Opioid initiation, continuity, and more

Summaries from ACP Hospitalist Weekly.


New opioid use as an inpatient associated with continued use

Receiving opioids for the first time while in the hospital, particularly right before discharge, was associated with long-term use of the drugs, a recent study found.

The retrospective cohort study used 2010 to 2014 electronic health record data from 12 community and academic hospitals in a single Pennsylvania health care system. It included 148,068 opioid-naive patients (191,249 admissions) with at least one outpatient encounter in the years before and after admission. Opioids were administered in 48% of the admissions. Patients who got opioids took them for a mean of 67.9% of their stay, with higher rates for surgical admissions. Whether patients first received opioids in the ED, ward, or ICU varied widely, as did use of nonopioid analgesics (which ranged from 7.9% to 22.2%, depending on admission type). Results were published by Annals of Internal Medicine on June 18 and appeared in the July 16 issue.

After adjustment, the study found that significantly more patients who received opioids in the hospital were taking them at 90 days after discharge compared to those who didn't receive them (5.9% vs. 3.0%; difference, 3.0 percentage points [95% CI, 2.8 to 3.2]). Receiving opioids less than 12 hours before discharge was also associated with 90-day use (7.5% vs. 3.9% of patients who were opioid-free 24 hours before discharge; difference, 3.6 percentage points [95% CI, 3.3 to 3.9]). The proportion of a hospital stay that included opioids was only modestly associated with later use. Similar associations were found when the researchers looked at opioid use a year after discharge.

The study has several key findings, according to the authors. It shows that opioid prescribing is common in medical admissions and rarely preceded by nonopioid analgesics; that any opioid use in the hospital, and the timing and duration of that use, is associated with postdischarge use; and that many patients receive opioids for a significant proportion of hospitalization, including during the last 12 hours. The authors called for research on changing inpatient prescribing practices, for example, by restricting the use of opioids before other analgesics or immediately before discharge or implementing protocols to guide pain assessment and treatment. The authors cautioned that the study's limitations included use of data from a single health system and lack of information on the appropriateness of the opioid prescriptions.

An accompanying editorial also supported the use of protocols to reduce unnecessary inpatient opioid prescriptions. “If opioids are to be used, prescribing them at the lowest dose for as few days as possible may reduce long-term use. Behavioral nudges, such as setting a lower default number of tablets in the electronic medical record, may help achieve this goal,” the editorial said. However, the editorialists also cautioned that opioid stewardship alone is not going to resolve the current opioid crisis. “Efforts to deliver harm reduction services and treatment for opioid use disorder are also urgently needed,” they wrote.

Physician continuity associated with speedier discharge

Physician continuity may influence hospital length of stay, according to a recent study.

The study included all patients admitted in 2015 to the general medicine service at a tertiary care teaching hospital in Canada. Attending inpatient physician continuity was measured as the consecutive number of days each patient was treated by the same physician. The study used generalized estimating equation methods to model the adjusted association of attending continuity with daily discharge probability. Results were published online on June 13 by the Journal of General Internal Medicine and appeared in the September issue.

Overall, the study assessed 6,301 admissions involving 41 internists, 5,134 patients (mean age, 68.0 years; 50.7% women), and 38,242 patient-days. Attending coverage was “notably fragmented,” with a median total duration of service of nine weeks among the 41 physicians (most also had varying degrees of outpatient care responsibilities). Each physician treated a median of 276 hospitalizations and a median of 1,035 patient-days during the study period.

Discharge likelihood significantly increased with greater attending continuity. Daily discharge probability increased for the average patient from 15.3% to 20.9% when the consecutive number of days the patient was treated by the same physician increased from one to seven. In contrast, the probability of daily discharge significantly decreased with greater severity of illness, higher annual mortality risk, and longer length of stay, as well as for elective admissions. Discharge was also less likely on the first day of admission (adjusted odds ratio, 0.40; 95% CI, 0.35 to 0.47) and on weekends and holidays (adjusted odds ratio, 0.61; 95% CI, 0.56 to 0.67).

Limitations of the study include its focus on only one patient outcome (daily discharge probability) and its single-center design, the study author noted. In addition, the analysis did not account for other factors that could influence the daily probability of discharge, including support from family and housestaff or nursing continuity. The findings “could be considered if physician or hospital administrators wish to classify hospital resource utilization as a factor when scheduling physician coverage,” the author wrote.

Costs of care among Medicare patients were similar at teaching vs. nonteaching hospitals

Medicare patients treated at major teaching hospitals had lower costs at 30 days and similar costs at 90 days compared to those treated at nonteaching hospitals, a recent study found.

Image by Getty Images
Image by Getty Images

The cross-sectional study assessed the costs of hospitalizations among Medicare beneficiaries age 65 years and older at teaching and nonteaching hospitals from Jan. 1, 2014, to Nov. 30, 2015, for 15 medical conditions and six surgical procedures. Indirect medical education (IME) payments were not included in the primary analysis because they were designed to help pay for the additional costs associated with teaching. Results were published on June 7 by JAMA Network Open.

