of care management goals or approaches may refer inappropriate patients to these programs.

Hybrid approaches that combine predictive modeling and clinician input have been suggested as a way to overcome the biases and limitations of the individual approaches.

 Prior research has explored the extent to which primary care clinician case finding agrees with predictive modeling case finding,

 but has not examined the added benefit of clinical review of a list of potentially high-risk patients generated from a predictive algorithm. We investigated the contribution of clinician input in the selection of high-risk patients within an active primary care-based CMP using semi-structured interviews of PCPs and their nurse care managers (CM). We were particularly interested in the criteria and patient factors clinicians used in practice when determining which potential high-risk patients within their panel should be targeted for care management. Specifically, we examined the extent to which clinicians considered: 1) patient characteristics that are and are not routinely available in clinical and billing data; and 2) existing patient linkages to other providers within the healthcare system when reviewing lists of potential high-risk patients, and identifying patients appropriate for practice-based care management. We hypothesized that clinician involvement enhanced the case selection process by allowing the inclusion of information about the patients not readily available in administrative data.

Description of Care Management Program and Patient Selection

The CMP is a primary care-based, nurse-led longitudinal care program designed to help high-risk patients (ie, chronically ill, medically complex patients) better manage their health and healthcare utilization. Approximately 400,000 Medicare and commercially insured patients 18 years or older were eligible for the CMP in 2013. CMP patient selection starts with the use of a billing claims-based predictive algorithm to identify an initial list of high-risk patients. As a first step, chronic conditions, utilization patterns and a predictive risk score for future total medical expense are generated using the prior year of paid claims using Optum ImpactPro version 6.0 software (Optum Inc, Eden Prairie, Minnesota). Combinations of specific chronic conditions and patterns of healthcare utilization, as well as overall risk score, are then used to identify the initial set of high-risk patients using an algorithm developed with input from an advisory committee of operational and clinical experts within the healthcare organization. Approximately 5% of patients are identified as potentially high risk based on the claims-based algorithm. These patients are then sorted into PCP-specific lists for clinical review. Lists provided to PCPs included the chronic conditions, utilization, and predictive risk score for each patient, as well as patient name and medical number so clinicians could easily recognize their patients. Each PCP, together with their CM, was responsible for reviewing between 1 and 39 potentially high-risk patients for 2 insurance-risk contracts in 2013, the second year of the program. Overall, clinicians selected approximately 61% of patients on the algorithm-generated list for the CMP. Clinicians were asked to classify patients on their list according to whether they were appropriate for the CMP, or they were not appropriate either because they were not considered high-risk or because the patient’s care needs were already being met. PCPs were given considerable latitude in how they classified and selected their patients for the CMP.

Using semi-structured interviews of PCP and CM dyads, we elicited and analyzed the patient characteristics that clinicians considered when selecting patients for CMP. This project was undertaken as a quality improvement initiative, and as such, was not formally supervised by Partners Healthcare’s Institutional Review Board per their policies.

We identified a convenience sample of 20 PCP and CM dyads from 18 ambulatory practices within an integrated healthcare system consisting of 270 primary care practices located in eastern Massachusetts. Physician organization leaders identified participants with the intention of maximizing variation in practice characteristics, such as the level of affiliation with the delivery system, location, and type (

). The CMP director then confirmed willingness to participate.

A physician researcher (CH) and project manager (VH) conducted all interviews, either by phone or in person, at the respective practices. Interviews lasted 45 to 60 minutes and were audio recorded with consent from all participants. All interviews were completed between May and September 2013. A research assistant (LM) transcribed the audiotapes verbatim.

Interviews followed a semi-structured, open-ended interview guide developed through an iterative process by the research team, and designed to obtain general programmatic quality improvement information, as well as information on the characteristics of patients selected for care management. Relevant interview questions included (see the 

] for the complete interview guide): 1) “How do you define a ‘high-risk’ patient and what criteria do you use to determine if a patient is ‘high-risk’?” 2) “What types of patients did you remove because they were ‘not high-risk’?” 3) “What types of patients did you remove because they have their care needs met elsewhere?”

