Predictors of HIV status among injecting drug users and health promotion

Two consecutive samples one year apart of injecting drug users (n = 754 and n = 345) were collected in Sydney, Australia and analysed for predictors of Human Immunodeficiency Virus (HIV) prevalence. Data indicated that similar variables were associated with HIV infection in both waves of the study. Risks for HIV infection included number of injections in last typical using month, acceptance of used injecting equipment from other injecting drug users (IDUs) who were known to be infected either before or after the sharing occurred, having sex with people known to be infected with HIV, and sexual orientation. It was not possible to determine whether sexual or equipment sharing with known HIV infected people preceded or followed HIV infection. These data confirm that predictors of HIV prevalence in Sydney are similar to those found in overseas studies and that sexual orientation appears to be the most powerful predictor. These data suggest both that sexual contact is an important route of infection in IDUs, and that sexual risks for HIV infection in IDUs need to be emphasised.

Physician profiling. An analysis of inpatient practice patterns in Florida and Oregon

BACKGROUND

Physician profiling is a method of cost control that focuses on patterns of care instead of on specific clinical decisions. It is one cost-control method that takes into account physicians' desire to curb the intrusion of administrative mechanisms into the clinical encounter. To provide a concrete example of profiling, we analyzed the inpatient practice patterns of physicians in Florida and Oregon.

METHODS

Data for 1991 from Medicare's National Claims History File were used to profile 12,720 attending physicians in Florida and 2589 in Oregon. For each attending physician, we determined the total relative value of all physicians' services delivered during each patient's hospital stay. Relative value was measured in relative-value units (RVUs), according to the resource-based relative-value scale used by Medicare in determining payments to physicians. The mean number of RVUs per admission was then adjusted for the physician's case mix according to the patients' assigned diagnosis-related groups. The influence of the physician's specialty and of selected types of services (such as imaging and endoscopy) was also examined.

RESULTS

Florida physicians used markedly more resources, on average, than their colleagues in Oregon (46 vs. 30 case-mix-adjusted RVUs per admission). The difference was apparent for all specialties and all types of service. To illustrate the profiling data potentially available to the medical staffs of individual hospitals, we examined specific data on individual attending physicians and for various types of service for three hospitals' staffs. Despite similar overall profiles that fell below the national mean, each staff had a different practice pattern and would require different efforts to improve efficiency.

CONCLUSIONS

In an effort to encourage further debate, we have described one method of physician profiling. Profiling data help identify and characterize differences in practice style to which individual physicians or hospital staffs can respond. Because profiling is not based on rigid rules, it is a cost-containment strategy that can easily accommodate legitimate exceptions; it is therefore preferable to methods in which the appropriateness of each clinical decision is judged separately.

Bromide pharmacokinetics in cystic fibrosis

OBJECTIVE

Individuals with cystic fibrosis (CF) have altered kinetics for a number of drugs, most often an increased volume of distribution (Vd) per body weight and increased clearance per body weight. To further evaluate those differences, we studied bromide kinetics (Vd, elimination rate constant, and clearance) and body mass index in eight adults with mild-to-moderate forms of CF, 21 obligate carriers of the CF gene, and 21 healthy controls. Bromide distribution approximates the extracellular fluid volume and bromide is excreted unchanged by the kidney.

DESIGN

Individuals were given a single oral dose of bromide (50 mg/kg), and serum bromide concentrations were measured over 4 weeks. Bromide pharmacokinetics (Vd, elimination rate constant, and clearance) were determined using a one-compartment model with first-order kinetics. Body mass index was determined for each individual.

