Miriam E.Tucker

LAS VEGAS – Emerging data support the practice of switching biologic agents after an initial biologic therapy fails in psoriasis, and even the possibility of rotating back to a biologic that didn’t work previously, according to Dr. Francisco Kerdel, director of the dermatology inpatient service at the University of Miami Hospital.

In one recent series of 747 psoriasis patients on biologics, the 4-year drug response was in the range of 40% for etanercept or adalimumab and 70% for infliximab (Br. J. Dermatol. 2011;164:1091-6). "Thus, there is a logical need for ‘switching’ [biologic] therapy," he noted at the SDEF Las Vegas Dermatology Seminar.

The reasons for treatment failure are not known. Antibody production has been suspected of playing a role, but the presence of antibodies seldom correlates with clinical response, he said.

In a 16-week, open-label trial, a Physician Global Assessment (PGA) rating of "clear" or "minimal" was achieved in 52% of 152 patients who had chronic plaque psoriasis and were switched to adalimumab following suboptimal responses to etanercept, methotrexate, or phototherapy (J. Am. Acad. Dermatol. 2011;64:671-81).

In the PSUNRISE study (a prospective, multicenter, open-label study of infliximab treatment in 215 patients with plaque psoriasis who had had a prior inadequate response to etanercept), a PGA score of 0 (clear) or 1 was achieved by week 10 in 65% of the 179 completers, and 60% remained at PGA 0-1 in weeks 14-26. These data have been submitted for publication, according to Dr. Kerdel, who was one of the study authors.

Good to excellent responses were achieved with ustekinumab at doses of either 45 mg or 90 mg, depending on body weight, in 9 of 11 psoriasis patients who had experienced treatment failures on multiple biologics, including infliximab (11 patients), etanercept (10), efalizumab (9), adalimumab (7), and golimumab (3) (Int. J. Dermatol. 2011;50:478-82).

There is even some evidence that patients can return with success to a biologic therapy that they had previously failed.

Dr. Kerdel and his associates have conducted an open-label study of etanercept re-treatment in 20 patients with moderate to severe psoriasis (defined as a PGA score of 3 or greater) who had had prior therapy with etanercept for a minimum of 6 months but had discontinued it because of loss of efficacy. The 10 men and 10 women had an average age of 49 years, with an average of 3.5 years between stopping and restarting etanercept. Five patients withdrew and were classified as treatment failures, regardless of the reason for withdrawal.

The proportion of responders (defined as those achieving a PGA score of 0 or 1) was 5 of 20 (25%) at week 8, and 8 of 20 (40%) at week 12. Body weight appeared to play a role. Among the 14 patients who had a PGA score of 2 or less at week 12, the average weight was 198.4 pounds, compared with 217.6 pounds for the 6 who did not have that response.

These preliminary findings suggest that rotational therapy is possible, at least in cases of secondary failure with etanercept. Rotational therapy may be possible with other biologics as well, but further studies are necessary, Dr. Kerdel concluded.

It had been thought that this practice would be of limited value because of the development of circulating antibodies, which were found in one study to range in prevalence from 5% of rheumatoid arthritis patients on ustekinumab to 87% of those treated with adalimumab (Clin. Rev. Allergy Immunol. 2010;38:82-9). However, the therapeutic relevance of non-neutralizing vs. neutralizing antibodies is currently unknown, he said.

In a recent prospective cohort study of 272 RA patients treated with adalimumab, 28% had developed antiadalimumab antibodies, which appeared within the first 28 weeks of treatment in two-thirds of them. Over 3 years of follow-up, 38% of the patients with antibodies discontinued adalimumab therapy because of treatment failure, compared with 14% of those who did not develop antibodies. Minimal disease activity was achieved in 48% without antibodies, compared with 13% who had antibodies (JAMA 2011;305:1460-8).

And in a retrospective observational study of 97 psoriasis patients who had been treated with anti–tumor necrosis factor–alpha therapy (60 on a first agent, 22 having failed one agent, 9 having failed two agents, and 6 having failed all three agents), antinuclear antibodies developed in 17%, 54%, 78% and 83%, respectively (Br. J. Dermatol. 2010;162:780-5). "So, antibodies do seem to play a role" in treatment failure, Dr. Kerdel commented.

Dr. Kerdel disclosed that he has been involved in clinical studies, has participated in advisory boards, and is a speaker for Amgen, Abbott, Janssen, Genentech, Astellas, Stiefel, Pfizer, Merck, Novartis, Eisai, and Celgene. SDEF and this news organization are owned by Elsevier.

LAS VEGAS – Emerging data support the practice of switching biologic agents after an initial biologic therapy fails in psoriasis, and even the possibility of rotating back to a biologic that didn’t work previously, according to Dr. Francisco Kerdel, director of the dermatology inpatient service at the University of Miami Hospital.

In one recent series of 747 psoriasis patients on biologics, the 4-year drug response was in the range of 40% for etanercept or adalimumab and 70% for infliximab (Br. J. Dermatol. 2011;164:1091-6). "Thus, there is a logical need for ‘switching’ [biologic] therapy," he noted at the SDEF Las Vegas Dermatology Seminar.

The reasons for treatment failure are not known. Antibody production has been suspected of playing a role, but the presence of antibodies seldom correlates with clinical response, he said.

In a 16-week, open-label trial, a Physician Global Assessment (PGA) rating of "clear" or "minimal" was achieved in 52% of 152 patients who had chronic plaque psoriasis and were switched to adalimumab following suboptimal responses to etanercept, methotrexate, or phototherapy (J. Am. Acad. Dermatol. 2011;64:671-81).

In the PSUNRISE study (a prospective, multicenter, open-label study of infliximab treatment in 215 patients with plaque psoriasis who had had a prior inadequate response to etanercept), a PGA score of 0 (clear) or 1 was achieved by week 10 in 65% of the 179 completers, and 60% remained at PGA 0-1 in weeks 14-26. These data have been submitted for publication, according to Dr. Kerdel, who was one of the study authors.

Good to excellent responses were achieved with ustekinumab at doses of either 45 mg or 90 mg, depending on body weight, in 9 of 11 psoriasis patients who had experienced treatment failures on multiple biologics, including infliximab (11 patients), etanercept (10), efalizumab (9), adalimumab (7), and golimumab (3) (Int. J. Dermatol. 2011;50:478-82).

There is even some evidence that patients can return with success to a biologic therapy that they had previously failed.

Dr. Kerdel and his associates have conducted an open-label study of etanercept re-treatment in 20 patients with moderate to severe psoriasis (defined as a PGA score of 3 or greater) who had had prior therapy with etanercept for a minimum of 6 months but had discontinued it because of loss of efficacy. The 10 men and 10 women had an average age of 49 years, with an average of 3.5 years between stopping and restarting etanercept. Five patients withdrew and were classified as treatment failures, regardless of the reason for withdrawal.

The proportion of responders (defined as those achieving a PGA score of 0 or 1) was 5 of 20 (25%) at week 8, and 8 of 20 (40%) at week 12. Body weight appeared to play a role. Among the 14 patients who had a PGA score of 2 or less at week 12, the average weight was 198.4 pounds, compared with 217.6 pounds for the 6 who did not have that response.

These preliminary findings suggest that rotational therapy is possible, at least in cases of secondary failure with etanercept. Rotational therapy may be possible with other biologics as well, but further studies are necessary, Dr. Kerdel concluded.

It had been thought that this practice would be of limited value because of the development of circulating antibodies, which were found in one study to range in prevalence from 5% of rheumatoid arthritis patients on ustekinumab to 87% of those treated with adalimumab (Clin. Rev. Allergy Immunol. 2010;38:82-9). However, the therapeutic relevance of non-neutralizing vs. neutralizing antibodies is currently unknown, he said.

