Enhancing Risk Assessment in Patients Receiving Chronic Opioid Analgesic Therapy Using Natural Language Processing.

OBJECTIVES: Clinical guidelines for the use of opioids in chronic noncancer pain recommend assessing risk for aberrant drug-related behaviors prior to initiating opioid therapy. Despite recent dramatic increases in prescription opioid misuse and abuse, use of screening tools by clinicians continues to be underutilized. This research evaluated natural language processing (NLP) together with other data extraction techniques for risk assessment of patients considered for opioid therapy as a means of predicting opioid abuse. DESIGN: Using a retrospective cohort of 3,668 chronic noncancer pain patients with at least one opioid agreement between January 1, 2007, and December 31, 2012, we examined the availability of electronic health record structured and unstructured data to populate the Opioid Risk Tool (ORT) and other selected outcomes. Clinician-documented opioid agreement violations in the clinical notes were determined using NLP techniques followed by manual review of the notes. RESULTS: Confirmed through manual review, the NLP algorithm had 96.1% sensitivity, 92.8% specificity, and 92.6% positive predictive value in identifying opioid agreement violation. At the time of most recent opioid agreement, automated ORT identified 42.8% of patients as at low risk, 28.2% as at moderate risk, and 29.0% as at high risk for opioid abuse. During a year following the agreement, 22.5% of patients had opioid agreement violations. Patients classified as high risk were three times more likely to violate opioid agreements compared with those with low/moderate risk. CONCLUSION: Our findings suggest that NLP techniques have potential utility to support clinicians in screening chronic noncancer pain patients considered for long-term opioid therapy.

Implementing patient-reported outcomes assessment in clinical practice: a review of the options and considerations.

PURPOSE: While clinical care is frequently directed at making patients "feel better," patients' reports on their functioning and well-being (patient-reported outcomes [PROs]) are rarely collected in routine clinical practice. The International Society for Quality of Life Research (ISOQOL) has developed a User's Guide for Implementing Patient-Reported Outcomes Assessment in Clinical Practice. This paper summarizes the key issues from the User's Guide. METHODS: Using the literature, an ISOQOL team outlined considerations for using PROs in clinical practice; options for designing the intervention; and strengths, weaknesses, and resource requirements associated with each option. RESULTS: Implementing routine PRO assessment involves a number of methodological and practical decisions, including (1) identifying the goals for collecting PROs in clinical practice, (2) selecting the patients, setting, and timing of assessments, (3) determining which questionnaire(s) to use, (4) choosing a mode for administering and scoring the questionnaire, (5) designing processes for reporting results, (6) identifying aids to facilitate score interpretation, (7) developing strategies for responding to issues identified by the questionnaires, and (8) evaluating the impact of the PRO intervention on the practice. CONCLUSIONS: Integrating PROs in clinical practice has the potential to enhance patient-centered care. The online version of the User's Guide will be updated periodically.

The prognostic value of measuring health-related quality of life in hospice patients.

PURPOSE: Better prognostic instruments are sorely needed for patients near the end of life. Health-related quality of life instruments designed for hospice patients have not been previously studied for their prognostic properties. This study evaluated the prognostic property of the Missoula-Vitas Quality of Life Index (MVQOLI) with hospice patients. METHODS: A prospective, cross-sectional cohort design included all consecutive patients admitted to a hospice over a 19-month period. At admission to hospice, patients were asked to complete an MVQOLI. In addition, hospice nurses completed three functional status instruments. All patients were followed until death, study closure, or loss to follow-up. RESULTS: The sample included 1047 patients, but only 231 (22%) were able to complete an MVQOLI at admission. Functional status data were collected on nearly all of the patients. The Karnofsky performance score, modified activity of daily living score, and descriptive symptom score were significantly associated with survival time. Using Cox regression models these functional status assessments were strongly associated with survival time (p<0.001). However, the MVQOLI scores were not significantly associated with survival time, except for the function subscale (p=0.045). CONCLUSION: The MVQOLI global, total, and four of the five weighted-domain baseline scores were not associated with survival time in hospice patients. Other methods for prognostication at the end of life are needed.

Rural-urban differences in cancer care: results from the Lake Superior Rural Cancer Care Project.

