Search PubMed⌕ Search

Biomedical subjects

G J Martin

Publications and source records attributed to G J Martin.

At least 55 records · Page 3Linked to original sources
PubMed1998

Measurement of walking endurance and walking velocity with questionnaire: validation of the walking impairment questionnaire in men and women with peripheral arterial disease.

OBJECTIVES: The Walking Impairment Questionnaire (WIQ) was designed to measure community walking ability in patients with peripheral arterial disease (PAD) and intermittent claudication. We compared the WIQ scores to objective measures of walking in a heterogeneous group of patients with and without PAD. METHODS: The study was designed as a cross-sectional study, with the setting in an academic medical center. The subjects were patients with PAD (n = 145) who were identified from a noninvasive vascular laboratory at an academic medical center. The patients without PAD (n = 65) were identified from a general medicine practice. The average number of comorbidities was 2.03 for patients with PAD and 1.52 for patients without PAD. Among the patients with PAD, 28% had classical intermittent claudication symptoms and 55% had exertional leg symptoms other than claudication. The main outcome measures were the WIQ estimates of the patient-reported walking distance and walking speed on a scale of 0 to 100. Walking endurance was measured objectively with the 6-minute walk. Walking velocity was measured with a 4-m walk. PAD and PAD severity were defined with the ankle brachial index. RESULTS: The Spearman rank correlation coefficients (rho) between the WIQ distance score and the 6-minute walk score were 0.557 among patients with PAD (P <.001) and 0.484 among patients without PAD (P <.001). The correlation coefficients between the WIQ speed score and the usual-paced 4-m walk score were 0.528 among patients with PAD (P <.001) and 0.524 among patients without PAD (P <.001). The correlations were not affected by the presence versus the absence of intermittent claudication, by PAD severity, or by the presence of 2 or more versus less than 2 comorbid illnesses. The WIQ scores in the highest and lowest quartiles were the most closely associated with the objective measures of function. CONCLUSION: The WIQ is a valid measure of community walking ability in a heterogeneous group of patients with and without PAD. The WIQ discriminates best among patients in the highest and the lowest quartiles of walking speed and endurance.

Aged↗
PubMed1998

Blood pressure and diurnal variation in sodium, potassium, and water excretion.

The objective of this study was to examine associations of blood pressure (BP) with ratios of overnight to 24-h urinary excretion of sodium, potassium, and water. Each of 125 men 27-64 years of age, not taking diuretics, had BP measured during the day on a Monday. Beginning Monday evening, each participant provided three carefully timed 24-h urine collections, divided into daytime and overnight (bedtime to awakening) specimens. Proportion of total 24-h excretion of sodium, potassium, and water in the overnight specimen, standardised for creatinine excretion, was determined for each 24-h period. Associations of systolic and diastolic BP (SBP/DBP) with these proportions were examined with control for age, body mass index, alcohol intake, and heart rate. Mean BP was 116/71 mm Hg; 15 men were on non-diuretic anti-hypertensive therapy. Mean 24-h urinary excretion was 168 mmol for sodium, 68 mmol for potassium, and 16 mmol for creatinine. Mean overnight to 24-h proportions averaged over the 3 days were 30.7% for sodium, 22.0% for potassium, 32.1% for urinary volume, and 33.2% for creatinine. Partial correlations of SBP and DBP with the 3-day averages were 0.257 (P < 0.01) and 0.210 (P < 0.05) for sodium; 0.223 (P < 0.05) and 0.222 (P < 0.05) for potassium; 0.127 and 0.091 for urinary volume; and -0.033 and 0.014 for creatinine. Correlations for sodium proportions were larger for the first 24-h period, compared to the second or third 24-h period. These results indicate that higher BP was associated with a relatively greater proportion of sodium and potassium excretion at night. Further work is needed to clarify temporal sequence, ie, whether a relatively greater sodium and potassium excretion at night is a risk factor for higher BP (eg, via renal mechanisms), or whether higher BP results in relatively greater sodium and potassium excretion at night, or both.

Adult↗
PubMed1998

A classification tree analysis of selection for discretionary treatment.

