Intervention group received laughter therapy twice a week for six weeks.

The control group did not take part in the laughter therapy program. This group received the routine care given by the nurses in the center.

Laughter therapy program was composed of four steps. The first step, which took 10 minutes, had warm-up exercises including gentle stretching and hand clapping. The second step, which included deep breathing exercises and hand clapping, took 5 minutes. The third step involved children's games and laughter exercises. In this step, the most common laughter exercises were milkshake laughter, bugi bugi laughter techniques, lion laughter, cell phone, hot soup laughter, hug laughter, bird laughter, dialogue with nonsense, speech exercises, laugh at one's own aches and pains exercises, argument laughter, brushing teeth and mouthwash exercises. The sessions included a combination of different laughter exercises. This step took 15 minutes. The last step, which included wishes and resting by breathing, took 10 minutes.

The De Jong Gierveld Loneliness Scale was originally developed by De Jong-Gierveld and Kamphuls (1985) and revised by De Jong Gierveld and Van Tilburg (1999). The 11-item DJGLS consists of two subscales: six negatively framed items measure emotional loneliness and five positively framed items measure social loneliness. The sum of these two subscales constitutes the general loneliness score. The scale is a Likert-type scale in which positive items are scored as 0 = yes, 1 = more or less, and 2 = no; and negative items are scored inversely as 2 = yes, 1 = more or less, and 0 = no. Total scores can range from 0 to 22, with a higher score denoting more severe loneliness. The DJGLS was tested for validity and reliability on Turkish population by Akgul and Yesilyaprak (2015), with Cronbach's alpha reliability coefficient of 0.85. In the present study, Cronbach's alpha was 0.87.

The Life Satisfaction Scale was developed by Diener et al. (1985). This five-point Likert scale, which has one factor structure, consists of 5 items and evaluates the life satisfaction of the individuals in general. The total score that could be obtained from the scale ranges from 5 to 25, and a higher scale score indicates a higher level of life satisfaction.

The LSS was tested for validity and reliability within the Turkish population by Daglı and Baysal (2016), with Cronbach's alpha reliability coefficient of 0.88. In the present study, Cronbach's alpha was 0.89.