In a recent prospective cohort study of 272 RA patients treated with adalimumab, 28% had developed antiadalimumab antibodies, which appeared within the first 28 weeks of treatment in two-thirds of them. Over 3 years of follow-up, 38% of the patients with antibodies discontinued adalimumab therapy because of treatment failure, compared with 14% of those who did not develop antibodies. Minimal disease activity was achieved in 48% without antibodies, compared with 13% who had antibodies (JAMA 2011;305:1460-8).

And in a retrospective observational study of 97 psoriasis patients who had been treated with anti–tumor necrosis factor–alpha therapy (60 on a first agent, 22 having failed one agent, 9 having failed two agents, and 6 having failed all three agents), antinuclear antibodies developed in 17%, 54%, 78% and 83%, respectively (Br. J. Dermatol. 2010;162:780-5). "So, antibodies do seem to play a role" in treatment failure, Dr. Kerdel commented.

Dr. Kerdel disclosed that he has been involved in clinical studies, has participated in advisory boards, and is a speaker for Amgen, Abbott, Janssen, Genentech, Astellas, Stiefel, Pfizer, Merck, Novartis, Eisai, and Celgene. SDEF and this news organization are owned by Elsevier.

LAS VEGAS – Emerging data support the practice of switching biologic agents after an initial biologic therapy fails in psoriasis, and even the possibility of rotating back to a biologic that didn’t work previously, according to Dr. Francisco Kerdel, director of the dermatology inpatient service at the University of Miami Hospital.

In one recent series of 747 psoriasis patients on biologics, the 4-year drug response was in the range of 40% for etanercept or adalimumab and 70% for infliximab (Br. J. Dermatol. 2011;164:1091-6). "Thus, there is a logical need for ‘switching’ [biologic] therapy," he noted at the SDEF Las Vegas Dermatology Seminar.

The reasons for treatment failure are not known. Antibody production has been suspected of playing a role, but the presence of antibodies seldom correlates with clinical response, he said.

In a 16-week, open-label trial, a Physician Global Assessment (PGA) rating of "clear" or "minimal" was achieved in 52% of 152 patients who had chronic plaque psoriasis and were switched to adalimumab following suboptimal responses to etanercept, methotrexate, or phototherapy (J. Am. Acad. Dermatol. 2011;64:671-81).

In the PSUNRISE study (a prospective, multicenter, open-label study of infliximab treatment in 215 patients with plaque psoriasis who had had a prior inadequate response to etanercept), a PGA score of 0 (clear) or 1 was achieved by week 10 in 65% of the 179 completers, and 60% remained at PGA 0-1 in weeks 14-26. These data have been submitted for publication, according to Dr. Kerdel, who was one of the study authors.

Good to excellent responses were achieved with ustekinumab at doses of either 45 mg or 90 mg, depending on body weight, in 9 of 11 psoriasis patients who had experienced treatment failures on multiple biologics, including infliximab (11 patients), etanercept (10), efalizumab (9), adalimumab (7), and golimumab (3) (Int. J. Dermatol. 2011;50:478-82).

There is even some evidence that patients can return with success to a biologic therapy that they had previously failed.

Dr. Kerdel and his associates have conducted an open-label study of etanercept re-treatment in 20 patients with moderate to severe psoriasis (defined as a PGA score of 3 or greater) who had had prior therapy with etanercept for a minimum of 6 months but had discontinued it because of loss of efficacy. The 10 men and 10 women had an average age of 49 years, with an average of 3.5 years between stopping and restarting etanercept. Five patients withdrew and were classified as treatment failures, regardless of the reason for withdrawal.

The proportion of responders (defined as those achieving a PGA score of 0 or 1) was 5 of 20 (25%) at week 8, and 8 of 20 (40%) at week 12. Body weight appeared to play a role. Among the 14 patients who had a PGA score of 2 or less at week 12, the average weight was 198.4 pounds, compared with 217.6 pounds for the 6 who did not have that response.

These preliminary findings suggest that rotational therapy is possible, at least in cases of secondary failure with etanercept. Rotational therapy may be possible with other biologics as well, but further studies are necessary, Dr. Kerdel concluded.

It had been thought that this practice would be of limited value because of the development of circulating antibodies, which were found in one study to range in prevalence from 5% of rheumatoid arthritis patients on ustekinumab to 87% of those treated with adalimumab (Clin. Rev. Allergy Immunol. 2010;38:82-9). However, the therapeutic relevance of non-neutralizing vs. neutralizing antibodies is currently unknown, he said.

In a recent prospective cohort study of 272 RA patients treated with adalimumab, 28% had developed antiadalimumab antibodies, which appeared within the first 28 weeks of treatment in two-thirds of them. Over 3 years of follow-up, 38% of the patients with antibodies discontinued adalimumab therapy because of treatment failure, compared with 14% of those who did not develop antibodies. Minimal disease activity was achieved in 48% without antibodies, compared with 13% who had antibodies (JAMA 2011;305:1460-8).

And in a retrospective observational study of 97 psoriasis patients who had been treated with anti–tumor necrosis factor–alpha therapy (60 on a first agent, 22 having failed one agent, 9 having failed two agents, and 6 having failed all three agents), antinuclear antibodies developed in 17%, 54%, 78% and 83%, respectively (Br. J. Dermatol. 2010;162:780-5). "So, antibodies do seem to play a role" in treatment failure, Dr. Kerdel commented.

Dr. Kerdel disclosed that he has been involved in clinical studies, has participated in advisory boards, and is a speaker for Amgen, Abbott, Janssen, Genentech, Astellas, Stiefel, Pfizer, Merck, Novartis, Eisai, and Celgene. SDEF and this news organization are owned by Elsevier.

SAN DIEGO – Recombinant human hyaluronidase combined with human regular insulin produced comparable glycemic responses to lispro insulin in a randomized open-label crossover study of 46 patients with well-controlled type 1 diabetes.

Recombinant human hyaluronidase (rHuPH20) is approved by the Food and Drug Administration to increase the dispersion and absorption of other injected drugs. This study investigated whether a combination of rHuPH20 with human regular insulin (Halozyme Therapeutics’ investigational Insulin-PH20) could produce glycemic control comparable to that of the currently available rapid-acting analogues. The ultimate aim is to combine it with those to make an even faster-acting analogue, Dr. Satish K. Garg said on June 25 at the annual scientific sessions of the American Diabetes Association.

“If we really want to achieve euglycemia in the postprandial phase, especially 1 and 2 hours after meals, we need an ultra–fast-acting insulin, and we don’t have that. So now, studies are on their way to look at using hyaluronidase with insulin lispro or another rapid-acting analogue to see if we can make it ultra–fast-acting insulin,” said Dr. Garg, professor of medicine and pediatrics at the University of Colorado, Denver.

The 46 patients had a mean age of 42 years, a mean body mass index of 26 kg/m², and a mean hemoglobin A_1c of 6.9%. They were randomized to either the Insulin-PH20 or insulin lispro for 2 consecutive 12-week periods, with twice-daily glargine as basal insulin in both groups. Forty-one of the 46 completed the trial.

The prespecified primary end point was a noninferiority margin of post-prandial glucose values not exceeding 21.6 mg/dL for three meals over 3 days. The difference in glycemic excursions between the two insulin formulations was 2.4 mg/dL, clearly meeting the end point, Dr. Garg said.

HbA_1c was maintained for both groups in the trial, 7.0% for Insulin-PH20 and 6.9% for lispro, meeting the commonly applied noninferiority margin of 0.4%. Continuous glucose monitoring during the last 2 weeks of each treatment period showed similar mean glucose values (153 vs. 143 mg/dL), with similar amounts of time spent in the target range of 70-130 mg/dL (39% vs. 44%).

