Effect of TDF-containing regimens on creatinine clearance in HIV patients in Namibia with a baseline CrCl <60ml/min; findings and implications

Background and aims: The advent of antiretroviral therapy (ART) and early diagnosis of the human immunodeficiency virus (HIV) has resulted in an appreciable reduction in morbidity and mortality among people infected with HIV. However, tenofovir disoproxil fumarate (TDF)-containing ART regimens are associated with a reduction in creatinine clearance (CrCl). No evaluation has been conducted in Namibia to date on the relationship between TDF-containing ART and CrCl among patients with moderate to severe reductions in CrCl to guide future practice. We aimed to address this.Methodology: Retrospective longitudinal study between January 2008 to December 2016 evaluating CrCl in patients with a baseline CrCl <60ml/min who were receiving TDF-containing ART in a leading hospital in Namibia. We identified patients who had experienced an improvement in CrCl and compared their characteristics with those whose CrCl did not improve. We assessed factors for an association with improvement in CrCl using binary logistic regression.Results: 389 patients were included, the majority were female (n = 294). Female vs. male assessments showed no difference in age (p = 0.340), weight (p = 0.920), number who experienced an improvement (105 vs 39, p = 0.349), or absence of improvement (189 vs. 56, p = 0.349). The improvement group (male and female) had a lower baseline CrCl (45.9 vs. 55.0, p < 0.001). The follow-up CrCl for the improvement and no improvement groups were 72.6 and 55.9 respectively. Multivariate analysis showed that the odds of improvement were: 0.905 (0.871-0.940, p < 0.001) for each unit rise in the baseline CrCl, and 0.904(0.880-0.923) for each year of follow-up.Conclusion: More improvement than decline in CrCl was observed. Improvement occurred more in patients with lower baseline CrCl, and occurred in the early period of ART with reduced odds of experiencing this with time. Our findings indicate that TDF may be used in patients with a low baseline CrCl.

Weekly miscarriage rates in a community-based prospective cohort study in rural western Kenya

OBJECTIVE

Information on adverse pregnancy outcomes is important to monitor the impact of public health interventions. Miscarriage is a challenging end point to ascertain and there is scarce information on its rate in low-income countries. The objective was to estimate the background rate and cumulative probability of miscarriage in rural western Kenya.

DESIGN

This was a population-based prospective cohort.

PARTICIPANTS AND SETTING

Women of childbearing age were followed prospectively to identify pregnancies and ascertain their outcomes in Siaya County, western Kenya. The cohort study was carried out in 33 adjacent villages under health and demographic surveillance.

OUTCOME MEASURE

Miscarriage.

RESULTS

Between 2011 and 2013, among 5536 women of childbearing age, 1453 pregnancies were detected and 1134 were included in the analysis. The cumulative probability was 18.9%. The weekly miscarriage rate declined steadily with increasing gestation until approximately 20 weeks. Known risk factors for miscarriage such as maternal age, gravidity, occupation, household wealth and HIV infection were confirmed.

CONCLUSIONS

This is the first report of weekly miscarriage rates in a rural African setting in the context of high HIV and malaria prevalence. Future studies should consider the involvement of community health workers to identify the pregnancy cohort of early gestation for better data on the actual number of pregnancies and the assessment of miscarriage.

Frequency and characteristics of cognitive services provided in response to a financial incentive

OBJECTIVE

To determine the effects of a financial incentive on the number and types of cognitive services (CS) provided by community pharmacies to Medicaid recipients in the State of Washington.

DESIGN

Prospective randomized trial. CS were reported using a problem-intervention-result coding system over a 20-month period.

SETTING AND SUBJECTS

Pharmacists practicing in 110 study (financial incentive) and 90 control community pharmacies.

RESULTS

Study pharmacists documented an average of 1.59 CS interventions per 100 prescriptions over a 20-month period, significantly more than controls, who documented an average of 0.67 interventions (P < .05) per 100 prescriptions. One-half (48.4%) of all CS were for patient-related problems, 32.6% were for drug-related problems, 17.6% were for prescription-related problems, and 1.4% were for other problems that did not involve drug therapy. A change in drug therapy occurred as a result of 28% of all CS documented in this demonstration. Changes were rarely (2.4%) due to generic or therapeutic substitution and almost always (90%) followed communication with the prescriber. The average self-reported time to perform CS was 7.5 minutes; 75% of interventions were < or = 6 minutes. Considerable differences existed between study and control groups in the types of problems identified, intervention activities performed, and results of interventions.

CONCLUSION

A financial incentive was associated with significantly more, and different types of, CS performed by pharmacists.

Antibiotics and Clostridium difficile diarrhea in the ambulatory care setting

OBJECTIVE

The goal of this study was to determine the prevalence of Clostridium difficile diarrhea (CDD) and the risk for CDD associated with different oral antibiotics commonly used in the ambulatory care setting.

METHODS

The prevalence of CDD was determined for enrollees in 4 UnitedHealth Group-affiliated health plans between January 1, 1992, and December 31, 1994. Cases were identified based on the presence of an inpatient or outpatient claim with a primary diagnosis of diarrhea, a pharmacy claim for a prescription drug used to treat CDD, or a physician or facility claim for the C. difficile toxin test, and were confirmed using full-text medical records. Within a retrospective cohort design, periods of risk for CDD were defined on the basis of duration of antibiotic therapy. To control for potential selection bias created by heterogeneous rates of C. difficile testing and to limit confounding due to multiple antibiotic exposures, we used a nested case-control design, restricting eligibility to subjects who underwent screening for C. difficile and who had been exposed to only 1 antibiotic risk period with a single antibiotic.

RESULTS

The global prevalence of CDD in 358,389 ambulatory care enrollees was 12 per 100,000 person-years. In the nested case-control study, after controlling for other risk factors, 2 antibiotics demonstrated an increased association with CDD: cephalexin (odds ratio [OR] = 7.5, 95% CI = 1.8 to 34.7) and cefixime (OR = 6.4, 95% CI = 1.2 to 39.0).

CONCLUSIONS

Although CDD is thought to occur primarily in hospitalized patients, it was found to be present in an ambulatory care population, but at a low frequency. In this population, it appeared to be associated with 2 cephalosporins but not with other types of antibiotics usually linked with nosocomial CDD. Because the frequency of C. difficile testing was shown to be more common with high-risk antibiotics, CDD may be underdiagnosed in the ambulatory care setting.

Influence of a financial incentive on cognitive services: CARE project design/implementation

OBJECTIVE

To describe the design and methods of the Washington State Cognitive Activities and Reimbursement Effectiveness (CARE) Project, a demonstration project in which community pharmacies were paid for cognitive services (CS) provided to Medicaid patients, its evaluation objectives, and the extent to which implementation objectives were achieved.

