Proactive symptom monitoring to initiate timely palliative care for patients with advanced cancer: a randomized controlled trial

doi:10.21203/rs.3.rs-4995145/v1

Purpose

To deliver timely palliative care in response to supportive and palliative care needs as they arise, we developed a model called ‘Supportive and Palliative care Review Kit in Locations Everywhere’ (SPARKLE), which comprises regular remote symptom monitoring using the Integrated Palliative care Outcome Scale (IPOS); early identification and prompt treatment of palliative care symptoms and concerns identified; referral to specialist palliative care if follow-up is required.

Methods

A prospective randomized controlled trial of SPARKLE versus usual care was conducted among patients with advanced cancers. The primary endpoint was Functional Assessment of Cancer Therapy – General (FACT-G) scores at 16 weeks post randomization.

Results

A total of 239 patients were randomized – 119 patients to usual care and 120 patients to SPARKLE intervention. There was no significant difference in total FACT-G score (baseline-adjusted difference 0.8, 95% CI -3.5 to 5.1, p = 0.73). Physical wellbeing was better in the SPARKLE intervention group compared to the usual care group (baseline-adjusted difference 1.9, 95%CI 0.4 to 3.4, p = 0.01). There were no significant differences in the other domain scores. There were no differences in palliative care referrals, occurrence of emergency department visits and/or hospital admissions.

Conclusion

Although there was no significant difference in overall quality of life, better physical symptom control could have been achieved through proactive identification and treatment of symptoms and concerns by the SPARKLE nurse. Further studies are warranted to evaluate how various components of palliative care service models impact clinical outcomes.

Trial registration

ClinicalTrials.gov identifier: NCT04242251

Palliative Care

Neoplasms

Patient Reported Outcome Measure

Hospice and Palliative Care Nursing

Quality of Life

To deliver timely palliative care, we evaluated SPARKLE – a model of care with proactive screening and triage assessment of palliative care concerns. Although there were no significant benefits in overall quality of life, those in the SPARKLE intervention group had better physical wellbeing at 16 weeks. This approach could be iterated and evaluated further to facilitate improvement of palliative care models to be more scalable yet still effective in improving patient outcomes.

When provided alongside standard oncology care, early access to palliative care improves quality of life and reduces acute healthcare utilisation and costs of patients with advanced cancers.[1, 7, 20, 24, 31] Both the American Society for Clinical Oncology (ASCO) and the European Society for Medical Oncology (ESMO) recommend that all patients with advanced cancer should receive palliative care from the time of diagnosis.[6, 12] However, this has not been implemented in routine clinical practice – a recent meta-analysis reported a median time from first palliative care consult to death of just 18.9 days.[13] Due to limited specialist palliative care resources, it is unrealistic to provide early palliative care to all patients with advanced disease.[15, 22, 23] Moreover, it may not be high value to do so because the palliative care needs of patients fluctuate over time and not all patients require palliative care early in their illness trajectory.[9] The research imperative has therefore shifted from early palliative care to timely palliative care.[11]

To deliver timely palliative care in response to supportive and palliative care needs as they arise, we developed an outpatient-based palliative care model, termed ‘Supportive and Palliative care Review Kit in Locations Everywhere’ (SPARKLE). SPARKLE is based on regular remote symptom monitoring using the Integrated Palliative care Outcome Scale (IPOS), early identification and prompt treatment of palliative care symptoms and concerns identified. SPARKLE starts with proactive monitoring for palliative care concerns in all patients with advanced cancer, escalating to a specialist palliative care consultation only when identified physical, psychological or social problems and concerns need addressing.

We conducted a randomized controlled trial to compare clinical outcomes between patients with advanced cancer receiving versus not receiving care from the SPARKLE care model. In the current oncologist-led model, there is no systematic screening for palliative care concerns, and only a minority of patients would be referred for a palliative care consultation. Therefore, we hypothesize that compared to usual care of no routine screening for palliative care access, the additional SPARKLE care model would result in improved quality of life in patients with advanced cancer, more palliative care referrals and reduced acute healthcare utilization within 16 weeks.