Overall hypoglycemia, defined as blood glucose value of 70 mg/dL or below, was 24.1 vs. 22.4 events per patient-month for Insulin-PH20 and lispro, respectively, a nonsignificant difference. There were no significant changes in anti-insulin and anti-lispro antibodies, Dr. Garg noted.

Studies are now investigating the combination of rHuPH20 and currently available rapid-acting analogues in type 1 and type 2 diabetes patients to see whether the course of action can be accelerated further.

Dr. Garg disclosed that he has received grants and honoraria from Halozyme Therapeutics, Sanofi-Aventis, Novo-Nordisk, Dexcom, and Eli Lilly. 

SAN DIEGO – Recombinant human hyaluronidase combined with human regular insulin produced comparable glycemic responses to lispro insulin in a randomized open-label crossover study of 46 patients with well-controlled type 1 diabetes.

Recombinant human hyaluronidase (rHuPH20) is approved by the Food and Drug Administration to increase the dispersion and absorption of other injected drugs. This study investigated whether a combination of rHuPH20 with human regular insulin (Halozyme Therapeutics’ investigational Insulin-PH20) could produce glycemic control comparable to that of the currently available rapid-acting analogues. The ultimate aim is to combine it with those to make an even faster-acting analogue, Dr. Satish K. Garg said on June 25 at the annual scientific sessions of the American Diabetes Association.

“If we really want to achieve euglycemia in the postprandial phase, especially 1 and 2 hours after meals, we need an ultra–fast-acting insulin, and we don’t have that. So now, studies are on their way to look at using hyaluronidase with insulin lispro or another rapid-acting analogue to see if we can make it ultra–fast-acting insulin,” said Dr. Garg, professor of medicine and pediatrics at the University of Colorado, Denver.

The 46 patients had a mean age of 42 years, a mean body mass index of 26 kg/m², and a mean hemoglobin A_1c of 6.9%. They were randomized to either the Insulin-PH20 or insulin lispro for 2 consecutive 12-week periods, with twice-daily glargine as basal insulin in both groups. Forty-one of the 46 completed the trial.

The prespecified primary end point was a noninferiority margin of post-prandial glucose values not exceeding 21.6 mg/dL for three meals over 3 days. The difference in glycemic excursions between the two insulin formulations was 2.4 mg/dL, clearly meeting the end point, Dr. Garg said.

HbA_1c was maintained for both groups in the trial, 7.0% for Insulin-PH20 and 6.9% for lispro, meeting the commonly applied noninferiority margin of 0.4%. Continuous glucose monitoring during the last 2 weeks of each treatment period showed similar mean glucose values (153 vs. 143 mg/dL), with similar amounts of time spent in the target range of 70-130 mg/dL (39% vs. 44%).

Overall hypoglycemia, defined as blood glucose value of 70 mg/dL or below, was 24.1 vs. 22.4 events per patient-month for Insulin-PH20 and lispro, respectively, a nonsignificant difference. There were no significant changes in anti-insulin and anti-lispro antibodies, Dr. Garg noted.

Studies are now investigating the combination of rHuPH20 and currently available rapid-acting analogues in type 1 and type 2 diabetes patients to see whether the course of action can be accelerated further.

Dr. Garg disclosed that he has received grants and honoraria from Halozyme Therapeutics, Sanofi-Aventis, Novo-Nordisk, Dexcom, and Eli Lilly. 

SAN DIEGO – Recombinant human hyaluronidase combined with human regular insulin produced comparable glycemic responses to lispro insulin in a randomized open-label crossover study of 46 patients with well-controlled type 1 diabetes.

Recombinant human hyaluronidase (rHuPH20) is approved by the Food and Drug Administration to increase the dispersion and absorption of other injected drugs. This study investigated whether a combination of rHuPH20 with human regular insulin (Halozyme Therapeutics’ investigational Insulin-PH20) could produce glycemic control comparable to that of the currently available rapid-acting analogues. The ultimate aim is to combine it with those to make an even faster-acting analogue, Dr. Satish K. Garg said on June 25 at the annual scientific sessions of the American Diabetes Association.

“If we really want to achieve euglycemia in the postprandial phase, especially 1 and 2 hours after meals, we need an ultra–fast-acting insulin, and we don’t have that. So now, studies are on their way to look at using hyaluronidase with insulin lispro or another rapid-acting analogue to see if we can make it ultra–fast-acting insulin,” said Dr. Garg, professor of medicine and pediatrics at the University of Colorado, Denver.

The 46 patients had a mean age of 42 years, a mean body mass index of 26 kg/m², and a mean hemoglobin A_1c of 6.9%. They were randomized to either the Insulin-PH20 or insulin lispro for 2 consecutive 12-week periods, with twice-daily glargine as basal insulin in both groups. Forty-one of the 46 completed the trial.

The prespecified primary end point was a noninferiority margin of post-prandial glucose values not exceeding 21.6 mg/dL for three meals over 3 days. The difference in glycemic excursions between the two insulin formulations was 2.4 mg/dL, clearly meeting the end point, Dr. Garg said.

HbA_1c was maintained for both groups in the trial, 7.0% for Insulin-PH20 and 6.9% for lispro, meeting the commonly applied noninferiority margin of 0.4%. Continuous glucose monitoring during the last 2 weeks of each treatment period showed similar mean glucose values (153 vs. 143 mg/dL), with similar amounts of time spent in the target range of 70-130 mg/dL (39% vs. 44%).

Overall hypoglycemia, defined as blood glucose value of 70 mg/dL or below, was 24.1 vs. 22.4 events per patient-month for Insulin-PH20 and lispro, respectively, a nonsignificant difference. There were no significant changes in anti-insulin and anti-lispro antibodies, Dr. Garg noted.

Studies are now investigating the combination of rHuPH20 and currently available rapid-acting analogues in type 1 and type 2 diabetes patients to see whether the course of action can be accelerated further.

Dr. Garg disclosed that he has received grants and honoraria from Halozyme Therapeutics, Sanofi-Aventis, Novo-Nordisk, Dexcom, and Eli Lilly. 

STOCKHOLM – The risk for developing type 2 diabetes is not the same for everyone with metabolic syndrome, but instead varies dramatically depending on individual factors.

In fact, hyperglycemia – with or without metabolic syndrome – was a much stronger predictor of incident type 2 diabetes than was metabolic syndrome without hyperglycemia in a 5-year observational analysis of 58,056 initially nondiabetic adults aged 30 years and older who were members of the managed care organization Kaiser Permanente Northwest, Gregory A. Nichols, Ph.D., said at the annual meeting of the European Association for the Study of Diabetes.

“In the absence of impaired fasting glucose, the definition of metabolic syndrome may be a misleading estimator of diabetes risk,” according to Dr. Nichols the lead investigator on the study (Diabetes Res. Clin. Pract. 2010;90:115-21).

He and a colleague examined the incidence of diabetes for all possible combinations of metabolic syndrome components using criteria defined in the National Cholesterol Education Program’s Adult Treatment Panel III report (ATP III) (Circulation 2004;109:433-8).

The one exception was the use of body mass index as a substitute for waist circumference, which is rarely measured clinically, explained Dr. Nichols of Kaiser Permanente’s Center for Health Research, Portland, Ore.

For the study, an individual was considered to have metabolic syndrome if they met three of the following five criteria:

• Impaired fasting glucose (greater than 100 mg/dL),

• Hypertension (130/85 mm Hg or greater),

• High triglycerides (150 mg/dL or greater),

• Low HDL cholesterol (less than 40 mg/dL for men, 50 mg/dL for women), and

• Body-mass index greater than 28.8 kg/m2. This BMI cut-point has been substituted for waist circumference in other published studies, he noted.