DESIGN

Prospective randomized trial. Community pharmacies were allocated to a documentation-and-payment group, documentation-only group, and "silent" control group. CS were reported using a problem-intervention-result classification system embedded within a pseudo-National Drug Code format. Management strategies included use of area coordinators.

SETTING

Pharmacies serving ambulatory Medicaid patients in the state of Washington, excluding staff-model health maintenance organization pharmacies and pharmacies predominantly serving long-term-care residents.

PARTICIPANTS

200 community pharmacies (110 treatment; 90 control), with another 100 randomly selected pharmacies as a silent control group.

INTERVENTIONS

A modest monthly stipend. The treatment group billed Medicaid for each documented CS associated with a drug therapy-related problem. All participants received training in documentation methods. A unique coding scheme allowed documentation of CS within the constraints of the Medicaid program. Data edit checks and feedback were used to ensure data quality and completeness. Area coordinators were used to facilitate training, compliance with study procedures, and participation.

MAIN OUTCOME MEASURES

Participation rates, documentation rates, coding scheme revision, data quality and completeness rates, and effectiveness of area coordinators.

RESULTS

Pharmacists documented more than 20,240 CS records. Approximately 89% of records passed edit checks, and 94% did so after modification. Nearly 83% could be linked to a paid drug or CS claim. The coding system was sufficient, with minor modifications, to account for all interventions documented. Area coordinators did not function as expected.

CONCLUSION

A system for documentation and payment of pharmacists' CS to Medicaid recipients was implemented successfully and relatively easily in community pharmacies.

Patients' perceived benefit from and satisfaction with asthma-related pharmacy services

OBJECTIVE

To determine whether patients targeted to receive intervention from an asthma management program reported receiving more services and had greater perceived benefit and satisfaction with those services compared with asthma patients not targeted by the program.

DESIGN

Mailed survey.

SETTING

Community pharmacy.

PATIENTS

471 community-based patients receiving asthma medications from 44 intervention pharmacies and 1,164 patients from 46 usual care (control) pharmacies.

MAIN OUTCOME MEASURES

Five-point agreement scale measuring asthma services received, perceived value of the services, and satisfaction.

RESULTS

Usable surveys were received from 39.0% of intervention patients and 42.4% of controls. There were no statistically significant differences between groups in the frequency of provision of listed services. Approximately 60% of respondents from both groups received written materials on asthma medications and 54% received inhaler counseling; both were rated high for perceived benefit. Fewer than 20% reported being counseled about asthma triggers. Fewer than 5% reported pharmacists talking to physicians on their behalf. General satisfaction with pharmacy services was high (78.2% agree or strongly agree), but not statistically different between groups. More than 65% believed that pharmacists spend enough time counseling patients. Several comments indicated that patients did not expect or ask for information because they were unaware that services were available and/or they had already been counseled by their physician. Responses to the statement "my asthma is better controlled because of help given to me by the pharmacist" were equivocal and not different between groups.

CONCLUSION

Overall, there were few differences between groups. General satisfaction with pharmacy services is high, but patients' perceived benefit and satisfaction with cognitive services is lower. Increased public awareness of pharmacists' capabilities and a more proactive approach to providing cognitive services is needed.

Comparison of self-report data and medical records data: results from a case-control study on prostate cancer

BACKGROUND

Self-report and review of medical records are the most common methods for the assessment of past exposures. However, information obtained from self-reports and medical records may not be consistent. This study compared information provided in a self-administered questionnaire with medical records data.

METHODS

Self-report and medical records data came from a case-control study on prostate cancer. Cases were 181 patients with primary prostate cancer and controls were 297 men without the disease, enrolled in Group Health Cooperative (GHC) in Seattle. The consistencies between the two data sources were examined.

RESULTS

In general, agreement between the two data sources was almost perfect for demographic and anthropometric variables, substantial for the history of inguinal hernia and kidney stones, and moderate for vasectomy, family history of prostate cancer, smoking and alcohol consumption. However, the two data sources generally were poorly concordant for prior genitourinary diseases that have less explicit diagnostic criteria such as benign prostatic hyperplasia and prostatitis. Analyses of discordant data showed that men were more likely to report an exposure or medical condition that could not be verified from medical records. No discernible patterns in the difference of agreement were found according to age, GHC membership length or case-control status.

CONCLUSIONS

This study suggests that agreement between self-reported data and medical records data varies depending upon the study variables. While both data sources are subject to some problems, self-report may provide more complete and comparable information, at least for variables unrelated to diagnosis.

Roles for pharmacists in the prevention and control of sexually transmitted diseases

As the most accessible and trusted health care professionals in the United States, pharmacists can be an important resource for sexually transmitted disease (STD) prevention and control. Pharmacists and pharmacies are located in every type of community throughout the United States. The profession of pharmacy is positioned well to make a meaningful contribution to STD prevention and control. The accessibility to the public, the large number of pharmacy locations, and the trust shared between pharmacists and the public they serve combine to afford a unique opportunity to reach millions of individuals with STD prevention and control messages and other strategies. The numerous innovative activities involving pharmacists and pharmacies in STD-related services suggest an expanded role for pharmacy.

Hypertension, antihypertensive medication use, and risk of renal cell carcinoma

To investigate whether diuretic medication use increases risk of renal cell carcinoma (RCC), the authors conducted a case-control study of health maintenance organization members in western Washington State. Cases (n = 238) diagnosed between January 1980 and June 1995 were compared with controls (n = 616) selected from health maintenance organization membership files. The computerized health maintenance organization pharmacy database provided information on medications prescribed after March 1977. Additional exposure information was collected from medical records. For women, use of diuretics was associated with increased risk of RCC (odds ratio (OR) = 1.8, 95% confidence interval (CI) 1.0-3.1), but the association was not independent of a diagnosis of hypertension (adjusted for hypertension, OR = 1.1, 95% CI 0.5-2.1). Similarly, nondiuretic antihypertensive use was associated with increased risk, but only when unadjusted for hypertension. For men, neither diuretic nor nondiuretic antihypertensive use was associated with risk of RCC. A diagnosis of hypertension was clearly associated with RCC risk for women (OR = 2.5, 95% CI 1.2-5.1), but not men (OR = 1.3, 95% CI 0.7-2.5). High systolic and diastolic blood pressures were associated with increased risk in both sexes. These results do not support the hypothesis that use of diuretic medication increases RCC risk; they are more consistent with an association between RCC and high blood pressure.

Prostate cancer in relation to the use of electric blanket or heated water bed

Using data from a case-control study conducted in Group Health Cooperative (GHC) of Puget Sound, we examined the relation between the use of electric blankets or heated water beds and the risk of prostate cancer. Cases were 175 prostate cancer patients ages 40-69 years. Controls were 258 male GHC members frequency matched to cases. The odds ratio (OR) for prostate cancer associated with the use of an electric blanket or heated water bed was 1.4 (95% confidence interval (CI) 0.9-2.2). The risk, however, did not tend to be higher with increasing months per year or years of use. This study did not provide clear evidence on the hypothesized association.

