This Web site is a component of the SAMHSA Health Information Network

Enter Keywords Choose Website This Site Mental Health at HHS All Mental Health

Online Publications Order Publications National Library of Medicine National Academies Press Publications Homepage

printer friendly page e-mail this page bookmark this page shopping cart  current or new account

This Web site is a component of the SAMHSA Health Information Network.

Clinical Preventive Services in Substance Abuse and Mental Health Update: From Science to Services

XIV. Appendix B: Policy and Management Issues and Guidelines

This appendix is intended as a primer for health care administrators, policymakers and fiscal officers—to set the stage for successful implementation of preventive behavioral-related services in health care delivery settings.

Additiona guidance on policy and management issues can be found in Appendix D, Implementation and Evaluation of Preventive Services. Appendix C provides guidelines for billing for preventive services. The following issues are addressed here:

Translation of Research Into Practice

• Deciding on benefit packages
• Interpreting the medical literature
• Projecting benefits and desired consequences

Unintended Benefits and Adverse Consequences

• Pareto’s Law (the “80/20” rule)
• Perceptions and biases often shared by administrators and physicians
• Time intervals from cost to benefit
• Build versus buy options—disease and demand management

Benefits of Preventive Services

• Member health status and quality of life
• Quality of care
• Employee productivity
• Cost containment
• Image/reputation of health plan

Need

• Incidence and period prevalence
• Case identification
• Severity of illness
• Severity of risk

Efficacy

• Projection and modeling
• Effectiveness and cost-efficiency

Infrastructure

• Surveillance and data systems
• Screening policies and procedures
• Follow-up protocols
• Counseling
• Psychoeducation
• Health education (individual, group, and Web site)
• Case management
• Call centers
• Home visitation

Translating Research Into Practice

Translating preventive behavioral research into health care practice is a complex matter. There are several questions to be addressed at policy, management, and clinical levels as well as perceptions to be addressed— perceptions that historically have limited patient access to behavioral services within health care delivery systems. The major questions can be summarized as follows:

• What preventive behavioral services should be considered for inclusion in benefit packages and why?
• How does a health care plan control the utilization of preventive behavioral services?
• How does a health care plan determine the need for preventive behavioral services?
• How does a health care plan measure the impact of preventive behavioral services on health care costs?
• How does a health care plan manage preventive behavioral services to assure quality, cost-efficiency, and effectiveness?

Deciding on Services and Benefit Packages
Health care plans and managed care organizations are mandated to provide mental health coverage by State and Federal authorities and to meet the National Committee for Quality Assurance’s Health Plan Employer Data and Information Set (HEDIS) guidelines. Even if this were not the case, the impact of such preventive services on other health care expenditures for selected preventive behavioral services make investment in such services a wise choice for health care plans and managed care organizations. Systems must be in place, however, to assure the quality and appropriate utilization of such services.

For most preventive behavioral services, there is little question as to the fiscal responsibility of health care plans and managed care organizations. However, some of the services discussed, which clearly are social or educational in nature, are intended to secure social and educational nonmedical benefits. Included are some of the services suggested for economically and socially disadvantaged women and their families. Although the value of these services is firmly established, the question remains as to who should pick up the cost. The answers to these questions will vary by health plan, depending on the public versus private orientation and the needs of membership.

Interpreting the Medical Literature
Randomized controlled trials present the strongest evidence for or against a suggested intervention, but the process of selecting both cases and controls almost always creates a situation in which the cases and controls differ in substantial ways from the enrollment of any given health care system. Cohort and cross-sectional studies usually have more typical patient and control populations, but they are weaker study designs.

When conducting a randomized controlled trial, the research team must carefully select both cases and controls to assure that the differences between these groups after the intervention can be reasonably attributed to the intervention. Health care plans wishing to implement services based on the research can consider the degree to which findings in each paper might or might not apply to their provider panel and membership. One major difference is the willingness of the patient to comply fully with prescribed regimens of care. Research subjects are selected for nearly 100 percent cooperation. Patients in conventional health care settings are far less compliant. Even with very careful selection of cases and controls, however, even the research studies still must address attrition and noncompliance.

