The hardest HEDIS measures to close (and why outreach is usually the bottleneck)
The hardest HEDIS measures to close are typically colorectal cancer screening, controlling high blood pressure, diabetes A1c control, follow-up after ED visits for mental illness, and medication adherence. In most organizations the bottleneck is not clinical: it is outreach capacity, scheduling friction, referral completion, and documentation that never makes it back into the record.
Ask a quality director which measures kept them up at night last measurement year and you will hear the same handful of names, almost regardless of organization type. That is not a coincidence. The measures that stay stubbornly below target share a structural trait: they require the patient to do something outside the exam room, often more than once, and often somewhere other than your clinic.
That structure is why the same intervention (another provider education session, another EHR alert) keeps failing to move them. The constraint is not that clinicians forget to order screenings. The constraint is that somebody has to reach the patient, get a test or visit scheduled, get the patient there, and get proof into the record, thousands of times per year.
This article ranks the measures organizations most often miss and names the specific operational bottleneck for each. HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA); measure definitions and specifications are published by NCQA at ncqa.org/hedis. If you want the fundamentals first, start with our HEDIS measures explained guide. And a standing caveat for everything below: whether any given screening, test, or medication is appropriate for a specific patient is a decision per the ordering provider. This article is about operations, not clinical judgment.
What makes a HEDIS measure hard to close
A HEDIS measure is hard to close when the distance between the clinical decision and the credited result is long. Four factors do most of the damage:
- Patient action outside the visit. The patient must complete something later, elsewhere: a colonoscopy, a lab draw, a follow-up appointment, a refill.
- Time windows. Some measures only credit action within days (behavioral health follow-up) or require sustained performance across the year (adherence, blood pressure control).
- Cross-organization handoffs. Screening and follow-up often route through external specialists, imaging centers, or labs, so completion depends on another organization's scheduling desk.
- Documentation distance. Even completed care fails the measure if the result never lands in a structured field or supplemental data feed the measure engine can read.
Measures with all four traits live at the top of the "hardest" list every year. Measures that can be closed inside a single visit (a BMI recorded, a screening question asked) rarely appear on it.
The 10 hardest measures, ranked
Rankings vary by population and contract mix, but across health plans, medical groups, and health centers, these ten come up most often. The table names the dominant bottleneck; the sections that follow explain the fix.
| Rank | Measure | Dominant bottleneck |
|---|---|---|
| 1 | Colorectal cancer screening | Referral and procedure completion |
| 2 | Controlling high blood pressure | Repeat visits and documentation |
| 3 | Glycemic control for patients with diabetes (A1c) | Lab completion and recurring outreach |
| 4 | Follow-up after ED visit for mental illness | Speed: 7-day and 30-day windows |
| 5 | Breast cancer screening | Imaging scheduling and outreach reach |
| 6 | Cervical cancer screening | Outreach sensitivity and long intervals |
| 7 | Childhood immunization status | Multi-dose series completion by age 2 |
| 8 | Well-child visits | Scheduling volume and family logistics |
| 9 | Eye exam for patients with diabetes | External specialist referral loop |
| 10 | Medication management and adherence measures | Sustained behavior across the year |
Screening measures: colorectal, breast, cervical
Colorectal cancer screening tops most lists because it has the longest completion chain in ambulatory quality: identify the due patient, reach them, discuss options, order, refer, get the prep instructions across, get the procedure scheduled and completed, and get the result documented. Every link leaks. The bottleneck is almost never the order; it is the referral that dies between primary care and gastroenterology. We wrote a full operational breakdown of that specific handoff in our guide to gastroenterology referral and colonoscopy scheduling. The fix is closed-loop referral management plus proactive outreach: track every screening referral to a booked date, chase the unbooked ones on a schedule, and offer stool-based testing pathways where the ordering provider deems them appropriate, since a mailed kit converts some patients a procedure never will.
Breast cancer screening fails earlier in the chain: patients simply never get reached or never get a mammogram slot booked. Imaging centers book out, work hours conflict, and a single voicemail is not an outreach program. The fix is multichannel, multilingual outreach with direct scheduling: a call or text that ends in a booked imaging slot, not in "please call radiology."
Cervical cancer screening adds sensitivity: outreach scripts matter, intervals are long (which multiplies false gaps from care completed elsewhere), and the visit itself is one many patients defer. The fix combines careful gap validation against outside records with respectful, persistent outreach that lets patients book into a visit they choose.
Chronic condition control: blood pressure and A1c
Controlling high blood pressure is hard because it is a results measure, not an activity measure. The patient must be seen, the reading must be controlled, and the reading must be documented in a way the measure logic accepts. Operationally that means two failure modes: patients with uncontrolled readings who never return for the recheck visit, and patients whose control was achieved but whose qualifying reading never made it into the right field. The fix is a recheck pipeline (every elevated reading generates an outreach task with a target date) plus a documentation audit so the measure engine sees what the clinicians saw.
