Steven J. Jacobsen, MD, PhD

We evaluated an alternative way to implement guidelines using an automated risk calculator and risk-based decision tool to calculate patients’ risk of cardiovascular disease (CVD) and recommend therapies. We compared such an approach with traditional guidelines.

A retrospective cohort study of 1,506,109 Kaiser Permanente Southern California members 35 years or older.

We estimated 3-year risks of fatal and nonfatal myocardial infarction and stroke using an independently developed risk calculator, then graphically compared risks with observed outcomes. We used the area under the receiver operating characteristics curve to assess discrimination, and the Hosmer-Lemeshow statistic to test fit. We compared the characteristics and outcomes of populations identified for medication therapy by the risk-based decision tool and traditional guidelines using bivariate statistics.

A risk score was obtained in 72% (1,082,158) of members. The risk calculator was fairly good in discrimination: the area under the curve was 0.774 (95% CI, 0.770-0.779) for myocardial infarction and 0.805 (95% CI, 0.801-0.808) for stroke. Predictiveness and fit was good based on graphical analysis and Hosmer-Lemeshow 

<.0001. The risk-based decision tool identified high-risk patients for treatment who were not identified by traditional guidelines (3.80% of all those identified for statins, 3.04% for antihypertensives), as well as low-risk patients who were identified by guidelines (3.80% for statins, 2.51% for antihypertensives).

The risk calculator provided risk estimates in most patients and demonstrated fairly good discrimination and predictiveness. The risk-based decision tool identified high-risk patients for treatment not identified by traditional guidelines, as well as low-risk patients for whom treatment may be unnecessary.

	Automated risk calculator and guideline decision support tools can move clinical care a step closer to systematically individualizing patient therapy and population healthcare.

		The proportion of the population for whom sufficient electronic data were available to make risk estimation was good.

		The risk-based decision tool performed well in discrimination and predictiveness.  

		The risk-based decision tool identified high-risk individuals missed by traditional guidelines, and low-risk individuals who were recommended treatment based on traditional guidelines.

Clinical practice guidelines traditionally have specified levels of biomarkers to determine whether or not patients should be treated and to specify treatment goals. For example, the Seventh Report of the Joint National Committee (JNC-7) guideline for cholesterol specifies 3 different risk strata based on LDL cholesterol levels.

 Traditional guidelines can include other risk factors that are typically secondary to the biomarker target, such as gender or smoking status. A notable exception to the traditional approach is the new American College of Cardiology (ACC)/American Heart Association (AHA) cholesterol guideline. Although not explicitly dropping the use of cholesterol treatment goals, these new guidelines endorse the use of a risk calculator to guide treatment.

 Depending on their risk of a cardiovascular event, as estimated by a calculator, patients with high cholesterol levels may not be treated and patients with low cholesterol levels may be treated.

We wished to understand the potential implications of using risk calculators and risk-based decision tools to help risk stratify a population and identify individual patients for treatment. In the first part of our study, we evaluated the population applicability, discrimination, and predictiveness of an independently developed cardiovascular disease (CVD) risk calculator.

 The risk calculator was part of a risk-based decision tool under consideration by Kaiser Permanente Southern California (KPSC) for use in place of standard clinical practice guidelines. The validation of the risk calculator against outcomes was considered an essential first step before comparing the risk-based decision tool to traditional care. In the second part of our study, we tested the potential population impact of the risk-based decision tool by comparing actual outcomes in patients selected for possible treatment by the risk-based decision tool versus those selected by traditional clinical practice guidelines.

We conducted a retrospective cohort study of KPSC members with active membership during 2007 to 2010. The ending time point was chosen because at the time of initial analyses, government mortality data were not available past 2010. In the first part of this study, we determined the proportion of an adult population to which the risk calculator could be applied, then we compared predicted and observed outcomes. In the second part of the study, we determined which patients would be selected for medication therapy by the risk-based decision tool and by traditional guidelines, then we compared observed outcomes in these groups. Archimedes, Inc developed the risk calculator and the risk-based decision tool, and was involved in study planning and interpretation, but not in data collection or analysis. No data from KPSC were used to create the risk calculator or the risk-based decision tool evaluated in this study. The study was approved by the KPSC Institutional Review Board.

KPSC is a prepaid, integrated health system with 3.5 million members of diverse race and ethnicity at 14 medical centers and 197 medical offices throughout southern California. Members have similar coverage benefits and a limited range of co-payments for services and medications. Subjects 35 years or older were selected from among health plan members who were alive on December 31, 2007, with no gap in enrollment of more than 45 days in the year prior to that date.

The electronic health record (EHR) was used to identify pre-existing conditions using 

International Classification of Diseases, Ninth Revision, Clinical Modification

) codes assigned during inpatient or outpatient healthcare encounters. We collected data on past cardiovascular risk—related disease history from January 1, 2004 (up to 4 years prior to study outcomes), including diabetes, stroke, myocardial infarction, atrial fibrillation, congestive heart failure, and chronic kidney disease. For 2007, we extracted all the clinical information needed for risk calculation from laboratory and pharmacy data files. Laboratory data included fasting plasma glucose, glycated hemoglobin, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, total cholesterol, triglycerides, serum creatinine, and measures of proteinuria (eg, microalbumin-to-creatinine ratio). We identified medication dispensing dates for antihypertensives (ie, angiotensin converting-enzyme inhibitors and angiotensin receptor blockers, calcium channel-blockers, thiazide diuretics, beta-blockers), cholesterol-lowering drugs (ie, statins and non-statins), and insulin. We identified additional key risk factors from the EHR, including recent smoking, nonprescription aspirin use, body mass index, and blood pressure measurements.

