Identifying Longitudinal Patterns for Individuals and Subgroups: An Example with Adherence to Treatment for Obstructive Sleep Apnea

To improve complex behaviors such as adherence to medical recommendations, a better understanding of behavior change over time is needed. The focus of this study was adherence to treatment for obstructive sleep apnea (OSA). Adherence to the most common treatment for OSA is poor. This study involved a sample of 161 participants, each with approximately 180 nights of data. First, a time series analysis was performed for each individual. Time series parameters included the mean (average hours of use per night), level, slope, variance, and autocorrelation. Second, a dynamic cluster analysis was performed to find homogenous subgroups of individuals with similar adherence patterns. A four-cluster solution was found, and the subgroups were labeled: Great Users (17.2%; high mean and level, no slope), Good Users (32.8%; moderate mean and level, no slope), Low Users (22.7%; low mean and level, negative slope), and Slow Decliners (moderate mean and level, negative slope, high variance). Third, participants in the identified subgroups were compared to establish external validity. These steps represent a Typology of Temporal Patterns (TTP) approach. Combining time series analysis and dynamic cluster analysis is a useful way to evaluate longitudinal patterns at both the individual level and subgroup level.     

Reversible Verbal Memory Integration Deficits in Obstructive Sleep Apnoea

When presented with novel but semantically related elements after learning verbal material, healthy participants tend to endorse these items as previously learned. This reflects the normal integration and association of novel verbal information into long-term memory. How obstructive sleep apnoea (OSA) negatively impacts verbal memory performance, and whether deficits are reversible following positive airway pressure (PAP) treatment, remain elusive. We investigated immediate and delayed OSA- and PAP treatment-related effects on verbal memory integration, using a false memory paradigm. Twenty-three patients with OSA learned lists of words semantically related to target non-presented words (1) at baseline after a polysomnography diagnosis night, (2) after a consecutive polysomnography night under PAP titration, and (3) after three months of compliant PAP treatment. At each session, participants learned 10 different lists of words, each list comprising 15 semantically related items. They had then to recognize 15 minutes later (after an intermediate vigilance task) previously learned words within a list including studied words (learned), unstudied but semantically related items (lures), and non-related unstudied items (controls). Sleep quality and fatigue questionnaires, and psychomotor vigilance tests (PVT) were administered at each session. PAP treatment led to OSA remission and improvement in objective and subjective sleep quality. Crucially, recognition of learned and lure words increased after the first night under treatment and remained stable three months later, suggesting successful memory integration and restoration of semantic processes. No treatment-related outcome was found on PVT performance. OSA exerts a detrimental but PAP-reversible effect on verbal learning and semantic memory integration mechanisms underlying the acquisition of novel memory representations.     

Predictors of Adherence to a Brief Behavioral Insomnia Intervention: Daily Process Analysis

Behavioral interventions for insomnia are effective in improving sleep, yet adherence is variable, and predictors of adherence have not been consistently replicated. The relationships between daily variations in state factors at the initiation of treatment and adherence have not been investigated. Using 2-week, self-report online logs, this study determined, among 53 college students with probable insomnia, the associations of pretreatment factors and daily factors during treatment on daily variations in adherence to one session of behavioral treatments for insomnia. These treatments included stimulus control therapy (SCT), sleep restriction therapy (SRT), and sleep hygiene (SH). Low self-efficacy was associated with poorer SCT and SH adherence. Participants with a “bed partner or pet” at least some of the time had better SCT adherence. Greater total sleep time and poorer sleep quality were associated with poor SCT and SRT adherence the following night. Greater sleep efficiency was related to greater next night SCT and SRT adherence. Alcohol consumption was related to poorer SRT and SH adherence the following night. Future studies should test the replicability of these findings. Adherence trials may want to test whether discouraging alcohol intake, enhancing treatment-related self-efficacy, and monitoring and providing feedback on sleep, early in treatment, affects adherence.     