OBJECTIVES: To study treatment bias in observational outcomes research, the authors present a nonlinear classification tree model of clinical and psychosocial factors influencing selection for interventional management (lower extremity bypass surgery or angioplasty) for patients with intermittent claudication. METHODS: The study sample includes 532 patients with mild to moderate lower extremity vascular disease, without prior peripheral revascularization procedures or symptoms of disease progression. All patients were enrolled in a prospective outcomes study at the time of an initial referral visit for claudication to one of the 16 Chicago-area vascular surgery offices or clinics in 1993-95. The influence of baseline sociodemographic, clinical, and patient self-reported health status data on subsequent treatment is analyzed. Study variables were derived from lower extremity blood flow records and patient questionnaires. Follow-up home health visits were used to ascertain the frequency of lower extremity revascularization procedures within 6 months of study enrollment. Hierarchically optimal classification tree analysis (CTA) was used to obtain a nonlinear model of treatment selection. The model retains attributes with the highest sensitivity at each node based on cutpoints that maximize classification accuracy. Experimentwise Type I error is ensured at P < 0.05 by the Bonferroni method and jackknife validity analysis is used to assess model stability. RESULTS: Seventy-one of 532 patients (13.3%) underwent interventional procedures within 6 months. Ten patient attributes were used in the CTA model, which had an overall classification accuracy of 89.5% (67.6% sensitive and 92.9% specific), achieving 57.7% of the theoretical possible improvement in classification accuracy beyond chance. Eleven model prediction endpoints reflected a 33-fold difference in odds of undergoing lower extremity revascularization. CONCLUSIONS: Initial ankle-brachial index (100%), leg symptom status over the previous six months (89%), self-reported community walking distance (74%) and prior willingness to undergo a lower extremity hospital procedure (39%) were used to classify most patients in the sample. These attributes are critical control variables for a valid observational study of treatment effectiveness.

Angioplasty↗
PubMed1998

The ankle brachial index independently predicts walking velocity and walking endurance in peripheral arterial disease.

OBJECTIVES: Maintaining function among older men and women is an important public health goal as the population lives longer with chronic disease. We report the relationships between lower extremity peripheral arterial disease (PAD), PAD severity, and PAD-related symptoms with walking velocity and endurance among men and women aged 55 and older. DESIGN: A cross-sectional design. SETTING: An academic medical center. PARTICIPANTS: Participants with PAD were men and women aged 55 and older identified from a blood flow laboratory or a general medicine practice (n = 158). Randomly selected controls without PAD were identified from the general medicine practice (n = 70). MEASUREMENTS: PAD was diagnosed and quantified using the ankle brachial index (ABI). Subjects were categorized according to whether they had severe PAD (ABI <0.40), mild to moderate PAD (ABI 0.40 to <0.90), or no PAD (ABI 0.90 to <1.50). Walking endurance was assessed with the 6-minute walk. Usual walking velocity and maximal walking velocity were assessed with "usual" and "maximal" paced 4-meter walks, respectively. RESULTS: Average distances achieved in the 6-minute walk were 1569+/-390 feet for subjects with ABI 0.90-1.50, 1192+/-368 feet for subjects with ABI 0.40 to <0.90, and 942+/-334 feet for subjects with ABI < 0.40 (trend P value < .001). Walking velocities for both the usual and maximal paced 4-meter walks were slowest among subjects with ABI < 0.40 and fastest among subjects with ABI 0.90 to <1.50. Subjects with PAD who had pain at rest had slower walking velocity and poorer walking endurance than other subjects with PAD. In multiple linear regression analyses that included subjects with PAD only, ABI level was an independent predictor of 6-minute walk performance (regression coefficient = 159 ft/0.40 ABI units, P = .011), usual paced 4-meter walk (regression coefficient = .095 meters/sec/0.40 ABI units, P = .031), and maximal paced 4-meter walk (regression coefficient = .120 meters/sec/0.40 ABI units, P = .050) adjusting for age, sex, race, leg symptoms, and comorbid diseases known to affect functioning. Pain at rest was associated independently with the maximally paced 4-meter walk (-0.201 meters/sec, P = .024), but not with the other walks. CONCLUSION: ABI level has a measurable and independent association with walking endurance and both usual and maximal walking velocity. These data suggest that PAD may impair lower extremity function by diminishing function of both Type I ("slow twitch") and Type II ("fast twitch") muscle fibers. Because walking velocity has important prognostic implications for functioning, these data also suggest that ABI may be used to identify patients at increased risk of mobility loss.

Aged↗
PubMed1998

Laparoscopic anterior spinal arthrodesis with rhBMP-2 in a titanium interbody threaded cage.

