Assessing trends and variability in outpatient dual testing for chronic kidney disease with urine albumin and serum creatinine, 2009-2018: a retrospective cohort study in the Veterans Health Administration System.

BACKGROUND: Simultaneous urine testing for albumin (UAlb) and serum creatinine (SCr), that is, 'dual testing,' is an accepted quality measure in the management of diabetes. As chronic kidney disease (CKD) is defined by both UAlb and SCr testing, this approach could be more widely adopted in kidney care. OBJECTIVE: We assessed time trends and facility-level variation in the performance of outpatient dual testing in the integrated Veterans Health Administration (VHA) system. DESIGN, SUBJECTS AND MAIN MEASURES: This retrospective cohort study included patients with any inpatient or outpatient visit to the VHA system during the period 2009-2018. Dual testing was defined as UAlb and SCr testing in the outpatient setting within a calendar year. We assessed time trends in dual testing by demographics, comorbidities, high-risk (eg, diabetes) specialty care and facilities. A generalised linear mixed-effects model was applied to explore individual and facility-level predictors of receiving dual testing. KEY RESULTS: We analysed data from approximately 6.9 million veterans per year. Dual testing increased, on average, from 17.4% to 21.2%, but varied substantially among VHA centres (0.3%-43.7% in 2018). Dual testing was strongly associated with diabetes (OR 10.4, 95% CI 10.3 to 10.5, p<0.0001) and not associated with VHA centre complexity level. However, among patients with high-risk conditions including diabetes, <50% received dual testing in any given year. As compared with white veterans, black veterans were less likely to be tested after adjusting for other individual and facility characteristics (OR 0.93, 95% CI 0.92 to 0.93, p<0.0001). CONCLUSIONS: Dual testing for CKD in high-risk specialties is increasing but remains low. This appears primarily due to low rates of testing for albuminuria. Promoting dual testing in high-risk patients will help to improve disease management and patient outcomes.

In-Hospital and 1-Year Mortality Trends in a National Cohort of US Veterans with Acute Kidney Injury.

BACKGROUND AND OBJECTIVES: AKI, a frequent complication among hospitalized patients, confers excess short- and long-term mortality. We sought to determine trends in in-hospital and 1-year mortality associated with AKI as defined by Kidney Disease Improving Global Outcomes consensus criteria. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: This retrospective cohort study used data from the national Veterans Health Administration on all patients hospitalized from October 1, 2008 to September 31, 2017. AKI was defined by Kidney Disease Improving Global Outcomes serum creatinine criteria. In-hospital and 1-year mortality trends were analyzed in patients with and without AKI using Cox regression with year as a continuous variable. RESULTS: We identified 1,688,457 patients and 2,689,093 hospitalizations across the study period. Among patients with AKI, 6% died in hospital, and 28% died within 1 year. In contrast, in-hospital and 1-year mortality rates were 0.8% and 14%, respectively, among non-AKI hospitalizations. During the study period, there was a slight decline in crude in-hospital AKI-associated mortality (hazard ratio, 0.98 per year; 95% confidence interval, 0.98 to 0.99) that was attenuated after accounting for patient demographics, comorbid conditions, and acute hospitalization characteristics (adjusted hazard ratio, 0.99 per year; 95% confidence interval, 0.99 to 1.00). This stable temporal trend in mortality persisted at 1 year (adjusted hazard ratio, 1.00 per year; 95% confidence interval, 0.99 to 1.00). CONCLUSIONS: AKI associated mortality remains high, as greater than one in four patients with AKI died within 1 year of hospitalization. Over the past decade, there seems to have been no significant progress toward improving in-hospital or long-term AKI survivorship.

Generalizability of an acute kidney injury prediction model across health systems.