Spousal Influence on CPAP Adherence: Applications of Health‐related Social Control

Obstructive sleep apnea (OSA) is one of the most prevalent sleep disorders in the United Sates and is a risk factor for poor health outcomes. Continuous positive airway pressure (CPAP) therapy is an effective treatment for OSA, but adherence rates are poor. Although the spouses of patients with OSA have high potential to influence adherence, there has been minimal research to date. The purpose of this review is to summarize key findings regarding spousal influence on health‐related behavior change and treatment adherence from other medical fields (e.g., type 2 diabetes and cardiovascular disease) in which this line of inquiry has been more thoroughly developed. Recommendations are presented to guide future research investigating spousal influence on CPAP adherence based on findings from other patient populations. In particular, we emphasize the use of spousal health‐related social control as it applies to adherence and provide guidance regarding conceptual and methodological moderators.     

Relationship between sleep disturbance,depression and anxiety in the 12 months following a cardiac event

We aimed to assess the prevalence of sleep disturbance in a cardiac patient population over a 12-month period and assess its relationship with treatment adherence, self-efficacy, anxiety and depression. A total of 134 patients consecutively admitted to two Australian hospitals after acute myocardial infarction (31%), or to undergo bypass surgery (29%) or percutaneous coronary intervention (40%) were interviewed at six weeks and four and 12 months. Sleep disturbance was measured using a recode of the Beck Depression Inventory (v.2) item 16. Anxiety and depression were assessed by the Hospital Anxiety and Depression Scale. Sleep disturbance was highly prevalent (69%) at 6 weeks but was not associated with 12-month psychological outcomes. Path analysis revealed that sleep disturbance at 4 months was, however, associated with reduced treatment adherence and self-efficacy, and higher anxiety and depression scores at 12 months. The high prevalence of sleep disturbance in this study and its association with psychological outcomes may have adverse prognostic implications and possibly impede cardiac rehabilitation efforts.     

Transitions in sleep problems from late adolescence to young adulthood: A longitudinal analysis of the effects of peer victimization

Adolescence is a developmental period with high vulnerability to sleep problems. However, research identifying distinct patterns and underlying determinants of sleep problems is scarce. This study investigated discrete subgroups of, changes in, and stability of sleep problems. We also examined whether peer victimization influenced sleep problem subgroups and transitions in patterns of sleep problems from late adolescence to young adulthood. Sex differences in the effects of peer victimization were also explored. In total, 1,455 male and 1,399 female adolescents from northern Taiwan participated in this longitudinal study. Latent transition analysis was used to examine changes in patterns of sleep problems and the effects of peer victimization on these changes. We identified three subgroups of sleep problems in males and two in females, and found that there was a certain level of instability in patterns of sleep problems during the study period. For both sexes, those with greater increases in peer victimization over time were more likely to change from being a good sleeper to a poor sleeper. The effects of peer victimization on baseline status of sleep problems, however, was only significant for males, with those exposed to higher levels of peer victimization more likely to be poor sleepers at baseline. Our findings reveal an important role of peer victimization in predicting transitions in patterns of sleep problems. Intervention programs aimed at decreasing peer victimization may help reduce the development and escalation of sleep problems among adolescents, especially in males.

     

Patterns of adolescent involvement in problem behaviors: Relationship to self-efficacy,social competence,and life events

Using a sample of 556 adolescents from a suburban community, patterns of various adolescent problem behaviors (e.g., delinquent behavior, smoking, use of alcohol or drugs) and their links to self-efficacy, social competence, and life events were examined. Cluster analysis was conducted to identify four subgroups of adolescents who showed distinct patterns of problem behaviors. These clusters were compared on the measures of self-efficacy, social competence, and life events. Overall results suggest there are meaningful links between adolescents’ problem behavior patterns and self-efficacy, the amount and quality of participation in various after school activities, and life events. For example, a subgroup of adolescents who showed uniformly low prevalence of all problem behaviors reported more positive academic self-efficacy, more active participation in sports and nonsports activities, more positive life events, and fewer negative events than adolescents who were involved in multiple problem behaviors. Implications for prevention and future research on adolescent problem behaviors are discussed.     

Factors related to sleep quality in healthy elderly women.

Elderly women in subjectively good health--free of acute illness and major sleep pathologies--who were self-identified as good (n = 22) and poor (n = 16) sleepers were compared on measures of physical health, psychological symptoms, psychosocial status, and life-style. Poor sleepers reported longer sleep latencies, less total sleep time, more nonrestorative sleep, and more daytime fatigue than did good sleepers. Sleep recordings confirmed subjective reports, with shorter total sleep times and trends for lower sleep efficiency, longer sleep latencies, and more wake-after-sleep onset among women with subjective poor sleep. Poor sleepers also were more frequent users of sedative-hypnotic medications in the past. Current medication use, alcohol and caffeine use, daytime napping, and exercise were equivalent in both groups. Psychosocial status failed to discriminate groups. Poor sleepers reported significantly more psychological symptoms than did good sleepers. The levels of both psychological symptoms and sleep disturbance were mild.     

