Here’s a draft that creates a clear paper trail without sounding confrontational:

Subject: Follow-up: Heating repair outstanding since [date reported] – [your address]

Dear [Landlord's Name],

I hope you're well. I'm writing to follow up on the heating at [your address], which I first reported as not working on [date].

Since then, I've followed up on [date of 2nd contact] and [date of 3rd contact, if applicable], but the heating still hasn't been repaired. It's now been three weeks without proper heating, and the property is becoming very cold, particularly at night, which is making it difficult to live in comfortably.

I appreciate that repairs and contractors can sometimes take time, and I'm keen to resolve this between us without any difficulty. However, I do need this to be treated as urgent now.

Could you please:

1. Confirm you've received this message

2. Let me know when an engineer will attend

3. Arrange for the repair to be completed by [specific date, e.g., this Friday]

If your usual contractor isn't available, I'd be grateful if you could arrange an alternative. I'm happy to be home to provide access — I'm available on [your availability].

Please could you reply by [tomorrow / date] with a confirmed plan? Thank you for your prompt help with this.

Kind regards, [Your Full Name] [Your Address] [Your Phone Number]

Tip: Replace the dates in brackets before sending — listing exactly when you first reported it and chased it is what makes this firm, while the cooperative tone keeps it from feeling like a threat.

import { useEffect, useState } from 'react';

/**
 * Debounces a value by the given delay.
 * Cancels the pending timeout on value change and on unmount.
 */
export function useDebouncedValue<T>(value: T, delay: number = 500): T {
  const [debouncedValue, setDebouncedValue] = useState<T>(value);

  useEffect(() => {
    const timer: ReturnType<typeof setTimeout> = setTimeout(() => {
      setDebouncedValue(value);
    }, delay);

    // Cleanup runs when `value`/`delay` changes AND on unmount,
    // cancelling the pending update.
    return () => {
      clearTimeout(timer);
    };
  }, [value, delay]);

  return debouncedValue;
}

Usage

function SearchInput() {
  const [query, setQuery] = useState<string>('');
  const debouncedQuery = useDebouncedValue<string>(query, 300);

  useEffect(() => {
    if (debouncedQuery) {
      // Only runs 300ms after the user stops typing
      fetchResults(debouncedQuery);
    }
  }, [debouncedQuery]);

  return (
    <input
      value={query}
      onChange={(e) => setQuery(e.target.value)}
      placeholder="Search..."
    />
  );
}

Short answer: only about 33% — roughly 1 in 3.

Most people guess 99%, or close to it. That intuition is wrong because it ignores how rare drug use is in the first place.

The calculation

Assume "99% accurate" means:

• If you do use the drug, the test is positive 99% of the time

• If you don't use it, the test is negative 99% of the time (so positive 1% of the time)

And only 0.5% of people use it.

By Bayes' theorem:

$$ P(\text{User} \mid +) = \frac{P(+ \mid \text{User})P(\text{User})}{P(+ \mid \text{User})P(\text{User}) + P(+ \mid \text{Non-user})P(\text{Non-user})} $$

$$ = \frac{0.99 \times 0.005}{(0.99 \times 0.005) + (0.01 \times 0.995)} = \frac{0.00495}{0.00495 + 0.00995} \approx 0.33 $$

Why it works: count real people

Imagine testing 10,000 random people:

So of the 149 positive results, only about 50 are real users:

$$49.5 / 149 \approx 33\%$$

False positives outnumber true positives about 2 to 1.

Why this surprises people

1. It reverses the probability. The test tells us $P(+ \mid \text{User}) = 99\%$ — the chance of a positive given you're a user. We want $P(\text{User} \mid +)$ — the chance you're a user given a positive. Those are very different things.

2. We neglect the base rate. Because non-users vastly outnumber users (9,950 vs. 50), even a tiny 1% false-positive rate applied to that huge group produces more positives than the 99% hit rate applied to the tiny group of users.

This is called base rate fallacy, and it's why screening for rare conditions almost always requires a second, confirmatory test — a single positive is much weaker evidence than the "99% accurate" label suggests.