Talking, not training, increased the accuracy of physicians' diagnosis of their patients' preferences for colon cancer screening

OBJECTIVE

Identify if primary care physicians (PCPs) accurately understand patient preferences for colorectal cancer (CRC) testing, whether shared decision making (SDM) training improves understanding of patient preferences, and whether time spent discussing CRC testing improves understanding of patient preferences.

METHODS

Secondary analysis of a trial comparing SDM training plus a reminder arm to a reminder alone arm. PCPs and their patients completed surveys after visits assessing whether they discussed CRC testing, patient testing preference, and time spent discussing CRC testing. We compared patient and PCP responses, calculating concordance between patient-physician dyads. Multilevel models tested for differences in preference concordance by arm or time discussing CRC.

RESULTS

382 PCP and patient survey dyads were identified. Most dyads agreed on whether CRC testing was discussed (82%). Only 52% of dyads agreed on the patient's preference. SDM training did not impact accuracy of PCPs preference diagnoses (55%v.48%,p = 0.22). PCPs were more likely to accurately diagnose patient's preferences when discussions occurred, regardless of length.

CONCLUSION

Only half of PCPs accurately identified patient testing preferences. Training did not impact accuracy. Visits where CRC testing was discussed resulted in PCPs better understanding patient preferences.

PRACTICE IMPLICATIONS

PCPs should take time to discuss testing and elicit patient preferences.

Decisions about adopting novel COVID-19 vaccines among White adults in a rural state, USA: A qualitative study

PURPOSE

Many people, especially in rural areas of the United States, choose not to receive novel COVID-19 vaccinations despite public health recommendations. Understanding how people describe decisions to get vaccinated or not may help to address hesitancy.

METHODS

We conducted semistructured interviews with 17 rural inhabitants of Maine, a sparsely populated state in the northeastern US, about COVID-19 vaccine decisions during the early rollout (March-May 2021). We used the framework method to compare responses, including between vaccine Adopters and Non-adopters.

FINDINGS

Adopters framed COVID-19 as unequivocally dangerous, if not personally, then to other people. Describing their COVID concerns, Adopters emphasized disease morbidities. By contrast, Non-adopters never mentioned morbidities, referencing instead mortality risk, which they perceived as minimal. Instead of risks associated with the disease, Non-adopters emphasized risks associated with vaccination. Uncertainty about the vaccine development process, augmented by social media, bolstered concerns about the long-term unknown risks of vaccines. Vaccine Adopters ultimately described trusting the process, while Non-adopters expressed distrust.

CONCLUSION

Many respondents framed their COVID vaccination decision by comparing the risks between the disease and the vaccine. Associating morbidity risks with COVID-19 diminishes the relevance of vaccine risks, whereas focusing on low perceived mortality risks heightens their relevance. Results could inform efforts to address COVID-19 vaccine hesitancy in the rural US and elsewhere.

PATIENT OR PUBLIC CONTRIBUTION

Members of Maine rural communities were involved throughout the study. Leaders of community health groups provided feedback on the study design, were actively involved in recruitment, and reviewed findings after analysis. All data produced and used in this study were co-constructed through the participation of community members with lived experience.

Promoting Informed Decisions About Colorectal Cancer Screening in Older Adults (PRIMED Study): a Physician Cluster Randomized Trial

BACKGROUND

For adults aged 76-85, guidelines recommend individualizing decision-making about whether to continue colorectal cancer (CRC) testing. These conversations can be challenging as they need to consider a patient's CRC risk, life expectancy, and preferences.

OBJECTIVE

To promote shared decision-making (SDM) for CRC testing decisions for older adults.

DESIGN

Two-arm, multi-site cluster randomized trial, assigning physicians to Intervention and Comparator arms. Patients were surveyed shortly after the visit to assess outcomes. Analyses were intention-to-treat.

PARTICIPANTS AND SETTING

Primary care physicians affiliated with 5 academic and community hospital networks and their patients aged 76-85 who were due for CRC testing and had a visit during the study period.

INTERVENTIONS

Intervention arm physicians completed a 2-h online course in SDM communication skills and received an electronic reminder of patients eligible for CRC testing shortly before the visit. Comparator arm received reminders only.

MAIN MEASURES

The primary outcome was patient-reported SDM Process score (range 0-4 with higher scores indicating more SDM); secondary outcomes included patient-reported discussion of CRC screening, knowledge, intention, and satisfaction with the visit.

KEY RESULTS

Sixty-seven physicians (Intervention n=34 and Comparator n=33) enrolled. Patient participants (n=466) were on average 79 years old, 50% with excellent or very good self-rated overall health, and 66% had one or more prior colonoscopies. Patients in the Intervention arm had higher SDM Process scores (adjusted mean difference 0.36 (95%CI (0.08, 0.64), p=0.01) than in the Comparator arm. More patients in the Intervention arm reported discussing CRC screening during the visit (72% vs. 60%, p=0.03) and had higher intention to follow through with their preferred approach (58.0% vs. 47.1, p=0.03). Knowledge scores and visit satisfaction did not differ significantly between arms.

CONCLUSION

Physician training plus reminders were effective in increasing SDM and frequency of CRC testing discussions in an age group where SDM is essential.

TRIAL REGISTRATION

The trial is registered on clinicaltrials.gov (NCT03959696).

"I know my body better than anyone else": a qualitative study of perspectives of people with lived experience on antimicrobial treatment decisions for injection drug use-associated infections

Background

People who inject drugs (PWID) are at risk for severe bacterial and fungal infections including skin and soft tissue infections, endocarditis, and osteomyelitis. PWID have high rates of self-directed discharge and are often not offered outpatient antimicrobial therapies, despite studies showing their efficacy and safety in PWID. This study fills a gap in knowledge of patient and community partner perspectives on treatment and discharge decision making for injection drug use (IDU)-associated infections.

Methods

We conducted semi-structured interviews with patients (n = 10) hospitalized with IDU-associated infections and community partners (n = 6) in the Portland, Maine region. Community partners include peer support workers at syringe services programs (SSPs) and outreach specialists working with PWID. We transcribed and thematically analyzed interviews to explore perspectives on three domains: perspectives on long-term hospitalization, outpatient treatment options, and patient involvement in decision making.

Results

Participants noted that stigma and inadequate pain management created poor hospitalization experiences that contributed to self-directed discharge. On the other hand, patients reported hospitalization provided opportunities to connect to substance use disorder (SUD) treatment and protect them from outside substance use triggers. Many patients expressed interest in outpatient antimicrobial treatment options conditional upon perceived efficacy of the treatment, perceived ability to complete treatment, and available resources and social support. Finally, both patients and community partners emphasized the importance of autonomy and inclusion in medical decision making. Although some participants acknowledged their SUD, withdrawal symptoms, or undertreated pain might interfere with decision making, they felt these medical conditions were not justification for health care professionals withholding treatment options. They recommended open communication to build trust and reduce harms.

Conclusion

Patients with IDU-associated infections desire autonomy, respect, and patient-centered care from healthcare workers, and may self-discharge when needs or preferences are not met. Involving patients in treatment decisions and offering outpatient antimicrobial options may result in better outcomes. However, patient involvement in decision making may be complicated by many contextual factors unique to each patient, suggesting a need for shared decision making to meet the needs of hospitalized patients with IDU-associated infections.

'I feel like they're actually listening to me': a pilot study of a hospital discharge decision-making conversation guide for patients with injection drug use-associated infections

Background

The prevalence of injection drug use (IDU)-associated infections and associated hospitalizations has been increasing for nearly two decades. Due to issues ranging from ongoing substance use to peripherally inserted central catheter safety, many clinicians find discharge decision-making challenging. Typically, clinicians advise patients to remain hospitalized for several weeks for intravenous antimicrobial treatment; however, some patients may desire other antimicrobial treatment options. A structured conversation guide, delivered by infectious disease physicians, intended to inform hospital discharge decisions has the potential to enhance patient participation in decisions. We developed a conversation guide in order to: (1) investigate its feasibility and acceptability and (2) examine experiences, outcomes, and lessons learned from use of the guide.

Methods

We interviewed physicians after they each piloted the conversation guide with two patients. We interviewed patients immediately after the conversation and again 4-6 weeks later. Two analysts indexed transcriptions and used the framework method to identify and organize relevant information. We conducted retrospective chart review to corroborate and contextualize qualitative data.

Results

Eight patients and four infectious disease physicians piloted the conversation guide. All patients (N = 8) completed antimicrobial treatment. Nearly all participants believed the conversation guide was important for incorporating patient values and preferences. Patients reported an increased sense of autonomy, but felt post-discharge needs could be better addressed. Physician participants identified the guide's long length and inclusion of pain management as areas for improvement.