Evaluating a brief measure of social self-efficacy among U.S. adolescents

Preliminary data were collected to evaluate the performance of a social self-efficacy measure among 4,061 public high school adolescents. Principal-axis factor analysis was followed by a 4-way between-groups analysis of variance (ANOVA) to test for differences in the Total score means on selected demographic estimates and their interactions. Relations between the Total score and selected risk behaviors were examined through a series of one-way ANOVAs and the Tukey HSD test. Factor analysis results suggested a one-factor model best explained the factor structure of the scale items (factor loading range = .64 - .77, eigenvalue = 4.05, h(2) = .51). Females reported a significantly higher mean Total social self-efficacy rating than males, while White students reported a significantly higher mean Total social self-efficacy rating than Black and Asian students. Statistically significant lower mean Total social self-efficacy ratings were also noted for those who reported physical fighting, avoiding school, and being bullied.     

Sleep patterns and aging: comparison of older adults with and without insomnia complaints

The nature of geriatric insomnia was studied by comparing older adults with (n = 42) and without (n = 30) insomnia complaints on measures of sleep, mood, life-style, health, and sleep-requirement expectations. Elderly persons with insomnia complaints reported longer sleep latency and more frequent and longer awakenings and used sleeping aids more often than those without insomnia complaints. Nocturnal sleep time was not a reliable discriminator. Poor sleepers showed greater discrepancies between their current sleep patterns and sleep-requirement expectations than did good sleepers. Elderly insomniacs acknowledged greater symptomatology of depression and anxiety than did good sleepers. Daytime napping and physical exercise were equivalent in both groups. Medical disorders, pain conditions, and drug usage (other than sleep aids) did not distinguish the two groups. Clinical implications for the treatment of geriatric insomnia are discussed.     

Child perceived motor competence as a moderator between physical activity parenting and child objectively measured physical activity

This longitudinal study aimed to examine how physical activity parenting (PAP) directly predicted objectively measured children’s moderate-to-vigorous physical activity (MVPA) and sedentariness over a three-year transitional period from early to middle childhood, and second, whether the children’s perception of motor competence (PMC) mediated or moderated the influence of PAP to children’s MVPA or sedentariness. At time 1 (T1), PAP and children’s (N = 396, mean age 5.80, SD 1.04) PA were assessed by parental questionnaire. Three years later, at time 2 (T2), children’s (N = 396, mean age 8.80, SD 1.04) PMC was measured by a validated pictorial scale, and MVPA and sedentariness were measured by accelerometers. All the analyses were conducted using the Mplus statistical package (Version 8.4). The models were adjusted for the following covariates: children’s PA (T1), gender (T1), age (T1), mean accelerometer measurement in hours per day (T2), and parents’ education level (T1). Results showed that PAP at T1 did not significantly predict level of MVPA or sedentary time at T2 and, therefore, PMC did not mediate the PAP-children’s MVPA or sedentary time relationship either. However, PMC significantly moderated the relationship between PAP and MVPA but not between PAP and sedentary time. The results suggested that parental support positively predicts children’s MVPA among children with low PMC but not among children with high PMC. This unique finding proposes that family-based PA interventions could benefit from screening of children with low PMC and provision of PA counselling to their parents.     

Impact of a Theory-Guided Encouragement Intervention on an Employee Walking Pilot Program

This study investigated the impact of encouraging e-mails on self-efficacy, adherence, and step-count in a 10-week walking program. Using a 3 × 3 mixed design, participants randomly received task-involving/caring (n = 37), ego-involving (n = 35), or neutral (n = 34) e-mails and completed pre-/mid-/post surveys. Exercise self-efficacy increased over time, regardless of e-mail content. More of the task/care group adhered (63%) than ego (41%) or neutral groups (50%). The task/care group members who adhered reported significantly more daily-steps than ego or neutral groups. Results suggest that the theoretical framework used to guide encouragement in walking programs may impact participants’ adherence and steps.     

