SEARCH · Search PubMed

Results for “longitudinal”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 253 records · Page 14Linked to original sources

PubMedDate not supplied

A multiple imputation strategy for incomplete longitudinal data.

Longitudinal studies are commonly used to study processes of change. Because data are collected over time, missing data are pervasive in longitudinal studies, and complete ascertainment of all variables is rare. In this paper a new imputation strategy for completing longitudinal data sets is proposed. The proposed methodology makes use of shrinkage estimators for pooling information across geographic entities, and of model averaging for pooling predictions across different statistical models. Bayes factors are used to compute weights (probabilities) for a set of models considered to be reasonable for at least some of the units for which imputations must be produced, imputations are produced by draws from the predictive distributions of the missing data, and multiple imputations are used to better reflect selected sources of uncertainty in the imputation process. The imputation strategy is developed within the context of an application to completing incomplete longitudinal variables in the so-called Area Resource File. The proposed procedure is compared with several other imputation procedures in terms of inferences derived with the imputations, and the proposed methodology is demonstrated to provide valid estimates of model parameters when the completed data are analysed. Extensions to other missing data problems in longitudinal studies are straightforward so long as the missing data mechanism can be assumed to be ignorable.

Bayes Theorem↗

PubMed1995

Topography and distribution of nerve fibers in the posterior longitudinal ligament of the rat: an immunocytochemical and electron-microscopical study.

The distribution and immunocytochemical characterization of nerve fibers and their terminals in the posterior longitudinal ligament of the rat lumbar vertebral column was studied in whole-mount preparations and serial semithin and ultrathin sections. Differences in the localization, distribution pattern and density of peptidergic and catecholaminergic nerve fibers were found in the vertebral and intervertebral regions of the posterior longitudinal ligament. For immunocytochemistry, free floating specimens were incubated with primary antibodies against protein gene product 9.5, substance P, calcitonin gene-related peptide, dopamine-beta-hydroxylase, vasoactive intestinal polypeptide and neuropeptide Y together with the avidin-biotin-peroxidase method. In whole-mount preparations, the neural marker protein gene product 9.5 is immunostained in all unmyelinated nerve fibers in the posterior longitudinal ligament, thus giving a panoramic view of the nerve fiber plexus. The most striking nerve fiber plexus is localized in the intervertebral region. In this region, the posterior longitudinal ligament is rich in capillaries that form a dense plexus within its ventral part and extend to the outer layer of the annulus fibrosus. The peptidergic and catecholaminergic innervation of the posterior longitudinal ligament is discussed in the context of pain syndromes related to the vertebral column and degenerative lumbar spine diseases.

Animals↗

PubMed2002

Changes in bone mineral density with age in men and women: a longitudinal study.

We performed a prospective study to evaluate the normal changes in bone mineral density (BMD) in the forearm, hip, spine and total body, and to study the agreement between changes in BMD estimated from cross-sectional data and the actual longitudinal changes. Six hundred and twenty subjects (398 women, 222 men; age 20-89 years) without diseases or medication known to affect bone metabolism undertook baseline evaluations, and 525 (336 women, 189 men) completed the study. BMD was measured twice 2 years apart by dual-energy X-ray absorptiometry. From cross-sectional evaluations the only premenopausal bone loss (<0.003 g/cm2/year) was found in the hip. In women after menopause and in men an age-related bone loss (0.002-0.006 g/cm2/year) was found at all sites. The data from the longitudinal evaluation showed a small bone loss in women before menopause at the hip and lumbar spine (<0.4%/year (<0.004 g/cm2/year)); this bone loss nearly tripled in the early postmenopausal years (<10 years since menopause), and thereafter decreased to the premenopausal rate for the hip, and to zero for the lumbar spine. The most pronounced bone loss after menopause occurred in the forearm (1.2%/year (0.006 g/ cm2/year)), and it remained constant throughout life. In men there was a small longitudinal bone loss in the hip throughout life, and a small bone loss in the distal forearm after the age of 50 years. In all groups, except for the early postmenopausal women, we found a small increase in total body BMD with age. When comparing the changes in BMD estimated from cross-sectional data with the longitudinal changes, only the hip and forearm generally displayed agreement, whereas the changes in the total body and spine generally were incongruous. In conclusion, the hip and forearm appear to be the sites with the best agreement between the cross-sectional estimated and the longitudinal age-related changes in BMD.

