Skip to main content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

The Planning and Operational Experience of the Channeling Projects

Publication Date

George Carcagno, Robert Applebaum, Jon Christianson, Barbara Phillips, Craig Thornton and Joanna Will

Mathematica Policy Research, Inc.

July 18, 1986

This report was prepared under contract #HHS-100-80-0157 between the U.S. Department of Health and Human Services (HHS), Office of Social Services Policy (now the Office of Disability, Aging and Long-Term Care Policy) and Mathematica Policy Research, Inc. For additional information about the study, you may visit the DALTCP home page at or contact the office at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, SW, Washington, DC 20201. The e-mail address is: The DALTCP Project Officer was Robert Clark.

In this Executive Summary we step back to consider the lessons learned from channeling concerning the design and management of programs providing case management in a community setting for a frail elderly clientele. The focus is on operational issues and considerations rather than broad issues of policy or financing. The intent is to provide recommendations for replicating channeling-type case management in a nondemonstration setting. Even though these lessons are drawn from the experience of a voluntary program with a limited caseload, many of them are also relevant for those concerned with programs which serve an entire community and which require mandatory participation (such as a nursing home preadmission screen).

Extrapolating from channeling's experience for this purpose is inevitably a matter of judgment. Where appropriate, we identify those aspects of channeling practice or experience that appear to have been largely determined by its status as a demonstration or by its relatively small scale. In many cases the recommendations in this chapter represent the consensus judgment of all those involved in channeling--from the local projects to the state and federal staff directly involved to the technical assistance and evaluation contractor staff. In other cases we make recommendations based on our own views or those of a subset of the actors involved in the project.

The first section considers organizational and management issues, the second identification and recruitment of clients, the third provision of case management. The final section addresses the arrangements between case management agencies and service providers.



If one were to consider establishing a publicly sponsored case management program, two major of issues would have to be confronted: what type of agency should be assigned responsibility for the program? What organizational structures and arrangements offer the greatest likelihood of success? Channeling's experience provides insight into these and related issues.

1. Selecting a Case Management Agency (Chapter III)

Selection of an appropriate host agency for a case management program depends in part on factors that vary from community to community. These factors include the structure of each community-based long term care system and the strengths and weaknesses of the potential host organizations available in an area. However, there do appear to be some general host agency characteristics that should receive weight in any selection process.

Agencies that are well known and respected in their communities make effective host agencies. This is hardly surprising. Their strong reputations can promote community acceptance of the case management program and facilitate the development of initial contacts with referral sources and service providers. Experience in contracting with service providers is particularly desirable for a host agency because contractual relationships between providers and the case management agency can be administratively complex, time consuming, and frustrating to the uninitiated.

It is also useful for host agencies to possess expertise in fiscal and budgetary matters. They can then provide support in the design and implementation of fiscal systems and/or assume responsibility for routine fiscal operations for the case management agency. This can result in a more efficient use of case management agency resources. In addition, flexibility in personnel matters is a valuable characteristic of host agencies. Staff of case management agencies must often work nontraditional hours and cope with job-related stress. Administrators, to respond creatively to personnel problems, must not be unduly restricted by host agency policies or procedures.

One issue which would seem extremely important in the establishment of any publicly funded program is whether it should be placed in a public or private sector host agency. The channeling experience does not permit a definitive recommendation on this score. Both disadvantages and advantages were perceived with respect to public agency auspices. Channeling project staff reported that procurement regulations and civil service procedures in public agencies presented obstacles in contracting with providers and recruiting and hiring staff. These problems were generally not encountered with private sector host agencies. However, they also noted positive aspects of being located in a public agency--such as enhanced potential to influence policy, demonstrated accountability for public funds, and greater access to political power.

