Skip to main content
Cross Validated

Return to Question

Notice removed Authoritative reference needed by CommunityBot
Bounty Ended with Sextus Empiricus's answer chosen by CommunityBot
added 302 characters in body
Source Link
processing_statistician
processing_statistician
  • 297
  • 2
  • 14

I have a dataset that includescomprising approximately 500 patients with variousabout 10 different diseases (potentially, potentially with correlated outcomes), and 200 healthy controls, with patient. Patient data is sourced directly from the hospital and, while control data obtainedcomes from volunteers (unmatched)and is not matched.

  We have detailedcomprehensive records of various life events with, including specific dates (e.g., age of first-time illegal drug use) and dates of diagnosis for both patients and controls. However, no sampling weights are available.

I have several questions.

(A) Since the cases are oversampled As a first step, I plan to runapply a Cox PH model on this datasetproportional hazards (in whichPH) model to each disease isindividually, using birth as the outcometime reference (time = 0), in which each sample is weighted somehow by prevalence/incidence rate found by outside resources. Is thisThe events are binary (whether a good idea? Ifdisease is diagnosed or not). The dataset includes around 10 potential predictors, what issuch as sex, race, and education, with some covariates potentially being time-dependent (e.g., marriage history and employment).

(A) Given that cases are oversampled, can I weight the recommended way to incorporate information likesample based on prevalence/incidence rate or incidence rates from external sources? (A link of previous papers doing this will be extremely helpful.).

(B) Can I treat this asstudy be considered a case-cohort study (even if, even though it isdoes not fit the traditional definition?)

(C) I know thatWhile logistic regression can beis typically used for a case-control studystudies (thoughyielding only the odds ratio is usableratios). Is, is there a counterpart to this foran analogous approach in survival analysis for dealing with this type of data?

Thank you so much!

I have a dataset that includes approximately 500 patients with various diseases (potentially with correlated outcomes) and 200 controls, with patient data sourced directly from the hospital and control data obtained from volunteers (unmatched).

  We have detailed records of various life events with specific dates (e.g., age of first-time illegal drug use) and dates of diagnosis for both patients and controls. However, no sampling weights are available.

I have several questions.

(A) Since the cases are oversampled, I plan to run a Cox PH model on this dataset (in which disease is the outcome), in which each sample is weighted somehow by prevalence/incidence rate found by outside resources. Is this a good idea? If not, what is the recommended way to incorporate information like prevalence/incidence rate?

(B) Can I treat this as a case-cohort study (even if it is not?)

(C) I know that logistic regression can be used for a case-control study (though only the odds ratio is usable). Is there a counterpart to this for survival analysis?

Thank you so much!

I have a dataset comprising approximately 500 patients with about 10 different diseases, potentially with correlated outcomes, and 200 healthy controls. Patient data is sourced from the hospital, while control data comes from volunteers and is not matched. We have comprehensive records of various life events, including specific dates (e.g., age of first-time illegal drug use) and dates of diagnosis for both patients and controls. However, no sampling weights are available.

As a first step, I plan to apply a Cox proportional hazards (PH) model to each disease individually, using birth as the time reference (time = 0). The events are binary (whether a disease is diagnosed or not). The dataset includes around 10 potential predictors, such as sex, race, and education, with some covariates potentially being time-dependent (e.g., marriage history and employment).

(A) Given that cases are oversampled, can I weight the sample based on prevalence or incidence rates from external sources? (A link of previous papers doing this will be extremely helpful.).

(B) Can this study be considered a case-cohort study, even though it does not fit the traditional definition?

(C) While logistic regression is typically used for case-control studies (yielding only odds ratios), is there an analogous approach in survival analysis for dealing with this type of data?

Thank you!

Notice added Authoritative reference needed by processing_statistician
Bounty Started worth 200 reputation by processing_statistician
edited title
Link
Roger V.
Roger V.
  • 4.5k
  • 15
  • 35

survival Survival analysis for survey (case-control?) data

Source Link
processing_statistician
processing_statistician
  • 297
  • 2
  • 14

survival analysis for survey (case-control?) data

I have a dataset that includes approximately 500 patients with various diseases (potentially with correlated outcomes) and 200 controls, with patient data sourced directly from the hospital and control data obtained from volunteers (unmatched).

We have detailed records of various life events with specific dates (e.g., age of first-time illegal drug use) and dates of diagnosis for both patients and controls. However, no sampling weights are available.

I have several questions.

(A) Since the cases are oversampled, I plan to run a Cox PH model on this dataset (in which disease is the outcome), in which each sample is weighted somehow by prevalence/incidence rate found by outside resources. Is this a good idea? If not, what is the recommended way to incorporate information like prevalence/incidence rate?

(B) Can I treat this as a case-cohort study (even if it is not?)

(C) I know that logistic regression can be used for a case-control study (though only the odds ratio is usable). Is there a counterpart to this for survival analysis?

Thank you so much!