To further complicate matters, real-life patients are likely to deny behavioral problems such as smoking, drinking, substance use disorder, and depression, and primary care practitioners in conventional health care settings are not likely to address these issues because of lack of comfort, desire not to weaken the patient/practitioner relationship, and lack of time.

Another problem is that enrollees might not be similar in demographics, cultural, or clinical profile to the subjects used in the research. Environmental factors also are important, with many patients and health care systems facing economic and administrative barriers that inhibit providerpatient communication (especially between clinic visits) and the high cost of some prescription medications.

Research studies present their results as if a cohort of cases and controls begin the intervention at the same time and are followed for the duration of the study, which then comes to an end. Health plan implementation must take place in the real world, where members are at all stages of illness at all times; where enrollment and processing must take place at all points in time; where concurrent comparable controls are not feasible; and where the intervention is anticipated to extend into the indefinite future. This severely complicates projections of costs and benefits, especially after the initial year of intervention. This set of complications requires establishment of tracking mechanisms that are not traditional in managed care plans—tracking mechanisms that at least initially may have to be structured as small-scale, separate, stand-alone data systems.

“Effectiveness” denotes whether a specified intervention will work under the conditions of a randomized controlled trial. “Efficacy” denotes whether this same intervention will work under conditions of routine health care delivery (Daumit et al., 2001). Depending on how it is managed, a theoretically effective intervention might not be efficacious within a given health care delivery system. The challenge for both health plan administrators and clinicians is how best to manage implementation of the intervention to secure the desired efficacy.

Finally, ethical considerations sometimes make randomization and/or control impossible to implement in research and health care delivery settings. The best example within the scope of this report is identifying the cause of poor pregnancy outcomes among pregnant women using illicit drugs (such as marijuana, cocaine, and heroin). The problem is that one cannot tell the degree to which the low birthweight and other “non-specific” poor pregnancy outcomes are due to the drug or due to the mother’s lack of prenatal care, poor nutrition, etc. Since it would not be ethical to treat the substance use disorder without providing prenatal care, or provide prenatal care while ignoring the substance use disorder, randomized controlled studies to differentiate the impact of prenatal care from the impact of substance use disorder counseling never will be done. This being the case, health care systems and individual physicians must establish some health policies and procedures without the benefit of randomized controlled trials.

Projecting Benefits and Desired Consequences
The researchers who generate the published literature generally avoid fiscal and management issues. The complexity of health care finance, pricing, and billing methodology, bundling and capitation, bad debt and cash flow issues, and annual changes in health plan enrollments all complicate attempts to link provision of preventive services with their impact on other health care expenditures. Although researchers can and often do document changes in health care utilization in response to effective delivery of preventive services, taking the next step and relating these changes to health plan expenditures usually is beyond the scope of their research protocols.

These fiscal and administrative issues are the issues of greatest importance to health care administrators and fiscal officers. From their perspective, morbidity and health care utilization are but two of many factors affecting the cost of health care delivery. To further complicate matters, fiscal incentives often are perverse in health care delivery, with cost savings for the health plan often being seen as revenue reductions for hospitals and providers.

The most tenuous aspect of documenting the benefit of preventive services is accurately and reliably projecting what would have occurred had the preventive service not been provided. This problem becomes more subjective with every passing year after initiation of the preventive service. Researchers can address this issue by dividing their subjects into cases and controls. This is something a health care system cannot do.

After the first few years of effective delivery of a new preventive service, there will be few differences in outcomes from year to year, as previously secured benefits are maintained on an ongoing basis. This lack of year-to-year improvements can leave the impression that the program has lost its effectiveness and thus lead to elimination of the preventive service. The literature indicates that preventing such premature program demise is best done by establishing baselines, benchmarks, and year-to-year projections prior to initiating the preventive programming, then tracking the programming against these projections. Precautions such as these are rarely taken when new preventive services are initiated. Failure to establish these baselines and benchmarks can lead to premature elimination of the preventive services when additional year-to-year reductions in health care costs cease to occur.

