The triple option choice: self-selection bias in traditional coverage, HMOs, and PPOs.

Health care costs and utilization by salaried employees and their dependents at a large self-insured midwestern industrial manufacturing corporation were analyzed for the year before employees were first offered a triple option choice. Members had the option of retaining traditional Blue Cross and Blue Shield of Michigan (BCBSM) coverage or switching to either a number of health maintenance organizations (HMOs) or a number of preferred provider organizations (PPOs). Members who switched to HMOs or PPOs were generally younger and had lower average expenses and utilization rates than those who retained the traditional BCBSM plan. The results suggest that a selection bias does occur in this population, as lower cost members were more attracted to the HMOs and PPOs than were more expensive members. Implications for the corporation as well as for the drive toward managed care alternatives are discussed.

Reliability and validity of utilization review criteria. Appropriateness Evaluation Protocol, Standardized Medreview Instrument, and Intensity-Severity-Discharge criteria.

A study was conducted to assess the reliability and validity of the Appropriateness Evaluation Protocol (AEP), the Standardized Medreview Instrument (SMI) and the Intensity-Severity-Discharge criteria set (ISD), three utilization review instruments used to determine whether inpatient care is required. Reliability and validity were assessed for retrospective application of these instruments to charts of a sample of 119 medical cases from 21 hospitals in the state of Michigan. The reliability of each instrument was determined by having the instrument applied by two different nurse reviewers to each hospital record. Results indicated that the AEP and ISD were moderately reliable, while the SMI had low reliability. The validity of each instrument was tested by comparing the judgments of nurse reviewers using the instruments with the judgment of a panel of physicians. The AEP and ISD were found to be moderately valid and the SMI was found to have low validity. Results suggested that the SMI should not be used. The modest level of validity of the other two instruments suggests that payment should never be denied on the basis of the instrument alone. Payment should be denied only if a physician confirms the judgment based on the instrument that inpatient care was not required.

Focused utilization review strategies.

Researchers from the Michigan Health Care Education and Research Foundation, a research affiliate of Blue Cross and Blue Shield of Michigan, compared three focused utilization review (UR) strategies to determine which method most effectively and efficiently identifies nonacute inpatient hospital admissions. Intensity, Severity, Discharge-Appropriateness (ISD-A) criteria were used to identify nonacute admissions in 8,973 cases in 73 Michigan hospitals. Significant proportions of nonacute admissions were found in medical, psychiatric, and substance abuse cases; surgical admissions had the lowest rates. Strategies involving the concentration ratio were most effective at indicating potential efficiency gains. Focused UR on Diagnosis-Related Groups (DRGs) with nonacute rates greater than 15% captured 41% of admissions and accounted for 85% of nonacute admissions, 85% of nonacute days, and 80% of potential dollar savings. This suggests that UR efforts focused primarily on DRGs with high nonacute rates would significantly improve the efficiency and effectiveness of the overall UR process.

Efficiency gains in utilization review.

A study was conducted to determine which Major Diagnostic Categories (MDCs) accounted for most of the nonessential hospital admissions in 73 Michigan hospitals in 1986. The Intensity, Severity, Discharge-Appropriateness (ISD-A) criteria set was used to identify nonessential admissions. Large concentrations of nonessential admissions were found in medical, psychiatric, and substance abuse cases; surgical admissions had low rates of nonessential hospitalization. Focusing utilization review on MDCs with nonessential rates exceeding 15% results in an examination of 44% of admissions, while capturing 77% of nonessential admissions and 73% of potential dollar savings. Implications for UR activities are discussed.

Association of dietary factors and other coronary risk factors with social class in women in five Indian cities.

The association between social classes, food intake and coronary risk factors was determined. Cross-sectional surveys were conducted in 6-12 urban streets in each of five cities, each one from five different regions of India using similar methods of dietary intakes and criteria of diagnosis. We randomly selected 3257 women aged 25-64 years inclusive, from Moradabad (n = 902), Trivandrum (n = 760), Calcutta (n = 410), Nagpur (n = 405) and Bombay (n = 780). All subjects, after pooling of data, were divided into social class 1 (n = 985), class 2 (n = 790), class 3 (n = 774), class 4 (n = 602) and class 5 (n = 206) based on various attributes of socioeconomic status. Social class 1 was the highest and 5 was the lowest social class. Social classes 1-3 had greater intake of pro-atherogenic foods; total visible fat, milk and milk products, meat and eggs, as well as sugar and confectionery, compared to social classes 4 and 5. The consumption of wheat, rice, millets, fruits, vegetables and legume/total visible fat ratio were inversely associated with social class. Mean body mass index (BMI), obesity, overweight, central obesity and sedentary lifestyle were also significantly more common among subjects from higher social classes. Spearman's rank correlation showed that bodyweight, BMI, wheat, rice, millets, total visible fat, milk and milk products, meat, eggs, sugar and jaggery intakes were significantly correlated with social class. Social class 5 subjects had a lower intake of all foods and a lower BMI, suggestive of a higher rate of undernutrition among them. The findings indicate that the consumption of pro-atherogenic foods and other coronary risk factors are more common in higher social classes compared to lower social classes.

Prevalence and determinants of central obesity and age-specific waist:hip ratio of people in five cities: the Indian Women's Health Study.

OBJECTIVE: To study the prevalence of central obesity and age-specific waist:hip ratio of urban women from five Indian cities. DESIGN AND SETTING: Cross-sectional surveys were conducted in 6-12 urban streets in different parts of India using similar methods of sample selection and criteria of diagnosis. SUBJECTS AND METHODS: We randomly selected 3212 women, aged 25-64 years, from Moradabad (n = 902), Trivandrum (n = 760), Calcutta (n = 365), Nagpur (n = 405), and Bombay (n = 780). Evaluation was by a questionnaire administered by a physician and a dietician, a physical examination, and anthropometric measurements. RESULTS: The overall prevalence of central obesity among the total number of women was 55.0%, with the highest prevalence in Calcutta (62.2%) and the lowest in Bombay (47.4%). Waist:hip ratio was 0.85 +/- 0.13 (mean +/- SD) with the highest ratio for women in Calcutta (0.87 +/- 0.12) and the lowest for women in Moradabad (0.84 +/- 0.16). After pooling of data from all five cities, multivariate logistic regression analysis showed that, regardless of age, body mass index (> 23 kg/m2; odds ratio 1.12), sedentary lifestyle (odds ratio 2.51), and family history of obesity (odds ratio 2.15) were strongly associated with central obesity. Excess intake of fat was weakly associated with central obesity but age was not a risk factor for central obesity, although the prevalence was highest among those aged over 55 years in Moradabad, Calcutta, and Nagpur. CONCLUSIONS: The overall prevalence of central obesity among the urban women of India has increased, more so in Calcutta and Trivandrum. Body mass index, sedentary lifestyle, and family history of excess intake of fat were significant risk factors for central obesity.