Adult↗

PubMed2007

Longitudinal in vivo reproducibility of cartilage volume and surface in osteoarthritis of the knee.

OBJECTIVE: The aim of this study was to evaluate the longitudinal reproducibility of cartilage volume and surface area measurements in moderate osteoarthritis (OA) of the knee. MATERIALS AND METHODS: We analysed 5 MRI (GE 1.5T, sagittal 3D SPGR) data sets of patients with osteoarthritis (OA) of the knee (Kellgren Lawrence grade I-II). Two scans were performed: one baseline scan and one follow-up scan 3 months later (96 +/- 10 days). For segmentation, 3D Slicer 2.5 software was used. Two segmentations were performed by two readers independently who were blinded to the scan dates. Tibial and femoral cartilage volume and surface were determined. Longitudinal and cross-sectional precision errors were calculated using the standard deviation (SD) and coefficient of variation (CV%=100x[SD/mean]) from the repeated measurements in each patient. The in vivo reproducibility was then calculated as the root mean square of these individual reproducibility errors. RESULTS: The cross-sectional root mean squared coefficient of variation (RMSE-CV) was 1.2, 2.2 and 2.4% for surface area measurements (femur, medial and lateral tibia respectively) and 1.4, 1.8 and 1.3% for the corresponding cartilage volumes. Longitudinal RMSE-CV was 3.3, 3.1 and 3.7% for the surface area measurements (femur, medial and lateral tibia respectively) and 2.3, 3.3 and 2.4% for femur, medial and lateral tibia cartilage volumes. CONCLUSION: The longitudinal in vivo reproducibility of cartilage surface and volume measurements in the knee using this segmentation method is excellent. To the best of our knowledge we measured, for the first time, the longitudinal reproducibility of cartilage volume and surface area in participants with mild to moderate OA.

Adult↗

PubMed1994

Ten-year change in the use of medical drugs among the elderly--a longitudinal study and cohort comparison.

Longitudinal changes and cohort differences from 1979 to 1989 in the use of prescribed and non-prescribed medical drugs were examined and the connections of drug use to various background variables were analysed in the framework of the Tampere Longitudinal Study of Ageing. In the longitudinal study random samples of men and women born in 1900-09 and 1910-19 (738 persons) were interviewed in 1979 and the survivors (62%) were interviewed again in 1989. In the cohort comparison the 60-69-year-olds studied in 1979 (364 persons) were compared with the 60-69-year-olds in 1989 (395 persons). In the longitudinal setting the number of prescribed drugs increased. In 1989 2 or 3 in 10 persons used at least five prescribed drugs simultaneously. In the cohort comparison there were no differences in the number of prescribed drugs between the two groups of 60-69-year-olds. Use was connected with multimorbidity and poor self-rated health. Both the longitudinal comparison in the group born in 1910-19 and the cohort comparison revealed a significant increase in the use of non-prescribed drugs. The use of analgesics was connected with sex, occupational class, self-rated health and feelings of loneliness. In the use of vitamins, no connection was found with health variables, but they were mostly used by white-collar employees and women. However, when all these variables were controlled for, both analgesics and, in particular, vitamins were used more in 1989 than in 1979. The results suggest that the increase in the use of non-prescribed drugs is mainly due to social and historical factors, by changes both in drug policies, health culture and health behaviour of elderly persons.

Age Factors↗

PubMed2006

Estimating dietary consumption patterns among children: a comparison between cross-sectional and longitudinal study designs.

