Cancer Pain: How Physiological Signals and Technology May Improve Assessment

Previous studies have shown a positive correlation between pain experienced in cancer and physiological markers such as heart rate, systolic blood pressure, and low-frequency to high-frequency component ratio (LF/HF).¹ With more research needed to determine the best ways to collect and interpret physiological signals to improve patient care, Serena Moscato, PhD, of University of Bologna, Italy, and colleagues conducted a systematic literature review to highlight future study considerations.¹

“Pain monitoring and assessment through physiological signals are still in an exploratory phase,” Looking ahead, some of these limitations may be overcome by the next generation of wearable devices,” wrote Moscato et al.¹

The team analyzed data from 14 studies published up to October 2020 encompassing a total of 528 individuals. Selected studies included physiological signals associated with pain, of which blood pressure and ECG were the most used. Nine of the 14 studies analyzed used scales or questionnaires to assess the validity of the physiological parameters, and the other 5 compared sensitivity to change of the physiological parameters after an intervention.

The researchers assessed 19 distinct physiological parameters in association with pain; 14 of these significantly related to pain in at least one study. Positive correlations between pain and heart rate, LF/HF, and systolic blood pressure were consistent across the studies.

Selected papers were limited by several factors including the heterogeneity of the type of cancer and source of pain being evaluated, and the range of assessment tools, the researchers noted. Other limitations included the lack of accepted guidelines for cancer pain management and the lack of methodological information (eg, sampling frequency) on the included studies. In addition, personal factors such as age, gender, and health status can impact physiological mechanisms, especially in people with cancer. “In some cases,” noted Dr. Moscato’s team, “cancer pathology itself can lead to a change in physiological mechanisms, which can be misinterpreted and related to the pain experience.”¹

Nonetheless, the review findings hold promise for the development of new pain assessment tools for people with cancer based on physiological signals, including the possibility of wearable sensors, concluded Moscato et al. The team is already working to design clinical trials that can be conducted among patients in residential facilities, “thus collecting the pain response when and where it is experienced.”¹

With regard to the Moscato review, Marcin Chwistek, MD, FAAHPM, director of the supportive oncology and palliative care program at Fox Chase Cancer Center, Philadelphia, said he was not surprised by the findings. “Clinicians know that some physiological parameters can potentially be helpful in assessing pain. Especially, if one has access to continuous monitoring over periods of time that can be then analyzed and interpreted,” he said. “I am excited by the possibility of using such data in the future, but we are not there yet.”

Looking ahead, “I think we need to look at a more comprehensive way to assess pain remotely, beyond collecting physiologic parameters,” said Dr. Chwistek. “Physiological parameters will become, in my view, a part of a more comprehensive assessment that also includes psycho-social-spiritual parameters as well as input from the patient’s environment or from their caregivers.”

However, “There is a danger in looking at pain only through the prism of physiology and therefore reducing it to nociception,” Dr. Chwistek said. “We know the pain experience is a complex phenomenon. Also, contrary to what the authors say in this study, collecting physiological data alone will not eliminate or simplify clinicians’ assessments. I am hoping that such data will assist us in formulating our assessments, not replacing them,” he explained.

Potential concerns related to the use of physiological data for chronic pain management include the complex nature of chronic pain, Dr. Chwistek said. “Physiologic monitoring, I believe, might be more practical and easier to interpret in the setting of acute or subacute pain, such as post-operative pain management.” Chronic cancer pain is especially complex, as physiologic data only tell a part of the story. “Effective cancer pain management takes into account the entire person,” he emphasized.

Other potential barriers to the use of physiologic data for cancer pain management include the need for artificial intelligence (AI) algorithms to process the massive amount of data that would be generated, and such algorithms have yet to be developed, explained Dr. Chwistek. Well-trained staff will be needed to receive and make sense of all the incoming information from continuous monitoring, he added, and privacy concerns that accompany continuous monitoring will remain challenges as well.

Remote Monitoring

One growing area of focus for assessing cancer pain is remote monitoring, noted Dr. Chwistek. “Despite significant progress in developing anti-cancer therapies, cancer pain remains a very significant clinical issue, but with relatively little progress in how we treat or assess patients who suffer from it,” he said. Clinicians have a strong desire to improve pain management in these individuals, he said, and “technological advances now make it possible to monitor physiological parameters remotely and relatively unobtrusively.”

The growth in AI and machine learning allows for analysis of large datasets of information, such as the constant flow of physiologic data, “that can be processed efficiently and may assist clinicians in the future,” Dr. Chwistek explained.

“Additionally, there is a growing interest in managing patients’ symptoms effectively at home to provide better care and minimize healthcare-related costs associated with visits to ER or unplanned hospitalizations due to poorly controlled pain,” he added.

Wearables

“The current model of cancer pain care relies on intermittent, often irregular visits to a clinician in the clinic (recently also through telehealth) during which decisions are made using an imperfect data set,” Dr. Chwistek explained. “We still rely on the patient’s ability to recall events, including their medication use, or on a caregiver’s report, if available.” Some patients are better at, or better able to, collect data; some do not collect it at all; “therefore, one could see that wearables could become a source of new valuable data that will supplement clinical assessments,” he said.

Looking Ahead

Areas for future research include identifying what physiologic parameters are most useful in different types of cancer-related pain, and how to interpret them; how to best analyze the data, so it produces meaningful results; and how to involve other aspects of the patients’ and their caregivers’ experience related to pain, Dr. Chwistek noted.