Study design

A prospective randomized controlled trial design was used to test the effectiveness of the SPARKLE model. We recruited patients with advanced cancers, defined as stage 4 solid tumors including lung, colorectal, nasopharyngeal and pancreatic cancers. Enrolled participants were randomized to usual care or SPARKLE intervention group in a 1:1 ratio using randomized permuted blocks method generated by an independent statistician, with block size kept unknown to the clinical investigators and site personnel as per ICH guidelines E9. (ClinicalTrials.gov identifier: NCT04242251). This study was conducted in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and received ethical approval from the SingHealth Centralized Institutional Review Board (CIRB 2019/2963).

Setting and participants

The study was conducted in the National Cancer Centre of Singapore, which has over 140,000 patient attendances a year, covering 65% of all cancer patients in the public sector in Singapore. Eligibility criteria were age 21 years and above and diagnosis of stage 4 solid tumor. Exclusion criteria were inability to complete the online patient-reported outcome measures and currently under the care of a specialist palliative care service.

Recruitment

The study team screened outpatient clinic lists of participating oncologists for eligible patient participants. Potential patient participants were approached in the oncology outpatient clinics to verify eligibility and obtain informed consent.

Usual care

In usual care, patients are managed by a primary oncologist who focuses on cancer-directed therapy such as chemotherapy. Although specialist training in oncology includes holistic care and communication skills, none of the medical oncologists in our healthcare institution are dually trained in specialist palliative medicine. Nonetheless, the oncologist may perform ad-hoc symptom assessment and treat any palliative care symptoms and concerns identified. If deemed beneficial, the oncologist may also decide to initiate a referral to a specialist palliative care service – either a palliative care outpatient clinic or home hospice. In Singapore, the term “hospice” is generally used to refer to palliative care services based in the community; patients may receive home hospice care concurrently with active cancer treatment. In our study setting, just under half of patients with advanced cancer receive specialist palliative care, with a median time from first specialist palliative care review to death of around 30 days.[27]

SPARKLE intervention group

In the intervention group, patients received the SPARKLE model of care in addition to usual oncologist-led care. In SPARKLE model, patients received a weekly invitation via a Short Message Service (SMS) text on their phone to complete a self-reported online Integrated Palliative care Outcome Scale (IPOS) questionnaire for 16 weeks. The IPOS is a 17-item questionnaire that assesses how a person is affected by symptoms and concerns experienced by patients with advanced cancer in three domains of physical (10 items), emotional (4 items), and communication and practical issues (3 items). There are two additional items: i) at the start of the questionnaire is a free text question about what their main problems and concerns have been; ii) at the end of the questionnaire is a question about how the patient completed the questionnaire – on their own or with help. Each item is rated on a Likert scale of 0 to 4. Higher scores indicate that symptoms and concerns affect the person more.[16] Both the English and translated Chinese versions of IPOS have been validated in Singapore. [14]

Depending on how participants completed the IPOS items, the SPARKLE nurse would take certain actions according to the workflow shown in Fig. 1. For this study, the SPARKLE nurse role was carried out by a palliative care nurse (NXH) with 3 years of working experience in specialist palliative care. According to the workflow, if any symptoms and concerns were identified by the online IPOS survey, the SPARKLE nurse would make a phone call to the patient within 3 or 5 working days to conduct a triage assessment of the identified symptoms and concerns.

If the presence of symptoms and concerns is confirmed, the SPARKLE nurse would initiate treatment of symptoms or addressing psychosocial concerns according to structured clinical pathways. If the symptoms and concerns were affecting the patient mildly, the nurse would provide management advice to the patient over the phone. If there were any symptoms and concerns affecting the patient moderately, severely or overwhelmingly, the patient be offered an in-person outpatient clinic consultation by the SPARKLE nurse, who would perform a comprehensive palliative care assessment and start appropriate treatment. In our setting, palliative care nurses are trained to assess and initiate treatment for palliative care issues. For the study, structured clinical pathways for each of the items in the IPOS questionnaire also guided clinical management by the SPARKLE nurse.

After initial assessment and management by the SPARKLE nurse, if any symptoms and concerns are not fully resolved or if further follow up is required, the patient would be referred to a specialist palliative care service – either a specialist palliative care physician outpatient clinic or home hospice. In our setting, ‘home hospice’ refers to specialist palliative care services based in the community where an interdisciplinary palliative care team would conduct home visits to review the patient in their own home.