Delays in the identification of acute kidney injury (AKI) in hospitalized patients are a major barrier to the development of effective interventions to treat AKI. A recent study by Tomasev and colleagues at DeepMind described a model that achieved a state-of-the-art performance in predicting AKI up to 48 hours in advance.1 Because this model was trained in a population of US Veterans that was 94% male, questions have arisen about its reproducibility and generalizability. In this study, we aimed to reproduce key aspects of this model, trained and evaluated it in a similar population of US Veterans, and evaluated its generalizability in a large academic hospital setting. We found that the model performed worse in predicting AKI in females in both populations, with miscalibration in lower stages of AKI and worse discrimination (a lower area under the curve) in higher stages of AKI. We demonstrate that while this discrepancy in performance can be largely corrected in non-Veterans by updating the original model using data from a sex-balanced academic hospital cohort, the worse model performance persists in Veterans. Our study sheds light on the importance of reproducing artificial intelligence studies, and on the complexity of discrepancies in model performance in subgroups that cannot be explained simply on the basis of sample size.

Predictors of kidney function recovery among incident ESRD patients.

BACKGROUND: ESRD is considered an irreversible loss of renal function, yet some patients will recover kidney function sufficiently to come off dialysis. Potentially modifiable predictors of kidney recovery, such as dialysis prescription, have not been fully examined. METHODS: Retrospective cohort study using United States Renal Data System (USRDS) data to identify incident hemodialysis (HD) patients between 2012 and 2016, the first 4 years for which dialysis treatment data is available. The primary outcome was kidney recovery within 1 year of ESRD and HD initiation, defined by a specific recovery code and survival off dialysis for at least 30 days. Patient and treatment characteristics were compared between those that recovered versus those that remained dialysis-dependent. A time-dependent survival model was used to identify independent predictors of kidney recovery. RESULTS: During the study period, there were 372,387 incident HD patients with available data, among whom 16,930 (4.5%) recovered to dialysis-independence. Compared to non-recovery, a higher proportion of patients with kidney recovery were of white race, and non-Hispanic ethnicity. Both groups had a similar age distribution. Patients with an acute kidney injury diagnosis as primary cause of ESRD were most likely to recover, but the most common ESRD diagnosis among recovering patients was type 2 diabetes (29.8% of recovery cases). Higher eGFR and lower albumin at ESRD initiation were associated with increased likelihood of recovery. When examining HD ultrafiltration rate (UFR), each quintile above the first quintile was associated with a progressively lower likelihood of recovery (HR 0.45, 95% CI 0.43-0.48 for highest versus lowest quintile, p < 0.001). CONCLUSIONS: We identified non-modifiable and potentially modifiable factors associated with kidney recovery which may assist clinicians in counseling and monitoring incident ESRD patients with a greater chance to gain dialysis-independence. Clinical trials are warranted to examine the impact of dialysis prescription on subsequent kidney function recovery.

Burden and Cost of Caring for US Veterans With CKD: Initial Findings From the VA Renal Information System (VA-REINS).

Kidney disease is a common, complex, costly, and life-limiting condition. Most kidney disease registries or information systems have been limited to single institutions or regions. A national US Department of Veterans Affairs (VA) Renal Information System (VA-REINS) was recently developed. We describe its creation and present key initial findings related to chronic kidney disease (CKD) without kidney replacement therapy (KRT). Data from the VA's Corporate Data Warehouse were processed and linked with national Medicare data for patients with CKD receiving KRT. Operational definitions for VA user, CKD, acute kidney injury, and kidney failure were developed. Among 7 million VA users in fiscal year 2014, CKD was identified using either a strict or liberal operational definition in 1.1 million (16.4%) and 2.5 million (36.3%) veterans, respectively. Most were identified using an estimated glomerular filtration rate laboratory phenotype, some through proteinuria assessment, and very few through International Classification of Diseases, Ninth Revision coding. The VA spent ∼$18 billion for the care of patients with CKD without KRT, most of which was for CKD stage 3, with higher per-patient costs by CKD stage. VA-REINS can be leveraged for disease surveillance, population health management, and improving the quality and value of care, thereby enhancing VA's capacity as a patient-centered learning health system for US veterans.

Assessment of Prescription Analgesic Use in Older Adults With and Without Chronic Kidney Disease and Outcomes.

