Correlation of self-reported pain severity and healthcare expenditures in older United States adults

Introduction

There are many definitions of pain, including the recent update from the International Association for the Study of Pain who define pain as “an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage” [1]. Determining a reliable estimate of pain prevalence in the United States (US) is challenging due to the variability of relevant data and definitions of pain [2]. For example, an Institute of Medicine report estimated that approximately 116 million US adults had chronic pain in 2011[2]. However, another study using the 2011 National Health Interview Survey (NHIS) estimated approximately 41 million (18.4%) of the adult population had chronic pain [3]. Data from the health and retirement study have indicated that the prevalence of pain in the US population has been increasing by 2–3% per year since 1992 and is likely to continue [4]. These increases in pain prevalence are even higher among older adults as increased age is a major risk factor for pain [5]. There are several types of pain, as recently indicated by International Classification of Disease (ICD)-11 codes [6], thus the nature of pain is important to consider when investigating pain. Studies have also shown that pain interferes with various activities, such as activities of daily living, ability to work, ability to concentrate, participate in leisure activities or social activities, and maintain relationships with others [7], [8], [9].

Pain severity is another factor that must be considered when investigating pain. A recent systematic review found that as a patient’s intensity of pain increased their Health-Related Quality of Life decreased [10]. Another study that used 2012 NHIS data concluded that individuals in the more severe pain categories had a higher probability of being in a poorer health state, to report being exhausted every day, taking medications to treat depression, and to be consistently worried, nervous, or anxious every day [11]. Meanwhile, a further study used Medical Expenditure Panel Survey (MEPS) data to identify factors that were associated with pain severity among older adults [12].

Estimates of the cost of pain also vary in the literature. For example, lost productivity from missed work due to pain amounted to $11.6 to $12.7 billion per year, and pain as a whole cost $560 to $635 billion per year in 2010 US dollars [2]. Another study using 2008 MEPS data found that individuals with severe pain had higher total and certain types of other healthcare service expenditures compared to those with low and moderate pain [13].

However, this study included adults aged 18 years and older [13]. There is a lack of data on how healthcare expenditures vary by pain severity among older adults with pain, which this current study sought to address. The objective of this study was to assess the correlation of self-reported pain severity and healthcare expenditures among older US adults with pain, using adjusted models to account for additional factors that may influence pain severity in this population.

Methods

Results

Figure 1 shows the study subject eligibility flowchart. There were 30,461 individuals included in the MEPS 2018 full year consolidated data file. Of these individuals, 5,123 met the inclusion criteria of the study (civilian, noninstitutionalized, US adults aged 50 years and older, alive for the full 2018 calendar year, with self-reported pain in the past four weeks that interfered with normal work inside and outside the home). Of the subjects included, 382 (6.8%) reported extreme pain, 1,035 (18.5%) reported quite a bit of pain, 1,121 (21.6%) reported moderate pain, and 2,585 (53.1%) reported a little bit of pain.

Figure 1:

Study subject eligibility flowchart.

Table 1 reports the characteristics of the study population stratified by the level of self-reported pain (little, moderate, quite a bit, extreme). The majority of participants were over the age of 65, female, white, non-Hispanic, were married, completed more than high school education, reported excellent/very good/good perceived health and mental health status, indicated the presence of limitation, reported not doing regular moderate/vigorous exercise, were non-smokers, unemployed, middle/high income, private health insurance coverage, and had less than five chronic conditions. Subjects most commonly resided in the Southern census region. There were significant differences (p<0.05) between all variables except census region, gender, and ethnicity.

Table 2 presents unadjusted and adjusted differences in healthcare expenditures between each level of self-reported pain for each category of healthcare expenditure. In the adjusted analyses, extreme pain had the greatest difference in costs compared to little pain, followed by quite a bit of pain and then moderate pain for total, office-based, and prescription medication expenditures. For outpatient expenditures, quite a bit of pain was greater than little pain, yet extreme pain or moderate pain was not significantly greater than little pain. For inpatient expenditures, extreme pain was greater than little pain and moderate pain was greater than little pain, yet quite a bit of pain was not significantly greater than little pain. For other expenditures, extreme pain and quite a bit of pain were greater than little pain, yet moderate pain was not significantly greater than little pain.

Discussion

This study aimed to assess the correlation of self-reported pain severity and adjusted healthcare expenditures among older US adults with pain. The key finding from this study was that a pattern emerged for the correlation of pain severity and adjusted healthcare expenditures for total, office-based, and prescription medication expenditures. For these expenditures, relative to little pain, extreme pain had the greatest difference in expenditures, followed by quite a bit of pain, followed by moderate pain. However, this pattern was not consistent for outpatient, inpatient, and other healthcare expenditures (where there were instances of no statistical correlation between pain severity and expenditures for some pain severity levels), and emergency room expenditures (where there was no statistical correlation between pain severity and expenditures). These findings provide new information to the literature, which is useful given the lack of contemporary existing studies that have investigated pain severity and healthcare expenditures. These findings will be further discussed below.

