Lanceley 2017.

Study characteristics
Methods	A randomised controlled, multi‐centre trial of individually tailored follow‐up (synonymously termed individualised or intervention treatment) led by a gynaecologic clinical nurse specialist versus conventional follow‐up. Cost, effects on QoL (using QLQ‐C30 and QLQ‐Ov28 questionnaire), mood (using HADS scale), and patient satisfaction were compared.
Participants	Eligible participants were women with the clinical diagnosis of ovarian cancer or fallopian tube or peritoneal cancer who had completed primary treatment by surgery alone or with chemotherapy irrespective of outcome with regard to remission, were expected to have a survival of more than 3 months, aged 18 years or older, and willing and able to participate. After randomisation, clinical or demographic characteristics of participants in the groups were similar and these included: stage at diagnosis, Eastern Cooperative Oncology Group performance status, coexisting diseases, ethnicity, marital and employment status and highest education level. Mean age of participants in the intervention arm was 62 years (range: 23 to 92 years), whereas in the conventional arm, the mean was 61 years (range: 21 to 85 years). At baseline, there was no significant treatment effect on the global QLQ‐C30 score (P = 0.3), global QLQ‐Ov28 score (P = 0.34), or global HADS score (P = 0.3).
Interventions	Individually tailored follow‐up led by a gynaecologic cancer nurse: women selected for individualised follow‐up were allocated to one of several gynaecologic cancer nurse specialists. They met with the nurse immediately after their end of treatment appointment to negotiate follow‐up to suit their individual situation. Contact was flexible, primarily by telephone at pre‐arranged mutually convenient times, although some women opted for face‐to‐face appointments, usually at the regular gynaecologic cancer clinic. In addition, contact with the nurse was made when necessary in the regular gynaecologic oncology clinic or by telephone without prearranged appointment. Women were assessed using a holistic guide to identify signs of disease progression, symptoms warranting intervention, and psychological issues. Unless the woman had worsening symptoms needing further treatment, the nurse was responsible for the care of women receiving individualised follow‐up. Conventional medical follow‐up: consisted of 1 posttreatment outpatient appointment with further appointments at 3 monthly intervals including complete clinical history and CA‐125 and radiological imaging when symptoms or signs appeared.
Outcomes	Global QLQ‐C30 score Outcome type: continuous outcome Direction: lower is better Scale: 0‐ to 100‐point scale Global QLQ‐Ov28 score Outcome type: continuous outcome Direction: lower is better Scale: 0‐ to 100‐point scale Global HADS score Outcome type: continuous outcome Direction: lower is better Scale: 0‐ to 100‐point scale Cost analysis Outcome type: continuous outcome Direction: lower is better
Identification	Sponsorship source: The work was partly funded by The Eve Appeal Gynaecological Cancer Charity and undertaken at UCLH/UCL within the NIHR UCLH/UCL Comprehensive Biomedical Research Center, supported by the Department of Health. The research activity of C.B. was partially supported by the FP7‐305280 MIMOmics European Collaborative Project, as part of the HEALTH‐2012‐INNOVATION scheme. Country: United Kingdom Authors name: Dr. Anne Lanceley Institution: Department of Women’s Cancer, The UCL Elizabeth Garrett Anderson Institute for Women’s Health, University College London Email: a.lanceley@ucl.ac.uk Address: 74 Huntley St, London WC1E 6AU
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random sequence generation was assured by using randomness derived from atmospheric noise through www.random.org
Allocation concealment (selection bias)	Unclear risk	Web‐based method of allocation was used, but no details were provided for the allocation sequence concealment.
Blinding (performance bias and detection bias) All outcomes	High risk	Participants and personnel were not blinded and the quality of life and psychological effect outcomes are likely to be influenced by lack of blinding. However, blinding would not be possible. Participants met with the nurse immediately after their end of treatment appointment to negotiate follow‐up to suit their individual situation. Knowledge of the assigned intervention is likely to influence patient‐reported outcomes, such as QoL and mood. However, the cost analysis outcome is unlikely to be influenced by lack of blinding.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	The proportion of missing data due to non‐compliance varied during the follow‐up period (range: 2% to 23%). However, there were no significant differences in the proportion of missing outcome data in the groups. Missing data in economic analysis outcome were imputed for all participants, independently of the intervention group, using appropriate methods.
Selective reporting (reporting bias)	Low risk	The study followed the prespecified protocol and is deemed to be low risk for reporting bias.
Other bias	Unclear risk	Despite the commitment to enrol consecutive women, more women than anticipated were deemed unsuitable for inclusion by their consultant, and some were simply judged too sick, with multiple comorbidity. Also, nurses trained to deliver the individualised follow‐up were likely to be invested in its success, and thus, they may have been more attentive and fulfilled women's expectations of continuity and responsiveness to their difficulties.