Importance: Pain is a common symptom among patients with kidney disease. However, little is known about use of analgesics among patients aged 65 years or older with chronic kidney disease (CKD) who do not receive dialysis treatment. Objective: To assess national trends and geographic variations in use of opioids and prescription nonsteroidal anti-inflammatory drugs (NSAIDs) in older adults with and without CKD in the US (2006-2015) and examine associations between use of opioids and patient outcomes. Design, Setting, and Participants: This cohort study used the 5% Medicare claims data (2005-2015) to select 10 retrospective annual cohorts of Medicare Part D beneficiaries aged 65 years and older from 2006 to 2015 and a retrospective longitudinal cohort. Data were analyzed in August 2019. Exposures: CKD status and other comorbidities identified using International Classification of Diseases, Ninth Revision, Clinical Modification codes. Main Outcomes and Measures: Analgesic use was measured by overall use (proportion of ever used opioids/NSAIDs), long-term use (prescribed >90 days), and cumulative use (total annual days' supply). Patient outcomes included progression to end-stage kidney disease (ESKD) and all-cause mortality. Results: A total of 6 260 454 beneficiaries (9.6% identified with CKD by claims) were selected in the annual cohorts and 649 339 beneficiaries (8.3% identified with CKD) were selected in the longitudinal cohort. There was significant growth in opioid use (31.2%-42.4%) and NSAID use (10.7%-16.6%) among patients aged 65 years and older with CKD from 2006 to 2015. Long-term use of opioids increased during 2006 to 2014 (25.8%-36.7%) but decreased through 2015 at 35.6%, while long-term use of NSAIDs remained stable. Opioid use was higher in patients with CKD, particularly CKD stages 4 to 5 (odds ratio [OR], 1.35; 95% CI, 1.33-1.37; P < .001) compared with non-CKD. NSAID use was lower in patients with CKD stages 4 to 5 (OR, 0.55; 95% CI, 0.54-0.56; P < .001). Substantial geographic variations in analgesic use were observed across states (opioid use in CKD: 24.7%-54.3%; NSAID use in CKD: 11.2%-20.8%, 2012-2015). Opioid use was associated with progression to ESKD (hazard ratio [HR], 1.10; 95% CI, 1.04-1.16; P = .001) and death (HR, 1.19; 95% CI, 1.18-1.20; P < .001) independent of CKD status and other covariates. There was an inverse association between NSAID use and death (HR, 0.84; 95% CI, 0.83-0.85; P < .001). Conclusions and Relevance: Among Medicare patients with CKD, use of prescription analgesics, both opioid and NSAID, increased from 2006 to 2015. Optimizing pain management in a complex condition such as kidney disease should remain a priority for clinicians and researchers alike.

Incidence of ESKD Among Native Hawaiians and Pacific Islanders Living in the 50 US States and Pacific Island Territories.

RATIONALE & OBJECTIVE: Native Hawaiians and Pacific Islanders (NHPI) have been reported to have the highest rates of incident end-stage kidney disease (ESKD) compared with other races in the United States. However, these estimates were likely biased upward due to the exclusion of nearly half the NHPI population that reports multiple races in the US Census. We sought to estimate the incidence rate of ESKD, including individuals reporting multiple races, and describe the clinical characteristics of incident cases by race and location. STUDY DESIGN: Health care database study. SETTING & PARTICIPANTS: US residents of the 50 states and 3 Pacific Island territories of the United States whose ESKD was recorded in the US Renal Data System (USRDS) between 2007 and 2016, as well as US residents recorded in the 2010 Census. PREDICTORS: Age, sex, race, body mass index, primary cause of ESKD, comorbid conditions, estimated glomerular filtration rate, pre-ESKD nephrology care, and hemoglobin A1c level among ESKD cases. OUTCOME: Initiation of maintenance dialysis or transplantation for kidney failure. ANALYTICAL APPROACH: Crude ESKD incidence rates (cases/person-years) were estimated using both single- and multiple-race reporting. RESULTS: Even after inclusion of multirace reporting, NHPI had the highest ESKD incidence rate among all races in the 50 states (921 [95% CI, 904-938] per million population per year)-2.7 times greater than whites and 1.2 times greater than blacks. Also using multirace reporting, the NHPI ESKD incident rate in the US territories was 941 (95% CI, 895-987) per million population per year. Diabetes was listed as the primary cause of ESKD most frequently for NHPI and American Indians/Alaska Natives. Sensitivity analysis adjusting for age and sex demonstrated greater differences in rates between NHPI and other races. Diabetes was the primary cause of ESKD in 60% of incident NHPI cases. Patients with ESKD living in the territories had received less pre-ESKD nephrology care than had patients living in the 50 states. LIMITATIONS: Different methods of race classification in the USRDS versus the US Census. CONCLUSIONS: NHPI living in the 50 US states and Pacific territories had the highest rates of ESKD incidence compared with other races. Further research and efforts are required to understand the reasons for and define how best to address this racial disparity.