Study measures and data collection

All participants completed the Functional Assessment of Cancer Therapy – General (FACT-G) questionnaire at baseline and 16 weeks from randomization. The FACT-G total score at 16 weeks was chosen as the primary endpoint as it is an established measure of health-related quality of life that has been well-validated in English and Chinese both internationally and locally for patients with many kinds of cancer.[2, 4, 5, 25] It covers four quality of life domains: physical well-being (7 items), social/family well-being (7 items), emotional well-being (6 items) and functional well-being (7 items), and each item has response choices in a 5-point Likert-type scale. Demographics, clinical characteristics and healthcare utilization data were extracted from the electronic medical records.

At the end of the 16-week period, patients randomized to the SPARKLE intervention group were invited to complete a feedback survey which included three items specifically about their satisfaction, perceived usefulness and disruptiveness of the SPARKLE model. Each item was rated on a 5-point Likert scale of “strongly disagree” to “strongly agree”.

Statistical analysis and sample size calculation

Analysis was by intention to treat, with patients analyzed in the arms to which they had been randomized. The official scoring method for FACT–G was used: After recoding negatively worded items so that for all items a higher score represents better quality of life, the scores for the four domains were obtained by summing individual item scores within each domain, subject to imputation for item nonresponse by the “half rule.” The total FACT–G score was computed by summing the four domain scores.

The FACT-G total score (primary endpoint) and domain scores (secondary endpoint) at 16 weeks were compared between the usual care and SPARKLE intervention groups with the Student’s T- test. Linear regression was used to compare the scores with the baseline score and other demographic and clinical variables as a covariates. The occurrence of palliative care referral, at least one emergency department visit and/or hospital admission within 16 weeks from enrolment were compared using chi-squared test. Results from the feedback survey administered at the end of 16 weeks were tabulated to show frequency and percentage of responses to the three items relating to the SPARKLE model. All analyses were performed using SAS OnDemand for Academics (SAS Institute Inc).

The target recruitment was 240 patients. Allowing for around 15% attrition rate, this would yield a sample size of 200 patient participants (100 per group), which would give 80% power, at 5% 2-sided type 1 error rate, to detect an effect size of 0.4 in the mean difference in the primary outcome of FACT-G total scores between the two groups. Based on our previous studies, the standard deviation of FACT-G in a similar population is around 18.[28] An effect size of 0.4 will then equate to an absolute difference in FACT-G score of 4.5, which has been shown to be a clinically significant difference.[3]

From February 2021 to January 2023, 240 patients provided written informed consent and completed the baseline FACT-G measure. One patient was subsequently excluded prior to randomization as the patient was referred to a specialist palliative care service; 119 patients were randomized to usual care and 120 patients were randomized to SPARKLE intervention. (Fig. 2)

There were no major differences in the demographic or clinical characteristics between groups at baseline. Around a third had gastrointestinal cancer and around a quarter had lung cancer; there were no patients with breast or gynecological cancer because we deconflicted with a concurrent study that was recruiting these patients. This study also involved routine proactive screening for problems causing distress, and would have contaminated the usual care arm. Majority had received chemotherapy at the time of enrolment. Quality of life were similar between both groups at baseline. (Table 1)

There was no significant difference in the total FACT-G scores between the SPARKLE intervention group and the usual care group (baseline-adjusted difference 0.8, 95% CI -3.5 to 5.1, p=0.73). Physical wellbeing was better in the SPARKLE intervention group compared to the usual care group (baseline-adjusted difference 1.9, 95%CI 0.4 to 3.4, p=0.01). There were no significant differences in the other domain scores. (Table 2) There were no significant differences in baseline FACT-G scores between the SPARKLE intervention and usual care groups for patients who were lost to follow-up.

During the 16 weeks after study enrolment, there were no differences between the SPARKLE intervention group and usual care group in palliative care referrals, occurrence of emergency department visits and/or hospital admissions. (Table 3) In both groups, around 7% of the patients were referred to specialist palliative care.

Table 4 shows the amount of time spent by the SPARKLE nurse on various activities related to reviewing and acting on the IPOS questionnaires submitted by patients. The most amount of time was spent on reviewing the IPOS responses and clinical notes, conducting phone consults, and documenting in the clinical notes.

