Statistical techniques for trend analysis:

1. Cochran-Armitage trend test

2. Cuzick's Test (also known as the Cuzick-Edwards test)

3. Jonckheere-Terpstra Test

The Cochran-Armitage Trend Test is a statistical technique primarily used for categorical data. This Test aims to identify trends in proportions across ordered groups. It is instrumental in studies involving dose-response or toxicology, where the response variable is binary (e.g., presence or absence of a disease). In case you want to determine whether the proportion of cases increases or decreases across ordered groups (such as drug dosage levels). Features of the Cochran-Armitage Trend Test include the following: - It is an extension of the chi-square test that looks explicitly for trends across ordered groups. - The Test is susceptible to linear trends in proportions. 

Cuzick's Test, developed by Jack Cuzick, is a non-parametric test that identifies trends across multiple groups. It is an extension of the Wilcoxon rank-sum test and is often used in clinical trials and epidemiological studies. Features of Cuzick's Test include the following: It can handle more than two groups, unlike the standard Wilcoxon test, which typically compares two groups. - It is suitable for ordinal data where the normality assumption may not hold. 

The Jonckheere-Terpstra Test is another relevant method in this context. It is a non-parametric statistical test used to evaluate the differences between several independent samples where the alternative hypothesis is an ordered difference among the medians. This Test is instrumental when the data is ordinal, and the hypothesis involves an ordered alternative (i.e., whether medians of groups are ordered in a particular direction). Features of the Jonckheere-Terpstra Test include the following: - It tests for trends across multiple independent samples, assuming ordinal data. - It provides a robust assessment of ordered alternatives, which is particularly useful in psychological, medical, or behavioral studies where treatments are hypothesized to have an ordered effect. These methods are valuable for identifying trends and associations in categorical and ordinal data, especially in medicine, public health, and biological sciences.