Supply and Distribution of Vascular Access Physicians in the United States: A Cross-Sectional Study.

BACKGROUND: Because functioning permanent vascular access (arteriovenous fistula [AVF] or arteriovenous graft [AVG]) is crucial for optimizing patient outcomes for those on hemodialysis, the supply of physicians placing vascular access is key. We investigated whether area-level demographic and healthcare market attributes were associated with the distribution and supply of AVF/AVG access physicians in the United States. METHODS: A nationwide registry of physicians placing AVFs/AVGs in 2015 was created using data from the United States Renal Data System and the American Physician Association's Physician Masterfile. We linked the registry information to the Area Health Resource File to assess the supply of AVF/AVG access physicians and their professional attributes by hospital referral region (HRR). Bivariate analysis and Poisson regression were performed to examine the relationship between AVF/AVG access physician supply and demographic, socioeconomic, and health resource conditions of HRRs. The setting included all 50 states. The main outcome was supply of AVF/AVG access physicians, defined as the number of physicians performing AVF and/or AVG placement per 1000 prevalent patients with ESKD. RESULTS: The majority of vascular access physicians were aged 45-64 (average age, 51.6), male (91%), trained in the United States (76%), and registered in a surgical specialty (74%). The supply of physicians varied substantially across HRRs. The supply was higher in HRRs with a higher percentage white population (ß=0.44; SEM=0.14; P=0.002), lower unemployment rates (ß=-10.74; SEM=3.41; P=0.002), and greater supply of primary care physicians (ß=0.18; SEM=0.05; P=0.001) and nephrologists (ß=15.89; SEM=1.22; P<0.001). CONCLUSIONS: Geographic variation was observed in the supply of vascular access physicians. Higher supply of such specialist physicians in socially and economically advantaged areas may explain disparities in vascular access and outcomes in the United States and should be the subject of further study and improvement.

Acute Kidney Injury Recovery Pattern and Subsequent Risk of CKD: An Analysis of Veterans Health Administration Data.

BACKGROUND: Studies suggest an association between acute kidney injury (AKI) and long-term risk for chronic kidney disease (CKD), even following apparent renal recovery. Whether the pattern of renal recovery predicts kidney risk following AKI is unknown. STUDY DESIGN: Retrospective cohort. SETTING & PARTICIPANTS: Patients in the Veterans Health Administration in 2011 hospitalized (> 24 hours) with at least 2 inpatient serum creatinine measurements, baseline estimated glomerular filtration rate > 60 mL/min/1.73 m², and no diagnosis of end-stage renal disease or non-dialysis-dependent CKD: 17,049 (16.3%) with and 87,715 without AKI. PREDICTOR: Pattern of recovery to creatinine level within 0.3 mg/dL of baseline after AKI: within 2 days (fast), in 3 to 10 days (intermediate), and no recovery by 10 days (slow or unknown). OUTCOME: CKD stage 3 or higher, defined as 2 outpatient estimated glomerular filtration rates < 60 mL/min/1.73m² at least 90 days apart or CKD diagnosis, dialysis therapy, or transplantation. MEASUREMENTS: Risk for CKD was modeled using modified Poisson regression and time to death-censored CKD was modeled using Cox proportional hazards regression, both stratified by AKI stage. RESULTS: Most patients' AKI episodes were stage 1 (91%) and 71% recovered within 2 days. At 1 year, 18.2% had developed CKD (AKI, 31.8%; non-AKI, 15.5%; P < 0.001). In stage 1, the adjusted relative risk ratios for CKD stage 3 or higher were 1.43 (95% CI, 1.39-1.48), 2.00 (95% CI, 1.88-2.12), and 2.65 (95% CI, 2.51-2.80) for fast, intermediate, and slow/unknown recovery. A similar pattern was observed in subgroup analyses incorporating albuminuria and sensitivity analysis of death-censored time to CKD. LIMITATIONS: Variable timing of follow-up and mostly male veteran cohort may limit generalizability. CONCLUSIONS: Patients who develop AKI during a hospitalization are at substantial risk for the development of CKD by 1 year following hospitalization and timing of AKI recovery is a strong predictor, even for the mildest forms of AKI.

