The anxiety- and pain-reducing effects of music interventions: a systematic review

AORN Journal,  April, 2008  by Ulrica Nilsson

• E-mail
• Print
• Link

QUALITY. No single included study was assessed to have the maximum quality score of 10 points. The total quality scores ranged from 1 to 9, with a mean score of 6. Two studies scored 9 points, (48, 51) and five studies scored 8 points. (17, 24, 42, 50, 52) All seven of these studies were published between 1995 and 2005. In 11 of the reviewed studies (ie, 26%), the quality was insufficient, meaning that the quality assessment method used in this review scored the studies three points or less. 19, 23, 26, 27, 29, 36, 39, 47, 54, 55, 58) These studies were published between 1995 and 2006.

A majority of the included studies (n = 30) had demonstrated validity and reliability measures in 90% of their outcomes. (17-21, 23-26, 28, 30-36, 38, 40-43, 46-48, 50, 51, 55-57). In the studies that demonstrated less or no validity and reliability, the most common reason was that the studies did not report the validity and reliability of the equipment used to measure blood pressure, heart rate, and respiratory rate.

Nine of the studies (24, 28, 33, 40, 43, 49-52) were single blind--that is, all patients used headphones, and the patients in the control group listened to a blank tape. The patients, thus, were blinded to their group assignment. Two studies were double blind, (42, 48) in that neither the patient nor the researcher knew the group assignment (ie, control group, music intervention). In these studies, the participants received the music intraoperatively while they were under general anesthesia. The remaining studies had no blinding, with the patients in the control groups receiving usual care or rest.

Only eight studies (17, 24, 44, 45, 48, 50-52) reported truly random allocation. Semisecure allocation (eg, flipping a coin, drawing a slip of paper) was reported in five studies. (18, 26, 35, 43, 57) In the remaining 21 studies, the randomization method was not explicit or was nonsecure (eg, patients were assigned to groups on random days or weeks). One hundred percent of follow up or intention-to-treat analysis was reported in 22 of the studies. (17, 25, 28-33, 35-38, 40-43, 46, 49, 51-54) Twelve studies reported 80% to 90% follow up, (18-21,24, 44, 45, 48, 50, 55, 56, 58) and in the rest of the studies, there was 79% or less follow up or the information was unclear. (22, 23, 26, 27, 34, 39, 47, 57) Sample size calculation was reported in 13 of the included studies. (17, 18,24, 28,30, 34, 40, 42, 48, 50-52, 57)

OUTCOME MEASURES

Various outcome measures were recorded in the articles reviewed. These included patient anxiety, pain, vital signs, and blood sample indicators (Table 2).

ANXIETY. A total of 24 studies evaluated the effect of music on patient anxiety. (17, 19, 24, 27, 30, 32-37, 39-43, 45-47, 49-51, 55, 57) The most common tool used to measure anxiety was the State-Trait Anxiety Inventory (STAI), which was used in 19 of the 24 studies. (17, 19, 24, 27, 32, 33, 35-37, 39-43, 46, 47, 49, 55, 57) a numeric rating scale was used in three studies, (17, 50, 51) and a visual analog scale was used in four studies. (30, 34,43,46) Perceived control over anxiety was measured in one study, (55) and the Observer's Assessment of Alertness/Sedation was used in one study. (49)