The Andersen Model of Health Services Utilization

 provided a basic framework for organizing interview responses according to whether they reflected patient predisposition to use services (predisposing factors), characteristics that enable or impede healthcare use (enabling factors), and patient need for healthcare (need). We coded and analyzed interview responses using QSR International’s software, NVivo version 9, a qualitative data management and analysis software tool (QSR International Pty Ltd, Doncaster, Australia). In the first level of analysis, an interviewer and a research assistant (VH and LM) independently read and thematically coded all interview transcripts—a process that allows coders to organize and evaluate similar quotations by themes.

 Through an iterative process, content-related minor themes or factors were collapsed into major factors, then minor and major factors were discussed until a mutual agreement was reached. We met with clinician researchers and operational leaders to solicit feedback, present disconfirming evidence, and validate the coding scheme.

 After coding the data into this framework, we retained and summarized the major and minor predisposing, enabling, and need factors mentioned in at least 20% of interviews. Exploratory analyses examined variations in themes by practice affiliation and model.

Although limited to a single large integrated delivery system, our convenience sample of interviewed practices represents a broad range of practice affiliations, locations, and models (Table 1). The major and minor factors that PCPs and CMs considered in deciding which patients would be appropriate for care management within their primary care practice are summarized in 

 and exemplary quotations are provided in 

Most interviewees (85%) considered a patient’s ability to comprehend medical instructions and navigate the healthcare system when determining whether patients were appropriate for the CMP. Specifically, more than half (55%) of interviewed PCPs and CMs reported that they considered patient need for education about their diseases and appropriate management as they selected high-risk patients. Almost half of interviewees suggested that activated patients (ie, patients who are good “advocates for themselves,” “well-versed and knowledgeable about what they need,” and “can take care of themselves”) may not benefit from the additional resources provided by care management. On the other hand, interviewees felt that patients with linguistic or cultural barriers, as well as individuals with impaired executive function, would be more likely to benefit from the services rendered by CMs.

More than half (60%) of interviewees considered whether their patients had physical vulnerabilities, including a high risk of falling, frailty, or advanced age, when determining appropriateness for CMP. Clinicians reported that, in some cases, they would select patients with relatively stable functional status if they had physical vulnerabilities because they felt care management might help facilitate patient care in the event of decompensation.

Patient insight regarding his/her health. 

Just over half (55%) of interviewees considered the extent to which patients accurately perceived their health status and their need for assistance. This included elderly patients who were reluctant to lose their independence, younger patients in denial of their medical conditions, or patients who minimize their health needs. Interviewees felt these types of patients lacked an accurate perception of their limitations and needs, potentially leading them to refuse certain types of medical care or assistance. Interviewees stated that they would include these patients in the program, but that it might be challenging to engage and impact these types of patients.

 All interviews mentioned that they considered their patients’ social and home environment when assessing appropriateness for the CMP. Interviewees observed that relatives often serve as patients’ advocates and caretakers, helping patients navigate the health system, monitoring medication regimens, and noting changes in the patient’s health status. A large majority (90%) of interviews considered the absence of patients’ familial and social supports as reasons to include patients in the CMP. Instability in a patient’s home environment was mentioned as a consideration in a large majority (75%) of interviews, with issues ranging from addiction and verbal abuse to hoarding, as well as spousal illness, hospitalization, or recent death. Some (30%) interviewees mentioned the social isolation of living alone as an added risk factor for elderly and frail chronically ill patients. Finally, lack of access to transportation, home delivery of food, or other community resources were also considerations for CMP inclusion (30%).

Over half (65%) of interviewees noted that they would select patients with underlying mild psychological concerns, such as health anxiety, who often seek reassurance from medical professionals, whether through frequent phone calls or urgent care visits. Such patients may benefit from a CM, who could act as a point-person to help them manage anxiety related to their medical conditions without consuming limited office resources.

A minority (35%) of interviewees mentioned socioeconomic instability (ie, chronic unemployment or unstable housing), and its effect on a patient’s ability to manage his or her chronic illness(es). Medication adherence issues related to financial instability were also considered.

All interviewees considered their patients’ chronic conditions, including the severity or stage of the patient’s disease, and the patient’s complexity or combination of conditions. Interviewees considered whether complicated medication regimens placed their patients at higher risk for decline, complications, and hospitalization. Interviewees specifically highlighted patients with co-occurring medical and psychiatric disorders, such as depression, which they felt further complicated patient care and the patient’s ability to manage his or her medical condition(s).