Most patients were satisfied with the SPARKLE model and felt that it was useful. Only 3% felt that being contacted by the SPARKLE nurse was disruptive to their life. (Table 5)

In our current climate where early palliative care is not routinely implemented in clinical practice partly due to scarce specialist palliative care resources, one possible way to direct limited resources to areas of greatest need is using symptom monitoring to identify patients with high symptom burden and who may therefore benefit most from timely palliative care.[17] This approach of symptom screening and targeted early palliative (STEP) was reported by Zimmermann et al to be both feasible and able delineate patients who subsequently had worse quality of life, depression and symptom control.[30] Even though the planned randomized trial of STEP was halted early due to COVID-19 pandemic, results favored STEP.[29]

Both the STEP model reported by Zimmermann et al and our SPARKLE stepped palliative care model recruited patients with advanced cancer in the outpatient medical oncology clinic setting. The main differences are: In STEP, symptom screening was done in both the STEP and usual care arms with electronic Edmonton Symptom Assessment System-revised version (ESAS-r), which was administered routinely at each outpatient visit in the waiting room of the oncology clinic before appointments. In SPARKLE, symptom screening was done only in the SPARKLE arm, with IPOS which was administered remotely via SMS text to the patient’s phone once per week regardless of whether the patient had an oncology clinic appointment or not. Therefore, STEP assesses the effect of a palliative care triage nurse; SPARKLE assesses the combined effect of both symptom screening and a palliative care nurse.

The main components of the SPARKLE model were proactive weekly screening for palliative care concerns; triage followed by comprehensive assessment and initial management of identified symptoms and concerns by a palliative care nurse; and formal referral to specialist palliative care services if further review or follow up is required. Our main study finding was that, compared to usual care, there was no significant difference in total FACT-G scores between the SPARKLE intervention group and usual care group. The lack of benefit could be due to a number of reasons: First, the SPARKLE model comprised a palliative care nurse rather than an interdisciplinary palliative care team, which have resulted in more positive outcomes in the literature compared to nurse-led interventions.[10] Second, the SPARKLE model did not include any form of palliative care review if the IPOS responses did not flag up the presence of symptoms and concerns. Third, the follow-up time period of 16 weeks may be too short to observe significant improvements in quality of life.

The SPARKLE model was associated with better physical wellbeing. Using the benchmark of 5% of the score range for meaningful improvement, the baseline-adjusted difference in physical wellbeing score of 1.9 out of a score range of 28 (7%) is clinically significant.[18] Better physical symptom control was probably achieved through proactive identification and treatment of symptoms and concerns by the SPARKLE nurse.

However, there were no significant differences in the other domains of emotional, functional and social wellbeing nor overall quality of life. This is despite the presence of symptoms and concerns in these psychosocial domains according to the weekly IPOS survey responses.[26] One of the possible reasons for the contrast between improvement in physical wellbeing and lack of improvement in the other psychosocial domains could be related to the background of the SPARKLE nurse. As a palliative care nurse who was mainly reviewing patients in the last weeks of life prior to joining this study, it could be that the skills required to manage physical symptoms such as pain and low appetite are similar whether the patient is in the last weeks of life or earlier in their illness trajectory. On the other hand, the types of social, emotional and functional symptoms and concerns experienced in the earlier phases of the illness trajectory may require a different approach from psychosocial symptoms and concerns in the last weeks of life.[8]

This has implications for early palliative care programs that are built on existing palliative care teams. Additional training may be required to equip healthcare professionals with competencies required to improve the psychosocial domains of quality of life when patients still have a longer expected survival of months to years. There could also be regular case discussions with a medical social worker and standardized workflows to refer these patients to relevant supportive care services such counselling or support groups.

In the 16 weeks after study enrolment, only a minority of around 7% of the patients were formally referred to specialist palliative care and there were no differences between the two groups. This was lower than expected. In a study of stepped palliative care where a higher intensity of palliative care visits was triggered by deterioration in quality of life as measured by FACT-L, 26.4% of patients were stepped up by 24 weeks and 36.4% were stepped up by 48 weeks.[21] In our study, perhaps more patients would have been referred to specialist palliative care if the weekly IPOS questionnaires and the SPARKLE intervention continued beyond 16 weeks. Furthermore, the completion rate of weekly IPOS was suboptimal, with only 48% of patients completing more than 70% (12 or more out of 16) of the questionnaires.[26] Future iteration of the SPARKLE model could include other triggers for a phone call assessment from the palliative care nurse, for example, non-completion of questionnaires or change in cancer treatment due to disease progression.