Over 5 years, 6% of the total study sample developed diabetes. Compared with those who did not develop diabetes, those who did were significantly older (59 vs. 57 years), and were more likely to be male (52% vs. 44%), non-white (10% vs. 8%) and a current smoker (15% vs. 12%).

Not surprisingly, the risk for developing diabetes was greater in the presence than in the absence of each individual component. The 5-year risk for diabetes rose with each component an individual had at baseline, from 0.3% for those with none to 1.2% with one, 3.5% with two, 8.4% with three, 16.9% with four, and 28.2% with five.

However, the risk for diabetes varied dramatically among the five individual components. The greatest risk was associated with impaired fasting glucose (IFG), with an incidence of 37.4/1,000 person-years. Low HDL cholesterol was next (21.6/1,000 person-years), followed by high triglycerides (20.6/1,000), obesity (19.5) and hypertension (16.2).

There was a clear separation between combinations of components that did and did not contain IFG. The combination of IFG and any one additional component – by definition, not meeting metabolic syndrome criteria – had a higher incidence rate of diabetes (16.5/1,000 person-years) than did any three- or four-component combination that did not include IFG (7.9 and 11.3 per 1,000 person-years, respectively), yet did meet the metabolic syndrome criteria.

The incidence of diabetes among those who had IFG and no other metabolic syndrome component was 10.2/1,000 person-years, just slightly less than the incidence for those with every component except IFG (11.3/1,000), Dr. Nichols reported.

“Better prediction tools that identify truly high-risk individuals would enhance diabetes prevention strategies,” he concluded at the meeting.

In the published article, Dr. Nichols and his coauthor Dr. Edward J. Moler noted that in addition to the association with diabetes, the clustering of risk factors that comprise metabolic syndrome also are precursors to cardiovascular disease (CVD), and that the relationship of individual components to CVD is unclear. “Because metabolic syndrome is also used to predict CVD risk, future research should assess the relative importance of metabolic syndrome components to predict incident CVD,” they said.

This study was funded by Tethys Bioscience Inc., where Dr. Moler is an employee. Dr. Nichols disclosed that he has also received research funding from GlaxoSmithKline, Novartis, Novo Nordisk, Takeda Pharmaceuticals, and Merck & Co.

STOCKHOLM – The risk for developing type 2 diabetes is not the same for everyone with metabolic syndrome, but instead varies dramatically depending on individual factors.

In fact, hyperglycemia – with or without metabolic syndrome – was a much stronger predictor of incident type 2 diabetes than was metabolic syndrome without hyperglycemia in a 5-year observational analysis of 58,056 initially nondiabetic adults aged 30 years and older who were members of the managed care organization Kaiser Permanente Northwest, Gregory A. Nichols, Ph.D., said at the annual meeting of the European Association for the Study of Diabetes.

“In the absence of impaired fasting glucose, the definition of metabolic syndrome may be a misleading estimator of diabetes risk,” according to Dr. Nichols the lead investigator on the study (Diabetes Res. Clin. Pract. 2010;90:115-21).

He and a colleague examined the incidence of diabetes for all possible combinations of metabolic syndrome components using criteria defined in the National Cholesterol Education Program’s Adult Treatment Panel III report (ATP III) (Circulation 2004;109:433-8).

The one exception was the use of body mass index as a substitute for waist circumference, which is rarely measured clinically, explained Dr. Nichols of Kaiser Permanente’s Center for Health Research, Portland, Ore.

For the study, an individual was considered to have metabolic syndrome if they met three of the following five criteria:

• Impaired fasting glucose (greater than 100 mg/dL),

• Hypertension (130/85 mm Hg or greater),

• High triglycerides (150 mg/dL or greater),

• Low HDL cholesterol (less than 40 mg/dL for men, 50 mg/dL for women), and

• Body-mass index greater than 28.8 kg/m2. This BMI cut-point has been substituted for waist circumference in other published studies, he noted.

Over 5 years, 6% of the total study sample developed diabetes. Compared with those who did not develop diabetes, those who did were significantly older (59 vs. 57 years), and were more likely to be male (52% vs. 44%), non-white (10% vs. 8%) and a current smoker (15% vs. 12%).

Not surprisingly, the risk for developing diabetes was greater in the presence than in the absence of each individual component. The 5-year risk for diabetes rose with each component an individual had at baseline, from 0.3% for those with none to 1.2% with one, 3.5% with two, 8.4% with three, 16.9% with four, and 28.2% with five.

However, the risk for diabetes varied dramatically among the five individual components. The greatest risk was associated with impaired fasting glucose (IFG), with an incidence of 37.4/1,000 person-years. Low HDL cholesterol was next (21.6/1,000 person-years), followed by high triglycerides (20.6/1,000), obesity (19.5) and hypertension (16.2).

There was a clear separation between combinations of components that did and did not contain IFG. The combination of IFG and any one additional component – by definition, not meeting metabolic syndrome criteria – had a higher incidence rate of diabetes (16.5/1,000 person-years) than did any three- or four-component combination that did not include IFG (7.9 and 11.3 per 1,000 person-years, respectively), yet did meet the metabolic syndrome criteria.

The incidence of diabetes among those who had IFG and no other metabolic syndrome component was 10.2/1,000 person-years, just slightly less than the incidence for those with every component except IFG (11.3/1,000), Dr. Nichols reported.

“Better prediction tools that identify truly high-risk individuals would enhance diabetes prevention strategies,” he concluded at the meeting.

In the published article, Dr. Nichols and his coauthor Dr. Edward J. Moler noted that in addition to the association with diabetes, the clustering of risk factors that comprise metabolic syndrome also are precursors to cardiovascular disease (CVD), and that the relationship of individual components to CVD is unclear. “Because metabolic syndrome is also used to predict CVD risk, future research should assess the relative importance of metabolic syndrome components to predict incident CVD,” they said.

This study was funded by Tethys Bioscience Inc., where Dr. Moler is an employee. Dr. Nichols disclosed that he has also received research funding from GlaxoSmithKline, Novartis, Novo Nordisk, Takeda Pharmaceuticals, and Merck & Co.

STOCKHOLM – The risk for developing type 2 diabetes is not the same for everyone with metabolic syndrome, but instead varies dramatically depending on individual factors.

In fact, hyperglycemia – with or without metabolic syndrome – was a much stronger predictor of incident type 2 diabetes than was metabolic syndrome without hyperglycemia in a 5-year observational analysis of 58,056 initially nondiabetic adults aged 30 years and older who were members of the managed care organization Kaiser Permanente Northwest, Gregory A. Nichols, Ph.D., said at the annual meeting of the European Association for the Study of Diabetes.

“In the absence of impaired fasting glucose, the definition of metabolic syndrome may be a misleading estimator of diabetes risk,” according to Dr. Nichols the lead investigator on the study (Diabetes Res. Clin. Pract. 2010;90:115-21).

He and a colleague examined the incidence of diabetes for all possible combinations of metabolic syndrome components using criteria defined in the National Cholesterol Education Program’s Adult Treatment Panel III report (ATP III) (Circulation 2004;109:433-8).

The one exception was the use of body mass index as a substitute for waist circumference, which is rarely measured clinically, explained Dr. Nichols of Kaiser Permanente’s Center for Health Research, Portland, Ore.

For the study, an individual was considered to have metabolic syndrome if they met three of the following five criteria:

• Impaired fasting glucose (greater than 100 mg/dL),

• Hypertension (130/85 mm Hg or greater),

• High triglycerides (150 mg/dL or greater),

• Low HDL cholesterol (less than 40 mg/dL for men, 50 mg/dL for women), and

• Body-mass index greater than 28.8 kg/m2. This BMI cut-point has been substituted for waist circumference in other published studies, he noted.

