Electronic Patient-Reported Outcome–Based Symptom Management Versus Usual Care After Lung Cancer Surgery: Long-Term Results of a Multicenter, Randomized, Controlled Trial

Abstract

Clinical trials frequently include multiple end points that mature at different times. The initial report, typically based on the primary end point, may be published when key planned co-primary or secondary analyses are not yet available. Clinical Trial Updates provide an opportunity to disseminate additional results from studies, published in JCO or elsewhere, for which the primary end point has already been reported.

We previously reported superior symptom control of electronic patient-reported outcome (ePRO)–based symptom management after lung cancer surgery for up to 1 month postdischarge. Here, we present the long-term results (1-12 months) of this multicenter, randomized trial, where patients were assigned 1:1 to receive postoperative ePRO-based symptom management or usual care daily postsurgery, twice weekly postdischarge until 1 month, and at 3, 6, 9, and 12 months postdischarge. Long-term patient-reported outcomes were assessed with MD Anderson Symptom Inventory-Lung Cancer module. Per-protocol analyses were performed with 55 patients in the ePRO group and 57 in the usual care group. At 12 months postdischarge, the ePRO group reported significantly fewer symptom threshold events (any of the five target symptom scored ≥4; median [IQR], 0 [0-0] v 0 [0-1]; P = .040) than the usual care group. From 1 to 12 months postdischarge, the ePRO group consistently reported significantly lower composite scores for physical interference (estimate, –0.86 [95% CI, –1.32 to –0.39]) and affective interference (estimate, –0.70 [95% CI, –1.14 to –0.26]). Early intensive ePRO-based symptom management after lung cancer surgery reduced symptom burden and improved functional status for up to 1 year postdischarge, supporting its integration into standard care.

INTRODUCTION

Most patients with lung cancer experience a variety of moderate-to-severe symptoms within the first year postsurgery.¹ However, under the current standard of care, symptom monitoring after discharge is often reactive and inefficient.² Electronic patient-reported outcome (ePRO)–based symptom monitoring, allowing for patient-centered, proactive, real-time, and remote symptom management, offers a way to address these challenges.³ Our primary analysis of a randomized trial revealed that intensive ePRO-based symptom management after lung cancer surgery led to a lower symptom burden than usual care for up to 1 month postdischarge.⁴ Here, we report the findings from 1 to 12 months postdischarge to address the gap in knowledge regarding the long-term effects of ePRO-based symptom management.

METHODS

Study Design

This multicenter, randomized trial was conducted at one tertiary cancer hospital and two tertiary general hospitals in China.⁴ Patients were enrolled between November 17, 2019, and July 15, 2020. Data collection for long-term outcomes was completed on August 7, 2021. The initial protocol and short-term results of this trial were reported previously.^4,5

Patients

Patients with clinical stage I-IIIA disease according to the eighth edition staging manual for lung cancer, age 18-75 years, consent for surgery, and ability and willingness to use personal electronic devices to report symptoms were eligible.⁴ Patients were excluded if they had a history of neoadjuvant therapy, malignancy, chest surgery, or chronic analgesic use or were unable to understand the study requirements. The study protocol was approved by the ethics committees of each study center. All patients provided written informed consent.

Procedures

Eligible patients were randomly assigned 1:1 to receive postoperative ePRO-based symptom management or usual care, stratified by participating hospitals. Patients self-reported their symptoms using the Chinese version of the MD Anderson Symptom Inventory-Lung Cancer (MDASI-LC) module^6,7 through their electronic devices at random assignment (presurgery); daily during postoperative hospitalization; twice weekly postdischarge until 4 weeks; and at 3, 6, 9, and 12 months postdischarge. Symptom reporting was discontinued when adjuvant therapy was initiated. The MDASI-LC contains 16 symptom items rated from 0 (no symptom) to 10 (most severe symptom) and six functional items also rated from 0 (no interference) to 10 (complete interference).

For the ePRO group, if any of the five target symptoms (pain, fatigue, disturbed sleep, shortness of breath, or coughing) scored ≥4 (indicating moderate-to-severe severity), the ePRO system immediately alerted the surgeon. Responses during hospitalization occurred during daily ward rounds, whereas postdischarge responses used text messages, voice messages, or calls. Surgeons had a 24-hour response window, guided by clinical guidelines. The usual care group received standard postoperative care. Patients completed the electronic MDASI-LC scale, but their reported symptoms did not trigger alerts.

