Differences Between Medicine and Dentistry Drive Plan Design

Medicine and dentistry have many differences, and sound dental plan design recognizes these. These differences include practice location, the nature of care, cost, and emphasis on prevention.

Location

The practice of the typical physician is hospital-based, while many dentists practice almost exclusively in individual offices. Partly because of these practice differences, physicians tend to associate with other physicians with greater frequency than dentists associate with other dentists. This isolation, along with the inherent differences in the nature of medical and dental care, tends to produce a greater variety of dental practice patterns than is the case in medicine. In addition, practicing in isolation does not afford the same opportunities for peer review and general quality control.

Nature of Care, Cost, and Prevention

Perhaps contributing more significantly to the differences in medicine and dentistry are the important differences between the nature of medical and dental care.

First and perhaps foremost, because of the importance of preventive dentistry, the need for dental care is almost universal to ensure sound oral hygiene. Many individuals sometimes require only preventive or no medical care for years. Individuals routinely visit their dentists for preventive dental care, but in medicine the patient typically visits a physician with certain symptoms—often pain or discomfort—and seeks relief.

Dental treatment, because of its emphasis on prevention, often is considered elective. Unless there is pain or trauma, dental care is sometimes postponed. The patient recognizes that life is not at risk and as a result has few reservations about postponing treatment. In fact, postponement may be preferable to some patients—perhaps because of an aversion to visiting the dentist, rooted many years in the past when dental technology was less developed.

Because major dental care is not life-threatening and time-critical, dentists' charges for major courses of treatment often are discussed in advance of the treatment when there is no pain or trauma. As with any number of other consumer decisions, the patient may opt to defer the treatment to a later time or spend the money elsewhere.

A second difference in the nature of care is that, while medical care is rarely cosmetic, dental care often is. A crown, for example, may be necessary to save a tooth, but it also may be used to improve the patient's appearance. Many people place orthodontics into the same category, although evidence exists that failure to obtain needed orthodontic care may result in problems ranging from major gum disease to temporomandibular joint (TMJ) disorders in later life.

A third major difference between the nature of medical and dental care is that dentistry often offers alternative procedures for treating disease and restoring teeth, many of which are equally effective. For example, a molar cavity might be treated by a two-surface gold onlay, which may cost 10 times as much as a simple amalgam filling. In these instances, the choice of the appropriate procedure is influenced by a number of factors, including the cost of the alternatives, the condition of the affected tooth and the teeth surrounding it, and the likelihood that a particular approach will be successful.

There are other significant differences in medical care and dentistry that will have an effect on plan design. These include the cost of the typical treatment and the emphasis on prevention.

Dental expenses generally are lower, more predictable, and budgetable. The average dental claim check is only about $139. Medical claims, on average, are much higher.

The last significant difference is the emphasis on prevention. The advantages of preventive dentistry are clearly documented. While certain medical diseases and injuries are self-healing, dental disease, once started, almost always gets progressively worse. Therefore, preventive care may be more productive in dentistry than in medicine. Certainly the value of preventive dentistry relative to its cost is acknowledged.

Evaluating Demand-Side Approaches

Demand-side approaches to improving quality of care can be considered under the broad heading of "demand management." Demand management has been defined as "the support of individuals so that they may make rational health and medical decisions based on a consideration of benefits and risks." Viewed in this way, traditional health promotion and disease prevention can be regarded as quality of care-related demand management. Much of the attention received by demand management has been directed at controlling utilization and cost of health care. Yet, there is intuitive appeal to the concept of modifying consumer behavior to improve quality of care. There is also some research evidence to suggest such an approach can be effective.

It has long been apparent that providing preventive services is an important element of quality health care. The U.S. Preventive Services Task Force, a panel of medical and health experts appointed by HHS, has published guidelines that have set the standard for quality in preventive care since 1989. Since that time, the NCQA has incorporated measures of delivery of selected preventive services into its HEDIS measures of MCO performance. Clearly, employers can improve the quality of care received by their employees by increasing employee demand for these preventive services.

Research also suggests that consumer-directed decision support, in the form of interactive video, can be effective in improving the appropriateness of medical treatment. This approach, referred to as shared decision-making programs, has produced dramatic changes in patient preferences for treatment of benign prostatic hypertrophy (BPH) or benign enlargement of the prostate gland. Patients with BPH participating in early shared decision-making programs showed a 44 percent to 60 percent reduction in surgery rates, opting more frequently for "watchful waiting" as an alternative. These results suggest the tremendous potential for targeted and well-designed demand management programs to improve quality. For more information on shared decision-making programs contact the Foundation for Informed Medical Decision-Making, Hanover, NH, atwww.healthdialog.com.

Another approach attempting to modify consumer care-seeking behavior has been the dissemination of information about provider quality. This approach has been used by the Minnesota Health Data Institute, the Cleveland Health Quality Choice program, the Foundation for Accountability (FAACT), the Pennsylvania Health Care Cost Containment Council, and others. Schneider and Epstein studied the impact of this approach, as implemented by the Pennsylvania Health Care Cost Containment Council in its Consumer Guide to Coronary Artery Bypass Graft (CABG) surgery. The Guide provided CABG mortality ratings of all cardiac surgeons and hospitals in the state. A telephone survey of patients who had undergone CABG in one of four hospitals included in the Guide revealed that only 12 percent of patients were aware of the Guide, and fewer than 1 percent knew the correct rating of their surgeon or hospital and reported that it had a moderate or major impact on their selection of provider. The authors concluded: "Efforts to aid patient decision-making with performance reports are unlikely to succeed without a tailored and intensive program for dissemination and patient education."