Glycemic control for patients with diabetes behaves similarly: the A1c draw is the unit of work, and patients who feel fine skip labs. Standing lab orders help; what moves the number is systematic outreach that converts an overdue lab into a booked draw, ideally attached to a visit the patient already wants. Point-of-care testing at any visit the patient attends, where clinically appropriate per the ordering provider, removes one trip entirely.
Both measures reward the same operating pattern: continuous, list-driven outreach across the whole year rather than a fourth-quarter scramble, because a controlled reading in November cannot rescue a patient who disengaged in March.
Behavioral health follow-up after the ED
Follow-up after an emergency department visit for mental illness credits follow-up within 7 and 30 days. This is the purest outreach-speed measure in the set: by the time a typical monthly gap file arrives, the 7-day window is long gone.
The bottleneck is latency at every step: learning about the ED visit (often via ADT feeds or payer notifications days later), reaching a patient population that is disproportionately hard to contact, and finding behavioral health appointment supply inside the window. The fix is event-driven rather than list-driven operations: treat each ED discharge notification as a same-day outreach task, attempt multiple channels within 24 to 48 hours, and hold protected rapid-access slots or telehealth options for this population. Organizations that keep working this measure off a monthly worklist will keep missing it, regardless of how motivated the staff are.
The related follow-up measures for substance use follow the same pattern and fail for the same reason: the window closes faster than manual outreach can move.
Pediatric measures: immunizations and well-child visits
Childhood immunization status requires completing multiple vaccine series by the second birthday. Operationally it is a series-completion problem: families start well and drift, one missed visit cascades into several missed doses, and records scatter across immunization registries, pharmacies, and prior practices. The fixes are registry reconciliation before outreach (many "missing" doses were given elsewhere), forward-booking the next visit before the family leaves the current one, and recall outreach that restarts drifted series with an actual booked appointment.
Well-child visits (in the first 30 months, and annually for children and adolescents) are a volume and logistics measure: working parents, transportation, and school schedules. The playbook mirrors annual wellness visit outreach on the adult side: proactive scheduling campaigns, evening and weekend slot inventory where feasible, text-first reminders, and immediate rebooking of every no-show. Community health centers, which report performance publicly through HRSA's Uniform Data System, consistently show that outreach staffing levels track directly with these rates; our care gap outreach benchmarks for FQHCs and CHCs covers realistic numbers.
Diabetes eye exams and the external referral loop
The retinal eye exam for patients with diabetes deserves its own entry because it fails in a distinctive way: the care happens in a different specialty, often in a different organization, and the result must travel back. Patients get referred to optometry or ophthalmology, never book, or complete the exam and the report never returns to the primary care chart, so the gap stays open despite completed care.
The fix has two halves. First, treat the referral like a tracked order with a completion deadline, not a suggestion: confirm booking, chase no-books, and rebook no-shows. Second, build the results return path: standing agreements with eye care partners on report delivery, plus in-clinic retinal imaging with remote reading where the organization can support it and the ordering provider directs. This measure is a preview of a general truth: any measure whose numerator lives in another organization's EHR will be as hard as your referral loop is leaky. (Where the specialist's input is a question rather than an exam, some organizations route it differently; our comparison of eConsults vs referrals covers when each pathway fits.)
Medication management and adherence measures
HEDIS includes medication measures such as antidepressant medication management and statin therapy measures with adherence components, and most value-based contracts add pharmacy adherence metrics alongside them. These are the hardest class of all because the "action" is continuous: the patient must fill and take medication across months, and every refill lapse erodes the metric.
The operational bottleneck is detection latency plus contact volume. Refill gaps surface in pharmacy data days or weeks after the miss, and each flagged patient needs a conversation, not a letter: side effects, cost, confusion about instructions, or a simple lapse each require different handling, and anything touching the regimen itself goes back to the prescriber. The fix is fast-cycle outreach on refill-gap triggers, barrier screening scripts that route clinical questions to clinicians and logistics problems (cost, transport, pharmacy switching) to staff or automation, and refill synchronization so one trip covers all medications.
The bottom line
The hardest HEDIS measures are hard for operational reasons, not clinical ones: they demand repeated patient contact, cross-organization scheduling, tight time windows, and airtight documentation. That is why the leverage sits in the outreach and coordination layer. The organizations that outperform on these measures run them like a production system: validated worklists, event-driven speed for windowed measures, multichannel and multilingual outreach that ends in a booked appointment, referral loops tracked to completion, and same-week documentation writeback. Provider education and EHR alerts are table stakes; care gap closure automation is where the remaining points live.
Frequently asked questions
What are the hardest HEDIS measures to close?
Why is colorectal cancer screening so hard to close?
Why do behavioral health follow-up measures have low rates?
Do documentation problems affect HEDIS scores?
Can AI or automation improve HEDIS measure performance?
Where are HEDIS measure definitions published?
Sources
