epidemiological
Content
Clinical Research Methods: Epidemiologic Studies
Epidemiology
Greek: EPI - Upon DEMOS - People LOGOS - Study of, Body of Knowledge
Definitions Of Epidemiology
• The study of the distribution and determinants of health-related states or events in specified populations, and the application of this study to control of health problems.
Epidemiology Has The Methodology To:
• Determine the extent of disease in the community • Study the natural history and prognosis of disease • Identify associations and potential etiology (causes) of a disease and risk factors for disease • Evaluate new preventive and therapeutic measures and new modes of health care delivery • Provide a foundation for developing public policy and regulatory decisions relating to environmental problems.
Descriptive Epidemiology
Purpose: To characterize the amount and distribution of disease within a population. In other words …. To identify health problems and patterns of disease that exist. Descriptive studies generally precede analytic studies designed to investigate determinants of disease. Thus, descriptive studies often help to generate research hypotheses.
Prevalence of High Blood Pressure in Americans Age 20 and Older by Age and Sex
90 80
70.3
59.9 68.8
84.1
Percent of Population
70 60 50 40 30 20 10
0 17.1 16 32.3 44.1
53
Men Women
30.5
0
0
20-34
NHANES IV: 1999-2000
35-44
45-54
55-64
65-74
75+
Ages
Note: “0.0” = data not available. Prevalence estimates for women ages 20-34 are considered unreliable.
Source: Health, United States, 2003, CDC/NCHS
Of Those With HBP
•
1/3 are unaware they have it 1/3 are on medication but uncontrolled, or aren’t on medication 1/3 are on medication and have it controlled
•
•
Analytic Epidemiology ------------Randomized trial Cohort Case-control Case-crossover Cross-sectional Case reports/series Observational Experimental
Association
Statistical dependence between two variables: • Exposure (risk factor, protective factor, predictor variable, treatment) (disease, event)
• Outcome
Advantages and Disadvantages Of The Case-Control Method
Advantages
Well suited to the study of rare diseases or those with long latency. Relatively quick to mount and conduct. Relatively inexpensive. Existing records can occasionally be used.
Disadvantages
Validation of information is difficult or sometimes impossible. Control of extraneous variables may be incomplete. Selection of an appropriate comparison group may be difficult.
Historical Cohort Cases Assembled Follow-up Concurrent Cohort Cases Assembled
Figure 5.2.
Follow-up
Historical and concurrent cohort studies
Advantages and Disadvantages Of Cohorts
Advantages
Less possibility of bias in obtaining information on exposure (e.g. recall bias) May allow use of concurrent measurements of exposure rather than historical estimates. Reveals potentially important information about the temporal relationship between exposure and disease
Disadvantages
May require a long time and is not suited to exposures that have long latency periods before the onset of disease. Usually very costly. Problem of loss of subjects during follow-up.
Measures Of Effect
• Risk Difference (RD) • Relative Risk (RR) ----Risk Ratio (RR) Rate Ratio (RR)
• Odds Ratio (OR)
Evaluating Associations
If we observe an exposure/disease association, we must consider:
1. Is the association valid? (do the study findings reflect the true relationship between the exposure and disease?) Is the association causal? (Is there sufficient evidence to infer that a causal association exists between the exposure and the disease?)
2.
Evaluating Associations
EVALUATING THE VALIDITY OF AN ASSOCATION: In any epidemiologic study, there are at least 3 possible explanations for the observed results: 1. CHANCE 2. BIAS 3. CONFOUNDING These explanations are not mutually exclusive -more than one can be present in the same study
Bias
BIAS: Systematic error in the design, conduct, or analysis of a study that results in a mistaken estimate of an exposure/disease relationship 1. SELECTION BIAS 2. INFORMATION BIAS * * * * Interviewer Recall Bias Reporting Bias Surveillance Bias
Confounding
DEFINITION: A third variable (not the exposure or outcome variable of interest) that distorts the observed relationship between the exposure and outcome. • Confounding is a confusion of effects that is a nuisance and should be controlled for if possible. • Age is a very common source of confounding.