Conclusions

A novel conversation guide to inform hospital discharge decision-making for patients with IDU-associated infections was feasible, acceptable, and fostered the incorporation of patient preferences and values into decisions. While we identified areas for improvement, overall participants believed that this novel conversation guide helped to improve patient care and autonomy.

Health care professional perspectives on discharging hospitalized patients with injection drug use-associated infections

Background:

Patients with injection drug use (IDU)-associated infections traditionally experience prolonged hospitalizations, which often result in negative experiences and bad outcomes. Harm reduction approaches that value patient autonomy and shared decision-making regarding outpatient treatment options may improve outcomes. We sought to identify health care professionals (HCPs) perspectives on the barriers to offering four different options to hospitalized people who use drugs (PWUD): long-term hospitalization, oral antibiotics, long-acting antibiotics at an infusion center, and outpatient parenteral antibiotics.

Methods:

We recruited HCPs (n = 19) from a single tertiary care center in Portland, Maine. We interviewed HCPs involved with discharge decision-making and other HCPs involved in the specialized care of PWUD. Semi-structured interviews elicited lead HCP values, preferences, and concerns about presenting outpatient antimicrobial treatment options to PWUD, while support HCPs provided contextual information. We used the iterative categorization approach to code and thematically analyze transcripts.

Results:

HCPs were willing to present outpatient treatment options for patients with IDU-associated infections, yet several factors contributed to reluctance. First, insufficient resources, such as transportation, may make these options impractical. However, HCPs may be unaware of existing community resources or viable treatment options. They also may believe the hospital protects patients, and that discharging patients into the community exposes them to structural harms. Some HCPs are concerned that patients with substance use disorder will not make ‘good’ decisions regarding outpatient antimicrobial options. Finally, there is uncertainty about how responsibility for offering outpatient treatment is shared across changing care teams.

Conclusion:

HCPs perceive many barriers to offering outpatient care for people with IDU-associated infections, but with appropriate interventions to address their concerns, may be open to considering more options. This study provides important insights and contextual information that can help inform specific harm reduction interventions aimed at improving care of people with IDU-associated infections.

Getting patients back for routine colorectal cancer screening: Randomized controlled trial of a shared decision-making intervention

Thousands of colonoscopies were canceled during the initial surge of the COVID-19 pandemic. As facilities resumed services, some patients were hesitant to reschedule. The purpose of this study was to determine whether a decision aid plus telephone coaching would increase colorectal cancer (CRC) screening and improve patient reports of shared decision making (SDM). A randomized controlled trial assigned adults aged 45-75 without prior history of CRC who had a colonoscopy canceled from March to May 2020 to intervention (n = 400) or usual care control (n = 400) arms. The intervention arm received three-page decision aid and call from decision coach from September 2020 through November 2020. Screening rates were collected at 6 months. A subset (n = 250) in each arm was surveyed 8 weeks after randomization to assess SDM (scores range 0-4, higher scores indicating more SDM), decisional conflict, and screening preference. The sample was on average, 60 years old, 53% female, 74% White, non-Hispanic, and 11% Spanish speaking. More intervention arm patients were screened within 6 months (35% intervention vs 23% control, p < 0.001). The intervention respondents reported higher SDM scores (mean difference 0.7 [0.4, 0.9], p < 0.001) and less decisional conflict than controls (-21% [-35%, -7%], p = 0.003). The majority in both arms preferred screening versus delaying (68% intervention vs. 65% control, p = 0.75). An SDM approach that offered alternatives and incorporated patients' preferences resulted in higher screening rates. Patients who are overdue for CRC screening may benefit from proactive outreach with SDM support.

Association of Vitamin D Prescribing and Clinical Outcomes in Adults Hospitalized with COVID-19

It is unclear whether vitamin D benefits inpatients with COVID-19. Objective: To examine the relationship between vitamin D and COVID-19 outcomes. Design: Cohort study. Setting: National COVID Cohort Collaborative (N3C) database. Patients: 158,835 patients with confirmed COVID-19 and a sub-cohort with severe disease (n = 81,381) hospitalized between 1 January 2020 and 31 July 2021. Methods: We identified vitamin D prescribing using codes for vitamin D and its derivatives. We created a sub-cohort defined as having severe disease as those who required mechanical ventilation or extracorporeal membrane oxygenation (ECMO), had hospitalization >5 days, or hospitalization ending in death or hospice. Using logistic regression, we adjusted for age, sex, race, BMI, Charlson Comorbidity Index, and urban/rural residence, time period, and study site. Outcomes of interest were death or transfer to hospice, longer length of stay, and mechanical ventilation/ECMO. Results: Patients treated with vitamin D were older, had more comorbidities, and higher BMI compared with patients who did not receive vitamin D. Vitamin D treatment was associated with an increased odds of death or referral for hospice (adjusted odds ratio (AOR) 1.10: 95% CI 1.05−1.14), hospital stay >5 days (AOR 1.78: 95% CI 1.74−1.83), and increased odds of mechanical ventilation/ECMO (AOR 1.49: 95% CI 1.44−1.55). In the sub-cohort of severe COVID-19, vitamin D decreased the odds of death or hospice (AOR 0.90, 95% CI 0.86−0.94), but increased the odds of hospital stay longer >5 days (AOR 2.03, 95% CI 1.87−2.21) and mechanical ventilation/ECMO (AOR 1.16, 95% CI 1.12−1.21). Limitations: Our findings could reflect more aggressive treatment due to higher severity. Conclusion: Vitamin D treatment was associated with greater odds of extended hospitalization, mechanical ventilation/ECMO, and death or hospice referral.

Cross-sectional Survey Examining Patient Attitudes and Preferences for Rescheduling Screening Colonoscopies Canceled due to the COVID-19 Pandemic

Background. Early in the COVID-19 pandemic colonoscopies for colorectal cancer (CRC) screening were canceled. Patient perceptions of the benefits and risks of routine screening relative to health concerns associated with the COVID-19 pandemic were unknown. Purpose. Assess patient anxiety, worry, and interest in CRC screening during the COVID-19 pandemic. Methods. A random sample of 200 patients aged 45 to 75 y with colonoscopy cancellation due to COVID-19 in March to May 2020 were surveyed. Anxiety, COVID-19 and CRC risk perceptions, COVID-19 and CRC worry, likelihood of following through with colonoscopy in the next month, and interest in alternatives to colonoscopy were assessed. Subsequent screening was tracked for 12 mo. Results. Respondents (N = 127/200, 63.5%) were on average 60 y old, female (59%), college educated (62% college degree or more), and White (91%). A substantial portion of patients (46%) stated they may not follow through with a colonoscopy in the next month. There was greater interest in stool-based testing than in delaying screening (48% v. 26%). Women, older patients, and patients indicating tolerance of uncertainty due to complexity reported they were less likely to follow through with colonoscopy in the next month. Greater interest in stool-based testing was related to lower perceptions of CRC risk. Greater interest in delaying screening was related to less worry about CRC and less tolerance of risk. Over 12 mo, 60% of participants completed screening. Patients who stated they were more likely to screen in the next month were more likely to complete CRC screening (P = 0.01). Conclusions. Respondents who had a colonoscopy canceled during the COVID-19 pandemic varied in interest in rescheduling the procedure. A shared decision-making approach may help patients address varying concerns and select the best approach to screening for them.

HIGHLIGHTS

In the wake of the first wave of the COVID-19 pandemic, almost half of patients stated they were not likely to follow through with a colonoscopy in the short term, about half were interested in screening with a stool-based test, and only one-quarter were interested in delaying screening until next year.Patients who perceived themselves at higher risk of colorectal cancer were less interested in stool-based testing, and patients who were more worried about colorectal cancer were less interested in delaying screening.A shared decision-making approach may be necessary to tailor screening discussions for patients during subsequent waves of the pandemic, other occasions where resources are limited and patient preferences vary, or where patients hold conflicting views of screening.

382. Vitamin D Supplementation and Covid 19: Results from the U.S. N3C Data Enclave

Background

It is estimated that 18% of adults in the U.S. take Vitamin D supplements. Some observational studies suggest that vitamin D supplementation activates the innate immune system and reduces the incidence and severity of viral infections. During the SARS-CoV-2 pandemic, vitamin D supplements were touted as a potential therapy to prevent the disease and/or complications. However, supportive evidence is lacking.

Methods

The National COVID Cohort Collaborative (N3C) enclave is the largest COVID-19 data base with nearly 1.4 million positive patients at 56 sites in the U.S. We performed a retrospective analysis of vitamin D supplementation, either prescribed before or during hospitalization for SARS-CoV-2.