Outcome Measures

The primary outcome was the number of symptom threshold events at 12 months postdischarge. Symptom threshold event refers to any target symptom (pain, fatigue, disturbed sleep, shortness of breath, and coughing) with a score of ≥4. The secondary outcomes included the composite physical interference score (average score of general activity, work, and walking), composite affective interference score (average score of mood, relations with others, and enjoyment of life), 16 single symptom item score, and six single functional item score in the MDASI-LC.

Statistical Analysis

MDASI-LC data at 1-(baseline), 3-, 6-, 9-, and 12-month postdischarge time points were analyzed, excluding patients who met the withdrawal criteria or received postoperative adjuvant therapy (Appendix Fig A1, online only). The number of symptom threshold events was compared between groups using the Wilcoxon-Mann-Whitney U test. The type I error rate for a two-sided test was set at 0.05. MDASI-LC composite and single item scores over time were analyzed using linear mixed-effects models. Two-sided P values for MDASI-LC composite and single item scores over time were adjusted for multiple comparisons using the Benjamini-Hochberg procedure, with a false discovery rate of 0.05. The hospital type (cancer hospital or general hospital) was included in all models for adjustment.⁸

RESULTS

Patient Characteristics

We screened 418 patients. with 166 (40%) patients eligible for inclusion.⁴ After excluding 54 patients, there were 55 patients in the ePRO group and 57 patients in the usual care group for the long-term outcome analysis (Appendix Fig A1). Baseline characteristics were comparable between groups (Table 1). Most patients were in early stage and received video-assisted thoracic surgery.

TABLE 1.

Baseline Characteristics

Characteristic	ePRO Group (n = 55)	Usual Care Group (n = 57)	P
Age, years, mean (SD)	51.6 (11.6)	51.2 (10.6)	.82
Age, years			.74
≤60	42 (76.4)	45 (78.9)
>60	13 (23.6)	12 (21.1)
Sex			.46
Female	33 (60.0)	38 (67.7)
Male	22 (40.0)	19 (33.3)
Educational level			.84
≤high school	26 (47.3)	28 (49.1)
>high school	29 (52.7)	29 (50.9)
Employment status			.86
Full or part-time	28 (50.9)	30 (52.6)
Not currently working	27 (49.1)	27 (47.4)
Annual household income			.35
<100,000 CNY	24 (43.6)	20 (35.1)
≥100,000 CNY	31 (56.4)	37 (64.9)
Karnofsky performance score			.70
100	45 (81.8)	45 (78.9)
≤90	10 (18.2)	12 (21.1)
Smoking status			.93
Current or former	10 (18.2)	10 (17.5)
Never	45 (81.8)	47 (82.5)
Comorbidity (CCI)			.69
0	23 (41.8)	26 (45.6)
≥1	32 (58.2)	31 (54.4)
Surgical approach			.43
Video-assisted thoracic surgery	51 (92.7)	55 (96.5)
Open surgery	4 (7.3)	2 (3.5)
Extent of surgery			.71
Sublobectomy	14 (25.5)	18 (31.6)
Lobectomy	35 (63.6)	32 (56.1)
Others	6 (10.9)	7 (12.3)
Histology			.36
Adenocarcinoma	52 (94.5)	56 (98.2)
Nonadenocarcinoma	3 (5.5)	1 (1.8)
Stage			>.99
0-I	53 (96.4)	55 (96.5)
II-IIIAa	2 (3.6)	2 (3.5)
Type of participating center			.49
Cancer hospital	50 (90.9)	54 (94.7)
General hospital	5 (9.1)	3 (5.3)
Postoperative length of stay, days, median (IQR)	5 (4-7)	4 (4-6)	.53

NOTE. Data are No. (%) unless otherwise indicated. No significant differences were found between the two groups (all P > .05).

Abbreviations: CCI, Charlson Comorbidity Index; CNY, Chinese yuan; ePRO, electronic patient-reported outcome; SD, standard deviation.

^a Four patients with stage II-IIIA disease did not undergo adjuvant therapy because of physical intolerance or refusal.

Response Rates and Symptom Alerts

The completion rates are listed in the Appendix (Table A1). At 3, 6, 9, and 12 months postdischarge, 18 (32.7%) patients in the ePRO group generated 50 symptom threshold events, leading to 34 alerts. Surgeons responded to 28 of them (82.4%).