In conclusion, SPARKLE – a model of care with proactive screening and triage assessment of palliative care concerns – did not result in significant benefits in overall quality of life, although those in the SPARKLE intervention group had better physical wellbeing at 16 weeks. Although the delivery of only selected components of proactive screening and targeted triage assessment in SPARKLE did not improve overall quality of life, the improvement in physical wellbeing may provide some motivation to explore this approach with further iteration and evaluation. More studies are warranted to evaluate how various components of palliative care service models impact clinical outcomes; decoding the core components of palliative care may facilitate the redesign the palliative care models that are more scalable yet still effective in improving patient outcomes.

Limitations

A limitation of this study is the patients were eligible for study participation if they had a diagnosis of stage 4 cancer, regardless of time of diagnosis. Consequently, recruited patients had a variable duration since stage 4 cancer diagnosis, which may affect interpretation of the study results. However, this may reflect real-world conditions where the screening approach could be implemented without additional administrative burden of determining time duration since diagnosis. Another limitation is that the SPARKLE model is a multi-component complex intervention and it is unclear which components worked for whom and in what circumstances.[19] This could be explored in future studies.

Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Funding

This study was supported by funds from the Singapore Ministry of Health’s (MOH) National Medical Research Council (NMRC) under the NMRC Transition Award (grant number: MOH-000336) and NCCS Cancer Fund. The funding sources were not involved in the study design in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

Author Contribution

GMY obtained grant funding, conceptualized the study, conducted data analysis, wrote the original draft and reviewed the final draft of the manuscript.YK conceptualized the study, edited the draft and reviewed the final draft of the manuscript.XHN conceptualized the study, participated in the study intervention as the SPARKLE nurse, edited the draft and reviewed the final draft of the manuscript.PSHN conceptualized the study, edited the draft and reviewed the final draft of the manuscript.YBC conceptualized the study, supervised the data analysis, edited the draft and reviewed the final draft of the manuscript.

Acknowledgements

This study was supported by funds from the Singapore Ministry of Health’s (MOH) National Medical Research Council (NMRC) under the NMRC Transition Award (grant number: MOH-000336) and NCCS Cancer Fund. The funding sources were not involved in the study design in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

Data Availability

The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request. Data are located in controlled access data storage at National Cancer Centre Singapore.