Results

137,399 people took vitamin D supplements out of 1.4 million. Females prescribed vitamin D outnumbered males by almost 2:1, whereas in non-users there were no sex differences. Most supplement users were older than 50. African Americans constituted 13% of the non-users, but 23% of those prescribed vitamin D. Infected individuals with any vitamin D supplementation, pre-Covid, post-Covid or both, had a 6.66% mortality rate vs 2% mortality in non-users. Similarly, nearly a third of the supplement users were hospitalized compared to 11% in the non-users. The Charlson Co-Morbidity Index was 3.0±3 (SD) in users vs 1.0±2 (SD) in non-users.

Conclusion

10% of SARS-CoV-2 infected patients were taking vitamin D. They tended to be older, more likely to be African American and have significant co-morbidities. Hospitalization and mortality were higher among those taking Vitamin D in this cohort. Vitamin D is widely used to prevent and treat SARS-CoV-2 but without evidence of efficacy.

Disclosures

Sally L. Hodder, M.D., Gilead (Advisor or Review Panel member)Merck (Grant/Research Support, Advisor or Review Panel member)Viiv Healthcare (Grant/Research Support, Advisor or Review Panel member)

Variation in additional testing and patient outcomes after stress echocardiography or myocardial perfusion imaging, according to accreditation status of testing site

BACKGROUND

The purpose of the present study was to determine whether patients receiving a stress echocardiogram or myocardial perfusion imaging (MPI) test have differences in subsequent testing and outcomes according to accreditation status of the original testing facility.

METHODS AND RESULTS

An all-payer claims dataset from Maine Health Data Organization from 2012 to 2014 was utilized to define two cohorts defined by an initial stress echocardiogram or MPI test. The accreditation status (Intersocietal Accreditation Commission (IAC), American College of Radiology (ACR) or none) of the facility performing the index test was known. Descriptive statistics and multivariate regression were used to examine differences in subsequent diagnostic testing and cardiac outcomes. We observed 4603 index stress echocardiograms and 8449 MPI tests. Multivariate models showed higher odds of subsequent MPI testing and hospitalization for angina if the index test was performed at a non-accredited facility in both the stress echocardiogram cohort and the MPI cohort. We also observed higher odds of percutaneous coronary interventions (PCI) performed (OR 1.68, 95% CI 1.13-2.50), if the initial MPI test was done in a non-accredited facility.

CONCLUSION

Cardiac testing completed in non-accredited facilities were associated with higher odds of subsequent MPI testing, hospitalization for angina, and PCI.

Prescription opioid policies and associations with opioid overdose and related adverse effects

BACKGROUND

United States (US) policies to mitigate the opioid epidemic focus on reducing access to prescription opioids to prevent overdoses. We examined the impact of state policies in Vermont (July 2017) and Maine (July 2016) on opioid overdoses and opioid-related adverse effects.

METHODS

Study population included patients 15 years and older in all-payer claims of Vermont (N = 597,683; Jan.2016-Dec.2018) and Maine (N = 1,370,960; Oct.2015-Dec.2017). We used interrupted time series analyses to assess the impact of opioid prescribing policies on monthly opioid overdose rate and opioid-related adverse effects rate. We used the International Classification of Disease-10-CM to identify overdoses (T40.0 × 1-T40.4 × 4, T40.601-T40.604, T40.691-T40.694) and adverse effects (T40.0 × 5, T40.2 × 5-T40.4 × 5, T40.605, T40.695).

RESULTS

Immediately after the policy, the level of Vermont's opioid overdose rate increased by 34% (95% confidence interval, CI: 1.09, 1.65) while the level of opioid-related adverse effects rate decreased by 29% (95% CI: 0.58, 0.87). In Maine, there was no level change in opioid overdose rate, but the slope of the adverse effects rate after the policy decreased by 3.5% (95% CI: 0.94, 0.99). These results varied within age and rurality subgroups in both states.

CONCLUSION

While the decrease in rate of adverse effects following the policy changes is promising, the increase in Vermont's opioid overdose rate may suggest there is an association between policy implementation and short-term risk to public health.

Influence of Opioid Prescription Policy on Overdoses and Related Adverse Effects in a Primary Care Population

BACKGROUND

In response to the opioid epidemic, many states have enacted policies limiting opioid prescriptions. There is a paucity of evidence of the impact of opioid prescribing interventions in primary care populations, including whether unintended consequences arise from limiting the availability of prescribed opioids.

OBJECTIVE

Our aim was to compare changes in opioid overdose and related adverse effects rate among primary care patients following the implementation of state-level prescribing policies.

DESIGN

A cohort of primary care patients within an interrupted time series model.

PARTICIPANTS

Electronic medical record data for 62,776 adult (18+ years) primary care patients from a major medical center in Vermont from January 1, 2016, to June 30, 2018.

INTERVENTIONS

State-level opioid prescription policy changes limiting dose and duration.

MAIN MEASURES

Changes in (1) opioid overdose rate and (2) opioid-related adverse effects rate per 100,000 person-months following the July 1, 2017, prescription policy change.

KEY RESULTS

Among primary care patients, there was no change in opioid overdose rate following implementation of the prescribing policy (incidence rate ratio; IRR: 0.64, 95% confidence interval; CI: 0.22-1.88). There was a 78% decrease in the opioid-related adverse effects rate following the prescribing policy (IRR: 0.22, 95%CI: 0.09-0.51). This association was moderated by opioid prescription history, with decreases observed among opioid-naïve patients (IRR: 0.18, 95%CI: 0.06-0.59) and among patients receiving chronic opioid prescriptions (IRR: 0.17, 95%CI: 0.03-0.99), but not among those with intermittent opioid prescriptions (IRR: 0.51, 95%CI: 0.09-2.82).

CONCLUSIONS

Limiting prescription opioids did not change the opioid overdose rate among primary care patients, but it reduced the rate of opioid-related adverse effects in the year following the state-level policy change, particularly among patients with chronic opioid prescription history and opioid-naïve patients. Limiting the quantity and duration of opioid prescriptions may have beneficial effects among primary care patients.

Abstract A117: Area deprivation index and rurality in relation to lung cancer prevalence and mortality in a rural state

Objectives: Lung cancer is the leading cause of cancer-related mortality. In rural areas, socioeconomic deprivation and geographic barriers to care may both influence lung cancer prevalence and outcomes. We sought to describe how socioeconomic deprivation and rurality are related to lung cancer prevalence and mortality. Methods: We conducted a population-based cross-sectional analysis of: 1) prevalent lung cancers identified in a statewide all-payer claims dataset between 2012 and 2016; 2) lung cancer deaths in Maine from 2012-2016, ascertained by the state death registry; 3) rurality; and 4) area deprivation index (ADI), a geographic area-based measure of socioeconomic deprivation. Analyses examined rate ratios for lung cancer prevalence and mortality according to rurality (small/isolated rural, large rural, or urban) and ADI (in quintiles, with highest reflecting the most deprivation) and after adjusting for age, sex, and area-level smoking rates as determined by Behavioral Risk Factor Surveillance System data. Results: Among 1,223,006 adults aged 20+ in the all-payer claims dataset during the 5-year observation period, 8300 received care for prevalent lung cancer and there were 4618 deaths from lung cancer between 2012 and 2016. Of the prevalent lung cancer cases, 36.1% resided in isolated or small rural areas, and 42.6% resided in the highest two quintiles for ADI. Increasing rurality was positively associated with lung cancer prevalence and mortality, but these associations did not persist after adjusting for age, sex and smoking rates. However, increasing ADI was positively associated with both lung cancer prevalence (rate ratio 1.41 (95% CI 1.30-1.54) for ADI quintile 5 vs. quintile 1) and mortality (rate ratio for ADI quintile 5 vs. quintile 1, 1.59 (95% CI 1.41-1.79) in multivariable models adjusted for age, sex, and smoking rates. Conclusion: Socioeconomic deprivation was associated with higher lung cancer prevalence and mortality, but rurality was not. These findings suggest that interventions aimed at improving access to lung cancer prevention, screening, and treatment services should target populations with socioeconomic deprivation, rather than rurality per se. Future research should examine the relationship between socioeconomic deprivation and other indicators of population health in rural areas.

Citation Format: Kathleen M Fairfield, Adam Black, Erika Ziller, Kimberly Murray, Lee Lucas, Leo B Waterston, Neil Korsen, Darlene Ineza, Paul KJ Han. Area deprivation index and rurality in relation to lung cancer prevalence and mortality in a rural state [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr A117.

Area Deprivation Index and Rurality in Relation to Lung Cancer Prevalence and Mortality in a Rural State

Background

We sought to describe lung cancer prevalence and mortality in relation to socioeconomic deprivation and rurality.