Long-Term Patient-Reported Outcomes

At 12 months postdischarge, the ePRO group reported significantly fewer symptom threshold events than the usual care group (median [IQR], 0 [0-0] v 0 [0-1]; P = .040). From 1 to 12 months postdischarge, the ePRO group reported significantly lower composite scores for physical interference (estimate, –0.86 [95% CI, –1.32 to –0.39]) and affective interference (estimate, –0.70 [95% CI, –1.14 to –0.26]) over time than the usual care group (Fig 1 and Appendix Tables A2 and A3). The ePRO group also reported significantly lower scores (lower symptom burden or better functioning) over time for fatigue, distress, and lack of appetite, work, enjoyment of life, walking, general activity, mood, and relations with others (Fig 2 and Appendix Tables A2 and A3). The interaction between intervention and time was not statistically significant for most symptom and functional items, except for fatigue (P = .039), lack of appetite (P = .043), and vomiting (P = .011; Appendix Table A2).

FIG 1.

Composite functional interference score over time. Mean scores of the (A) composite physical interference and (B) composite affective interference from 1 to 12 months postdischarge measured using the MDASI-LC. Low scores indicate lower symptom burden or better functioning. Error bars are the 95% confidence intervals. ePRO, electronic patient-reported outcome; MDASI-LC, MD Anderson Symptom Inventory-Lung Cancer module.

FIG 2.

Single symptom and functional item score over time. Mean scores of (A) fatigue, (B) distress, (C) lack of appetite, (D) work, (E) enjoyment of life, (F) walking, (G) general activity, (H) mood, and (I) relations with others from 1 to 12 months postdischarge measured using the MDASI-LC. Low scores indicate lower symptom burden or better functioning. Error bars are the 95% confidence intervals. ePRO, electronic patient-reported outcome; MDASI-LC, MD Anderson Symptom Inventory-Lung Cancer module.

DISCUSSION

This randomized study demonstrated that ePRO-based symptom management after lung cancer surgery reduced symptom burden at 12 months postdischarge and improved symptom control and daily functioning from 1 to 12 months postdischarge compared with usual care. These long-term findings underscore the sustained effects of early intensive ePRO-based management. Even in the era of minimally invasive surgery, it supports patients' early return to normal life.

The infrequent symptom monitoring (every 3 months) from 1 to 12 months postdischarge suggests that the long-term benefits may stem from the enduring effects of early intensive interventions, supported by the insignificant intervention-time interactions for most symptoms and functions. The infrequent interventions provided later may help alleviate fatigue and lack of appetite, as evidenced by the significant interaction effects for these symptoms. Additionally, although the ePRO group maintained consistently low fatigue and appetite scores, a significant gap between the two groups persisted, supporting the effects of early intensive symptom management for long-term control. Therefore, symptom monitoring may be unnecessary after 1 month postdischarge or as frequent as within 1 month postdischarge. However, drawing firm conclusions is challenging because of the study design that involved symptom alerts at four time points from 1 to 12 months postdischarge.

The use of ePRO-based care in clinical practice faces challenges, primarily due to the increased workload of managing frequent alerts over an extended duration.^9,10 As this study found that the long-term effects were primarily attributed to early intensive ePRO-based symptom management, it may be cost-effective to implement intensive ePRO monitoring only in the early phase (eg, 1 month) after lung cancer surgery. This approach could facilitate the integration of ePRO-based patient care into routine practice.

In conclusion, early intensive ePRO-based symptom management after lung cancer surgery reduced symptom burden and improved functional status for up to 1 year postdischarge, supporting its integration into standard care.

APPENDIX

FIG A1.

CONSORT diagram. ePRO, electronic patient-reported outcome.

TABLE A1.

Completion Rates of the MDASI-LC at Each Time Point Postdischarge

Time Point	ePRO Group (n = 55)		Percentage	Usual Care Group (n = 57)		Percentage
	Available	Scheduled		Available	Scheduled
	No. of Patients			No. of Patients
1 month	55	55	100.0	57	57	100.0
3 months	55	55	100.0	57	57	100.0
6 months	53	55	96.4	55	57	96.5
9 months	53	55	96.4	54	57	94.7
12 months	53	55	96.4	55	57	96.5

Abbreviations: ePRO, electronic patient-reported outcome; MDASI-LC, MD Anderson Symptom Inventory Lung Cancer module.

TABLE A2.