We identified fatal and nonfatal myocardial infarction and cerebrovascular (ie, stroke) events from January 1, 2008, to December 31, 2010. These 3-year outcomes were identified from the EHR using 

 hospital discharge diagnosis codes (myocardial infarction: 410; and stroke: 346.6x, 430, 431, 432.x, 433.01, 433.11, 433.21, 433.31, 433.81, 433.91, 434.x, 436). Hospitalization discharge diagnosis codes were collected from claims data if the hospitalization occurred at a non-KPSC facility. The date and cause of death were determined from the California State Death Statistical Master Files and the Multiple Cause of Death Files for all subjects, regardless of health plan membership status.

Risk Calculator and Risk-Based Decision Tool

 We predicted CVD risk for all patients using an updated version of a previously developed and described risk calculator.

 The models used in the risk calculator were in the form of a proportional hazards survival model and were independently derived using publicly available data sets from the following studies: Framingham, Framingham Offspring, Atherosclerosis Risk in Communities Study (ARIC), and Cardiovascular Health Study (CHS).

 The risk calculator was designed to determine the 5-year risk of fatal and nonfatal myocardial infarction, and, as a separate estimate, fatal and nonfatal stroke. Multiple statistical imputation is used to handle missing data when possible. Confidence guardrails prevent presentation of risk information when the data for a particular patient are insufficient for imputation and risk calculation.

The risk-based decision tool is the set of risk calculators and additional models that allow computations of health outcomes and the effects of interventions on risk at a person-specific level. This tool has a software application that utilizes EHR data to create an interactive visual display that quantifies health risks and the benefits of different healthcare interventions for individual patients.

 The risk-based decision tool interface can be used to assist collaborative provider-patient decision making in place of standard clinical practice guidelines. The risk-based decision tool sets a threshold of expected benefits and, for each patient, calculates a Health Impact score—the expected gain in 5-year quality-adjusted life-years, depending on whether or not they follow a treatment recommendation (see 

] for details). Information on the expected risks and the outcomes associated with possible therapies is presented to patients and physicians for individualized decision making.

 In a population management context, the risk-based decision tool can be used to prioritize patients by the potential benefit of actionable interventions and rank interventions in order of expected benefit. Some of the models for intervention impact on risk were derived independently, and others were derived from risk equations used in the Archimedes Model, which allows computations of health outcomes and effects of interventions on a person-specific level.

 The model is validated on a continuous basis to remain relevant and comply with new medical guidelines.

All of these models were derived from and validated against data from multiple studies and public sources, but not data from Kaiser Permanente.

Identification of Patients for Medication Therapy

Traditional guidelines and the risk-based decision tool were each used to identify patients for medication therapy. KPSC traditional guidelines were based on the JNC-7 blood pressure guideline and the Adult Treatment Panel III (ATP III) cholesterol guideline, which incorporated use of the Framingham risk calculator for borderline cholesterol values. These traditional guidelines were converted into an automated algorithm that could be applied to the entire population. For the risk-based decision tool, patients were chosen using a Health Impact score that resulted in the same number of patients being selected for treatment as by traditional guidelines (the “same-size” population threshold).

Our focus for this evaluation was therapy with statins (HMG-Co-A reductase inhibitors) and antihypertensives for CVD event prevention. For both traditional clinical practice guidelines and the risk-based decision tool, a patient was not identified for a medication therapy if they were already on it or there was a history of allergy or prior adverse events based on EHR data.

For the first part of this study, we determined the proportion of patients for whom a risk calculation was possible. The calculation was not performed if the data required by the calculator were not present in the EHR and could not be imputed. For comparisons with actual 3-year outcomes, we linearly interpolated the risk calculations from the risk-based decision tool from 5 to 3 years. We used the area under the receiver operating characteristic curve (AROC) to assess how well the risk calculator sorted those who had an outcome from those who did not (ie, discrimination accuracy). We used graphical methods, as described by Pepe et al,

 to assess the risk calculator predictiveness across a range of risk percentiles, which provided a visual calibration and goodness of fit assessment. We used the Hosmer-Lemeshow statistic to test risk calculator model goodness of fit.

For the second part of the study, we compared the characteristics of the “same-sized” populations identified for medication therapy by the risk-based decision tool and traditional guidelines using the 

 test. Any of 1 or more antihypertensive medications were counted as a single medication recommendation. We determined the outcomes (3-year events/1000 persons) of patients who were identified for possible treatment or not based on the risk-based decision tool, with those identified for treatment or not based on traditional guidelines. We also determined the outcomes in “indeterminate” patients for whom the risk-based decision tool did not calculate a Health Impact score due to insufficient data to calculate risk or to support a treatment recommendation. These indeterminate cases also included patients with medical conditions not modeled (about 5% of all patients): advanced heart failure, late-stage kidney disease, and a CVD event that occurred under age 45. When implementing traditional guidelines, if there was insufficient data to apply the guidelines, patients were assigned to “no treatment” by default since traditional guidelines only specified when to treat, and this approach was consistent with population-based implementation of guidelines.

The study cohort consisted of 1,506,109 patients who met inclusion criteria. 

 describes selected characteristics of the study cohort. A risk calculation was possible for 1,082,158 patients (72% of the cohort). The mean (SD) follow-up time for patients in which a calculation was possible was 2.61 (0.86) years.

The risk-based decision tool sorted patients fairly well by their propensity to have a myocardial infarction or stroke: the AROC was 0.774 (95% CI, 0.770-0.779) for myocardial infarction, and 0.805 (95% CI, 0.801-0.808) for stroke. The Hosmer-Lemeshow fit statistic indicated good fit (

 shows the predictiveness curves of the risk calculator for fatal and nonfatal myocardial infarction (graph A) and stroke (graph B) for 20 different ranked risk groups. The predicted rates were slightly higher than observed rates for myocardial infarctions in the top 15th risk percentile. The predicted rates for strokes were slightly lower than the observed rates in several high-risk groups, although not the highest-risk group.