Bakitas M, Tosteson T, Li Z, Lyons K, Hull J, Li Z, Dionne-Odom J, Frost J, Dragnev K, Hegel M, Azuero A, Ahles T (2015) Early Versus Delayed Initiation of Concurrent Palliative Oncology Care: Patient Outcomes in the ENABLE III Randomized Controlled Trial. J Clin Oncol 33: 1438–1445
Brucker P, Yost K, Cashy J, Webster K, Cella D (2005) General population and cancer patient norms for the Functional Assessment of Cancer Therapy-General (FACT-G). Eval Health Prof 28: 192–211
Cella D, Eton D, Lai J, Peterman A, Merkel D (2002) Combining anchor and distribution-based methods to derive minimal clinically important differences on the Functional Assessment of Cancer Therapy (FACT) anemia and fatigue scales. J Pain Symptom Manage 24: 547–561
Cella D, Tulsky D, Gray G, Sarafian B, Linn E, Bonomi A, Silberman M, Yellen S, Winicour P, Brannon J, et a (1993) The Functional Assessment of Cancer Therapy scale: development and validation of the general measure. J Clin Oncol 11: 570–579
Cheung Y, Goh C, Thumboo J, Khoo K, Wee J (2005) Variability and sample size requirements of quality-of-life measures: a randomized study of three major questionnaires. J Clin Oncol 23: 4936–4944
Ferrell BR, Temel JS, Temin S, Smith TJ (2017) Integration of Palliative Care Into Standard Oncology Care: ASCO Clinical Practice Guideline Update Summary J Oncol Pract 13: 119–121
Haun MW, Estel S, Rucker G, Friederich HC, Villalobos M, Thomas M, Hartmann M (2017) Early palliative care for adults with advanced cancer Cochrane Database Syst Rev 6: CD011129
Hoerger M, Greer JA, Jackson VA, Park ER, Pirl WF, El-Jawahri A, Gallagher ER, Hagan T, Jacobsen J, Perry LM, Temel JS (2018) Defining the Elements of Early Palliative Care That Are Associated With Patient-Reported Outcomes and the Delivery of End-of-Life Care J Clin Oncol 36: 1096–1102
Hui D, Bruera E (2016) Integrating palliative care into the trajectory of cancer care Nat Rev Clin Oncol 13: 159–171
Hui D, Bruera E (2020) Models of Palliative Care Delivery for Patients With Cancer J Clin Oncol 38: 852–865
Hui D, Heung Y, Bruera E (2022) Timely Palliative Care: Personalizing the Process of Referral Cancers (Basel) 14
Jordan K, Aapro M, Kaasa S, Ripamonti CI, Scotte F, Strasser F, Young A, Bruera E, Herrstedt J, Keefe D, Laird B, Walsh D, Douillard JY, Cervantes A (2018) European Society for Medical Oncology (ESMO) position paper on supportive and palliative care Ann Oncol 29: 36–43
Jordan RI, Allsop MJ, ElMokhallalati Y, Jackson CE, Edwards HL, Chapman EJ, Deliens L, Bennett MI (2020) Duration of palliative care before death in international routine practice: a systematic review and meta-analysis BMC Med 18: 368
Long VJE, Cheung YB, Qu D, Lim K, Lee G, Yee ACP, Guo P, Harding R, Yang GM (2021) Validity and reliability of the English and translated Chinese versions of the Integrated Palliative care Outcome Scale (IPOS) in Singapore BMC Palliat Care 20: 40
Mulvey CL, Smith TJ, Gourin CG (2016) Use of inpatient palliative care services in patients with metastatic incurable head and neck cancer Head Neck 38: 355–363
Murtagh FE, Ramsenthaler C, Firth A, Groeneveld EI, Lovell N, Simon ST, Denzel J, Guo P, Bernhardt F, Schildmann E, van Oorschot B, Hodiamont F, Streitwieser S, Higginson IJ, Bausewein C (2019) A brief, patient- and proxy-reported outcome measure in advanced illness: Validity, reliability and responsiveness of the Integrated Palliative care Outcome Scale (IPOS) Palliat Med 33: 1045–1057
Petrillo LA, Jones KF, El-Jawahri A, Sanders J, Greer JA, Temel JS (2024) Why and How to Integrate Early Palliative Care Into Cutting-Edge Personalized Cancer Care Am Soc Clin Oncol Educ Book 44: e100038
Ringash J, O'Sullivan B, Bezjak A, Redelmeier DA (2007) Interpreting clinically significant changes in patient-reported outcomes Cancer 110: 196–202
Skivington K, Matthews L, Simpson SA, Craig P, Baird J, Blazeby JM, Boyd KA, Craig N, French DP, McIntosh E, Petticrew M, Rycroft-Malone J, White M, Moore L (2021) A new framework for developing and evaluating complex interventions: update of Medical Research Council guidance BMJ 374: n2061
Temel J, Greer J, Muzikansky A, Gallagher E, Admane S, Jackson V, Dahlin C, Blinderman C, Jacobsen J, Pirl W, Billings J, Lynch T (2010) Early palliative care for patients with metastatic non-small-cell lung cancer. N Engl J Med 363: 733–742
Temel JS, Jackson VA, El-Jawahri A, Rinaldi SP, Petrillo LA, Kumar P, McGrath KA, LeBlanc TW, Kamal AH, Jones CA, Rabideau DJ, Horick N, Pintro K, Gallagher Medeiros ER, Post KE, Greer JA (2024) Stepped Palliative Care for Patients With Advanced Lung Cancer: A Randomized Clinical Trial JAMA
Teo I, Singh R, Malhotra C, Ozdemir S, Dent RA, Kumarakulasinghe NB, Yeo WL, Cheung YB, Malhotra R, Kanesvaran R, Yee ACP, Chan N, Wu HY, Chin SM, Allyn HYM, Yang GM, Neo PSH, Nadkarni NV, Harding R, Finkelstein EA (2018) Cost of Medical Care of Patients with Advanced Serious Illness in Singapore (COMPASS): prospective cohort study protocol BMC Cancer 18: 459
Uppal S, Rice LW, Beniwal A, Spencer RJ (2016) Trends in hospice discharge, documented inpatient palliative care services and inpatient mortality in ovarian carcinoma Gynecol Oncol 143: 371–378
Vanbutsele G, Pardon K, Van Belle S, Surmont V, De Laat M, Colman R, Eecloo K, Cocquyt V, Geboes K, Deliens L (2018) Effect of early and systematic integration of palliative care in patients with advanced cancer: a randomised controlled trial The Lancet Oncology 19: 394–404
Webster K, Cella D, Yost K (2003) The Functional Assessment of Chronic Illness Therapy (FACIT) Measurement System: properties, applications, and interpretation. Health Qual Life Outcomes 1: 79
Yang GM, Lee Y, Ke Y, Neo PSH, Cheung YB (2024) Feasibility of Weekly Integrated Palliative Care Outcome Scale (IPOS) Within a Stepped Care Model J Pain Symptom Manage
Yang GM, Lim C, Zhuang Q, Ong WY (2021) Prevalence and timing of specialist palliative care access among advanced cancer patients and association with hospital death Proceedings of Singapore Healthcare 31: 20101058211055279
Yang GM, Tan YY, Cheung YB, Lye WK, Lim SHA, Ng WR, Puchalski C, Neo PSH (2017) Effect of a spiritual care training program for staff on patient outcomes Palliat Support Care 15: 434–443
Zimmermann C, Pope A, Hannon B, Bedard PL, Rodin G, Dhani N, Li M, Herx L, Krzyzanowska MK, Howell D, Knox JJ, Leighl NB, Sridhar S, Oza AM, Lheureux S, Booth CM, Liu G, Castro JA, Swami N, Sue AQR, Rydall A, Le LW (2023) Symptom screening with Targeted Early Palliative care (STEP) versus usual care for patients with advanced cancer: a mixed methods study Support Care Cancer 31: 404
Zimmermann C, Pope A, Hannon B, Krzyzanowska MK, Rodin G, Li M, Howell D, Knox JJ, Leighl NB, Sridhar S, Oza AM, Prince R, Lheureux S, Hansen AR, Rydall A, Chow B, Herx L, Booth CM, Dudgeon D, Dhani N, Liu G, Bedard PL, Mathews J, Swami N, Le LW (2022) Phase II Trial of Symptom Screening With Targeted Early Palliative Care for Patients With Advanced Cancer J Natl Compr Canc Netw 20: 361–370.e365
Zimmermann C, Swami N, Krzyzanowska M, Hannon B, Leighl N, Oza A, Moore M, Rydall A, Rodin G, Tannock I, Donner A, Lo C (2014) Early palliative care for patients with advanced cancer: a cluster-randomised controlled trial. Lancet 383: 1721–1730

