Search Results

You are looking at 1 - 4 of 4 items for

  • Author or Editor: George B. Ploubidis x
Clear All Modify Search

Valid inference from the investigation of mental health relies – among others – on the assumption of no measurement error. However, it is well known that data from self-reported measures are likely to be biased by some process that is driven by the respondent’s personality and/or circumstances. We capitalised on data available in two nationally representative birth cohorts, the National Child Development Study (1958 birth cohort) and the 1970 British Cohort Study to formally test the longitudinal measurement equivalence of the nine-item version of the Malaise Inventory, a measure of psychological distress. The inclusion of identical assessments of mental health in adulthood in both cohorts allowed us to evaluate their measurement properties and investigate whether the passage of time has differentially affected the interpretation of mental health assessments. To do so, we employed methods within the generalised latent variable measurement modelling framework and related extensions for formally testing measurement invariance. We found that the passage of two decades and more in both cohorts have not influenced how participants respond to the short version of the Malaise Inventory. The observed scalar invariance of the short version of the Malaise Inventory implies that potential sources of bias such as age effects, survey design, period effects, or cohort specific effects did not influence the way participants in the two cohorts respond to the symptoms described in the Malaise Inventory. Our results offer some reassurance for the extent to which self-reported mental health survey questions are affected by systematic sources of error.

Restricted access

Background:

Life course trajectories of affective symptoms (depression and anxiety) are heterogenous. However, few studies have investigated the role of early life risk factors in the development of these trajectories. The present study aimed to: (1) derive latent trajectories of affective symptoms over a period of more than 50 years (ages 13–69), and (2) examine early life risk factors for associations with specific life course trajectories of affective symptoms.

Method:

Participants are from the MRC National Survey of Health and Development (NSHD) (n = 5,362). Affective symptoms were measured prospectively at ages 13, 15, 36, 43, 53, 60–64 and 69. A latent variable modelling framework was implemented to model longitudinal profiles of affective symptoms. Twenty-four prospectively measured early life predictors were tested for associations with different symptom profiles using multinomial logistic regression.

Results:

Four life course profiles of affective symptoms were identified: (1) absence of symptoms (66.6% of the sample); (2) adolescent symptoms with good adult outcome (15.2%); (3) adult symptoms only (with no symptoms in adolescence and late life) (12.9%); (4) symptoms in adolescence and mid adulthood (5.2%). Of the 24 early life predictors observed, only four were associated with life course trajectories, with small effect sizes observed.

Conclusions:

People differ in their life course trajectories of anxiety and depression symptoms and that these differences are not largely influenced by early life factors tested in this study.

Full Access

Non-response is common in longitudinal surveys, reducing efficiency and introducing the potential for bias. Principled methods, such as multiple imputation, are generally required to obtain unbiased estimates in surveys subject to missingness which is not completely at random. The inclusion of predictors of non-response in such methods, for example as auxiliary variables in multiple imputation, can help improve the plausibility of the missing at random assumption underlying these methods and hence reduce bias. We present a systematic data-driven approach used to identify predictors of non-response at Wave 8 (age 25–26) of Next Steps, a UK national cohort study that follows a sample of 15,770 young people from age 13–14 years. The identified predictors of non-response were across a number of broad categories, including personal characteristics, schooling and behaviour in school, activities and behaviour outside of school, mental health and well-being, socio-economic status, and practicalities around contact and survey completion. We found that including these predictors of non-response as auxiliary variables in multiple imputation analyses allowed us to restore sample representativeness in several different settings, though we acknowledge that this is unlikely to universally be the case. We propose that these variables are considered for inclusion in future analyses using principled methods to explore and attempt to reduce bias due to non-response in Next Steps. Our data-driven approach to this issue could also be used as a model for investigations in other longitudinal studies.

Open access