UW Notes - 6 - Public Health
UW Notes - 6 - Public Health
UW Notes - 6 - Public Health
1
Malpractice: it is a legal term not medical error, occur when lower than
standard of care medication given to the patient. (it is not a category of medical
error)
Biostatistics
Mortality
Case fatality rate: can be used as measure of severity of the disease. Calculated as
proportion fatal cases / total number of cases (fatal / all)
Calculating probability:
- Independent event probability of getting all same result = multiply
- While probability of at least 1 event different = 1 – P
- Probability of getting 3 heads in row → 0.5 X 0.5 X 0.5 = 0.125 , the probability
that at least on of them is tail = 1 – 0.125
What is meant by attack rate in case of outbreak investigation
- Number become ill / total number exposed to this risk
How to calculate cumulative incidence:
- Total number of new cases of a disease over specific period divided by
number of people at risk (number of already diseased are not
included) at the beginning of the period
- N.B. deaths are not considered under any cause (they are already included in
people at risk)
- N.B2 Time is not incorporated at the denominator, unlike incidence
Incidence: number of new cases / population at risk over a period of time
Prevalence: point prevalence as opposed to period prevelance
Prevalence = incidence X duration of disease
↑↑ survival from a disease will ↑↑ prevelance with no effect on
incidence
2
Meaning of RRR → e.g. RRR = 50% means that a drug ↓↓ the incidence of disease
from 50% to 25%
How to calculate NNT (number needed to treat):
Number of patients needed to be treated to prevent an additional adverse event
We use absolute risk reduction
NNT= 1/ ARR
The ideal NNT is 1 → all patients in treatment group will benefit from the
treatment , so NNT the lower, the better
What is the meaning of odds ratio and How to calculate it ∷
Odds of event = probability (not ratio) of the event happening / probability of
event not happening
Used in case control studies, with outcome occurring based on exposure status.
First do 2 X 2 table and take care
= الطولoutcome = العرضexposure
A B
C D
Take care of the site of letters
Calculate odds (probability) of event to occur → a/b (not a / (a+b)
Odds of event not to occur → c / d
For simplification odds ratio = AD / BC
When odds < 1 → risk of developing adverse effect is lesser in exposure than non
exposure
How to calculate NNH (number needed to harm):
Number of patients needed to be treated before an additional adverse event
occurs
3
Simply, we use absolute risk increase (attributable risk)
To calculate absolute risk increase → you must calculate the absolute event rate in
both groups (number of adverse event / total number treated), and subtract both
numbers
NNH = 1 / absolute risk increase (if NNH = 40, so we need to treat 40 patient
before we see additionl one adverse effect)
4
- If 95% CI contain null value → so P-value > 0.005 → statistical
insignificance
- If 95% CI Don’t contain null value → P-value < 0.005 → statistical significance
Ecological study: when the frequency of a given character and outcome are studied
using population data not individual data, these studies geerate hypothesis
and associations but unable to make conclusions regarding individuals (ecological
fallacy)
Cross over study:
- the patients serve as their own controls
- Cons: effects of onettt may carry oer and alter the effect o fthe subsequent
treatment → so washout phase with no drug is advised and it is long
enough
Case (diseased) control (non diseased) recruitment not recruit about risk
stratification which is cohort, so first of all select subject according to disease not
5
risk exposure, also cases and controls are matched, the subject selection don’t
depend on exposure
6
Chi-square test:
- Need two categorical variables (both dependent and independent)
Analysis of Variance (ANOVA):
- Used to compare ≥ 2groups of quantitative variables (by comparing means in
them)
Importance of two sample t-test:
- Used if the dependent variable is quantitative one (≥ 2 population → use ANOVA)
- Used to determine if the means of 2 populations are equal or not
- The basic requirements → mean values, sample variances, sample sizes
- Calculate t statistic → P-value if <0.005 → reject the null hypothesis and the two
means are from different populations, and the 2 means are statistically significant
7
Effect modification:
- effect modification appear when external variable affect the previous
association when stratified against(outcome is modified by another variable), so
effect modifier; initially there is association and after stratification there is
also difference measures of association either positive or negative
8
- Recall bias: result from inaccurate recall of past exposure; mostly occur in
retrospective studies (as case control study). this bias type is reduced by
prospective studies
- Observer bias (expectancy bias):
o occur when the investigator (observer) misclassify data due to preconceived
expectations or prior knowledge about the study
o this important when the outcomes are subjective (radio, patho
interpretation)
o this bias can be reduced by blinded study in which the observers are
unaware of study details, and multiple observers encode and verify the
recorded data
- Observer bias effect (Hawthorne effect): tendency of the study subjects to
change his behavior as a result of awareness that they are being studied, common
in behavioral outcomes / changes → affect the validity of the study
o Study subjects can be kept unaware that they are being studied, but this can
occasionally pose ethical problems
- Lead time bias: should be considered when evaluating any screening test.