2. Organization and Internal Management (Chapter IV)

There was considerable variation in the organizational structures and internal management practices of the channeling projects. This provided the opportunity to compare the operation of centralized versus decentralized projects and also of projects where assessment and case management functions were integrated functions (i.e., both assigned to the same person) versus projects where the two functions were handled by different staff members.1

Screening, Assessment, and Case Management. In the case of the screening, assessment, and case management functions, applying channeling's experience to other programs and settings is not straightforward. In channeling, to avoid inducing behavioral changes in the control group, the screening unit was kept administratively separate from the rest of the project. The intent was to keep the control group insulated from the case management staff. There was universal agreement among respondents to the site visit protocols, however, that in a nondemonstration setting the screening unit should be an integral part of the case management program. Channeling project staff believed that an integrated structure would both facilitate supervision and enhance screener morale.

In general, the project staff also preferred combining assessment and case management functions into one position. Even at projects where assessment and case management functions were split, a majority of respondents concluded that the combined approach was preferable. The problems with the split function approach stem from the fact that clients have to deal with both an assessor and a case manager, resulting in a potential loss of continuity, a need to reestablish rapport, and the danger of giving clients conflicting information.

Automated Support. The ability to pay for services from pooled funds requires the capacity to keep track of services authorized and delivered and to bill providers promptly and accurately. Furthermore, case managers and agency management must be able to monitor service costs on a timely basis. The experience of channeling indicates the magnitude of these tasks and the importance of a well-designed, automated system to perform them. For example, calculating the average service cost for the caseload, one element of channeling's cost controls, would be very tedious to do manually. However, care should be taken not to oversell the capabilities of such a system to the staff who will use it. An automated system will not relieve them of all tedious paperwork. Moreover, automated systems are difficult and time consuming to design and implement. In short, while an automated system is critically important to the operation of a case management program, it poses new requirements and challenges for case managers and management staff. Staff are likely to underestimate these and become disillusioned with the system unless warned of the difficulties beforehand.

Service Audit/Program Review. Initially, channeling projects were to implement an ongoing service audit/program review activity. The service audit involved a review of case records to determine whether care plans were adequate to meet clients' needs and to help ensure that the records themselves were up to professional standards. The program review was meant to be a broader examination of channeling project procedures and functions. The service audit/program review was made an optional program feature by DHHS, and most projects did not implement either of them as a regular activity. Despite this, staff of all projects indicated their view that an independent service audit of records of agency clients should be instituted as a matter of standard procedure for a case management agency. They felt the results of this audit are useful as a management tool for improving case management performance. Moreover, they believed that in a nondemonstration public program some type of service audit and program review is necessary to establish accountability to the public and to ensure the adequate quality of the case management.

Advisory Committees. The usefulness and structure of advisory committees will depend in large part on the style of the director of a case management agency and the existing long term care environment. The channeling experience suggests that such committees could be particularly useful where community acceptance of the agency is at issue, since they can play an important role in facilitating acceptance.



In this section we first consider the implications of channeling's experience for identifying a population at risk of nursing home placement. We then consider what we learned about the actual process of screening applicants for a community care program.

1. Eligibility Criteria (Chapters VI and VII)

Channeling set out to identify a population at high risk of nursing home placement. Based on the experiences of prior community care demonstrations, channeling planners believed that to be cost effective channeling had to divert people from institutional placement. The eligibility criteria were developed to identify such an at-risk population. Despite the fact that these eligibility criteria were implemented as planned, the channeling control group made relatively little use of nursing homes. (Not more than 13-14 percent were in a nursing home after 12 months; see Wooldridge and Schore 1986.) Furthermore, with one exception, analysis of subgroup outcomes conducted to date does not show any subgroups for whom channeling was particularly effective (Grannemann, Grossman, and Dunstan 1986). The one exception was the small group in a nursing home at enrollment. Nursing home use among this group was much higher than the full sample, and the reduction in nursing home use substantially larger.

In evaluating channeling's experience in this area, it does not appear that the target criteria were conceptually flawed, or that improvements in instrumentation or data collection procedures would have identified in the demonstration. It is possible that if a larger proportion of referrals had come from nursing home preadmission screening units a group at greater risk of institutionalization would have been identified. This is, however, speculative. In the opinion of channeling staff, once an elderly person and his family had made the difficult decision to apply for entrance to a nursing home, it was usually not possible to convince them to change their minds. A mandatory screening program obviously has more leverage than channeling, and could require an applicant to remain in the community. But whether such a mandatory program will target more effectively over the long run is an unanswered question. The South Carolina demonstration, for example, used the preadmission screen approach and reported high use of nursing homes among control group members and substantial reductions among treatment group members. The danger is that referral sources and families will, with the passage of time, come to view the nursing home screen as the mechanism whereby one gains access, not to nursing homes but to community care services. In effect, the characteristics of people applying for nursing homes might shift and become more like those of channeling applicants.