Despite the proliferation of physician report cards, there are few studies indicating that they influence consumer behavior. In a survey of employees in firms participating in the Minneapolis-based Buyers Health Care Action Group, health care consumers reported they were using employer-provided information on satisfaction and service-quality for physicians. A more recent survey of individuals with employer-sponsored health benefits reported that patients remain largely passive consumers of physician services.

More general approaches to demand management have produced suggestive, though less well-documented results. One such approach is telephonic nurse counseling. These services offer telephone access to nurses to discuss health issues in general and answer clinical questions in particular. Vendors of these services purport to be effective in reducing costs and improving appropriateness of health care, and they appear to have convinced a growing number of employers and health plans.

Telephonic nurse case management is also being targeted to patients with specific medical conditions, such as congestive heart failure, diabetes, and asthma. A variety of organizations offer this type of service, including pharmacy benefit management firms, MCOs, hospitals, and others. This approach appears to hold promise for improving compliance with state-of-the-art treatment through improved self-care and patient-provider communication.

The explosive growth of the Internet and its widespread use in the arena of health, suggests that it may be a medium that can contribute to health care quality improvement. Yet, its growth and use have raised a number of new quality-related issues. One study of Internet-derived information on clinical questions found that:

• Eighty-nine percent of retrieved pages were not applicable to the question that prompted the search.
• Fewer than 1 percent of pages consisted of original research or systematic reviews.
• Sixty-nine percent of pages did not indicate an author.
• Only 1 percent of pages provided information on financial or other conflicts of interest.
• Fewer than 18 percent of pages gave the date they were posted or most recently updated.

While the Internet represents a tool with great promise for health care quality improvement, consumers, purchasers and providers should employ the same rigor in evaluating its application as we do for other quality improvement interventions.

The U.S. Department of Health and Human Services, Agency for Health Care Research and Quality has useful resources for consumer decision-making about health care quality. These resources and internet links can be found at www.ahcpr.gov. Another source—NCQA—has developed a suite of tools for employers to assist employees in accessing information on quality of care and using the information to make more informed patient choices. HealthChoices^TM offers an Internet portal, data on quality ratings, custom report cards, and other employee communication tools. These tools can be found at the NCQA web site at www.ncqa.org.

Demand management represents a wide variety of concepts and products with potential application to quality improvement. The most cost-effective of these are likely to be focused on well-defined, measurable target behaviors, and to include education and skill-building, monitoring, and reinforcement of target behaviors. Effective integration of such demand-management programs with supply-management programs will likely bring about the greatest impact on quality improvement.

Evaluating Supply-Side Approaches

The resources required to significantly change provider behavior, whether at the level of the physician, hospital, or MCO, make it unlikely that relatively small purchasers of health care (e.g., individuals and small businesses) acting alone will be successful in driving this approach to quality improvement. However, by banding together in purchasing or policy making, a supply-side quality improvement agenda can be advanced. Business coalitions on health care have proliferated throughout the United States, most with a focus on controlling costs. Many, however, also have addressed issues of quality of care, with some effect. In Michigan, for example, the Southwest Michigan Healthcare Coalition championed the adoption of a uniform hospital database for analyzing severity of illness, health care outcomes, and cost. The information derived and published from the database has been used to identify deficiencies in quality and to inform and monitor quality improvements in area hospitals. The coalition is also active in promoting the concept of a statewide uniform provider database.

Providing feedback on hospital and medical staff performance, with encouragement to initiate quality improvement activities, can produce significant results. This approach was applied by Medicare in its Cooperative Cardiovascular Project, yielding improvements in the use of state-of-theart care for acute myocardial infarction and reducing mortality for this condition.

Large employers and purchasing coalitions can use quality data to selectively contract with providers. A survey of business coalitions found that 35 percent directly contract with providers, and 20 percent contract with "centers of excellence" for high-cost and/or high-risk conditions or procedures. The value of such contracts may be enhanced by incorporating pay-for-performance provisions. JCAHO has published principles that are instructive in designing pay-for-performance programs. These principles can be found atwww.jcaho.org.

While guidelines can be a useful tool in quality improvement, a number of concerns have been raised about the ways in which guidelines are currently developed and implemented. A study of clinical practice guidelines found that fully half of those published did not adhere to established methodological standards.

For information on business coalition activity in your area, contact the National Business Coalition on Health at www.nbch.org.

Larger businesses and health care purchasing cooperatives may have the ability to influence quality of care more directly through their managed care purchasing decisions. By increasing the numbers of covered lives at stake in a managed care bid process, large employers and purchasing cooperatives can generally enhance the responsiveness of MCOs to the quality evaluation described above. This can help ensure the selection of an MCO with superior quality. Ensuring that the MCO will maintain or improve quality of care, however, may require the purchaser to take additional steps.