Methods

We conducted a population-based cross-sectional analysis of prevalent lung cancers from a statewide all-payer claims dataset from 2012 to 2016, lung cancer deaths in Maine from the state death registry from 2012 to 2016, rurality, and area deprivation index (ADI), a geographic area-based measure of socioeconomic deprivation. Analyses examined rate ratios for lung cancer prevalence and mortality according to rurality (small and isolated rural, large rural, or urban) and ADI (quintiles, with highest reflecting the most deprivation) and after adjusting for age, sex, and area-level smoking rates as determined by the Behavioral Risk Factor Surveillance System.

Results

Among 1 223 006 adults aged 20 years and older during the 5-year observation period, 8297 received lung cancer care, and 4616 died. Lung cancer prevalence and mortality were positively associated with increasing rurality, but these associations did not persist after adjusting for age, sex, and smoking rates. Lung cancer prevalence and mortality were positively associated with increasing ADI in models adjusted for age, sex, and smoking rates (prevalence rate ratio for ADI quintile 5 compared with quintile 1 = 1.41, 95% confidence interval [CI] =1.30 to 1.54) and mortality rate ratio = 1.59, 95% CI = 1.41 to 1.79).

Conclusion

Socioeconomic deprivation, but not rurality, was associated with higher lung cancer prevalence and mortality. Interventions should target populations with socioeconomic deprivation, rather than rurality per se, and aim to reduce lung cancer risk via tobacco treatment and control interventions and to improve patient access to lung cancer prevention, screening, and treatment services.

Association between the food and physical activity environment, obesity, and cardiovascular health across Maine counties

BACKGROUND

Accounting for nearly one-third of all deaths, cardiovascular disease is the leading cause of mortality and morbidity in the United States. Adverse health behaviors are major determinants of this high incidence of disease. Examining local food and physical activity environments and population characteristics in a poor, rural state may highlight underlying drivers of these behaviors. We aimed to identify demographic and environmental factors associated with both obesity and overall poor cardiovascular health (CVH) behaviors in Maine counties.

METHODS

Our cross-sectional study analyzed 40,398 Behavioral Risk Factor Surveillance System (BRFSS) 2011-2014 respondents alongside county-level United States Department of Agriculture (USDA) Food Environment Atlas 2010-2012 measures of the built environment (i.e., density of restaurants, convenience stores, grocery stores, and fitness facilities; food store access; and county income). Poor CVH score was defined as exhibiting greater than 5 out of the 7 risk factors defined by the American Heart Association (current smoking, physical inactivity, obesity, poor diet, hypertension, diabetes, and high cholesterol). Multivariable logistic regression models described the contributions of built environment variables to obesity and overall poor CVH score after adjustment for demographic controls.

RESULTS

Both demographic and environmental factors were associated with obesity and overall poor CVH. After adjustment for demographics (age, sex, personal income, and education), environmental characteristics most strongly associated with obesity included low full-service restaurant density (OR 1.34; 95% CI 1.24-1.45), low county median household income (OR 1.31; 95% CI 1.21-1.42) and high convenience store density (OR 1.21; 95% CI 1.12-1.32). The strongest predictors of overall poor CVH behaviors were low county median household income (OR 1.30; 95% CI 1.13-1.51), low full-service restaurant density (OR 1.38; 95% CI 1.19-1.59), and low fitness facility density (OR 1.27; 95% CI 1.11-1.46).

CONCLUSIONS

In a rural state, both demographic and environmental factors predict overall poor CVH. These findings may help inform communities and policymakers of the impact of both social determinants of health and local environments on health outcomes.

Association Between Rurality and Lung Cancer Treatment Characteristics and Timeliness

BACKGROUND

Lung cancer is the leading cause of cancer-related mortality in the United States, and rural states bear a greater burden of disease.

METHODS

We analyzed tumor registry data to examine relationships between rurality and lung cancer stage at diagnosis and treatment. Cases were from the Maine Cancer Registry from 2012 to 2015, and rurality was defined using rural-urban commuting areas. Multivariable models were used to examine the relationships between rurality and treatment, adjusting for age, sex, poverty, education, insurance status, and cancer stage.

RESULTS

We identified 5,338 adults with incident lung cancer; 3,429 (64.2%) were diagnosed at a late stage (III or IV). Rurality was not associated with stage at diagnosis. For patients with early-stage disease (I or II), rurality was not associated with receipt of treatment. However, for patients with late-stage disease, residents of large rural areas received more surgery (10%) compared with metropolitan (9%) or small/isolated rural areas (6%), P = .01. In multivariable analyses, patients in large rural areas received more chemotherapy (OR 1.48; 95% CI: 1.08-2.02) than those in metropolitan areas. Patients with early-stage disease residing in small/ isolated rural areas had delays in treatment (median time to first treatment = 43 days, interquartile range [IQR] 22-68) compared with large rural (34 days, IQR 17-55) and metropolitan areas (35 days, IQR 17-60), P = .0009.

CONCLUSION

Rurality is associated with differences in receipt of specific lung cancer treatments and in timeliness of treatment.

Variations in Hospice Utilization and Length of Stay for Medicare Patients With Melanoma

CONTEXT

Timely hospice referral is an indicator of high-quality end-of-life care for cancer patients. Variations in patient characteristics associated with hospice utilization and length of stay have been demonstrated in studies of other malignancies but not melanoma.

OBJECTIVES

We sought to understand hospice utilization and patient characteristics associated with variability in use for the older melanoma population.

METHODS

We used the Surveillance, Epidemiology, and End Results-Medicare database to identify 13,393 melanoma patients aged 65+ years at time of diagnosis between 2000 and 2009, who died by 12/31/10. The primary outcome was enrollment in hospice with secondary outcome of hospice duration. Patient characteristics associated with variations in hospice enrollment were examined.

RESULTS

Among 13,393 patients who died with melanoma, 5298 (40%) received hospice care. Of these, 17% were enrolled in hospice for three days or less, while 13% had ≥90 days of hospice care. Despite improvements over time in the proportion of patients who received hospice and those who received at least 90 days of hospice care, late hospice enrollments did not change. Multivariable analysis revealed that patients of older age, with distant disease at time of diagnosis, and residing in rural areas or in census tracts with higher rates of high school completion were more likely to enroll in hospice.

CONCLUSION

Rates of hospice enrollment increased over time but remained under accepted quality benchmarks with variations evident in those who receive hospice services. Efforts to increase access to earlier hospice care for all patients dying with melanoma are essential.

Behavioral Risk Factors and Regional Variation in Cardiovascular Health Care and Death

INTRODUCTION

Reducing the burden of death from cardiovascular disease includes risk factor reduction and medical interventions.

METHODS

This was an observational analysis at the hospital service area (HSA) level, to examine regional variation and relationships between behavioral risks, health services utilization, and cardiovascular disease mortality (the outcome of interest). HSA-level prevalence of cardiovascular disease behavioral risks (smoking, poor diet, physical inactivity) were calculated from the Behavioral Risk Factor Surveillance System; HSA-level rates of stress tests, diagnostic cardiac catheterization, and revascularization from a statewide multi-payer claims data set from Maine in 2013 (with 606,260 patients aged ≥35 years), and deaths from state death certificate data. Analyses were done in 2016.

RESULTS

There were marked differences across 32 Maine HSAs in behavioral risks: smoking (12.4%-28.6%); poor diet (43.6%-73.0%); and physical inactivity (16.4%-37.9%). After adjustment for behavioral risks, rates of utilization varied by HSA: stress tests (28.2-62.4 per 1,000 person-years, coefficient of variation=17.5); diagnostic cardiac catheterization (10.0-19.8 per 1,000 person-years, coefficient of variation=17.3); and revascularization (4.6-6.2 per 1,000 person-years; coefficient of variation=9.1). Strong HSA-level associations between behavioral risk factors and cardiovascular disease mortality were observed: smoking (R²=0.52); poor diet (R²=0.38); and physical inactivity (R²=0.35), and no association between revascularization and cardiovascular disease mortality after adjustment for behavioral risk factors (R²=0.02). HSA-level behavioral risk factors were also strongly associated with all-cause mortality: smoking (R²=0.57); poor diet (R²=0.49); and physical inactivity (R²=0.46).

CONCLUSIONS

There is substantial regional variation in behavioral risks and cardiac utilization. Behavioral risk factors are associated with cardiovascular disease mortality regionally, whereas revascularization is not. Efforts to reduce cardiovascular disease mortality in populations should focus on prevention efforts targeting modifiable risk factors.

Knowledge and values for cancer screening decisions: Results from a national survey

BACKGROUND

Guidelines recommend shared decision making (SDM) for cancer screening decisions. SDM requires providers to ensure that patients are informed about screening issues and to support decisions that are concordant with patient values. We evaluated decision-quality factors for breast, colorectal, and prostate cancer screening decisions.

METHODS

We conducted a national, population-based Internet survey of adults aged 40+ to characterize perceptions about about cancer screening, the importance of information sources, cancer screening knowledge, values and preferences for screening, and the most influential drivers of decisions.