Even if channeling selection criteria did not identify a group at-risk of nursing home placement, did channeling, nevertheless, provide information that would enable us to define more refined criteria for admission to community care? Functional disabilities and impairments as measured by activities and instrumental activities of daily living are a necessary component of any eligibility process that seeks to identify a frail population with the potential to remain in the community. The experience in channeling and other studies, however, indicates that these are not sufficient. As documented in the comparisons in this report to both the current nursing home population and a simulated national sample eligible for channeling, there are many individuals in the community with levels of impairment comparable to those of persons in a nursing home. Why some individuals enter nursing homes while others with similar characteristics do not is not well understood. In addition to disability or impairment, there appear to be other factors which hold the answers to the targeting question. Some of them may be essentially unmeasurable. Measurable factors such as living arrangement, marital status, proximity to children, living environment, family and client attitudes to nursing home care, ethnicity and cultural background, income, and type of climate have been shown to contribute to the decision to choose nursing home care. However, using these factors as operational eligibility criteria for entry into a public program is not a very plausible option. Applicants can hardly be turned away in our society because of their marital status, attitudes, living arrangement, ethnic group, and so on.

It is true that the channeling eligibility criteria were developed several years ago. In our judgment, however, the current state of knowledge would lead to a very similar set of criteria. It would, therefore, be incorrect to conclude that development of better targeting criteria would be a straightforward way of enhancing our ability to identify an at-risk population. Moreover, the entire effort to identify such a group is based on the assumption that community care can be cost effective only if it succeeds in diverting to the community people who would otherwise be institutionalized. Perhaps this in itself is misguided. It has been argued that such a strategy may be infeasible (Weissert 1985) and that attention should instead be directed to the beneficial aspects of improving the quality of community residence for clients and caregivers.

2. The Screening Process (Chapters VI and VII)

In the channeling context, screening means a brief telephone interview conducted with applicants before an inperson (baseline) assessment is performed. The purpose of such a screen is to determine whether an applicant is likely to meet a program's eligibility criteria. Experience with the channeling demonstration indicated that some form of applicant screening is necessary because of the expense and staff effort that would otherwise be required to give detailed baseline assessments to all applicants, including a substantial proportion who would turn out to be ineligible for the program. In the majority of instances (80 percent) the channeling eligibility decisions on the basis of the telephone screen agreed with those on the basis of the inperson assessment. Data were not available on those applicants rejected on the telephone screen, so we could not test whether applicants who would have been eligible on the baseline were declared ineligible on the screen. Given the pressure to build caseloads in the demonstration, however, it is unlikely that substantial numbers of eligibles were incorrectly screened out. Channeling demonstrated that a systematic telephone screen with well established criteria can be implemented, thereby yielding efficiencies in the eligibility determination process.

One reason why the telephone screen was feasible was that proxy respondents were accepted. While there was some concern expressed about proxies overstating need, in general proxy respondents were reported to be a good source of information. The use of proxies also substantially improved the efficiency of the screening process, because in cases where an applicant did not have access to a telephone or was unable to use one, the use of proxies made the telephone option possible in the overwhelming majority of cases. Even so, the channeling experience made it clear that a program should build in the flexibility to conduct inperson screens in the rare cases they are needed.

Our experience in operationalizing the screen criteria indicates that a telephone screen can successfully apply criteria related to physical functioning. However, in many cases it proved quite difficult to use a short telephone screen to adequately measure such factors as the future availability of informal supports, unmet needs, and cognitive impairments. We recommend that, although these factors may be important in determining final eligibility, they should not be used on a telephone screen but rather saved for a later stage. The screen should focus on the functional ability of the applicant as the major criterion.