Anterior lumbar interbody arthrodesis is commonly performed for conditions involving infection, deformity, and instability. The purpose of this investigation was to determine the effective dose of recombinant human bone morphogenetic protein-2 (rhBMP-2) as a bone graft substitute inside a titanium threaded interbody fusion cage using a nonhuman primate model of laparoscopic anterior lumbar interbody arthrodesis. Eight adult rhesus monkeys underwent laparoscopic exposure of the lumbosacral spine followed by insertion of a hollow titanium threaded cylindrical cage (Sofamor-Danek, Memphis, TN, U.S.A.). Before insertion, the chamber of the cage was filled with a collagen sponge delivery vehicle soaked with either 0 mg/ml (sham, buffer only), 0.75 mg/ml (low dose), or 1.5 mg/ml (high dose) of rhBMP-2 (Genetics Institute, Cambridge, MA, U.S.A.). Fusions were evaluated in a blinded fashion with plain radiographs and computed tomography (CT) scans 12 and 24 weeks after surgery, and by manual palpation and histology after euthanasia 24 weeks after surgery. All five monkeys treated with a cage filled with rhBMP-2 obtained a solid fusion as assessed by manual palpation. The two monkeys that received no growth factor did not achieve solid fusions. Plain radiographs were of limited value, with fusions best assessed on sagittally reconstructed CT scans. Scans from the two animals treated without growth factor showed ingrowth of bone only into the outer edges of the cage, but not through the center. Scans from the rhBMP-2-treated animals demonstrated arthrodesis with continuous bone growth through the cage. Histologic analysis demonstrated normal mature trabecular bone surrounding and growing through the cages, which correlated with the CT scan findings. We conclude that rhBMP-2 delivered in a threaded titanium interbody cage can serve as a bone graft substitute in a nonhuman primate model. Sagittal reconstructed CT may be a better method to assess for fusion with this device.

Animals↗
PubMed1998

The Marshall R. Urist Young Investigator Award. Gene expression during autograft lumbar spine fusion and the effect of bone morphogenetic protein 2.

A prospective animal study of posterolateral lumbar spine arthrodesis was performed to determine the temporal and spatial pattern of gene expression and to determine the effect of recombinant human bone morphogenetic protein 2 on the gene expression pattern of a healing spine fusion mass. In Group 1, 20 adult New Zealand rabbits underwent L4-L5 posterolateral intertransverse process arthrodesis using autograft alone. Two rabbits were euthanized at each of the following points: 0, 2, and 4 days, and 1, 2, 3, 4, 5, 6, and 10 weeks after surgery. The same surgical technique was used for 16 rabbits in Group II, except that the autograft first was soaked in a solution of recombinant human bone morphogenetic protein 2 before implantation. Ribonucleic acid was extracted from different regions of the fusion mass at each point and analyzed for expression of bone and cartilage related genes using reverse transcription polymerase chain reaction. A reproducible temporal sequence and spatial pattern of gene expression was found in healing spine fusions. In the central portion of the fusion mass a temporal lag in gene expression was observed that parallels the lag in healing within the central zone previously observed in histologic studies. Treatment of bone graft with recombinant human bone morphogenetic protein 2 resulted in an increase in the early expression of bone morphogenetic protein 6 which was associated with expression of higher levels of Type I collagen, osteocalcin, and other bone related genes. These findings suggest that central nonunion may be associated with delayed expression of osteoblast related genes in the central region of the forming fusion mass. The growth factor, recombinant human bone morphogenetic protein 2, increased the level of bone related gene expression throughout the fusion mass, eliminated the delay in healing within the central zone, and may decrease the likelihood of a nonunion.

Animals↗
PubMed1997

15N enrichments of casein and plasma protein amino acids in cows ingesting 15N-labelled ammonium sulphate.

The aim of this work was to determine by ion-exchange liquid chromatography and isotope ratio mass spectrometry the specific 15N enrichment of amino acids in casein and plasma proteins in cows receiving three successive daily oral doses (300, 150 and 150 g) of (15NH4)2SO4 (10 atom per cent isotopic enrichment) and to examine the 15N enrichments obtained with regard to nitrogen transport and metabolism in the lactating cow. To investigate the 15N distribution in amino acids in casein and in plasma proteins, samples of 15N-labelled casein and plasma proteins were extracted either from a pool of several milkings (36-96 h after starting to administer the tracer) or from pooled venous blood (removed on the fourth day after the start of administration) from the four lactating cows. 15N enrichments of the proteins studied, expressed as atoms percent excess, were 0.2509 for casein and 0.0577 for plasma protein. Chromatographic fractionation of the amino acid mixture (protein hydrolysates) resulted in nine groups containing between one and four amino acids: Asp, Ser and Thr; Glu; Pro; Gly; Ala; Val and Met; Ileu and Leu; Tyr; Phe; His and Lys; and Arg. High 15N incorporation was demonstrated in all individual or groups of amino acids studied. In both proteins, Glu appeared to be the most enriched amino acid, Phe and Arg the least enriched. Most aliphatic molecules with a single amino group were highly enriched. The much lower (3.5-7.7-fold) enrichments in plasma protein compared with casein suggest considerable intracellular dilution at the site of liver protein synthesis. Finally, the amino acid separation methods are discussed and suggestions for improving them considered.

