The growth of health-related meta-analyses published from 1980 to 2000.

A MEDLINE search was conducted to estimate trends in the growth of health-related meta-analyses published during the past two decades. Employing a more specific than sensitive search strategy, and not supplementing these results with known sources of published meta-analyses or manual search strategies, 3,025 probable meta-analyses were selected from the 5,128 citations identified. The data showed a definitive upward (and generally linear) trend across time with no evidence for this genre of research either leveling off or decreasing.

Demographic and health-related correlates to visits to complementary and alternative medical providers.

OBJECTIVES: The objective of this study was to ascertain the extent to which demographic and health-related variables are related to visits to a complementary or alternative medicine (CAM) practitioner. METHODS: This study reports a secondary analysis of visits to CAM practitioners during the year before the 1996 Medical Expenditure Panel Survey (n = 16,038). RESULTS: Overall visits to CAM providers (9%) were lower than reported in widely cited surveys but quite consistent with a previous Robert Wood Johnson study that used a similar sampling procedure. Gender, education, age, geographic location, and race (Hispanics and African Americans proved to be less likely to visit CAM providers than whites) were statistically significant predictors of visits to CAM providers. Individuals in poorer health and those suffering from mental, musculoskeletal, and metabolic disorders also tended to be more likely to have visited a CAM provider. CONCLUSIONS: Although the choice of alternative versus orthodox treatment appears to be a complex phenomenon, these data suggest that the heaviest users of CAM therapies tend to be individuals with comorbid, non-life-threatening health problems. This finding may help to ameliorate concerns that this type of care is being used in lieu of therapies with more definitive efficacy evidence.

Measures of disease control in Medicare beneficiaries with diabetes mellitus.

OBJECTIVE: To investigate measures of disease control for Medicare beneficiaries with diabetes and their outpatient care in the fee-for-service setting. DESIGN: Retrospective cohort study. SETTINGS: Office practices in Alabama, Iowa, and Maryland of 293 primary care physicians (PCPs) who volunteered to participate in the Ambulatory Care Quality Improvement Project. PARTICIPANTS: A total of 5980 patients with an average age of 75.2 years. MEASUREMENTS: For an 18-month period (1/1/94-6/30/95), medical records were abstracted for clinical parameters, including up to four blood glucose values; two blood pressure measurements; one total cholesterol value; two serum creatinine values; medication use, including antihypertensives, angiotensin-converting enzyme (ACE) inhibitors, and lipid-lowering agents; and frequency of glycosylated hemoglobin (GHb) determinations. RESULTS: During the study, 44% of patients received at least one GHb determination, 94% received at least one blood glucose, 68% at least one total serum cholesterol, 74% at least one serum creatinine test, and 97% at least one blood pressure measurement. Ten percent of patients had mean blood glucose levels > or = 250 mg/dL. Eighty-five percent had evidence of hypertension. Of this group of hypertensive patients with blood pressure readings available, 70% had blood pressure readings > or = 140/90 mm Hg, even though there were on medication that could have been prescribed for hypertension. Thirty-six percent of those who had evidence of hypertension were taking an ACE inhibitor. Thirty-two percent of those taking lipid-lowering medication had a total serum cholesterol value > or = 240 mg/dL. Statistically significant differences were noted for age and gender, with men and patients more than 85 years old generally having better measures of disease control. CONCLUSIONS: Many older Medicare patients with diabetes did not achieve recommended target levels of blood glucose, blood pressure, and lipids. GHb and serum cholesterol are not being monitored at recommended intervals. Significant opportunities exist to improve diabetes care for this population.

An intervention for changing high-risk HIV behaviors of African American drug-dependent women.

The purpose of this study was to test the effectiveness of an AIDS education intervention for methadone-dependent, African American women. The women were randomly assigned to experimental (n=107) or control (n=97) group. The experimental group participated in a peer counseling and leadership training program conducted by two experienced nurse counselors over an 8-week period, followed by 8 weeks of reinforcement. The program was designed to reduce AIDS high-risk sexual behavior, increase self-esteem, decrease depressive affect, and increase the women's community-based AIDS prevention communication activities. A total of 130 women completed all phases of the study, including longitudinal Posttests at 2, 4, and 7 months after enrollment. Compared to the control group, there were statistically significant differences in three of the outcomes for the experimental group: The experimental group reported an increased number of safer sexual behaviors (p=.029), showed decreases in depression (p=.001), and reported engaging in more AIDS-related, community-based communication activities regarding prevention (p=.005).

The empowering potential of quality improvement data.

Data are being used to redefine, transform, and empower the roles of providers, payers, and patients. The empowering potential of data on these three groups, and their changing roles are reviewed. The use of data to transform care and increase equality is not new, yet the comprehensive application of principles of continuous quality improvement to the delivery processes is just beginning. The approach to the quality improvement processes employed by peer review organizations includes idea capture, development of intuitively appealing quality indicators, formulating the intervention, formalizing the research architecture, and quantifying the impact of the interventions. The multifaceted interventions for improvement are discussed. The future of medicine includes a greater use of data and the quantitative sciences to inspire improvement across the health care delivery continuum.

AIDS prevention in high-risk African American women: behavioral, psychological, and gender issues.

A three-year longitudinal intervention study was implemented to reduce high-risk drug and sexual behaviors in methadone-dependent African American women. Participants were recruited from four inner-city methadone maintenance programs and randomly assigned either to an eight-week peer counseling and leadership training group or to a control group. The 107 trainees and 97 controls completed pretests and posttests at two, four, and seven months. This paper focuses on final data related to the subjects' sexual beliefs, attitudes, knowledge, and behaviors that put them at risk for HIV/AIDS. Reasons for not using condoms are categorized and discussed. Despite the women's awareness of the seriousness of AIDS, perceived powerlessness to negotiate condom use, negative attitudes about the use of condoms, influence of drugs, and unavailablility of condoms interfered with safer sex practices. The inability of education alone to prevent many high-risk sexual behaviors suggests that more serious consideration be given to expanded distribution of condoms as well as needle exchange programs and legalization of illicit drugs.