o Lead time → time between initial detection of the disase and specific
outcome / end-point
o Lead time bias is apparent ↑↑ in survival time among patients undergoing
screening when they actually have unchanged prognosis, this only occur
because The diseases is detected earlier
o To detect the effectiveness, you should follow up the patients for longer
periods than the apparent ↑↑ in survival an compare mortality.
9
What is the difference between sample mean & population mean, how to calculate
variance in both:
- When applying rules of normal distribution curve, mean is the average of results
of the sample, while SD is the distribution of results around that mean
- When 95% lies between ± 1.96 SD → I mean that 95% of result lies between
(mean± 1.96 * SD)
10
- But this rule cannot be applied on distribution of the population, as the sample
mean is different from the true population mean (the variability between
sample means)
- So how to calculate variability around the unknown true population mean?? It is
by usage of standard error (SE) =
SD/√n (n is size of sample)
e.g. (sample mean ± 2 * SE) means
that I am sure by 95% that the true
population mean lies at this range
→ meaning of confidence interval
- So confidence interval of mean =
mean ± [z-score] * SE (SD/√n)
- Sample size (n), SD of the sample
determines the magnitude of
variability due to sampling. So ↑↑
sample size → narrower & precise
CI, ↑↑ SD → ↑↑ SE → wider CI
Remember that above 2 SD → there
is only 2.5% of sample size, while 5%
are lie outside the 2 SD
- α value typically = 0.05, it is the P-value, not directly affected by sample size
- β value related to study power (power = 1 - β):
o ↑↑ sample size → ↑↑ study power (ability of the study to detect difference
when there is true difference)
o ↓↓ sample size → ↓↓ power → ↑↑ β → type II error
12
Reliability & accuracy:
- PRrecision (Reliabilty) → Repeat exam show the same results (Reproductive)
- Accuracy (Validity) → if compare to the gold standard test, it have equivalent
result (measure what is supposed to measure)
- Poor accuracy or prescision can limit (low) the specificity and sensitivity of the
tests if results from health and diseased are so close
13
Ethics
Patient confidentiality must be respected at all levels, in medical and non-medical
situations, with colleagues or not, even telling your colleague that someone is your
patient. If you want to talk with another doctor in the hospital, make sure that you
use private area (even in the hospital)
Physicians should always provide the life threatening therapy to minor in emergency
regardless of parents’ wishes
Effective discharge planning:
- During hospitalization, social worker can contact the involved family members
and assess whether there are any social factors affecting the patient’s ability to
return home and develop strategy for effective discharge
- It needs the interdisciplinary collaboration among the social workers, nurse, and
physician.
