Claire C. Bourguet

A comparison of the National Ambulatory Medical Care Survey (NAMCS) measurement approach with direct observation of outpatient visits.

BackgroundThe National Ambulatory Medical Care Survey (NAMCS) informs a wide range of important policy and clinical decisions by providing nationally representative data about outpatient practice. However, the validity of the NAMCS methods has not been compared with a reference standard. MethodsOffice visits of 549 patients visiting 30 family physicians in Northeastern Ohio were observed by trained research nurses. Visit content measured by direct observation was compared with data reported by physicians using the 1993 NAMCS form. ResultsOutpatient visit physician reports of procedures and examinations using the NAMCS method showed generally good concordance with direct observation measures, with kappas ranging from 0.39 for ordering a chest x-ray to 0.86 for performance of Pap smears. Concordance was generally lower for health behavior counseling, with kappas ranging from 0.21 for alcohol counseling to 0.60 for smoking cessation advice. The NAMCS form had high specificity (range, 0.90–0.99) but variable (range, 0.12–.84) sensitivity compared with direct observation, with the lowest sensitivities for health behavior counseling. The NAMCS physician report method overestimated visit duration in comparison with direct observation (16.5 vs. 12.8 minutes). ConclusionsCompared with direct observation of outpatient visits, the NAMCS physician report method is more accurate for procedures and examinations than for health behavior counseling. Underreporting of behavioral counseling and overreporting of visit duration should lead to caution in interpreting findings based on these variables.

Antigenic stimulation and multiple myeloma. A prospective study

Background. A causal relationship between antigenic conditions and multiple myeloma was suggested by case reports. Although controlled studies identified associations with individual conditions, they failed to give overall support to the hypothesis. Using a prospective cohort representative of the U.S. population, the authors hypothesized that immune‐stimulating conditions are a risk factor for multiple myeloma.

Use of the Centers for Disease Control and Prevention childhood lead poisoning risk questionnaire to predict blood lead elevations in pregnant Women

Objective To assess the accuracy of a questionnaire developed by the Centers for Disease Control and Prevention (CDC), given to pregnant women for identification of children at risk for lead poisoning. Methods The study population consisted of all 314 new prenatal patients enrolled in health department clinics in 1990–1992. Lead was measured in venous blood, and patients completed written questionnaires to gather information about lead expoure risk factors. The relationship between elevated maternal blood lead levels (at or greater than 10 μg/dL or 0.483 μmol/L) and responses to the CDC questionnaire and other questions were examined using χ2 statistical analysis. Results Two hundred ninety-nine women provided responses to questions about lead exposure risk. Thirty-nine women (13%) had elevated blood lead levels. A woman with a positive response to at least one CDC question was more likely to have elevated blood lead than a woman who answered negatively to all four CDC questions (relative risk = 2.39,95% confidece interval 1.17–4.89; P = .01). Using the CDC definition of high risk (“yes” to at least one question), the questionnaire had a sensitivity of 75.7% and a negative predictive value of 93.1%. A questionnaire that combined housing conditions, smoking status, and high consumption of canned foods had a sensitivity of 89.2% and a negative predictive value of 96.4%. A high prevalence of elevated blood lead in children living with women with elevated blood lead was observed. Conclusion Querying pregnant women about risk factors for lead exposure can aid in assessing prenatal lead exposure risk. The sensitivity and negative predictive value of the CDC questionnaire, when used with high-risk women, are comparable to its reported accuracy in young children.

Recognition and management of obesity in a family practice setting.

Background: Research on the diagnosis and management of obesity in primary cure is limited. Our study goals were to describe the rate of obesity in a primary care setting, to identify factors associated with clinically recognized obesity, and to ascertain the level of diet and exercise counseling for obesity. Methods: Medical records from a private group practice were used for a historical cohort study of 276 patients (aged 40 years and older) who were provided care for a maximum 4.5-year follow-up period. Results: forty-six percent of the study patients (95 percent confidence interval = 0.43, 0.49) received an obesity diagnosis according to medical record notations. The diagnosis of obesity, in turn, was predicted by body mass index (BMI) quartile (P < 0.001) and a positive family history of cardiovascular disease (P < 0.01). Those patients with a diagnosis of obesity had a higher mean level of subsequent weight and diet counseling (P = 0.0001) but the same level (P = 0.11) of exercise counseling as nonobese patients. Weight and diet counseling was also predicted by diabetes (P = 0.0001) and hypercholesterolemia (P = 0.0003). Conclusions: The clinical recognition of obesity was not determined by BMI alone. Although weight and diet counseling was initiated for those individuals described as obese, there was a relatively low level of exercise counseling among these patients. Additional research could provide ways of reducing both physician and patient barriers to exercise counseling.