A model for improving medication use in home health care patients

OBJECTIVES

(1) To develop a model for the identification and resolution of problems associated with suboptimal medication use in elderly patients receiving home health care; (2) To select the most important identifiable problems and develop structured procedures for their resolution.

DESIGN

Expert panel review, problem selection, and development of a problem resolution model and guidelines.

SETTING

Home health care.

PARTICIPANTS

A panel with expertise in home health nursing, pharmacy, clinical pharmacology, gerontology, pharmacoepidemiology, and health services research.

INTERVENTIONS

A list of potential problems associated with the most frequently used classes of drugs was compiled for review by the panel. Problems that were controversial or that could not be identified in the home care setting were excluded. Panel members individually ranked the remainder. Detailed procedures for identification and resolution of the 15 top-ranking problems were developed.

MAIN OUTCOME MEASURES

Not applicable.

RESULTS

Potential medication problems were defined by both drug use and symptoms or clinical signs associated with specific adverse effects, to ensure that clinically relevant problems would be identified. The model developed for problem assessment and resolution was centered on the drug utilization review (DUR) coordinator and the attending home health nurse. Following guidelines developed by the panel, the DUR coordinator advises the home health nurse about identified problems and how to resolve them. One of these practitioners, usually the nurse, then contacts the attending physician to explain their concerns, offer potential solutions, and request instructions.

CONCLUSION

A potentially useful model for the identification and resolution of medication problems in the home health care setting was developed. This model is currently being evaluated in a randomized controlled trial.

Principal findings from the Washington State cognitive services demonstration project

In addition to dispensing, pharmacists are ideally positioned to provide cognitive services that are targeted at optimizing drug therapy through identification and resolution of drug therapy problems. The Washington Cognitive Activities and Reimbursement Effectiveness project sought to determine: (1) if pharmacists would respond to a financial incentive by performing more cognitive services, and (2) the effect on drug cost of cognitive services they performed.

Risk factors for incident and recurrent condylomata acuminata among men. A population-based study

BACKGROUND AND OBJECTIVES

The rapid increase in the number of physician office visits for condylomata acuminata and the association of human papillomavirus and cancer has prompted renewed interest in the epidemiology of this sexually-transmitted disease. Few epidemiologic studies have examined what risk factors are associated with condylomata acuminata in men.

GOAL

To determine what factors may predispose a man to the occurrence of condylomata acuminata.

STUDY DESIGN

A population-based case-control study was conducted among male members of a health maintenance organization. Patients were men 18 years or older who were seen for condyloma at one of four primary care clinics of Group Health Cooperative of Puget Sound between April 1, 1987 and September 30, 1991. Control subjects were frequency matched to the patients on clinic site, race, and age. In-person interviews were used to ascertain exposure histories from both patients and control subjects.

RESULTS

Recurrent condyloma was reported by about one third of our patients. Patients with multiple partners were strongly associated with developing the disease. Several factors were either more strongly or only associated with recurrent disease. Other behavioral measures, such as recreational drug use, were also related the occurrence of condyloma.

CONCLUSION

These results confirm the sexual-transmitted mechanism of condyloma in men. Exposure to multiple partners was associated with elevated risk of both recurrent and incident disease. Other cofactors may be involved in the etiology of condyloma.

Risk factors for incident and recurrent condylomata acuminata among women. A population-based study

BACKGROUND

Condylomata acuminata is one of the most common sexually transmitted diseases (STDs) diagnosed in the United States, yet relatively little research has been conducted on the determinants of this disease in well-defined populations.

GOAL

To determine the exposures that predispose a woman to the development of condylomata acuminata or genital warts.

STUDY DESIGN

A population-based case-control study was conducted among enrollees of Group Health Cooperative of Puget Sound. Patients (94 women with incident and 55 women with recurrent condyloma) were diagnosed between April 1, 1987 and September 30, 1991. Control subjects were 133 women without a history of genital warts. An in-person interview was conducted to collect information on subject characteristics, exposures, and on all episodes of genital warts.

RESULTS

Women with five or more partners within the 5 years before reference date were over seven times more likely to have incident condyloma (relative risk [RR], 7.5; 95% confidence interval [CI], 3.1-18.1) and over 12 times more likely to have recurrent condyloma (RR, 12.8; 95% CI, 4.2-38.9) compared with women with only one sexual partner during this time period. An increased risk of incident condyloma was also associated with a history of any STD (RR, 2.6; 95% CI, 1.1-5.8), a history of oral herpes (RR, 2.2; 95% CI, 1.1-4.4), and a history of allergies (RR, 2.0 95% CI, 1.0-3.8). Our data did not support a strong association between risk of condyloma and smoking or recent use of oral contraceptives.

CONCLUSION

Our results suggest that risk of condyloma is primarily related to sexual behavior. We did not observe a strong association between risk of condyloma and many of the exposures considered to be potential cofactors for anogenital cancers associated with other types of human papillomaviruses.

A randomized controlled trial of a drug use review intervention for sedative hypnotic medications

OBJECTIVES

Drug use review is used by both the public and private sector to influence prescribing behavior and patient drug use. Past interventions mailed to prescribers have had mixed results. The objective was to evaluate the effect of a one-time, mailed intervention on subsequent use of sedative hypnotic medication.

METHODS

An experimental design was used. The intervention contained guidelines for the use of sedative hypnotics, a prescriber profile detailing sedative hypnotic prescribing, and a patient profile. Clustering of patients and their shared prescribers was done to avoid contamination bias and statistical problems associated with a lack of independence of observations. Subjects were 189 Washington State Medicaid recipients who had received at least one tablet per day of a sedative hypnotic medication for 1 year and their prescribing physicians or (when information about the physician was lacking) the dispensing pharmacy.

RESULTS

A significant reduction in the use of targeted sedative hypnotic medications was measured in the intervention group (-27.6%) versus the control group (-8.5%). In the intervention group, 9.4% of patients began a new prescription for a benzodiazepine not targeted by the drug use review, whereas no control patients had new use of nontarget benzodiazepines.

CONCLUSIONS

The intervention achieved a statistically significant decrease in targeted drug use, and the amount of reduction is likely to have decreased the risk of fractures associated with benzodiazepine use. This study adds to the recent evidence that mailed drug use review interventions can have a desirable impact on patient drug use.

Vaginal douching as a risk factor for cervical Chlamydia trachomatis infection

OBJECTIVE

To evaluate the association between vaginal douching and cervical Chlamydia trachomatis infection.