RESULTS

Individuals with CF had a significantly greater lean body mass per kilogram as estimated by body mass index compared with individuals in the obligate carrier and control groups. The mean (+/- SD) Vd per kilogram for the CF group (311 +/- 29 mL/kg) was significantly greater than that of the obligate carrier group (261 +/- 26 mL/kg) and the control group (274 +/- 30 mL/kg). However, the mean (+/- SD) Vd per square meter for the three groups was similar. The mean elimination rate constant for the CF group (3.55 +/- 0.98 x 10(-3)/h) was significantly greater compared with the mean elimination rate constant for the obligate carrier group (2.55 +/- 0.36 x 10(-3)/h) and the control group (2.58 +/- 0.49 x 10(-3)/h). The mean (+/- SD) clearance per kilogram was also significantly greater for the CF group (1095 +/- 283 microL/kg per hour) compared with the obligate carrier group (664 +/- 100 microL/kg per hour) and the control group (700 +/- 115 microL/kg per hour).

CONCLUSIONS

These findings indicate that individuals with CF have a greater Vd per kilogram for bromide and drugs that distribute in the extracellular fluid volume because of their greater lean body mass per kilogram. The findings also suggest that individuals with CF have a greater renal clearance of bromide and presumably of other anionic drugs excreted by the kidney. The results emphasize the importance of body composition in drug disposition.

Statistical prediction of the locus of endoproteolytic cleavage of the nascent polypeptide in glycosylphosphatidylinositol-anchored proteins

Existing methods of identifying the cleavage site of the nascent polypeptide and the C-terminal residue to which the glycosylphosphatidylinositol (GPI) anchor is attached in mature GPI-anchored proteins are technically difficult and labour-intensive. We tested the hypothesis that it was possible to predict this locus using data from the cDNA-deduced amino acid sequence and amino acid composition of GPI-anchored proteins. We employed a statistical approach which allowed repeated chi 2 comparisons between the proportions of residual amino acids in the major body of the cDNA-deduced polypeptide (minus the N-terminal signal peptide) after repeated computer-generated progressive exoproteolysis from its C-terminus one amino acid at a time and the fixed proportion of amino acids obtained from amino acid analysis of the mature GPI-anchored protein. Initial comparison of the two parameters invariably revealed a relatively high chi 2 statistic which progressively lowered to a minimum point at which the amino acid proportions of progressively exoproteolysed polypeptide and fixed endoproteolysed polypeptides of the mature GPI-anchored protein were in closest agreement. This objectively defined and unique minimum point of closest agreement accurately identified the locus of post-translational endoproteolytic cleavage of the nascent polypeptide in several tissue-specific single-gene-encoded GPI-anchored proteins. Thus the C-terminal amino acid to which the GPI anchor is attached can be rapidly identified using data from the cDNA sequence and the amino acid composition of proteins suspected to be GPI-anchored.

Proposals to control high-cost hospital medical staffs

Several health care reform bills would limit Medicare payments to high-cost medical staffs, that is, physicians in hospitals with a high volume of physician services per admission. In a given year, Medicare's payment to the physicians on each hospital's medical staff could not collectively exceed a limit defined as a certain percentage above the national median. Limits of various forms are used in other parts of the Medicare program. This policy would combine cost containment incentives with a clear organizational structure. In addition, medical staffs could be provided with detailed information on their practice styles.

Carboplatin (CBDCA) plus alpha interferon in metastatic non-small cell lung cancer. A Hoosier Oncology Group phase II trial

Interferons have been shown to increase in vitro cytotoxicity of platinum compounds. The Hoosier Oncology Group has conducted a Phase II clinical trial to determine if interferon alpha-2a (IFN-alpha-2a) given in combination with carboplatin (CBDCA) can increase response rates or survival in patients with metastatic or recurrent inoperable non-small-cell lung cancer. Forty-four patients with no prior chemotherapy and high KPS (80-100) were enrolled. CBDCA 400 mg/m2 was given intravenously on day 1 and IFN-alpha-2a 9 million units was given subcutaneously on days 1, 3, and 5. Treatment was administered every 4 weeks until onset of progressive disease or to a maximum of 4 courses: 37 patients (84%) received at least 2 courses, whereas only 16 (36%) received the full 4 courses. Dose-limiting toxicities were leukopenia (27%) and thrombocytopenia (20%) attributable to CBDCA. Grade 2-3 anemia occurred in 32%. Only 4-7% of patients experienced severe fever, fatigue, or flu-like symptoms attributable to interferon administration. Of 41 patients evaluable for response, there were no complete responses and only 3 (7.3%) partial remissions. The overall median survival was 6 months. The combination of CBDCA and IFN-alpha-2a given in this dose and schedule does not appear to have superior activity compared to CBDCA alone in patients with non-small-cell lung cancer.