The better weight-better sleep study: a pilot intervention in primary care.

OBJECTIVE To explore the feasibility of integrating sleep management interventions with dietary and exercise interventions for obesity in a 12-week randomized trial. METHODS We randomized 49 overweight or obese adult patients either to a better weight (BW) cognitive behavioral intervention, or to a combination of the BW intervention and a better sleep intervention, better weight-better sleep (BWBS). RESULTS The BWBS group lost weight faster (P=.04), and coping self-efficacy accelerated (P=.01). CONCLUSIONS These preliminary results merit replication in a larger primary care-based trial with a longer continuous intervention and follow-up period.

Predictors of Nonattendance at the First Newborn Health Supervision Visit

Failure to attend the first newborn health supervision visit is an important problem for the Continuity Care Clinic of Childrens Hospital Medical Center of Akron, Ohio. The goal of this study was to use objective data from the neonatal record to identify newborns at high risk of failure to attend. Clinical and social risk factors of the mother and newborn were abstracted from the neonatal progress notes of 319 infants. The relative risk (RR) of nonattendance was calculated for each factor, and rules for predicting failure to attend were evaluated. The best predictors were multiparous mother (RR = 2.4, P = .01), no telephone in home (RR = 2.6, P= .002), and unmarried teenage mother (RR = 5.8, P= .05). Newborns who had a medical problem and had an adult mother were more likely to attend (RR = 0.4, P = .02) . These risk factors were easily identifiable from the medical record at birth. Because interventions may be labor-intensive, it is important to target the families at the highest risk.

A method for obtaining an unbiased sample of family medicine patients for research purposes.

Introduction: Obtaining a representative patient sample for research purposes can be challenging. Classic probability sampling can be trusted, but these approaches are not always feasible; yet alternatives may introduce bias. We summarize relevant literature, the need for new approaches, and illustrate a practical hybrid approach that could consistently produce representative patient samples. Methods: Valid approaches shift sampling decisions from fallible interviewers to less fallible reproducible processes. In lieu of the interviewers inclination to select particular patients, we used the day of the week, the appointment time, and the sequence of the patients last name in the alphabet to select a sample for a consent process and a survey. Illustrative Use: Characteristics of the study sample (n = 225) were compared with the characteristics of the population (N = 1964) that had an office visit during the recruiting period. The data suggested that the study sample was highly representative of the population in this illustrative case. Discussion: A hybrid sampling approach, in the context of a brief consent process, and a nonthreatening interview produced a representative study sample, but formal evaluation via simulation is needed to validate the hybrid approach. Convenience samples of consecutive patients should be avoided to minimize bias.

Vaginosonographic Measurement of Endometrial Thickness in the Evaluation of Amenorrhea

The progestin-induced withdrawal bleeding test has often been used in patients with amenorrhea to assess endogenous estrogen (E) production. The endometrial thickness measured by vaginal ultrasonography with a 5 MHz transducer is also dependent on E stimulation of the endometrium. In this study, 70 consecutive patients were evaluated with both the progestin-induced withdrawal bleeding test and a measurement of the endometrial thickness by vaginosonography. An endometrial thickness of 1.5 mm or less was selected to predict absence of bleeding after a progesterone (P) challenge test. This resulted in a sensitivity of 94% with a 95% confidence interval (CI) of 0.70 to 1.00 and a specificity of 93% with a 95% CI of 0.82 and 0.98. The positive and negative predictive values were 79% and 98%, respectively. This study shows that the endometrial thickness measured by vaginosonography can predict the results of the P challenge test in patients with amenorrhea.