Longitudinal Symptom and Functional Scores From 1 to 12 Months Postdischarge Using MDASI-LC

Item	ePRO Group (n = 55) v Usual Care Group (n = 57)
	Group			Time			Group × Time
	Estimate	95% CI	P	Estimate	95% CI	P	Estimate	95% CI	P
Composite items
Composite five target symptomsa	–0.42	–0.80 to –0.04	.031	–0.12	–0.17 to –0.06	<.001	0.03	–0.05 to 0.10	.482
Composite five most severe symptomsb	–0.42	–0.81 to –0.03	.037	–0.04	–0.10 to 0.01	.135	–0.01	–0.09 to 0.07	.846
Composite physical interferencec	–0.86	–1.32 to –0.39	<.001	–0.27	–0.37 to –0.18	<.001	0.13	–0.01 to 0.26	.064
Composite affective interferenced	–0.70	–1.14 to –0.26	.002	–0.16	–0.26 to –0.07	.001	0.09	–0.04 to 0.23	.178
Symptom itemse
Disturbed sleep	–0.63	–1.21 to –0.06	.031	–0.07	–0.17 to 0.04	.208	0.11	–0.04 to 0.25	.154
Shortness of breath	–0.38	–0.83 to 0.07	.096	–0.09	–0.17 to –0.01	.027	–0.04	–0.16 to 0.07	.473
Coughing	–0.21	–0.71 to 0.28	.398	–0.17	–0.26 to –0.08	<.001	–0.10	–0.22 to 0.03	.122
Problem with remembering things	–0.29	–0.72 to 0.13	.175	0.21	0.12 to 0.31	<.001	–0.10	–0.24 to 0.03	.136
Fatigue	–0.59	–1.05 to –0.14	.010	–0.10	–0.17 to –0.03	.004	0.10	0.01 to 0.20	.039
Dry mouth	–0.41	–0.85 to 0.03	.065	0.07	–0.001 to 0.14	.052	–0.08	–0.18 to 0.02	.106
Distress	–0.58	–1.03 to –0.13	.011	–0.06	–0.13 to 0.02	.126	–0.02	–0.12 to 0.09	.776
Sadness	–0.43	–0.84 to –0.02	.038	–0.02	–0.09 to 0.05	.534	–0.05	–0.15 to 0.05	.357
Drowsiness	–0.23	–0.64 to 0.18	.279	–0.004	–0.07 to 0.07	.902	0.04	–0.06 to 0.14	.405
Constipation	–0.21	–0.60 to 0.18	.292	0.12	0.04 to 0.20	.006	–0.09	–0.21 to 0.03	.143
Pain	–0.26	–0.64 to 0.11	.166	–0.14	–0.21 to –0.07	<.001	0.05	–0.04 to 0.15	.270
Lack of appetite	–0.63	–0.99 to –0.27	<.001	–0.09	–0.17 to –0.02	.015	0.11	0.004 to 0.21	.043
Numbness	0.06	–0.28 to 0.40	.739	–0.03	–0.09 to 0.03	.299	–0.06	–0.14 to 0.03	.175
Sore throat	–0.08	–0.40 to 0.24	.615	0.03	–0.05 to 0.10	.495	–0.02	–0.12 to 0.09	.758
Nausea	–0.10	–0.36 to 0.16	.466	–0.02	–0.07 to 0.03	.419	–0.01	–0.08 to 0.07	.880
Vomiting	0.09	–0.11 to 0.28	.374	0.02	–0.03 to 0.06	.471	–0.08	–0.14 to –0.02	.011
Functional itemse
Work	–1.00	–1.53 to –0.47	<.001	–0.29	–0.40 to –0.19	<.001	0.10	–0.05 to 0.25	.203
Enjoyment of life	–0.74	–1.25 to –0.24	.004	–0.21	–0.33 to –0.09	<.001	0.07	–0.10 to 0.23	.412
Walking	–0.81	–1.31 to –0.30	.002	–0.23	–0.34 to –0.12	<.001	0.13	–0.02 to 0.28	.093
General activity	–0.76	–1.23 to –0.28	.002	–0.29	–0.39 to –0.19	<.001	0.14	–0.01 to 0.28	.063
Mood	–0.72	–1.18 to –0.26	.002	–0.11	–0.21 to –0.005	.041	0.07	–0.07 to 0.22	.311
Relations with others	–0.61	–1.06 to –0.16	.009	–0.17	–0.26 to –0.07	.001	0.12	–0.02 to 0.26	.083

NOTE. P values were adjusted for participating centers. Lower scores represent better symptom control or better functioning. The bold entries indicate that the P value is less than .05.

Abbreviations: ePRO, electronic patient-reported outcome; MDASI-LC, MD Anderson Symptom Inventory-Lung Cancer module.

^a Composite five target symptom score means the average score of the five target symptoms (pain, fatigue, disturbed sleep, shortness of breath, and coughing).

^b Composite five most severe symptom score means the average score of the five most severe symptoms (disturbed sleep, shortness of breath, coughing, problem with remembering things, and fatigue).