When contracting with an MCO, the following approaches to promoting CQI are recommended:

• Identify key deficiencies in the MCO's QA/CQI and stipulate that they be remedied in a specified reasonable period. Failure to remedy deficiencies in the agreed-upon period should result in financial penalties to the MCO. In a self-insured, administrative-services-only arrangement, this penalty may be a significant portion of the MCO's administrative fee (e.g., 10 percent). In a fully insured arrangement, the penalty may be cost sharing by the MCO in noninsured, employer health-related costs (e.g., worksite health promotion/disease prevention).
• Specify reliable and valid measures to be used to track MCO quality over the life of the contract. Ideally, these will be measures already tracked by the MCO and will include appropriateness of care, excellence of care, and satisfaction. It may be necessary to stipulate that the MCO adopt new measures, or to hire an independent organization to do the MCO quality measurement.
• Require periodic reporting of the above quality measures and track the MCO's performance. Arrange to meet with key MCO staff to review the reports. Financial penalties and rewards should be specified in the contract for failing to meet or exceeding agreed-upon targets for improved performance, respectively.

By monitoring MCO performance in routine reports, providing feedback in periodic meetings, and reinforcing CQI with financial rewards and penalties, employers can continue to enhance the value of their health care expenditures over the life of an MCO contract. This approach has been taken by the Pacific Business Group on Health in negotiating more than two dozen performance guarantees with 13 California HMOs. Of more than $8 million at risk for meeting performance targets, nearly $2 million was refunded for sub-par performance. Eight of 13 plans missed their targets in the area of childhood immunization. Most plans met or exceeded their targets in such areas as satisfaction, cesarean section rates, mammography, Pap smear, and prenatal care.

Evaluating Managed Care Organization Quality

One of the great potential advantages of purchasing health care through a managed care organization is the cost-effective ongoing quality assurance and continuous quality improvement that these plans can provide. The question for the employer/purchaser of an MCO is how to evaluate the quality of its supplier's QA/CQI programs. One approach to this question is to look for accreditation by an independent organization that has evaluated the quality of the MCO. Today, there are two major accrediting organizations for MCOs: the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) and the National Committee for Quality Assurance (NCQA).

The NCQA is the most experienced of the MCO accrediting organizations. It has reviewed a majority of MCOs in the United States. The NCQA accreditation process involves a review of MCO quality-related systems, including quality improvement, processes for reviewing and authorizing medical care, quality of provider network, and members' rights and responsibilities. Documentation of these processes provided by the MCO are analyzed, and a site survey is conducted involving both physician and administrative reviewers.

In addition to its process-oriented assessments, the NCQA has developed the Health Plan Employer Data and Information Set (HEDIS®) to help standardize the measurement and reporting of health plan performance. HEDIS measures have become the basis of performance measures produced by many health plans and purchasing coalitions.HEDIS has measures applicable to commercial, Medicaid, and Medicare plans. In the area of effectiveness of care, the measures include such items as breast cancer screening, controlling high blood pressure, and follow-up after hospitalization for mental illness. In the area of access/availability of care, measures include getting needed care and getting care quickly. In the area of satisfaction, HEDIS incorporates the Consumer Assessment of Health Plans (CAHPS) instrument—a reliable and valid survey and reporting kit developed by a consortium of Harvard Medical School, the RAND Corporation, and the Research Triangle Institute under the sponsorship of the Agency for Health Care Policy and Research.

Based upon these reviews, and MCO performance on HEDIS and CAHPS, the MCO is granted one of the following levels of accreditation status:

• Excellent: Demonstrated performance that meets or exceeds NCQA requirements for consumer protection and quality improvement (QI), and HEDIS results among the highest scoring plans nationally or regionally.
• Commendable: Demonstrated performance that meets or exceeds NCQA requirements for consumer protection and QI.
• Accredited: Performance meeting most of NCQA's requirements for consumer protection and QI.
• Provisional: Compliance with some, but not all, of NCQA's consumer protection and QI requirements.
• Denied: Failure to meet NCQA requirements for consumer protection and QI requirements
• Suspended: NCQA accreditation for a plan has been withdrawn to investigate and implement corrective action
• Under Review: Accreditation is under review at the request of the plan
• Discretionary Review: NCQA is assessing the appropriateness of a plan's current accreditation status.

NCQA has a similar process for accreditation of PPO plans based upon assessments of access, service, and qualified providers. PPO accreditation levels are Full, One-Year, Provisional, and Denied.

The performance of health plans can be compared using accreditation level and HEDIS measures. Individual plan performance and performance benchmarks can be accessed through NCQA's Quality Compass (see www.ncqa.org). NCQA has also created an economic model for projecting the comparative performance of health plans. The Quality Dividend Calculator^TM projects cost savings that an individual employer can expect from choosing a high quality MCO. Projections are based upon how health care quality as measured by HEDIS reduces absenteeism and increases productivity among employees. The calculator is also available through the NCQA web site. It should be noted that not all health plans collect and publish HEDIS data, and although NCQA audits HEDIS data, the data are collected and analyzed by the health plans themselves, with the potential for bias that is inherent in this approach.

The approach of the Joint Commission on Accreditation of Healthcare Organizations to accrediting MCOs is comparable to that of the NCQA and results in assignment of an MCO to one of the following categories of accreditation: provisional accreditation, accreditation with commendation, accreditation with or without recommendations, conditional accreditation, or nonaccreditation.

In assessing health plan quality of care, it would be worthwhile to ask the following questions:

• Has your MCO applied for accreditation from either NCQA or the Joint Commission on Accreditation of Healthcare Organizations?
• If so, when was your most recent review, and what category of accreditation did your MCO receive?
• Will the MCO provide a summary of the findings of the accreditation process?