Estimating dietary intake for children is an essential component of conducting pesticide exposure assessments given the fact that children are predominantly exposed to certain pesticides, such as organophosphorus pesticide, through dietary intake. Different study designs and their respective sampling methodology utilized to estimate food consumption patterns can significantly alter the parameter estimates and the variability in the values obtained. This study investigated the impacts of study design on overall estimates of dietary intake by applying the temporal sampling characteristics used in cross-sectional approaches, as in The Continuing Survey of Food for Intakes by Individuals (CSFII), to food consumption data collected in a longitudinal manner via a bootstrap sampling technique. We examined the precision of time-averaged dietary intake estimates under various sampling schemes and explored the contribution of seasonality toward the dietary patterns. A comparison between the estimates of food consumption obtained from the bootstrap replicates and the longitudinal study estimates indicate that variability is significantly decreased when employing a longitudinal study design. Moreover, both between and within-subject variability decreases when individuals are followed over an increasing number of days. Finally, within the longitudinal study cohort, we observed a seasonal component to dietary intake for fruits and grains. Our findings suggest that longitudinal dietary surveys offer substantial improvements for exposure assessment compared to a standard cross-sectional design.

Child↗

PubMed2025

Association Between Ratio of Triglycerides to HDL-C and Cognitive Impairment: A Longitudinal Population-Based Analysis and Mendelian Randomization Study.

OBJECTIVES: This study aimed to explore the longitudinal association between the triglyceride to high-density lipoprotein cholesterol (TG/HDL-C) ratio and cognitive impairment in older adults and further assess potential causality using Mendelian randomization (MR). DESIGN: Longitudinal population-based analysis combined with two-sample MR. Data from Waves 6-8 (2015-2019) of the Survey of Health, Ageing, and Retirement in Europe (SHARE) were analyzed using hierarchical regression models and mixed linear effects models. MR utilized genome-wide association studies (GWAS) summary data. PARTICIPANTS: 11,444 adults aged ≥60 years from SHARE Wave 6, with longitudinal follow-up in Waves 7 (N = 2,775) and 8 (N = 5,469). MEASUREMENTS: TG/HDL-C ratio, cognitive function (orientation, immediate/delayed recall, verbal fluency, numeracy), and cognitive impairment (defined as scores >1.5 SD below age-group mean). Covariates included socio-demographics, health behaviors, comorbidities, and national-level factors. MR employed genetic variants associated with TG/HDL-C as instrumental variables. RESULTS: Higher TG/HDL-C ratios were negatively associated with total cognitive scores (β = -0.115, χ² = 15.44, df = 1, p < 0.001), immediate recall (β = -0.029, χ² = 12.34, df = 1, p < 0.001), delayed recall (β = -0.028, χ² = 7.33, df = 1, p < 0.001), verbal fluency (β = -0.038, χ² = 11.53, df = 1, p < 0.001), and numeracy (β = -0.019, χ² = 4.55, df = 1, p < 0.05) in fully adjusted models. Longitudinal analysis revealed increased cognitive impairment risk in the highest TG/HDL-C quartile (OR=1.43, 95% CI:1.01-2.03, χ² = 9.24, df = 1) over 4 years. MR supported a causal link between elevated TG/HDL-C and cognitive decline. CONCLUSIONS: Elevated TG/HDL-C ratios are longitudinally associated with cognitive decline in older adults. Managing lipid metabolism may mitigate cognitive impairment, highlighting the importance of TG/HDL-C as a modifiable risk factor in aging populations.

Humans↗

PubMed2001

A longitudinal study of children's juice intake and growth: the juice controversy revisited.