The training & support programme for parents of children with ataxia: a pilot study

The aim of the study was to evaluate the impact of the Training and Support Programme among parents of children with ataxia. A total of 39 parents expressed an interest in the TSP, 27 (mean age 41; range 25-58) returned baseline questionnaires and completed the study. Twenty-four children (mean age 12.5 years, SD=12.4) received the TSP. Data were collected by self-administered questionnaires mailed to parents immediately before attending the TSP and at 4-month follow-up. Interviews were conducted with 10 parents immediately following TSP completion. Comparisons over time revealed significant decreases in parental anxious mood (p=.011), depressed mood (p=.046) and perceived stress (p=.020) and significant improvements in generalized self-efficacy (p=.010), satisfaction with life (p=.045) and parental health status (p=.020). Parents reported improvements in children's mobility, "jumping legs", sleep patterns, energy and activity levels, relaxation, and happiness (e.g., more smiles). Parents felt closer to their children and one parent had become "more accepting" of their child's ataxia. Results indicate that the TSP may enhance the psychosocial well-being of parents of children with ataxia although a randomized controlled trial would be necessary to determine whether the changes reported here are in fact due to attending the TSP.     

Heterogeneity of Incentives for Physical Activity and Self-Efficacy in Highly Active and Moderately Active Women Exercisers

The purpose of this investigation was to examine whether persons exercising 2 times per week or less versus 3 times per week or more, with intentions to maintain current level of activity form a homogenous group in terms of motivational features. A group of 86 women recruited from local fitness clubs reported on their exercise incentives, self-efficacy, exercise behavior, and exercise intentions. A discriminant function analysis revealed that self-efficacy and incentives for stress reduction and mental health discriminated highly active from moderately active women. The implications of these findings are discussed in terms of stage theory and research in exercise adherence.     

Poor Sleep as a Risk Factor for Nonsuicidal Self-Injury in Adolescent Girls

The purpose of this study was to examine the relation between poor sleep and nonsuicidal self-injury (NSSI), and to test the hypothesis that poor sleep is a risk factor for the development of NSSI in young adolescents. Questionnaire data were used from a 2-wave longitudinal study of a community sample of 881 young Swedish adolescents. The results showed that 7 % of the girls reported poor sleep (never or seldom sleeping well), and 20–26 % of the girls reported repeated NSSI (at least 5 instances). Poor sleep was associated prospectively with NSSI among girls, but not among boys. Of girls who responded that they seldom or never slept well at T1, 77 % reported repeated NSSI 1 year later. Poor sleep at T1 was found to predict the incidence of new cases of repeated NSSI in girls at T2, independently of their degree of psychopathology. No similar relationship between poor sleep and NSSI was found in boys. The present results suggest that screening for poor sleep in adolescents may serve to identify a subgroup of girls at risk for developing NSSI. It is concluded that poor sleep in young girls should be taken seriously, even in the absence of other self-reported psychological problems, and that interventions targeted at sleep disturbances may be important for prevention.     

Sleep,resilience, and psychological distress in United States military Veterans

Sleep problems are prevalent among Veterans. Left untreated, such problems may elevate psychological distress and increase risk of subsequent mental health disorders. Psychological resilience may buffer against negative psychological outcomes, yet the relationship between sleep and resilience has not been studied. This study explored poor sleep, resilience, and psychological distress using questionnaires collected as part of the Study of Post-Deployment Mental Health. Participants (N = 1,118) had served in the US military since September 11, 2001, had one or more overseas deployments, and were free from a past-month DSM-defined mental health disorder. Hierarchical linear regression was used to examine the association between poor sleep quality (Pittsburgh Sleep Quality Index total score) and psychological distress (Global Symptom Index; Symptom Checklist-90-R), controlling for demographic and health characteristics. Moderation analyses tested for a potential buffering effect of resilience (Connor-Davidson Resilience Scale). Poor sleeping Veterans had worse physical and psychological health, lower resilience, and endorsed more lifetime traumatic events. Poor sleep was associated with greater psychological distress controlling for health and demographic characteristics. Both resilience factors—adaptability and self-efficacy—had significant buffering effects on the relationship between poor sleep and psychological distress, suggesting that resilience may protect against negative outcomes in poor sleepers. Additional research is warranted to better understand the relationships between sleep, resilience, and psychological distress. Such research may inform pertinent prevention efforts, including interventions that improve sleep, enhance resilience, and protect against incident mental health diagnoses.     