Evaluating Causal Associations
CAUSATION: A philosophical concept merged with practical guidelines
• The presence of a valid statistical association does not imply causality • Biologic plausibility • Temporality • Consistency
Clinical Trials
I.
Definition:
A planned experiment in humans designed to assess the efficacy and safety of an intervention in one group with those observed in a comparable group receiving a control intervention concurrently. The ideal clinical trial is one that is randomized and double-blind.
Effectiveness of inpatient and outpatient treatment strategies among subgroups of women with pelvic inflammatory disease from the PEACH Randomized Trial
Roberta B. Ness, MD, MPH Gail Trautmann, MA Holly E. Richter, PhD, MD Hugh Randall, MD Jeffrey F. Peipert, MD, MPH Deborah B. Nelson, PhD
Diane Schubeck, MD S. Gene McNeeley, MD Wayne Trout, MD Debra C. Bass, MS David E. Soper, MD
Etiologic Agents
A. N. gonorrhoeae B. C. trachomatis C. Mixed aerobes and anaerobes
Why worry about PID
Morbid Common Difficult to accurately diagnose
PID and Infertility
One fifth of women who have at least one episode of PID may become infertile Half of women with three or more episodes of PID may become infertile
PID and Chronic Pelvic Pain
One fourth of women with acute PID may experience subsequent chronic pelvic pain
PID and Ectopic Pregnancy
One in ten women may have an ectopic pregnancy in their first pregnancy following PID Tubal pregnancy is the leading cause of firsttrimester, pregnancy-related deaths in American women
Prevalence of PID
8% of all American women of reproductive age are reported to have received treatment for PID in 1995 Over one million American women seek treatment for PID annually
“…Then in 1975, I had a bout of painful cramps which, as it turned out, were caused by a severe infection. My horrified gynecologist took out the IUD and prescribed large doses of antibiotics. He also warned me that I might have trouble having children.”
From: “A cash settlement but no apology” NY Times 2/99
“He was right, though I didn’t find out for more than a decade. I had to fall in love, get married and spend a few years trying to conceive before an X-ray showed that my fallopian tubes were scarred and completely blocked. My husband and I tried in-vitro fertilization three times but were unsuccessful in having children.”
From: “A cash settlement but no apology” NY Times 2/99
PID Evaluation and Clinical Health (PEACH) Study
Study Effectiveness of Outpatient Treatment of PID
Importance The first randomized clinical trial to evaluate the effectiveness of the most common treatment for PID
Eligibility of Patients into the Study
Inclusion Exclusion
(Must meet all of the following criteria) (Must not meet any of the following criteria) 1. 37 years of age or younger 1. Currently pregnant by urine testing 2. Willing to participate 2. Tubo-ovarian abscess 3. Presenting with a history of pelvic 3. Appendicitis, hemorrhagic ovarian cyst discomfort for <30 days. (This does or other condition requiring surgery by not need to be the chief complaint) ultrasound or laparoscopy 4. Experiencing pelvic organ 4. Nausea and vomiting after a trial of tenderness (uterine or adnexal metoclopramide tenderness) on bimanual 5. Antimicrobial therapy within the past 7 examination days 5. Leukorrhea and/or Mucopurulent 6. Delivery, abortion or gynecologic Cervicitis and/or untreated known surgery within the past 30 days +GC or CT 7. Prior hysterectomy or bilateral salpingectomy *Leukorrhea: white cells > epithelial 8. Allergy to penicillins, cephalosporins or cells in 4 fields at high power. Can tetracyclines use LPF to find fields with 9. Homeless substantial WBC’s.
