Availability of Nutrition Services for Medicaid Recipients in the Northeastern United States: Lack of Uniformity and the Positive Effect of Managed Care

Objectives: To evaluate third-party coverage of medical nutrition therapy for the Medicaid population, among whom obesity is a common health problem, and to compare coverage in managed care Medicaid programs with that in fee-for-service state Medicaid programs.

Methods: Fifty-four Medicaid organizations in 14 states were surveyed by telephone and asked about their provision of nutrition counseling to members.

Results: Overall, similar percentages of state Medicaid programs and Medicaid managed care organizations (MCOs) reimbursed for nutrition counseling; significantly more Medicaid MCOs reimbursed for this service for treatment of obesity alone. Analysis of Medicaid MCO responses by plan size failed to show a difference in the nutrition services offered.

Conclusions and Recommendations: Nutritionist consultation, an inexpensive and effective intervention for treatment of obesity, is not uniformly available to Medicaid patients. The inclusion of managed care in Medicaid has increased access to this service in the northeastern United States. We propose that all Medicaid recipients should have access to visits with a registered dietician or certified nutritionist either as part of a defined benefit structure or through a disease management program for obesity.

(Am J Manag Care 2003;9:817-821)

Obesity is a common health problem among the low socioeconomic status (SES) populations of developed nations. The World Health Organization defines a body mass index (BMI) of 25.1-29.9 kg/m2 as "overweight" and a BMI of â„¢30 as "obese" for adult men and women.1 In the United States, the recent guidelines on overweight and obesity published by the National Heart, Lung, and Blood Institute concur with this definition.2 Consultation with a registered dietician (RD) or certified nutritionist (CN) is a low-cost intervention that, as part of a weight management program, can help obese patients lose weight.^3-5 As our Medicaid managed care organization (MCO) in Philadelphia cares for a low SES population, we wished to evaluate access to nutrition services for Medicaid patients, both in Pennsylvania and in neighboring states.

METHODS

State Medicaid organizations and Medicaid MCOs in 14 states in Centers for Medicare & Medicaid Services regions 2 and 3 (New England and the mid-Atlantic states, with the exception of Puerto Rico and the Virgin Islands) were surveyed by telephone. Of the 14 states surveyed, 13 had MCOs that administer care to at least 25% of the Medicaid patients in their state. Calls were made by a single interviewer to state Medicaid offices and to either the customer services department or the provider relations department of Medicaid MCOs in each of the states. When possible, department managers at the MCOs were surveyed. Multiple MCOs were surveyed in each state so that we accounted for at least 50% of the total Medicaid enrollment in each state. Each Medicaid organization was asked the same 3 questions:

1. Is consultation with an RD/CN available for Medicaid patients in your state/in your plan (yes or no)?

2. Is obesity alone, without another diagnosis such as diabetes, hypertension, or high cholesterol, sufficient for referral to an RD/CN (yes or no)?

3. Are other non-pharmacologic services offered to your obese patients as part of a disease management program for obesity, such as discounted access to exercise facilities or to commercial weight loss classes such as Weight Watchers (yes or no)?

Statistical Analysis

P

P

P

For each 2×2 comparison, Fisher's exact test of equality for 2 binomial distributions (2-sided) was performed by using StatXact Software version 4.01 (Cytel, Inc, Cambridge, Mass). Because of the small sample size, data were characterized as being consistent with the null hypothesis (no difference in frequency of "yes" answers) whenever the associated value exceeded .20. Data were characterized as inconsistent with the null hypothesis when the value was ˜.05 and as suggestive of a possible association whenever the value fell between 0.05 and 0.20.

RESULTS

P

Of the 14 state Medicaid programs, 8 (57%) provided consultation with an RD/CN, but only 1 (7.1%) provided the service with obesity alone as the diagnosis. Of the 40 Medicaid MCOs, 31 (77.5%) provided RD/CN consultation; 16 of these (40%) provided the service for obesity alone. One state Medicaid program (7.1%) and 11 Medicaid MCOs (27.5%) stated that they reimbursed for either Weight Watchers or gym membership. A small number of Medicaid MCOs (<10) refused to release information on services reimbursed because of concerns about competitive intelligence gathering. See Table 1 for a comparison of state Medicaid programs and Medicaid MCOs with percentages and accompanying values.