The analysis of longitudinal polytomous data: generalized estimating equations and connections with weighted least squares

In recent years, methods have been developed for modelling repeated observations of a categorical response obtained over time on the same individual. Although situations in which the repeated response is binary or Poisson have been studied extensively, relatively little attention has been given to polytomous categorical response variable. In this paper, we extend the estimating equations initially developed for clustered discrete data by Liang and Zeger (1986, Biometrika 73, 13-22), and subsequently extended by Prentice (1988, Biometrics 44, 1033-1048), to polytomous response variables. Under certain assumptions, we illustrate that these estimating equations simplify to the weighted least squares (WLS) equations formalized by Koch et al. (1977, Biometrics 33, 133-158). This connection provides a formal framework for obtaining iterated weighted least squares model parameter estimates. Cumulative logit models are developed and applied to a representative longitudinal data set. Simulation results comparing WLS, an iterative form of WLS, and independence estimating equations using a robust estimate of the variance are presented.

Posterior pelvic fixation using a transiliac 4.5-mm reconstruction plate: a clinical and biomechanical study

Operative management of unstable pelvic fractures is directed toward stabilization of the posterior lesion. We describe a technique of posterior pelvic fixation that uses a 4.5-mm reconstruction plate as a transiliac tension band. The plate is inserted through the posterior superior iliac spines with screw fixation to the ilium. Our initial clinical experience was gained using this technique in 15 patients who had unstable pelvic ring injuries with sacral fractures. Stable fixation was achieved in all patients with this low-profile plate. There were no infections, no wound complications, and no failures of fixation. Comparative biomechanical testing using cadaveric and artificial pelvises demonstrated that the strength of the transiliac plate method was equal to that of other techniques of posterior pelvic fixation.

Requiring physicians to respond to computerized reminders improves their compliance with preventive care protocols

OBJECTIVE

To improve compliance with computer-generated reminders to perform fecal occult blood testing (FOBT), mammography, and cervical Papanicolaou (Pap) testing.

DESIGN

Six-month prospective, randomized, controlled trial.

SETTING

Academic primary care general internal medicine practice.

SUBJECTS

Thirty-one general internal medicine faculty, 145 residents, and 5,407 patients with scheduled visits who were eligible for any of the three cancer screening protocols.

INTERVENTION

Primary care teams of internal medicine residents and faculty received either routine computer reminders (control) or the same remainders to which they were required to circle one of four responses: 1) "done/order today," 2) "not applicable to this patient," 3) "patient refused," or 4) "next visit."

RESULTS

Intervention physicians complied more frequently than control physicians with all remainders combined (46% vs 38%, respectively, p = 0.002) and separately with remainders for FOBT (61% vs 49%, p = 0.0007) and mammography (54% vs 47%, p = 0.036) but not cervical Pap testing (21% vs 18%, p = 0.2). Intervention residents responded significantly more often than control residents to all reminders together and separately to reminders for FOBT and mammography but not Pap testing. There was no significant difference between intervention and control faculty, but the compliance rate for control faculty was significantly higher than the rate for control residents for all reminders together and separately for FOBT but not mammography or Pap testing. The intervention's effect was greatest for patients > or = 70 years old, with significant results for all tests, together and singly, for residents but not faculty. Intervention physicians felt that the reminders were not applicable 21% of the time (due to inadequate data in patient's electronic medical records) and stated that their patients refused 10% of the time.

CONCLUSIONS

Requiring physicians to respond to computer-generated reminders improved their compliance with preventive care protocols, especially for elderly patients for whom control physicians' compliance was the lowest. However, 100% compliance with cancer screening remainders will be unattainable due to incomplete data and patient refusal.