Treatment Stratification by the Risk-Based Decision Tool and Traditional Guidelines

Table 1 shows the characteristics of 3 subgroups identified for possible new medication treatment by: 1) the risk-based decision tool but not by traditional guidelines, 2) traditional guidelines but not by the risk-based decision tool, and 3) both. For almost all characteristics, there were statistically significant differences (

 <.0001) between the groups. The most notable difference was the older age of patients identified for possible treatment by the risk-based decision tool alone versus those identified by traditional guidelines alone. Using the “same size” population identified for treatment by both approaches, more than 50% of patients to whom medications were recommended by the risk-based guidelines only were 70 years or older. The corresponding percentage was less than 5% for those to whom medications were recommended by traditional guidelines only. Compared with traditional guidelines, the risk-based decision tool also identified proportionally more men, more patients with high-normal blood pressures, and more smokers.

 shows the percentage of the total cohort eligible for new medication treatment based on the risk-based decision tool, as well as the percentage based on traditional guidelines. The risk-based decision tool made statin recommendations in 34.85% of patients (8.73% for, 26.12% against), and antihypertensive recommendations in 18.20% of patients (4.82% for, 13.38% against). Traditional guidelines made statin recommendations in 9.69% of patients and antihypertensive recommendations in 4.94% of patients, with all others not recommended treatment by default. The CVD event rates per 1000 members among those to whom statins were recommended were similar using either approach, and somewhat higher among those to whom antihypertensives were recommended using the risk-based decision tool (72.0 vs 63.1). The event rates per 1000 members identified for treatment by both the risk-based decision tool and traditional guidelines were greater than the event rates identified by either approach alone.

Although there was overlap in the patients identified for treatment by the risk-based decision tool and traditional guidelines, the risk-based decision tool identified a number of patients who were at high risk, but who were not identified for treatment by traditional guidelines. For statins, the risk-based decision tool identified an additional 3.80% of the population (beyond the 9.69% identified by traditional guidelines) who were not identified for treatment by traditional guidelines but who had a CVD event rate (45.3 events/1000 patients) that was not much less than that of patients to whom treatment was recommended by traditional guidelines (50.1/1000 patients). Similarly, for antihypertensives, the risk-based decision tool identified an additional 3.04% of the population who were not identified for treatment by traditional guidelines, but who had a similar CVD event rate (65.9/1000 patients) to those for whom treatment was recommended by traditional guidelines (63.1/1000 patients).

The risk-based decision tool also made recommendations against therapy for a number of patients for whom treatment was recommended by traditional guidelines. For statins, the risk-based decision tool made recommendations against treatment for 3.80% of the population. These patients had a CVD event rate that was relatively low (15.5/1000 patients), and virtually identical to the risk among patients to whom treatment was not recommended by traditional guidelines (14.3/1000 patients). For antihypertensives, the risk-based decision tool made recommendations against treatment in 2.51% of the population; this group’s CVD event rate (19.9/1000 patients) was actually lower than that of patients to whom treatment was not recommended by traditional guidelines (22.7/1000 patients).

We undertook this study to evaluate an alternative way to implement guidelines using an automated risk calculator and decision support tool that captures EHR data to calculate patients’ risk of CVD outcomes, and identifies the impact of potential treatments. Our objectives were to determine its applicability, the usefulness of its predictions, and how patient identification for possible treatment compares with traditional guidelines. In doing so, we hoped to assess the implications of using a risk-based decision tool for both individual and population care.

The risk calculator was widely applicable to the health plan’s general adult population (72% had a risk estimate given 1 year of clinical data) and demonstrated fairly good discrimination and predictiveness. Our results indicate that when tuned to treatment in the same number of patients as traditional guidelines, the risk-based decision tool identifies a group of individuals who are likely to benefit from treatment but who are missed by current guidelines, even when those guidelines include Framingham risk. It also identifies a group of individuals for whom treatment is recommended by current guidelines, but have risks that are sufficiently low enough that they may safely be left untreated.

In a similar type of comparison, the newer ACC/AHA guideline for statin therapy based on a risk calculator was compared with ATP III guidelines with its cholesterol targets supplemented by Framingham risk in an existing community-based cohort.

 This study compared a different risk-based calculator (ACC/AHA) with essentially the same traditional guidelines as used by KPSC. The findings mirror ours in that the calculator had “greater accuracy and efficiency in identifying increased risk of incident CVD and subclinical coronary artery disease”

—similarly, more higher-risk patients were identified. Additionally of note, the risk-based approach of the ACC/AHA guidelines was deemed, in a simulation study, to have an acceptable cost-effectiveness profile.

 In this simulation study, the cost-effectiveness of the ACC/AHA risk-based approach varied with risk cut points.

 In our study, the Health Impact score was used to introduce the variation of risk to the population affected in place of pure risks.

A significant problem facing all risk-based calculators and decision support tools that use EHR data is that they might not be able to perform the desired calculations, either because of missing data or limitations in patient groups to which they are applicable (eg, late-stage kidney disease). Some of the patients who are excluded from risk models are at higher risk due to relatively infrequent medical conditions; in these situations, decisions are left to the discretion of the treating physician. Hopefully, future risk calculators will address this restricted scope of application. Risk-based decision tools also suffer from the same lack of evidence about treatment as do traditional guidelines. This is especially evident when considering that 72% of patients had a risk calculation, but among patients potentially eligible for a medication, 65% (statins) and 82% (antihypertensives) could not be definitively classified into treatment or no treatment groups.

Our study has several limitations, including that the KPSC system may differ in patient outcomes from other health delivery models. Our determination of who would be recommended for treatment was based solely on EHR data and automated guideline implementation rather than clinical practice. Use in practice may increase or reduce the identification of high-risk patients and the differences between approaches. Not all potentially relevant characteristics of a patient, or clinical actions, are captured in the EHR. Additionally, prediction accuracy was not adjusted for subsequent changes in risk factors or clinical treatments. Our results may vary somewhat over shorter or longer follow-up time frames. The risk-based decision tool identified some patients at increased risk who were not identified by traditional guidelines; however, we did not assess the separate contributions of imputation, calculator variables, and estimator accuracy in determining this difference in identification.

Articles

The authors evaluated the clinical applicability, accuracy, and implications of using an automated risk calculator and risk-based decision tool in an integrated health system.