Unintended Benefits and Unanticipated Adverse Consequences

Prevention programming can have unintended benefits and unanticipated adverse consequences more substantial than the direct costs and intended benefits. An example of an unintended benefit would be a health education program to motivate patients to quit smoking, which might also result in lifestyle enhancements such as a more sensible diet or less binge drinking. An example of an unanticipated adverse consequence would be the process by which high-quality depression management programming by a managed care plan might result in physicians urging patients with depression to switch to that plan. In this case, the adverse consequence would be to the health plan and take the form of adverse patient selection. Health plan managers should try to project possible unintended benefits and adverse consequences of preventive services, and plan to measure them for purposes of program planning, program evaluation, and future policy development.

Pareto’s Law (The “80/20” Rule)
Pareto was a sixteenth century Italian economist who stated that in any human activity, a small percentage of the participants will account for most of the action. As commonly interpreted, in any given year, approximately 20 percent of the enrollees in a health care plan will be expected to account for approximately 80 percent of the cost, with the top 5 percent accounting for approximately 50 percent of the total health plan expenditures. In this context, the challenge to health care delivery systems is to identify those individuals who are likely to become high-cost patients in the near future, then take the action necessary to reduce the number of individuals at risk of falling into these high-cost groups.

An example appears in the behavioral literature where Simon and Untzer published a study in 1999 (Simon & Unutzer, 1999) presenting health care utilization and costs among patients treated for bipolar disorder in an insured population. Five percent of the patients accounted for approximately 40 percent of costs for (outpatient) specialty mental health and substance abuse services; 90 percent of inpatient costs for specialty mental health and substance abuse services; and 95 percent of out-of-pocket costs for inpatient care.

Health care systems can address both these risk profiles in two ways. One is through the use of the IOM “universal” screening procedures presented in this report, which are to be followed by the “selective” preventive services to confirm or deny the finding of the initial screening and identify those who could benefit from more intensive “indicated” services.

A major key to success in implementing the more expensive and more individualized preventive behavioral services is the ability of the health care delivery systems to successfully target the services to those most at risk, while avoiding provision of the more expensive and more individualized services to members not in need of these more expensive interventions. For most of these interventions, this depends on the skill and level of training of the primary care practitioners.

The second way for health care systems to identify individual and group risk involves data mining and use of predictive modeling software; in other words, skilled manipulation of claims, pharmacy, and other data. Unfortunately, these procedures do not lend themselves to randomized controlled trials. Unlike the screening procedures, the guidelines and software are proprietary and not subject to the scrutiny of peer-reviewed journals. Because of the lack of high-quality peer reviewed literature in this arena, these data mining and predictive modeling interventions were considered outside the scope of this report.

Perceptions and Biases Often Shared by Administrators and Practicing Physicians
Bias against prevention:

There is a widely held belief that prevention sounds good but just does not work.

Bias against behavioral/mental health services:

Behavioral needs and benefits are seen as “soft” since they usually are not verifiable by laboratory testing or physical examination.

Stratification and discrimination:

Stratification of both subpopulations and individuals is critical to costefficient provision of selective and indicated preventive services. In this context, “stratification” refers to the process by which a health care delivery system might identify groups of members to receive specific preventive services not offered to other members. Health care system managers often are uncomfortable with any form of stratification because of their perception that it is not ethical to offer some services to some members, but not others, when all are paying the same premium. They also worry about accusations of discrimination, stereotyping, and inappropriate use of confidential data. This issue and these conflicts can be very carefully explored. Failure to do so may severely inhibit the costefficiency of preventive services.

Uncertainty:

Administrators and fiscal officers dislike uncertainty. The inability to directly document what might have occurred had a preventive service not been provided creates a situation where there often is substantial uncertainty about the benefit to be secured by almost any preventive service. The most effective way to address this uncertainty may be by generating epidemiologically sound projections of likely costs and benefits, then measuring the outputs and outcomes so they can be compared with the original projections.

Competition and cost:

Health care systems and health insurance plans are highly costcompetitive. From their perspective, preventive services sometimes are perceived as overhead costs of no benefit to the plan. As a result, some managed care plans may feel that they cannot afford the cost of providing preventive behavioral services unless such services are imposed as a requirement or promise to reduce other health care costs within a year of program implementation. With members frequently switching managed care plans on the basis of premium cost, some health care systems are reluctant to invest in preventive services where they will pick up the cost of the preventive service, but a competing health care system will enjoy the fiscal benefit of the illness prevented. To address this issue at the national level, certifying and regulatory bodies now require the provision of selected preventive services.