RESULTS

Among 1452 participants who completed the survey, the mean age was 60, and 94% were insured. Most participants reported feeling well informed about cancer screening, though only 21% reported feeling extremely well informed. Most participants correctly answered about 50% of the knowledge questions, with the majority markedly overestimating lifetime risk of cancer diagnoses and mortality. Participants rated health care providers as the most important source of information.

CONCLUSION

Although respondents considered themselves well informed about cancer they performed poorly on knowledge questions. This discordance suggests the potential for poor-quality decision making.

PRACTICE IMPLICATIONS

To improve the quality of decision making, providers need training to utilize decision support tools and time to carry out SDM.

Decisions about medication use and cancer screening across age groups in the United States

OBJECTIVE

To describe decision process and quality for common cancer screening and medication decisions by age group.

METHODS

We included 2941 respondents to a national Internet survey who made at least one decision about colorectal, breast, and prostate cancer screening, blood pressure or cholesterol medications. Respondents were queried about decision processes.

RESULTS

Across the five decisions considered, decision process scores were similar (and generally low) across age groups for medication and cancer screening, indicating that all groups had poor involvement in medical decision making. Overall knowledge scores were low across age groups, with elderly (75+) having slightly higher knowledge about medications vs. younger respondents. Elderly respondents reported similar goals and concerns when making decisions, though placed greater importance of having peace of mind from a normal result for cancer screening vs. younger respondents.

CONCLUSION

Across age groups, respondents reported poor decision processes about common medications and cancer screening, despite little evidence of benefit for some interventions (cancer screening, cholesterol lowering medicines in low risk elderly) and possibility of harm in the elderly.

PRACTICE IMPLICATIONS

Particular care should be taken to help patients understand both benefit and risk of screening tests and routine medications.

Intraperitoneal chemotherapy among women in the Medicare population with epithelial ovarian cancer

BACKGROUND

Intraperitoneal combined with intravenous chemotherapy (IV/IP) for primary treatment of epithelial ovarian cancer results in a substantial survival advantage for women who are optimally debulked surgically, compared with standard IV only therapy (IV). Little is known about the use of this therapy in the Medicare population.

METHODS

We used the Surveillance, Epidemiology, and End Results (SEER)-Medicare database to identify 4665 women aged 66 and older with epithelial ovarian cancer diagnosed between 2005-2009, with their Medicare claims. We defined receipt of any IV/IP chemotherapy when there was claims evidence of any receipt of such treatment within 12 months of the date of diagnosis. We used descriptive statistics to examine factors associated with treatment and health services use.

RESULTS

Among 3561 women with Stage III or IV epithelial ovarian cancer who received any chemotherapy, only 124 (3.5%) received IV/IP chemotherapy. The use of IV/IP chemotherapy did not increase over the period of the study. In this cohort, younger women, those with fewer comorbidities, whites, and those living in Census tracts with higher income were more likely to receive IV/IP chemotherapy. Among women who received any IV/IP chemotherapy, we did not find an increase in acute care services (hospitalizations, emergency department visits, or ICU stays).

CONCLUSION

During the period between 2005 and 2009, few women in the Medicare population living within observed SEER areas received IV/IP chemotherapy, and the use of this therapy did not increase. We observed marked racial and sociodemographic differences in access to this therapy.

Disparities in hospice care among older women dying with ovarian cancer

BACKGROUND

Timely hospice referral is an essential component of quality end-of-life care, although a growing body of research suggests that for patients with various types of cancer, hospice referrals often occur very late in the course of care, and are marked by sociodemographic disparities. However, little is known about the ovarian cancer patient population specifically. We examined the extent and timing of hospice referrals in ovarian cancer patients over age 65, and the factors associated with these outcomes.

METHODS

We used the Surveillance, Epidemiology, and End Results (SEER)-Medicare database to identify 8211 women aged 66+ with ovarian cancer who were diagnosed between 2001 and 2005 and died by December 31, 2007. We excluded women who were not eligible for Medicare A continuously during the 6 months prior to death. Outcomes studied included overall hospice use in the last 6 months of life and late hospice enrollment, defined as within 3 days of death. We examined variations in these two measures based on year of diagnosis and sociodemographic characteristics (age, race, marital status, rural residence, income, education) and type of Medicare received (fee-for-service vs. managed care).

RESULTS

Among 8211 women in the cohort who died from ovarian cancer, 39.7% never received hospice care (3257/8211). Overall hospice care increased over the period of observation, from 49.7% in 2001 to 63.6% [corrected] in 2005, but the proportion of women receiving hospice care within 3 days of death did not improve. Among those who received hospice care, 11.2% (556/4954) and 26.2% (1299/4954) received such care within 3 and 7 days of death, respectively. A higher proportion of black women (46.5% vs. 38.4% among whites), women in the lowest income group (42.8% vs. 37.0% in the highest income group), and those receiving fee-for-service Medicare (41.3% vs.33.5% for women in managed care) never received hospice care. In multivariable models, factors associated with lack of hospice care included age younger than 80 years (OR 1.27, 95% CI 1.15-1.40), non-white race (OR 1.44, 95% CI 1.26-1.65), low income (OR 1.17, 95% CI 1.04-1.32) and enrollment in fee-for-service Medicare compared with managed care (OR 1.39, 95% CI 1.24-1.56).

CONCLUSION

More older women with ovarian cancer are receiving hospice care over time, however, a substantial proportion receive such care very near death, and sociodemographic disparities in hospice care exist. Our data also support the need to target lower-income and minority women in efforts to increase optimally timed hospice referrals in this population. Our finding that ovarian cancer patients enrolled in managed care plans were more likely to receive hospice care suggests the importance of health care system factors in the utilization of hospice services.

Completion of Adjuvant Chemotherapy and Use of Health Services for Older Women With Epithelial Ovarian Cancer

Purpose

This analysis identifies factors associated with completion of adjuvant chemotherapy for patients with ovarian cancer and subsequent use of health services.

Patients and Methods

We used the Surveillance, Epidemiology, and End Results (SEER) –Medicare database to identify 4,617 women age 65 years or older with ovarian cancer diagnosed from 2001 to 2005. By using multivariable analyses with completion of chemotherapy as the outcome of interest, we describe factors associated with completion of treatment, including age, race, marital status, comorbidities, and sociodemographic factors. Use of health services was captured from Medicare claims.

Results

Among 4,617 patients with untreated ovarian cancer, 1,329 (28.8%) received no chemotherapy, 1,139 (24.7%) received a partial course of chemotherapy, and 2,149 (46.5%) completed chemotherapy. Women age 75 years or older were at greater risk of incomplete chemotherapy versus women age 65 to 74 years (odds ratio [OR], 1.64; 95% CI, 1.33 to 2.04). Having two or more comorbidities was also significantly associated with incomplete chemotherapy (OR, 1.83; 95% CI, 1.34 to 2.50). Among women who received either a partial or complete course of chemotherapy, we did not find an increase in use of health services (hospitalizations, emergency department visits, or physician visits) for the oldest women (age 80 years or older) compared with younger women.

Conclusion

There is considerable room for improvement in helping older patients with ovarian cancer initiate and complete chemotherapy. The oldest women who completed chemotherapy in this study did not use health services more than younger women did. Treatment teams for older patients with ovarian cancer should include expertise in geriatric assessment, should carefully identify medical and psychosocial barriers to completing treatment, and should support patients throughout treatment.

Regional variation in cancer-directed surgery and mortality among women with epithelial ovarian cancer in the Medicare population

BACKGROUND

Regional differences in health services can point to disparities in access to healthcare. The authors performed a population-based cohort study to examine differences in ovarian cancer treatment and mortality according to geographic region.

METHODS

The Surveillance, Epidemiology, and End Results (SEER)-Medicare database was used to identify 4589 women aged ≥65 years with ovarian cancer diagnosed between 1998 and 2002 who had Medicare claims filed from 1998 to 2005. Hospital Referral Region (HRR) was assigned according to patient zip code. The authors calculated the proportion of women in each HRR who underwent cancer-directed surgery. With HRR as the predictor of interest, mortality and the receipt of cancer-directed surgery were described in multivariate analyses.

RESULTS

Among 4589 women with ovarian cancer, 3286 underwent cancer-directed surgery. The receipt of cancer-directed surgery varied by HRR (range, 53%-88%). Women were less likely to undergo cancer-directed surgery if they were older, nonwhite, had higher stage disease, or had more comorbidities. For example, white women were more likely to undergo such surgery (odds ratio, 1.41; 95% confidence interval, 1.10-1.82) compared with all nonwhite women. HRR was a significant predictor of cancer-directed surgery (P = .01). A significant correlation was observed between HRR and all-cause mortality (P = .02); however, after adjusting for cancer-directed surgery, that correlation was no longer significant (P = .10).