Potential of Risk-Based Population Guidelines to Reduce Cardiovascular Risk in a Large Integrated Health System

To assess the impact of using the combined measles, mumps, rubella, and varicella vaccine (MMRV) instead of the measles, mumps, and rubella vaccine and the varicella vaccine separately (MMR+V) on uptake of vaccines recommended by the Advisory Committee for Immunization Practices (ACIP) in young children.

Electronic immunization records of children 12 to 15 months of age who received either MMR+V (n = 29,716) or MMRV (n = 29,711) between November 2003 and June 2007 while enrolled in Kaiser Permanente Southern California were evaluated. Proportions of children who received either MMRV or MMR+V and other ACIP-recommended vaccines either simultaneously or during recommended age intervals were compared.

Concomitant administration of DTaP (76.3 vs 75.6%), PCV7 (89.1 vs 78.6%), Hib (92.3 vs 91.3%), and hepatitis A vaccine (HAV) (44.2 vs 0.1%) and on-time administration of PCV7 (91.7 vs 78.7%), Hib (93.9 vs 93.2%), and HAV (86.8 vs 24.3%) were greater with MMRV than with MMR+V. On-time administration of DTaP (91.4 vs 90.5%) and IPV (91.3 vs 90.0%) was greater with MMR+V than with MMRV.

Concomitant and on-time administration of most ACIP-recommended vaccines was greater with MMRV than with MMR+V despite the addition of HAV to the recommended schedule after introduction of MMRV. However, the modest compliance increases suggest that focusing on other barriers to compliance may be important to increasing adherence to future vaccination recommendations.

Vaccination compliance improved slightly following the introduction of MMRV combination vaccine in place of the MMR and varicella vaccines given separately.

		This increase was detected in spite of addition of another vaccine to the immunization schedule at the same time that MMRV was introduced.

		The increase in compliance was small, suggesting that other barriers to vaccination compliance may need to be addressed in order to optimize adherence to future vaccinationrecommendations.

		Combination vaccines will continue to be important to maintaining vaccine complianceas new vaccines are added to the recommended immunization schedule.

Childhood immunizations have markedly reduced an increasing number of vaccine-preventable diseases.

 However, vaccination rates among children remain less than optimal, with nearly 30% of children failing to receive all recommended vaccines by 2 years of age

 and as many as 90% of children failing to receive all recommended doses at age-appropriate intervals.

In an attempt to improve childhood vaccination coverage rates, the standards for pediatric immunization practices direct providers to administer simultaneously all vaccines for which children are eligible at each visit.

 However, concerns of parents and providers about the pain and psychological trauma associated with the administration of multiple injections result in a reluctance to administer more than 2 to 3 injections simultaneously.

This frequently results in vaccine deferral,

 which substantially reduces vaccine coverage rates.

The importance of simultaneous administration of vaccines will grow as the number of vaccines included in the childhood vaccine schedule increases. Between 1991 and 2000, the number of injections necessary to provide all recommended vaccinations for children through age 23 months increased from 11 to 17, assuming the use of all combination vaccines then licensed in the United States.

However, development of combination vaccines may decrease the number of injections required to provide the increasing number of recommended vaccines, potentially improving immunization rates.

ProQuad (MMRV; Merck and Co, Inc, Summit, New Jersey), a combination of measles, mumps, and rubella (MMR) and varicella vaccines, was developed primarily to increase varicella vaccine coverage in children. However, we hypothesized that decreasing the number of injections required for children at 12 to 15 months of age, when the recommended vaccination schedule is particularly injection dense,

 might increase the simultaneous uptake of other childhood vaccines recommended by the Advisory Committee for Immunization Practices (ACIP). The objective of this study was to assess the impact of combining the MMR and varicella vaccines on the uptake of other vaccines given at the same time as MMRV or the MMR vaccine and varicella vaccine separately (MMR+V) in children 12 to 15 months of age.

Subjects for this study had been included in a phase-IV post-licensure retrospective cohort study that was conducted by Kaiser Permanente Southern California (KPSC) to evaluate the safety of a combined MMRV vaccine licensed in the United States in 2005 as previously described.

 Briefly, children 12 to 60 months of age who received a first dose of MMRV between February 2006 and June 2007 or MMR+V between November 2003 and January 2006 in KPSC were included in the study. Subjects who received MMRV were optimally matched without replacement on age, sex, and month and day of vaccination to children who received MMR+V. Children were excluded if they had a history of measles, mumps, rubella, or varicella disease, or a history of vaccination for any of these diseases. To allow validation of these criteria, subjects were required to be continuously enrolled in the Kaiser Foundation Health Plan from age 6 months until 90 days following vaccination. We compared the proportions of children aged 12 to 15 months who received either MMRV or MMR+V and the vaccinations recommended in the routine childhood vaccination schedule

 either simultaneously (concomitantly) or during recommended age intervals (on time). Study subjects were members of KPSC, the largest managed care organization in Southern California, serving 3.4 million members who are broadly representative of the diverse racial, ethnic, and socioeconomic background of the source population in Southern California. Vaccinations may be administered to members without charge at any visit including no-cost nurse visits. Immunizations are tracked by KPSC through the Kaiser Immunization Tracking System, one of the largest immunization tracking systems in the United States.

Several changes in the schedule of recommended childhood vaccines occurred during the study that affected measurements of compliance with vaccination recommendations (

). Due to a nationwide shortage of pneumococcal conjugate vaccine (PCV7), the fourth and the third PCV7 doses were suspended for low-risk children in February 2004 and March 2004, respectively; as the shortage resolved, these doses were reinstituted in July 2004 and September 2004, respectively.

Therefore, compliance with recommendations to administer PCV7 dose 4 in the MMR+V recipients was corrected by excluding children who received MMR+V between February 2004 and September 2004 when PCV7 dose 4 was suspended.

The recommendation to administer hepatitis A vaccine (HAV) at 12 months of age was made in January 2006

 just prior to introduction of MMRV at KPSC on February 1, 2006.