METHODS

We analyzed cross-sectional data from a study conducted at Group Health Cooperative of Puget Sound, a nonprofit health maintenance organization in western Washington state. Participants were nonpregnant women Group Health enrollees between the ages of 18 and 34 years who were attending two primary care clinics either for nonurgent visits, primarily routine preventive health visits, or in response to an invitation from the study. Before the clinical examination, all completed a self-administered survey assessing demographic and behavioral characteristics, including the timing, frequency, products used, and reasons for douching. Chlamydial infection was ascertained via cell culture isolation of C trachomatis from endocervical specimens obtained at the same visit.

RESULTS

Chlamydia trachomatis was isolated from cervical cultures in 58 (3.4%) of 1692 study participants. Women who reported douching in the 12 months before their clinic visit had an increased likelihood of chlamydial infection compared with women who did not douche (prevalence odds ratio [OR] 2.29, 95% confidence interval [CI] 1.22, 4.30, after adjusting for confounding factors). The likelihood was higher for women who reported douching more often: OR 2.60 (95% CI 1.29, 5.24) for women who douched one to three times per month, and OR 3.84 (95% CI 1.26, 11.70) for those douching four times or more per month. These associations were slightly stronger when women who reported douching because of an infection were excluded from the analysis.

CONCLUSION

These results support the hypothesis that vaginal douching predisposes to acquisition of cervical chlamydial infection and are compatible with previous studies that report associations between douching and sequelae of chlamydial infection, including pelvic inflammatory disease, ectopic pregnancy, and infertility.

The role of cost-consequence analysis in healthcare decision-making

A greater understanding of value associated with new pharmaceutical products should lead to better decision-making. Most commonly cost-effectiveness ratios (CERs) are used to indicate value; however, researchers have recently shown that CER estimates are rarely used by decision-makers in making formulary decisions. In this article, a cost-consequence approach to estimating the value for money of a new treatment for a specific disease is described. Using a cost-consequence approach, the impact of the new treatment on lifetime resource use and costs (including specific healthcare service use and costs, and productivity losses) and health outcomes (including disease symptoms, life expectancy and quality of life) for an individual or group of individuals is estimated and presented in a tabular format. The cost-consequence format is more likely to be approachable, readily understandable and applied by healthcare decision-makers than a simple CER. The decision-maker may use selected items from the cost-consequence analysis to compute composite measures of drug value, such as cost per life-year gained or cost per quality-adjusted life-year (QALY) gained. In general, the cost-consequence approach, by making the impact of the new treatment as comprehensive and transparent as possible, will enable decision-makers to select the components most relevant to their perspective and will also give them confidence that the data are credible to use as the basis for resource allocation decisions.

Antecedent antimicrobial use increases the risk of uncomplicated cystitis in young women

To examine whether antecedent antimicrobial use influenced subsequent relative risk of urinary tract infection (UTI) in premenopausal women, data were analyzed from two cohorts of women observed prospectively for 6 months to determine risk factors for UTI. Using a Cox proportional hazards model to adjust for covariates, we found that 326 women in a University cohort and 425 women in a health-maintenance organization cohort were at increased risks for UTI (2.57 [95% confidence interval (CI), 1.24-5.32] and 5.83 [95% CI, 3.17-10.70], respectively) if antimicrobials had been taken during the previous 15-28 days but not during the previous 3, 7, or 14 days. The increased risks were noted both for women whose antimicrobial use was for treatment of a previous UTI and for women who received antimicrobials for other illnesses. These results suggest that recent antimicrobial use increases a woman's risk of UTI, perhaps by altering the indigenous urogenital flora and predisposing to vaginal colonization with uropathogens.

Vasectomy and prostate cancer: a case-control study in a health maintenance organization

A case-control study was conducted within Group Health Cooperative of Puget Sound to evaluate the relation between prior vasectomy and the risk of prostate cancer. Cases consisted of 175 men newly diagnosed with histologically confirmed prostate cancer during 1989-1991. A total of 258 controls, matched to cases on birth year and membership status, were randomly selected from the general membership of the plan. Information was collected from mailed questionnaires and medical records on medical history, including prior vasectomy, anthropometric measures, family history of prostate cancer, personal habits, and medical care utilization, and demographic factors. Conditional logistic regression analyses showed that the odds ratio for prostate cancer associated with vasectomy was 0.86 (95% confidence interval 0.57-1.32) after adjustment for confounders. The odds ratio estimate did not differ substantially by age at vasectomy or time since vasectomy. However, the odds ratio estimate for prostate cancer associated with vasectomy tended to be increased among men who had a father or brother with prostate cancer. Nevertheless, the increased risk may be related to detection bias or differential participation rates due to both vasectomy status and a family history of prostate cancer. These results suggest no overall association between vasectomy and prostate cancer.

Association between use of spermicide-coated condoms and Escherichia coli urinary tract infection in young women

Diaphragm/spermicide use increases the risk of urinary tract infection (UTI). To determine whether spermicide-coated condoms are also associated with an increased risk of UTI, the authors conducted a case-control study at a large health maintenance organization in Seattle, Washington. Cases were sexually active young women with acute UTI caused by Escherichia coli, identified from computerized laboratory files during 1990-1993. Age-matched controls were randomly selected from the enrollment files of the plan. Of 1,904 eligible women, 604 cases and 629 controls (65%) were interviewed. During the previous year, 40% of the cases and 31% of the controls had been exposed to any type of condom. The unadjusted odds ratio for UTI increased with frequency of condom exposure from 0.91 (95% confidence interval (CI) 0.65-1.28) for weekly or less during the previous month to 2.11 (95% CI 1.37-3.26) for more than once weekly. Exposure to spermicide-coated condoms conferred a higher risk of UTI, with odds ratios ranging from 1.09 (95% CI 0.58-2.05) for use weekly or less to 3.05 (95% CI 1.47-6.35) for use more than once weekly. In multivariate analyses, intercourse frequency (odds ratio (OR) = 1.14 per weekly episode), history of UTI (OR = 2.64), and frequency of spermicide-coated condom exposure (OR = 3.34 for more than once weekly and 5.65 for use more than twice weekly) were independent predictors of UTI. Spermicide-coated condoms were responsible for 42% of the UTIs among women who were exposed to these products.

A prospective study of risk factors for symptomatic urinary tract infection in young women

BACKGROUND

Although acute urinary tract infections are common in young women, the associated risk factors have not been defined prospectively.

METHODS

We recruited sexually active young women who were starting a new method of contraception at a university health center or a health maintenance organization (HMO) and monitored them for six months for symptomatic urinary tract infections. Daily diaries and serial interviews were used to collect data on potential risk factors.