OBJECTIVES: To determine associations between children's longitudinal juice intake and growth parameters at age 72 months and to determine children's beverage intake patterns over time. SUBJECTS/SETTING: White children (n = 72) residing in the vicinity of a Southern US city participated in a longitudinal study. METHODS: Seven in-home interviews were conducted per child when each child was between 24 and 72 months of age. The 7 sets of 3-day dietary data were analyzed for beverage intake, categorized as juice (100% juice only), milk, carbonated beverages, and other drinks (e.g., lemonade, tea, juice drinks). The following growth parameters were determined for each child at age 72 months: height, weight, body mass index (measured as kg/m2), and ponderal index (measured as kg/m3); the 3 latter are measures of overweight. STATISTICAL ANALYSES: Associations between longitudinal juice intake and each growth parameter were tested with general linear models. Repeated measures analysis of variance tested changes in beverage intakes over time. RESULTS: There were no statistically significant associations between juice and children's height, weight, or body mass index, as tested by general linear models. Children's longitudinal juice intake was negatively related to ponderal index (beta = 0.065, P = .050). Children's juice intake decreased significantly between ages 2 and 6 years, from 6.8 to 3.6 oz/day (P = .0001); intakes of carbonated beverages and other drinks increased from 1.7 to 3.8 oz/day (P = .0016) and from 2.7 to 6.2 oz/day (P = .0001), respectively. APPLICATIONS/CONCLUSIONS: Children's longitudinal juice intake was not associated with either short stature or overweight. As juice consumption decreased, intakes of less nutritious beverages increased. Consumption of 100% juices should be encouraged by health professionals and parents/caregivers.

Analysis of Variance↗

PubMed2003

Imputation of missing longitudinal data: a comparison of methods.

BACKGROUND AND OBJECTIVES: Missing information is inevitable in longitudinal studies, and can result in biased estimates and a loss of power. One approach to this problem is to impute the missing data to yield a more complete data set. Our goal was to compare the performance of 14 methods of imputing missing data on depression, weight, cognitive functioning, and self-rated health in a longitudinal cohort of older adults. METHODS: We identified situations where a person had a known value following one or more missing values, and treated the known value as a "missing value." This "missing value" was imputed using each method and compared to the observed value. Methods were compared on the root mean square error, mean absolute deviation, bias, and relative variance of the estimates. RESULTS: Most imputation methods were biased toward estimating the "missing value" as too healthy, and most estimates had a variance that was too low. Imputed values based on a person's values before and after the "missing value" were superior to other methods, followed by imputations based on a person's values before the "missing value." Imputations that used no information specific to the person, such as using the sample mean, had the worst performance. CONCLUSIONS: We conclude that, in longitudinal studies where the overall trend is for worse health over time and where missing data can be assumed to be primarily related to worse health, missing data in a longitudinal sequence should be imputed from the available longitudinal data for that person.

Aged↗

PubMed2001

Near infra-red interactance for longitudinal assessment of nutrition in dialysis patients.

OBJECTIVE: Serial nutritional assessment of dialysis patients is important because of the high incidence and prevalence of malnutrition in these patients. Near-infrared interactance (NIR) technology may provide a practical and reliable method to evaluate body fat and its changes over time in dialysis patients. DESIGN: Longitudinal study consisting of 2 cross-sectional measurements, 2 months apart. SETTING: Outpatient dialysis unit affiliated to a tertiary care community medical center. PATIENTS: Seventy-one dialysis patients (35 men, 36 women), 57 +/- 15 years old, who have been on dialysis between 5 months and 11 years (43 +/- 30 months). Twelve additional patients with similar features were studied during the second round. INTERVENTION: None. MAIN OUTCOME MEASURES: NIR was used to estimate the body fat percentage. Other simultaneous measurements included subjective global assessment, anthropometric indices including midarm circumference, triceps and biceps skinfold thickness, and body mass index, and some laboratory values including albumin, transferrin, and cholesterol. NIR measurement was performed by placing a Futrex sensor on the nonaccess upper arm for several seconds, after logging the required individual data (sex, weight, height, and body frame), along with uniform physical activity levels for all patients, into a mini-computer. RESULTS: Seventy-one dialysis patients underwent nutritional and laboratory measurements. A second measurement round was performed 8 to 9 weeks after the first one and included 12 additional patients. Within each cross-sectional round, Pearson correlation coefficients (r) between the NIR score and nutritionally relevant variables were significant for anthropometric values (0.56 to 0.82) as well as low cholesterol and creatinine (0.22 to 0.30). The two serial NIR measurements on the same patients were highly consistent over the 2-month study interval (r = 0.96), whereas anthropometric values showed greater variability. The within-person coefficient of variation for NIR was low, indicating high consistency between 2 measurements. Moreover, the timing of the NIR measurement (predialysis v postdialysis) did not have any impact on consistency of the NIR results. The longitudinal changes of NIR had significant correlations with anthropometric and laboratory changes over time. CONCLUSION: The NIR, which can be performed within seconds, may serve as a reliable and practical tool for objective measurements of nutritional status in hemodialysis patients. The NIR not only seems to have a high degree of reproducibility but may also be an optimal tool to detect longitudinal changes in body fat over time. The NIR measurement is independent of the fluid status in dialysis patients. More comparative and longitudinal studies are needed to confirm the validity of NIR measurements in longitudinal evaluation of dialysis patients.