A list of health plans reviewed by NCQA for accreditation is available on-line at http://www.ncqa.org. While reviewing the results of these accreditation processes can be informative, the accreditation organizations explicitly warn that they do not warranty any third parties (e.g., employers) regarding the quality of care of an MCO. In addition, many MCOs have not yet undergone accreditation. Therefore, whenever an employer or employee is purchasing MCO services, it would be advisable to do some additional evaluation, including contacting your state departments of insurance and/or public health, reviewing some minimal documentation related to MCO quality, and making a site visit.

State governments generally have some regulatory authority over MCOs operating within their borders. This regulatory authority may reside with the department of public health, the department of insurance, or some combination of these. A call to one or both of these agencies in your state, asking for information about the status of a particular MCO, can be informative. If the MCO of interest is an HMO, you may want to ask for a copy of the HMO's annual report, which must be filed with the state department of insurance.

Requesting and reviewing the following information from the MCO also can be helpful:

• Credentialing criteria/processes for network physicians, hospitals, and ancillary providers (e.g., laboratory, X-ray, home health agencies): Do these criteria and processes include those mentioned above under physician and hospital quality? Are provider credentials verified by the MCO, or do they accept a provider's self-report? How frequently are providers recredentialed? Does the recredentialing process include routine, systematic consideration of member complaints, member satisfaction, and other quality indicators?
• A copy of the most recent quality assurance, quality management, or CQI plan and annual report (individual provider and patient identifiers can be removed to protect confidentiality): Does the plan include reliable and valid measures and standards of appropriateness of care, excellence in care, and satisfaction with care as described above? Are providers educated about these measures and standards? Are performance measures documented and routinely fed back to providers? Is meaningful reinforcement and support provided for performance improvement? Are there credible, specific documented examples of performance improvement over the preceding year?
• Routine provider quality profiles (i.e., sample reports on provider performance routinely analyzed by the MCO): How reliable, valid, and useful to quality improvement are the data contained in the reports? To what extent has the quality performance monitoring described in the QA plan been incorporated into MCO reporting systems?
• Reimbursement formula for physicians in the MCO: Are there substantial financial incentives for physicians to withhold necessary care? Conversely, are there substantial financial incentives for physicians to provide quality care? (It has been this author's observation that MCOs providing such financial incentives are more likely to have reliable and valid measures of physician quality and systems for monitoring and feedback of these measures.)
• Preventive care programs offered and participation rates: What preventive care programs does the MCO offer, at what location, and with what frequency? What member cost sharing, if any, is required? What are the participation and success rates for these programs?
• Plan-wide measures of quality: Will the MCO provide the most recent report of performance using HEDIS measures? Did it use survey instruments recommended in HEDIS for assessing member satisfaction and health status? If not, how did it ensure the reliability and validity of the instruments? What were the response rates to these surveys?

An additional step that can be immensely helpful in assessing the quality of an MCO is to conduct a brief site visit to "kick the tires." In this author's experience, it is not uncommon to come away from such a visit with an entirely different assessment of MCO quality than is conveyed in written material from the organization. Consultants with some knowledge of managed care can be helpful but are not necessary. For a site visit to be most helpful, the following guidelines are recommended:

• Allow four to eight hours for the visit.
• Try to limit the time devoted to marketing and formal presentations.
• Arrange to meet key staff, including the medical director and the heads of member services, quality assurance, utilization management, and finance: What is their relevant training and experience? Are they credible and involved? What is their level of commitment?
• Devote the most time on site to direct observation and questioning of MCO operations staff, and listening to staff on the telephone in member services, claims administration, and utilization management: What is their relevant training and experience? What is their level of commitment? What is the quality of their customer service? Do they document members' complaints, concerns, and questions, and follow up? Do you see signs of a pervasive CQI program with posted performance standards and measures?
• Discuss quality-related information provided prior to the site visit (see above). What are the processes for collection and quality control of data? What were the most successful improvement initiatives in the preceding year? Review minutes of the most recent quality assurance committee meetings.
• Assess the philosophy of the MCO: Is it a good fit with your own and that of your organization? Is the MCO interested in you as a customer, your quality concerns, and your business needs?

Patient, member, and/or physician confidentiality should not be a barrier to conducting a site visit as long as reviewers are willing to sign confidentiality agreements.

For larger employers, all of the above elements can be incorporated into a formal competitive bid process involving multiple MCOs. A nationwide effort, spearheaded by the National Business Coalition on Health and Watson Wyatt Worldwide, has been joined to gather MCO responses to a standardized Request For Information (RFI). Information on this initiative, named eValue8, is available at www.evalue8.org.

Other resources to consider when evaluating MCO quality include the following:

• National Committee for Quality Assurance (NCQA), Washington, DC. www.ncqa.org
• America's Health Insurance Plans (AHIP), Washington, DC. www.ahip.org
• National Coalition on Health Care, Washington, DC. www.nchc.org
• Institute for Health Care Improvement (IHI), Roxbury, MA. www.ihi.org
• Centers for Medicare & Medicaid Services, Baltimore, MD. www.cms.hhs.gov
• Agency for Healthcare Research and Quality (AHRQ), Rockville, MD. www.ahrq.gov
• Foundation for Accountability, Portland, OR. www.facct.org

Evaluating MCO quality, like physician and hospital quality assessment, can be a time-consuming process. Yet, this may be a relatively small investment of time when weighed against the resources spent by employer and employee on health care and the risks posed by the purchase of poor-quality health care.