RESULTS

Among 796 women, the incidence of urinary tract infections per person-year was 0.7 in the university cohort (mean age, 23 years; n = 348) and 0.5 in the HMO cohort (mean age, 29; n = 448). In both cohorts, there were strong dose-response relations between the risk of infection and both recent use of a diaphragm with spermicide (respective relative risks for one, three, and five days of use in the past week, 1.42, 2.83, and 5.68 in the university cohort, P<0.001; and 1.29, 2.14, and 3.54 in the HMO cohort, P=0.04) and recent sexual intercourse (respective relative risks for one, three, and five days with intercourse in the past week, 1.37, 2.56, and 4.81 in the university cohort, P<0.001; and 1.24, 1.91, and 2.96 in the HMO cohort, P=0.002). The risk of acute infection was also associated with a history of recurrent infection (relative risk, 5.58 in the university group and 2.10 in the HMO group) but not with cervical-cap use, ABO-blood-group nonsecretor phenotype, or delayed postcoital voiding.

CONCLUSIONS

Among sexually active young women the incidence of symptomatic urinary tract infection is high, and the risk is strongly and independently associated with recent sexual intercourse, recent use of a diaphragm with spermicide, and a history of recurrent urinary tract infections.

Prevention of pelvic inflammatory disease by screening for cervical chlamydial infection

BACKGROUND

Chlamydia trachomatis is a frequent cause of pelvic inflammatory disease. However, there is little information from clinical studies about whether screening women for cervical chlamydial infection can reduce the incidence of this serious illness.

METHODS

We conducted a randomized, controlled trial to determine whether selective testing for cervical chlamydial infection prevented pelvic inflammatory disease. Women who were at high risk for disease were identified by means of a questionnaire mailed to all women enrollees in a health maintenance organization who were 18 to 34 years of age. Eligible respondents were randomly assigned to undergo testing for C. trachomatis or to receive usual care; both groups were followed for one year. Possible cases of pelvic inflammatory disease were identified through a variety of data bases and were confirmed by review of the women's medical records. We used an intention-to-screen analysis to compare the incidence of pelvic inflammatory disease in the two groups of women.

RESULTS

Of the 2607 eligible women, 1009 were randomly assigned to screening and 1598 to usual care. A total of 645 women in the screening group (64 percent) for chlamydia; 7 percent tested positive and were treated. At the end of the follow-up period, there had been 9 verified cases of pelvic inflammatory disease among the women in the screening group and 33 cases among the women receiving usual care (relative risk, 0.44; 95 percent confidence interval, 0.20 to 0.90). We found similar results when we used logistic-regression analysis to control for potentially confounding variables.

CONCLUSIONS

A strategy of identifying, testing, and treating women at increased risk for cervical chlamydial infection was associated with a reduced incidence of pelvic inflammatory disease.

Current use of thiazide diuretics and prevention of femur fractures

A case control study of a defined population from The Netherlands was performed to evaluate the risk of femur fractures associated with the use of thiazide diuretics. Included were 386 patients hospitalized for femur fractures between 1986 and 1990 who were residents and 45 years of age and older. Per case, one age-, sex-, pharmacy-, and general practitioner-matched control was chosen from the general population. Drug use was ascertained from computerized pharmacy records. The adjusted odds ratio of current use of thiazide diuretics was 0.5 (95% confidence interval, 0.3-0.9). The protective effect of thiazide diuretics was greatest for use of 1 year or longer at relatively high doses of thiazides (odds ratio, 0.3; 95% confidence interval, 0.1-0.9). We also found that patients who discontinued thiazide use longer than 2 months were not protected against femur fractures. These results support the hypothesis that use of thiazide diuretics protects against femur fractures.

Contraceptive methods and the risk of Chlamydia trachomatis infection in young women

To evaluate the relation between contraceptive methods and cervical Chlamydia trachomatis infection, the authors studied a population-based sample of 1,779 nonpregnant women aged 15-34 years who underwent cell culture diagnostic testing for the detection of C. trachomatis at a health maintenance organization. Barrier contraceptive method users were classified as those who reported using one of the following methods at time of testing: condom, diaphragm, cervical cap, spermicidal sponge, foam, or vaginal spermicidal suppositories. Barrier methods were associated with a reduction in the risk of chlamydial infection in women aged 25 years or older when compared with all other women in the same age category (adjusted prevalence odds ratio = 0.15, 95% confidence interval (CI) 0.04-0.66). When compared with only noncontracepting women, the adjusted prevalence odds ratio was 0.34 (95% CI 0.06-1.99). The protective effect of barrier methods was not evident in women younger than age 25 years. Oral contraceptive use was not associated with the risk of C. trachomatis infection using either referent group; the adjusted prevalence odds ratio was 0.99 (95% CI 0.57-1.73) compared with all other women, and 0.88 (95% CI 0.44-1.79) compared with noncontracepting women. These findings suggest that present patterns of use of barrier methods differ by age and afford only selective protection against cervical C. trachomatis infections.

Overview of cost-consequence modeling in outcomes research

Outcomes research has developed in response to the need for information on costs, risks, and benefits of clinical treatments, including data regarding the effectiveness of prescription drugs. It attempts to consider more than the biologic effects of pharmaceuticals, that is, to encompass wider measures of the results of their use, issues that are not routinely addressed during clinical trials. Cost-effectiveness analysis compares the outcome of different treatment options in terms of monetary cost per unit of effectiveness. Examples of measures of effectiveness are years of life saved, number of days of hospitalization avoided, and number of treatment successes. Cost-consequence models, also referred to as cost-outcome models, deal with costs and a variety of outcomes ranging from clinical to humanistic. Direct medical costs are those for prevention, detection, treatment, and rehabilitation; they are amounts spent to treat an illness, including hospitalization, professional services, pharmaceuticals, and medical supplies. Indirect medical costs are associated with changes in productivity, such as earnings lost because of illness. Humanistic outcomes deal primarily with functional status, quality of life, and satisfaction, and may include pain, anxiety, self-esteem, ability to carry out normal activities, and overall impressions. Since it is not possible to study all effects of treatments with clinical trials, modeling techniques are useful in making therapeutic decisions.

Nonsteroidal antiinflammatory drug discontinuation in patients with osteoarthritis

OBJECTIVE

To determine whether discontinuation patterns differed among nonsteroidal antiinflammatory drugs (NSAID) prescribed to treat osteoarthritis (OA).

METHODS

In a retrospective cohort study of Health Maintenance Organization enrollees, 1405 patients with OA aged 45 and older who received a new prescription for one of 4 NSAID were followed for 12 months. Survival analysis was used to evaluate time to discontinuation, used here as a relative measure of both drug efficacy and tolerability.