Table 1 – Baseline characteristics

Usual care (n=119)

SPARKLE intervention (n=120)

Male gender, n (%)

73 (61.3)

69 (57.5)

Race, n (%)

Chinese

Malay

Indian

Others

103 (86.6)

12 (10.1)

4 (3.4)

0

99 (82.5)

13 (10.8)

2 (1.7)

6 (5.0)

Age, years, mean (SD)

60.6 (11.2)

61.2 (11.8)

Marital status, n (%)

Married

Single

Divorced/ separated

Widowed

88 (74.0)

21 (17.7)

7 (5.9)

3 (2.5)

85 (70.8)

18 (15.0)

9 (7.5)

8 (6.7)

Highest education level, n (%)

No formal education

Primary school

Secondary school

Post-secondary school

(e.g. diploma, degree, etc.)

1 (0.8)

16 (13.5)

56 (47.1)

46 (38.7)

7 (5.8)

16 (13.3)

49 (40.8)

48 (40.0)

Employment status, n (%)

Working full-time

Working part-time

Retired

Homemaker

Unemployed

42 (35.3)

4 (3.4)

51 (42.9)

10 (8.4)

12 (10.1)

39 (32.5)

5 (4.2)

46 (38.3)

13 (10.8)

17 (14.2)

Functional status (ECOG), n (%)

0

1

2

3

54 (45.4)

59 (49.6)

6 (5.0)

0

45 (37.5)

66 (55.0)

6 (5.0)

3 (2.5)

Primary cancer, n (%)

Gastrointestinal

Lung

Head and neck

Genitourinary

Prostate

Others

45 (37.8)

29 (24.4)

11 (9.2)

9 (7.6)

14 (11.8)

11 (9.2)

42 (35.0)

28 (23.3)

13 (10.8)

11 (11.7)

14 (11.7)

12 (10.0)

Time from advanced cancer diagnosis to enrolment, mean days (SD)

Anticancer therapy at baseline, n (%)

Radiotherapy

Chemotherapy

Immunotherapy

1019 (1068)