Injecting drug users and HIV/AIDS: risk behaviours and risk perception

This paper reports on the incidence of risk taking behaviours, and the relationship between risk perception and risk behaviours in a sample of 1245 Sydney injecting drug users (IDUs). Almost all respondents reported engaging in behaviours that placed them at risk of HIV infection: 32.9% through unsafe injecting, 84.4% because of unsafe sexual behaviour and 89.2% because of either injecting or sexual behaviour. Injecting and sexual behaviour were poorly correlated. This study also found that risk perception is unrelated to injecting or sexual behaviours, previous history of sexually transmitted diseases, a range of demographic characteristics including age and gender, and the number of times tested for HIV. Social policy and prevention programs should aim to change unsafe injecting and sexual behaviours directly, rather than attempting to achieve change indirectly by changing risk perception.

A comparison of drug use and HIV infection risk behavior between injecting drug users currently in treatment, previously in treatment, and never in treatment

We compared three groups of injecting drug users in a large cross-sectional study on HIV/AIDS and risk behaviors in Sydney, Australia, to determine what differences existed between those who had never been in treatment (n = 458), had previously been in treatment (n = 387), and were currently in treatment (n = 367). Drug use for those currently in treatment was assessed during the last typical using month before treatment. Those previously and currently in treatment were similar in terms of drug use patterns and HIV risk-related injecting behaviors. Those never in treatment were younger, had a lower level of HIV risk-related injecting behaviors, and reported lower drug use and less involvement in the drug subculture. There was little difference between the three groups on HIV risk-related sexual behaviors. These data suggest that those never in treatment are less dysfunctional and possibly less involved in drug-using careers, whereas there appears to be a close relationship between being previously and currently in treatment.

Estimating physician costliness. An empirical Bayes approach

Outpatient data obtained from the general medicine practice of an urban, health care facility are used to provide an application of empirical Bayes techniques in the estimation of physician "costliness." The results illustrate that application of the simplest empirical Bayes estimation procedure can provide more reasonable estimates of physician's utilization of resources than a standard estimation procedure. Empirical Bayes estimates are shown to adjust for potential instability in standard estimates that may arise from either a physician treating a small number of patients or an inappropriate case-mix adjustment. Using simulation, it is demonstrated that the empirical Bayes procedure can provide overall better estimates using fewer data than the standard procedure. This application, although somewhat limited in scope, should provide impetus for increased utilization of the numerous Bayesian and empirical Bayes techniques that currently exist in the statistical literature and pertain to small area estimation techniques.

Geographic variation in expenditures for physicians' services in the United States

BACKGROUND

The national volume-performance standard recently implemented by Medicare does not account for geographic variation in expenditures for physicians' services. To study this variation, we examined expenditures for physicians' services in all metropolitan areas in the United States.

METHODS

We used Medicare claims data for 1989 to measure rates of service use for beneficiaries living in the 317 U.S. metropolitan statistical areas (MSAs). The variables investigated were rates of admission to the hospital, payments to physicians for inpatient care per admission and per beneficiary, payments to physicians for outpatient care per beneficiary, and overall payments to physicians per beneficiary. Expenditures were measured in terms of allowed charges as adjusted to reflect prevailing charges in each MSA. Rates of use were adjusted for age and sex, with the exception of the variable for payments to physicians for inpatient care per admission, which was adjusted for case mix.

RESULTS

Expenditures for the delivery of physicians' services to Medicare beneficiaries varied markedly among MSAs, with those for the areas with the lowest and the highest rates differing at least twofold on each measure. The measures for specific areas varied in parallel: areas with high rates of admission tended to have high levels of payment to physicians for inpatient care per admission, and areas with high payments for inpatient services tended to have high payments for outpatient services. Expenditures were not related to the number of physicians per capita but were lower in MSAs with a high proportion of primary care practitioners. The variation persisted when the 25 largest MSAs were examined; for total payments to physicians per beneficiary, there was a twofold difference between the area with the lowest rate and that with the highest, San Francisco ($872) and Miami ($1,874). The states with the highest overall payments to physicians per beneficiary were Florida, Louisiana, and Michigan.