In order to estimate the impact of combining MMR and varicella vaccines on varicella vaccine uptake and to estimate the relative compliance of the source population for each cohort, we compared the uptake of MMRV with the uptake of MMR+V in 1-year samples of children drawn from the source population from which the cohorts were drawn for the phase-IV postlicensure study described previously.

 Vaccinated children were required to have had continuous membership from 6 months of age to the time of vaccination. The percentage of children born between March 1, 2005, and February 28, 2006, who received the first dose of MMRV at age 12 to 15 months  was compared with the percentage of children born between March 1, 2003, and February 28, 2004, who received the first dose of MMR+V at age 12 to 15 months.

Race and ethnicity data were available from health plan files for 82% of members included in this study and were complete for 90% after surname analysis and geocoding imputation.

Chi-square tests were used to compare the percentages. All tests were conducted using SAS software (SAS Enterprise Guide, version 4.1, SAS Institute Inc, Cary, North Carolina). This study was approved by the KPSC Institutional Review Board.

Of the 31,298 children each in the MMRV and MMR+V cohorts,18 29,711 and 29,716 were between 12 to 15 months of age, respectively. The demographics of the cohorts were similar (

Compliance Among MMRV and MMR+V Recipients

The on-time and concomitant uptake of recommended childhood vaccines among children who received MMRV or MMR+V between 12 and 15 months of age is shown in 

. Concomitant administration of the booster doses of diphtheria, tetanus, and pertussis (DTaP), PCV7, and H

(Hib) was greater with MMRV than with MMR+V (

 <.04). However, most differences were small (0.7%-10.5%). Concomitant administration of inactivated polio vaccine (IPV) dose 3 was low for MMRV and MMR+V (1.83%-1.92%). Similarly, on-time administration of booster doses of PCV7 and Hib vaccines was greater with MMRV than with MMR+V, though the differences were small (0.7%-13.0%). On the other hand, on-time administration of DTaP booster and IPV dose 3 was greater with MMR+V than MMRV (

<.001), but the differences were small (0.8%- 1.3%). The hepatitis A vaccine was given more often both concomitantly and on-time with MMRV than with MMR+V (

Combined on-time administration of all recommended vaccines (DTaP, PCV7, Hib, and IPV) was greater with use of MMRV than with MMR+V. This result held both with inclusion and exclusion of HAV from analysis (64.6% vs 18.5% and 74.1% vs 60.5%, respectively; both 

Uptake of MMRV vs MMR+V in Young Children

In 1-year samples of children from the cohort source populations, 81.4% of children who were born between March 1, 2005, and February 28, 2006, received MMRV at 12 to 15 months of age and 81.7% of children born between March 1, 2003, and February 28, 2004, received MMR+V at 12 to 15 months of age (

The recommended vaccine schedule has become increasingly injection dense, particularly at the 1-year visit. Strategies to address this situation have included giving vaccines at other ages and developing combination vaccines. However, many vaccines cannot be given at less than 12 months of age due to diminished efficacy.

 In addition, deferring vaccination results in decreased uptake.

 Combination vaccines make it possible to administer vaccines on schedule while reducing the number of injections given to patients.

 In this study, we evaluated the potential effect of combining MMR and varicella vaccines on the uptake of other recommended vaccines in children 12 to 15 months of age. While an increase was seen in the combined uptake of vaccines given concomitantly as well as on time with MMRV compared with MMR+V, the statistically significant but small increase in uptake of individual vaccines noted in this study is likely of limited clinical significance.

There are several possible reasons why a greater increase in vaccines given with MMRV compared with MMR+V was not seen. First, the recommendation to give HAV at the 12-month visit was made nearly simultaneously with the use of MMRV instead of MMR+V. Therefore, the number of injections recommended at the 12- to 15-month visit for nearly the entire MMRV cohort was same as that recommended for the MMR+V cohort. This is likely why on-time uptake of all recommended vaccines, including HAV, when MMRV was used (64.6%) was similar to on-time uptake of all recommended vaccines, not including HAV, when MMR+V was used (60.5%). However, while the number of injections received by each cohort was the same (n = 6), MMRV recipients were protected against 1 additional illness.

In addition, we compared 2 populations that received MMR-containing vaccine, making them more likely to be compliant with receipt of other vaccines.

 A slight reduction in the number of injections administered to compliant cohorts may have had less impact than it would have had in populations more reluctant to accept vaccines. Furthermore, the high compliance seen in both cohorts in this managed care setting may have had a ceiling effect, limiting further achievable increases in compliance. Moreover, vaccines are provided within KPSC without copayment and at no-cost visits. Hence, the reduced cost to caregivers of 1 less injection resulting from the use of MMRV rather than MMR+V seen in some settings was not a factor in this study setting.

 For these reasons, the results of this study conducted in a single managed care organization may not be generalizable to other populations and more diverse settings.

Finally, school and day care vaccine requirements may have increased compliance in both cohorts, thereby blunting any increase in compliance due to a decreased number of doses of vaccines that might otherwise have been seen in the MMRV cohort.

Combination vaccines have been found to increase compliance with vaccine administration recommendations.

 However, the minimal increase in uptake of individual vaccines and the continued suboptimal combined on-time uptake of all recommended vaccines seen following the introduction of MMRV in place of MMR+V suggest that other barriers to vaccination compliance need to be addressed. In addition, complications may arise as a result of combining vaccines. Shortly after completion of this study, ACIP withdrew its recommendation to administer MMRV preferentially as the first dose of MMR-containing vaccine in young children as a result of an increased incidence of febrile seizures following administration of MMRV compared with MMR+V.

Although vaccination compliance increased slightly following use of combination MMRV instead of separate MMR and varicella vaccines, additional barriers to improved compliance warrant future evaluation.

Impact of MMRV Combination Vaccine on Childhood Vaccination Compliance

To examine potential weight-related disparities in receipt of preventive screening exams and to compare several quality indicators and health behaviors among overweight/obese men and healthy-weight men enrolled in 2 large managed care plans.