Some community care programs may wish to exclude persons who are judged to be too disabled to remain in the community, either because they might not be safely cared for or because the cost of their care might be excessive. Such determinations involve careful consideration of the availability and capabilities of informal caregivers and the development of a care plan. Consequently, we recommend that telephone screens not be used to judge whether persons are too disabled for community care.

With respect to the skills of screening staff, the majority of screeners had some human service experience prior to channeling and administrative staff felt this was helpful. Of greater importance, however, were the reports of screening and administrative staff that standardized screening training was essential. The screening training was considered to have two beneficial effects. First, it ensured to the extent possible that screeners implemented the criteria in a standardized manner. Second, the training provided information on listening and telephone skills which screeners reported as important. (Training manuals and other technical assistance reports are listed at the end of this report.)



The provision of case management services was the common element of both channeling models. In this section we consider the implications of channeling's experience for assessment and care planning, for access to discretionary, gap-filling funds, for the design and execution of cost control mechanisms, and for the qualifications of case managers and supervisors.

1. Assessment and Care Planning (Chapter VIII)

Channeling planners placed great emphasis at the beginning of the demonstration on a standardized structured assessment and a formalized structured care planning process. One motivation for this was the research requirement that key elements of the intervention be uniformly implemented across projects. But a more important motivation was the substantial clinical gains seen by channeling planners from this approach to assessment and care planning.

At the end of the demonstration our belief is unshaken that a standardized assessment and formalized care planning process are important in designing a care plan for the chronically impaired individual. But we share the view with a majority of care managers who suggested that the research assessment instrument used in channeling is not practical for regular program use. In our judgment a shorter version of the structured baseline assessment instrument--possibly resembling the clinical baseline assessment instrument used after the end of random assignment, or even the seven-page assessment summary--can feasibly be incorporated into long term care programs if adequate training is provided, and will reap important clinical benefits.

The importance of extensive training in the use of the assessment instrument appears to be overlooked in general practice. This should include a detailed walk-through of the assessment instrument, instruction in the basic principles of interviewing, and close supervision of the initial assessments completed by a new assessor.

A formal reassessment (with associated training) should be undertaken at regular intervals using an instrument closely patterned on the intake assessment instrument but omitting one-time client information. Initially the first reassessment under channeling was scheduled three months after program entry. This proved burdensome for case managers and also unnecessary, for most case managers were in frequent contact with clients during the care planning and service initiation process, which extended over the first several weeks in any case. Subsequently, reassessments were scheduled every six months. They provided case managers with a comprehensive status report on each client and were important in helping ensure the adequacy of care plans over time.

We also conclude, as channeling evaluators, that a structured care planning process which requires case managers to design a comprehensive plan of care and to modify that plan as needed is important in meeting the needs of long term care clients. Channeling staff concurred with this general recommendation. But they felt strongly that ways should be found to reduce the substantial paperwork burden. There were, for example, many more service order modifications to be prepared than had been anticipated by channeling planners when the financial control model automated system was designed. In sites with competitive bidding where only one provider of a given service could be selected, a change in providers required rewriting every service order. As a consequence, one of the strong recommendations of channeling staff was to automate the client care planning process. We believe efforts to automate should proceed with caution. Care planning is a complex judgmental process that will be difficult to reduce successfully to the precision of an automated system. Furthermore, even when an automated system is implemented it is not likely to do away with the clerical burdens--service orders must still be prepared and data entered to the system. A more realistic approach might be to provide additional case aide and clerical support to the case management staff, as was done in some channeling projects.

Case managers' interactions with the medical community, particularly physicians, was the subject of considerable comment by channeling staff. Many persons in need of long term care services have medical needs that require physician attention. Physicians often are asked for advice about long term car options and their recommendations can have a major influence on the decisions of older persons and their families. Physicians also serve as gatekeepers to long term care services both for home health care and institutional services. Despite their important role in the long term care system, communication between channeling case managers and physicians was minimal. Numerous reasons were given for this lack of communication ("physicians are not knowledgeable about community care," "case managers lack enough medical training to interact with physicians"). We conclude from the channeling experience that an ongoing long term care case management program should develop explicit mechanisms to increase physician involvement in the exploration of community based long term care alternatives. In the absence of such special efforts, it is quite likely that the needed communication simply will not occur, as was the case in channeling. Some channeling staff even suggested that reimbursement for physicians who assisted in arranging for community care might be a fruitful strategy.