RESULTS

Rates of NSAID discontinuation during the study period were high; only 15 to 20% of those started on a study NSAID were still using the same drug at the end of the 12 month followup period. Using a proportional hazards model to adjust for covariates, the risk of discontinuation did not differ when comparing the agent with the longest duration of use, piroxicam (the referent), to enteric coated aspirin [relative risk (RR) 1.10, 95% confidence interval (CI) 0.93 to 1.30]. Adjusted rates of discontinuation were significantly higher for ibuprofen (RR 1.43, 95% CI 1.22 to 1.69) and for naproxen (RR 1.40, 95% CI 1.19 to 1.65) when compared to piroxicam.

CONCLUSION

NSAID discontinuation rates are high among patients with OA and risk of discontinuation differed between NSAID, even after controlling for the effects of such variables as age, disease severity, and concomitant therapy.

Postmenopausal estrogen replacement therapy and the risk of Alzheimer's disease: a population-based case-control study

Preliminary animal and human data suggest that estrogens may be protective against Alzheimer's disease in women. In a population-based case-control study at Group Health Cooperative of Puget Sound, Seattle, Washington, the authors compared the exposure of estrogen replacement therapy of 107 female Alzheimer's disease cases with 120 age- and sex-matched controls by using computerized pharmacy data. The cases were obtained from the Alzheimer's Disease Patient Registry of the University of Washington, Seattle, Washington, which is based on the enumerated health plan population from 1987 to 1992. Newly recognized cases of probable Alzheimer's disease according to standardized diagnostic criteria were ascertained, evaluated, and enrolled in the Registry. The controls were selected from the same defined population by stratified random sampling. When the authors applied logistic regression, ever use of estrogens did not show an association with Alzheimer's disease (adjusted odds ratio = 1.1, 95 percent confidence interval 0.6-1.8). Oral and vaginal estrogens yielded similar results. In conclusion, this study provides no evidence that estrogen replacement therapy has an impact on the risk of Alzheimer's disease in women.

Cost-effectiveness of unenhanced MR imaging vs contrast-enhanced CT of the abdomen or pelvis

OBJECTIVE

The purpose of this study was to compare the cost of unenhanced MR imaging with that of CT with high- or low-osmolality contrast agents for abdominal or pelvic imaging in patients at risk for nephrotoxic effects induced by contrast material.

MATERIALS AND METHODS

Using decision analysis, we evaluated direct medical care costs associated with the use of CT with high- or low-osmolality contrast agents vs MR imaging. We used data from a variety of sources and assumed that the three imaging techniques provide nearly equivalent diagnostic information. The base-case analysis assumed that patients were at low risk (i.e., 2%) for development of nephrotoxic effects. Nephrotoxic effects were defined as increases in the serum level of creatinine of 50% or more above baseline. Our analysis took the perspective of the hospital and used estimated hospital costs, not charges. Sensitivity analyses were performed on risk estimates for development of nephrotoxic effects and for the estimates of medical care costs.

RESULTS

For the base case, CT with a high-osmolality contrast agent is the least costly imaging strategy. When the risk of nephrotoxic effects exceeds 5% for high-osmolality contrast agents or 2.6% for low-osmolality contrast agents, then MR imaging is the preferred strategy. The model is relatively insensitive to treatment costs.

CONCLUSION

From a hospital's perspective, MR imaging of the abdomen or pelvis is cost minimizing in patients at high risk for nephrotoxic effects induced by contrast agents. Use of low-osmolality contrast agents must reduce the frequency of nephrotoxic effects in high-risk patients by at least 50% to be less costly than MR imaging.

Diet and genital warts: a case-control study

BACKGROUND AND OBJECTIVES

Clinical observations support a substantial role for impaired immunity in the development of human papillomavirus (HPV) infections. Intake of vitamins, especially vitamins A and C, and alcohol consumption have been reported to influence immune response.

GOAL OF THE STUDY

To examine the relationship between nutritional risk factors, including alcohol consumption, and the risk of genital warts.

STUDY DESIGN

A case-control study was conducted among enrollees at four clinics of Group Health Cooperative in western Washington state. A total of 188 cases diagnosed with condyloma from April 1, 1987 to September 30, 1991 and 245 controls completed a semi-quantitative food frequency questionnaire.

RESULTS

After adjustment for socioeconomic indicators, total energy intake, smoking and sexual behavior, a weekly consumption of two to four alcoholic drinks was associated with an almost doubled risk of genital warts (OR = 1.9, 95% confidence interval [CI] = 1.0-3.6). Consuming five or more alcoholic drinks per week was even more related to the risk of genital warts (OR = 2.4, 95% CI = 1.2-5.1). The risks tended to increase with the number of alcoholic drinks (P = 0.006). Vitamin A and C intakes as measured by a food frequency questionnaire did not alter the risk of condyloma.

CONCLUSION

Moderately high consumption of alcohol is associated with increased risk of condyloma. Further biological and epidemiological studies are needed to explain this association.

Cigarette smoking and functional ovarian cysts

This population-based case-control study assessed the effect of current cigarette smoking on the risk of functional ovarian cyst development. Cases were all 15-39 year-old enrollees of Group Health Cooperative of Puget Sound (GHC) who had either an inpatient primary diagnosis of functional ovarian cyst in 1988 or 1989 (n = 61) or an outpatient primary diagnosis of functional ovarian cyst at five GHC primary care clinics (n = 37). Controls were randomly selected enrollees matched to cases on age, primary care clinic, and enrollment date (n = 239). A total of 36.7% of cases and 20.5% of controls were determined by medical record review to be current smokers. Compared with women who were not current smokers, the relative risk of a diagnosed functional ovarian cyst among smokers was 2.0 (95% confidence interval 1.1-3.5). This finding, in an area of little previous research, provides a promising lead in the investigation of the etiology of functional ovarian cysts and the effects of smoking on ovarian function in general.

Accuracy of technicians and pharmacists in identifying dispensing errors

The accuracies with which pharmacists and technicians checked medications in a unit dose distribution system were compared. The study was conducted at three large hospitals in Washington State. From August through October 1991, technicians filled unit dose medication drawers and pharmacists verified the accuracy of each fill (pharmacist verification period). From November 1991 through January 1992, technicians who had undergone special training verified the accuracy with which medication drawers were filled by other technicians (technician verification period). For each study period, two error rates were estimated: the frequency with which the pharmacists or technicians identified dispensing errors and the frequency of verification errors identified by the investigators in a final independent check. A total of 143,952 unit doses were dispensed during the pharmacist verification period, of which 49,718 were randomly analyzed for accuracy. A total of 151,721 doses were dispensed during the technician intervention period, of which 55,470 were assessed. The mean +/- S.E. daily rates of dispensing-error identification by pharmacists (0.0125 +/- 0.0069%) and technicians (0.0119 +/- 0.0001) did not differ significantly. While pharmacists overlooked more errors (107) than technicians (50), the percentage of such missed errors classified as potentially serious did not differ significantly between the groups (25.2% versus 32.0%, respectively). Pharmacy technicians who underwent special training were able to verify medications in a unit dose distribution system without compromising the accuracy of dispensing.