- You should take into account the cognitive status, activity, family support
What is the best communication model developed in medical practice:
- Communication error (vague, unclear commands) are one of the most common
factors involved in malpractice claims of medical errors\
- The best approach is closed loop communication → the doctor transmit the
message to nurse, the nurse should repeat the message again in clear sentnces,
the doctor should confirm with yes
Certification of disability:
- Disability is insurance with financial support, certification of disability must be
signed by the physician before the benefit is granted
- Take care of possible malingering, exaggeration to obtain benefits
- You must take detailed History, examination, testing to assess the
condition and don’t depend on the previously signed one
14
How to deal with patient in acute intoxication:
- patient autonomy for deciding treatment & management must be in a decision
making capacity → i.e. patient must communicate a choice, understand
information given, appreciate consequences.
- Intoxicated patients or blunted patients have temporarily lack capacity so
1) Never let the patient leave the ED as it may endanger his life
2) Keep the patient and reassess him when he is clinically sober as reassess his
descicion making capacity
3) After the patient gain the capacity, if he refuse the treatment → discharge the
patient
Release of medical information:
- HIPAA state that release of medical info must be done only after verbal or
(better) written authorization for relase of information the family member.
- This authorization must be privately discussed, preferred to be written.
- If the patient is incapacitated (ori ED situation); only basic information can be
shared (like; the patient is stable now)
How to reduce the “wrong site surgery” error:
- Causes by failure to mark the site, emergency operations, poor communication,
surgeon fatigue
- To reduce this most frequent error → marking the operative site,
independent verification
- Independent verification of the patient, site, procedure mustbe done independely
by 2 HCW, also use of surgical timeout is important
It is the patient right to refuse receive diagnostic information and must be
respected by the doctor, the physician must communicate with the patient why he
ask this
What is the appropriate decision if there is medical error occur (regardless
presence of harm)
- Full disclosure of the medical error should occur in timely manner → associated
with neutral - positive effect on patient relationship
15
- The disclosure should include apology for what occurred
Discharge checklist:
- Transition from the hospital to outpatient is high risk for the patients, so the
most effective strategy for decreasing the adverse outcome and avoid
readmission in discharge checklist
- It is different from discharge summary which is written by the hospital and
directed to other outpatient physician and written in difficult technical language
- Discharge list should e written in appropriate language and take patient education
and literacy into account, it huld include:
1) Detailed instructions about medications
2) Follow up appointments
3) Any medication changes
Advance directive:
- Advance care planning should begin by conversation between the physician and
patient in outpatient care
- As a part of admission process (either emergency / ward) → hospital should
inquire about the advance directive in the event that the patient become unable to
make decisions
- Advance directive are consists of : living will: patient’s end of life wishes about
resuscitation, intubation … etc. health care proxy: allow the patient to
designate individual to take the decisions according to living will
Sequence of patients’ informed consent / making medical decisions:
1) The best, is from patient himself
2) Advance directives, health care proxy
3) Family member (act as surrogate decision maker)
4) If no family member, person who care about and knows the patients’ wishes → if
there is struggle between family member → ethical committee → last resort, court
All adolescent visits should include an opportunity to interview the patient alone to
discuss topics such as drugs, alcohol, tobacco and sexual activity
If you suspect child abuse, take History with open ended questions then ask
specific questions, the next step usually to interview the child alone and properly
examine him and take detailed history … if suspect abuse call child protective
services immediately
16
- The first step usually to collect data by interviewing multiple staff members about
the events → after gathering informations, key solution may be important after
determine the cause
Transcription errors are preventive medical error, largely reduced by educations to
reduce use of abbreviations and avoid trailing zero e.g. use “2 mg not 2.0 mg”
What is the proper response to history of sexual abuse:
- 2 things must occur simultaneously, empathy and support
- First, acknowledge the sexual abuse (this help the patient fell that she is
understood better)
- Second; gently ask the patient if she would like to discuss it further (if the patient
not ready to discuss that → convey willingness to discuss it when the patient is
ready)
What is the proper managemt of an emergency case with NO money:
- According to Law EMTALA, three primary requirements in hospitals that provide
emergency services.