Study Treatments
Inpatient 2g Cefoxitin IV every 6 hours 100mg Doxycycline IV every 12 hours Outpatient 2g Cefoxitin > 1 shot 1g Probenecid 100 Doxycycline oral twice a day
Outcomes
Time to Pregnancy
Outcomes
Short Term: ■ Time to Clinical Improvement ■ Microbiologic Cure on Repeat Cervical Culture and Endometrial Biopsy ■ Treatment Adherence Long Term: ■ Infertility ■ Tubal Occlusion of Women with Involuntary Infertility ■ Repeat Episodes of PID ■ Ectopic Pregnancy ■ Functional Decline due to Pelvic Pain ■ Frequency of Health Services use and Indirect PID-Related Costs; Cost Utility Analysis
PEACH: Long-term outcomes by treatment group
Outpatient n Pregnancy† Infertile‡ Self-reported recurrent pelvic inflammatory disease† Ectopic pregnancy† Tubal obstruction§ Chronic Pelvic Pain|| 51 4 7 128 12.4% 1.0% 41.2% 33.7% 66 16.6% 1 110 0.3% 29.8% 4 33.3% P = .11 P = .37 P = .72 P = .27 0.69 (0.43 to 1.09) 3.66 (0.40 to 33.12) 0.61 (0.10 to 3.63) 1.24 (0.87 to 1.77) 172 71 % 42.0% 18.4% n Inpatient % Statistical significance P = 1.00 P = .85 Adjusted OR (95% CI)* 1.00 (0.71 to 1.39) 1.32 (0.86 to 2.04)
166 41.7% 67 17.9%
*Odds ratios were adjusted for tubal ligation (yes/no), bacterial vaginosis (yes/no), and IUD in place (yes/no).
Source: Ness RB, et al. Effectiveness of inpatient and outpatient treatment strategies for women with pelvic inflammatory disease: Results from the Pelvic Inflammatory Disease Evaluation and Clinical Health (PEACH) Randomized Trial. Am J Obstet Gynecol 2002;186:929-37.
PEACH: Secondary Results
No Difference between treatment groups: ■ Among women with endometritis ■ Among women without previous PID ■ Among women who got IV doxycycline as inpatients
Barrier contraceptive use and relative risks for recurrent PID, chronic pelvic pain and infertility after PID.
Recurrent PID
(N) (%)
Condoms No 324 (16.7) 1.0 0.8 (0.5-1.5) 0.5 (0.3-0.9) 300 (36.7) 132 (54.5) 142 (31.0) 0.8 (0.5-1.3) 187 (26.7) 0.7 (0.5-1.2) 1.0 59 (45.8) 0.8 46 (34.8) 0.6 0.7 (0.4-1.5) 0.4 (0.2-0.9)
Chronic Pelvic Pain
(N) (%) Adjusted RR (N) (%)
Infertility
Adjusted RR
Adjusted RR+
< 5/10 times 156 (16.0) > 6/10 times 204 (8.8)
+ Adjusted for age, number of live births, race, non-monogamy at baseline, new partner at baseline, gonococcal or chlamydial cervicitis at baseline, education, number of study visits, and all other forms of contraception other than that under consideration ++ Diaphragms, spermicides, cervical cap, or female condom Source: Ness RB. Barrier contraception and the risk of recurrent pelvic inflammatory disease (PID), chronic pelvic pain, and infertility, following PID. Am J Pub Health, 2004.
Risk of Reproductive Outcomes Flollowing Endometritis and/or Upper Genital Tract Infection (UGTI)a or No Endometritis/No UGTI Among Women with Clinical PID
No Endometritis/ No UGTI (n=258) Outcome Pregnant Infertile Recurrent PID Chronic Pelvic Pain N 103 50 50 112 % 40.2% 17.8% 19.5% 44.4% Endometritis and/ or UGTI (n=356) N 142 53 46 98 % 40.8% 16.4% 13.2% 29.5% 0.8 (0.6-1.2) 1.0 (0.6-1.6) 0.6 (0.4-0.9) 0.6 (0.4-0.9)
b
Adjusted OR (95% CI)
a UGTI was defined as isolation of gonorrhea and/or chlamydia from endometrium. b Adjusted for age, PID history, race, and education. Pregnancy and infertility are additionally adjusted for reported baseline
infertility (one year of unprotected intercourse without resulting pregnancy)
Effect of Chronic Pelvic Pain Intensity on Physical and Mental Health (Means + sd are presented).