In identifying patients appropriate for CMP, interviewees considered factors in isolation, but also described the interplay between patient factors, and how the combination of these factors either supported or impeded a patient’s ability to effectively manage his or her health.

Once clinicians determined whether patients were high-risk, interviewees then considered whether the CMP would be the right program for the patient. Interviewees considered whether the patient was already connected with other clinicians, and if the resources available through the CMP would address the patient’s needs. Specifically, they reported considering existing relationships with specialist physicians (eg, psychiatrists, cardiologists) or specialty clinics (eg, transplant, oncology), private CMs, or institutional care providers (eg, group homes, assisted living). Decisions regarding whether care needs were met relied on clinicians’ perceptions of whether there were additional resources or services the CMP could provide, such as transitions to end-of-life care. Moreover, interviewees’ understanding of the CMP also impacted their decisions about whether to include certain subgroups of patients. For instance, some interviewees excluded patients whose primary diagnosis was a psychiatric condition or substance abuse because they felt the CMP did not yet have the resources to meet these patients’ needs.

Our study evaluated the factors clinicians considered important in determining whether potentially high-risk patients within their practice would benefit from a practice-based CMP. Similar to others, we found that clinicians consider patients’ predisposing factors, including health literacy and patient self-efficacy, as well as enabling factors, such as home and family supports, when identifying patients appropriate for care management.

 We found that these factors remained important, even in our hybrid approach to case finding, in which clinician decision making is limited to a set of risk contract patients with prior high-cost care patterns and complex chronic illness. Additionally, clinicians in our investigation considered whether existing medical and community resources were already addressing patient needs and whether CMP would enhance or interfere with existing patient care networks.

Many of the factors clinicians considered in their decision making are not routinely captured in clinical or billing data, and are therefore unavailable for case finding.

 For example, patient home environment and social support, as well as patient health literacy, were factors considered by a large majority of clinicians when identifying patients appropriate for CMP. This information is not contained in billing records and is not routinely captured in electronic health records. 

Previous research found only a small subgroup of patients concordantly selected by PCPs and predictive modeling for care management.

 Compared with PCP-based case selection, a predictive modeling approach tends to identify patients who are at a higher risk of healthcare utilization and have a higher morbidity burden,

 but are less likely to be helped by care management, either because their utilization is not preventable or they are unwilling or unable to participate.

The hybrid approach in this study builds on these findings. Starting with a predictive model to identify a subset of patients that may be appropriate for CMP effectively limits the review burden on clinicians, and focuses the clinical review on the subset of patients who are enrolled in a risk contract and have chronic conditions and patterns of prior healthcare utilization that may be amenable to care management.

More systematic capture of clinical and demographic data may become standard in newer electronic medical records or health information exchanges,

 but many of the nonclinical, nuanced factors clinicians reported considering are unlikely to become standard elements of clinical documentation, either because of privacy concerns or because of their limited relevance to standard clinical care.

Incorporating clinical input in patient selection addresses some of the limitations associated with the use of claims or medical record-based algorithms for case finding, but might also introduce new biases. Physicians and practices may vary in their threshold for CMP referral or how they weigh specific criteria. For example, some practices may have existing programs or supports, such as special programs for patients with heart failure or diabetes, or close ties with cancer centers that might impact their patient selection. Variations in decision making may also be related to differences in knowledge or acceptance of team approaches to patient care. For instance, we found that some clinicians were unsure whether the CMP would have the necessary resources for patients with co-occurring mental health and substance use. More training and education as to the types of patients who would be a good fit for CMP may improve the patient selection process.

Clinicians consider a number of patient predisposing and enabling characteristics not typically available in clinical data systems when selecting high-risk patients for care management.

Clinician Considerations When Selecting High-Risk Patients for Care Management

: To determine if the type of insurance arrangement, specifically health maintenance organization (HMO) vs fee-for-service (FFS), affects cancer outcomes for Medicare beneficiaries with disabilities.