1) Provide appropriate screening medical exam to anyone who comes to ED
seeking medical care
2) Stabilize & treat the emergency situation
3) Not transfer an individual with emergency medical condition that has not been
stabilized
Medical history taking, should start wiith open ended questions that make the
patient answer and describe the condition in his own word without any interference
from the doctor
Types of prevention in heath
care:
- Health promotion → Process
of enabling people to ↑↑
control over their health and
its determinants, it is type of
primary prevention
Never be judgmental when
taking with patient, always show
empathy, acknowledge, maintain
confidentiality and autonomy
18
Contact precautions with infectious diseases:
- Standard precautions (for all patients) → handwashing, proper disposal +
- Suspected C.difficile handwashing by soap & water as alcohol don’t kill
the spores
- Contaminated secretions gown for any patient contact, non-sterile gloves
- Dedicated devices as BP cuff …
- Simple facemask → for infections transpmitted by droplets >5μ for 6 – 10 ft.
19
- Burnout → emotional exhaustion, ↓↓ sense of personal accomplishment which
lead to medical errors, but here errors result from lack of concern and callousness
toward patients
- Fatigue → sleep deprivation which may lead to forgetting do job.
o ≥ 17 hours of wakefulness, impairment of cognitive performance as those
who are seen in alcohol intoxication
Dealing with low level literate patient:
- These patients always have lower quality of medical care, as they fail to
understand both written & spoken language and medical device.
- Detecting those patients is difficult as they are always ashame, should be very
suspicious
- The physician should have alternative modes of learning those
patients like videos or drawing is the best ones
Normal stages of grief in terminal illness:
- not all patients will pass thorugh all of them
- if denial is significant (interfere with care) → should be confronted
21
- To reduce these risks, structured handoffs that include key elements have been
shown to reduce adverse effects
- Key element e.g. systemic procedure for sign-out, checklists of tasks that need to
be completed, standardization approach, don’t overwhelm with many details
rather be systematic
Developmental mile stones in toddlers:
What are circumstances in which minors (<18 y) can provide their own consent
22
What is the proper response in the following situation:
“if one’s physician disagree with another’s practices & physicians”
- If the practice within the standard of care → never criticize the physician in
front of the patient
- If the practice is grossly negligent or treatment far outside the standard
of care (dangerous)→ the doctor should be criticized
- Privately discuss the patient with the referring physician to understand the
reason for his medications and explain the change
23
“ senior attending order a wrong drug to the patient”
- As art of education process it is essential to understand the clinical
reasoning behind the team member decisions, as it is the best interest for
both junior & senior staff
- So respectfully discuss the issue directly and ask why the decision
was made, don’t order the medication until you ask him
“the intern should take informed consent about procedure he didn’t know about”
- Informed consent is not a paper needed to be signed, it is a dialogue between the
provider and patient about the procedure
- The doctor must inform the patient about risk, benefits, alternative treatment
- The ideal physician who take informed consent is the one who will do the
procedure
24
“you are an attractive doctor, I wish to go for date with you”
- Romantic and sexual relationships with current patients are always unethical
due to potential interference with the physician role as a doctor
- Romantic relationships between doctors (non-psychiatrists)– patients (after
termination the doctor patient relationship) may be acceptable, but it is
not the role
25
- Before helping your friend, consider ethical issues. There are potential problems
that results from inadequate assessment.
Each person should have monthly premium ( )قسط شهريto cover the insurance
plans, but expenditures are usually higher than those premium
So health insurance try to ↓↓ expenditures to be lower than premiums by 1)
patient share of cost for the services they receive (copayment, deductibles) 2)
limiting range of services the patient receive
26
Health maintenance organization (HMO) plan:
o Low monthly premium, low copayments, low cost for the patients تأمين صحي
مصري اصيل
o It reduce utilisations by some conditions:
1) Confiding patients to a limited panel of providers (cannot go outside it)
2) Specialized consultation need referral from primary care provider
3) Any service don’t meet EBM → denied by health insurance
27