Using topic modeling to develop multi-level descriptions of naturalistic driving data from drivers with and without sleep apnea

One challenge in using naturalistic driving data is producing a holistic analysis of these highly variable datasets. Typical analyses focus on isolated events, such as large g-force accelerations indicating a possible near-crash. Examining isolated events is ill-suited for identifying patterns in continuous activities such as maintaining vehicle control. We present an alternative approach that converts driving data into a text representation and uses topic modeling to identify patterns across the dataset. This approach enables the discovery of non-linear patterns, reduces the dimensionality of the data, and captures subtle variations in driver behavior. In this study topic models were used to concisely described patterns in trips from drivers with and without untreated obstructive sleep apnea (OSA). The analysis included 5000 trips (50 trips from 100 drivers; 66 drivers with OSA; 34 comparison drivers). Trips were treated as documents, and speed and acceleration data from the trips were converted to “driving words.” The identified patterns, called topics, were determined based on regularities in the co-occurrence of the driving words within the trips. This representation was used in random forest models to predict the driver condition (i.e., OSA or comparison) for each trip. Models with 10, 15 and 20 topics had better accuracy in predicting the driver condition, with a maximum AUC of 0.73 for a model with 20 topics. Trips from drivers with OSA were more likely to be defined by topics for smaller lateral accelerations at low speeds. The results demonstrate topic modeling as a useful tool for extracting meaningful information from naturalistic driving datasets.     

Temporal relationships of self-efficacy and social support as predictors of adherence in a 6-month strength-training program for older women.

The present study investigated how self-efficacy and social support predicted adherence to a strength training program for elderly women over two time periods in the initial 6 mo. of the program. Participants were 30 elderly women volunteers aged 75 to 80 who completed measures of barrier self-efficacy and general social support at baseline and 3 mo. later. Social support from the program was also measured at 3 mo. Adherence to the program was measured by attendance. Hierarchical regression equations were utilized to identify the contributions of self-efficacy and social support for adherence at 0 to 3 mo. and 4 to 6 mo. For prediction of the first 3 mo. of adherence, both self-efficacy and social support contributed significant unique variance towards the total explained variance of 36%. For the 4- to 6-mo. period, self-efficacy explained significant (12%) variance in adherence even when controlling for the previous 3-mo. adherence. Inclusion of general social support and social support from the program, however, did not account for significant variance. Researchers must continue to examine self-efficacy and social support in exercise adherence within various time periods among older adults to develop effective intervention strategies.     

Depression in adults with type 2 diabetes: the role of adherence, body mass index, and self-efficacy.

Considerable evidence links depression with the development and worsening of diabetes, but the factors contributing to this link have not been established. The authors examined the role of adherence, body mass index (BMI), and self-efficacy. Adult patients with Type 2 diabetes (N = 56) completed self-report measures of diet and exercise adherence, diet and exercise self-efficacy, and depression. BMI was obtained from medical records. Path and mediation analyses indicated that both adherence and BMI independently contributed to self-efficacy. Self-efficacy mediated both the association between adherence and depression and the association between BMI and depression. These findings are consistent with the proposal that lower self-efficacy in reaction to adherence failure and higher BMI contributes to depression in adults with diabetes.     

Diagnostic test evaluation of a nasal flow monitor for obstructive sleep apnea detection in sleep apnea research

In this diagnostic test evaluation of a nasal flow monitoring device for obstructive sleep apnea (OSA), 34 patients referred for polysomnography were studied at home for three consecutive nights with the monitor. The mean age of subjects (+/-SD) was 41.9+/-10.3 years, and their mean apnea-hypopnea index (AHI) was 31.5+/-27.2. The difference between the average AHI from three nights at home on the monitor and the polysomnogram (PSG) result was 1.8+/-17.1. The area under the receiver operating characteristic curve (AUC) for PSG AHI > or = 10 was .96. With a threshold AHI of 18 on the flow monitor, sensitivity was .92, specificity .86, positive predictive value .96, and negative predictive value .75. For detecting severe OSA (AHI > or = 30), the AUC was .85. With knowledge of appropriate thresholds and the pretest risk of OSA, the flow monitor can be used to detect or exclude OSA for sleep-related research, as well as to identify severe cases needing priority for further evaluation.