Composite
Physical Health
110 100 90 80 70 60 50 N=
346
No CPP
79
Mild-Mod
123
Mod-Severe
Epidemiology
Greek: EPI - Upon DEMOS - People LOGOS - Study of, Body of Knowledge
Definitions Of Epidemiology
• The study of the distribution and determinants of health-related states or events in specified populations, and the application of this study to control of health problems.
Epidemiology Has The Methodology To:
• Determine the extent of disease in the community • Study the natural history and prognosis of disease • Identify associations and potential etiology (causes) of a disease and risk factors for disease • Evaluate new preventive and therapeutic measures and new modes of health care delivery • Provide a foundation for developing public policy and regulatory decisions relating to environmental problems.
Descriptive Epidemiology
Purpose: To characterize the amount and distribution of disease within a population. In other words …. To identify health problems and patterns of disease that exist. Descriptive studies generally precede analytic studies designed to investigate determinants of disease. Thus, descriptive studies often help to generate research hypotheses.
Prevalence of High Blood Pressure in Americans Age 20 and Older by Age and Sex
90 80
70.3
59.9 68.8
84.1
Percent of Population
70 60 50 40 30 20 10
0 17.1 16 32.3 44.1
53
Men Women
30.5
0
0
20-34
NHANES IV: 1999-2000
35-44
45-54
55-64
65-74
75+
Ages
Note: “0.0” = data not available. Prevalence estimates for women ages 20-34 are considered unreliable.
Source: Health, United States, 2003, CDC/NCHS
Of Those With HBP
•
1/3 are unaware they have it 1/3 are on medication but uncontrolled, or aren’t on medication 1/3 are on medication and have it controlled
•
•
Analytic Epidemiology ------------Randomized trial Cohort Case-control Case-crossover Cross-sectional Case reports/series Observational Experimental
Association
Statistical dependence between two variables: • Exposure (risk factor, protective factor, predictor variable, treatment) (disease, event)
• Outcome
Advantages and Disadvantages Of The Case-Control Method
Advantages
Well suited to the study of rare diseases or those with long latency. Relatively quick to mount and conduct. Relatively inexpensive. Existing records can occasionally be used.
Disadvantages
Validation of information is difficult or sometimes impossible. Control of extraneous variables may be incomplete. Selection of an appropriate comparison group may be difficult.
Historical Cohort Cases Assembled Follow-up Concurrent Cohort Cases Assembled
Figure 5.2.
Follow-up
Historical and concurrent cohort studies
Advantages and Disadvantages Of Cohorts
Advantages
Less possibility of bias in obtaining information on exposure (e.g. recall bias) May allow use of concurrent measurements of exposure rather than historical estimates. Reveals potentially important information about the temporal relationship between exposure and disease
Disadvantages
May require a long time and is not suited to exposures that have long latency periods before the onset of disease. Usually very costly. Problem of loss of subjects during follow-up.
Measures Of Effect
• Risk Difference (RD) • Relative Risk (RR) ----Risk Ratio (RR) Rate Ratio (RR)
• Odds Ratio (OR)
Evaluating Associations
If we observe an exposure/disease association, we must consider:
1. Is the association valid? (do the study findings reflect the true relationship between the exposure and disease?) Is the association causal? (Is there sufficient evidence to infer that a causal association exists between the exposure and the disease?)
2.
Evaluating Associations
EVALUATING THE VALIDITY OF AN ASSOCATION: In any epidemiologic study, there are at least 3 possible explanations for the observed results: 1. CHANCE 2. BIAS 3. CONFOUNDING These explanations are not mutually exclusive -more than one can be present in the same study
Bias
BIAS: Systematic error in the design, conduct, or analysis of a study that results in a mistaken estimate of an exposure/disease relationship 1. SELECTION BIAS 2. INFORMATION BIAS * * * * Interviewer Recall Bias Reporting Bias Surveillance Bias
Confounding
DEFINITION: A third variable (not the exposure or outcome variable of interest) that distorts the observed relationship between the exposure and outcome. • Confounding is a confusion of effects that is a nuisance and should be controlled for if possible. • Age is a very common source of confounding.