The timeliness of care was another important issue highlighted by the channeling experience. Obviously, an ongoing program should be designed so a timely determination can be made as to whether an applicant or new client has an immediate need for care and, if so, is structured so that care is provided on a timely basis. Channeling's assessment and care planning process took longer than expected--elapsed time from screening to initiation of services averaged about a month. Some referral sources adjusted to this by making dual referrals--to channeling and to another agency. A new client's immediate needs could then be met, for example, by a home health agency while a case manager developed a care plan for the longer term. While such arrangements could be formalized in an ongoing program, they appear cumbersome and potentially duplicative. A sounder approach, in our judgment, would be to design the case management intake process so that clients with immediate needs could be readily accommodated. Initial assessments must be completed promptly. Furthermore, some system must be built into the program so that clients with an urgent immediate need can have services put in place before completion of the overall care plan. This necessitates a special simplified service ordering process. To guard against misuse there should be mandatory supervisory review of at least a random sample of these immediate-need cases.

2. Purchasing Services and Controlling Costs (Chapter VIII)

Channeling's experience with direct service discretionary spending--in the form of the limited gap-filling dollars for the basis projects and the more substantial funds pools for the financial control projects--highlighted two important issues for replication. First, the overwhelming needs of the chronically impaired or disabled population are home health/personal care needs. Hands-on personal care, home health care, homemaking, and meal preparation were the dominant services purchased under both models, rather than the medically oriented services ordinarily funded under Medicare and Medicaid. Respondents reported that these types of personal care services were in short supply in the existing system, yet they were the backbone of the direct services component of the channeling intervention. The second issue is the flexibility associated with the funds. Even though the majority of expenditures was for personal care services, case managers emphasized how important they found the flexibility to purchase other types of services. Channeling project staff also reported the importance of providing caregiver respite by purchasing weekend and evening care for clients, and suggested that this care was often not available under current funding mechanisms. Channeling project staff generally felt strongly that having access to flexible discretionary funds, even if the amount of available funds is small, should be a component of an ongoing program.

Cost control mechanisms are needed to guide case manager behavior in the use of such funds. Such mechanisms should also be a component of the care planning process, in order to encourage case managers to be sensitive to the costs of the care they prescribe. Channeling provided the opportunity to set up cost control measures and gain experience using them. The channeling staff reported that they became increasingly sensitive to the costs of care through this experience, and learned to deal with the tradeoffs among types of care, volume of services, and costs.

At the start of the demonstration, financial control model project staff were concerned that the average caseload expenditure cap at 60 percent of the nursing home rate was too low. Many also voiced opposition to any form of client cost sharing. By the end of the demonstration, the channeling staff were convinced of the value of cost control mechanisms and were recommending more stringent cost sharing requirements than were used in the demonstration. This suggests that well-managed cost control and cost sharing procedures can gain the full support of human services staff.

Cost control mechanisms (such as costing out care plans and caps on service costs) should be part of any case management program. An average caseload cap, as used in channeling, is one such mechanism, although the 60 percent channeling cap was probably too high. More operating experience is needed to evaluate how case managers behave when handling a caseload with a cap much closer to actual care plan costs. In addition to monitoring estimated care plan costs, actual service costs should be monitored to ensure compliance with cost control measures.

Within the cost control mechanism special consideration should be given to accepting clients who have high cost care plans when there is a reasonable prospect that the costs will decline over time or be offset by clients with low cost care plans. The strong tendency for channeling case managers was to make the average caseload cost control cap the maximum for individuals. This suggests that high cost cases are likely to be excluded, unless explicit steps are taken to ensure that inclusion. Obviously, such cases should also be subject to special review.