Adipose Tissue↗

PubMed1996

Longitudinal studies in organizational stress research: a review of the literature with reference to methodological issues.

Demonstrating causal relationships has been of particular importance in organizational stress research. Longitudinal studies are typically suggested to overcome problems of reversed causation and third variables (e.g., social desirability and negative affectivity). This article reviews the empirical longitudinal literature and discusses designs and statistical methods used in these studies. Forty-three longitudinal field reports on organizational stress were identified. Most of the investigations used a 2-wave panel design and a hierarchical multiple regression approach. Six studies with 3 and more waves were found. About 50% of the studies analyzed potential strain-stressor (reversed causation) relationships. In about 33% of the studies there was some evidence of reverse causation. The power of longitudinal studies to rule out third variable explanations was not realized in many studies. Procedures of how to analyze longitudinal data are suggested.

Humans↗

PubMedDate not supplied

Venturing a 30-year longitudinal study.

Longitudinal inquiry has long been recognized as a uniquely powerful method for seeking understanding of psychological development. A 30-year longitudinal venture is described--its theoretical motivation, methodological rationale, and details of implementation. Some of the novel and implicative findings the study has generated are briefly described. Common to all of the results is an absolute reliance on long-term, widely ranging, independent data. Although specific aspects of the study have appeared over the years, its intentions and scope are recounted only here. By and large, the organizing constructs of ego-control and ego-resiliency find impressive support in various empirical inquiries, here quickly described. Methodologically, a number of savvy research procedures useful and perhaps even necessary in longitudinal research are conveyed. The troublesome burdens but ever-alluring attractions of longitudinal inquiry are noted. A forthcoming Web site will contain the extensive 30-year longitudinal data bank together with explanatory information. Psychological investigators may find these imminently available data resources useful.

Ego↗

PubMed1999

Cross-sectional and longitudinal analyses of nutrition guidance by primary care physicians.

OBJECTIVE: To investigate in primary care physicians (PCPs) the determinants of a nutrition guidance practice ('noticing patients' overweight and guidance of treatment'), as well as their mechanism of action, in a cross-sectional and a longitudinal approach. DESIGN: Mixed longitudinal design. Five years follow up study of a previous cross-sectional study in October 1992. SUBJECTS: A representative sample of 675 Dutch PCPs, in practice for 5 up to 20y. INTERVENTIONS: A shortened version of the Wageningen PCPs Nutritional Practices Questionnaire was mailed to the subjects in August 1997. MAIN OUTCOME MEASURE: To obtain with the LISREL-program a model of the mechanism of action of determinants of the dependent variable 'noticing patients' overweight and guidance of treatment' with an adequate fit of the empirical data, both in the cross-sectional and in the longitudinal approach. RESULTS: The same set of predisposing factors and intermediary factors explains the dependent variable both in two different representative cross-sectional study populations of PCPs, and in a cohort cross-sectional study at two points in time. Two dynamic LISREL-models were developed (the 'determinant-longitudinal approach' and the 'early behaviour longitudinal approach') which explain the dependent variable. The latter model has, as added value, a gain in explained variance. In 5 y time, the dependent variable decreased significantly (P < 0.001). CONCLUSIONS: This study reconfirms that PCPs' nutritional guidance practices are determined partly directly by predisposing factors, and indirectly via driving forces and barriers. However this study also reveals that an important nutrition guidance practice of PCPs, 'noticing patients' overweight and guidance of treatment', shows a significant decrease over the last 5 y. At the same time, two of the four predisposing factors and two of the three driving factors also decreased significantly. As research findings indicate that the role of diet in health and disease becomes of greater influence PCPs need to be activated to apply their responsibility in this field within a multi-faceted approach.