Race and other risk factors for incident proteinuria in a national cohort of HIV-infected veterans.

BACKGROUND: Proteinuria in human immunodeficiency virus (HIV)-infected individuals has been associated with poorer outcomes. We examined risk factors associated with the development of proteinuria in a national registry of HIV-infected veterans. METHODS: A total of 21,129 HIV-infected veterans of black and white race without preexisting kidney disease were receiving health care in the Veterans' Health Administration (VHA) medical system between 1997 and 2011. Using the VHA electronic record system, we identified kidney-related risk factors (hypertension, diabetes, and cardiovascular disease) and HIV-related risk factors (CD4 lymphocyte count, HIV RNA level, hepatitis C virus, and hepatitis B virus) for developing proteinuria. Proteinuria was defined by 2 consecutive dipstick measures of 1 or higher. The Fine-Gray competing risk model was used to estimate association between clinical variables and incident proteinuria, while accounting for intervening mortality events. RESULTS: During follow-up (median = 5.3 years), 7031 patients developed proteinuria. Overall, black race compared with white race was associated with a higher risk of proteinuria {hazard ratio [95% confidence interval (CI)] = 1.51 [1.43 to 1.59]}, but the association was stronger at younger ages (P interaction <0.001). Age-stratified risk of proteinuria for blacks relative to whites was greatest among veterans <30 years [2.19 (1.66 to 2.89)] and the risk diminished with increasing age [1.14 (0.97 to 1.34) for >60 years]. We found the race difference to be stronger for the outcome of 2 or higher proteinuria [2.13 (1.89 to 2.39)]. Both HIV-related and traditional risk factors were also associated with incident proteinuria (P < 0.05). CONCLUSIONS: Compared with whites, risk of proteinuria was higher in black veterans with HIV infection, particularly at younger ages. In both races, HIV- and kidney-related risk factors were associated with higher proteinuria risk.

Comparison of the long-term outcomes of kidney transplantation: USA versus Spain.

BACKGROUND: The long-term outcomes of kidney transplantation are suboptimal because many patients lose their allografts or experience premature death. Cross-country comparisons of long-term outcomes of kidney transplantation may provide insight into factors contributing to premature graft failure and death. We evaluated the rates of late graft failure and death among US and Spanish kidney recipients. METHODS: This is a cohort study of US (n = 9609) and Spanish (n = 3808) patients who received a deceased donor kidney transplant in 1990, 1994, 1998 or 2002 and had a functioning allograft 1 year after transplantation with follow-up through September 2006. Ten-year overall and death-censored graft survival and 10-year overall recipient survival and death with graft function (DWGF) were estimated with multivariate Cox models. RESULTS: Among recipients alive with graft function 1 year after transplant, the 10-year graft survival was 71.3% for Spanish and 53.4% for US recipients (P < 0.001). The 10-year, death-censored graft survival was 75.6 and 76.0% for Spanish and US recipients, respectively (P = 0.73). The 10-year recipient survival was 86.2% for Spanish and 67.4% for US recipients (P < 0.001). In recipients with diabetes as the cause of ESRD, the adjusted DWGF rates at 10 years were 23.9 and 53.8 per 1000 person-years for Spanish and US recipients, respectively (P < 0.001). Among recipients whose cause of ESRD was not diabetes mellitus, the adjusted 10-year DWGF rates were 11.0 and 25.4 per 1000 person-years for Spanish and US recipients, respectively. CONCLUSIONS: US kidney transplant recipients had more than twice the long-term hazard of DWGF compared with Spanish kidney transplant recipients and similar levels of death-censored graft function. Pre-transplant medical care, comorbidities, such as cardiovascular disease, and their management in each country's health system are possible explanations for the differences between the two countries.