Reluctance to develop supplemental data systems:

Supplemental data systems to track individuals and small groups often are required for the cost-efficient implementation of preventive and disease management services. Supplemental systems can be seen as costly, as violations of patient and physician confidentiality, and as an inappropriate use of premium dollars to fund research. Despite this reluctance, supplemental data systems along these lines now are being widely implemented in health care systems to address the overlapping needs of quality assurance, disease management, and preventive service programming.

The supplemental information systems often can be developed and initially managed on desktop computers using off-the-shelf spreadsheet or database software until such time as the data can be incorporated into larger claims-based or electronic medical record (EMR) systems.

The perceptions and biases noted above often may be best addressed directly within individual health care delivery systems if preventive services and therapeutic behavioral services are to be effectively and efficiently implemented with full accountability for costs and outcomes.

Time Intervals From Cost to Benefit
Changes in a health care system’s client base can dramatically alter both costs and perceptions of benefit. To track both costs and benefits, the possibility of such changes can be considered in the initial design of the systems used to track costs and benefits. Since there is little published literature on the time-related issues noted above, guidelines may be based on personal/professional opinion until such time as the needed research can be done.

Time-related issues are critically important given the need for administrators and fiscal officers to look at costs and benefits by calendar-quarter and fiscal year. There is a time delay between the decision to initiate the service and the time the preventive program comes online. There are options concerning how quickly the plan or medical center will saturate the initial need for the service in question, with rapid coverage usually requiring an intensity of staffing that need not be maintained in future years. There also are delays between the provision of the service and reduction in subsequent health care costs that must be well understood.

Build vs. Buy Options: Disease and Demand Management
Health plans may wish to carefully consider whether they wish to build their own disease management systems in-house, purchase the services from disease management and data system vendors, or pursue a hybrid approach. As a rule of thumb, one can generally expect a system purchased from an outside vendor to cost more than it would cost to provide the service in-house. In return for this higher cost, the purchase option offers the advantages of very little, if any, development time, access to welldeveloped and debugged systems, and a much greater chance to secure health care savings in excess of program cost within 12 months of initial program implementation. Vendors also offer expertise not otherwise available to many health care delivery systems.

Once developed, the health plan can plan to maintain the preventive/disease management program on a long-term basis. Even with vendored programs, much work is required to get the patients and physicians into the program and used to the system. Initially, the tracking systems will be the means by which the health care system documents the savings and other investments from the preventive programming, by tracking year-to-year changes in the outcome parameters. Three to 5 years into the program, however, the year-to-year differences will disappear, creating a situation where the tracking system becomes even more important to documenting maintenance of the desired benefit.

A major problem is that the dramatic reductions in health care costs from year to year will phase out over the first 3 to 5 years, as the benefit is achieved and maintained. This creates a situation where the higher costs of fully vendored systems will be harder and harder to justify to financial managers, with each passing year. This problem may be anticipated when such programs are initiated, either in-house, vendored, or hybrid, with understandings reached among program advocates, program staff, the vendor(s) and the financial managers as to how these issues are to be addressed in subsequent years. One possibility is to plan to transition the vendored services in-house over the first 3 to 5 years. If this is to be considered, attention must be paid to the issue of software licensure at the time the vendored service is initiated.

Benefits of Preventive Services
Member Health Status and Quality of Life
Major depression is one of a number of behavioral health disorders reviewed in this report with major impact on overall health status (especially in members with concurrent major chronic diseases), quality of life, and workplace productivity. The most effective and objective way for a health care system to measure functional health status and overall quality of life is through use of questionnaires, such as the SF-36 (Jordan- Marsh, 2002; Ferguson, Robinson, & Splaine, 2002; deHaan, 2002). The SF-36 is one of a number of currently available health status questionnaires that measures current physical health and mental health status in terms of what a patient is able to do and how the patient feels on a day-to-day basis. Such questionnaires can help the clinician identify undetected illness and depression and can help the health care system track the overall quality of care provided to patients with medical and behavioral chronic diseases.