Over 5 years, 6% of the total study sample developed diabetes. Compared with those who did not develop diabetes, those who did were significantly older (59 vs. 57 years), and were more likely to be male (52% vs. 44%), non-white (10% vs. 8%) and a current smoker (15% vs. 12%).

Not surprisingly, the risk for developing diabetes was greater in the presence than in the absence of each individual component. The 5-year risk for diabetes rose with each component an individual had at baseline, from 0.3% for those with none to 1.2% with one, 3.5% with two, 8.4% with three, 16.9% with four, and 28.2% with five.

However, the risk for diabetes varied dramatically among the five individual components. The greatest risk was associated with impaired fasting glucose (IFG), with an incidence of 37.4/1,000 person-years. Low HDL cholesterol was next (21.6/1,000 person-years), followed by high triglycerides (20.6/1,000), obesity (19.5) and hypertension (16.2).

There was a clear separation between combinations of components that did and did not contain IFG. The combination of IFG and any one additional component – by definition, not meeting metabolic syndrome criteria – had a higher incidence rate of diabetes (16.5/1,000 person-years) than did any three- or four-component combination that did not include IFG (7.9 and 11.3 per 1,000 person-years, respectively), yet did meet the metabolic syndrome criteria.

The incidence of diabetes among those who had IFG and no other metabolic syndrome component was 10.2/1,000 person-years, just slightly less than the incidence for those with every component except IFG (11.3/1,000), Dr. Nichols reported.

“Better prediction tools that identify truly high-risk individuals would enhance diabetes prevention strategies,” he concluded at the meeting.

In the published article, Dr. Nichols and his coauthor Dr. Edward J. Moler noted that in addition to the association with diabetes, the clustering of risk factors that comprise metabolic syndrome also are precursors to cardiovascular disease (CVD), and that the relationship of individual components to CVD is unclear. “Because metabolic syndrome is also used to predict CVD risk, future research should assess the relative importance of metabolic syndrome components to predict incident CVD,” they said.

This study was funded by Tethys Bioscience Inc., where Dr. Moler is an employee. Dr. Nichols disclosed that he has also received research funding from GlaxoSmithKline, Novartis, Novo Nordisk, Takeda Pharmaceuticals, and Merck & Co.

STOCKHOLM – The risk for developing type 2 diabetes is not the same for everyone with metabolic syndrome, but instead varies dramatically depending on individual factors.

In fact, hyperglycemia – with or without metabolic syndrome – was a much stronger predictor of incident type 2 diabetes than was metabolic syndrome without hyperglycemia in a 5-year observational analysis of 58,056 initially nondiabetic adults aged 30 years and older who were members of the managed care organization Kaiser Permanente Northwest, Gregory A. Nichols, Ph.D., said at the annual meeting of the European Association for the Study of Diabetes.

“In the absence of impaired fasting glucose, the definition of metabolic syndrome may be a misleading estimator of diabetes risk,” according to Dr. Nichols the lead investigator on the study (Diabetes Res. Clin. Pract. 2010;90:115-21).

He and a colleague examined the incidence of diabetes for all possible combinations of metabolic syndrome components using criteria defined in the National Cholesterol Education Program’s Adult Treatment Panel III report (ATP III) (Circulation 2004;109:433-8).

The one exception was the use of body mass index as a substitute for waist circumference, which is rarely measured clinically, explained Dr. Nichols of Kaiser Permanente’s Center for Health Research, Portland, Ore.

For the study, an individual was considered to have metabolic syndrome if they met three of the following five criteria:

• Impaired fasting glucose (greater than 100 mg/dL),

• Hypertension (130/85 mm Hg or greater),

• High triglycerides (150 mg/dL or greater),

• Low HDL cholesterol (less than 40 mg/dL for men, 50 mg/dL for women), and

• Body-mass index greater than 28.8 kg/m2. This BMI cut-point has been substituted for waist circumference in other published studies, he noted.

Over 5 years, 6% of the total study sample developed diabetes. Compared with those who did not develop diabetes, those who did were significantly older (59 vs. 57 years), and were more likely to be male (52% vs. 44%), non-white (10% vs. 8%) and a current smoker (15% vs. 12%).

Not surprisingly, the risk for developing diabetes was greater in the presence than in the absence of each individual component. The 5-year risk for diabetes rose with each component an individual had at baseline, from 0.3% for those with none to 1.2% with one, 3.5% with two, 8.4% with three, 16.9% with four, and 28.2% with five.

However, the risk for diabetes varied dramatically among the five individual components. The greatest risk was associated with impaired fasting glucose (IFG), with an incidence of 37.4/1,000 person-years. Low HDL cholesterol was next (21.6/1,000 person-years), followed by high triglycerides (20.6/1,000), obesity (19.5) and hypertension (16.2).

There was a clear separation between combinations of components that did and did not contain IFG. The combination of IFG and any one additional component – by definition, not meeting metabolic syndrome criteria – had a higher incidence rate of diabetes (16.5/1,000 person-years) than did any three- or four-component combination that did not include IFG (7.9 and 11.3 per 1,000 person-years, respectively), yet did meet the metabolic syndrome criteria.

The incidence of diabetes among those who had IFG and no other metabolic syndrome component was 10.2/1,000 person-years, just slightly less than the incidence for those with every component except IFG (11.3/1,000), Dr. Nichols reported.

“Better prediction tools that identify truly high-risk individuals would enhance diabetes prevention strategies,” he concluded at the meeting.

In the published article, Dr. Nichols and his coauthor Dr. Edward J. Moler noted that in addition to the association with diabetes, the clustering of risk factors that comprise metabolic syndrome also are precursors to cardiovascular disease (CVD), and that the relationship of individual components to CVD is unclear. “Because metabolic syndrome is also used to predict CVD risk, future research should assess the relative importance of metabolic syndrome components to predict incident CVD,” they said.

This study was funded by Tethys Bioscience Inc., where Dr. Moler is an employee. Dr. Nichols disclosed that he has also received research funding from GlaxoSmithKline, Novartis, Novo Nordisk, Takeda Pharmaceuticals, and Merck & Co.

STOCKHOLM – The risk for developing type 2 diabetes is not the same for everyone with metabolic syndrome, but instead varies dramatically depending on individual factors.

In fact, hyperglycemia – with or without metabolic syndrome – was a much stronger predictor of incident type 2 diabetes than was metabolic syndrome without hyperglycemia in a 5-year observational analysis of 58,056 initially nondiabetic adults aged 30 years and older who were members of the managed care organization Kaiser Permanente Northwest, Gregory A. Nichols, Ph.D., said at the annual meeting of the European Association for the Study of Diabetes.

“In the absence of impaired fasting glucose, the definition of metabolic syndrome may be a misleading estimator of diabetes risk,” according to Dr. Nichols the lead investigator on the study (Diabetes Res. Clin. Pract. 2010;90:115-21).

He and a colleague examined the incidence of diabetes for all possible combinations of metabolic syndrome components using criteria defined in the National Cholesterol Education Program’s Adult Treatment Panel III report (ATP III) (Circulation 2004;109:433-8).

The one exception was the use of body mass index as a substitute for waist circumference, which is rarely measured clinically, explained Dr. Nichols of Kaiser Permanente’s Center for Health Research, Portland, Ore.

For the study, an individual was considered to have metabolic syndrome if they met three of the following five criteria:

• Impaired fasting glucose (greater than 100 mg/dL),

• Hypertension (130/85 mm Hg or greater),

• High triglycerides (150 mg/dL or greater),

• Low HDL cholesterol (less than 40 mg/dL for men, 50 mg/dL for women), and

• Body-mass index greater than 28.8 kg/m2. This BMI cut-point has been substituted for waist circumference in other published studies, he noted.