: We used the Surveillance, Epidemiology, and End Resultsâ€“Medicare linked dataset to identify beneficiaries older and younger than 65 years entitled to Medicare benefits because of disability (Social Security Disability Insurance) who subsequently were diagnosed as having breast cancer (n = 6839) or nonâ€“small cell lung cancer (n = 10,229) from 1988 through 1999. We categorized persons according to Medicare insurance arrangement (continuous FFS, continuous HMO, or mixed FFS/HMO) during the periods 12 months before diagnosis and 6 months after diagnosis. Using a retrospective cohort design, we examined stage at diagnosis, cancer-directed treatments, and survival. 

: Women with continuous HMO insurance had earlier-stage breast cancer diagnosis (adjusted relative risk, 0.77; 95% confidence interval, 0.65-0.91) and were more likely to receive radiation therapy following breast-conserving surgery (adjusted relative risk, 1.11; 95% confidence interval, 1.03-1.19). Women having continuous HMO insurance had better breast cancer survival, primarily resulting from earlier-stage diagnosis. Among persons with nonâ€“small cell lung cancer, those having mixed FFS/HMO insurance were more likely to receive definitive surgery for early-stage disease (adjusted odds ratio, 1.23; 95% confidence interval, 1.02-1.49) and to have better overall survival but not significantly better lung cancer survival.

: When diagnosed as having breast cancer or nonâ€“small cell lung cancer, some Medicare beneficiaries with disabilities fare better with managed care compared with FFS insurance plans.

It is unknown whether the type of Medicare insurance arrangement, specifically health maintenance organization (HMO) vs fee-for-service (FFS), affects cancer outcomes for Medicare beneficiaries with disabilities, a vulnerable population. We found that Medicare beneficiaries with disabilities had better breast cancer outcomes if they were continuously enrolled in HMOs. 

The HMO vs FFS insurance arrangement had little effect on the care and outcomes of lung cancer among Medicare beneficiaries with disabilities.

In 2005, 1 in 6 Medicare beneficiaries (6.5 million persons) was entitled to receive Medicare benefits because of disability.

 Medicare beneficiaries with disabilities seem to be at risk for increased cancer mortality,

 even when diagnosed at the same stage or at an earlier stage, compared with persons without disabilities.

 In addition, persons with disabilities may receive different cancer treatment than persons without disabilities.

Medicare beneficiaries may receive care within a health maintenance organization (HMO) or within the fee-for-service (FFS) sector. It is uncertain whether the type of health insurance arrangement (HMO vs FFS) affects the quality of care for Medicare beneficiaries with disabilities.

 beneficiaries with disabilities were less satisfied with managed care plan performance and were more likely to disenroll. However, other evidence indicates that beneficiaries with disabilities receiving care in HMO plans perceive better access to primary care services and greater affordability of health services than those with traditional Medicare coverage.

 Medicare beneficiaries who are enrolled in HMO plans are more likely to undergo cancer screening,

generally have cancers diagnosed at an earlier stage,

The Surveillance, Epidemiology, and End Results (SEER) cancer registries merged with Medicare data have been used to study health disparities among persons with disabilities.

 We used merged SEER-Medicare data to evaluate whether the type of Medicare insurance arrangement (HMO or FFS) affects cancer outcomes for Medicare beneficiaries with disabilities. We studied 2 high-volume cancers, breast cancer and lung cancer. We chose breast cancer because it is amenable to screening and because experiences of these patients would capture potential disparities in early detection and treatment. In contrast, screening is not recommended to detect lung cancer, although surgery and radiation treatment may improve survival.

We used the SEER-Medicare dataset, which links SEER registry information to Medicare claims data.

SEER consists of 11 population-based tumor registries representing approximately 14% of the US population.

 SEER collects patient information on demographic characteristics, primary tumor site, stage at diagnosis, tumor size, histologic findings, tumor grade, hormone receptor status, initial course of treatment, and vital status. SEER tracks vital status annually, and death certificates are used to capture underlying cause of death.

We identified all persons 21 years and older within the SEER-Medicare dataset having a pathologically confirmed first diagnosis of breast cancer (n = 62,315) or non&#8211;small cell lung cancer (n = 55,770) from January 1, 1988, through December 31, 1999. We then restricted our sample to those persons who originally qualified for Medicare coverage because of Social Security Disability Insurance (6839 with breast cancer and 10,229 with lung cancer). Therefore, our sample includes persons younger than 65 years who have Social Security Disability Insurance and persons 65 years and older whose Social Security Disability Insurance has been automatically converted to Old Age Survivors Insurance. As described elsewhere, we focused exclusively on individuals with Medicare when newly diagnosed with cancer, eliminating persons disabled by cancer.