Quality of Care
“Quality of care” has multiple domains. One domain is physician, nurse, and other health professional compliance with nationally recognized guidelines for the process of care. Another domain is health care system performance, as assessed using industrialtype measures of quality, consistency, and efficiency of administrative, logistical, and support services. Yet other domains include member health outcomes, member and physician satisfaction, employer-as-a-client satisfaction, and financial performance.

Because of the impact of preventive/disease management programming on member outcomes, and the manner in which HEDIS has framed preventive/disease management programming in the name of health care quality, all the guidelines in this report can be seen as quality-control measures. Increasingly, more and more preventive/disease management programming will be required of managed care plans wishing to score well on quality of care “score cards.”

Employee Productivity
Most of the commercial market for managed care plans is employer-based. There is ample literature to document that poor health can adversely affect employee productivity, with behavioral problems among the most costly (Williams & Strasser, 1999). Ironically, preventive/disease management programming does not commonly cover these disorders, presumably because they do not result in large numbers of emergency room visits and hospitalizations.

Cost Containment
The overwhelming importance of financial concerns in the management of health care systems has resulted in almost single-minded focus on “return on investment,” which usually is calculated on the basis of program costs and reductions in other health care costs within each 12-month period after program initiation. As noted above, these calculations can be very problematic.

With behavioral health services, favorable cost-effectiveness, as calculated above, can be reliably secured for services related to tobacco, drugs, and alcohol for pregnant women, and for early detection and treatment of depression in patients with diabetes, asthma, congestive heart failure, and other major chronic illnesses. The services to pregnant women reduce the need for costly newborn intensive care unit (NICU) services. The services to persons with chronic illness reduce emergency room and hospital use. Another service that the literature shows would almost certainly be cost-effective within 12 months of program initiation is psychoeducation for patients scheduled for surgery.

Image/Reputation of Health Care System
The image/reputation of the health care system is important for recruiting employers, members, and medical staff. Voluntary certifying agencies, such as the National Committee for Quality Assurance (NCQA), the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), and the Utilization Review Accreditation Commission, track preventive programming and selected health outcomes as measures of health care quality. Thus, preventive programming that does not offer immediate cost-effectiveness still can have substantial favorable fiscal impact by virtue of its value for health plan marketing.

Need
Incidence and Period Prevalence
For practical purposes, in a managed care setting, incidence can be considered the rate of new cases per 100 or per 1,000 members per year. New cases can arise from initial diagnosis/onset of new illness in previously enrolled members, and through enrollment of persons with preexisting illness. Period prevalence is the percentage of members known to have a specified disease within any given year. There is interplay among incidence, prevalence, and severity of illness that needs to be understood.

When dealing with behavioral disorders and major chronic diseases, there always is a hidden burden of illness that has not yet been diagnosed but is present within the enrolled population. For behavioral disorders and diseases such as hypertension and diabetes, the number of not-yetdiagnosed cases may equal or exceed those known to the plan through claims data. The purpose of screening programming is to aid the early diagnosis of these diseases so that they may be inexpensively managed before the progress to a more severe stage of illness. Screening programming increases incidence and prevalence while reducing average severity of illness and future costs. In many cases, the “future” is only 6 to 24 months away, creating a situation where such programming often can pay for itself within 12 to 24 months.

Case Identification
Both individuals and groups with priority needs can be identified in many ways. For some, claims or pharmacy data can be used to identify persons with diabetes, asthma, and other chronic diseases who may need priority screening for depression and other mental illnesses. Others can be identified through chart review. For some, however, identification will be difficult because the member may not be forthright about behaviors seen as stigmatizing, or out of a desire to please the doctor who may be seen as an authority figure. This frequently is seen relative to tobacco use and alcohol and illicit drugs, where less than half of those using these substances may voluntarily admit to their use. Some research protocols have included blood or urine tests. With rare exceptions, blood or urine screening for these substances is not appropriate for routine medical care. As a result, from a health care delivery perspective, there is no easy or complete solution for this problem.