 Cross-sectional analysis nested within a diverse cohort of men participating in the California Men’s Health Study (CMHS) (N = 80,771).

We extracted utilization of serum cholesterol, triglycerides, glucose, glycosylated hemoglobin, sigmoidoscopy exams, and prostatespecific antigen tests from health plan electronic sources. CMHS survey data provided information about diet and physical activity. Adjusted odds ratios and 95% confidence intervals were estimated to assess the association of screening exams and behaviors with categories of body mass index (BMI).

 Tests for cholesterol, glucose, and diabetes control increased across categories of BMI, while overweight and obese men were less likely to undergo screening exams for colorectal and prostate cancer. Smoking and alcohol consumption were less frequent among overweight/obese men; however, they reported diets higher in fat and lower in fruits and vegetables, and were much less likely to report moderate/vigorous activity and much more likely to be sedentary.

Managed care organizations might reduce weight-related health risks and disparities in care with targeted efforts to promote cancer screenings, healthy diets, and physical activity among overweight and obese patients.

	Concerns persist about quality of care and health behaviors that contribute to the elevated health risks of overweight/obese patients. Among men enrolled in managed care

		 Overweight/obese men were more likely to be screened for serum cholesterol, glucose, and among patients with diabetes, blood sugar control.

		 In contrast, overweight/obese men were less likely to be screened for colorectal and prostate cancer.

		 While overweight/obese men were less likely to consume alcohol or smoke, they reported less healthy diets and much less physical activity.

	The dramatic increase in the prevalence of overweight and obesity presents a serious threat to public health. Two-thirds of US adults are overweight (body mass index [BMI] 

), and approximately one-third are classified as obese (BMI 

 Overweight and obesity are associated with multiple serious and costly health outcomes, including coronary heart disease, high blood cholesterol, hypertension, stroke, type 2 diabetes, gallbladder disease, osteoarthritis, sleep apnea, and some forms of cancer such as kidney and colorectal.

 Adding to this substantial health burden are weightrelated social and psychological consequences that can adversely affect the quality of care received by overweight and obese patients, and promote behaviors that are harmful to their health.

	Negative attitudes and stereotypes about overweight and obesity are widespread and may be especially acute in the medical setting.

 Heavier patients report that physicians are one of the most common sources of weight bias.

 National surveys among large random samples of primary care physicians found more than 50% viewed obese patients as unattractive and noncompliant.

 Numerous studies find that health professionals describe overweight and obese patients as unmotivated and nonadherent to prevention and healthcare recommendations. 

 Consequently, physicians may have low expectations for heavier patients, find it less satisfying to treat them, and be less likely to promote preventive services or spend time on health education counseling.

 Weight stigma can further impact health through patients’ negative responses to perceived bias including delays in seeking care, missed appointments, unhealthy eating behaviors, and exercise avoidance.

	While the pervasiveness of weight-related stereotypes is well documented, it is unclear whether and how weight stigma translates into disparities in receipt of healthcare services. In response to persistent concerns about the quality of care received by overweight and obese patients, recent studies were conducted among Medicare recipients and US veterans.

 The purpose of this study conducted among socioeconomically diverse middle-aged and older men enrolled in 2 large managed care plans is 2-fold. First, we examined potential weightrelated disparities in receipt of preventive screening exams that are commonly included in quality performance measures. 

 Secondly, we compared health behaviors across categories of BMI to identify potential opportunities to reduce some of the increased risks of overweight and obesity.

	This study was conducted among subjects in the California Men’s Health Study (CMHS). Participants were recruited from the Kaiser Permanente Southern and Northern California health plans. Together, these nonprofit managed care organizations provide comprehensive healthcare to more than 6 million members. The study protocol was approved by the institutional review boards of both organizations.

	Eligible participants included all 850,000 male health plan enrollees between 45 and 69 years of age. Between 2002 and 2003, 84,170 men joined the cohort. Participants were comparable to similar-aged men enrolled in the health plans (and to men residing in California) with respect to sociodemographic characteristics and BMI.

Details of the cohort design and recruitment methods were described previously by Enger and colleagues.

	In all analyses, we excluded men for whom we could not calculate a valid BMI (n = 2863) or who reported a BMI of less than 18.5 kg/m

 (n = 536), resulting in a sample of 80,771. When examining cancer screenings, we excluded men younger than 50 years, those who had been a member of the health plans for fewer than 5 years, and those at aboveaverage risk for prostate or colorectal cancer (CRC).

	We extracted data on membership, primary care utilization, laboratory tests, and cancer screenings from health plan electronic databases. The CMHS baseline survey collected sociodemographic information (race/ethnicity, age, educational attainment, household income, marital status, country of birth), height and weight, personal and family histories of disease, and health behaviors.

	Body mass index was calculated as weight in kilograms divided by the square of height in meters (kg/m

) and classified according to guidelines of the National Heart, Lung, and Blood Institute.

 We defined underweight as a BMI of less than 18.5 kg/m

, healthy weight as a BMI of 18.5-24.9 kg/m

 or greater. We further categorized obesity into class I obesity (BMI 30.0-34.9 kg/m

) and class II/III obesity (BMI 35.0 kg/m

	We examined 6 quality performance measures for common preventive services related to cardiovascular disease, diabetes, and cancer screening. Recommendations for these tests are not weight related. Time intervals for the screening tests were based on the health plans’ clinical practice guidelines in place when the cohort was established. Utilization of laboratory tests was obtained for the 2 years prior to the baseline survey, including serum glucose, cholesterol, triglycerides, and among participants with diabetes, glycosylated hemoglobin (A1C). We assessed CRC screening with use of sigmoidoscopy, the first-line screening test for CRC recommended by the health plans in the 5 years preceding the baseline survey.