CONCLUSIONS

There was regional variation in mortality among Medicare recipients with ovarian cancer, and access to cancer-directed surgery explained some of that variation. Improving access to high-quality cancer surgery for ovarian cancer may improve outcomes, particularly for minorities and for older women.

The effect of overnight in-house attending coverage on perceptions of care and education on a general medical service

BACKGROUND

An increased emphasis on patient safety has led to calls for closer supervision of medical trainees. It is unclear what effect an increased degree of faculty presence will have on educational and clinical outcomes. The aim of this study was to evaluate resident and attending attitudes and preferences regarding overnight attending supervision.

METHODS

This study was a cross-sectional electronic survey of physicians. Participants were resident and faculty physicians recently on inpatient service rotations after implementation of an overnight attending coverage system.

RESULTS

Of 58 total respondents, most faculty (91%) and resident (92%) physicians reported they were satisfied with the overall quality of care delivered and believed the quality of care delivered overnight improved with an in-house attending system (90% and 85%, respectively). Most resident physicians (82%) believed the educational experience improved with the system of increased attending availability. Nearly all faculty (95%) and resident (97%) physicians preferred the in-house attending system to the traditional system of attendings being available by pager. The implementation of such coverage resulted in increased cost to the hospital for compensating covering hospitalist physicians.

CONCLUSION

In-house attending coverage was acceptable to both residents and faculty, with perceived improvements in quality and educational experience.

Breast, cervical, and colorectal cancer screening rates amongst female Cambodian, Somali, and Vietnamese immigrants in the USA

INTRODUCTION

Minority women, particularly immigrants, have lower cancer screening rates than Caucasian women, but little else is known about cancer screening among immigrant women. Our objective was to assess breast, cervical, and colorectal cancer screening rates among immigrant women from Cambodia, Somalia, and Vietnam and explore screening barriers.

METHODS

We measured screening rates by systematic chart review (N = 100) and qualitatively explored screening barriers via face-to-face questionnaire (N = 15) of women aged 50-75 from Cambodia, Somalia, and Vietnam attending a general medicine clinic (Portland, Maine, USA).

RESULTS

Chart Review - Somali women were at higher risk of being unscreened for breast, cervical, and colorectal cancer compared with Cambodian and Vietnamese women. A longer period of US residency was associated with being screened for colorectal cancer. We observed a 7% (OR 1.07, 95% CI 1.01-1.13, p = 0.01) increase in the odds that a woman would undergo a fecal occult blood test for each additional year in the US, and a 39% increase in the odds of a woman being screened by colonoscopy or flexible sigmoidoscopy for every five years of additional US residence (OR 1.39, 95% CI 1.21-1.61, p = 0.02). We did not observe statistically significant relationships between odds of being screened by mammography, clinical breast exam or papanicolaou test according to years in the US. Questionnaire - We identified several barriers to breast, cervical, and colorectal cancer screening, including discomfort with exams conducted by male physicians.

DISCUSSION

Somali women were less likely to be screened for breast, cervical, and colorectal cancer than Cambodian and Vietnamese women in this population, and uptake of colorectal cancer screening is associated with years of residency in this country. Future efforts to improve equity in cancer screening among immigrants may require both provider and community education.

Association of oral contraceptive use, other contraceptive methods, and infertility with ovarian cancer risk

Although oral contraceptives are protective for ovarian cancer, it is unclear how long this protection persists. The authors prospectively assessed this question as well as associations of other, less studied contraceptive methods (tubal ligation, rhythm method, diaphragm, condoms, intrauterine device, foam, spousal vasectomy) and infertility with ovarian cancer risk among 107,900 participants in the US Nurses' Health Study. During 28 years of follow-up (1976-2004), 612 cases of invasive epithelial ovarian cancer were confirmed. Duration of oral contraceptive use was inversely associated with risk (p-trend = 0.02), but no clear trend was observed for years since last use. However, for women using oral contraceptives for >5 years, the rate ratio for ovarian cancer for <or=20 years since last use was 0.58 (95% confidence interval (CI): 0.39, 0.87), with no association found for >20 years since last use (rate ratio (RR) = 0.92, 95% CI: 0.61, 1.39). Tubal ligation (RR = 0.66, 95% CI: 0.50, 0.87) was associated with decreased ovarian cancer risk, whereas intrauterine device use (RR = 1.76, 95% CI: 1.08, 2.85) and infertility (RR = 1.36, 95% CI: 1.07, 1.75) were associated with an increased risk. Results suggest that the beneficial effect of oral contraceptives on ovarian cancer risk attenuates after 20 years since last use. Furthermore, tubal ligation, intrauterine device use, and infertility were associated with ovarian cancer risk.

A prospective study of dietary lactose and ovarian cancer

The milk sugar lactose is an hypothesized risk factor for epithelial ovarian cancer because of possible direct toxic effects of its metabolites on oocytes or by compensatory gonadotropin stimulation. Women are presently encouraged to consume dairy products as a source of calcium to prevent osteoporosis. The objective of our study was to prospectively assess lactose, milk and milk product consumption in relation to ovarian cancer risk among 80326 participants in the Nurses' Health Study who had no history of cancer other than nonmelanoma skin cancer. Participants in the Nurses' Health Study reported on known and suspected ovarian cancer risk factors in questionnaires mailed biennially from 1976 to 1996. Food frequency questionnaires were included in the years 1980, 1984, 1986 and 1990. Newly reported ovarian cancer was documented by review of medical records. During 16 years of follow-up (1980-1996), 301 cases of invasive epithelial ovarian cancer were confirmed. Pooled logistic regression was used to control for age, body mass index (kg/m(2)), caffeine intake, oral contraceptive use, smoking history, parity and tubal ligation. For all subtypes of invasive ovarian cancer combined, we observed a nonsignificant 40% greater risk for women in the highest category of lactose consumption compared to the lowest (multivariate relative risk (RR) 1.40, 95% confidence interval (CI), 0.98-2.01). We observed a 2-fold higher risk of the serous ovarian cancer subtype among those in the highest category of lactose consumption compared to the lowest (RR 2.07, 95% CI, 1.27-3.40). For each 11-gram increase in lactose consumption (the approximate amount in one glass of milk), we observed a 20% increase in risk of serous cancers (RR 1.20, 95% CI, 1.04-1.39). Skim and low-fat milk were the largest contributors to dietary lactose. Women who consumed one or more servings of skim or low-fat milk daily had a 32% higher risk of any ovarian cancer (RR 1.32, 95% CI, 0.97-1.82) and a 69% higher risk of serous ovarian cancer (RR 1.69, 95% CI, 1.12-2.56) compared to women consuming 3 or less servings monthly. Controlling for fat intake did not change our findings. Our findings provide some support for the hypothesis that lactose intake increases risk of epithelial ovarian cancer. However, the observed excess risk appeared limited to the serous subtype of ovarian cancer in our study.

Colon cancer risk counseling by health-care providers: perceived barriers and response to an internet-based cancer risk appraisal instrument

BACKGROUND

Risk appraisal instruments may be helpful in reinforcing prevention messages, although little is known about physician acceptance of such instruments.

OBJECTIVE

We explored perceived barriers to colon cancer risk counseling and responses to the colon cancer component of an Internet-based risk appraisal instrument.

METHODS

We qualitatively assessed provider perceptions of barriers to colon cancer prevention and screening, and their responses to the Harvard Cancer Risk Index using focus groups of primary care providers.

RESULTS

Many providers commented that the risk appraisal instrument may be most helpful to reinforce messages by a health-care provider. The tool may increase awareness about modifiable risk factors for cancer and help patients prioritize changes as well as improve screening acceptance. With regard to barriers to counseling patients about colon cancer prevention and screening, providers expressed concerns that behaviors are too difficult to change. In addition, they were frustrated by limited time for prevention counseling and poor reimbursement.

CONCLUSIONS

The Internet-based risk index was well accepted, although providers thought it would be most effective when used to complement provider messages about prevention. Use of an Internet-based risk index along with physician counseling could help improve cancer prevention practices and cancer screening acceptance.

Aspirin, other NSAIDs, and ovarian cancer risk (United States)

OBJECTIVE

We sought to evaluate the association between ovarian cancer risk and use of aspirin and nonsteroidal anti-inflammatories.

