The purpose of this 5-year interventional study is to design, implement, and evaluate the effectiveness of an intervention aimed at facilitating prognosis communication and end-of-life care decision-making to increase the extent of congruence between the patient's preferred and actual EOL care received and patients' and family caregivers' QOL and psychological well-being, reduce futile aggressive healthcare resources utilization at end-of-life, and facilitate bereavement adjustment.

A randomized controlled trial with a tailored, multifaceted intervention will be conducted on a convenience sample of 231 dyads of terminally ill cancer patients and caregivers with the same number of attention controlled group to evaluate the intervention's effectiveness.

Sample size calculation: There was no interventional study directly investigating the effectiveness in increasing the extent of congruence between patient preferred and actual end-of-life (EOL) care received by terminally ill cancer patients; therefore sample size estimation will be based on the effectiveness of holding EOL care discussions between patients and physicians on the extent to increase congruence between the patient's preferred and actual EOL care received. Patients who reported having EOL care discussions with their physicians significantly more likely to receive EOL care consistent to their preferences (OR=2.26; p<.0001) and received significantly fewer aggressive medical interventions near death, with lower rates of CPR (0.8% vs 6.7%; AOR, 0.16; 95% CI, 0.03-0.80), mechanical ventilation support (1.6% vs 11.0%; AOR, 0.26; 95% CI, 0.08-0.83), and ICU care (4.1% vs 12.4%; AOR, 0.35; 95% CI, 0.14-0.90), respectively. A sample size of 124-195 dyads per group achieves 80% power to detect a difference between the interventional and control group by a two-sided hypothesis test with a significance level of 0.05. In compensating the 18.5% of attrition rate found in our previous longitudinal study, 147-231 dyads per group are needed. The proposed sample size will be targeted on 231 dyads per group to ensure adequate power to detect the hypothesized effects of the proposed intervention.

Approximately 8-10 new dyads of terminally ill cancer patients and their family caregivers were recruited in each month. In order to recruit the targeted 462 dyads (231 dyads in each treatment group) of subjects, after development of detailed study protocol, subjects will start to enter into the study at the beginning of the 6th month of the study through the 54 th month and be followed through the 5th year of the proposed study period. We will enroll and randomly assign eligible dyads of terminally ill cancer patients and their family caregivers into intervention or attention controlled group in a 1:1 fashion without stratification.

Data collection procedures: Assessments will be performed prospectively and continue until patient death, loss to follow-up, study withdrawal, or when the patient can no longer be interviewed. An every-3-week time frame will be used in this study for repeated QOL and other outcomes assessments for both patients and their family caregivers bases on the review of literature to cover the most rapid change of patient physical conditions and demanding period of caregiving until the death of the patient. After each patient's death, a chart review and postmortem interview with patients' caregivers will be performed to confirm the type of medical care received at the EOL. Bereavement interviews will be conducted 1, 3, 6 and 13 months after the time of death in order to avoid contamination on the basis of anniversary grief reactions.1

Several strategies will be taken to ensure research fidelity.

1. The proposed intervention requires professionals skilled in the content, techniques, and delivery of the intervention. Built on her previous oncology and hospice care experiences, a competency-based educational approach will be used to train the facilitator. The training will include an overview of the study protocol and procedures, a review of the developed workbook and video, and instructions in motivational assessments. The principal investigator (PI) will serve as a role model for the facilitator to coach her how to assess and motivate participants at different stage of the readiness for engagement in ACP and cooperate with physicians in coordinating and facilitating EOL care discussions. The facilitator needs to successfully demonstrate predefined competencies and consistency to delivery of the ACP intervention after training and receipt of individual feedback from the PI before formal interventions can be conducted. Thereafter the study team will meet biweekly to review the facilitator's notes on interventional sessions and to provide feedback on difficult subject management issues.
2. Interventions provided to subjects assigned to the intervention group will be compared with those received by those who assigned to the control group every 3 months. Randomly selected dyads of terminally ill cancer patients and their family caregivers in each treatment group will be interviewed by the PI to check the extent to which treatments provided to them consistent with the protocol for each group.
3. To avoid bias, separate data collectors will be hired and trained for the study to collect data for the intervention and the control group independently. Training will include screening subjects, obtaining consents, and administering the project instruments without offering information; research assistants (RAs) will be blinded to treatment conditions. Procedures will be implemented to insure the standardization of data collection. RAs reliability will be established by comparing data collected by the RAs with that recorded by the PI on five pilot cases. Ninety-five percent agreements will be required before data collection for main study can begin. Failure to reach this agreement rate will require additional training until the acceptable reliability level is reached. Intermittent reliability checks will be carried out by the PI throughout the study to maintain a 95% agreement.
4. An independent Data and Safety Monitoring Board (DSMB) will be organized to ensure continuing patient safety as well as the validity and scientific merits of the trial. The DSMB will constitute a biostatistician experienced in statistical methods for clinical trials, a nursing researcher experienced in conducting RCTs, and a physician-scientist experienced in cancer care.

