The Ninety-Minute Wait
At one Ethiopian referral hospital, the average outpatient waited 90 minutes to be seen — and only 14% were seen inside the 50-minute mark. The measured causes are not clinical. They are clerical.

A patient's opinion of your clinic is formed before they meet a clinician. It's formed in the queue.
We tend to treat waiting as an unavoidable by-product of demand — too many patients, too few clinicians, nothing to be done. The Ethiopian evidence says otherwise. When researchers actually measure where the minutes go, most of them are not lost to consultation time. They are lost to finding a card, taking a payment, and nobody telling the patient what happens next.
What the queue actually costs
At Debre Tabor Comprehensive Specialized Hospital, researchers timed 1,058 adult outpatients across April 2024. The average wait for consultation was 90 minutes (SD 36.1; 95% CI 87.9–92.4). Measured against the 50-minute standard the authors reference, only 13.8% of patients were seen in acceptable time.
Ninety minutes is not one long delay. It's a dozen small ones stacked end to end — and the study named them.
- Patients whose charts were moved by dedicated porters were far less likely to wait long (AOR 0.17, 95% CI 0.05–0.57). Record transport was one of the biggest levers in the building.
- Patients served by punctual staff waited less (AOR 0.54, 95% CI 0.32–0.89).
- Patients who received a brief explanation of how the outpatient department works waited less (AOR 0.34).
Earlier work at Debre Markos and Felege Hiwot hospitals put numbers on the same theme: long card searches accounted for 28.9% and 31.5% of the delay reasons respectively, alongside long registration times and high patient-to-doctor ratios.
Read that list again. Card searches. Chart transport. Registration. Being told what's happening. Not one of those requires another physician.
Satisfaction follows the queue
A 2025 meta-analysis pooling 16 Ethiopian studies and 6,580 outpatients found overall satisfaction with outpatient services at 61.95% (95% CI 53.0–70.9). Roughly two in five patients leave dissatisfied.
The factors most strongly associated with satisfaction in that review are worth memorising, because they are all operational rather than clinical:
| What the patient experienced | Odds of being satisfied |
|---|---|
| Provider respected their privacy | OR 6.58 (3.32–9.83) |
| Prescribed drugs were actually in the pharmacy | OR 4.04 (3.46–4.62) |
| Consultation room was clean | OR 2.20 (1.68–2.71) |
| Was told how to prevent their illness recurring | OR 2.12 (1.68–2.56) |
Privacy tops the list. Not equipment, not seniority of the clinician — whether the patient felt their information was handled with discretion. A crowded registration desk where the person behind you can hear your history is a satisfaction problem and a confidentiality problem, and patients register it as one thing.
There's a second gap worth noting. In Hadiya Zone, 65.8% of insured outpatients were satisfied compared with 44.95% of uninsured ones. The same building, the same staff, a twenty-point difference — much of it in how smoothly the paperwork of paying moves.
If a patient's wait is ninety minutes and their consultation is twelve, you are not running a clinical service with an administrative annexe. You are running an administrative service with a clinical appointment at the end of it.
Four fixes that don't require hiring
1. Stop searching for records. A card that has to be found is a card that can't be found. When the record opens with the patient's name — at registration, in the consultation room, at the cashier — the single largest measured source of delay disappears rather than being managed.
2. Make registration one step, not three. Every hand-off between desks is a queue. Capture the patient once, and let payment and the clinical note read from the same record instead of re-creating it.
3. Tell people what's happening. Patients briefed on the workflow waited less and rated the wait better. Some of that is real sequencing; some of it is that an explained twenty minutes feels shorter than an unexplained ten. Both count.
4. Watch the number. Time from arrival to consultation, measured weekly, for a sample of ten patients. You cannot manage a queue you have never timed — and every clinic that has timed it has been surprised.
The part that compounds
Long waits don't just annoy people. They quietly select which patients come back. The Lithuanian data the Debre Tabor authors cite makes the link explicit: longer waits correlated with patients resolving their problem at a different facility or not attending at all. In a chronic-care caseload, that patient doesn't reappear as a complaint. They reappear as a complication, two years later, or not at all.
Ninety minutes is not a fact of Ethiopian healthcare. It's an accumulation of clerical steps, each individually reasonable, that nobody has yet been asked to justify in minutes. Time yours, and start with the cards.
Sources
- Setie A, Mekete G, Asnakew A, Abeje G. Assessment of Adult Outpatient Waiting Time for Consultation and its Associated Factors at Debre Tabor Comprehensive Specialized Hospital, Northwest Ethiopia, 2025. Journal of Nursing & Healthcare 11(1):1–12, 2026. Facility-based cross-sectional study, 1,058 clients, 1 April – 1 May 2024.
- Tura MR, et al. Patient Satisfaction With Outpatient Department Health Service and Associated Factors at Public Hospitals in Ethiopia. Inquiry 2025. Systematic review and meta-analysis, 16 studies, 6,580 participants.
- Disparities in satisfaction among insured and uninsured adult outpatient department service users in Southern Ethiopia, 2022. Comparative cross-sectional study, Hadiya Zone.
Figures are quoted with their study year and setting because they are facility-level findings, not national statistics. A single hospital's mean wait is evidence about that hospital, and a useful benchmark for yours — not a national average.
See Bloom in your clinic
Fewer no-shows, less paper, cleaner billing — in one calm system your team will actually use. Book a short demo and we'll show you how.