METHODS

We prospectively assessed use of aspirin, nonsteroidal anti-inflammatories (NSAIDs), and acetaminophen use in relation to ovarian cancer risk among 76,821 participants in the Nurses' Health Study who had no history of cancer other than non-melanoma skin cancer. Women reported known and suspected ovarian cancer risk factors in biennial mailed questionnaires from 1976 to 1996, along with new diagnoses of ovarian cancer. Aspirin use was assessed in 1980, 1982, 1984, and 1988-1994. We assessed NSAID use in 1980, and both NSAID and acetaminophen use in 1990, 1992, and 1994. During 16 years of follow-up and 1,222,412 person-years, 333 cases of invasive epithelial ovarian cancer were confirmed. We used pooled logistic regression to control for age, body mass index, oral contraceptive use, smoking history, parity, postmenopausal hormone use, tubal ligation, and other potential ovarian cancer risk factors.

RESULTS

Aspirin use was not associated with ovarian cancer risk overall (RR for users compared with nonusers, 1.00, 95% confidence interval (CI 0.80-1.25). We found no association between aspirin dose (in number of weekly tablets) and ovarian cancer risk (RR for those taking 15 or more tablets weekly compared with nonusers, 0.98, 95% CI 0.63-1.52). Similarly, duration of aspirin use was not associated with risk (RR for aspirin use of 20 or more years, 0.99, 95% CI 0.69-1.43). In separate models assessing the relation between NSAID use and ovarian cancer risk we found a 40% reduction in risk among NSAID users versus nonusers (RR 0.60, 95% CI 0.38-0.95). However, when we examined this relationship in terms of days of NSAID use per month, we did not observe a dose-response with increasing NSAID use.

CONCLUSIONS

We observed no association between aspirin use, dose, or duration and epithelial ovarian cancer risk. Although we found a modest reduction in risk associated with NSAID use, there was no dose-effect.

Obesity, weight gain, and ovarian cancer

OBJECTIVE

To investigate how adipose tissue alters endogenous hormone levels and may affect events at the ovarian tissue level.

METHODS

We assessed current weight, weight at age 18, and adult weight change in relation to ovarian cancer risk among 109,445 participants in the Nurses' Health Study. Women reported ovarian cancer risk factors and new ovarian cancer diagnoses in biennial mailed questionnaires from 1976 to 1996. Height and weight were queried in 1976, current weight was updated biennially, and weight at age 18 was ascertained in 1980. During 20 years of follow-up and 1,703,474 person-years, 402 cases of epithelial ovarian cancer were confirmed. We used pooled logistic regression to control for age, oral contraceptive use, smoking history, parity, age at menarche, and tubal ligation.

RESULTS

We found no evidence of an association between recent body mass index (BMI, kg/m(2)) and ovarian cancer risk. The multivariable relative risk for women with BMI of 30 kg/m(2) or higher versus BMI less than 21 kg/m(2) was 1.05 (95% confidence interval 0.73, 1.51). For BMI at age 18, there was no association with ovarian cancer risk overall, but a two-fold increase in premenopausal ovarian cancer risk associated with having a BMI at age 18 of 25 kg/m(2) or higher versus BMI less than 20 kg/m(2) (relative risk 2.05, 95% confidence interval 1.07, 3.93, P for trend =.01). Adult weight gain was not associated with ovarian cancer risk.

CONCLUSION

We found no evidence of an association between recent BMI or adult weight change and ovarian cancer risk. Higher BMI in young adulthood was associated with an increased risk of premenopausal ovarian cancer. If confirmed, these findings suggest an additional reason for avoiding adolescent obesity.

Vitamins for chronic disease prevention in adults: scientific review

CONTEXT

Although vitamin deficiency is encountered infrequently in developed countries, inadequate intake of several vitamins is associated with chronic disease.

OBJECTIVE

To review the clinically important vitamins with regard to their biological effects, food sources, deficiency syndromes, potential for toxicity, and relationship to chronic disease.

DATA SOURCES AND STUDY SELECTION

We searched MEDLINE for English-language articles about vitamins in relation to chronic diseases and their references published from 1966 through January 11, 2002.

DATA EXTRACTION

We reviewed articles jointly for the most clinically important information, emphasizing randomized trials where available.

DATA SYNTHESIS

Our review of 9 vitamins showed that elderly people, vegans, alcohol-dependent individuals, and patients with malabsorption are at higher risk of inadequate intake or absorption of several vitamins. Excessive doses of vitamin A during early pregnancy and fat-soluble vitamins taken anytime may result in adverse outcomes. Inadequate folate status is associated with neural tube defect and some cancers. Folate and vitamins B(6) and B(12) are required for homocysteine metabolism and are associated with coronary heart disease risk. Vitamin E and lycopene may decrease the risk of prostate cancer. Vitamin D is associated with decreased occurrence of fractures when taken with calcium.

CONCLUSIONS

Some groups of patients are at higher risk for vitamin deficiency and suboptimal vitamin status. Many physicians may be unaware of common food sources of vitamins or unsure which vitamins they should recommend for their patients. Vitamin excess is possible with supplementation, particularly for fat-soluble vitamins. Inadequate intake of several vitamins has been linked to chronic diseases, including coronary heart disease, cancer, and osteoporosis

Vitamins for chronic disease prevention in adults: clinical applications

Vitamin deficiency syndromes such as scurvy and beriberi are uncommon in Western societies. However, suboptimal intake of some vitamins, above levels causing classic vitamin deficiency, is a risk factor for chronic diseases and common in the general population, especially the elderly. Suboptimal folic acid levels, along with suboptimal levels of vitamins B(6) and B(12), are a risk factor for cardiovascular disease, neural tube defects, and colon and breast cancer; low levels of vitamin D contribute to osteopenia and fractures; and low levels of the antioxidant vitamins (vitamins A, E, and C) may increase risk for several chronic diseases. Most people do not consume an optimal amount of all vitamins by diet alone. Pending strong evidence of effectiveness from randomized trials, it appears prudent for all adults to take vitamin supplements. The evidence base for tailoring the contents of multivitamins to specific characteristics of patients such as age, sex, and physical activity and for testing vitamin levels to guide specific supplementation practices is limited. Physicians should make specific efforts to learn about their patients' use of vitamins to ensure that they are taking vitamins they should, such as folate supplementation for women in the childbearing years, and avoiding dangerous practices such as high doses of vitamin A during pregnancy or massive doses of fat-soluble vitamins at any age.

Use of a computerized risk-appraisal instrument for cancer prevention education of medical students

BACKGROUND

Interactive computer-based tools are increasingly used for patient and medical student instruction.

METHOD

To assess medical students' responses to a computer-based risk-appraisal instrument aimed at teaching concepts of risk and cancer prevention, and to evaluate perceived barriers to using such tools in clinical practice and knowledge about risk factors for colon cancer.

DESIGN

An electronic survey of 133 second-year medical students elicited their responses to the instrument after using it for a case-based learning exercise.

RESULTS

The students identified several potential barriers to use, including inadequate time during the office visit (75% of students) and problems with literacy (48%). They were generally knowledgeable about risk factors for colon cancer. 84% found the instrument "very helpful" or "somewhat helpful" in understanding a case patient's health and setting priorities for behavioral changes.

CONCLUSIONS

Medical students may benefit from exposure to interactive, computer-based tools such as health-appraisal instruments when learning skills in risk counseling and cancer prevention. Second-year students were already concerned about having adequate time during office visits to use such instruments.

Risk of ovarian carcinoma and consumption of vitamins A, C, and E and specific carotenoids: a prospective analysis

BACKGROUND

Antioxidant vitamins may decrease risk of cancer by limiting oxidative DNA damage leading to cancer initiation. Few prospective studies have assessed relations between antioxidant vitamins and ovarian carcinoma.

METHODS

The authors prospectively assessed consumption of vitamins A, C, and E and specific carotenoids, as well as fruit and vegetable intake, in relation to ovarian carcinoma risk among 80,326 participants in the Nurses' Health Study who had no history of cancer other than nonmelanoma skin carcinoma. Women reported on known and suspected ovarian carcinoma risk factors including reproductive factors, smoking, and use of vitamin supplements on biennial mailed questionnaires from 1976 to 1996. Food frequency questionnaires were included in 1980, 1984, 1986, and 1990. The authors confirmed 301 incident cases of invasive epithelial ovarian carcinoma during 16 years of dietary follow-up (1980-1996). Pooled logistic regression was used to control for age, oral contraceptive use, body mass index, smoking history, parity, and tubal ligation.

RESULTS

The authors observed no association between ovarian carcinoma risk and antioxidant vitamin consumption from foods, or foods and supplements together. The multivariate relative risks (95% confidence intervals [CIs]) for ovarian carcinoma among women in the highest versus lowest quintile of intake were 1.04 (95% CI, 0.72-1.51) for vitamin A from foods and supplements; 1.01 (95% CI, 0.69-1.47) for vitamin C; 0.88 (95% CI, 0.61-1.27) for vitamin E; and 1.10 (95% CI, 0.76-1.59) for beta-carotene. Among users of vitamin supplements, the authors found no evidence of an association between dose or duration of any specific vitamin and ovarian carcinoma risk, although the authors had limited power to assess these relations. No specific fruits or vegetables were associated significantly with ovarian carcinoma risk. The authors found no association between ovarian carcinoma and consumption of total fruits or vegetables, or specific subgroups including cruciferous vegetables, green leafy vegetables, legumes, or citrus fruits. Women who consumed at least 2.5 total servings of fruits and vegetables as adolescents had a 46% reduction in ovarian carcinoma risk (relative risk, 0.54, 95% CI, 0.29-1.03; P value for trend 0.04).