^c Composite physical interference score means the average score of general activity, work, and walking in the MDASI-LC.

^d Composite affective interference score means the average score of mood, relations with others, and enjoyment of life in the MDASI-LC.

^e Ranked by the average scores during the 1-12 months after discharge in descending order.

TABLE A3.

Significance of Between-Group Differences in Longitudinal Symptom and Functional Scores From 1 to 12 Months Postdischarge Using MDASI-LC With Multiple Comparison Adjustment

Item	ePRO Group (n = 55) v Usual Care Group (n = 57)
	P	Rank	(i/m)Q
Lack of appetite	<.001	1	0.002
Work	<.001	2	0.004
Composite physical interference	<.001	3	0.006
Walking	.002	4	0.008
General activity	.002	5	0.009
Mood	.002	6	0.011
Composite affective interference	.002	7	0.013
Enjoyment of life	.004	8	0.015
Relations with others	.009	9	0.017
Fatigue	.010	10	0.019
Distress	.011	11	0.021
Disturbed sleep	.031	12	0.023
Composite five target symptoms	.031	13	0.025
Composite five most severe symptoms	.037	14	0.027
Sadness	.038	15	0.029
Dry mouth	.065	16	0.031
Shortness of breath	.096	17	0.033
Pain	.166	18	0.035
Problem with remembering things	.175	19	0.037
Drowsiness	.279	20	0.038
Constipation	.292	21	0.040
Vomiting	.374	22	0.042
Coughing	.398	23	0.044
Nausea	.466	24	0.046
Sore throat	.615	25	0.048
Numbness	.739	26	0.050

NOTE. Two-sided P values were adjusted for multiple comparisons using the Benjamini-Hochberg procedure. The individual P values were arranged in ascending order, from smallest to largest. The smallest P value has a rank of i = 1, then the next smallest has i = 2, and so on. Each individual P value was compared with its Benjamini-Hochberg critical value, (i/m)Q, where i represents the rank, m represents the total number of tests set to 26, and Q represents the false discovery rate set at 0.05. The largest P value that was less than its (i/m)Q is distress, which has a P value of .011 and a (i/m)Q of 0.021. Thus, this test and all tests (the first 11 items) with smaller P values were considered significant. Lower scores represent better symptom control or better functioning.

Abbreviations: ePRO, electronic patient-reported outcome; MDASI-LC, MD Anderson Symptom Inventory-Lung Cancer module.

DATA SHARING STATEMENT

Deidentified data reported in this article can be made available by the corresponding author upon reasonable request. Access to the data sets will be granted for nonprofit research purposes or regulatory decision making, subject to a signed agreement.

AUTHOR CONTRIBUTIONS

Conception and design: Wei Dai, Xin Shelley Wang, Cecilia Pompili, Yuanqiang Zhang, Guibin Qiao, Qiang Li, Qiuling Shi

Financial support: Wei Dai, Guibin Qiao, Qiang Li, Qiuling Shi

Administrative support: Xiaoqin Liu, Guibin Qiao, Qiang Li

Provision of study materials or patients: Xing Wei, Yuqian Zhao, Bangrong Cao, Yuanqiang Zhang, Fang Liu, Jin Zhou, Juan Li, Shaohua Xie, Run Xiang, Xiang Wang, Bo Tian, Xiaozun Yang, Bin Hu, Tianpeng Xie, Xiaojun Yang, Xiang Zhuang, Guibin Qiao

Collection and assembly of data: Wei Dai, Yaqin Wang, Jia Liao, Xing Wei, Wei Xu, Bangrong Cao, Qifeng Wang, Wenhong Feng, Yuanqiang Zhang, Shaohua Xie, Xiang Wang, Xiaozun Yang, Bin Hu, Xiaoqin Liu, Xiaojun Yang, Xiang Zhuang, Guibin Qiao

Data analysis and interpretation: Wei Dai, Yaqin Wang, Jia Liao, Xing Wei, Zhen Dai, Wei Xu, Yangjun Liu, Xin Shelley Wang, Cecilia Pompili, Hongfan Yu, Yang Pu, Yuqian Zhao, Yuanqiang Zhang, Fang Liu, Yuanle Deng, Jin Zhou, Juan Li, Run Xiang, Xiang Wang, Bo Tian, Tianpeng Xie, Xiang Zhuang, Guibin Qiao, Qiuling Shi

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Electronic Patient-Reported Outcome–Based Symptom Management Versus Usual Care After Lung Cancer Surgery: Long-Term Results of a Multicenter, Randomized, Controlled Trial

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).