Client cost sharing should be incorporated into regular ongoing case management programs. Cost sharing can be designed so that it is not unduly burdensome on clients and their families. The cost-sharing requirements under channeling were constrained by its status as a demonstration--a special rule was implemented that clients could not be asked to pay for services already available at no cost in their community. Since there were many such services and the cost sharing threshold was intentionally set at a relatively high level, only 5 percent of the channeling clients were involved in cost sharing. In an ongoing program many more clients could be expected to share in the costs of their care.

3. Qualifications of Case Managers and Supervisors (Chapter VIII)

Strongly held and divergent views were expressed by some of the channeling project staff as to the appropriate background of a case manager. Some argued that a nursing background was a must; others argued as strongly fora social services background. However, the majority believed that either discipline provides an adequate background, but that each requires some experience of the other. Social service case managers need some medical experience, for example, and nurse case managers need some social service experience. All agreed that at least some staff representation of both areas in some capacity is essential. The channeling experience suggests that supervision can be provided by either experienced nurse or social work staff; however, if the supervision is not by nursing staff, nursing input and review of care plans by a consultant was viewed as essential. It is clear that the consultative and supportive role of the supervisor is of paramount importance to the ongoing case management function, and that the support and encouragement of supervisors was important in helping many channeling case managers cope with the stresses of their job.



1. Provider Selection (Chapters IX and XI)

The channeling projects, particularly in the financial control model, had to develop procedures for selecting providers. Several variants of a competitive bidding process were used. Based on the experience of the channeling projects, it is not possible to recommend a single "best" approach to provider selection and rate determination for other case management agencies. The appropriate procedure for any given case management agency will depend to some degree on the availability of providers, the procurement rules of the organization of which the agency is a part, and the budget pressures confronted by the agency. In communities where there are few competing services providers, it is unlikely that formal, competitive selection processes will be successful. Competitive contracting processes are more likely to be effective in containing costs where there are a number of potential bidders and relatively larger numbers of clients to be served under contract. Even under these circumstances, the overall impact of the contracting process chosen will depend crucially on its design.

Bidding processes that result in contracts with a single provider for a large number of clients appear to have strong cost-containment incentives. However, they also are administratively difficult and costly to manage and may lock case management agencies into contracts with unproven providers. (Such contracts can require significant resources to enforce and monitor.) Furthermore, it an incumbent contractor has to be replaced, the transition can be very costly and disruptive, and can even put clients at risk of harm. Bidding processes that result in the award of multiple contracts, and allow greater consideration of nonprice factors in making contract awards, seem to contain weaker incentives for price restraint, although as discussed below, this conclusion is somewhat speculative within the limited time frame of the channeling demonstration. Multiple award procurements do offer the greater likelihood of including proven and/or familiar service providers as options for clients.

In choosing an approach to provider contracting, case management agencies will need to evaluate these tradeoffs in light of their own circumstances. Where the achievement of low unit prices is the primary objective of the contracting process, and an ample number of existing service providers are available or where it is relatively easy for new providers to enter the market, then the adoption of competitive bidding with a single source contract award per service may be an attractive option. However, estimates of savings in direct service costs from this approach must be qualified by the likelihood of higher administrative costs and the potential difficulties faced in changing contractors. For these reasons, awarding multiple contracts might be preferable to some agencies. Where agency administrative resources are tightly constrained, it may be desirable to pursue less structured provider selection processes that place more discretion in the hands of agency case managers. One result of this approach, however, could be higher average prices paid for services.

These observations, along with the discussion of the effect of bidding system design on bid prices, raise an important issue regarding extrapolation of the demonstration findings to an ongoing program. If, in such a program, the case management agency were the largest purchaser of services and awarded contracts only to the lowest winning bidders, it could risk driving nonprofit providers from the industry and creating near-monopoly market positions for winning bidders. By affecting the structure of the home health care market in this manner, the channeling agency could reduce competition in subsequent rounds of bidding and increase program costs in the long run, particularly if there were barriers to new firms entering the market. Conversely, in an ongoing program, where a case management agency controlled larger numbers of clients, price competition among bidders might be more intense under multiple winning bidder systems than observed in the demonstration. Thus, the apparent cost-containment advantage of the single winning bidder versus multiple winner system might be less in an ongoing program than suggested by the experience of the channeling projects in the demonstration. The channeling demonstration did not operate long enough, nor did it involve large enough numbers of clients, to provide empirical evidence on this issue.