Evaluating Causal Associations
CAUSATION: A philosophical concept merged with practical guidelines
• The presence of a valid statistical association does not imply causality • Biologic plausibility • Temporality • Consistency
Clinical Trials
I.
Definition:
A planned experiment in humans designed to assess the efficacy and safety of an intervention in one group with those observed in a comparable group receiving a control intervention concurrently. The ideal clinical trial is one that is randomized and double-blind.
Effectiveness of inpatient and outpatient treatment strategies among subgroups of women with pelvic inflammatory disease from the PEACH Randomized Trial
Roberta B. Ness, MD, MPH Gail Trautmann, MA Holly E. Richter, PhD, MD Hugh Randall, MD Jeffrey F. Peipert, MD, MPH Deborah B. Nelson, PhD
Diane Schubeck, MD S. Gene McNeeley, MD Wayne Trout, MD Debra C. Bass, MS David E. Soper, MD
Etiologic Agents
A. N. gonorrhoeae B. C. trachomatis C. Mixed aerobes and anaerobes
Why worry about PID
Morbid Common Difficult to accurately diagnose
PID and Infertility
One fifth of women who have at least one episode of PID may become infertile Half of women with three or more episodes of PID may become infertile
PID and Chronic Pelvic Pain
One fourth of women with acute PID may experience subsequent chronic pelvic pain
PID and Ectopic Pregnancy
One in ten women may have an ectopic pregnancy in their first pregnancy following PID Tubal pregnancy is the leading cause of firsttrimester, pregnancy-related deaths in American women
Prevalence of PID
8% of all American women of reproductive age are reported to have received treatment for PID in 1995 Over one million American women seek treatment for PID annually
“…Then in 1975, I had a bout of painful cramps which, as it turned out, were caused by a severe infection. My horrified gynecologist took out the IUD and prescribed large doses of antibiotics. He also warned me that I might have trouble having children.”
From: “A cash settlement but no apology” NY Times 2/99
“He was right, though I didn’t find out for more than a decade. I had to fall in love, get married and spend a few years trying to conceive before an X-ray showed that my fallopian tubes were scarred and completely blocked. My husband and I tried in-vitro fertilization three times but were unsuccessful in having children.”
From: “A cash settlement but no apology” NY Times 2/99
PID Evaluation and Clinical Health (PEACH) Study
Study Effectiveness of Outpatient Treatment of PID
Importance The first randomized clinical trial to evaluate the effectiveness of the most common treatment for PID
Eligibility of Patients into the Study
Inclusion Exclusion
(Must meet all of the following criteria) (Must not meet any of the following criteria) 1. 37 years of age or younger 1. Currently pregnant by urine testing 2. Willing to participate 2. Tubo-ovarian abscess 3. Presenting with a history of pelvic 3. Appendicitis, hemorrhagic ovarian cyst discomfort for <30 days. (This does or other condition requiring surgery by not need to be the chief complaint) ultrasound or laparoscopy 4. Experiencing pelvic organ 4. Nausea and vomiting after a trial of tenderness (uterine or adnexal metoclopramide tenderness) on bimanual 5. Antimicrobial therapy within the past 7 examination days 5. Leukorrhea and/or Mucopurulent 6. Delivery, abortion or gynecologic Cervicitis and/or untreated known surgery within the past 30 days +GC or CT 7. Prior hysterectomy or bilateral salpingectomy *Leukorrhea: white cells > epithelial 8. Allergy to penicillins, cephalosporins or cells in 4 fields at high power. Can tetracyclines use LPF to find fields with 9. Homeless substantial WBC’s.
