Two Laboratories, One Advisory: Water Contamination in a Municipal Supply
You chair the District Water Quality Committee in a district of 1.4 million people. A sampling round covered 4 of the 11 municipal tubewells. Two samples returned total dissolved solids of 2,300–2,600 mg/L and nitrate of 60–68 mg/L; one of those also showed faecal coliform. The municipal laboratory calls the bacteriological result a 'sampling error' and offers to re-test in eight days. A private laboratory paid for by a residents' association reports coliform contamination at a third well. Over the same ten days, the teaching hospital recorded 47 acute diarrhoea admissions in children under five against a seasonal baseline of 12, and two union councils have begun storing water in open drums. The TMA water account holds PKR 4.2 million. Chlorination units for three wells cost PKR 1.9 million in capital; a tanker service for both union councils costs PKR 900,000 per month. A sitting MNA has publicly called the water 'perfectly safe' and has asked you to hold any public advisory until after a by-election in eleven days.
1. Set out your first-72-hour action sequence and justify the order in which you take each step.
Read the worked answer
Ordering principle: stop irreversible harm first, then establish facts, then fix the system. Diagnosis can wait; exposure cannot.
Hours 0–6 - interrupt exposure, on precaution. The hospital signal alone justifies action: 47 admissions against a baseline of 12 is roughly a fourfold excess concentrated in under-fives, and it coincides with the sampling round. You do not need a settled laboratory verdict to issue a boil-water notice and close the two wells with nitrate above 50 mg/L. Closing them is not merely cautious - nitrate cannot be removed by boiling (boiling concentrates it) and is a methaemoglobinaemia risk in infants, so the advisory must say boil for microbiological safety but switch infants to tankered or bottled water entirely.
Hours 6–24 - resample properly. Split samples to the municipal lab, the private lab and the provincial reference laboratory, with sterile bottles, ice, signed seals and a witness from the residents' association.
Hours 24–48 - start tankering and sanction chlorination. The cost comparison is decisive: tankering at PKR 900,000 per month exhausts the PKR 4.2 million account in under five months and fixes nothing; PKR 1.9 million of chlorination pays for itself against tankering within about two months and continues to protect the supply afterwards.
Hours 48–72 - build the line list. Map every hospital case by union council and date against the well catchments; that is what will tell you whether the third well is genuinely implicated.
2. The two laboratories disagree. How do you resolve the conflict in a way that would survive an audit and public scrutiny?
Read the worked answer
Do not adjudicate between the two reports on the strength of who wrote them, and do not simply pick the more convenient one. Design a test that discriminates between the competing explanations.
- Split-sample protocol. One sample, divided on site into three sealed aliquots, dispatched the same day, blind-coded, to the municipal lab, the private lab and the provincial reference laboratory. Same bottle, same day, same chain of custody removes sampling as a variable.
- Chain of custody on paper. Named courier, dispatch time, temperature on arrival, seal numbers, receiving officer's signature. Without this the exercise proves nothing.
- Test the 'sampling error' claim rather than dismissing it. If the fault lay in sampling, a properly taken resample clears the well. If the fault lies in the municipal laboratory's media, incubation or analyst technique, the third lab will not reproduce the municipal result. The protocol is what makes the two hypotheses separable.
- Refuse the eight-day resample as the sole test. It is neither independent nor timely; eight days exceeds the incubation window of the illness you are investigating.
- Check the municipal laboratory's own records - media batch, incubator log, calibration, analyst qualification - without turning the exercise into a blame inquiry.
- Publish the outcome either way, including if the private lab turns out to be wrong. Credibility after a scare depends on being seen to report inconvenient results.
The audit trail is the dated memos, the seal numbers and the third-party result, not your verbal reasoning.
3. How do you handle the MNA's request to delay the advisory until after the by-election?
Read the worked answer
Decline the delay, but decline it institutionally rather than personally.
- Acknowledge the request and record it. Note the date, the request and your reasons for not complying in a written file note. An undocumented conversation later becomes your word against his.
- Apply the asymmetry test. A delayed advisory risks illness and possibly death in infants; an early advisory risks some political embarrassment and a temporary drop in turnout. The first harm is irreversible, the second is not. That asymmetry, not political preference, is the stated reason.
- Route the decision upward. Copy the note to the Secretary, Health and the Director-General Health so the decision belongs to the department and not to an individual officer who can be transferred or pressured.
- Give him a constructive role. Offer a briefing and invite him to appear alongside you at the announcement. Converting an opponent into a co-signatory is worth more than winning the argument privately.
- Keep the advisory factual. No attribution of blame, no reference to the by-election, no speculation about the cause of contamination before the third-lab result. A neutral, technically framed notice is harder to attack and does not politicise the department.
- Escalate in writing if pressure continues, and note that you have done so.