Amino Acids↗
PubMed1997

Effectiveness of a skin cancer control educational intervention for internal medicine housestaff and attending physicians.

OBJECTIVES: To evaluate the effects of a brief educational program on beliefs, knowledge, and behaviors related to skin cancer control among internal medicine housestaff and attending physicians. DESIGN: Randomised controlled trial. SETTING: Urban academic general medicine practice. PARTICIPANTS: Internal medicine housestaff and attending physicians with continuity clinics at the practice site. INTERVENTION: Two 1-hour educational seminars on skin cancer control conducted jointly by a general internist and a dermatologist. MEASUREMENTS AND MAIN RESULTS: Self-reported attitudes and beliefs about skin cancer control, ability to identify and make treatment decisions on 18 skin lesions, and knowledge of skin cancer risk factors were measured by a questionnaire before and after the teaching intervention. Exit surveys of patients at moderate to high risk of skin cancer were conducted 1 month before and 1 month after the intervention to measure physician skin cancer control practices reported by patients. Eighty-two physicians completed baseline questionnaires and were enrolled in the study, 46 in the intervention group and 36 in the control group. Twenty-five physicians attended both sessions, 11 attended one, and 10 attended neither. Postintervention, the percentage of physicians feeling adequately trained increased from 35% to 47% in the control group (p = .34) and from 37% to 57% in the intervention group (p = .06). Intervention physicians had an absolute mean improvement in their risk factor identification score of 6.7%, while control physicians' mean score was unchanged (p = .06). Intervention and control physicians had similar increases in their postintervention lesion identification and management scores. Postintervention, the mean proportion of patients per physician stating they were advised to watch their moles increased more among intervention physicians than control physicians (absolute difference of 19% vs -8%, p = .04). Other changes in behavior were not significant. CONCLUSIONS: Although we observed a few modest intervention effects, overall this brief skin cancer education intervention did not significantly affect primary care physicians' skin cancer control attitudes, beliefs, knowledge, or behaviors. A more intensive intervention with greater participation may be necessary to show a stronger impact on attitudes and knowledge about skin cancer control among primary care physicians.

Clinical Competence↗
PubMed1996

Effect of lower extremity blood pressure on physical functioning in patients who have intermittent claudication. The Chicago Claudication Outcomes Research Group.

PURPOSE: Claudication patients' perceptions of walking impairment often influence recommendations for peripheral bypass and angioplasty. The actual relationship between lower extremity blood flow and physical functioning, however, has rarely been explicitly studied. METHODS: Patients were enrolled at a visit to one of 16 vascular surgery offices and clinics that participated in a prospective outcomes study. A total of 555 patients (445 men and 110 women) with an abnormal ankle-brachial index (ABI), none of whom had had previous leg revascularization or symptoms of rest pain, skin ulcers, or gangrene, completed the SF36 Health Survey and the Peripheral Arterial Disease Walking Impairment Questionnaire (WIQ). Stepwise multiple regression analysis was used to test the statistical significance and strength of association between patients' ABI level and SF36 physical functioning (PF) and WIQ community walking distance scores, controlled for sociodemographic characteristics and the presence and severity of comorbid conditions. RESULTS: Univariate correlations with ABI were modest but significant (PF score, r = 0.12, p = 0.004; WIQ distance score, r = 0.18, p < 0.001). ABI was a very significant predictor of both PF (b = 18.8; p = 0.001) and WIQ scores (b = 0.33; p < 0.0001) in the multiple regression analysis. Other positive predictors of PF scores were high-school graduation and male sex. Negative predictors of PF scores were heart, lung, and cerebrovascular disease; knee arthritis and chronic back pain; and enrollment at a Veterans Administration clinic rather than a private community or academic office. CONCLUSION: Cross-sectional findings indicate that a 0.3 improvement in ABI is associated with an average improvement of 5.6% in PF or 10.3% in WIQ distance score. However, proper selection of individual candidates for interventional therapy, that is, those patients who have lower ABIs, lower initial functioning, and fewer disabling comorbidities would be predicted to produce a much greater functional benefit. Surgeons should make a rigorous functional evaluation when recommending interventional management of claudication.