 To differentiate screening from diagnostic sigmoidoscopy we used 

Current Procedural Terminology, 4th Edition

 codes and a validated automated data algorithm

 and limited this assessment to men at average risk (ie, those with no personal history of CRC or colorectal polyps, and no family history of CRC). Serum prostate-specific antigen (PSA) testing was extracted from laboratory databases during the 5 years preceding the baseline survey. To differentiate screening from diagnostic tests, we excluded men whose PSA tests were ordered by a urologist, and men who reported they had completed a PSA test due to symptoms or as follow-up to previous abnormal results.

	Dietary factors were assessed with a detailed semiquantitative food-frequency questionnaire developed for the Women’s Health Initiative and other studies

 We calculated total energy intake, percentage of calories from fat, and daily servings of fruits and vegetables. Consumption of alcoholic beverages included beer, wine, and liquor.

	Physical activity was determined with questions adapted from the Coronary Artery Risk Development in Young Adults (CARDIA) Physical Activity History,

 which asked about the frequency and duration of participation in specific types of moderate and vigorous recreational, household, and work-related activities. The CARDIA Physical Activity History has indirect validity against aerobic capacity and percentage of fat, and a strong inverse relationship with most cardiovascular disease risk factors.

Regular moderate/vigorous physical activity was defined as 3.5 hours per week or more spent in activities of 3 or more metabolic equivalents. Participation in vigorous physical activity was defined as 3.5 hours per week or more spent in activities of 6 or more metabolic equivalents. We defined sedentary behavior outside of work with reports of the total time spent watching television, sitting at a computer, or reading 6 or more hours per day.

	We determined the frequencies of sociodemographic characteristics, preventive screening exams, and health behaviors to describe these factors across categories of BMI. Given the large sample size, and therefore the probability of detecting clinically irrelevant differences, analyses focused on the adjusted magnitude of the differences found for overweight/obese men compared with healthy-weight men. We fit multivariable logistic regression models to calculate the adjusted odds ratio (AOR) and 95% confidence interval (CI) estimating the effect of BMI categories on utilization of screening exams and health behaviors. Models were adjusted for race/ethnicity, age, education, income, marital status, country of birth, number of outpatient visits, department of primary care provider, and health plan. Models were run with BMI categorized as described above, as well as a continuous factor, to test for trends in utilization of screening exams and health behaviors with increasing BMI. Analyses were conducted using SAS version 9.1.3 (SAS Institute Inc, Cary, North Carolina).

	The majority of the subjects were white non-Latino, were less than 60 years old, had at least some college, had a household income of $60,000 or more, and were married or living with a partner. In the previous 5 years, a quarter had 31 or more outpatient visits (

). Almost two-thirds had a primary care physician in internal medicine, and approximately half were members of Kaiser Permanente Southern California. Most had been born in the United States (data not shown).

	A quarter of the men were in the healthy weight range (BMI 18.5-24.9 kg/m

), almost half were in the overweight range (BMI 25.0-29.9 kg/m

), a fifth were in the class I obesity range (BMI 30.0-34.9 kg/m

), and 8% were in the class II/III obesity category (BMI 35.0 kg/m

). Proportionately, fewer Latinos and African Americans and more Asians were in the healthy category compared with whites (16%, 18%, and 51% compared with 25%, respectively). Healthyweight men more often had a college education and were less likely to be born in the United States. As expected, healthyweight men had less outpatient utilization (Table 1).

describes utilization of preventive services. The overall frequency of testing for serum glucose, cholesterol, and triglycerides in the previous 2 years was 52%, 70%, and 63%, respectively. Among men with diabetes, 70% had an A1C test. Across all these tests, adjusted results found overweight and obese men were significantly more likely to have been tested. Less than one-third of men 50 years of age or older at average risk for CRC had undergone screening sigmoidoscopy in the preceding 5 years. In contrast to results found for laboratory tests, compared with healthy-weight men, CRC screening decreased significantly across categories of BMI (overweight men, AOR 0.93, 95% CI 0.88-0.99; class I obesity, AOR 0.84, 95% CI 0.78-0.90; class II/III obesity, AOR 0.72, 95% CI 0.65-0.80). Similarly, while almost three-fourths of men had a PSA test in the previous 5 years, those with class I obesity were 10% less likely and men with class II/ III obesity were 20% less likely than healthy-weight men to have been tested.

displays men’s reported health behaviors by BMI categories. As expected, total daily average calorie intake increased with BMI. Less than one-third of the men met national recommendations for dietary fat intake (

30% calories from fat), and the likelihood of meeting the recommendations decreased significantly with increasing BMI. Compared with healthy-weight men, obese men were less than half as likely to eat a lower fat diet. Similarly, fewer than one-third of all men met the national recommendation for consuming at least 5 daily servings of fruits and vegetables, with overweight and obese men about 10% less likely than healthy-weight men to meet this recommendation. Although almost two-thirds of the men reported participating in regular moderate/vigorous activity, overweight men were almost 20% less likely and class I and II/III obese men were 49% and 69%, respectively, less likely than were healthyweight men to be active. Similar results were found for vigorous activity. Conversely, the percentage of men reporting 6.5 or more sedentary hours each day increased significantly from the healthy to obese weight categories. On the positive side, heavier men were less likely to report drinking more than 5 alcoholic beverages per week or to be a current smoker, and were more likely to have never smoked. Of note, similar results were found when the analyses for preventive services and health behaviors were run with BMI as a continuous variable, and when men with class II/III obesity were included in or excluded from the models.

	In a diverse cohort of middle-aged and older men, we assessed utilization of preventive screening exams and health behaviors by BMI ranging from healthy weight to class II/III obesity in the managed care setting. We found no evidence for weight-related disparities in laboratory tests associated with cardiovascular disease and diabetes. However, despite equivalent access to care, overweight and obese men were less likely to be screened for CRC and prostate cancer. In addition, while heavier men reported less consumption of alcoholic beverages and were less likely to smoke, their diets and levels of physical activity more often failed to meet national recommendations for health.