Over 5 years, 6% of the total study sample developed diabetes. Compared with those who did not develop diabetes, those who did were significantly older (59 vs. 57 years), and were more likely to be male (52% vs. 44%), non-white (10% vs. 8%) and a current smoker (15% vs. 12%).

Not surprisingly, the risk for developing diabetes was greater in the presence than in the absence of each individual component. The 5-year risk for diabetes rose with each component an individual had at baseline, from 0.3% for those with none to 1.2% with one, 3.5% with two, 8.4% with three, 16.9% with four, and 28.2% with five.

However, the risk for diabetes varied dramatically among the five individual components. The greatest risk was associated with impaired fasting glucose (IFG), with an incidence of 37.4/1,000 person-years. Low HDL cholesterol was next (21.6/1,000 person-years), followed by high triglycerides (20.6/1,000), obesity (19.5) and hypertension (16.2).

There was a clear separation between combinations of components that did and did not contain IFG. The combination of IFG and any one additional component – by definition, not meeting metabolic syndrome criteria – had a higher incidence rate of diabetes (16.5/1,000 person-years) than did any three- or four-component combination that did not include IFG (7.9 and 11.3 per 1,000 person-years, respectively), yet did meet the metabolic syndrome criteria.

The incidence of diabetes among those who had IFG and no other metabolic syndrome component was 10.2/1,000 person-years, just slightly less than the incidence for those with every component except IFG (11.3/1,000), Dr. Nichols reported.

“Better prediction tools that identify truly high-risk individuals would enhance diabetes prevention strategies,” he concluded at the meeting.

In the published article, Dr. Nichols and his coauthor Dr. Edward J. Moler noted that in addition to the association with diabetes, the clustering of risk factors that comprise metabolic syndrome also are precursors to cardiovascular disease (CVD), and that the relationship of individual components to CVD is unclear. “Because metabolic syndrome is also used to predict CVD risk, future research should assess the relative importance of metabolic syndrome components to predict incident CVD,” they said.

This study was funded by Tethys Bioscience Inc., where Dr. Moler is an employee. Dr. Nichols disclosed that he has also received research funding from GlaxoSmithKline, Novartis, Novo Nordisk, Takeda Pharmaceuticals, and Merck & Co.

STOCKHOLM – The risk for developing type 2 diabetes is not the same for everyone with metabolic syndrome, but instead varies dramatically depending on individual factors.

In fact, hyperglycemia – with or without metabolic syndrome – was a much stronger predictor of incident type 2 diabetes than was metabolic syndrome without hyperglycemia in a 5-year observational analysis of 58,056 initially nondiabetic adults aged 30 years and older who were members of the managed care organization Kaiser Permanente Northwest, Gregory A. Nichols, Ph.D., said at the annual meeting of the European Association for the Study of Diabetes.

“In the absence of impaired fasting glucose, the definition of metabolic syndrome may be a misleading estimator of diabetes risk,” according to Dr. Nichols the lead investigator on the study (Diabetes Res. Clin. Pract. 2010;90:115-21).

He and a colleague examined the incidence of diabetes for all possible combinations of metabolic syndrome components using criteria defined in the National Cholesterol Education Program’s Adult Treatment Panel III report (ATP III) (Circulation 2004;109:433-8).

The one exception was the use of body mass index as a substitute for waist circumference, which is rarely measured clinically, explained Dr. Nichols of Kaiser Permanente’s Center for Health Research, Portland, Ore.

For the study, an individual was considered to have metabolic syndrome if they met three of the following five criteria:

• Impaired fasting glucose (greater than 100 mg/dL),

• Hypertension (130/85 mm Hg or greater),

• High triglycerides (150 mg/dL or greater),

• Low HDL cholesterol (less than 40 mg/dL for men, 50 mg/dL for women), and

• Body-mass index greater than 28.8 kg/m2. This BMI cut-point has been substituted for waist circumference in other published studies, he noted.

Over 5 years, 6% of the total study sample developed diabetes. Compared with those who did not develop diabetes, those who did were significantly older (59 vs. 57 years), and were more likely to be male (52% vs. 44%), non-white (10% vs. 8%) and a current smoker (15% vs. 12%).

Not surprisingly, the risk for developing diabetes was greater in the presence than in the absence of each individual component. The 5-year risk for diabetes rose with each component an individual had at baseline, from 0.3% for those with none to 1.2% with one, 3.5% with two, 8.4% with three, 16.9% with four, and 28.2% with five.

However, the risk for diabetes varied dramatically among the five individual components. The greatest risk was associated with impaired fasting glucose (IFG), with an incidence of 37.4/1,000 person-years. Low HDL cholesterol was next (21.6/1,000 person-years), followed by high triglycerides (20.6/1,000), obesity (19.5) and hypertension (16.2).

There was a clear separation between combinations of components that did and did not contain IFG. The combination of IFG and any one additional component – by definition, not meeting metabolic syndrome criteria – had a higher incidence rate of diabetes (16.5/1,000 person-years) than did any three- or four-component combination that did not include IFG (7.9 and 11.3 per 1,000 person-years, respectively), yet did meet the metabolic syndrome criteria.

The incidence of diabetes among those who had IFG and no other metabolic syndrome component was 10.2/1,000 person-years, just slightly less than the incidence for those with every component except IFG (11.3/1,000), Dr. Nichols reported.

“Better prediction tools that identify truly high-risk individuals would enhance diabetes prevention strategies,” he concluded at the meeting.

In the published article, Dr. Nichols and his coauthor Dr. Edward J. Moler noted that in addition to the association with diabetes, the clustering of risk factors that comprise metabolic syndrome also are precursors to cardiovascular disease (CVD), and that the relationship of individual components to CVD is unclear. “Because metabolic syndrome is also used to predict CVD risk, future research should assess the relative importance of metabolic syndrome components to predict incident CVD,” they said.

This study was funded by Tethys Bioscience Inc., where Dr. Moler is an employee. Dr. Nichols disclosed that he has also received research funding from GlaxoSmithKline, Novartis, Novo Nordisk, Takeda Pharmaceuticals, and Merck & Co.

WASHINGTON — Reimbursement for routine, universal HIV screening will prove challenging in both the private and public sectors, Dr. Michael Horberg and Ms. Christine Lubinski said in separate presentations at a meeting on HIV diagnosis and prevention and access to care.

In September 2006, the Centers for Disease Control and Prevention recommended that diagnostic HIV testing and “opt-out” HIV screening be made a part of routine clinical care in all health care settings for patients aged 13–64 years (MMWR 2006;55[RR-14]). Kaiser Permanente, the country's largest staff-model HMO, is “grappling with this now. We have to look at the implications,” said Dr. Horberg, director of HIV/AIDS Policy, Quality Improvement, and Research at Kaiser.

“Yes we have the capacity to do it, and yes, we have the will to do it. But it is a lot of money,” said Dr. Horberg.

As for the public sector, “There are significant roadblocks … The Centers for Medicare and Medicaid Services and the [Bush] administration have little commitment to expand the federal contribution to the Medicaid program in any way, shape, or form,” said Ms. Lubinski, executive director of the HIV Medicine Association.

The HIV Medicine Association is a multidisciplinary arm of the Infectious Diseases Society of America that represents medical professionals involved in HIV care.

However, a few states—most notably New Jersey—have committed their Medicaid funds to cover broad-based HIV testing for low-income beneficiaries, Ms. Lubinski noted.

The Kaiser Permanente/Group Health Cooperative system covers approximately 3% of the entire U.S. population, including more than 16,000 active HIV-infected patients. The numbers vary widely by region, from about 180 patients in Ohio to nearly 5,500 in California.