Severity of Illness
Behavioral and chronic medical illnesses have a natural history of progressing from asymptomatic to symptomatic, to more severely ill stages, with or without medical complications. The purpose of both preventive and therapeutic services is to delay or interrupt this progression—perhaps delaying it for the remainder of the natural life of the person. Given this model of illness, the secret to success is either prevention of the illness or early identification to prevent progression of illness. Tracking onset and severity of illness will require data not found in most claimsbased data systems. Some of these data will not even be in the medical record and will require screening and follow-up questionnaires.

Severity of Risk
For purposes of quality assurance and disease management programming, “severity of risk” is a characteristic of each enrolled member separate from severity of illness, and not directly ascertainable from claims data. One major dimension of severity of risk can be defined in terms of a member’s willingness and ability to adhere to prescribed regimens of lifestyle and management of current illness. In this context, “high-risk” members are those who do not quit smoking, take their pills, control their diet, etc. Basically, the higher the risk, the more health plan resources need to be invested to encourage and enable the member to do what is needed to prevent future illness and prevent deterioration of current illness. The common practice of developing educational and fitness programming, but not aggressively marketing it to those most in need of such programming, utterly fails to reach the higher risk members. High-quality case management and preventive programs, including psychoeducation, can be very effective in reaching out to these higher risk members, eliminating barriers to their participation, and enabling their participation.

Efficacy
Projection and Modeling
Assessing the efficacy of preventive services involves projecting what would have happened had the service not been provided. In most cases, this is best done by estimating likely outcomes based on the published literature and the experience of others. Projection and modeling does not necessarily imply reliance on elegant mathematical modeling procedures.

The most practical ways involve use of carefully selected baselines and benchmarks. The usual baseline is incidence or prevalence data from within the health plan. Benchmarks can be secured from a variety of sources, including but not limited to HEDIS, Healthy People 2010, the published literature, and databases presenting State and national averages and State and national survey data. Unfortunately, from the perspective of managed care plans, many of the most useful benchmarks are not parameters discernable from claims data. They require special surveys, medical record reviews, or data that might be secured from electronic medical record systems. Much of the ascertainment of HEDIS compliance is based on highly structured reviews of randomly sampled medical records.

It is generally considered best to establish the baseline and define the benchmarks and objectives prior to initiating new preventive services. Running the preventive program first, then trying to reconstruct the baseline can be difficult and can significantly erode confidence in the results.

Use of fixed baselines and benchmarks is critical to maintaining preventive services after the stage has been reached in which annual differences in process and outcome measures no longer occur, or have been substantially curtailed by the previous success of the preventive programming.

Effectiveness and Cost-Efficiency
Cost-efficiency generally is assessed in two stages. The first is limited to process measures (persons educated, calls, visits, prescriptions filled, etc.) that usually are easy to gather and track on a concurrent basis. These measures will document the activity needed to improve the outcomes.

The second stage focuses on member outcomes (cost-benefit). Given the delays between the delivery of the preventive service and the capture of the benefit in terms of reducing illness and reduction in hospital and emergency room utilization, total reliance on outcome data can give distorted pictures of program efficacy, both positive and negative. Both process and outcome measures may be tracked and validly interpreted. Total reliance on costeffectiveness calculations based on costs of programming and claims-related health care use data can be distorted by changes in membership, delays in submission or processing of health care claims, changes in rate structure, and movement between feefor- service and capitated billing.

Infrastructure

From a management perspective, similar infrastructure elements are needed for preventive services (both medical and behavioral), disease and demand management programming, quality assurance, and utilization review.

A review of the literature indicates the following four major components as most promising for a health plan to develop and maintain the best possible quality assurance and preventive/disease management services:

• Medical leadership, preferably with one or more physicians on a full-time or part-time basis, with personnel who are trained and experienced in epidemiology and population-health management
• Management information support services, surveillance and data systems (MIS) for need ascertainment, program management, and program evaluation
• Staff capacity to manage and interpret the quality assurance and preventive/disease management data
• Financial support and staffing for screening and survey work, provider and patient health education, preventionoriented case management, and prevention-oriented outreach and home visitation.