New-onset diabetes mellitus in kidney transplant recipients discharged on steroid-free immunosuppression.

BACKGROUND: New-onset diabetes after transplant (NODAT) is a serious complication after kidney transplantation. We studied the relationship between steroid-free maintenance regimens and NODAT in a national cohort of adult kidney transplant patients. METHODS: A total of 25,837 previously nondiabetic kidney transplant patients, engrafted between January 1, 2004, and December 31, 2006, were included in the study. Logistic regression analysis was used to compare the risk of developing NODAT within 3 years after transplant for patients discharged with and without steroid-containing maintenance immunosuppression regimens. The effect of transplant program-level practice regarding steroid-free regimens on the risk of NODAT was studied as well. RESULTS: The cumulative incidence of NODAT within 3 years of transplant was 16.2% overall; 17.7% with maintenance steroids and 12.3% without (P<0.001). Patients discharged with steroids had 42% greater odds of developing NODAT compared with those without steroids (adjusted odds ratio [AOR]=1.42, 95% confidence interval [CI]=1.27-1.58, P<0.001). The maintenance regimen of tacrolimus and mycophenolate mofetil or mycophenolate sodium was associated with 25% greater odds of developing NODAT (AOR=1.25, 95% CI=1.08-1.45, P=0.003) than the regimen of cyclosporine and mycophenolate mofetil or mycophenolate sodium. Several induction therapies also were associated with lower odds of NODAT compared with no induction. Patients from programs that used steroid-free regimens for a majority of their patients had reduced odds of NODAT compared with patients from programs discharging almost all of their patients on steroid-containing regimens. CONCLUSION: The adoption of steroid-free maintenance immunosuppression at discharge from kidney transplantation in selected patients was associated with reduced odds of developing NODAT within 3 years.

Steroid-free maintenance immunosuppression in kidney transplantation: is it time to consider it as a standard therapy?

Steroid-free immunosuppression in kidney transplantation has been gaining popularity over the past decade, as documented by a continuous and steady rise in the number of kidney transplant patients discharged on steroid-free regimens. This increased interest in steroid-free immunosuppression is fueled by the recognition that half of transplant loss is related to patient death due to cardiovascular disease and/or infectious complications and that the long-term use of steroids contributes to such elevated cardiovascular morbidity and mortality. The availability of newer and more potent immunosuppressive agents has furthered such interest. Many clinical trials over the past two decades have demonstrated the feasibility of steroid-free regimens, at the expense of a slight increase in the rate of acute rejection, which is an important end point in any clinical trial of relatively short duration. The largest epidemiological study to date has reassured the transplant community that the selective use of steroid-free immunosuppression in kidney transplant patients provides no inferior outcome in patient and graft survival at intermediate term. Steroid-free regimens have the potential to improve cardiovascular risk profile. The challenges that remain are to identify the subset of kidney transplant patients who may not benefit from steroid-free immunosuppression and to demonstrate the survival advantage of steroid-free immunosuppresion in suitable kidney transplant candidates.

Epidemiology of regular prescribed opioid use: results from a national, population-based survey.