PIP: The US Centers for Disease Control report that although African-American women comprise only 12% of the US female population, they account for 56.9% of AIDS cases in women. In 1994, 41% of women with AIDS reported IV-drug use and 38% reported heterosexual contact with a partner at risk for or known to have HIV infection or AIDS. Findings are presented from a 3-year study conducted to assess and reduce high-risk drug and sex behavior among methadone-dependent African-American women in Baltimore, Maryland. Through random assignment, 107 women received 8 weeks of peer counseling and leadership training, and 97 women formed the control group. The women were 20-59 years old of mean age 35.5. 84-90% were unemployed and 16% had been homeless within the preceding 6 months. Subjects completed pretests and post-tests at 2, 4, and 7 months. Despite the women's awareness of the seriousness of AIDS, their perceived powerlessness to negotiate condom use, negative attitudes about condom use, the influence of drugs, and the unavailability of condoms interfered with the practice of safer sex behavior. The authors believe that the inability of education alone to prevent many high-risk sexual behaviors is cause to give more serious consideration to expanding the distribution of condoms as well as needle exchange programs and the legalization of currently illicit drugs.

The growth of meta-analytic literature from 1980 to 1993.

A search of the literature was undertaken to estimate the number of meta-analytic studies that have been published in the social and health sciences to date. Altogether, 1,874 published quantitative reviews were located by the end of 1994: 892 in the health sciences, 982 in the social sciences. Trends from 1980 to 1993 indicate that, while the social sciences embraced the technique earlier, health researchers are now publishing almost twice as many meta-analyses as their social science colleagues.

Realizing the meta-analytic potential--a survey of experts.

A survey of 99 authors of meta-analyses conducted between 1988 and 1993 was undertaken to identify policies and products that might be helpful in the realization of the procedure's ultimate potential. Although a number of differences were found between medical and social/nonmedically related health scientists, the greatest degree of overall enthusiasm for both groups was observed for the need to educate journal editors and primary researchers regarding information that needs to be reported in an empirical study. The respondents were also in general agreement that a consensus should be developed and disseminated on minimum standards for published meta-analyses. A number of the other proposals were less popular, although an argument is made that some of these have actually come to fruition since the original survey was conducted.

Preventive behaviour: attitudes and compliance of nursing students.

As a professional, the nurse is called upon to serve as a role model for positive health behaviours. Because the practise of these behaviours may effect one's effectiveness as a role model, it is important to determine whether nurses lead a so-called preventive lifestyle and to identify what factors are predictive of compliance. Previous researchers have tended to examine individual behaviours. In this study, a prevention index was used to examine an overall preventive lifestyle. For a sample of senior year nursing students, the preventive behaviours were considered to be important. However, when compared to a national sample of females in the same age range, nursing students were found to be significantly less compliant for 12 of 19 behaviours and more compliant for only three behaviours. Both the desire to practise preventive behaviours and the perceived difficulty in doing so were the factors predictive of the level of compliance. Additional study is needed to determine how these factors can be mediated.

Alcohol use and its relationship to other addictive and preventive behaviors.

Immoderate consumption of alcohol was found to be related to three other potentially addictive behaviors (illicit drug use, smoking, and caffeine consumption) in a randomly drawn sample (n = 1253) of American adults. In addition, alcohol consumption was found to be related to nutritional, life style, safety, and health monitoring preventive behaviors. Although the underlying mechanism for these behaviors is not clear, possibilities include the subsumption of both addictive and other preventive behaviors under a more generalized risk taking (or risk avoidance) dimension.

Public perceptions regarding the AIDS epidemic: selected results from a national poll.

A telephone interview of 1256 adults age 18 and over was conducted using a random digit dialing procedure. Participants were queried about their perceptions of being at risk for contracting the acquired immune deficiency syndrome (AIDS), optimism-pessimism about the controllability of the epidemic, whether they take special steps to avoid catching AIDS, and how they would prioritize three possible courses of action. The sample was about evenly divided between optimism and pessimism about controlling AIDS within five years; 10% perceived they were at least at some risk of contracting AIDS (with such "at risk" persons overrepresented among residents in the east, among college graduates, and within the 30-39 age bracket). Forty-one percent indicated taking special steps to avoid AIDS, with higher percentages of Blacks, Hispanics, and persons under age 30 reporting such precautions. Personal preventive action was ranked first among the three possible courses of action by 51%, unlimited governmental spending to find a cure or vaccine and governmental restrictions of certain homosexual behaviors were each ranked first by about 20% of the respondents. The results also indicated that perception of being at risk mediates some opinions about AIDS.

Perceived quality of hospital care.

Physicians, nurses, and consumers in the New York City area were asked to indicate which of 76 acute-care hospitals they would be willing to be patients in if they became seriously ill. Percentages of favorable judgments ranged from a high of 95% for one university hospital to a low of less than 10% for a community hospital. Over half of all responding physicians, nurses, and consumers indicated that they would not wish to be a patient in the majority of the 76 hospitals sampled. Although a high degree of consistency (r = .93) was observed among all three types of respondents, physicians were on the average significantly more pessimistic (X percentage of favorable judgments = 34) than nurses and consumers. All three groups indicated a statistically significant (p less than .001) preference for larger hospitals as well as institutions that were affiliated with a medical school in some way.