1. (Bayesian). A new test for multinucleoside-resistant (MNR) human immunodeficiency virus type 1 (HIV-1) variants was recently developed. The test maintains 96% sensitivity, meaning that, for those with the disease, it will correctly report “positive” for 96% of them. The test is also 98% specific, meaning that, for those without the disease, 98% will be correctly reported as “negative.” MNR HIV-1 is considered to be rare (albeit emerging), with about a .1% or .001 prevalence rate. Given the prevalence rate, sensitivity, and specificity estimates, what is the probability that an individual who is reported as positive by the new test actually has the disease? If the median cost (consider this the best point estimate) is about $100,000 per positive case total and the test itself costs $1000 per administration, what is the total first-year cost for treating 100,000 individuals?

Solution:

Given,

sensitivity = 96%, specificity = 98% , false positive, P(FR) = 2%, chances of MNR-HIV-1, P(MNR) = 0.1%,

median cost per positive case = $100,000,

total first year for treating 100,000 = ?

\[\begin{equation} P(MNR|+) = \frac {P(+|MNR).P(MNR)}{P(+|MNR).P(MNR) + P(+|FR).P(FR)} \\ \\ = \frac {(0.96)*(0.001)} {(0.96*0.001) + 0.02*0.98} \\ = \frac {0.00096}{0.00096+0.0196} \\\\ = \frac {0.00096}{0.02056} \\\\ \approx 0.047 \end{equation}\]

Out of 100,000 individuals, the rate of being positive is 0.1% i.e. approximately 100 people will test positive

So, \[\text {The cost of testing 100,000 individuals costs is } 100,000*1000\] \[=100,000,000\] test for positive case \[\begin{equation} = 100 * 100,000 \\\\ = 10,000,000 \\\\ \end{equation}\]

So, \[\text {total cost} = 100,000,000+10,000,000\] \[ = \$ 110,000,000\] 2. (Binomial). The probability of your organization receiving a Joint Commission inspection in any given month is .05. What is the probability that, after 24 months, you received exactly 2 inspections? What is the probability that, after 24 months, you received 2 or more inspections? What is the probability that your received fewer than 2 inspections? What is the expected number of inspections you should have received? What is the standard deviation?

Solution:

Given,

Probability of joint commission inspection P(inspection) = 0.05 probability after 24 months = ?

n = 24

for exactly 2 inspection, \[\begin{equation} P(X=2) = ^nC_x P^x(1-P)^{n-2} \\\\ = ^{24}C_2(0.05)^2*(1-0.05)^{24-2} \\\\ =(0.05)^2*(0.95)^22 \\\\ \approx 0.246 \end{equation}\]

for 2 or more inspection \[\begin{equation} P(X \geq 2) = 1 - P(X \leq 2) \\\\ = 1 - P(X = 0) - P(X=1) \\\\ = 1 - (0.005)^0*0.95^{24} - (0.005)^1 * (0.95)^{23} \\ \approx 0.0317 \end{equation}\]

for less than 2 inpection, \[\begin{equation} P(x<2) = P(X=0) + P(X=1) \\\\ = (0.005)^0 * (0.95)^{24} + (0.005)^1 * (0.95)^{23} \\\\ \approx 0.754 \end{equation}\]

Standard Deviation \[\begin{equation} \sqrt{n*p*(1-p)} \\\\ \sqrt{24*0.05*0.95} \\\\ \approx 0.974 \end{equation}\]

  1. You are modeling the family practice clinic and notice that patients arrive at a rate of 10 per hour. What is the probability that exactly 3 arrive in one hour? What is the probability that more than 10 arrive in one hour? How many would you expect to arrive in 8 hours? What is the standard deviation of the appropriate probability distribution? If there are three family practice providers that can see 24 templated patients each day, what is the percent utilization and what are your recommendations?

Solution:

Given,

\[\mu = 10/hr\] \[ x_1 = 3 \]

So, \[\begin{equation} P(X = x_1) = \frac {\mu^x*e^{-\mu}}{x!} \\\\ P(X = 3) = \frac {10^3*e^{-10}}{3!} \\\\ = \frac {0.0454}{3!} \approx 0.00757 \end{equation}\]

For more than 10 patient in 1 hr,

\[\begin{equation} P(X>10) = 1 - P(X<10) \\\\ = 1 - e^\mu(\sum_{1-9} \frac {\mu^x}{x!}) \\\\ = 1 - e^{-10}(\frac {10^1}{1!} .... \frac {10^9}{9!}) \\\\ = 1 - e^{-10}(10085.57319) \\\\ = 1-0.4578 \approx 0.5421 \end{equation}\]

