Sample selection may bias the outcome of an adolescent mental health survey: results from a five-year follow-up of 4171 adolescents.

OBJECTIVES: The representativeness of the data is one of the main issues in evaluating the significance of research findings. Dropping out is common in adolescent mental health research, and may distort the results. Nevertheless, very little is known about the types of systematic bias that may affect studies in a) the informed consent phase and b) later in follow-up phases. STUDY DESIGN: The authors addressed this gap in knowledge in a five-year follow-up study on a sample of adolescents aged 13-18 years. METHODS: The data were collected using self-report questionnaires. The baseline sample consisted of 4171 adolescents, 1827 (43.8%) of whom gave consent to be contacted for a follow-up survey, but only 797 (19.1%) participated in the follow-up. Binary logistic regression models were used to explain the participation. RESULTS: Young age, female gender, a high number of hobbies, good performance at school in the native language and general subjects, family disintegration such as divorce, high parental employment, and symptoms of depression and anxiety were associated with both consent and participation. However, the effect of mental health aspects was smaller than the effect of age and gender. CONCLUSIONS: This study confirmed the possibility of systematic selection bias by adolescents' sociodemographic characteristics. The representativeness of the study sample might have been improved by more intense recruitment strategies.

Fear of childbirth in nulliparous and multiparous women: a population-based analysis of all singleton births in Finland in 1997-2010.

OBJECTIVE: To identify risk factors for fear of childbirth (FOC) according to parity and socioeconomic status, and to evaluate associations between FOC and adverse perinatal outcomes. DESIGN: A cohort study. SETTING: The Finnish Medical Birth Register. POPULATION: All 788 317 singleton births during 1997-2010 in Finland. METHODS: Fear of childbirth was defined according to the International Classification of Diseases code O99.80, and its associations with several risk factors and perinatal outcomes were analysed by multivariable logistic regression. MAIN OUTCOME MEASURES: Prevalence of, risk factors for and outcomes of FOC. RESULTS: Fear of childbirth was experienced by 2.5% of nulliparous women and 4.5% of multiparous women. The strongest risk factors for FOC in nulliparous women were depression [adjusted odds ratio (aOR), 6.35; 95% confidence interval (CI), 5.25-7.68], advanced maternal age (aOR, 3.78; 95% CI, 3.23-4.42) and high or unspecified socioeconomic status. In multiparous women, the strongest risk factors for FOC were depression (aOR, 5.47; 95% CI, 4.67-6.41), previous caesarean section (CS) (aOR, 3.02; 95% CI, 2.93-3.11) and high or unspecified socioeconomic status. Among both nulliparous and multiparous women, FOC was associated with higher rates of CS (3.3-fold and 4.5-fold higher, respectively) and a lower incidence of low birthweight (<2500 g), small for gestational age babies, preterm birth and low Apgar scores at 1 minute. CONCLUSIONS: High and unspecified socioeconomic status, advanced maternal age and depression are predisposing factors for FOC regardless of parity. Among multiparous women, a previous CS increases vulnerability to FOC. FOC is associated with increased rates of CS, but does not adversely affect other pregnancy outcomes.

The effect of long-term life dissatisfaction on health-related quality of life among general population subjects.

The aim of this population-based study (n = 329) was to explore how long-term life dissatisfaction (LS burden) and concurrent life dissatisfaction are associated with the concurrent health-related quality of life (HRQL) (RAND-36), and how long-term life dissatisfaction predicts HRQL in the general population. The sum of the life satisfaction scores in 1998, 1999, 2001 (LS burden) and the concurrent life satisfaction score (LS) in 2005 were used to categorize the study participants into satisfied, intermediate and dissatisfied groups. Differences in RAND-36 dimensions in 2005 were investigated with respect to the LS burden and concurrent life dissatisfaction. The predictive power of the LS burden for HRQL dimensions was assessed with logistic regression models. Both a high LS burden and concurrent life dissatisfaction were strongly associated with HRQL and were risk factors for poor HRQL, regardless of its dimensions. The LS burden predicted all of the RAND-36 dimensions, except for physical functioning. Screening of life dissatisfaction can be used to identify service users whose HRQL should be further investigated. Assessment of HRQL provides information on the domains and factors that require mental health nursing intervention. This knowledge could assist mental health nurses in both the alleviation of disease consequences and promotion of well-being of service users.