Medicare data indicate for each month whether persons were eligible for Part A and Part B and whether they were enrolled in an HMO insurance arrangement. To examine possible effects of insurance structure on early detection of cancer, we constructed a variable that defined insurance arrangement before diagnosis. We determined the type of insurance arrangement during the month of diagnosis and the previous 12 months. For this period, we assigned cases to 1 of the following 3 insurance categories: FFS for persons continuously enrolled in traditional FFS Medicare, HMO for persons continuously enrolled in HMO plans, and mixed FFS/HMO for persons enrolled in both FFS and HMO plans during the period. To examine treatments following diagnosis, we designated similar postdiagnosis insurance variables for persons continuously eligible for Medicare Part A and Part B during the month of diagnosis and the 6 months after diagnosis (or until death if survival was <6 months). For analyses of survival, we assigned cases to similar insurance categories covering the prediagnosis and postdiagnosis periods combined.

SEER determines stage at diagnosis based on a combination of pathologic surgical and clinical assessments available within 2 months of diagnosis.

 Stage at diagnosis is recorded using the American Joint Committee on Cancer (AJCC) staging system (stage 0, I, II, III, or IV). In our analysis of stage at diagnosis, we excluded persons whose cancers were unstaged (346 with breast cancer and 1182 with lung cancer).

We examined survival (all-cause and cancer-specific mortality) following diagnosis. We measured survival time as the number of days from diagnosis until death or December 31, 2001, whichever came first. For all-cause mortality analyses, we censored observations of persons alive at the end of followup. We studied breast cancer&#8211;specific and lung cancer&#8211;specific deaths, censoring observations of subjects alive at the end of follow-up or who died from causes other than breast cancer or lung cancer.

 lists the characteristics of our sample. Persons having HMO insurance at diagnosis tended to be older and more likely to reside in census tracts having higher median household income. Consistent with market penetration of HMOs, patients having HMO insurance were more likely to originate from SEER registries in California and in Seattle, Washington. Among patients with breast cancer, those having FFS insurance were more likely to have missing information on tumor grade and on estrogen receptor status and progesterone receptor status.

Insurance type was sometimes statistically significantly associated with cancer-directed treatments for breast cancer and for lung cancer (

). Women having HMO insurance were more likely to receive radiation therapy following breast-conserving surgery. There was a statistically nonsignificant trend for women having HMO insurance to receive breast-conserving surgery rather than mastectomy. Insurance status had no effect on the likelihood of axillary lymph node dissection. Among persons diagnosed as having non&#8211;small cell lung cancer, those having mixed FFS/HMO insurance were more likely to receive definitive surgery for early-stage tumors.

Results were similar when subjects having missing information on tumor characteristics were excluded from the sample and multivariate analysis was repeated (data not shown). To examine the possibility that the effects of HMO insurance varied over time, we repeated our analyses separately for 2 periods, persons diagnosed from 1989 through 1994 and persons diagnosed from 1995 through 1999. Results were similar for both periods. We calculated an attributable fraction for the 831 breast cancer deaths observed in the cohort (769 with FFS insurance and 62 with HMO insurance). We estimated that 17% (144 deaths) would theoretically have been prevented if patients having FFS insurance had a mortality experience comparable to that of patients having HMO insurance.

associations for patients with disabilities diagnosed as having lung cancer (eg, no association with stage). 

Among Medicare beneficiaries, patients belonging to HMOs are more likely to be screened for cancer

 and are more likely to have cancers diagnosed at an earlier stage.

This may in part explain our finding of earlier breast cancer diagnosis and improved survival. Higher rates of cancer screening within HMOs may result from greater emphasis on delivery of preventive care32 and from increased focus on primary care rather than on subspecialty care.

 In some studies, beneficiaries with disabilities in HMOs perceived better access to primary care services8 and were more likely to undergo cancer screening tests.

 Greater use of preventive services in HMOs may be the result in part of favorable selection in which healthier patients are differentially enrolled in HMOs.