CONCLUSIONS

The marked variation among metropolitan areas in payments to physicians underscores the lack of consensus among physicians about which services are required. Moreover, the practice style in a given community appears to be influenced not by the aggregate supply of physicians but rather by the mixture of primary care physicians and specialists.

Hypnotic analgesia: dissociated experience or dissociated control?

High-hypnotizable subjects (n = 18) were superior to low-hypnotizable subjects (n = 18) in the extent of pain reduction produced by hypnotic analgesia and by a stress-inoculation procedure. However, stress inoculation but not hypnotic analgesia impaired performance on a cognitively demanding task that competed with pain reduction for cognitive resources. This outcome implies that hypnotic analgesia occurs with little or no cognitive effort to reduce pain and challenges the social psychological model of hypnosis. The findings are also inconsistent with the notion of dissociated experience, which proposes that pain and the cognitive efforts to reduce it are cut off from consciousness by an amnesialike barrier. However, the results do support the notion of dissociated control, which proposes that suggestions for hypnotic analgesia directly activate pain reduction and thereby avert the need for cognitive strategies to reduce pain.

Physician inpatient order writing on microcomputer workstations. Effects on resource utilization

OBJECTIVE

To assess the effects on health care resource utilization of a network of microcomputer workstations for writing all inpatient orders.

DESIGN

Randomized controlled clinical trial.

SETTING

Inpatient internal medicine service of an urban public hospital.

SUBJECTS

A total of 5219 internal medicine patients and the 68 teams of house officers, medical students, and faculty internists who cared for them.

INTERVENTION

Microcomputer workstations, linked to a comprehensive electronic medical record system, for writing all inpatient orders.

MAIN OUTCOME MEASURES

Total inpatient charges for each admission and charges for specific categories of orders. A time-motion study of selected interns assessed the ordering system's time consumption.

RESULTS

Intervention teams generated charges that were $887 (12.7%) lower per admission than did control teams (P = .02). Significant reductions (P < .05) were demonstrated separately for bed charges, diagnostic test charges, and drug charges. Reductions of similar proportion and statistical significance were found for hospital costs. The mean length of stay was 0.89 day shorter for intervention resident teams (P = .11). Interns in the intervention group spent an average of 33 minutes longer (5.5 minutes per patient) during a 10-hour observation period writing orders than did interns in the control group (P < .0001).

CONCLUSIONS

A network of microcomputer workstations for writing all inpatient orders significantly lowered patient charges and hospital costs. This would amount to savings of more than $3 million in charges annually for this hospital's medicine service and potentially tens of billions of dollars nationwide. However, the system required more physician time than did the paper charts. Research at other sites and system advances to reduce time requirements are warranted.

Platelet adhesion at low shear rate: study of a normal population

The use of oral contraceptives (OCs) has been associated with vascular complications. The mechanism(s) by which OCs predispose to thrombotic events remains unclear. Recent studies have demonstrated that postmenopausal (PM) women who take estrogen replacement therapy (ERT) have a decreased incidence of myocardial infarction compared to those who do not take ERT. This study was undertaken to determine if healthy individuals have differences in platelet adhesion depending on hormonal status. Men, PM women taking ERT, PM women not taking ERT, OC users, and premenopausal women not taking any medications were studied. Platelet studies were performed in a Hele-Shaw flow chamber at a low shear rate using platelet-rich plasma. The platelet adhesion process to subendothelial components: fibronectin, collagen I and collagen III was recorded using a 35 mm camera mounted on an inverted microscope. Photographs were taken at 30 second intervals for a total of 12 minutes and analyzed using a modified computer program which provided a numerical account of platelet adhesion. OC users had significantly higher platelet adherence to fibronectin, collagen I and collagen III compared to all other groups. All other study groups had similar platelet adhesion independent of hormonal status. These findings suggest that OCs cause increased platelet adhesion in some individuals and this may be one of the mechanisms by which OCs contribute to thrombotic events.