Chronic pain occurs commonly and accounts for significant suffering and costs. Although use of opioids for treatment of chronic pain is increasing, little is known about patients who use opioids regularly. We report data from the second wave of the Healthcare for Communities survey (2000-2001), a large, nationally representative household survey. We compared regular users of prescription opioids to nonusers of opioids and calculated the percentage of individuals within a given demographic or disease state that reported chronic opioid use. Approximately 2% of the 7,909 survey respondents reported use of opioid medications for at least a month, which the Healthcare for Communities survey defined as "regular use." Opioid users were more likely than nonusers to report high levels of pain interference with their daily lives and to rate their health as fair or poor. Arthritis and back pain were the most prevalent chronic, physical health conditions among users of opioids, with 63% of regular users of opioids reporting arthritis and 59% reporting back pain. The majority of regular users of opioids had multiple pain conditions (mean=1.9 pain conditions). Regular opioid users appear to have an overall lower level of health status and to have multiple, chronic physical health disorders.

Do users of regularly prescribed opioids have higher rates of substance use problems than nonusers?

OBJECTIVE: To determine whether individuals who use prescribed opioids for chronic noncancer pain have higher rates of any opioid misuse, any problem opioid misuse, nonopioid illicit drug use, nonopioid problem drug use, or any problem alcohol use, compared with those who do not use prescribed opioids. METHODS: Respondents were from a nationally representative survey (N = 9,279), which contained measures of regular use of prescribed opioids, substance use problems, mental health disorders, physical health, pain, and sociodemographics. RESULTS: In unadjusted models, compared with nonusers of prescription opioids, users of prescription opioids had significantly higher rates of any opioid misuse (odds ratio [OR] = 5.48, P < 0.001), problem opioid misuse (OR = 14.76, P < 0.001), nonopioid illicit drug use (OR = 1.73, P < 0.01), nonopioid problem drug use (OR = 4.48, P < 0.001), and problem alcohol use (OR = 1.89, P = 0.04). In adjusted models, users of prescribed opioids had significantly higher rates of any opioid misuse (OR = 3.07, P < 0.001) and problem opioid misuse (OR = 6.11, P < 0.001) but did not have significantly higher rates of the other outcomes. CONCLUSIONS: Users of prescribed opioids had higher rates of opioid and nonopioid abuse problems compared with nonusers of prescribed opioids, but these higher rates appear to be partially mediated by depressive and anxiety disorders. It is not possible to assign causal priority based on our cross-sectional data, but our findings are more compatible with mental disorders leading to substance abuse among prescription opioid users than prescription opioids themselves prompting substance abuse iatrogenically. In patients receiving prescribed opioids, clinicians need to be alert to drug abuse problems and potentially mediating mental health disorders.

Risk factors for clinically recognized opioid abuse and dependence among veterans using opioids for chronic non-cancer pain.

A central question in prescribing opioids for chronic non-cancer pain (CNCP) is how to best balance the risk of opioid abuse and dependence with the benefits of pain relief. To achieve this balance, clinicians need an understanding of the risk factors for opioid abuse, an issue that is only partially understood. We conducted a secondary data analysis of regional VA longitudinal administrative data (years 2000-2005) for chronic users of opioids for CNCP (n=15,160) to investigate risk factors for the development of clinically recognized (i.e., diagnosed) opioid abuse or dependence among these individuals. We analyzed four broad groups of possible risk factors: (i) non-opioid substance abuse disorders, (ii) painful physical health disorders, (iii) mental health disorders, and (iv) socio-demographic factors. In adjusted models, a diagnosis of non-opioid substance abuse was the strongest predictor of opioid abuse/dependence (OR=2.34, p<0.001). Mental health disorders were moderately strong predictors (OR=1.46, p=0.005) of opioid abuse/dependence. However, the prevalence of mental health disorders was much higher than the prevalence of non-opioid substance abuse disorders (45.3% vs. 7.6%) among users of opioids for CNCP, suggesting that mental health disorders account for more of the population attributable risk for opioid abuse than does non-opioid substance abuse. Males, younger adults, and individuals with greater days supply of prescription opioids dispensed in 2002 were more likely to develop opioid abuse/dependence. Clinicians need to carefully screen for substance abuse and mental health disorders in candidates for opioid therapy and facilitate appropriate treatment of these disorders.