Past studies have shown significant differences between rural and urban cancer patients in many measures of cancer care. There is little recent information about this disparity, which generally has shown disadvantages in rural populations. This study reports the rural and urban differences in cancer care using data from the Lake Superior Rural Cancer Care Project. The study used a prospective, population-based design that included all incident cases of breast, colorectal, lung, and prostate cancers diagnosed in northeastern Minnesota, northwestern Wisconsin, and the western portion of Michigan's Upper Peninsula from 1992 to 1997. The outcome measures were 9 endpoints that represented state-of-the-art cancer care during the study. Rural cancer patients as compared with their urban counterparts were disadvantaged in proportion staged, stage at diagnosis, initial management procedures, post-treatment surveillance testing, and participation in cancer clinical trials. These findings are similar to previously published studies. Further research is needed to determine more clearly the barriers in rural cancer care and to find more effective strategies.

Improving rural cancer patients' outcomes: a group-randomized trial.

CONTEXT: Significant barriers exist in the delivery of state-of-the-art cancer care to rural populations. Rural providers' knowledge and practices, their rural health care delivery systems, and linkages to cancer specialists are not optimal; therefore, rural cancer patient outcomes are less than achievable. PURPOSE: To test the effects of a strategy targeting rural providers and their practice environment on patient travel for care, satisfaction, economic barriers, and health-related quality of life. METHODS: A group-randomized trial was conducted with 18 rural communities in the north-central United States. Twelve of these communities were included and defined as the unit of analysis for the patient outcomes portion of the study. The intervention targeted rural providers and their practice environment. The subjects were patients with breast, colorectal, lung, and prostate cancers from the rural communities. The main outcomes were patients' travel to obtain health care, satisfaction with care, perceptions of economic barriers to care, and health-related quality of life. In total, 881 patients were included. RESULTS: Group randomization was balanced. Travel for health care was significantly reduced in the community group exposed to the intervention during months 13 to 24 following cancer diagnosis. The mean miles traveled per patient were 1,326 (SE = 306) for the experimental group and 2,186 (SE = 347) for the control group (P = 0.03). No significant differences in satisfaction with care, economic barriers to care, or health-related quality of life were found. CONCLUSIONS: The intervention significantly reduced cancer patient travel for health care, which suggests that access to care improved in the experimental group. The results of this study do not allow conclusion that there was no effect on other patient outcomes. The results supported the study's conceptual framework and many of its hypotheses.

Chronic pain, depression, and quality of life: correlations and predictive value of the SF-36.

OBJECTIVE: Depression is a major barrier to effective pain relief. The SF-36 Health Survey may be useful as an outcome measure for chronic pain patients with and without depression. The study purpose was to determine the correlation between the SF-36 Mental Composite Scale t-score and depression type in chronic pain patients and the positive predictive value of the SF-36 in classifying depression type in chronic pain patients. DESIGN: Cross-sectional survey of chronic pain patients at baseline assessment. SETTING: Interdisciplinary pain management center in the North Central United States. PATIENTS: Two hundred forty-two consecutive, chronic noncancer pain patients. INTERVENTIONS: Assessments of patient-reported health-related quality of life (SF-36), pain, pain type and diagnoses, mental health diagnoses, and patient demographics. OUTCOME MEASURES: SF-36 Health Survey, pain diagnoses, and depression diagnoses. RESULTS: Study participants were 160 women (66%) and 82 men (34%), 95% Caucasian, whose mean age was 46 (+/-0.8 SD) years (range: 19-83). All types and diagnoses of chronic pain were represented. The prevalence of major depressive disorder was 52%. The type of depression was highly correlated with SF-36 score (r=-0.567; P < 0.001). All chronic pain patients had very low SF-36 scores. Compared with U.S. population norms, chronic pain patients with and without depression had significantly lower SF-36 scores as measured by z-scores. Chronic pain patients with major depressive disorder had a significantly lower Mental Composite Score t-score than those with minor or no depression-34.1 and 47.6, respectively (P < 0.001). The positive predictive value of the SF-36 for differentiating major depression from minor or no depression was 98% (sensitivity=84.4%, specificity=93.9%). CONCLUSIONS: The SF-36 Mental Composite Score and all subscales were highly correlated with depression type in chronic pain patients. The positive predictive value of the SF-36 in classifying depression type was high. The SF-36 may be a useful clinical tool to measure health-related quality of life in chronic pain patients. In addition, the SF-36 was able to detect major depression and demonstrate a dose-effect relationship between depression type (severity) and health-related quality of life in chronic pain patients.