Research strongly suggests that universal preventive services, both medical and behavioral, will have to be implemented by primary care physicians and their staff assistance. These services represent an additional burden for them to carry and must be recognized in terms of clinical productivity expectations and reimbursement if such services are to be effectively and universally implemented. In the case of the universal preventive behavioral screening procedures, depending on the type of patient and whether one is dealing with an initial or follow-up visit, such screening can be expected to lengthen the clinical encounter between 30 seconds and approximately 2 or 3 minutes. A percentage of the patients will give positive responses to the screening questions and will then require between 2 and 10 minutes of additional time for patient interview and in some cases to arrange follow-up referrals to other programs and professionals.

Quality assurance (QA) and preventive/disease management (DM) services both require the same populationbased approach and same types of data systems. Since they both contribute to member outcomes, and since NCQA, JCAHO, and peer-review organizations consider them together, it is probably best to address them with a combined initiative.

Surveillance and Data Systems
Surveillance is the process by which health care systems identify those in need of preventive services, then follow up to assess the effectiveness of such services. Depending on the disease, the surveillance system might be entirely claims-based, might include detailed pharmacy and laboratory data, and might be by chart review, by member survey, or a combination of these measures.

Not all health care delivery systems have integrated and computerized claims and medical record systems. Given this circumstance, the literature suggests that the most cost-efficient approach to preventive/disease management data systems may be to have small, separate, dedicated systems that can be developed in-house, using spreadsheet or database software, secured free of charge from selected pharmaceutical manufacturers or purchased from a variety of vendors.

According to the research, the most practical approach in most cases will be to develop a registry of patients to be considered for preventive services for each disease or health condition, then track appropriate health status measures, appointments, etc. This would enable the plan to follow up on missed appointments, cue primary care physicians as to needed periodic diagnostic and treatment procedures, track medication compliance, and flag those who may need special attention due to a deteriorating health or risk profile. Such information systems could be created in-house for a single physician group for a single disease, then, after the bugs have been worked out of the system, expanded to other physician groups and other diseases.

Screening Policies and Procedures
Screening is a process intended to identify preclinical illness so that it can be treated early. Sometimes reviewing claims data or charts can do this. Most of the time, however, it requires direct member participation to collect questionnaire data, xrays, and/or laboratory specimens. Screening programs and related health fairs are widespread, but many fail to secure desired health benefits and cost savings for lack of adequate follow-up.

Screening for behaviors or disorders with social stigma raises the issue of invasion of privacy and the issue of unwanted intrusiveness into the life of the member. Such behaviors and disorders include use of tobacco, alcohol and drugs, presence of mental illness in the member or their family, AIDS, sexually transmitted diseases, selfinflicted injury, or injury due to criminal behavior. When addressing these issues, some percentage of members voluntarily will provide this information on interview. For those who will not, proceeding with blood or urine testing or other means of investigation may be warranted in selected cases. Such more intrusive screening probably does not appear justified on a routine basis. This situation, in turn, creates circumstances in which the health care system knowingly will fail to identify a substantial percentage of the cases.

Follow-Up Protocols Postscreening
Research indicates that in the context of preventive and disease management programming, screening without wellestablished postscreening follow-up protocols is likely to be a waste of time (for anything beyond marketing and community relations). Desired follow-up usually entails more detailed clinical assessment, with treatment in some form initiated for those found to be “positive” for the disease.

Counseling
Counseling, which entails providing objective information and helpful advice, is what every health professional does daily in his or her contact with patients. It is an essential part of the care process. Unfortunately, a quick and cursory explanation by the physician or nurse simply will not suffice, even when accompanied by written instructions. In addition, in most health care systems, if the patient has a question after they reach home, it is difficult, if not impossible, to secure the answer from the provider.

Because the usual practice of physician/nurse counseling often is inadequate to assure patient adherence to prescribed preventive measures and regimens of care, each health care delivery system should consider its needs for health education, psychoeducation, case management, call centers, and home visitation to improve patient adherence to medical recommendations for selected groups of high-risk and chronic disease patients.