Currently, nearly two-thirds of HIV-infected patients within Kaiser are not diagnosed until they meet AIDS criteria, “which means our case-finding is not very good,” Dr. Horberg remarked.

Once diagnosed, however, more than 90% enter into care within 120 days of diagnosis. Last year, more than 70% of those patients were on highly active antiretroviral therapy, he said.

Kaiser has been performing about 340,000 HIV antibody tests a year, which account for 15% of its target population aged 13–65 years. The majority are pregnant women, of whom more than 90% are currently tested. If Kaiser were to adopt the CDC guidelines, it would mean about 5 million more tests—and 1,773 newly identified cases—at a cost of at least $26,599,450 annually.

Aside from cost, other potential barriers to expanded HIV screening in managed care include the fact that many managed care organizations follow recommendations from the U.S. Preventive Services Task Force, not the CDC, in determining what type of tests to cover.

The USPSTF has not yet issued guidelines on universal HIV screening. Although most managed care organizations do support targeted screening for pregnant women and for individuals with high-risk behavior, they have not yet generated broader screening policies. “Most are probably waiting for the USPSTF,” Dr. Horberg said.

The CDC's provision that prevention counseling should not be required as part of HIV screening is already posing problems in states that require informed consent for HIV testing, including many of the states that Kaiser now serves.

Kaiser differentiates between “screening,” defined as testing without counseling, and “testing,” which includes the HIV antibody test, pre- and posttest counseling, and patient education.

“Testing in [Kaiser Permanente/Group Health Cooperative] is the desired norm…. We are uncomfortable screening without a proper testing process,” Dr. Horberg said. However, he added, despite the potential roadblocks, “We are confident we can handle all new HIV-infected patients identified.”

The public sector is another story. It would take an act of Congress before Medicare, which has only recently begun to cover any preventive health services, would cover HIV screening. Because the upper target age of the CDC recommendation is 64 years, the only people for whom Medicare would cover screening are the 6.8 million current beneficiaries under age 65 who qualify by disability. And that number includes about 100,000 who have already been diagnosed with HIV/AIDS, Ms. Lubinski said.

Thus, the bulk of the reimbursement for HIV screening would fall to Medicaid, which currently provides health coverage to about half of all people with AIDS in the United States and a significant number of those newly diagnosed with HIV. In an analysis done in 25 states, 22% of HIV-infected individuals were already Medicaid eligible at the time of diagnosis.

Federal law allows HIV screening to be covered by states either under fee-for-service or Medicaid managed care. This service is “optional,” and thus depends on the individual state's policy.

A recent study by George Washington University's Center for Health Services Research and Policy found that Medicaid programs in 32 of the 48 states surveyed covered targeted HIV testing and counseling, with 19 of those also covering prenatal and perinatal counseling. A few state programs also covered services such as HIV risk assessment and case management.

But as yet, with the exception of New Jersey, most state Medicaid programs have not adopted routine HIV testing. California has employed a special waiver to provide broad family planning services including HIV testing and counseling for men and women of childbearing age up to 200% of the poverty level.

However, that type of waiver is unlikely to be granted elsewhere, Ms. Lubinski noted. States could opt to cover HIV screening under a “diagnostic, screening, preventive, and rehabilitative” (DSPR) benefit. The state would need to broaden the definition of medical necessity to allow for preventive services such as HIV screening, which is what Massachusetts has done. There, a service is “medically necessary if it is reasonably calculated to prevent, diagnose, prevent the worsening of, alleviate, correct, or cure conditions in the member that endanger life, or cause suffering or pain,” the definition says.

Such definitions could theoretically make HIV testing and counseling eligible for reimbursement, Ms. Lubinski said.

She said she believes that the federal government will need to contribute more to Medicaid for the CDC guidelines to be fully implemented: “Medicaid, with its significant reach into low-income populations and ethnic/racial minorities, must be part of the financing mix.”

'Yes we have the capacity to do it, and yes, we have the will to do it. But it is a lot of money.' DR. HORBERG

ELSEVIER GLOBAL MEDICAL NEWS

ELSEVIER GLOBAL MEDICAL NEWS

WASHINGTON — Reimbursement for routine, universal HIV screening will prove challenging in both the private and public sectors, Dr. Michael Horberg and Ms. Christine Lubinski said in separate presentations at a meeting on HIV diagnosis and prevention and access to care.

In September 2006, the Centers for Disease Control and Prevention recommended that diagnostic HIV testing and “opt-out” HIV screening be made a part of routine clinical care in all health care settings for patients aged 13–64 years (MMWR 2006;55[RR-14]). Kaiser Permanente, the country's largest staff-model HMO, is “grappling with this now. We have to look at the implications,” said Dr. Horberg, director of HIV/AIDS Policy, Quality Improvement, and Research at Kaiser.

“Yes we have the capacity to do it, and yes, we have the will to do it. But it is a lot of money,” said Dr. Horberg.

As for the public sector, “There are significant roadblocks … The Centers for Medicare and Medicaid Services and the [Bush] administration have little commitment to expand the federal contribution to the Medicaid program in any way, shape, or form,” said Ms. Lubinski, executive director of the HIV Medicine Association.

The HIV Medicine Association is a multidisciplinary arm of the Infectious Diseases Society of America that represents medical professionals involved in HIV care.

However, a few states—most notably New Jersey—have committed their Medicaid funds to cover broad-based HIV testing for low-income beneficiaries, Ms. Lubinski noted.

The Kaiser Permanente/Group Health Cooperative system covers approximately 3% of the entire U.S. population, including more than 16,000 active HIV-infected patients. The numbers vary widely by region, from about 180 patients in Ohio to nearly 5,500 in California.

Currently, nearly two-thirds of HIV-infected patients within Kaiser are not diagnosed until they meet AIDS criteria, “which means our case-finding is not very good,” Dr. Horberg remarked.

Once diagnosed, however, more than 90% enter into care within 120 days of diagnosis. Last year, more than 70% of those patients were on highly active antiretroviral therapy, he said.

Kaiser has been performing about 340,000 HIV antibody tests a year, which account for 15% of its target population aged 13–65 years. The majority are pregnant women, of whom more than 90% are currently tested. If Kaiser were to adopt the CDC guidelines, it would mean about 5 million more tests—and 1,773 newly identified cases—at a cost of at least $26,599,450 annually.

Aside from cost, other potential barriers to expanded HIV screening in managed care include the fact that many managed care organizations follow recommendations from the U.S. Preventive Services Task Force, not the CDC, in determining what type of tests to cover.

The USPSTF has not yet issued guidelines on universal HIV screening. Although most managed care organizations do support targeted screening for pregnant women and for individuals with high-risk behavior, they have not yet generated broader screening policies. “Most are probably waiting for the USPSTF,” Dr. Horberg said.

The CDC's provision that prevention counseling should not be required as part of HIV screening is already posing problems in states that require informed consent for HIV testing, including many of the states that Kaiser now serves.

Kaiser differentiates between “screening,” defined as testing without counseling, and “testing,” which includes the HIV antibody test, pre- and posttest counseling, and patient education.

“Testing in [Kaiser Permanente/Group Health Cooperative] is the desired norm…. We are uncomfortable screening without a proper testing process,” Dr. Horberg said. However, he added, despite the potential roadblocks, “We are confident we can handle all new HIV-infected patients identified.”

The public sector is another story. It would take an act of Congress before Medicare, which has only recently begun to cover any preventive health services, would cover HIV screening. Because the upper target age of the CDC recommendation is 64 years, the only people for whom Medicare would cover screening are the 6.8 million current beneficiaries under age 65 who qualify by disability. And that number includes about 100,000 who have already been diagnosed with HIV/AIDS, Ms. Lubinski said.