How does lifestyle intervention affect depressive symptoms? Results from the Finnish Diabetes Prevention Study.

AIMS: To assess the effect of lifestyle intervention on depressive symptoms during a 36-month randomized clinical trial designed to prevent Type 2 diabetes. METHODS: A total of 522 middle-aged participants, who were overweight or obese and had impaired glucose tolerance, were randomized to the lifestyle intervention or control group in the Finnish Diabetes Prevention Study. The intervention group received individualized counselling aimed at reducing weight and increasing physical activity. Depressive symptoms were measured using the Beck Depression Inventory among a subgroup of 140 participants. RESULTS: On study entry, the mean Beck Depression Inventory scores ± sd were 6.8 ± 5.6 in the intervention group and 6.7 ± 5.5 in the control group. Beck Depression Inventory scores reduced during the intervention study: the mean ± sd (95% CI) reduction was 0.90 ± 4.54 (-1.99 to -0.19) scores in the intervention group and 0.75 ± 4.47 (-1.80 to 0.31) in the control group, with no difference between the groups. In a stepwise linear multivariate regression analysis, the variables with the strongest associations with the change in Beck Depression Inventory scores were baseline Beck Depression Inventory scores, marital status, weight change and change of total energy intake (R(2) = 0.209, P < 0.001). CONCLUSIONS: Participation in the study lowered depression scores, with no specific group effect. Among the lifestyle changes, particularly successful reduction of body weight was associated with the greater reduction of depressive symptoms. Thus, regardless of the intensity of the treatment, the success in executing alterations in one's lifestyle and behaviour is associated with beneficial changes in mood.

Serum adiponectin and resistin levels in major depressive disorder.

OBJECTIVE: To examine the role of the adipose-tissue-derived low-grade inflammation markers adiponectin and resistin in major depressive disorder (MDD) in a population-based sample. METHOD: Serum levels of adiponectin and resistin were measured from 70 DSM-IV MDD subjects and 70 healthy controls. Depression severity was assessed with the 29-item Hamilton Depression Rating Scale. RESULTS: The MDD group had lowered serum adiponectin levels. Regression modelling with adjustments for age, gender, overweight, several socioeconomic and lifestyle factors, coronary heart disease and metabolic syndrome showed that each 5.0 microg/ml decrease in serum adiponectin increased the likelihood of MDD by approximately 20% (P = 0.01). The resistin levels correlated with atypical (P = 0.02), but not with typical depressive symptoms (P = 0.12). CONCLUSION: Our findings suggest that the lowered adiponectin levels in MDD are depression-specific and not explained by conventional low adiponectin-related factors such as such as coronary heart disease and metabolic disorders.

Association of depressive symptoms and metabolic syndrome in men.

OBJECTIVE: To explore the relationship between several indicators of depression and metabolic syndrome (MetS). METHOD: A population-based sample with high (HMS group) or low (LMS group) levels of mental symptoms, including those of depression, in three follow-ups participated in a clinical examination in 2005 (n = 223). MetS was determined according to the NCEP criteria. RESULTS: The prevalence of MetS was 49% in men and 21% in women. Men with MetS had higher rates of major depressive disorder than other men. They also displayed higher Hamilton Rating Scale for Depression (HDRS) scores and more often signs of suicidality. In logistic regression analyses, higher HDRS scores (OR 1.31, 95% CI 1.04-1.64) and belonging to the HMS group (OR 10.1, 95% CI 1.98-51.3) were independent associates for MetS but only in men. CONCLUSION: The results highlight that there is an association between long-term depressive symptoms and the emergence of MetS, especially in men.