The sample included about 1.25 million hospitalizations at 3,064 hospitals (232 [7.6%] major teaching, 837 [27.3%] minor teaching, and 1,995 [65.1%] nonteaching hospitals). Overall, treatment at a major teaching hospital was associated with slightly lower total 30-day adjusted standardized costs ($18,605 vs. $18,793 at minor teaching hospitals and $18,873 at nonteaching hospitals; difference between major and nonteaching hospitals, −$268 [95% CI, −$456 to −$80]; P=0.005). Thirty-day total costs were lower at major teaching hospitals compared with nonteaching hospitals for 12 of 21 conditions and procedures. By 90 days, there was no significant difference in costs by teaching status ($24,982 at major teaching hospitals, $24,959 at minor teaching hospitals, and $25,044 at nonteaching hospitals).

Treatment at a major teaching hospital was associated with significantly higher spending for the index hospitalization ($8,529 at major, $8,370 at minor, and $8,180 at nonteaching hospitals), but also with significantly lower physician costs ($677 at major, $725 at minor, and $728 at nonteaching hospitals). Postacute care costs at 30 days were lowest at major teaching hospitals ($6,015 at major, $6,239 at minor, and $6,260 at nonteaching hospitals). When including IME payments, which increased Medicare spending by about $1,200 at 30 days, major teaching hospitals had higher total spending at 30 and 90 days.

The study was limited to hospitalizations among Medicare patients, so the results may not be generalizable to other populations, particularly commercially insured patients, the authors noted. In addition, the study did not consider how out-of-pocket spending for patients may differ by hospital teaching status and may have had unmeasured confounders (e.g., differences in coding intensity), they said.

The findings “may seem unexpected given a general consensus that teaching hospitals are more expensive and that the involvement of trainees in patient care is relatively inefficient,” the authors wrote. “This study suggests that although costs are somewhat higher for the initial hospitalization at major teaching hospitals, spending after hospital discharge, particularly on post–acute care services, is generally lower.”

Transition program reduced admissions but not 90-day mortality

There was no significant reduction in 90-day probability of death or additional hospital encounters among patients randomized to a patient-centered transitional care intervention, while there were significant decreases in measures of inpatient admissions over 180 days, a study found.

To evaluate effects of a transitional care practice that comprehensively addresses patients' medical and psychosocial needs after hospital discharge, researchers designed a pragmatic, randomized, comparative effectiveness trial among adults discharged from an initial ED, observation, or inpatient hospital encounter with no trusted usual source of care.

Transitional care included a scheduled postdischarge appointment within 10 days of discharge, and care from a multidisciplinary team that assessed patients' medical and psychosocial needs, addressed modifiable barriers, and linked patients to a new primary care source. Routine care involved help in scheduling a postdischarge appointment with a primary care clinician who often partnered with the hospital.

The study's primary outcome was death or additional hospital encounters within 90 days of discharge. Secondary outcomes included hospital encounters and utilization over a longer period and patient-reported outcomes. Results were published by the Journal of General Internal Medicine on May 29 and appeared in the September issue.

Four hundred ninety patients were randomized to transitional care, and 164 were assigned to routine care. There was no significant difference between study arms in 90-day probability of death or additional hospital encounters. However, transitional care patients had lower probability of any inpatient admission over 90 days (relative risk, 0.63; 95% CI, 0.43 to 0.91) and 180 days (relative risk, 0.65; 95% CI, 0.47 to 0.89). Over 180 days, transitional care patients had 42% fewer inpatient admissions (incidence rate ratio, 0.58; 95% CI, 0.37 to 0.90).

“Multiple factors could have contributed to the TC [transitional care] intervention's effects on inpatient admissions,” the authors wrote, noting that the program occurred entirely after discharge, in contrast to some other efforts that also had predischarge components. “More explicit integration of the TC practice with predischarge transitional care could lead to increased practice attendance and, in turn, increased intervention effects.”

Electronic registry of CRE carriers may help reduce regional spread

Implementing a registry that tracks patients carrying carbapenem-resistant Enterobacteriaceae (CRE) may help reduce regional spread of the extensively drug-resistant organism, even with modest participation among inpatient facilities, a recent study found.

Researchers developed a model of patient flow at all inpatient health care facilities (including acute care hospitals, long-term acute care hospitals, skilled nursing facilities, and ventilator-capable skilled nursing facilities) in the Chicago metropolitan area and surrounding communities using the Regional Healthcare Ecosystem Analyst software platform. Scenarios in the model explored the regional impact of an electronic public health registry that tracked patients carrying CRE, and results were published online on May 9 by Clinical Infectious Diseases.

When all Illinois facilities (n=402) participated, the registry reduced the number of new CRE carriers by 11.7% and decreased region-wide CRE prevalence by 7.6% over a three-year period. With 75% participation among the largest facilities (n=304), registry use led to an 11.6% relative reduction in new carriers (16.9% and 1.2% among participating and nonparticipating facilities, respectively) and a 5.0% relative reduction in CRE prevalence. When 50% of the largest facilities (n=201) participated, the relative reductions in incident carriers and prevalence were 10.7% and 5.6%, respectively. Finally, with 25% participation among the largest facilities (n=101), there was a 9.1% relative reduction in incident carriers (20.4% and 1.6% in participating and nonparticipating facilities, respectively) and a 2.8% relative reduction in prevalence.

The biggest gains from the registry were seen among hospitals compared to other facilities and when participation in the registry among the largest facilities increased from 0% to 25%, the study authors noted. Limitations of the study include its regional, model-based design and the use of only CMS patient data, which may not represent transfer patterns of all patients, they noted.

The results show that an electronic public health registry may be an effective tool in combating the spread of extensively drug-resistant organisms between facilities in a region, the authors said.