Thus, the bulk of the reimbursement for HIV screening would fall to Medicaid, which currently provides health coverage to about half of all people with AIDS in the United States and a significant number of those newly diagnosed with HIV. In an analysis done in 25 states, 22% of HIV-infected individuals were already Medicaid eligible at the time of diagnosis.

Federal law allows HIV screening to be covered by states either under fee-for-service or Medicaid managed care. This service is “optional,” and thus depends on the individual state's policy.

A recent study by George Washington University's Center for Health Services Research and Policy found that Medicaid programs in 32 of the 48 states surveyed covered targeted HIV testing and counseling, with 19 of those also covering prenatal and perinatal counseling. A few state programs also covered services such as HIV risk assessment and case management.

But as yet, with the exception of New Jersey, most state Medicaid programs have not adopted routine HIV testing. California has employed a special waiver to provide broad family planning services including HIV testing and counseling for men and women of childbearing age up to 200% of the poverty level.

However, that type of waiver is unlikely to be granted elsewhere, Ms. Lubinski noted. States could opt to cover HIV screening under a “diagnostic, screening, preventive, and rehabilitative” (DSPR) benefit. The state would need to broaden the definition of medical necessity to allow for preventive services such as HIV screening, which is what Massachusetts has done. There, a service is “medically necessary if it is reasonably calculated to prevent, diagnose, prevent the worsening of, alleviate, correct, or cure conditions in the member that endanger life, or cause suffering or pain,” the definition says.

Such definitions could theoretically make HIV testing and counseling eligible for reimbursement, Ms. Lubinski said.

She said she believes that the federal government will need to contribute more to Medicaid for the CDC guidelines to be fully implemented: “Medicaid, with its significant reach into low-income populations and ethnic/racial minorities, must be part of the financing mix.”

'Yes we have the capacity to do it, and yes, we have the will to do it. But it is a lot of money.' DR. HORBERG

ELSEVIER GLOBAL MEDICAL NEWS

ELSEVIER GLOBAL MEDICAL NEWS

WASHINGTON — Reimbursement for routine, universal HIV screening will prove challenging in both the private and public sectors, Dr. Michael Horberg and Ms. Christine Lubinski said in separate presentations at a meeting on HIV diagnosis and prevention and access to care.

In September 2006, the Centers for Disease Control and Prevention recommended that diagnostic HIV testing and “opt-out” HIV screening be made a part of routine clinical care in all health care settings for patients aged 13–64 years (MMWR 2006;55[RR-14]). Kaiser Permanente, the country's largest staff-model HMO, is “grappling with this now. We have to look at the implications,” said Dr. Horberg, director of HIV/AIDS Policy, Quality Improvement, and Research at Kaiser.

“Yes we have the capacity to do it, and yes, we have the will to do it. But it is a lot of money,” said Dr. Horberg.

As for the public sector, “There are significant roadblocks … The Centers for Medicare and Medicaid Services and the [Bush] administration have little commitment to expand the federal contribution to the Medicaid program in any way, shape, or form,” said Ms. Lubinski, executive director of the HIV Medicine Association.

The HIV Medicine Association is a multidisciplinary arm of the Infectious Diseases Society of America that represents medical professionals involved in HIV care.

However, a few states—most notably New Jersey—have committed their Medicaid funds to cover broad-based HIV testing for low-income beneficiaries, Ms. Lubinski noted.

The Kaiser Permanente/Group Health Cooperative system covers approximately 3% of the entire U.S. population, including more than 16,000 active HIV-infected patients. The numbers vary widely by region, from about 180 patients in Ohio to nearly 5,500 in California.

Currently, nearly two-thirds of HIV-infected patients within Kaiser are not diagnosed until they meet AIDS criteria, “which means our case-finding is not very good,” Dr. Horberg remarked.

Once diagnosed, however, more than 90% enter into care within 120 days of diagnosis. Last year, more than 70% of those patients were on highly active antiretroviral therapy, he said.

Kaiser has been performing about 340,000 HIV antibody tests a year, which account for 15% of its target population aged 13–65 years. The majority are pregnant women, of whom more than 90% are currently tested. If Kaiser were to adopt the CDC guidelines, it would mean about 5 million more tests—and 1,773 newly identified cases—at a cost of at least $26,599,450 annually.

Aside from cost, other potential barriers to expanded HIV screening in managed care include the fact that many managed care organizations follow recommendations from the U.S. Preventive Services Task Force, not the CDC, in determining what type of tests to cover.

The USPSTF has not yet issued guidelines on universal HIV screening. Although most managed care organizations do support targeted screening for pregnant women and for individuals with high-risk behavior, they have not yet generated broader screening policies. “Most are probably waiting for the USPSTF,” Dr. Horberg said.

The CDC's provision that prevention counseling should not be required as part of HIV screening is already posing problems in states that require informed consent for HIV testing, including many of the states that Kaiser now serves.

Kaiser differentiates between “screening,” defined as testing without counseling, and “testing,” which includes the HIV antibody test, pre- and posttest counseling, and patient education.

“Testing in [Kaiser Permanente/Group Health Cooperative] is the desired norm…. We are uncomfortable screening without a proper testing process,” Dr. Horberg said. However, he added, despite the potential roadblocks, “We are confident we can handle all new HIV-infected patients identified.”

The public sector is another story. It would take an act of Congress before Medicare, which has only recently begun to cover any preventive health services, would cover HIV screening. Because the upper target age of the CDC recommendation is 64 years, the only people for whom Medicare would cover screening are the 6.8 million current beneficiaries under age 65 who qualify by disability. And that number includes about 100,000 who have already been diagnosed with HIV/AIDS, Ms. Lubinski said.

Thus, the bulk of the reimbursement for HIV screening would fall to Medicaid, which currently provides health coverage to about half of all people with AIDS in the United States and a significant number of those newly diagnosed with HIV. In an analysis done in 25 states, 22% of HIV-infected individuals were already Medicaid eligible at the time of diagnosis.

Federal law allows HIV screening to be covered by states either under fee-for-service or Medicaid managed care. This service is “optional,” and thus depends on the individual state's policy.

A recent study by George Washington University's Center for Health Services Research and Policy found that Medicaid programs in 32 of the 48 states surveyed covered targeted HIV testing and counseling, with 19 of those also covering prenatal and perinatal counseling. A few state programs also covered services such as HIV risk assessment and case management.

But as yet, with the exception of New Jersey, most state Medicaid programs have not adopted routine HIV testing. California has employed a special waiver to provide broad family planning services including HIV testing and counseling for men and women of childbearing age up to 200% of the poverty level.

However, that type of waiver is unlikely to be granted elsewhere, Ms. Lubinski noted. States could opt to cover HIV screening under a “diagnostic, screening, preventive, and rehabilitative” (DSPR) benefit. The state would need to broaden the definition of medical necessity to allow for preventive services such as HIV screening, which is what Massachusetts has done. There, a service is “medically necessary if it is reasonably calculated to prevent, diagnose, prevent the worsening of, alleviate, correct, or cure conditions in the member that endanger life, or cause suffering or pain,” the definition says.

Such definitions could theoretically make HIV testing and counseling eligible for reimbursement, Ms. Lubinski said.

She said she believes that the federal government will need to contribute more to Medicaid for the CDC guidelines to be fully implemented: “Medicaid, with its significant reach into low-income populations and ethnic/racial minorities, must be part of the financing mix.”

'Yes we have the capacity to do it, and yes, we have the will to do it. But it is a lot of money.' DR. HORBERG

ELSEVIER GLOBAL MEDICAL NEWS

ELSEVIER GLOBAL MEDICAL NEWS