Attitude of Health Personnel↗

PubMed2005

Information and informed consent in a longitudinal screening involving children: a questionnaire survey.

This empirical study explores participants' perceptions of information and understanding of their children's and their own involvement in a longitudinal screening, the ABIS Study. ABIS (All Babies In Southeast Sweden) is a multicentre, longitudinal research screening for Type 1 diabetes and multifactorial diseases involving 17 005 children and their families. For this study, a random selection of mothers was made, using perinatal questionnaire serial numbers from the ABIS study. In total, 293 of these mothers completed an anonymous questionnaire (response rate 73.3%). Our findings from the questionnaire indicate a marked difference between the reported satisfaction with and understanding of the information provided on the one hand and the significant lack of knowledge of some of the aims and methods of the ABIS screening on the other, namely concerning high-risk identification of involved children, potential prevention and future questionnaires. Two questions evoked by our results are: (1) what information is required for participants in longitudinal studies involving children? and (2) how do we ensure and sustain understanding, and thus in a prolonging, informed consent in these studies? This study underlines the importance of an increased understanding of the ethical issues that longitudinal research on children raise and the need to discuss how information and informed consent strategies should be analysed and designed in longitudinal studies.

Adult↗

PubMed1999

Lifestyle and obesity in adolescence and young adulthood: results from the Amsterdam Growth And Health Longitudinal Study (AGAHLS).

OBJECTIVE: To investigate if there is a longitudinal relationship between the development of fat mass in males and females (aged between 12-28 y) and their lifestyle, with respect to diet and physical activity. DESIGN: In the Amsterdam Growth And Health Longitudinal Study (AGAHLS), a group of 500 boys and girls are being followed from the age of 13 y (mean age 13.5 y), over a period of 20 y until the age of 32 y (mean age 32.5 y). SUBJECTS AND MEASUREMENTS: Data from AGAHLS are analysed of six repeated measurements of growth (body height, body mass, four skinfolds), health parameters with respect to cardiovascular disease (CVD) (obesity, hypertension, hypercholesterolaemia) and two important lifestyle tractors: physical activity (PA) (weighted energy output) and dietary intake (DI) (total energy intake, contribution of fat, carbohydrate and protein) of about 200 males and females between the ages of 13-27 y. RESULTS: The longitudinal results of PA show a steep decrease in the energy expenditure from the age of 13 y from about 4500 Mets per week to 3000 Mets per week at the age of 27 y in both sexes. The longitudinal results of DI also show a decrease in energy intake per kilogram body mass from about 225 kJ per day at the age of 13 y to about 155 kJ at the age of 27 y. During adolescence, boys show a 15% higher energy intake and a 20% higher energy expenditure than girls. At the age of 27 y, the difference between the sexes in energy intake is reduced to 10% and the difference in energy expenditure disappears. Results of tracking analyses over the period of 15y indicate a low stability coefficient of PA (0.34; 0.19-0.49), but higher coefficients of DI (0.55; 0.45-0.64) and fat mass (0.63; 0.56-0.71). The search for important lifestyle factors that can discriminate high- from low-risk participants for a high fat mass, estimated from the sum of four skinfolds and body mass (BM), resulted in an Odds Ratio (OR) of 1.5 (1.2-1.8) with daily intake of proteins and an OR of 0.81 (0.69-0.96) with PA. Surprisingly the fat mass is negatively related with the daily intake of energy per kg BM (OR: 0.37; 0.28-0.49). The longitudinal relation between fat mass and PA (corrected for DI) between the ages of 13-27 y, showed a significant inverse relationship (P<0.01) if fat mass was estimated from the sum of four skinfolds, but not if estimated from the body mass index (BMI). CONCLUSIONS: Over the adolescent and young adult period from 13-27 y, fat mass indicates a fairly good predictability. A high PA in both sexes is related with a low fat mass. Therefore promotion of habitual physical activity in the adolescent period seems effective in the early prevention of obesity.