The finding that adjusted total, office-based, and prescription medication expenditures were greater with greater levels of pain severity is an expected finding that is unsurprising from a clinical perspective (i.e., those with more severe pain are likely to have overall greater use of the healthcare system, including more office-based visits and greater use of prescription medications, as they seek to manage their pain). A previous study using data from the Mint (minimal invasive treatment) study [17] found that pain severity was significantly and positively correlated with total healthcare expenditures in both unadjusted and adjusted analyses [18], which parallel the findings from the current study.

The finding that adjusted outpatient, inpatient, and other healthcare expenditures were not always associated with greater levels of pain severity is perhaps more surprising. In the case of adjusted outpatient expenditures, quite a bit of pain was correlated with greater expenditures than little pain, yet extreme pain and moderate pain were not. Whereas, in the case of adjusted inpatient expenditures, extreme pain and moderate pain were correlated with greater expenditures than little pain, yet quite a bit of pain was not. For adjusted other healthcare expenditures, extreme pain and quite a bit of pain were correlated with greater expenditures than little pain, yet moderate pain was not. In addition, there was no correlation between pain severity and emergency room expenditures. A previous study using 2008–2011 MEPS data found a similar pattern to the findings in the current study for outpatient expenditures and other healthcare expenditures [13]. Interestingly, the findings for inpatient expenditure in the 2008–2011 MEPS increased with increasing pain severity [13], more akin to the findings for the total, office-based, and prescription medications in the current study. However, there were differences in the definition of pain severity and differences in the variables adjusted for in analyses between the two studies, which likely contributed to the different results observed. With regards to the emergency room expenditures, emergency room utilization is typically low (a previous study found 18% of US adults aged 18–64 used the emergency room at least one or more times in 2014) [19], which may not be influenced by pain severity. Future research is needed to understand why some types of healthcare expenditures are positively correlated with greater pain severity levels, yet others are not.

The current study found most of the personal characteristics (outlined in Table 1) were associated with pain severity. This is perhaps an expected finding, given that many of these characteristics have been shown to be associated with pain in other studies. For instance, in keeping with the findings of this study, previous research has demonstrated that individuals had greater odds of reporting quite a bit or extreme pain if they were aged 50–64 vs. ≥65 years, other vs. white race, married vs. other marital status, had fair/poor vs. excellent/very good/good perceived health, had a limitation vs. no limitation, did not do regular exercise vs. did regular exercise, were smokers vs. non-smokers, unemployed vs. employed, and had a lower vs. higher income [12]. These characteristics should therefore be considered by researchers investigating pain and clinicians helping individuals manage their pain.

The findings of the current study suggest the need for interventions to help reduce pain severity, or prevent pain from becoming severe in the first place, which in turn may help to reduce healthcare expenditures (although no causal inference can be assumed based on the study design). This may be done through improving or optimizing pain management strategies. For instance, studies have identified that individuals with pain often use a variety of pain management strategies beyond prescription medications [8, 20]. Some organizations use enhanced recovery after surgery (ERAS) techniques to help reduce the time to discharge. ERAS uses multimodal techniques to help enhance pain control, including adjunctive non-pharmacological methods such as acupuncture, music therapy, transcutaneous electrical nerve stimulation, and hypnosis [21]. Improving pain management and reducing pain severity may have implications beyond healthcare expenditures, for instance, one study found that those with lower pain severity reported improved quality of life [22].

The findings of this study represent civilian, noninstitutionalized, US adults aged 50 years and older with self-reported pain in the past four weeks that interfered with normal work inside and outside the home. However, limitations of this study included the retrospective nature of the study design, which prohibited the determination of a cause-and-effect relationship between pain severity and healthcare expenditures. Another limitation of this study is the use of self-reported pain in the past four weeks, which varies from other definitions of pain in the literature and may be subject to recall bias. Pain is also a subjective condition and therefore one of the more difficult healthcare conditions to assess objectively.

Conclusions

This study found that total, office-based, and prescription medication healthcare expenditures increased as self-reported pain severity increased, yet this pattern was not consistent when analyzing other subcategories of healthcare expenditures (i.e., outpatient, inpatient, emergency room, and other healthcare expenditures). These findings suggest that reducing pain severity, or preventing pain from becoming severe, could improve healthcare expenditures. Future research is needed to understand why some types of healthcare expenditures are positively correlated with greater pain severity levels, yet others are not, and should evaluate clinical or public health interventions to reduce and prevent pain.