Selective screening for Chlamydia trachomatis infection in a primary care population of women

The authors used cell cultures to assess the prevalence of cervical infection caused by Chlamydia trachomatis in a population-based sample of nonpregnant women aged 15 to 34 years who attended two primary care clinics at Group Health Cooperative of Puget Sound, Seattle, Washington, between January 1, 1988, and June 30, 1989. C. trachomatis was isolated from 67 of 1,804 women (3.7%), including 13% of those who were less than 20 years of age. Seven patient characteristics were independently predictive of chlamydial infection by stepwise multivariate logistic regression analysis: being unmarried, examination showing cervical ectopy, black race, douching, nulliparity, age of 24 years or less, and intercourse with two or more partners within the preceding year. Testing all women who had a score of 5 or more (28% of women) on a weighted index based on risk factors would detect 77% of all infections with a positive predictive value of 9%. These results suggest that it should be feasible to develop a risk factor-based program to screen for cervical infection with C. trachomatis in populations where its prevalence is low.

Vaginal douching as a risk factor for acute pelvic inflammatory disease

OBJECTIVE

To conduct a population-based evaluation of vaginal douching as a risk factor for acute pelvic inflammatory disease (PID), emphasizing timing, frequency, and reasons for douching.

METHODS

A population-based case-control study was conducted at Group Health Cooperative of Puget Sound, a staff-model health maintenance organization located in western Washington state. The cases (N = 131) were women 18-40 years of age who experienced a first episode of clinically diagnosed acute PID. Both hospitalized and ambulatory-care patients were identified. Medical records were reviewed for clinical inclusion criteria and for additional evidence of inflammation/infection. Controls (N = 294) were chosen from a population-based series of randomly selected women from a concurrent Group Health study of ectopic pregnancy. Of the women identified, 72.4% of cases and 73.4% of controls agreed to participate.

RESULTS

Relative to women who reported never having douched, women who douched during the previous 3 months had a risk of PID of 2.1 after controlling for other measured risk factors (95% confidence interval [CI] 1.2-3.9). Women who douched at least once a week had a higher estimated risk (odds ratio 3.9, 95% CI 1.4-10.9) than those who douched less often (odds ratio 1.8, 95% CI 1.0-3.4). The risk was highest in the small group of women who gave infection as the reason for douching (odds ratio 7.9, 95% CI 2.6-24.2). However, exclusion of this group did not eliminate the association among the remaining women (odds ratio 3.0, 95% CI 1.0-9.1 for douching at least once a week).

CONCLUSION

These population-based data lend added support to the hypothesis that vaginal douching can predispose a woman to PID.

Epidemiology of the noncontraceptive effects of oral contraceptives

Epidemiologic studies have made major contributions to our understanding of oral contraceptive safety. The findings of major studies of the risk of cardiovascular disease and neoplasia associated with oral contraceptive use are reviewed, with discussion of the many factors to be considered in the interpretation of conflicting results. Mortality data that were based on earlier oral contraceptive formulations, dosage, and usage patterns may not be relevant to current practices. Recent studies suggest that reduced doses of estrogens and progestins in current oral contraceptive formulations and better screening of patients have reduced the risks of cardiovascular disease associated with oral contraceptive use, with the most important exception being an increased risk among older women who smoke. While most epidemiologic studies demonstrate no overall association between oral contraceptives and breast cancer, unanswered questions remain concerning this relationship.

Medical care and cost outcomes after pentoxifylline treatment for peripheral arterial disease

We assessed the medical outcomes and costs associated with the pharmacologic treatment of patients with peripheral arterial disease (PAD) in a population-based historical cohort study of patients enrolled in a health maintenance organization. For up to 2 years, we compared 58 patients who used therapeutic amounts of pentoxifylline with a comparison group of 112 patients who received a minimal subefficacious trial of pentoxifylline. Medical records data were used to assess and control for the severity of PAD and other potentially confounding factors. Continuous use of a therapeutic amount of pentoxifylline during an initial 120-day period significantly reduced the incidence of PAD-related invasive therapeutic and diagnostic procedures in the first year of follow-up (adjusted relative risk, 0.35; 95% confidence interval, 0.12 to 0.99). However, there were no significant differences in the risk of a PAD-related hospitalization or cost of PAD-related care between continuous pentoxifylline users and the comparison group. Pentoxifylline therapy may reduce the risk of vascular surgery while not increasing the total cost of PAD care.

Tubal sterilization and risk of subsequent hospital admission for menstrual disorders

OBJECTIVE

Our objective was to investigate tubal sterilization and subsequent hospitalization for menstrual disorders.

STUDY DESIGN

Automated discharge data were used in a population-based cohort study of 7253 women aged 20 to 49 years with tubal sterilization (1968 through 1983) at Group Health Cooperative of Puget Sound. Comparisons were with an age-matched cohort of 25,448 nonsterilized women and a nonmatched cohort of 5283 spouses of men with vasectomies.

RESULTS

In the sterilization cohort, 282 had hospitalization for menstrual disorders (curettage, n = 191; hysterectomy, n = 66; nonsurgical, n = 25). Risk of hospitalization for menstrual disorders was 2.4 times greater after tubal sterilization (95% confidence interval 2.0 to 2.9). This risk was 6.1 times greater for sterilized women aged 20 to 24 years (95% confidence interval 0.72 to 3.2). Compared with the risk for nonsterilized women whose spouses had a vasectomy, the risk was 1.6 times greater (95% confidence interval 1.3 to 2.1). Hospitalization for menstrual disorders was not more common after unipolar sterilization than after other methods, as might have been expected if the menstrual disorder was related to impaired uteroovarian circulation.

CONCLUSIONS

Tubal sterilization is associated with a greater risk of hospitalization for menstrual disorders. A biologic association is not supported by these results.

Functional ovarian cysts in relation to the use of monophasic and triphasic oral contraceptives

This population-based case-control study assessed the effect of current use of monophasic or triphasic oral contraceptives (OCs) on the risk of functional ovarian cyst development. The cases were all 15-39-year-old enrollees in the Group Health Cooperative of Puget Sound who had either an inpatient primary diagnosis of functional ovarian cyst in 1988 or 1989 (N = 67) or an outpatient primary diagnosis of functional ovarian cyst from March 1988 through August 1989 at one of five Group Health Cooperative primary care clinics (N = 39). Controls were randomly selected enrollees matched to the cases for age, primary care clinic, and enrollment date (N = 255). Subjects with previous hysterectomy or oophorectomy were excluded from this analysis. Pharmacy and medical record review showed that 16% of cases and 19% of controls were currently using monophasic OCs, whereas 11% of cases and 9% of controls were using triphasic OCs. Compared with women not using hormonal contraception, the relative risks of a diagnosed functional ovarian cyst among women currently using OCs were 0.8 (95% confidence interval [CI] 0.4-1.8) for users of monophasic OCs and 1.3 (95% CI 0.5-3.3) for users of triphasic OCs. In contrast to previous studies of monophasic OCs containing higher steroid dosages, the results of this study suggest that current use of low-dose monophasic OCs does not substantially decrease a woman's risk of functional ovarian cyst formation. In addition, our results do not support recent speculation that current use of triphasic OCs appreciably increases the risk of functional ovarian cysts.