Psychoeducation
Psychoeducation is health education combined with behavioral counseling. The counseling component of psychoeducation deals with emotions, perceptions, coping, relaxation, and self-care. Psychoeducation is of value for high users of health care services, those with major chronic diseases, and persons scheduled to undergo major surgical procedures. In the case of surgical patients, psychoeducation can enhance early mobility and control of pain.

Health Education (Individual, Group, and Web Site)
Health education is the least intensive of these ancillary services. Although it is sometimes provided one-on-one, it is more often provided as distribution of written material or in group sessions at the health facility or a community site. The goal of health education is to provide the patient with the information needed to secure and maintain the best of health, and to motivate the participants to adhere to the recommendations.

With each passing year, more and more health education is being provided over the Internet, or by other self-paced electronic means. This can be highly effective and costefficient for patients with the computer literacy and motivation to use these modalities.

Seeking health-related information is a major activity on the Internet. The problem from a physician and health-plan perspective is that the information secured may be of dubious quality, may conflict with physician advice, or may simply be wrong. The research indicates that for these reasons, health plans are well advised to develop and maintain their own Web sites, contract with a vendor of such service, and provide health-education pages, with linkage to sources of health information that they consider to be accurate, reliable, upto- date, and consistent with what their physicians are advising their members.

Case Management
Case management is a service usually provided by a nurse or social worker to identify the full range of health needs for a given enrollee and to assure that the service package is both complete and cost-efficient. The most common use of case managers is to control the cost of care for persons with exceptionally high health care costs. Case management usually is conducted in inpatient settings or for patients on longterm home care. Prevention-oriented case management, however, frequently is provided on an outpatient basis and consists of an occasional in-depth assessment and relatively light continued contact with the patient and/or family to assure that appointments are being kept, medications are taken, and questions are answered. Such case management usually is reserved for those with very complex or severe illness, or those who need additional assistance and motivation to reasonably comply with prescribed regimens of care.

Call Centers
Call centers provide several types of services. Most commonly, they provide members with 24-hour telephone access for health advice and guidance. Members can call any time of the day or night with questions concerning how to handle current medical situations or issues. These call centers also can be configured to place outgoing calls for purposes of health education, case management, gathering of survey data, etc. Like most of the other components on this list, a health plan or medical center can either set up their own call center or purchase this service from a national vendor who will customize it to meet the needs of the client health plan or medical center.

Home Visitation
Preventive/anticipatory home visitation for high-risk pregnant women and infants was a common practice in many local health departments and some public sector health care delivery systems through the mid-1970s. Since then, it has all but disappeared in the face of cost containment and a lack of peerreviewed literature demonstrating the efficacy of such services. Over the past 20 years, a new body of literature has been published that clearly demonstrates the value of such home visitation, provided that the mothers and infants are carefully selected, the services are provided in accordance with strict protocols, and the goals of the service are clearly articulated (Eckenrode et al., 2000; Kitzman et al., 1997, 2000; Olds et al., 1986, 1988, 1992, 1993, 1994, 1995, 1997, 1998).

Home visitation is the most expensive service discussed in this report. Research indicates that to be cost-efficient, it must be reserved for only a small number of patients whose specific needs and risk profiles justify this level of service. Home visitation allows the nurse or social worker to observe the home environment, and by doing so, does a better job of providing counseling, education, and case management to empower and enable that patient and family to better manage their risk profile. Routine home visitation is recommended for lowincome, economically and socially vulnerable first-time pregnant women for prenatal and postnatal visits shortly after birth, their newborn infants, and infants born to mentally retarded mothers. Occasional home visits may be helpful to support caregivers in the home caring for patients with dementia or other serious long-term mental or physical ailments. The home visit option might also be considered for all other classes of patients with extreme risk profiles and/or difficulty adhering to prescribed regimens of care.

The home visitation envisioned herein is very different from the home health services currently provided by most health care systems to facilitate early hospital discharge or to substitute for nursing home placement. Home visitation uses nurses or social workers to provide in-home assessment and guidance in dealing with a wide range of medical, social, financial, psychological, and educational issues. This home visitation may include physical examination and health assessment, but it does not involve giving injections, changing bandages, or any other hands-on therapeutic medical service.

TOC | Previous | Next