Aged↗
PubMed1996

Assessing functional status of elderly adults via microcomputer.

Geriatric adults (at least 65 years of age) completed the Functional Status Questionnaire twice, first using the original paper-and-pencil format and again using one of six computer-based interactive methods. Scoring via original and computer-based methods was identical. Satisfaction with computers was high. Of the interactive methods, touchscreen required least training and questionnaire completion time and was rated most useful.

Activities of Daily Living↗
PubMed1995

Epidemiology of Norwalk virus during an outbreak of acute gastroenteritis aboard a US aircraft carrier.

A large outbreak of acute gastroenteritis occurred over a 5-week period aboard an aircraft carrier. The estimated cumulative attack rate was 13% among the 4,500-man crew. Eight percent of the crew sought medical attention, nearly all of whom missed 1 day or more of work. The risk of developing illness was 2 to 3 times greater for individuals living in more crowded sleeping quarters (> 50 persons per compartment). Occurrence of gastroenteritis was associated with a fourfold or more rise in Norwalk virus antibody levels, as measured by an enzyme-linked immunoassay utilizing a baculovirus expressed recombinant antigen. In addition, 27 nm Norwalk virus-like particles were visualized in two of six stools examined by immune electron microscopy. The presence of a low (< 1:50) or a high (> or = 1:6,400) pre-illness antibody level was associated with a lower incidence of illness. This investigation indicates that Norwalk virus can adversely impact operations of a military vessel and that crowding is a major risk factor in transmission.

Acute Disease↗
PubMed1995

Alcohol consumption among HIV-infected patients.

This prospective, cohort study analyzed the prevalence of alcoholism and patterns of alcohol intake over time in a cohort of HIV-infected patients, predominantly homosexual/bisexual men. One hundred eleven HIV-positive subjects were recruited from a comprehensive HIV clinic associated with a large Midwestern university hospital. Each participant completed the Michigan Alcoholism Screening Test (MAST) survey and a standardized quantity-frequency questionnaire on alcohol intake at enrollment. The quantity-frequency scale was repeated every six months for a total of 30 months. Forty-five of the 111 subjects (41%) met the criteria for alcoholism, as defined by a MAST score 5 or higher. There was a significant decrease in alcohol consumption over time, from 6.4 drinks/week in the initial time period to 3.9 drinks/week by the final time period (p < 0.001).

Alcohol Drinking↗
PubMed1995

Skin cancer control practices among physicians in a university general medicine practice.

Physician counseling about sun protection and routine screening for skin cancer in high-risk individuals have been widely recommended. The purpose of this study was to assess the skin cancer control practices and knowledge among physicians in a university-based general medicine practice. Fifty-two physicians completed a survey on attitudes toward, behaviors in, and knowledge of skin cancer control. In addition, the ability of general medicine residents and attending physicians to correctly identify and make biopsy recommendations for ten photographed skin lesions was compared with that of third-year medical students and dermatology residents and attendings. The results of the survey illustrate a need for improving primary care physicians' knowledge and identification of skin cancer risk factors, and increasing the frequency and consistency with which they perform skin cancer prevention counseling and complete skin examination in high-risk patient groups.

Counseling↗
PubMed1995

Application of multivariable optimal discriminant analysis in general internal medicine.

OBJECTIVE: To illustrate the use of multivariable optimal discriminant analysis (MultiODA). DESIGN: Data from four previously published studies were reanalyzed using MultiODA. The original analysis was Fisher's linear discriminant analysis (FLDA) for two studies and logistic regression analysis (LRA) for two studies. MEASUREMENTS AND MAIN RESULTS: In Study 1, FLDA achieved an overall percentage accuracy in classification (PAC) for the training sample of 69.9%, compared with 73.5% for MultiODA. In Study 2, the LRA model required three attributes to achieve a 76.1% overall PAC for the training sample and a 79.4% overall PAC for the hold-out sample. Using only two attributes, the MultiODA model achieved similar values. In Study 3, the FLDA model achieved an overall PAC of 82.5%, compared with 87.5% for the MultiODA model. In Study 4, MultiODA identified a two-attribute model that achieved a 93.3% overall training PAC, when an LRA model could not be developed. CONCLUSIONS: MultiODA identified: a superior training model (Study 1); a more parsimonious model that achieved superior overall training and identical hold-out PAC (Study 2); a model that achieved a higher hold-out PAC (Study 3); and a two-attribute model that achieved a relatively high PAC when a multivariable LRA model could not be obtained (Study 4). These findings suggest that MultiODA has the potential to improve the accuracy of predictions made in general internal medicine research.