Evaluating Hospital Quality

Some of the same approaches to quality assessment described for physicians can be applied to hospitals. A useful starting place for assessing a hospital's quality is its accreditation. Accreditations to look for include these:

• Current, unrestricted license from the state.
• Current, unrestricted, nonprobationary accreditation from the Centers for Medicare & Medicaid Services for participation in Medicare and Medicaid.
• Current, unrestricted, nonprobationary accreditation from the Joint Commission on Accreditation of Healthcare Organizations.

The Joint Commission on Accreditation of Healthcare Organizations has an extensive process for assessing hospital quality with an on-site survey. Beginning January 1, 1995, the Joint Commission made available summaries of the results of its new surveys. These summaries, however, are brief and offer only general information. In addition, the Joint Commission's surveys were at one time criticized by the Inspector General of the U.S. Department of Health and Human Services as "unlikely to detect patterns, systems, or incidents of substandard care." Among the improvements implemented by the Joint Commission in an effort to address these concerns are the inclusion of outcomes measures in its review process, such as acute myocardial infarction, congestive heart failure, and complications of surgery.

General information on hospital facilities, personnel, and services are published annually by the American Hospital Association. This information can sometimes be helpful in making inferences about quality for particular conditions or procedures. For example, if you are having a high-risk delivery, you may wish to choose a hospital that has an advanced level nursery, including a dedicated neonatal intensive care unit. Or if you are planning a percutaneous transluminal coronary angioplasty, you might be well-advised to choose a hospital that offers high-quality emergency coronary artery bypass graft surgery, in the event it may be required.

The Centers for Medicare & Medicaid Services makes data publicly available on hospital performance through its Medicare Provider Analysis and Review (MEDPAR) files. In some states (e.g., Pennsylvania and New York), data are publicly available on hospital performance for specific conditions and procedures. These data can include the volume of cases, outcomes (mortality and complication rates), average length of stay, and average cost per case. Whether or not such data are publicly available for the condition or procedure of interest to you, you may wish to consider approaching the hospital administration directly with the following questions:

• What is the hospital's volume of admissions for the condition/procedure of interest?
• What is the complication/mortality rate for the condition/ procedure as performed at the hospital?
• What is the success rate for the treatment/procedure at the hospital?
• What is the average length of stay for the condition/procedure?
• What are the results of your patient satisfaction survey for the most recent period (including response rate)?
• Does the hospital participate in any managed care networks (e.g., HMO, PPO, or POS plans)?
• Has the hospital been designated as a center of excellence for the condition/procedure by a health plan?

The final question will apply to only a small number of conditions/ procedures and hospitals. Nevertheless, one can find designated regional and national centers of excellence for high-risk, high-cost conditions/ procedures, such as organ transplantation, open-heart surgery, and burns. The National Institutes of Health also designates research centers for selected conditions. One might postulate that these centers are more likely to provide quality care for these conditions because of their successful research programs.

One variable to consider in assessing hospital quality is whether it is a major teaching hospital (defined as more than 0.097 teaching residents per hospital bed set up and staffed for patient care). Such hospitals have been found to have a lower risk of death than other hospitals, when evaluated for mortality due to hip fracture, stroke, coronary heart disease, and congestive heart failure.

The question pertaining to volume of patients treated with a particular condition or procedure can be extremely useful as a surrogate measure of quality. More than 20 years of research and dozens of published studies have linked better outcomes to hospitals and doctors delivering higher volumes of particular health care services. Research has confirmed the link of high volume to better outcomes in acute myocardial infarction (hospitals with more than 6.3 Medicare patients with acute myocardial infarction per week on average), major cancer surgery (hospitals with more than one Medicare patient per year on average for a given procedure), and carotid endarterectomy, surgery removing blockages from the carotid arteries to prevent stroke, (hospitals with more than 62 Medicare patients undergoing the procedure per year).

In interpreting hospital satisfaction survey results, it is important to consider the validity of the survey instrument and response rates. Ask whether the survey is based on a standard instrument that has been evaluated for its reliability and validity. If the survey has a response rate of less than 50 percent, the results should be considered suspect. Research suggests that nonrespondents to such surveys have lower levels of satisfaction than respondents.

HealthGrades.com Inc. is an organization that provides hospital quality rating information on the world wide web (www.healthgrades.com). Much of the HealthGrades quality rating system relies on publicly available Medicare data. A study of the ability of HealthGrades ratings to discriminate between individual hospitals in the processes and outcomes of their care was published by a team of researchers at Yale University. They found that HealthGrades ratings could accurately identify groups of hospitals that performed better in quality than other hospital groups. HealthGrades ratings did poorly, however, when it came to discriminating between two individual hospitals on their processes of care or mortality performance. The Leapfrog Group is an organization of large purchasers of health care that strives to create big leaps in health care safety, quality and value. The group recognizes hospitals conforming to their standards, publishing their performance on a number of measures of safety, quality and value on their website at www.leapfroggroup.org.