The mean Medicaid enrollment for the 40 Medicaid MCOs surveyed was 56,081. Eighteen (45%) of the MCOs had enrollments larger than this ("large plans"). Twenty-two (55%) of the MCOs had enrollments smaller than this ("small plans"). When responses of large and small MCOs were compared, there were no statistically significant differences between the level of services offered (see Table 2).

DISCUSSION

Within developed countries, obesity is more prevalent in low SES populations. Defined as a BMI of â„¢30, the prevalence of obesity among adults in the United States, estimated conservatively by telephone survey, is 19.8%; among those with less than a high school education, the prevalence is estimated to be 26.1%.6 The prevalence of obesity in the Medicaid population is likely to be closer to this second estimate. An additional 36.6% of the general population is estimated to be overweight (BMI of 25-29.9).⁶ Low SES children are more likely to be obese than the general pediatric population.⁷ Obesity is associated with significant comorbidity,⁸ high healthcare costs,⁹ and excess mortality.¹⁰

It has been shown that nutritional counseling as part of a multidisciplinary weight management intervention can help obese patients lose weight.^3-5 In the recently published Diabetes Prevention Program study, nutrition counseling as part of an intensive program of lifestyle intervention in overweight subjects was effective in decreasing the cumulative incidence of diabetes. The lifestyle intervention included counseling to increase physical activity and modify eating behavior, as well as counseling on nutrition and diet.⁵

Although a weight management program (e.g., the program used in the Diabetes Prevention Program) can be visit intensive, the methods used by nutrition counselors are simple and inexpensive. These include teaching patients to read labels and to count calories; counselors also teach patients to monitor and change their diets through the use of food diaries. Self-monitoring has been shown to be a predictor of weight loss when incorporated as part of an obesity treatment program.¹¹ Dietary education (e.g., learning to read food labels and to count calories of different food types) is more difficult with low SES patients. However, nutritionists have developed reliable and valid tools for nutritional assessment in low SES persons.¹²

Nutritional education and counseling can have positive effects on other clinical end points besides overweight and obesity. The sixth report of the Joint National Committee on High Blood Pressure states that moderate restriction of dietary sodium and a weight loss of 10 pounds are considered first-line therapy for hypertension.¹³ Nutrition therapy has been shown to be effective for treatment of non—insulin-dependent diabetes¹⁴ and hypercholesterolemia.^15,16

It is often assumed that third-party payers may hesitate to reimburse for nutrition counseling because of the unknown long-term cost effectiveness of treatment.¹⁷ However, the recently published cost-effectiveness analysis of the Diabetes Prevention Program (with a mean follow-up of 2.8 years) provides evidence that group nutrition counseling, as part of a lifestyle modification program, is cost-effective.¹⁸ Nutrition therapy also has been shown to be cost effective for treatment of non—insulin-dependent diabetes,¹⁴ cost effective and even cost saving for the treatment of hypercholesterolemia, ^15,16 and cost effective as part of the treatment of overweight and obesity with a computerized weight management program.¹⁹

Another concern of payers is appropriately estimating the number of eligible people who would take advantage of nutrition counseling, given the high prevalence of obesity in the United States. However, clinicians know that as with smoking cessation, only a motivated subset of patients will attend nutrition counseling and weight management programs. For example, it has been shown that the take-up rate of a voluntary nutrition therapy program implemented in a company with 80,000 employees was only 0.5%.20 In this study, patients were eligible for nutrition therapy for a wide range of diagnoses, including obesity; consultation was restricted to patients who were already under the care of a physician and who called on their own for an appointment. In addition, patients had to answer "a series of questions regarding their readiness to change to determine if nutrition intervention is appropriate."²⁰