Medicaid case management: Kentucky's Patient Access and Care Program

Since 1981, States have been experimenting with Medicaid managed care programs to improve access and continuity of care and to contain costs by reducing inappropriate and unnecessary utilization. To determine the impact of primary care case management (PCCM) on utilization, the authors examine data from the Kentucky Patient Access and Care program (KenPAC). Using monthly utilization data from 1984 to 1989 and an interrupted time-series research design, the authors find that PCCM reduces the use of independent laboratory, physician, emergency department, and outpatient hospital services. PCCM does not appear to affect utilization of inpatient hospital services or prescription drugs.

How do Medicare physician fees compare with private payers?

Under the new fee schedule, Medicare physician fees are 76 percent of private fees. Consistent with the intent of payment reform, Medicare physician fees more closely approximate private fees for visits (93 percent) than for surgery (51 percent) and in rural areas as compared with large metropolitan areas. Variation in private fees across the country is considerably greater than it is for Medicare fees. Consequently, Medicare fees are most generous in areas that compare least favorably with the private market because private fees in these areas are well above average. These results shed light on the impact of the fee schedule and on the implications of using Medicare payment methods as part of a broad-based health reform.

Validation of probabilistic predictions

Current advances in high-speed computing and increased availability of statistical software have led to widespread use of statistical methods for the development of computerized protocols predictive of binary health outcomes. If these predictive algorithms are to be used in settings other than those for which they were developed, e.g., applied in a different geographic setting or extrapolated for use in a slightly different population, then they should be carefully validated to ensure appropriate application. Miller et al. (Stat Med. 1991) provided a comprehensive methodology for external validation of logistic prediction models, and applied these methods in a temporal validation setting. In this article, the authors emphasize how these methods can be applied to general forms of probabilistic predictions and provide several SAS macros for computation of the desired statistics.

Medicare inpatient physician charges: an econometric analysis

To control Medicare physician payments, Congress in 1989 established volume performance standards (VPS) that tie future physician fee increases to the growth in expenditures per beneficiary. The VPS risk pool is nationwide, and many observers believe it is too large to affect behavior. VPS could be modified by defining a separate risk pool for inpatient physician services and placing each hospital medical staff at risk for those services. Using a national random sample of 500,000 Medicare admissions, we explore the determinants of medical staff charges and comment on the policy implications. Multivariate analysis shows that charges increase with case mix and bed size but, surprisingly, decrease with the level of teaching activity. The teaching result is explained by the substitution of residents for physicians in these hospitals.

Physician charges in the hospital. Exploring episodes of care for controlling volume growth

Medicare physician payments are growing rapidly. At least 40% of the annual growth is due to volume increases. Reforms passed in 1989 include volume performance standards that attempt to control volume by linking future physician fee increases to volume growth. There is concern that defining the entire nation as the risk pool will result in an unworkable volume performance standard. One way to improve incentives is to create a separate volume performance standard for in-hospital physician services, define bundles of services related to the hospital stay, and place the medical staff of the hospital at risk for volume growth. To forestall the unbundling of services outside the stay, windows could be defined around the stay. This study reports physician services during the stay and in windows around the stay. In so doing, the study creates the knowledge base necessary to design better volume control policies and judge among alternative window definitions. Using 1987 data, this study presents average physician charges by type of service during: 1) the hospital stay; and 2) 1-month windows before and after the stay. For all admissions, 85% of charges occur during the stay and 15% occur during the windows (windows for surgical admissions and medical admissions are 9% and 23%, respectively). Pre- and postwindows are roughly symmetrical and average charges per day gradually increase before the admission and decline after discharge. A small physician panel commented on the clinical appropriateness of the one month windows. The panel indicates that defining in-hospital episodes of physician care is feasible.

Gigantic osteoma of the mandible: report of a case

A case of a very large solitary osteoma of the right posterior mandible in a 22-year-old man is presented. The tumor was asymptomatic despite its location and large size. It was removed via an extraoral Risdon approach without complication. The importance of differentiating a large solitary osteoma from a parosteal osteogenic sarcoma is emphasized. Any patient presenting with a solitary osteoma also should be evaluated for Gardner's syndrome.