Again,

in 8 hrs,

\[\mu_2 = ? \\\\ \mu_1 = 10\\\\ t_2 = 8hrs\\\\ t_1 = 1 \] \[\begin{equation} \frac {\mu_2} {\mu1} = 8 \\\\ \frac {\mu_2}{10} = 8 \\\\ \mu_2 = 80 \end{equation}\]

So, we expect 80 patients in 8 hrs

\[\begin{equation} \sigma = \sqrt{\mu_2} \\\\ = \sqrt {80} \approx 8.94 \end{equation}\]

If there are 3 family practice provider care and see 24 template patient, the utilization is **3*24 = 72**

Currently there are only 10 patient / hr

so,

\[\frac {72}{10} = 7.2 hrs \text{ of patient}\] this means \[\frac {7.2}{24} = 30 \text{ % utilization}\]

Suggestion: increase number of patient that can be seen in each hour. Some new services can be added to attract more patient.

  1. Your subordinate with 30 supervisors was recently accused of favoring nurses. 15 of the subordinate’s workers are nurses and 15 are other than nurses. As evidence of malfeasance, the accuser stated that there were 6 company-paid trips to Disney World for which everyone was eligible. The supervisor sent 5 nurses and 1 non-nurse. If your subordinate acted innocently, what was the probability he/she would have selected five nurses for the trips? How many nurses would we have expected your subordinate to send? How many non-nurses would we have expected your subordinate to send?

Solution:

Given,

total population (N) = 30 total nurse (A) = 15 total # of employee picked for Disney vacation (n) = 6 total nurses picked out of 6 (k) = 5

we know that, \[\begin{equation} P(K=k) = \frac {{A \choose k} {N-A \choose n-k}} {N \choose n} \\\\ P(K = 5) = \frac {{15 \choose 5} {15 \choose 1}} {30 \choose 6} \\\\ = \frac {3003.15}{593775} \\\\ \approx 0.0758 \end{equation}\]

To find the expected # of nurses selected for a tryp, we can use fact that the expected value of a hypergeometric distribution is given by:

\[\begin{equation} E(X) = n * \frac {k} {N} \\\\ = 6 * \frac {15}{30} \\\\ = 3 \end{equation}\]

for non-nurse, we know that there are 15 staffs, so, \[\begin{equation} E(X_2) = n * \frac {(N-k)}{N} \\\\ = 6 * \frac {15}{30} = 3 \end{equation}\]

therefore, we would have expected the supervisor to send 3 non-nurses to the trip.

  1. The probability of being seriously injured in a car crash in an unspecified location is about .1% per hour. A driver is required to traverse this area for 1200 hours in the course of a year. What is the probability that the driver will be seriously injured during the course of the year? In the course of 15 months? What is the expected number of hours that a driver will drive before being seriously injured? Given that a driver has driven 1200 hours, what is the probability that he or she will be injured in the next 100 hours?

Solution:

In this case, p = 0.001 and X is the number of hours the driver must drive before being seriously injured. The distribution is geometric because each hour is independent of the others and has the same probability of a serious injury occurring.

The probability of the driver not being seriously injured in one hour is \[1 - p = 0.999 \]

Therefore, the probability of the driver not being seriously injured in 1200 hours is \[(0.999)^{1200} \approx 0.818\] This means that the probability of being seriously injured at some point during the year is approximately \[1 - 0.818 = 0.182 \text { or 18.2%} \]

To find the probability of being seriously injured during the course of 15 months, we need to calculate the probability of not being seriously injured in the first 1080 hours (which is 9 months) and being seriously injured in the next 360 hours (which is 3 months). The probability of not being seriously injured in 1080 hours is \[(0.999)^{1080} \approx 0.361\]

Therefore, the probability of being seriously injured in the next 360 hours is \[1 - 0.361 = 0.639 \text { or about 63.9%}\]

The expected number of hours a driver will drive before being seriously injured is 1/p, which is 1000 hours.

Given that a driver has driven 1200 hours, the probability of being seriously injured in the next 100 hours is the probability that the first serious injury occurs between hour 1201 and hour 1300. This probability can be calculated as:

\[\begin{equation} P(1201 \leq X \leq 1300 | X > 1200) = [1 - (1 - p)^{100}] / (1 - (1 - p)^{1200}) \\\\ \approx 0.009 \end{equation}\]

Therefore, the probability of being seriously injured in the next 100 hours, given that the driver has driven 1200 hours, is approximately 0.009, or about 0.9%.