There was some evidence that Medicare beneficiaries with disabilities enrolled in HMOs were more frequently treated with breast-conserving surgery as shown by a statistically significant odds ratio in the unadjusted model but not in the adjusted model. These HMO enrollees were more often treated with radiation therapy following breast-conserving surgery (the treatment combination recommended by National Institutes of Health consensus panels) and had better breast cancer survival. Persons having HMO insurance were more likely to have tumor grade and hormone receptor status documented for their cancers. Findings from previous studies suggest that in general Medicare beneficiaries belonging to HMOs are more likely to undergo breast-conserving surgery,

 to receive adjuvant radiation therapy following breast-conserving surgery,

 and to have improved breast cancer survival.

Our study extends these findings to Medicare beneficiaries with disabilities.

The reasons for treatment differences among patients having HMO insurance vs FFS insurance could not be ascertained in this study but may result from variations in practice structure. Health maintenance organizations, especially staff-model and group-model forms, have resources and organizational structures that can disseminate standards of care and ensure that current practice patterns are consistent with these standards.

Improved breast cancer survival among HMO recipients seemed to be primarily the result of earlier stage at diagnosis.

While HMO insurance vs FFS insurance arrangement was statistically significantly associated with breast cancer outcomes, lung cancer outcomes showed few effects. The subset of persons changing between HMO and FFS plans seemed to have better lung cancer outcomes. Among 137 patients with lung cancer who changed insurance type between diagnosis and 6-month follow-up, most (65%) changed from HMO insurance to FFS insurance. This group had a greater likelihood of undergoing surgery for early-stage disease and had better overall survival, with a statistically nonsignificant trend toward better lung cancer survival.

Changing between HMO and FFS Medicare plans might indicate problems in accessing care or dissatisfaction with care. Patients having disabilities are generally more likely to report dissatisfaction or problems with their healthcare plan

 and are more likely to disenroll from their HMO, often changing to an FFS plan.

 Forced disenrollment from a Medicare HMO plan has been associated with problems in accessing needed care.

In our study, Medicare beneficiaries with disabilities had remarkably stable HMO and FFS insurance status during their follow-up, similar to other studies.

Among persons continuously eligible for Medicare and followed up until their death, 95% of persons with lung cancer and 92% of persons with breast cancer had continuous coverage within FFS or HMO arrangements from the 12 months before diagnosis until their death. In addition, patients who changed between HMO and FFS plans generally had similar or better outcomes compared with persons continuously enrolled in FFS.

Our study had several important limitations. We did not have Medicare claims data for persons enrolled in HMOs and were unable to examine cancer screening or to supplement SEER information on treatment using Medicare claims. Our lack of Medicare claims for persons in HMOs prevented us from assessing comorbidity. SEER does not release data on chemotherapy, so we were unable to evaluate this aspect of cancer treatment, which is especially critical in breast cancer. Medicare data did not include details about the specific HMO plan, so we were unable to assess the particular financial arrangements for the HMO plan, nor could we capture patient movement between HMO plans. Our sample was restricted to persons who originally qualified for Medicare coverage because of Social Security Disability Insurance, and our results may not generalize to the greater population of persons having disabilities. We studied persons who were diagnosed as having cancer through the end of 1999, and it is possible that trends may have changed since that time. Finally, this was an observational study that did not randomize subjects to insurance types. Statistical methods, such as propensity scores, can only adjust for measured characteristics within the cohort. As a result, it is possible that our results were in part due to unmeasured patient characteristics that differed between HMO patients and FFS patients and not because of the specific insurance arrangement.

In conclusion, Medicare beneficiaries with disabilities diagnosed as having breast cancer generally had more favorable outcomes within HMO arrangements. Health maintenance organization vs FFS insurance status had little effect on lung cancer outcomes. Changes between HMO and FFS insurance types were not associated with poor cancer outcomes.

This study used the linked SEER-Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of several groups responsible for the creation and dissemination of the linked database, including the Applied Research Branch, Division of Cancer Control and Population Sciences, National Cancer Institute; the Office of Information Services and the Office of Strategic Planning, Centers for Medicare and Medicaid Services; Information Management Services, Inc; and the SEER Program Tumor Registries. 

