Page 74 - AN-2-3

Advanced Neurology Sleep and lifestyle factors in young adults with childhood TBI

studies, an actigraphy sleep efficiency score below 85% reported previously, with results showing significantly
20
indicated poorer objective sleep . more males (χ [1, n = 172] = 9.33, P = 0.002) and higher
[37]
2
SES (median = 4.30, U = 2541.50, P = 0.032) in the non-
2.4. Statistical analysis participants compared to those who participated in this
All analyses were performed using IBM SPSS Statistics 20-year follow-up.
(Version 29). Data were checked to ensure compliance Table 1 presents results from demographic comparisons
with statistical assumptions using frequencies, distribution for this 20-year follow-up sample. Results show statistically
plots, skewness and kurtosis values, and with Shapiro–Wilk significant differences in age at injury among the TBI
statistics. Differences in demographic factors between the severity groups (P = 0.035), with significant differences
TBI and TDC groups, and paired contrasts were conducted found between the mild and moderate TBI groups
between the TDC and TBI severity groups using χ tests, (P = 0.032). Age at follow-up was also significantly higher
2
Mann–Whitney U tests, analysis of variance (ANOVA), in the TBI compared to the TDC group (P = 0.047). The
and independent sample t-tests.
four group contrasts showed these significant differences
Before exploring the relationships between lifestyle factors in age at follow-up: TDC < mild TBI (P = 0.002) and mild
2
and sleep outcomes, χ tests and Mann–Whitney U test were TBI > moderate TBI (P = 0.039). The proportion of people
used to compare the TBI severity groups to the TDC group with higher levels of education was also higher in the TDC
on the sleep outcomes and lifestyle factors. Although we compared to severe TBI group (P = 0.037).
have previously reported similar comparisons in this sample
for the sleep outcomes [20,33] , the current analysis extends the 3.2. Differences in lifestyle factors and sleep
previous reports by looking at group differences in lifestyle outcomes: TBI severity and TDC groups
factors and teasing out the differences in outcome between Table 2 presents analyses comparing the TBI and TDC
each TBI severity group and the control group. groups and the four groups (i.e., TDC, mild, moderate, and
To address the study aim, the relationships between severe TBI), on lifestyle factors and sleep outcomes. There
subjective and objective sleep and lifestyle factors in the were significantly more parents in the TBI compared to
TBI group were explored using generalized linear models, TDC group (P = 0.049). All participants who have children
since outcomes were mostly not normally distributed. were in the TBI group, and the majority of them (92%)
Separate models were run for the subjective and objective had children below 8 years old. Pain medication use was
sleep outcomes, with each model including these lifestyle higher in the TDC compared to the TBI group (P = 0.048).
factors: Caffeine use (morning, afternoon, evening, and Subjective and objective sleep quality were not statistically
total), screentime, nap duration, chronotype, substance use different between the TBI and TDC groups, and neither
(alcohol and tobacco), parenting status, and medication use were the proportion of participants presenting with poor
(stimulants, antidepressants, and pain medications). Age at subjective sleep quality (39% and 15%, P = 0.109), and
follow-up was included in each model to control for its effect poor objective sleep efficiency (67% and 81.8%, P = 0.327)
on the relationships assessed since age emerged as a potential significant in these group, respectively.
confounding variable in the demographic comparisons In the four group comparisons, there was a greater
and has been associated with sleep outcomes . Follow-up proportion of parents in the mild (P = 0.037) and severe
[38]
generalized linear models were conducted in just the TBI (P = 0.013) TBI groups, compared to the TDC group.
participants with poor subjective (PSQI > 5) or objective The proportion of young adults using pain medication
(sleep efficiency <85%) sleep outcomes to verify if similar was higher in the TDC compared to severe TBI group
factors predicted outcomes in this subgroup. To verify if (P = 0.018), and stimulant medication use was higher in the
the identified relationships were specific to the TBI group, mild compared to severe TBI group (P = 0.037). Evening
similar analyses were conducted in the control group using caffeine use was significantly higher in the severe compared
Spearman correlations (since the small sample size was to mild TBI group (P = 0.037). Finally, subjective sleep
not suitable for generalized linear models). A statistical
significance threshold of P < 0.05 was used for all analyses. quality was significantly poorer in the mild TBI compared
to TDC (P = 0.042), and in the moderate TBI compared to
3. Results the TDC (P = 0.012) and severe TBI (P = 0.025) groups.
3.1. Demographic characteristics 3.3. Relationships between sleep outcomes and
lifestyle factors in the TBI group
Demographic differences between 20-year follow-up
TBI participants and non-participants (i.e., those who Results from generalized linear models assessing the
did not participate in this 20-year follow-up) have been relationships between sleep outcomes and lifestyle factors,

Volume 2 Issue 3 (2023) 4 https://doi.org/10.36922/an.0876