An aspect of provider selection in channeling that was unanticipated was the fact that about 20 percent of sample members had services provided by privately contracted individuals. These arrangements must be carefully evaluated by the case manager in the care planning process. Where privately contracted individuals are providing dependable, adequate service they can be incorporated in the client's care plan. In instances where they prove undependable or their services are substandard, case managers may wish to increase their monitoring efforts or encourage the substitution of other providers.

The experience of some of the channeling projects demonstrated that individually contracted providers can be a valuable client resource when trained and supervised by the case management agency as part of a formal program. They can be used successfully to fill gaps in the existing service delivery system and are generally well received by clients.

2. Monitoring the Provision of Community Care (Chapter X)

Channeling planners and project staff did not anticipate the importance of monitoring service providers and underestimated the amount of effort required to do it effectively. We suggest that provider monitoring and quality assurance activities by explicitly identified in agency budgets, since they are likely to consume substantial administrative and case manager resources.

Funder demands for agency accountability mean that data on provider performance must be collected routinely and systematically through a variety of procedures. These basic data should include, at a minimum, frequency of complaints about service quality, tardiness in service delivery, failure to keep service appointments and abusive behavior. This sort of evidence is essential to detect patterns of inadequate service delivery and, where formal contracts are present, to enforce contractual penalties.

Top management staff need to be committed to monitoring and quality assurance activities, and to be involved in them on a regular basis. The commitment is needed for successful agency implementation of monitoring procedures; the regular involvement is required to resolve differences with service providers.

Cooperation among community case management agencies in provider monitoring and quality assurance activities can improve its effectiveness. Since there are no well-defined existing models for case management agencies to follow, information exchange among agencies can be useful in identifying exemplary approaches. Also, pooling of information can facilitate the detection and correction of service delivery problems associated with particular providers.

Although channeling was a voluntary demonstration program serving only a portion of the eligible population for a limited period of time, it provides a wealth of information about conducting a case managed community care program. In this chapter we sought to summarize the most general lessons learned. More detailed discussions can be found in the other chapters of this report. Obviously these insights can only provide guidance to those concerned with designing and implementing case management programs. And even for case management, other program in other communities will have to take into account their differences from as well as similarities to channeling before applying these results to their own situations.



Grannemann, Thomas W., Jean Baldwin Grossman, and Shari Miller Dunstan. Differential Impacts Among Subgroups of Channeling Enrollees. Princeton, NJ: Mathematica Policy Research, 1986.

Thornton, Craig, Joanna Will, and Mark Davies. The Evaluation of the National Long Term Care Demonstration: Analysis of Channeling Project Costs. Princeton, NJ: Mathematica Policy Research, Inc., 1986.

Weissert, William G. "Seven Reasons Why it is so Difficult to Make Community Based Long-Term Care Cost Effective." Health Services Research, Vol. 20, No. 4, October 1985, pp. 423-433.

Wooldridge, Judith and Jennifer Schore. Channeling Effects on Hospital, Nursing Home, and Other Medical Services. Princeton, NJ: Mathematica Policy Research, 1986.



  1. The costs of case management were also examined (see Chapter V). The estimates of channeling case management costs are well within the range of costs of similar demonstrations that have been subjected to systematic project cost analysis. They are slightly higher than they would be for an identical program that is not a demonstration. Our analysis suggests that the overestimate is about out percent of total costs. They may also tend to be higher than a regular ongoing program with the same case management components because of the relatively small scale of the specific channeling projects. We found no evidence, however, that these considerations were any more important than cost differences associated with target populations served, program goals, case management models adopted, internal management and organization of functions, relationships with referral agencies and service providers, regional prices, and the local service environment. See Thornton, Will, and Davies 1985 for more details.