Additional resources to consider when evaluating hospital quality include the following:

• American Hospital Association, Chicago. www.aha.org
• Centers for Medicare & Medicaid Services, Baltimore, MD. www.cms.hhs.gov
• Joint Commission on Accreditation of Healthcare Organizations, Oakbrook Terrace, IL. www.jcaho.org

Assessing hospital quality, both initially and on an ongoing basis, can be a labor-intensive process. As in the case of physician quality assessment, this kind of assessment and more should be obtainable with economies of scale through a quality health plan offering a provider network

Evaluating Physician Quality

Most assessments of physician quality begin with the physician's training, experience, and professional certifications. The literature on the link between these factors and quality is limited. Nevertheless, these characteristics can serve as a starting point for evaluating a physician's level of knowledge and skills, which we might postulate would be related to the appropriateness and excellence of his or her practices. In addition, a review of physician credentials might reveal that small proportion of physicians for whom glaring quality-of-care problems have been identified. Characteristics to consider in this assessment include the following:

• Current unrestricted license to practice in your state.
• Current unrestricted license to dispense prescription drugs from the state and the Federal Drug Enforcement Administration.
• Certification by a specialty board recognized by the American Board of Medical Specialties.
• Current active, unrestricted hospital staff privileges.

The latter of these criteria may not apply to physicians who choose not to see patients in a hospital setting. It may be difficult, however, to determine if a physician's privileges were dropped as a result of his or her own choice or because of a quality-driven decision by the hospital. The advantages of using a physician with hospital privileges include having continuity of both inpatient and outpatient care and having the benefit of the hospital's QA and/or CQI program apply to your physician. This latter benefit includes hospital access to the National Practitioner Data Bank, a national database on physician quality problems that is not accessible to the public.

Conspicuously absent from the above list is malpractice experience. There are questions about the extent to which malpractice experience is a reflection of physician quality. On the other hand, research indicates that any history of malpractice claims, paid or unpaid, is associated with an increased likelihood of future claims. Therefore, it may be worth evaluating a physician's malpractice claim history, if only to reduce your risk of being involved in a future malpractice claim.

A physician's credentials can be evaluated directly by employees, by benefit managers, or by health plans. The following are some of the resources for employees and benefit managers to consider in conducting such an evaluation:

• American Board of Medical Specialties. www.abms.org
• American Medical Association, www.ama-assn.org
• HealthGrades, www.healthgrades.com
• Federation of State Medical Boards, www.docinfo.org

The state physician licensing board is a good place to look for answers to questions about the state licensing status of individual physicians. Most states have such information available on-line. A review of information available from state licensing boards has been assembled by Public Citizen Health Research Group and can be found at www.citizen. org/hrg. Some malpractice claim information on individual physicians may be available from the court clerk in the jurisdiction(s) where the physician has practiced.

The quality-related issues described above pertain to all physicians, regardless of their specialty. When assessing physician quality as it relates to specific diagnoses or conditions, additional factors should be considered. For example, physicians being evaluated for their quality in performing a particular surgical procedure should be asked such questions as these:

• What kind of advanced training and/or certification has the physician had in performing the procedure?
• What is the annual volume of the procedure performed by the physician?
• What is the complication/mortality rate for the procedure as performed by the physician?
• What is the success rate for the procedure as performed by the physician?
• What is the average length of hospital stay for the procedure?
• What is the average length of disability following the procedure?

The applicability of these and other questions will vary by specialty, condition, and procedure. Generally speaking, however, the quality of a physician's performance, as in the example of percutaneous transluminal coronary angioplasty described above, is related to the frequency with which he or she performs the procedure. For some conditions and procedures, there may be regional or national research centers or centers of excellence. Helpful resources in learning about such centers, and obtaining consumer information about various health issues include the following:

• National Cancer Institute, Cancer Information Service. Tel: 800-4-CANCER
• American Cancer Society local affiliates
• American Heart Association local affiliates 
• American Lung Association local affiliates
• National Institute of Mental Health. Tel: 800-421-4211

The National Committee for Quality Assurance, in collaboration with the American Heart Association, American Stroke Association and American Diabetes Association, has developed programs to recognize physicians demonstrating that they provide high quality care for patients with selected common chronic conditions, including diabetes mellitus, cardiac conditions, and stroke. Information about these programs and a database of recognized physicians is available at www.ncqa.org. These recognitions are being tied to rewards in four metropolitan areas through Bridges to Excellence—a coalition of employers, physicians, health plans and patients. Coalition members agree to financial incentives to be paid to physicians for providing high quality care as demonstrated through the physician recognition programs described above. These programs are part of a movement toward pay-for-performance in health care, that can be expected to grow in breadth and depth in the years to come. The Centers for Medicare & Medicaid Services, for example, is conducting the Physician Group Practice demonstration combining Medicare fee-for-service payments with a bonus pool to reward improvements in the management of care and services.

If your physician participates in any managed care programs, he or she may receive periodic performance report cards from the MCO and may be willing to share the results with you. Some of these results may be published. For example, the Pacific Business Group on Health has published performance data on medical groups that make up California health plans (though the data are not broken out by individual physician). The Pennsylvania Health Care Cost Containment Council published heart attack mortality data for physician groups. While "report cards" on physicians promise to be increasingly available, they should be interpreted with caution. A number of potential pitfalls with such reports have been identified. For example, multiple physicians may participate in a patient's care, making it difficult to assign primary responsibility for the patient's outcome to any one physician or medical group. In a study of physician report cards for diabetes care, Hofer and colleagues found that they were unable to reliably detect true practice differences among physicians at three practice sites. They also found that physicians could easily "game" the reporting system by avoiding or deselecting patients with high prior cost or with poor adherence or poor response to treatment.