Acquired Immunodeficiency Syndrome↗
PubMed1995

Late survival risk factors for abdominal aortic aneurysm repair: experience from fourteen Department of Veterans Affairs hospitals.

BACKGROUND: This study evaluates late survival risk factors for patients who underwent elective abdominal aortic aneurysm surgical procedures performed at 14 Department of Veterans Affairs hospitals across the United States between 1985 and 1987. METHODS: Preoperative risk factors for a representative sample of 280 male veterans were obtained from an extensive Department of Veterans Affairs Office of Quality Management study and subsequent chart review. The National Death Index was used to determine survival through December 1991. RESULTS: Mortality at 30 days was 2.9%. Kaplan-Meier survival probabilities were 89% (+/- 2%) at 1 year and 64% (+/- 3%) at 5 years. Multivariate hazards models indicated significantly poorer survival for patients with age greater than 69 years, chronic obstructive pulmonary disease, cerebrovascular disease, and left ventricular hypertrophy. A history of coronary artery disease including previous myocardial infarction or bypass operation did not predict late survival for this cohort. CONCLUSIONS: Given the substantial burden of comorbidity of veterans who use Department of Veterans Affairs facilities, the overall survival experience of this all male cohort compares well with previously published series and with overall U.S. male life expectancy. The fact that a history of coronary artery disease did not predict survival for this cohort may be related to selection bias; however, a more likely explanation is the presence of unsuspected coronary disease among patients without a documented history of angina or myocardial infarction.

Age Factors↗
PubMed1995

Cross-sectional psychometric assessment of the Functional Status Questionnaire: use with geriatric versus nongeriatric ambulatory medical patients.

OBJECTIVE: The Beth Israel/UCLA Functional Status Questionnaire (FSQ) is a multidimensional self-report instrument used for assessing the physical, social, and psychological status of children and adults. This study assessed the appropriateness of the FSQ for use with geriatric (at least 65 years of age) ambulatory medical patients. METHOD: Concurrent cohort convenience samples of forty geriatric and eighty-five nongeriatric ambulatory patients were drawn from a general internal medicine clinic. Patients completed the FSQ while waiting to see their physician. Analyses were conducted in order to: 1) address the magnitude and pattern of missing responses that are generated when the FSQ is administered to small groups; 2) assess whether, as was intended, each of the different FSQ subscales is unidimensional; 3) evaluate the magnitude and pattern of variances and covariances of items constituting FSQ subscales; 4) compare profiles of FSQ scores between geriatric versus nongeriatric samples; and 5) investigate whether FSQ subscales correlate with different self-report outcome measures in a convergent or divergent manner. RESULTS: For both samples, FSQ subscales were internally consistent and moderately variable. Cross-sectional comparisons revealed both convergent and divergent relationships between FSQ subscales and measures of social support and satisfaction with health. CONCLUSIONS: The FSQ appears to be appropriate for use with geriatric ambulatory medical patients. Findings highlighted the positive influence of intermediate activities of daily life--involving moderate levels of physical activity--on satisfaction with health, and the negative effect of depression on physical activity.

Activities of Daily Living↗
PubMed1994

Heart rate variability and susceptibility for sudden cardiac death: an example of multivariable optimal discriminant analysis.

The statistical classification problem motivates the search for an analytical procedure capable of classifying observations accurately into one of two or more groups on the basis of information with respect to one or more attributes, and constitutes a fundamental challenge for all scientific disciplines. Although there are many classification methodologies, only optimal discriminant analysis (ODA) explicitly guarantees that the discriminant classifier will maximize classification accuracy in the training sample. This paper presents the first example of multivariable ODA (MultiODA) in medicine, for an application in which we employ three attributes (age and two measures of heart rate variability) to predict susceptibility to sudden cardiac death for a sample of 45 patients. MultiODA outperformed logistic regression analysis on every classification performance index (overall accuracy, sensitivity, specificity, and positive and negative predictive values). In fact, the worst performance result achieved by MultiODA (in total sample or leave-one-out validity analysis) exceeded the best performance achieved by logistic regression analysis. We conclude that ODA offers promise as a methodology capable of improving the classification performance achieved by medical researchers, and that clearly merits investigation in future research.

Arrhythmias, Cardiac↗