The DSMB will monitor and address the following issues: (1) sufficient and appropriate enrollment of subjects, including compliance with the eligibility criteria for each dyad of terminally ill cancer patients and family caregivers enrolled in the trial, (2) appropriate implementation of randomization, (3) comparability of baseline data between treatment groups, (4) protocol compliance, including treatments delivered to each treatment group and the data collection schemes,(5) adverse events (AEs), quality assurance for data validation and registry procedures for Clinical Trial registration at ClinicalTrials.gov.

A databet that records number of recruitment each month and the baseline information will be maintained to ensure sufficient and appropriate enrollment of subjects.

A dataset that captures the values the personnel used to randomize participants will be maintained, thereby allowing the process of treatment allocation to be reproduced and verified later. The DSMB will monitor the appropriate implementation of randomization by reviewing this dataset.

Interventions delivered to subjects assigned to the intervention and the attention-controlled group will be compared every 6 months by interviewing randomly selected dyads of terminally ill cancer patients and their family caregivers in each treatment group to check treatments provided to them.

AEs will be recorded and submitted by the study team in written to the DSMB monthly, with immediate reporting of serious AEs to the DSMB and an oncologist with expertise in cancer care. Site reports AEs/SAEs to its IRB will be dictated by local requirements.

No data will be analyzed before the study ended. Access to interim results, including results according to study arm will be limited to the DSMB members and the statistician who prepared the reports only. The DSMB will review study data every 6 months and the DSMB' summary recommendations will be directed to the PI. A summary of the review of reported AEs/SAEs will be sent to the local IRB to ensure that the participating center will be informed of any pertinent safety issues.

The biostatistician will be responsible for ensuring quality of data submitted to the registry against the predefined range of each independent and dependent variable and assessing the accuracy and completeness of registry data by comparing the submitted data to the original data.

Data analysis and interpretation:

1. All data will be scored and entered into a computer spreadsheet by an administrative assistant blinded to the group allocation of the participant.
2. In order to test for baseline group equivalence, differences between study groups in baseline characteristics and identified outcomes will be assessed with the use of two-sided Fisher's exact tests and chi-square tests for categorical variables and independent-samples Student's t-tests for continuous variables.
3. Intention-to-treat regression analyses will be performed with generalized estimating equations (GEE) to examine the effectiveness of intervention. In the intention- to-treat regression analyses, all participants will be analysed in the treatment group to which they are initially allocated regardless of whether they complete or withdraw from the treatment until the patient dies.
4. The extend of congruence of the terminally ill cancer patients' preferred and actual EOL care received will be obtained by comparing the agreement between the preferences of EOL care elicited at the last assessment and actual EOL care received by the patient by the percentage of overall agreement and kappa coefficients to assess the extent of congruence to correct for the amount of agreement that can be expected to occur by chance alone.
5. Multivariate logistic regression by the GEE will be used to examine the impact of intervention on improving the extend of congruence between terminally ill cancer patients' preferred and actual EOL care received, and secondary outcomes, namely increasing prognosis awareness, EOL care discussion, and the extent of patient-family agreement on preferences of EOL care, decreasing use of futile aggressive EOL care, and facilitating hospice use and early hospice referral, with simultaneously adjusting for confounding factors. The moderation effects of prognosis awareness, EOL care discussion, and patient-family agreement on preferences of EOL care on extend of congruence between terminally ill cancer patients' preferred and actual EOL care received, decreasing use of futile aggressive EOL care, and facilitating hospice use and early hospice referral between interventional group and the attention controlled group also will be examined in the GEE model.

7. Multivariate multiple regression by the GEE will be used to test the effectiveness of intervention in improving patients' and family caregivers' QOL and psychological well-being (anxiety and depression), and family bereavement outcomes (including QOL, depression, and grief reactions), with simultaneously adjusting for confounding factors. The moderation effects of prognosis awareness, EOL care discussion, and patient-family agreement on preferences of EOL care on the differences in patients' and family caregivers' QOL and psychological well-being before the patient's death, and family bereavement outcomes between intervention group and the attention controlled group also will be examined in the GEE model. How limiting futile aggressive EOL care and the extent of congruence between terminally ill cancer patients' preferred and actual EOL care received will moderate the intervention's effectiveness on improving patients' and family caregivers' QOL and psychological well-being before the patient's death and family bereavement outcomes also will be examined in the GEE model.

Inclusion Criteria:

Terminally ill cancer patients are they:

1. have a disease at a terminal stage which continues to progress with distant metastases and is unresponsive to current curative cancer treatment as judged by their oncologists
2. are cognitively competent
3. can communicate with data collectors
4. age ≧ 20
5. have a designated family caregiver who agrees to participate in the survey.

Family caregivers will be recruited if they are:

1. family members of cancer patients with a terminally ill disease as defined by this proposed study
2. identified by the patients as the persons primarily conduct the patient's care needs without receiving financial reimbursement for the care they provide
3. age≧20
4. who agree to participate and can communicate with data collectors.

Exclusion Criteria:

1. Patients or family caregivers who are not able to communicate coherently.
2. Only one party in the patient-family dyad agrees to participate.