44 (37.0)

101 (84.9)

32 (26.9)

864 (998)

56 (46.7)

98 (81.7)

28 (23.3)

Quality of life at baseline (FACT-G) [score range], mean (SD)

Physical wellbeing [0-28]

21.2 (5.9)

20.7 (6.1)

Social wellbeing [0-28]

21.8 (5.7)

21.3 (5.5)

Emotional wellbeing [0-24]

17.6 (4.9)

17.7 (5.3)

Functional wellbeing [0-28]

18.5 (6.7)

19.2 (6.8)

Total FACT-G score [0-108]

79.2 (17.4)

78.9 (17.9)

Abbreviations: ECOG. Eastern Cooperative Oncology Group; FACT-G, Functional Assessment of Cancer Therapy – General; SD, standard deviation

Table 2 – Means and differences (baseline-adjusted) in total and domain scores of FACT-G at 16 weeks

FACT-G [score range]	Mean score (95% CI)		P value ^a	Baseline-adjusted difference (95% CI)	P value ^b
FACT-G [score range]	Usual Care (n=73)	SPARKLE intervention (n=84)	P value ^a	Baseline-adjusted difference (95% CI)	P value ^b
Physical wellbeing [0-28]	20.2 (18.8 to 21.6)	22.0 (20.9 to 23.1)	0.05	1.9 (0.4 to 3.4)	0.01
Social wellbeing [0-28]	20.8 (19.6 to 22.1)	19.8 (18.4 to 21.1)	0.24	-0.9 (-2.5 to 0.6)	0.24
Emotional wellbeing [0-24]	17.3 (16.1 to 18.5)	17.6 (16.5 to 18.7)	0.68	0.3 (-1.0 to 1.5)	0.66
Functional wellbeing [0-28]	18.4 (16.9 to 19.8)	17.7 (16.2 to 19.2)	0.54	-0.6 (-2.3 to 1.1)	0.49
Total FACT-G score [0-108]	76.7 (72.6 to 80.8)	77.1 (73.3 to 80.9)	0.89	0.8 (-3.5 to 5.1)	0.73

Abbreviations: CI, confidence interval; FACT-G, Functional Assessment of Cancer Therapy – General

^aStudent’s T-test

^b Linear regression with baseline score as covariate

Table 3 – Healthcare utilisation between enrolment and 16 weeks

N (%)	Usual care (n=119)	SPARKLE intervention (n=120)	p-value^a
Palliative care referral	8 (6.7)	9 (7.5)	0.82
At least 1 emergency department visit	40 (33.6)	52 (43.3)	0.23
At least 1 hospital admission	27 (22.7)	31 (25.8)	0.57

^achi-square test

Table 4 – SPARKLE nurse activity

Activity	Number of episodes	Time spent on each episode (minutes)	Total amount of time spent (hours)
Review IPOS and clinical notes	1072	6	109.1
Phone consult	527	22	195.3
Face-to-face consult	95	23	36.3
Sending and receiving text messages	373	9	56.1
Sending and receiving emails	107	10	18.2
Documenting in clinical notes	778	17	222.2
Discuss case with doctor	31	11	5.8
Arrange for clinic appointments	11	16	3.0
Arrange for referrals	8	12	1.6

Table 5 – Responses to feedback survey (n=79)

Frequency (%)	Strongly disagree	Disagree	Neutral	Agree	Strongly Agree
I feel satisfied with the SPARKLE model of care	2 (3)	1 (1)	11 (14)	37 (47)	28 (35)
I feel that the SPARKLE model of care is useful	1 (1)	3 (4)	14 (18)	33 (42)	28 (35)
Being contacted by the SPARKLE nurse was disruptive to my life	35 (44)	29 (37)	13 (16)	0	2 (3)

No competing interests reported.

Proactive symptom monitoring to initiate timely palliative care for patients with advanced cancer: a randomized controlled trial

Status:

Version 1

Abstract

Purpose

Methods

Results

Conclusion

Trial registration

Figures

Relevance to inform research, policies and/or programs

Background

Methods

Study design

Setting and participants

Recruitment

Usual care

SPARKLE intervention group

Study measures and data collection

Statistical analysis and sample size calculation

Results

Discussion

Limitations

Declarations

Competing interests

Funding

Author Contribution

Acknowledgements

Data Availability

References

Tables

Additional Declarations

Status:

Version 1