Successful implementation of a guideline program for the rational use of lipid-lowering drugs

Following the National Cholesterol Educational Program's (NCEP) 1988 screening and treatment recommendations, an educational and behavior-change program at Group Health Cooperative of Puget Sound (GHC) was developed to guide the use of lipid-lowering drugs within the larger context of cardiac risk reduction. The program has been successful in advancing a rational program to enhance care and manage costs of the use of lipid-lowering agents at GHC. Cost savings have been significant over the past two years. The educational design of the program includes training and ongoing education of a core group of "lipid gurus," who educate colleagues in area medical centers in a rational approach to hyperlipidemia. Patient education and patient participation in decision-making was emphasized. Program evaluation has demonstrated that physicians and patients are satisfied with the program, and inappropriate drug expenditures have been prevented. Key elements of the program include a critical review of outcome studies in the medical literature, use of information systems, algorithms and written materials organized into a well-designed, ongoing educational program, and development of a core group of physicians and pharmacists to administer the program at the clinic level.

Results and effect of refusal recontact in a case-control study of ectopic pregnancy

In a population-based study of ectopic pregnancy, a card-back system of initial subject contact yielded a positive response from 85.1% of potential cases and 65.0% of potential controls. We subsequently recontacted by telephone potential controls who had declined to participate and offered them $10 compensation for participation. We recontacted 205 of the 330 control women who had refused to participate. Twenty-eight were found to be ineligible, and 75.7% of the remaining 177 agreed to participate on recontact. The additional 125 women were not recontacted, 9 because of firm initial refusals and 116 because the study ended before contact could be sought or made. Controls who initially declined to participate but subsequently agreed differed from those who initially agreed on a number of demographic and reproductive characteristics, and estimates of ectopic pregnancy risk associated with induced abortion and tubal sterilization changed slightly with the inclusion of initial nonparticipants. We conclude that use of a card-back system of subject contact yields a low level of positive response among reproductive-aged women and that many initial nonparticipants can be converted into participants by personal contact and the offer of compensation.

Diuretic drug use and the risk for hip fracture

OBJECTIVE

To test the hypothesis that use of thiazide diuretics prevents hip fracture and to study the risk for hip fracture associated with furosemide use.

DESIGN

A case-control study.

SETTING

Hospitals owned by a health maintenance organization in Washington.

PATIENTS

Elderly patients (n = 462) hospitalized because of a hip fracture between 1977 and 1983 and an equal number of age- and sex-matched population-based control patients.

MEASUREMENTS

Use of thiazide diuretics and furosemide was ascertained from medical records and computerized pharmacy records. The relative risk for hip fracture associated with diuretic use was calculated and adjusted for the potentially confounding effects of nursing home residence; previous hospitalizations; a history of stroke, alcoholism, or the organic brain syndrome; body weight; leg paralysis; and use of phenobarbital, corticosteroids, or other diuretics. Current and former users of diuretics were analyzed separately.

MAIN RESULTS

The adjusted risk for hip fracture was 1.6 (95% CI, 1.0 to 2.5) for current thiazide users. The adjusted risk for hip fracture for current furosemide use was 3.9 (CI, 1.5 to 10.4).

CONCLUSIONS

According to this study, use of thiazide diuretics did not protect against hip fracture and cannot be recommended for fracture prevention. Current furosemide use was also associated with hip fracture.

Maternal cigarette smoking and the risk of tubal pregnancy

The association between maternal smoking and the occurrence of tubal pregnancy was evaluated in a population-based case-control study of members of the Group Health Cooperative of Puget Sound, Seattle, Washington. Women hospitalized with tubal pregnancy from October 1981 through September 1986 (n = 274) were compared with reproductive-age women who were at risk of becoming pregnant during the same time period (n = 727). The relative risk of tubal pregnancy associated with ever having smoked cigarettes was 1.3 (95% confidence interval (CI) 1.0-1.8). Compared with women who had never smoked, those who smoked at the time of conception had a 40% increase in the risk of tubal pregnancy (95% CI 1.0-2.0). These results support earlier epidemiologic and nonepidemiologic reports of a greater risk of tubal pregnancy associated with current or recent maternal cigarette smoking.

Vaginal douching and the risk of tubal pregnancy

To explore the possible association between vaginal douching and tubal pregnancy, we interviewed 273 women who were diagnosed with tubal pregnancy at Group Health Cooperative between September 31, 1981 and October 1, 1986. Their responses were compared with responses of a random sample of 722 female members of Group Health Cooperative who were assumed to be at risk of becoming pregnant at the time the cases conceived. After adjusting for differences between cases and controls with regard to other measured risk factors, we found a modest increase in risk associated with having douched more than two times per year in the past (RR = 1.3, 95% CI: 0.9-1.8). Among women who had more than one sexual partner during their lifetime, however, the risk for those who had douched more than twice per year was somewhat higher (RR = 1.6, 95% CI: 1.1-2.3). There was an indication that women who had been exposed to Chlamydia trachomatis, as indicated by elevated antibody titers, may further increase their risk for tubal pregnancy by douching (RR = 2.4, 95% CI: 0.8-7.3). The associations found in other studies between douching and pelvic inflammatory disease, and between pelvic inflammatory disease and subsequent tubal pregnancy, argue that a relation between douching and tubal pregnancy might be anticipated. Our results offer further support for this hypothesis.

Tubal sterilization and the long-term risk of hysterectomy

To assess the effect of tubal sterilization on the risk of hysterectomy, we studied 7414 women aged 20 to 49 years who had had a tubal sterilization at a health maintenance organization between January 1, 1968, and December 31, 1983. Compared with a population-based cohort of nonsterilized women, women sterilized while 20 to 29 years old were 3.4 times more likely to have had a subsequent hysterectomy (95% confidence interval, 2.4 to 4.7). Adjustment for the effects of potential confounders with a subset of 276 women did not appreciably alter this association. For multivariate comparisons with 5323 wives of vasectomized men, there was no significant elevation in the risk of hysterectomy following sterilization among women sterilized while 20 to 29 years old. Tubal sterilization was not associated with hysterectomy for married women who underwent tubal sterilization at age 30 or older. These results do not support a biological basis for the relationship between tubal sterilization and hysterectomy.