Finally, there is a large body of research suggesting that physician– patient communication is related to the quality and outcome of care. Perhaps the best way to evaluate a physician's communication skills is to do so firsthand, scheduling an office visit to get to know a physician you may not already be familiar with. If you make such a visit, it may be helpful to prepare both general questions and questions particular to your circumstances in advance of your appointment.

Investigating even this minimum set of criteria for physician quality care requires a significant investment of time and resources. And such assessments should be repeated periodically to ensure that there has been no change in physician status. The extensive nature of this undertaking points to one advantage of purchasing medical care from a health plan that includes a network of providers. The various aspects of physician quality described above and others can be consistently and rigorously assessed by the plan on an ongoing basis, with associated economies of scale.

Is Quality Of Care Important?

Quality of health care is an important issue for employer and employee purchasers for a number of reasons. First, there are widespread documented errors in the delivery of health care services. Second, there is substantial evidence for extensive overuse and underuse of various health care services. Third, poor quality of care erodes the value of health care purchases. Fourth, failure to exercise due diligence in evaluating quality of care may impact an employer's liability for a bad outcome of care. And, finally, lack of attention to quality of care can have negative consequences for an employer in employee relations and relationships with providers and others in the local business community. 

Errors in the Delivery of Health Care Services

More than a decade ago, the Harvard Medical Practice Study found that injuries caused by medical management occurred in 3.7 percent of hospital admissions in New York State. Among these injuries were drug complications, wound infections, and technical complications. Fully 27.6 percent of these injuries were the result of negligence, and 13.6 percent of the injuries led to death. Extrapolating these results to all U.S. hospital admissions in 1997, as many as 98,000 Americans may have died because of errors during their hospitalization in a single year. Other studies have confirmed the order of magnitude of this estimate for injuries during hospital admissions. Yet, we must assume this is a gross understatement of the impact of medical errors, given that it does not include injuries because of outpatient care.

The Institute of Medicine focused attention on the impact of errors in medicine through the work of its Quality of Health Care in America Project. Its first published report—To Err is Human: Building a Safer Health System—notes: "More people die in a given year as a result of medical errors than from motor vehicle accidents (43,458), breast cancer (42,297), or AIDS (16,516)." The group estimated that preventable adverse events resulted in total national costs of between $17 billion and $29 billion, over one-half of which are direct health care costs. More recently, a study of 20 percent of all Medicare hospital admissions in the year 2000 found that patient safety errors resulted in excess lengths of stay ranging from 1.34 days for accidental puncture or laceration to 10.89 days for post-operative sepsis. The excess charges associated with these errors were $8,271 and $57,727, respectively. Excess mortality was also associated with these patient safety errors, from 2.16 percent for accidental puncture or laceration to more than 21 percent for postoperative sepsis. Clearly, medical errors have a significant negative impact on employer health care costs, as well as employee health outcomes and productivity. These findings beg the question of what employers and individuals can do to help minimize the likelihood and the impact of medical errors in the health care services they purchase and receive a question to be taken up later

Overuse of Health Care Services

Investigators have long noted dramatic geographic variations in the use of health care services, without apparent differences in the health of the populations being served. For example, one study showed that Medicare hospitalization rates were 60 percent higher in Boston than in New Haven, yet Medicare mortality rates did not differ between the two cities. A recent study of Medicare end-of-life spending found that beneficiaries in higher spending regions of the United States received approximately 60 percent more care than beneficiaries in lower spending regions, without finding consistent differences between these groups in quality of care or access to care.

There is a large and growing body of research on the extent of medical care that is inappropriate or unnecessary. Studies of appropriateness of care have found that as much as 32 percent of selected procedures are inappropriate. An excellent example of research supporting this estimate is the series of studies commissioned by the State of New York Cardiac Advisory Committee on the appropriateness of various cardiac procedures in New York State. Evaluation of coronary angiographies (inserting a catheter into coronary arteries and injecting contrast material) found that 20 percent were of uncertain appropriateness and 4 percent were clearly inappropriate. When percutaneous transluminal coronary angioplasty (PTCA) (using a balloon catheter to open blood flow through a coronary artery) was evaluated, 38 percent were of uncertain appropriateness and 4 percent were clearly inappropriate. At some hospitals, as many as 57 percent of PTCAs were either inappropriate or of uncertain appropriateness. In a companion study, inappropriate and uncertain use of coronary artery bypass graft surgery was found to be 2.4 percent and 7 percent, respectively. Though these rates may appear relatively low, they have significant health implications, given that the average mortality rate for patients undergoing surgery in the study was 2 percent, and the complication rate was 17 percent.

These are but a few examples of research suggesting that inappropriate and unnecessary medical care has substantial negative consequences both for employee health and the cost of health care.

Underuse of Health Care Services

Another deficiency identified in health care quality is the failure to apply services known to be beneficial in improving health. In a study of patients hospitalized for acute myocardial infarction (heart attack), Marciniak and colleagues found that between 11 percent and 68 percent of patients nationwide did not receive particular standard treatments for this condition, despite being "ideal candidates" for therapies. An earlier study found that internists and family physicians were less knowledgeable about, and less inclined to practice, state-of-the-art advances in treatment of acute myocardial infarction than were cardiologists.