	The majority of Americans are at elevated risk for numerous diseases as a result of their higher-than-recommended weight. The widespread and increasing prevalence of perceived weight discrimination in the United States

 raises concern about the healthcare that heavier patients receive. Unhealthy behaviors that add to the adverse effects of overweight/obesity are common and may be exacerbated in response to perceived weight bias. Our findings highlight factors that healthcare organizations could address to reduce the risk associated with overweight and obesity.

	Previously, the limited studies of weight-related disparities in care found mixed results, perhaps due to small sample size and reliance on self-report.

 In addition, those studies were mainly focused on cancer screening in women.

Recently, 2 large studies assessed quality indicators among overweight and obese patients during comparable years and with time intervals for screening exams similar to those in our study. Like us, Chang and colleagues

 found higher rates of A1C and lipid screening among overweight and obese male Medicare beneficiaries and US veterans. However, their examination of lipid screening was limited to patients with diabetes. They found no differences in receipt of CRC screening by weight category; however, their overall rates were higher (52% and 73% in the Medicare and veteran samples, respectively, compared with 32% in this study), possibly due to the older age of the subjects, inclusion of fecal occult blood testing as a CRC screening test, and reliance on self-reports from many subjects. Also in contrast to our study, Yancy and colleagues

 found higher rates of CRC and PSA testing among obese male veterans compared with those at healthy weights. Although they too included more screening tests for CRC in addition to sigmoidoscopy, overall rates were similar to those found in our analysis. The rate of PSA testing, on the other hand, was less than half of the rate observed among our health plan members, perhaps due to services received outside the US Department of Veterans Affairs system. Unlike previous studies, we attempted to exclude tests for CRC and prostate cancer that had a high probability of being ordered for diagnostic purposes instead of screening. This may account for differences in our findings. Overweight and obese men are at greater risk for colon cancer

 and possibly more aggressive forms of prostate cancer,

 and have higher prostate cancer mortality.

 Managed care organizations should monitor the use of cancer screening exams among heavier patients, raise provider awareness about potential disparities, and promote use among all patients. Future studies of weight-related disparities need to examine cancer screening among patients of similar age, avoid reliance on self-reports of health service utilization, and distinguish between screening and diagnostic exams.

, 2005 recommended a diet limiting fats to 20% to 35% of total calories and consumption of at least 5 servings of fruits and vegetables per day.

 While less than one-third of men met those dietary goals, overweight and obese men were significantly less likely to consume a diet that decreased their risk for cardiovascular disease, type 2 diabetes, and some cancers, as well as weight gain and obesity. Similarly, physical activity is associated with multiple health benefits

 and prevents weight gain and increases weight loss, especially when combined with reduced caloric intake. Of concern, this study found obese men were at least half as likely as healthyweight men to report recommended levels of moderate/vigorous physical activity. Despite long-standing calls for physicians to advise patients about the risks of excess weight,

 fewer than half of obese patients report receiving such advice.

 Yet when patients are advised by their physicians, they are more likely to try to lose weight, demonstrating once again the importance of physician recommendations.

We found overweight and obese men made substantially more primary care visits, thus providing clinicians added opportunities to counsel patients about health behaviors, as well as cancer screening.

	A recent review described the growing evidence for the impact of weight bias on healthcare utilization as a result of both physician and patient factors.

 Misconceptions about the causes of obesity and the motivation of heavier patients are commonly identified in surveys and experimental studies conducted among health  professionals.

On the other hand, overweight and obese patients may be reluctant to seek care due to embarrassment and perceived antifat bias.

Provider education can help reduce negative attitudes about obesity

and consequently might reduce weight-related disparities in care. Education also can help clinicians be sensitive to the stigmatizing experiences heavier patients deal with in the medical setting, and the importance of their attitude and choice of words when treating these patients.

Future studies need to identify effective interventions for counseling patients about weight-related health issues. Healthcare organizations must ensure providers have and spend sufficient time in counseling their overweight and obese patients.

	A strength of this study is the availability of health plan data, as patients who feel that physicians are negatively biased may also have a greater tendency to underreport care. Other strengths include the large, multiethnic cohort of men and the examination of healthcare utilization without the confounding effect of insurance coverage. In contrast to previous reports, we distinguished diagnostic from screening cancer exams and adjusted results for outpatient utilization.

	While responses to the CMHS survey are subject to the potential limitations of self-reported cross-sectional data, it is reassuring that the prevalences of overweight and obesity were similar to those in reports with men living in California

 in comparable years. In addition, we were able to validate the self-reported heights and weights with the electronic medical records of nearly 20,000 men.

 Overall reports of dietary data and levels of physical activity are consistent with national surveys.

 Other limitations include our inability to explain why overweight and obese men were less likely to be screened for CRC and prostate cancer. Perhaps urgent problems or management of comorbidities were the focus of physician concern during medical encounters. Tests might have been recommended, but patients might have refused or did not follow through with exams that were ordered. In addition, these results might not generalize to women or to younger adults, for whom the stereotypes and stigma associated with obesity may be more salient than in this older cohort.

 Future research into weight-related disparities in care and health behaviors should include reports from both patients and providers about delivery and adherence to health recommendations. Effective patient counseling and interventions that reduce barriers to care need to be developed and tested among overweight and obese patients.

Healthcare organizations may reduce weight-related health risks and disparities in care among overweight/obese patients through promoting cancer screening exams, healthier diets, and physical activity.

Steven J. Jacobsen, MD, PhD

Articles

Potential of Risk-Based Population Guidelines to Reduce Cardiovascular Risk in a Large Integrated Health System

Impact of MMRV Combination Vaccine on Childhood Vaccination Compliance

Preventive Care and Health Behaviors Among Overweight/Obese Men in HMOs

Correlates of Prostate-Specific Antigen Testing in a Large Multiethnic Cohort

Agreement Between Patient-reported Symptoms and Their Documentation in the Medical Record

Electronic Medical Records for Clinical Research: Application to the Identification of Heart Failure

The Impact of a Simulated Immunization Registry on Perceived Childhood Immunization Status