: Department of Family Medicine, University of South Florida (RGR), and H. Lee Moffitt Cancer Center & Research Institute (RGR, TNC, KJW), Tampa; and Divisions of General Medicine and Primary Care (EPM, LHN, DL) and Hematology and Oncology (RED), Beth Israel Deaconess Medical Center, Department of Medicine, Harvard Medical School (EPM, LHN, RED, LII), and Institute for Health Policy, Division of General Medicine, Massachusetts General Hospital (LII), Boston.

: This study was supported by grant R01 CA100029 from the National Cancer Institute.

: The authors (RGR, TNC, KJW, EPM, LHN, DL, RED, LII) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article. 

: Concept and design (RGR, TNC, EPM); acquisition of data (TNC, EPM, DL, LII); analysis and interpretation of data (RGR, TNC, KJW, EPM, DL, RED, LII); drafting of the manuscript (RGR, TNC, KJW, RED); critical revision of the manuscript for important intellectual content (RGR, TNC, EPM, RED, LII); statistical analysis (RGR, TNC, KJW, LHN, DL); and obtaining funding (TNC, LII).

: Richard G. Roetzheim, MD, MSPH, Department of Family Medicine, University of South Florida, 12901 Bruce B. Downs Blvd, MDC 13, Tampa, FL 33612. E-mail: 

1. Centers for Medicare and Medicaid Services Web site.

 Medicare enrollment: disabled beneficiaries: as of July 2005. 

http://www.cms.hhs.gov/MedicareEnRpts/Downloads/05Disabled.pdf

3. McCarthy EP, Ngo LH, Chirikos TN, et al.

 Cancer stage at diagnosis and survival among persons with Social Security Disability Insurance on Medicare. 

 Disability and managed care frenzy: a cautionary note. 

 Financial vulnerability among Medicare managed care enrollees. 

9. Baker LC, Phillips KA, Haas JS, Liang SY, Sonneborn D.

 The effect of area HMO market share on cancer screening. 

Type of health care coverage and the likelihood of being screened for cancer. 

13. Lee-Feldstein A, Feldstein PJ, Buchmueller T, Katterhagen G.

 Breast cancer outcomes among older women: HMO, fee-for-service, and delivery system comparisons. 

15. Riley GF, Potosky AL, Lubitz JD, Brown ML.

 Stage of cancer at diagnosis for Medicare HMO and fee-for-service enrollees. 

17. National Comprehensive Cancer Network Web site

. NCCN Clinical Practice Guidelines in Oncology: nonâ€“small cell lung cancer, V.2.2008. 

http://www.nccn.org/professionals/physician_gls/PDF/nscl.pdf

19. Potosky AL, Riley GF, Lubitz JD, Mentnech RM, Kessler LG.

 Potential for cancer related health services research using a linked Medicareâ€“tumor registry database. 

 Summary Staging Guide: Cancer Surveillance, Epidemiology, and End Results Reporting. Bethesda, MD: Public Health Service, US Dept of Health and Human Services, National Institutes of Health; 1977. Publication 86-2313.

23.Virnig BA,Warren JL, Cooper GS, Klabunde CN, Schussler N, Freeman J.

 Studying radiation therapy using SEER-Medicareâ€“linked data. 

. Radical radiotherapy for stage I/II nonâ€“small cell lung cancer in patients not sufficiently fit or declining surgery (medically inoperable): a systematic review. 

 Reducing bias in observational studies using subclassification on the propensity score. 

.The effectiveness of adjustment by subclassification in removing bias in observational studies. 

 Use and misuse of population attributable fractions. 

33. Phillips KA, Haas JS, Liang SY, et al.

 Are gatekeeper requirements associated with cancer screening utilization? 

35. Morgan RO,Virnig BA, DeVito CA, Persily NA.

 The Medicare-HMO revolving door: the healthy go in and the sick go out. 

This study showed better outcomes for disabled Medicare patients with breast cancer but not those with lung cancer when they were enrolled in HMOs.

Lisa I. Iezzoni, MD, MSc

Articles

Early Experiences With the Acute Community Care Program in Eastern Massachusetts

Clinician Considerations When Selecting High-Risk Patients for Care Management

Managed Care and Cancer Outcomes for Medicare Beneficiaries With Disabilities