The Helpline and the Missing OPD Patients
In March a district hospital launched a 12-seat tele-consultation helpline staffed by two medical officers and four nurses, working 9 a.m. to 5 p.m. Between March and August it logged 18,400 calls, of which 11,200 were closed with advice and 2,100 were referred to a specialist. In the same six months OPD footfall at the hospital fell from 96,000 to 74,900, a drop of 22%. The hospital's finance officer has written: 'The helpline has clearly diverted unnecessary visits from the OPD. We should therefore cancel the ₹1.8 crore OPD extension block and put the money into 20 more helpline seats.' The medical superintendent is uneasy: a new private hospital 4 km away opened in April, the OPD registration fee rose from ₹10 to ₹50 in May, and the helpline closes at 5 p.m. while the OPD runs until 8 p.m. She has asked for a one-page appraisal before the governing body meeting.
1. State the unstated assumption on which the finance officer's recommendation depends, and explain why the argument collapses without it.
Read the worked answer
The argument runs: visits fell after the helpline opened → the helpline caused the fall → the visits it removed were unnecessary → therefore shift ₹1.8 crore from OPD capacity to helpline seats.
The load-bearing assumption is that nothing else of comparable size changed over the same six months, so that the helpline is the only available explanation for a fall of 21,100 visits.
Why it is load-bearing: the recommendation is not 'study the helpline' but 'cancel a capital block'. That step needs two separate claims to be true - that the helpline caused the fall, and that what it displaced was low-value care. The passage offers evidence for neither. The arithmetic is also weaker than it looks: 18,400 calls is a large fraction of 21,100 lost visits, but calls are not visits, one patient can call twice, and a call closed with advice may have replaced nothing at all rather than an OPD attendance. Note also that 11,200 'closed with advice' calls plus 2,100 referrals is only 72% of total calls; the remaining 5,100 were not resolved, which sits awkwardly with the claim that the helpline is efficiently absorbing demand.
2. Give three competing explanations for the fall and, for each, name one piece of data - already in the passage or easily obtained - that would test it.
Read the worked answer
1. The new private hospital (opened April, 4 km away). A facility that close can absorb a large share of walking patients. Test: map the lost visits by ward or by catchment area. If the fall is concentrated in the wards nearest the private hospital and is near zero in far wards, the helpline is not the driver. Alternatively, obtain the private hospital's own OPD counts for the same months.
2. The fee rise from ₹10 to ₹50 (a 400% increase, May). Price is the classic cause of a fall in attendance, and the timing (May, inside the window) fits. Test: split footfall into paying and fee-exempt patients. If attendance by exempt patients (free-treatment scheme, emergency cases) held steady while paying attendance collapsed, the fee, not the helpline, explains the drop.
3. Seasonality and the evening gap. March–August spans summer and the monsoon; compare with March–August of the previous year. Test: if last year's same six months also fell 15–20%, the helpline explains very little. Separately, plot the lost visits by hour: the helpline stops at 5 p.m. and the OPD runs to 8 p.m., so any loss concentrated after 5 p.m. cannot be attributed to the helpline at all.
A defensible appraisal presents these as rival hypotheses and says which one the hospital can test within a fortnight - the fee split and the time-of-day plot can both be produced from registers already in the building.
3. Using only the figures given, assess how far the evidence supports buying '20 more seats' specifically, and state what must be measured before ₹1.8 crore is committed.
Read the worked answer
Volume check. Six months at roughly 26 working days a month is about 156 days. 18,400 ÷ 156 ≈ 118 calls a day, over an 8-hour day, i.e. about 15 calls an hour. Spread across 12 seats that is roughly 1.2 calls per seat per hour. A 12-seat line at that load is not obviously saturated; the constraint may be the two medical officers who must sign off advice and referrals, or the 5 p.m. closing time, or public awareness - not seat count.
What the figures cannot show. Nothing in the passage reports call abandonment, waiting time, average call duration, or repeat callers. Without those, '20 more seats' is a guess: if average handling time is long or abandonment is low, extra seats add cost without adding consultations.
Before committing ₹1.8 crore, measure: (i) abandonment rate and queue time by hour of day; (ii) average call duration and the share of repeat callers; (iii) the time-of-day profile of demand, which will show whether the binding constraint is 5 p.m. closing rather than seats; (iv) where the 21,100 lost OPD visits actually went. A cheap first step - extending hours to 8 p.m. with the existing 12 seats for three months - would test the capacity question at a fraction of the capital cost.