Buy this Article for $10.95

Have a coupon or promotional code? Enter it here:

When you buy this you'll get access to the ePub version, a downloadable PDF, and the ability to print the full article.


end of life, lung cancer, symptoms, symptom cluster



  1. Gift, Audrey G.
  2. Stommel, Manfred
  3. Jablonski, Anita
  4. Given, William


Background: Patients with lung cancer present late in the disease and have multiple symptoms. Previous research has shown the symptom cluster of fatigue, weakness, weight loss, appetite loss, nausea, vomiting, and altered taste to be present at time of lung cancer diagnosis.


Objectives: The study determined whether the symptom cluster identified at the time of diagnosis remained 3 and 6 months later, and whether there was a difference in the mean number of symptoms and the mean level of symptom severity over time. The relation of the severity rating for individual symptoms at the time of diagnosis and at 3 and 6 months after diagnosis was examined. Predictors for the number of symptoms and whether the symptom cluster was predictive of death were determined.


Methods: Secondary analysis of an existing data set for 112 patients with newly diagnosed lung cancer assessed at diagnosis and at 3 and 6 months was performed and determined whether they were alive or dead 19 months after diagnosis.


Results: The cluster of seven symptoms had internal consistency that remained at 3 and 6 months. The mean symptom severity and the number of symptoms at diagnosis were correlated with later ratings, but decreased in severity over time. A similar decrease in severity rating was seen for the individual symptoms in the cluster. The stage of cancer at diagnosis was the most predictive of the number of cluster symptoms reported. Death 6 to 19 months after diagnosis was predicted by age, stage of cancer at diagnosis, and symptom severity at 6 months.


Conclusions: The symptom cluster remains over the course of lung cancer and is an independent predictor of the patient's death. Symptom severity, the number of symptoms reported, and the severity of the individual symptoms decreased over time. The stage of cancer at diagnosis is the best predictor of symptoms later in the disease.


Patients with lung cancer, whether newly diagnosed or more advanced, experience more severe symptoms than patients with other cancers at a similar stage in their disease (Cooley, 2000;Degner & Sloan, 1995;Sarna & Brecht, 1997;Weisman & Worden, 1976). Few studies of patients with lung cancer have examined symptoms over time to determine whether and how they change during the course of the disease. The purpose of this secondary analysis was to study symptoms over time in patients with lung cancer. The evaluation analyzed symptoms in a cluster, individual symptoms, mean symptom severity, and the number of symptoms reported over time. Factors present at the time of diagnosis were explored as predictors of symptom severity later in the disease course and as predictors of death. Such a prediction would allow healthcare providers to identify those most at risk for poor outcomes, and to design interventions targeted at alleviating suffering in those high-risk patients.


The current analysis builds on previous work in which symptoms reported by patients with lung cancer were examined. A factor analysis was performed at the time of diagnosis on the symptoms reported by 220 patients with lung cancer using the Physical Symptom Experience tool (Given et al., 1993). Seven symptoms, namely, fatigue, weakness, nausea, vomiting, loss of appetite, weight loss, and altered taste symptoms, formed a unique factor (Gift, Jablonski, Stommel, & Given, in press). The current study examined that cluster of symptoms in the same patients over the course of their disease.


The study of symptoms as clusters reported by the individual patient is a recent focus. One of the first research teams to take this approach was Dodd, Miaskowski, and Paul (2001). These researchers did not, however, study patients with lung cancer, but rather included 93 patients with breast or colorectal cancer who were undergoing chemotherapy. They studied fatigue, pain, and sleep insufficiency, stating that they were the most distressing symptoms for patients. The intercorrelations among the three symptoms were low (r = .06 to .22). These researchers did not study the symptoms over time.


Other researchers have examined symptoms at different stages of the lung cancer disease process, finding them to be greater in those with more advanced disease (Degner & Sloan, 1995;Hopwood & Stephens, 1995;Krech, Davis, Walsh, & Curtis, 1992;Vainio, Aurinen, & Members of Symptom Prevalence Group, 1996). However, the use of a cross-sectional design does not allow the study of symptoms in the same patient over time.


Cooley, Short, and Moriarty (2002) are the only known researchers to focus on symptom clusters in patients with lung cancer over the course of the disease. In their study of 117 patients, symptom distress was moderately high at the time of diagnosis, decreased at 3 months after diagnosis, and then increased at 6 months. Women had higher symptom distress than men. Those with more advanced disease had more symptoms. Kurtz, Kurtz, Stommel, Given, and Given (2000), however, did not find greater symptom distress in patients with more advanced disease.


Exploration of other factors contributing to symptom severity has resulted in similar contradictory findings. Age has been found by some to influence symptom reporting (Degner & Sloan, 1995;McMillan, 1989;Walsh, Donnelly, & Rybicki, 2000), but not by others (Kurtz, Given, Kurtz, & Given, 1994;Kurtz et al., 2000). Some researchers have found that women report more symptoms than men (Degner & Sloan, 1995;Hopwood & Stephens, 1995;Krech et al., 1992), whereas others have not found such a gender difference (Kurtz et al., 2000;Ouellette & Kobasa, 1998). Factors influencing the symptom experience warrant further study.


Outcomes of symptom distress also are important. Degner and Sloan (1995) found that symptom distress is the best predictor of survival time, and that it is a stronger predictor than the stage of illness. This finding should be replicated in a sample that includes more patients with advanced disease.