Adolescent↗

PubMed1998

Use of cohorts with extensive longitudinal data in investigating the molecular genetics of asthma.

Recent studies have associated several chromosomal regions and specific polymorphisms with asthma, atopy and airway hyperresponsiveness. Most of these studies have been cross-sectional which substantially limits their power to analyse genotype/phenotype associations. Asthma is a such a highly variable condition that future genotype/phenotype studies should use longitudinal data. Populations with extensive long-term longitudinal data will facilitate more precise definition of phenotype and allow analysis of asthma's natural history. Longitudinal information also means that each phenotypic attribute can be quantified and considered with respect to age. Collecting longitudinal data prospectively would be extremely expensive and take many years. Using existing longitudinal data sets would clearly be much quicker and more economical, but there are very few suitable data sets to be found. Studying a normal and an asthmatic population is ideal and would provide complementary genotype/phenotype information. This approach is a logical step to using the genotypic information obtained from cross-sectional molecular genetic studies to more critically establish the effect of genotype on phenotype.

Asthma↗

PubMed2002

Longitudinal assessment of the development of diabetic polyneuropathy and associated risk factors.

AIMS: To longitudinally assess risk factors for diabetic polyneuropathy (DPN) severity, and to longitudinally assess risk factors for the change of DPN severity during 2-4 years of follow-up. METHODS: From 1995 to 1999, 486 Type 2 diabetes patients in general practice were examined annually with regard to DPN severity and its possible risk factors. DPN severity was assessed with a clinical neurological examination (CNE) which included pinprick sense, light touch sense, vibration sense and ankle jerk. Longitudinal (multivariate) linear associations of (change of) CNE score and predicting variables were analysed using multilevel analyses. RESULTS: In this population, 50% of participants were men and had a mean age of 65.4 years, almost one-third (31.7%) of the participants had a CNE score > 4 at baseline and were classified with DPN. CNE score significantly increased during follow-up. Among participants not graded with DPN at baseline, 21.3% progressed towards a CNE score > 4 after 3 years of follow-up. Longitudinal multivariate analyses showed that age, diabetes duration, HbA1c, height, body mass index and ankle-arm index together best predicted CNE score during follow-up. Change of CNE score during follow-up was best predicted by age, diabetes duration and HbA1c, with the latter being the strongest predictor. CONCLUSIONS: Although several factors are longitudinally associated with DPN, HbA1c, age and diabetes duration were the best predictors of CNE change during follow-up. Therefore, improving glycaemia remains an important amenable factor in preventing worsening of diabetic polyneuropathy.

Age of Onset↗

PubMed1997

Different statistical models to analyze epidemiological observational longitudinal data: an example from the Amsterdam Growth and Health Study.

With the development of new statistical techniques [such as generalized estimating equations (GEE)] it became possible to analyze longitudinal epidemiological relations, using all available longitudinal data. However, there are different possibilities in modeling longitudinal relations. In this paper four possible models were compared. (1) A simple model in which the actual values of the outcome and predictor variables were related (Y(it) = beta0 + beta1X(it)...); (2) A model with a time lag between outcome and predictor variables (Y(it) = beta0 + beta1X(it-1)...); (3) A model in which not the actual values, but changes in values between different time points were related ([Y(it)-Y(it-1)] = beta0 + beta1 [X(it)-X(it-1)]...); and (4) A first-order autoregressive model in which the actual value of the outcome variable at time point t is not only related to the actual value of the predictor variable at time point t, but also to the value of the outcome variable at t-1 (Y(it) = beta0 + beta1X(it) + beta2Y(it-1) +...). In this paper the use of the possible models was discussed by means of an example with data from the Amsterdam Growth and Health Study. In this longitudinal observational study six repeated measurements were carried out over a period of 15 years on subjects with an initial age of 13 years. It can be concluded that each model reflects different parts of the longitudinal relationships and the choice for a particular model must be based on logical considerations. However, in most cases epidemiologists should use the results of different models to obtain a more accurate answer to the particular epidemiological question.

Epidemiologic Methods↗