Reducing Antibiotic and Bronchodilator Use in Infant Bronchiolitis: A Small-Numbers Quality Problem
You are the lead NP in a four-provider pediatric clinic that sees about 90 bronchiolitis visits each season. At a staff meeting a colleague presents an audit: of 62 infants aged 2-11 months seen for bronchiolitis last winter, 41% received antibiotics, 37% had a chest radiograph, and 29% were treated with albuterol; 8% were referred to the emergency department and 2 were admitted. She proposes a mandatory "no albuterol" policy with individual chart audits and feedback letters. Another provider objects that two winters ago an infant with bronchiolitis was later diagnosed with pneumonia, and she worries about missing something. The medical director wants a plan before the season starts in six weeks. Baseline: there is no bronchiolitis order set, no standardized discharge handout, and triage nurses use a homegrown respiratory score that has never been validated. Four providers see most of the infants, and two are part-time.
1. What change ideas and measures would you choose, and why not a mandate with individual feedback?
Read the worked answer
Start from the evidence, then change the environment. The AAP bronchiolitis guideline recommends against routine chest radiography, bronchodilators, systemic corticosteroids, and antibiotics in uncomplicated bronchiolitis. Treatment is supportive: suctioning, hydration, and oxygen when saturation falls below 90%.
Change ideas: an EHR order set and note template that offer suctioning, a feeding plan, and return precautions first, so the default is the guideline; a one-page family handout placed in the visit packet; a required free-text field when a radiograph or antibiotic is ordered, which produces data rather than punishment; and nursing education tied to one validated score instead of the homegrown one.
Measures in three families, not one:
- Outcome - antibiotic prescribing rate, ED referrals, admissions.
- Process - chest radiograph rate, albuterol use, documented risk assessment, handout given.
- Balancing - return visits within 7 days, ED visits after a clinic visit, family satisfaction, and clinician time per visit.
Skip the mandate. A "no albuterol" policy with individual audit-and-feedback letters punishes the judgment you are trying to improve, and it guarantees the resistance you already heard in the room.
2. How do you interpret the baseline data, and what would you track monthly?
Read the worked answer
Sixty-two visits is a small denominator. Forty-one percent is 25 infants; a shift of three patients moves the rate by five percentage points, so a before-and-after comparison will produce either a celebration or a panic that the data do not support.
Use a run chart or p-chart with monthly rates and control limits, annotated with the date of each intervention. Look for a shift - six or more consecutive points on one side of the median - or a trend, not for a single month's dip. Keep the denominator next to every percentage.
Case mix matters: log the age distribution and the proportion of visits under 3 months or with comorbid conditions, because those drive both prescribing and referral and will confound your interpretation.
Keep collection cheap: a two-field EHR query of diagnosis code plus antibiotic order, pulled monthly and reviewed by one person in 15 minutes. If collecting the data takes longer than that, it will stop by February.
Set an honest target: not zero. Aim for a 25-30% relative reduction in the first season, then reassess. Add a small qualitative arm - five chart reviews of antibiotic-treated cases - to learn the actual reasoning, which is what tells you whether the pathway is being ignored or is simply wrong.
3. How do you address the colleague's fear of missing pneumonia, and how do you sustain the change?
Read the worked answer
Her concern is legitimate and should be named before the data are presented. Bronchiolitis and bacterial pneumonia overlap at the edges, and a clinician who has missed one will be risk-averse for years. The way to help her is to standardize what "something else" looks like rather than telling her to worry less.
Agree on explicit red flags that justify imaging or antibiotics: apnea, toxic appearance, focal findings that persist after suctioning, temperature at or above 38°C in an infant under 3 months, insufficient oral intake or decreased urine output, or worsening after day 3-5 of illness. Put that list on the order set, so the pathway supports clinical judgment instead of replacing it.
Give uncertainty somewhere to go: a same-day recheck slot or a curbside second opinion, so the alternative to a prescription is not "send them home and hope."
Sustain it: recruit a clinician champion - ideally her, once she has watched the pathway work - give 10-minute monthly run-chart feedback at staff meetings, and plan in advance how a bad outcome will be reviewed, using a structured system-focused review that asks what the system allowed rather than who failed.
Check the system, not just the clinician: if triage assigns 20-minute slots to wheezing infants, the pathway will fail for reasons that have nothing to do with knowledge.