CONCLUSIONS

These data do not support an important relation between consumption of antioxidant vitamins from foods or supplements, or intake of fruits and vegetables, and incidence of ovarian carcinoma in this cohort. However, modest associations cannot be excluded, and the authors' finding of an inverse association for total fruit and vegetable intake during adolescence raises the possibility that the pertinent exposure period may be much earlier than formerly anticipated.

Brief communication: detecting depression: providing high quality primary care for HIV-infected patients

Depression is common among HIV-infected patients, but little is known about risk factors for depression in this population. Several studies before protease inhibitors became available have reported inconsistent associations between depression and disease severity. Delivering high quality HIV care includes adequate detection and treatment of depression. The objective of this study was to describe the prevalence and correlates of depression among a contemporary group of HIV-infected patients. The setting and design for the study was a chart abstraction for HIV-infected patients in a primary care practice in Boston, Mass, in June 1997. Among 275 HIV-infected patients, depression was documented in 147 patient charts (53%), half of whom (n = 73, 27%) also received antidepressant medications. We used multivariable logistic regression to identify risk factors for depression among patients with both a chart diagnosis of depression and current antidepressant medication use. We observed increased risk of depression among patients with a history of substance use (odds ratio 2.7, 95% confidence interval 1.5-4.7), recent medical hospitalization (2.6, 1.4-5.0), and homosexual risk behavior (2.1, 1.1-4.2). Depression remains a common problem for HIV-infected patients, particularly among those with history of substance abuse, medical hospitalization, or homosexual risk behavior. Routine screening for depression in this population with special attention to those at higher risk may offer opportunities for earlier diagnosis and treatment.

Impact of relational coordination on quality of care, postoperative pain and functioning, and length of stay: a nine-hospital study of surgical patients

BACKGROUND

Health care organizations face pressures from patients to improve the quality of care and clinical outcomes, as well as pressures from managed care to do so more efficiently. Coordination, the management of task interdependencies, is one way that health care organizations have attempted to meet these conflicting demands.

OBJECTIVES

The objectives of this study were to introduce the concept of relational coordination and to determine its impact on the quality of care, postoperative pain and functioning, and the length of stay for patients undergoing an elective surgical procedure. Relational coordination comprises frequent, timely, accurate communication, as well as problem-solving, shared goals, shared knowledge, and mutual respect among health care providers.

RESEARCH DESIGN

Relational coordination was measured by a cross-sectional questionnaire of health care providers. Quality of care was measured by a cross-sectional postoperative questionnaire of total hip and knee arthroplasty patients. On the same questionnaire, postoperative pain and functioning were measured by the WOMAC osteoarthritis instrument. Length of stay was measured from individual patient hospital records.

SUBJECTS

The subjects for this study were 338 care providers and 878 patients who completed questionnaires from 9 hospitals in Boston, MA, New York, NY, and Dallas, TX, between July and December 1997.

MEASURES

Quality of care, postoperative pain and functioning, and length of acute hospital stay.

RESULTS

Relational coordination varied significantly between sites, ranging from 3.86 to 4.22 (P <0.001). Quality of care was significantly improved by relational coordination (P <0.001) and each of its dimensions. Postoperative pain was significantly reduced by relational coordination (P = 0.041), whereas postoperative functioning was significantly improved by several dimensions of relational coordination, including the frequency of communication (P = 0.044), the strength of shared goals (P = 0.035), and the degree of mutual respect (P = 0.030) among care providers. Length of stay was significantly shortened (53.77%, P <0.001) by relational coordination and each of its dimensions.

CONCLUSIONS

Relational coordination across health care providers is associated with improved quality of care, reduced postoperative pain, and decreased lengths of hospital stay for patients undergoing total joint arthroplasty. These findings support the design of formal practices to strengthen communication and relationships among key caregivers on surgical units.

Delays in protease inhibitor use in clinical practice

OBJECTIVE

To determine the clinical factors associated with delayed protease inhibitor initiation.

DESIGN

Chart review and telephone survey.

SETTING

General medicine practice at an academic medical center in Boston, Mass.

PATIENTS

One hundred ninety patients living with HIV and a viral load of more than 10,000 copies/ml.

MEASUREMENTS AND MAIN RESULTS

The main outcome measurement was time to first protease inhibitor prescription after first elevated HIV viral load (>10,000 copies/ml). In this cohort, 190 patients had an elevated viral load (median age 39; 87% male; 12% history of injection drug use; 63% AIDS; 53% with depression; 17% history of pneumocystis pneumonia; 54% CD4 <200). In Cox proportional hazards modeling, significant univariate correlates for delayed protease inhibitor initiation were higher CD4 cell count (hazard ratio [HR] 2. 38 for CD4 200-500 compared with <200, 95% confidence interval [CI] 1.59, 3.57; and HR 8.33 for CD4> 500; 95% CI 2.63, 25.0), higher viral load (HR 0.43 for each 10-fold increase; 95% CI 0.31, 0.59), injection drug use (HR 2.08; 95% CI 1.05, 4.17), AIDS (HR 0.24; 95% CI 0.15, 0.36), and history of pneumocystis pneumonia (HR 0.32; 95% CI 0.21, 0.49). In multivariate models adjusted for secular trends in protease inhibitor use, factors significantly associated with delay of protease inhibitor initiation (p <.05) were higher CD4 cell count (for CD4 200-500, HR 2.63; 95% CI 1.61, 4.17; for CD4> 500, HR 11.11; 95% CI 3.57, 33.33), higher viral load (HR 0.66 for each 10-fold increase; 95% CI 0.45, 0.98), history of pneumocystis pneumonia (HR 0.57; 95% CI 0.37, 0.90), history of depression (HR 1. 49; 95% CI 1.03, 2.13), and history of injection drug use (HR 2.70; 95% CI 1.35, 5.56).

CONCLUSIONS

HIV-infected patients with higher CD4 cell counts or a history of depression or history of injection drug use have significant and lengthy delays of protease inhibitor therapy. Although some delays may be clinically appropriate, enhancement of provider and patient education might prove beneficial. Further research should examine reasons for delays in protease inhibitor initiation and their appropriateness.

Patterns of use, expenditures, and perceived efficacy of complementary and alternative therapies in HIV-infected patients

BACKGROUND

Complementary and alternative medicine (CAM) use is common in the general population, accounting for substantial expenditures. Among patients with human immunodeficiency virus (HIV) infection, few data are available on the prevalence, costs, and patterns of alternative therapy use.

METHODS

We carried out detailed telephone surveys and medical chart reviews for 289 active patients with HIV in a general medicine practice at a university-based teaching hospital in Boston, Mass. Data were collected on prevalence and patterns of CAM use, out-of-pocket expenditures, associated outcomes, and correlates of CAM use.

RESULTS

Of 180 patients who agreed to be interviewed, 122 (67.8%) used herbs, vitamins, or dietary supplements, 81 (45.0%) visited a CAM provider, and 43 (23.9%) reported using marijuana for medicinal purposes in the previous year. Patients who saw CAM providers made a median of 12 visits per year to these providers compared with 7 visits per year to their primary care physician and nurse practitioner. Mean yearly out-of-pocket expenditures for CAM users totaled $938 for all therapies. For the main reason CAM was used, respondents found therapies "extremely" or "quite a bit" helpful in 81 (81.0%) of 100 reports of supplement use, in 76 (65.5%) of 116 reports of CAM provider use, and in 27 (87%) of 31 reports of marijuana use. In multivariable models, college education (odds ratio [OR]=3.7, 95% confidence interval [CI]=1.9-7.1) and fatigue (OR=2.7, 95% CI=1.4-5.2) were associated with CAM provider use; memory loss (OR=2.3, 95% CI=1.1-4.8) and fatigue (OR=0.4, 95% CI=0.2-0.9) were associated with supplement use; and weight loss (OR=2.6, 95% CI=1.2-5.6) was associated with marijuana use.

CONCLUSIONS

Patients with HIV infection use CAM, including marijuana, at a high rate; make frequent visits to CAM providers; incur substantial expenditures; and report considerable improvement with these treatments. Clinical trials of frequently used CAMs are needed to inform physicians and patients about therapies that may have measurable benefit or measurable risk.