  1. You are working in a hospital that is running off of a primary generator which fails about once in 1000 hours. What is the probability that the generator will fail more than twice in 1000 hours? What is the expected value?

This situation can be modeled using a Poisson distribution with \[\lambda = 1/1000 \text{ failures/hour.}\] The probability of the generator failing more than twice in 1000 hours is: \[\begin{equation} P(X > 2) = 1 - P(X \leq 2) \\\\ = 1 - (P(X=0) + P(X=1) + P(X=2)) = 1 - (e^{(-\lambda)} * (\frac {\lambda^0} {0!}) + e^{(-\lambda)} * (\frac {\lambda^1}{1!}) + e^{(-\lambda)} * (\frac {\lambda^2} {2!})) \\\\ = 1 - (e^{(-1)} * (\frac {1^0} {0!}) + e^{(-1)} * (\frac {1^1}{1!}) + e^{(-1)} * (\frac {1^2}{2!})) \\\\ = 1 - (0.3679 + 0.3679 + 0.1839) \\\\ \approx 0.0803 \end{equation}\]

So the probability of the generator failing more than twice in 1000 hours is 0.0803.

The expected value for the number of failures in 1000 hours is:

\[\begin{equation} E(X) = \lambda * t = (1/1000) * 1000 = 1 \end{equation}\]

So we can expect 1 failure in 1000 hours.

  1. A surgical patient arrives for surgery precisely at a given time. Based on previous analysis (or a lack of knowledge assumption), you know that the waiting time is uniformly distributed from 0 to 30 minutes. What is the probability that this patient will wait more than 10 minutes? If the patient has already waited 10 minutes, what is the probability that he/she will wait at least another 5 minutes prior to being seen? What is the expected waiting time?

Solution:

The waiting time for the surgical patient is uniformly distributed from 0 to 30 minutes. Let X denote the waiting time in minutes.

The probability that the patient will wait more than 10 minutes is:

\[\begin{equation} P(X > 10) = \frac {(30 - 10)} {30} = \frac {2}{3} = 0.6667 \end{equation}\]

So the probability that the patient will wait more than 10 minutes is 0.6667.

If the patient has already waited 10 minutes, the remaining waiting time is uniformly distributed from 0 to 20 minutes. Let Y denote the remaining waiting time in minutes.

The probability that the patient will wait at least another 5 minutes prior to being seen is:

\[\begin{equation} P(Y \geq 5 | X = 10) = P(X + Y \geq 15 | X = 10) = P(Y \geq 5) = \frac {(30 - 5)}{30} = \frac {5}{6} \approx 0.8333 \end{equation}\]

So the probability that the patient will wait at least another 5 minutes prior to being seen, given that they have already waited 10 minutes, is 0.8333.

The expected waiting time is:

\[E(X) = \frac{(0 + 30)} {2} = 15\]

So the expected waiting time is 15 minutes.

  1. Your hospital owns an old MRI, which has a manufacturer’s lifetime of about 10 years (expected value). Based on previous studies, we know that the failure of most MRIs obeys an exponential distribution. What is the expected failure time? What is the standard deviation? What is the probability that your MRI will fail after 8 years? Now assume that you have owned the machine for 8 years. Given that you already owned the machine 8 years, what is the probability that it will fail in the next two years?

Solution:

Since the failure of MRIs obeys an exponential distribution, the probability density function (PDF) for the lifetime of the MRI is given by:

\[f(x) = \lambda*e^{-λx}\]

where λ is the failure rate (i.e., the number of failures per unit time) and x is the time.

Since the expected lifetime of the MRI is 10 years, we know that λ = 1/10 = 0.1 per year.

The expected failure time (i.e., the mean) is given by:

E(X) = 1/λ = 10 years

The standard deviation is given by:

SD(X) = 1/λ = 10 years

The probability that the MRI will fail after 8 years is given by:

\[P(X > 8) = \int_8^\infty \lambda e^{-\lambda x} dx = e^{(-0.8)} \approx 0.4493\]

Now assume that you have owned the machine for 8 years. Given that you already owned the machine for 8 years, the probability that it will fail in the next two years is given by:

\[\begin{equation} P(X \leq 2 | X > 8) = \frac {P(8 < X \leq 10)} {P(X > 8)} \\\\ = \frac {\int_8^{10} \lambda e^{-\lambda x} dx} {\int_8^\infty \lambda e^{-\lambda x} dx} \\\\ = \frac{ [e^{(-0.8)} - e^{(-1)}]} {e^{(-0.8)}} \\\\ \approx 0.2587 \end{equation}\]