Our results show that nutrition consultation is not available to a large percentage of Medicaid patients, within both Medicaid fee-for-service and Medicaid MCO reimbursement structures. Medicaid MCOs reimbursed nutrition counseling more frequently than fee-for-service Medicaid for obesity without comorbidity. There also were associations between managed care status and more overall coverage of nutrition counseling, and between managed care status and more frequent coverage of other modalities for weight loss. These associations do not reach conventional levels of significance, perhaps due to small sample size and under-powering. Our relatively small sample size was due to the limitation of 1 person carrying out the telephone surveys. However, these trends seem to reflect an important difference between the 2 types of Medicaid structures in terms of policy. That is, Medicaid managed care is more likely than state Medicaid to reimburse for preventive services for the treatment of obesity. Although we hypothesized that MCOs with larger numbers of Medicaid enrollees might be more likely to provide nutrition services, we did not find such a trend. This may be due to small sample size, or it may be due to a random trend among managed care companies for the reimbursement of these treatments.

Future studies that may be useful include a larger and more complete survey of Medicaid organizations or insurance organizations in general regarding therapies reimbursed for the treatment of obesity. Finally, the greater access to nutrition services provided by MCOs may be partially attributable to the mix of patients. Nationwide, fee-for-service enrollment favors a higher percentage of chronically ill members compared with MCO enrollment. The Medicaid managed care population may, therefore, be more able to benefit from nutrition counseling than the fee-for-service population, for whom preventive treatments often are ranked relatively low in priority.

Our study has several important limitations. First, we did not ask about numbers of RD/CN visits covered for treatment of obesity. Studies that have demonstrated clinical and cost effectiveness for nutritional counseling, such as the Diabetes Prevention Program, have used intensive interventions with frequent follow-up visits. An effective program of nutritional counseling would almost certainly cover multiple visits. Although our results show that many Medicaid organizations are not covering this service at all, more information about number and frequency of visits would give insight into whether the program would be likely to be effective. Second, we did not ask whether the RDs/CNs providing the counseling were certified as obesity experts. However, all RDs and CNs are trained to educate patients with a wide range of diseases, including obesity, about healthy diets and to facilitate diet change. We believe that this expertise constitutes a standard of care that is potentially effective for obese patients, if followed correctly. Third, we did not make a complete assessment of reimbursement for more comprehensive weight management programs such as the program used in the Diabetes Prevention Program. The responses to question 3 of our survey show that reimbursement for commercial weight loss programs and for exercise facilities is limited. However, this result does not answer the question of how often a comprehensive weight loss program is reimbursed for an overweight or obese patient. Fourth, we did not inquire about the level of coverage for nutrition counseling visits. It is certainly possible that in a low SES population, incomplete reimbursement or copayments would be potential barriers to access.

In June 2000, there were 7,899,642 persons enrolled in Medicaid in the 14 states surveyed, 3,983,977 of them in Medicaid managed care.²¹ Based on these numbers and on a conservative prevalence of obesity in the Medicaid population of 20%, close to 2 million Medicaid recipients in these states are affected by policies regarding RD/CN visits. The cost of treating obesity-related illness in the Medicaid population, with its higher prevalence of obesity, is financed by public monies. It makes sense to apply preventive measures in this circumstance, especially those that are clinically effective and cost-effective.

CONCLUSIONS AND RECOMMENDATIONS

In summary, we have shown that (1) access to nutrition counseling for Medicaid patients in the northeastern United States is not uniform and (2) managed care Medicaid organizations provide this service more frequently than do state Medicaid organizations.

Given the high prevalence of obesity in the United States, the disproportionate prevalence in the Medicaid population, and the existing evidence for the clinical and cost effectiveness of nutrition therapy (both by itself and as part of a lifestyle modification intervention), we recommend that all Medicaid recipients should have access to RD/CN visits as part of a weight management program. The exact form of such a program should be modeled after the interventions in clinical trials such as the Diabetes Prevention Program, and the program should be available to Medicaid recipients under both the traditional fee-for-service structure and managed care.

Download PDF: Availability of Nutrition Services for Medicaid Recipients in the Northeastern United States: Lack of Uniformity and the Positive Effect of Managed Care