You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.

Primary Care Corner with Geoffrey Modest MD: Antibiotic Overprescribing

3 Aug, 15 | by EBM

By: Dr. Geoffrey Modest

An array of recent articles highlighted the issue of antibiotic overuse (and the increasing potential for antibiotic resistance).

Background — the CDC in 2013 released a report elaborating the burden of antibiotic resistance:  2 million antibiotic-resistant illnesses and 23,000 deaths yearly in the US.

  1. CDC researchers published a study looking at outpatient prescriptions dispensed in 2011, using the IMS Health Xponent database which contains >70% of all outpatient prescriptions in the US, including all payers from community pharmacies and nongovernmental mail service pharmacies (see Clinical Infectious Diseases 2015;60(9):1308–16). >60% of antibiotic expenditures are in the outpatient setting. 58% of all antibiotic prescriptions in the outpatient setting are for respiratory infections that are predominantly viral.  A total of 262.5 million courses of outpatient antibiotics were prescribed in 2011, an astounding 842 prescriptions per 1000 persons in that year.

addiction-71573_640

Results:

  • Antibiotic prescriptions: penicillin – 60.3 million, macrolides 59.1 million, cephalosporins 35.6 million, quinolones 27.6 million, b-lactams with increased activity 21.6 million, tetracyclines 21.1 million, trimethoprim/sulfa 20.3 million.
  • Top five agents: azithromycin 54.1 million, amoxacillin 52.9 million, amox/clavulanate 21.2 million, ciproflox 20.9 million, cephalexin 20.0 million
  • By gender (rate per 1000 persons): female 990, male 672
  • Census region (rate per 1000 persons):  south 931, midwest 897, northeast 848, west 647
  • Age (rate per 1000 persons): under 3yo — 1287, 3-9yo — 1018, 10-19yo — 691, 20-39yo — 685, 40-64yo — 790, >65yo — 1048
  • By provider type (rate per 1000 persons): family practice 667, pediatrics 598,  emergency medicine 427, internal medicine 383
  • By general assessment of demographics by counties: adjusted odds ratio of 0.6 in those with highest % with 4 years of college, 0.5 for highest 1/3 of per capita income, 1.7 for those with more obese adults
  1. A retrospective cross-sectional review of all patients seen in the VA system between 2005 and 2012 for acute respiratory infections (ARIs) — (see Ann Intern Med. 2015;163(2):73-80​). Finding:
  • VA network includes 6.5-8.5 million veterans seen yearly at 1700 clinical and 152 hospitals with approximately 13 million primary care visits/yr, all using the same electronic medical record.
  • 1.045 million people (85.8% men, median age 61, 98% without fever, 62.5% seen by MD/24.5% by midlevel provider, median provider age 50, 72.4% seen in primary care clinic and 30.1% in community-based outpatient clinic, 22.9% in ER, 19.6% from the western US/28.4% central/35.6% south/ 16.5% northeast),  with diagnosis of nasopharyngitis, pharyngitis, sinusitis, acute bronchitis, upper respiratory infection, and others (laryngitis, tonsillitis), excluding those with diagnoses of pneumonia, influenza, urinary tract infection, or with serious comorbidities (HIV, neoplasia, diabetes, chronic lung disease, end-stage renal disease, transplantation, other immunocompromise).

Results:

  • Overall increase in use of antibiotics from 67.5% in 2005 to 69.2% in 2012 (p<0.001).
  • Increase in macrolide prescriptions form 36.8% to 47.0% in same time period (p<0.001), with decrease in penicillins (36.0% to 32.1%; p<0.001) and fluoroquinolones (15.0% to 12.7%; p<0.001).
  • Antibiotics were prescribed for 68.4% of ARIs, with antibiotics given for 86% of those with sinusitis, 85% bronchitis, 78% with T>102. antibiotics given in 75% of urgent care visits, slightly more by midlevel providers than MDs (70% vs 68%), slightly higher in VA clinics than community-based ones (70% vs 64%).
  • Macrolides prescribed for 51% of bronchitis and 49% of upper respiratory infections (macrolides are not recommended as first-line therapy for either pharyngitis or sinusitis; and there is an increase in macrolide-resistant pneumococcal disease as well as potential cardiotoxicity).
  • The greatest variability in prescribing was by provider (instead of temperature, setting type, geographical region), with 10% of providers prescribing antibiotics >95% of all ARI visits, and 10% in <40% of these visits.
  1. The MMWR just published a report on the knowledge and attitudes of adult patients and health care providers regarding antibiotic usage (seehttp://www.cdc.gov/mmwr/preview/mmwrhtml/mm6428a5.htm?s_cid=mm6428a5_w). They surveyed 4701 US consumers in 2012 (response rate of 86%), 4420 consumers in 2013  (response rate 79%), 2609 Hispanic consumers (response rate 38%), and 3149 health care providers (response rate 48%). results:

For consumers overall:

  • 17% felt that when they have a cold, they should take antibiotics to prevent getting sicker
  • 25% felt that when they have a cold, antibiotics help them get better more quickly
  • Approx 20% thought that antibiotics had common side-effects (nausea/vomiting, diarrhea, headache, rash), and only 16% thought antibiotics had none of these adverse effects
  • 20% had taken antibiotics from sources other than clinics/providers (mostly left-over ones, some from family members, some from neighborhood grocery stores)
  • Only 26% expected an antibiotic when they saw a provider for a cough or cold. 35% expected suggestions for symptom relief. 42% just wanted to make sure they had nothing more serious going on
  • Hispanic consumers were different in a few areas –more likely to think antibiotics helped (around 45%), more likely to get antibiotics from outside of clinic/provider (54% of Hispanics overall: especially leftover antibiotics and from the neighborhood grocery store — the local “bodega”), and more of them expected antibiotics to be prescribed (41%)

For health care providers:

  • 54% thought parents/patients expected an antibiotic
  • 77% thought they wanted symptom relief and 72% wanted reassurance that it wasn’t anything more serious going on
  • In terms of deterrents to prescribing antibiotics, 94% were concerned about antibiotic resistance and 71% about adverse effects. also 58% were concerned about killing “good bacteria” ​

For prior blogs on antibiotic overprescribing, see http://blogs.bmj.com/ebm/2015/01/25/primary-care-corner-with-geoffrey-modest-md-antibiotic-overprescribing/​ which assesses 2 studies, one finding large-scale antibiotic overprescribing for kids with pharyngitis, and the other looking at giving antibiotics for adults with respiratory infections in the Partners system in Boston finding that overall there was overprescribing, but that there were more prescriptions later in a clinic session, suggesting provider fatigue. For a rather sobering blog on the recent WHO report on worldwide antibiotic resistance, see http://blogs.bmj.com/ebm/2014/07/11/primary-care-corner-with-geoffrey-modest-md-whos-remarkable-scary-report/

One positive development is that there have been major gains in decreasing the use of antibiotics in chickens (use of antibiotics in animals leads to increased size of animal/profit but at the considerable expense of increasing antibiotic-resistant bacteria. In 2011 there was 29.9 million pounds of antibiotics sold in the US for meat/poultry, and 7.7 million pounds for people). Tyson just announced that they would eliminate routine use of antibiotics within 2 years. McDonalds (which uses lots of Tyson chicken) and Chipotle Mexican Grill are eliminating chickens raised with antibiotics. Purdue and Pilgrim’s Pride are decreasing antibiotic usage. The first 2 studies above confirm that we as providers overall seem to be increasing the use of antibiotics for non-indicated reasons, and we are using more broad-spectrum antibiotics which create more wide-spread antibiotic resistance. And, it seems that most people with cough/cold are not expecting antibiotics. In this regard, it is pretty striking that health care providers lag behind patients in how they view antibiotic prescribing for predominantly viral illnesses, with more than twice as many thinking parents/patients expect antibiotics than actually do!!! My own experience is that in the vast majority of cases, confidently telling the patient that “the good new is that you have a viral infection, which will get better on its own and antibiotics will not help” really works. and my sense over time is that there are many fewer patients expecting antibiotics or unsatisfied with that statement.

By submitting your comment you agree to adhere to these terms and conditions
You can follow any responses to this entry through the RSS 2.0 feed.
EBM blog homepage

Evidence-Based Medicine blog

Analysis and discussion of developments in Evidence-Based Medicine Visit site



Creative Comms logo

Latest from Evidence-Based Medicine

Latest from EBM