Study Treatments
Inpatient 2g Cefoxitin IV every 6 hours 100mg Doxycycline IV every 12 hours Outpatient 2g Cefoxitin > 1 shot 1g Probenecid 100 Doxycycline oral twice a day
Outcomes
Time to Pregnancy
Outcomes
Short Term: ■ Time to Clinical Improvement ■ Microbiologic Cure on Repeat Cervical Culture and Endometrial Biopsy ■ Treatment Adherence Long Term: ■ Infertility ■ Tubal Occlusion of Women with Involuntary Infertility ■ Repeat Episodes of PID ■ Ectopic Pregnancy ■ Functional Decline due to Pelvic Pain ■ Frequency of Health Services use and Indirect PID-Related Costs; Cost Utility Analysis
PEACH: Long-term outcomes by treatment group
Outpatient n Pregnancy† Infertile‡ Self-reported recurrent pelvic inflammatory disease† Ectopic pregnancy† Tubal obstruction§ Chronic Pelvic Pain|| 51 4 7 128 12.4% 1.0% 41.2% 33.7% 66 16.6% 1 110 0.3% 29.8% 4 33.3% P = .11 P = .37 P = .72 P = .27 0.69 (0.43 to 1.09) 3.66 (0.40 to 33.12) 0.61 (0.10 to 3.63) 1.24 (0.87 to 1.77) 172 71 % 42.0% 18.4% n Inpatient % Statistical significance P = 1.00 P = .85 Adjusted OR (95% CI)* 1.00 (0.71 to 1.39) 1.32 (0.86 to 2.04)
166 41.7% 67 17.9%
*Odds ratios were adjusted for tubal ligation (yes/no), bacterial vaginosis (yes/no), and IUD in place (yes/no).
Source: Ness RB, et al. Effectiveness of inpatient and outpatient treatment strategies for women with pelvic inflammatory disease: Results from the Pelvic Inflammatory Disease Evaluation and Clinical Health (PEACH) Randomized Trial. Am J Obstet Gynecol 2002;186:929-37.
PEACH: Secondary Results
No Difference between treatment groups: ■ Among women with endometritis ■ Among women without previous PID ■ Among women who got IV doxycycline as inpatients
Barrier contraceptive use and relative risks for recurrent PID, chronic pelvic pain and infertility after PID.
Recurrent PID
(N) (%)
Condoms No 324 (16.7) 1.0 0.8 (0.5-1.5) 0.5 (0.3-0.9) 300 (36.7) 132 (54.5) 142 (31.0) 0.8 (0.5-1.3) 187 (26.7) 0.7 (0.5-1.2) 1.0 59 (45.8) 0.8 46 (34.8) 0.6 0.7 (0.4-1.5) 0.4 (0.2-0.9)
Chronic Pelvic Pain
(N) (%) Adjusted RR (N) (%)
Infertility
Adjusted RR
Adjusted RR+
< 5/10 times 156 (16.0) > 6/10 times 204 (8.8)
+ Adjusted for age, number of live births, race, non-monogamy at baseline, new partner at baseline, gonococcal or chlamydial cervicitis at baseline, education, number of study visits, and all other forms of contraception other than that under consideration ++ Diaphragms, spermicides, cervical cap, or female condom Source: Ness RB. Barrier contraception and the risk of recurrent pelvic inflammatory disease (PID), chronic pelvic pain, and infertility, following PID. Am J Pub Health, 2004.
Risk of Reproductive Outcomes Flollowing Endometritis and/or Upper Genital Tract Infection (UGTI)a or No Endometritis/No UGTI Among Women with Clinical PID
No Endometritis/ No UGTI (n=258) Outcome Pregnant Infertile Recurrent PID Chronic Pelvic Pain N 103 50 50 112 % 40.2% 17.8% 19.5% 44.4% Endometritis and/ or UGTI (n=356) N 142 53 46 98 % 40.8% 16.4% 13.2% 29.5% 0.8 (0.6-1.2) 1.0 (0.6-1.6) 0.6 (0.4-0.9) 0.6 (0.4-0.9)
b
Adjusted OR (95% CI)
a UGTI was defined as isolation of gonorrhea and/or chlamydia from endometrium. b Adjusted for age, PID history, race, and education. Pregnancy and infertility are additionally adjusted for reported baseline
infertility (one year of unprotected intercourse without resulting pregnancy)
Effect of Chronic Pelvic Pain Intensity on Physical and Mental Health (Means + sd are presented).
Composite
Physical Health
110 100 90 80 70 60 50 N=
346
No CPP
79
Mild-Mod
123
Mod-Severe