Similarly, a study of patients with diabetes treated in primary care offices found that between 55 percent and 84 percent of these patients did not receive optimal services recommended for their condition according to national guidelines in use. Optimal use of services varied by location of practice by as much as 238 percent. A recent study of 439 indicators of quality of care for 30 conditions and preventive care among a random sample of adults in 12 U.S. metropolitan areas found that only 54.9 percent received recommended care. Quality of care varied by medical condition, from a high of 78.7 percent of recommended care received for senile cataract to a low of 10.5 percent of recommended care for alcohol dependence.

Some studies suggest that physicians are more likely to underuse health care services when treating women, particularly black women. Research by Roger, et al. found that women with unstable angina (chest pain from blockages in arteries supplying blood to heart muscle) were 27 percent less likely to undergo non-invasive cardiac tests, and a startling 72 percent less likely to receive invasive cardiac procedures. Even within Medicare managed care health plans, black plan members were found to receive poorer quality of care than white plan members, specifically for eye examinations for patients with diabetes, for beta-blocker use after heart attack, and for follow up to hospitalization for mental illness.

There is also evidence to suggest that underuse varies by type of health plan. For example, a study found that Medicare patients with joint pain who were enrolled in HMOs reported less improvement in symptoms than similar fee-for-service Medicare beneficiaries. Yet, other research suggests no significant difference between quality of care in HMO and fee-for-service environments in such areas as hypertension and diabetes.

Failure to apply services known to be beneficial in improving health is a substantial and widespread problem. Clearly, this type of quality problem has negative implications for employee health and productivity. The implications for cost of care are more variable, because some of the underused services may result in a net increase in direct medical care costs, despite being effective in preventing negative and costly health outcomes. Nevertheless, in purchasing health care for ourselves or for employees, these are services we would want to receive as part of state-of-the-art quality in health care delivery. Whether one looks at quality from the perspective of individual providers, practices, or health plans, these landmark studies shed new light on deficiencies in quality of care, and suggest how appropriate to health care is the maxim: "Let the buyer beware."

Employer Liability

One reason employers should be concerned about health care quality is their potential liability for managed care programs they may purchase. The evidence for payer liability for managed care stems from two legal cases: Wickline v. State of California and Wilson v. Blue Cross of Southern California. In Wickline, the court concluded that a third-party payer can be legally liable for negligence in utilization review decisions. In Wilson the court determined that a third-party payer cannot escape liability for negligent utilization review based on the argument that the treating physician bears all legal responsibility for a hospital discharge decision.

In Fox v. Health Net, the estate of Nelene Fox was awarded damages from a managed care organization for its refusal to cover bone marrow transplantation and high-dose chemotherapy for advanced breast cancer. The total jury award was $89 million, $77 million of which was punitive damages. Some observers believed that the impact of this case on the liability of employers and MCOs was limited by the Employee Retirement Income Security Act (ERISA).

In the case of Goodrich v. Aetna U.S. Healthcare, a jury awarded the widow of David Goodrich $120 million for delays in approving coverage of high-dose chemotherapy and bone marrow transplantation for a form of stomach cancer. Although ERISA did not apply to this case because Goodrich was covered under a state-sponsored health plan, the size of the award and legislative action in 10 states to limit ERISA protection raised concerns about the exposure of MCOs and employer-sponsored health plans to litigation and resulting damage awards. More recent developments have mitigated these concerns. A 2003 study of the impact of state managed care liability statutes found that these statutes had produced no appreciable increase in liability exposure. In a 2004 ruling by the U.S. Supreme Court in Aetna v. Davila, the court essentially voided these statutes and found that employer-sponsored health plans cannot be held liable for damages for denial of coverage. While these developments represent a shift toward reduced liability of third-party payers for poor quality, negligent managed care processes, it nevertheless seems prudent for employers to address the issue of quality of care in their medical benefits plans as a matter of risk management.

Employee, Provider, and Community Relations

When medical benefits decisions are made without substantive consideration of quality of care, employees can take away the message that their health and well-being are not valued by their employer. This message can undermine one of the key objectives of offering medical benefits: to promote the recruitment and retention of employees. Incorporating quality assurance and continuous quality improvement (CQI) processes into medical benefits decisions, and effectively communicating these processes to employees, can help avoid the employee relations pitfall of employee dissatisfaction with their medical benefits.

Disillusioned providers also can undermine the extent to which employees value their medical benefits. Employee opinion may be influenced by negative assessments from physicians about the quality of an employer's health plan. In addition, physician performance may be adversely affected by a poor quality health plan, with consequences for employee health and productivity.

Failing to demonstrate a commitment to quality assurance and CQI in health care decisions can leave employers vulnerable to the charge of neglecting corporate social responsibility as well. This can have obvious negative implications for community relations.

Value of Medical Care Expenditures

The value of health care services can be defined as the health benefit per dollar spent. Chassin and the National Roundtable on Health Care Quality observed that errors in the delivery of health care services, as well as overuse of services, can reduce the value of health care services by both decreasing the numerator and increasing the denominator of this equation. Conversely, by reducing errors and overuse, the value of health care services can be increased. (The impact of underuse on value is more variable, as it tends to move the numerator and denominator in the same direction.) Most businesses would not view as prudent the practice of purchasing from suppliers based upon price alone. When viewing health plans and providers as you would view other suppliers to your business, considerations of quality and service, as well as cost, become essential components of the value equation.