Mishel 2002.

Study characteristics
Methods	Setting: central and eastern North Carolina, USA Recruitment: eligible patients were African American and Caucasian men who were diagnosed with localised prostate cancer and were within 2 weeks of catheter removal post surgery and/or < 3 weeks of starting radiation therapy, had no major cognitive impairment, had no concurrent other cancer, had access to a telephone, had a family member willing to participate in the study, and planned to reside in their current community for 12 months. Participants were recruited from 9 treatment facilities. Members of the research team introduced the study to participants during a clinic visit, and interested participants were subsequently mailed an information package and were called to determine willingness to participate Randomisation: RCT using 3 × 2 block randomisation
Participants	239 men with prostate cancer
Interventions	Intervention: based on the Theory of Uncertainty in Illness, which assumes that uncertainty occurs when participants lack the information or knowledge needed to fully understand their illness and treatments. Through cognitive re‐framing, participants can learn to cope better with uncertainty. The intervention consisted of 8 × weekly phone calls, which were delivered in different ways in the 2 intervention arms: (1) direct, in which only the participant received the intervention, and (2) supplemented, with supplemented delivery to a close family member. An assessment was made by telephone to identify the participant's cancer‐related concerns. In the supplemented arm, the spouse or designated family support person also received a weekly phone call for 8 weeks from a nurse who conducted a similar assessment of family members’ concerns about the participant using a list similar to that used for the participants. Based on identified concerns, an appropriate intervention was then selected from a standardised list, which included validating and reinforcing views and behaviours, providing information, activating resources, teaching symptom management strategies, structuring expectations, applying problem‐solving, and teaching assertion techniques for communicating with health care providers. These interventions were applied to concerns about diagnosis, treatment, response to treatment, living with cancer, caring for oneself, and social/lifestyle issues. Printed materials, audiotapes, and videotapes for managing specific problems were mailed to participants after each weekly call. Any materials that were sent to the participant were reviewed at the next phone call Control group: usual care, which included printed general health information not related to prostate cancer or the side effects of treatment Interventionist: trained nurses matched with the participant and the family member by ethnicity and gender
Outcomes	Psychological symptoms ‐ symptom distress Uncertainty ‐ uncertainty and uncertainty management were measured Methods for assessing outcomes: Mini‐Mental State Examination (MMSE) (α = 0.85 to 0.98 in various studies) Mishel's Uncertainty in Illness Scale, shortened version (α = 0.82 to 0.88 in this sample) Self‐Control Scale ‐ Problem‐Solving and Cognitive Re‐framing subscales (α = 0.86 to 0.90 in this sample) Cancer Knowledge Scale (KR‐21) (α = 0.64 to 0.73 in this sample) Symptom Distress Scale (α = 0.68 to 0.74 in this sample)
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The generation of random sequence is not described
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding (performance bias and detection bias) All outcomes	Unclear risk	"Data collectors who were not involved in the delivery of the intervention collected all data". It is unclear if participants were blinded to group
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Low loss to follow‐up, but reasons for this are not described
Selective reporting (reporting bias)	Low risk	Results of all measures are reported, including non‐significant findings. However, the paper focuses slightly more on significant differences found at T2 than on non‐significant differences at T3
Other bias	Low risk	The trial appears to be free of other problems that could put it at high risk of bias