Results:
Table 1. shows the relationship between the baseline information of participants and the score of Sum sleep. Sum sleep score is divided in to two groups (<5 hours and >5 hours). None of the baseline characteristics had significant difference with sum sleep score.
Table 2. shows the association between the baseline information of participants and time in bed. time in bed is divided in to two groups (<6 hours and >6 hours). None of the baseline characteristics had significant difference with time in bed.
Table 3. shows the results and relationship between DASH score and sum sleep and time in bed. Our results showed that there was a negative relationship between DASH score and sum sleep, but the relationship was not significant. The data for DASH score were divided into quartiles, and the first quartile was considered as reference (anti-inflammatory quartile). Data were analyzed in two models, the first model was the non-adjusted model and the second one was adjusted for energy intake, age, BMI. In none adjusted model, there was not any significant relationship between any of the quartiles and sum sleep. But, the p-value for second quartile was 0.06 that showed the relationship could be strong. Also, the results showed that there is a strong positive association between DASH score and time in bed, in adjusted and none adjusted model (P<0.001).
Table 4 shows the association between HEI score and sum sleep and Time in bed, in adjusted and non-adjusted model. The negative relationship between HEI and sum sleep and time in bed, in adjusted and non-adjusted model was strongly significant (P<0.001). The data for HEI score were divided into quartiles, and the first quartile was considered as reference. There was not any significancy in the quartiles.
The relationship between DII score and time in bed and sum sleep is presented in Table5. The results showed that there is a strong negative relation between DII and time in bed and sum sleep, in adjusted and non-adjusted model (P<0.001). Like the other scores, DII scores were divided into quartiles, and the relationship between each quartile was examined, and no significancy was observed.