6.3 Variable Relationships & Basic Probability

6.3 Variable Relationships & Basic Probability

In nursing practice and clinical research, understanding how healthcare variables relate to one another and quantifying the probability of patient outcomes are essential competencies. The ATI TEAS 7 Mathematics section evaluates your ability to analyze bivariate variable relationships (independent vs. dependent variables, scatter plots, correlation vs. causation) and calculate basic single-event and compound probabilities.

Whether evaluating the relationship between drug dosage and therapeutic serum level or calculating the probability of inherited genetic traits and diagnostic test accuracy, this section provides the mathematical framework required for healthcare statistics.

---

Independent vs. Dependent Variables

When investigating clinical relationships between two numerical quantities, variables are classified into two distinct roles:

1. Independent Variable (x-axis):

   • The variable that is manipulated, controlled, or selected as the predictor/cause.
   • Graphically plotted on the horizontal x-axis.
   • Clinical Examples: Medication dosage (mg), hours of exercise, patient age, elapsed infusion time.

2. Dependent Variable (y-axis):

   • The outcome, response, or effect that changes in response to the independent variable.
   • Graphically plotted on the vertical y-axis.
   • Clinical Examples: Systolic blood pressure (mmHg), serum drug concentration, resting heart rate, recovery time.

  DEPENDENT VARIABLE (y-axis)
    ^  Outcome / Response
    |    (e.g., Blood Pressure mmHg)
    |
    +----------------------------------> INDEPENDENT VARIABLE (x-axis)
                                         Cause / Predictor (e.g., Sodium Intake g)

Worked Example 1: Identifying Clinical Variables

A pharmacology study measures the effect of varying daily doses of an antihypertensive drug (0 mg, 10 mg, 20 mg, 40 mg) on patient mean arterial pressure (MAP).

Question: Identify the independent variable and dependent variable in this study.

• Independent Variable (x): Daily drug dosage in milligrams (controlled input).
• Dependent Variable (y): Mean arterial pressure in mmHg (measured clinical outcome).

---

Scatter Plots, Correlation, & Lines of Best Fit

A scatter plot displays bivariate data as individual coordinate points (x, y) on a two-dimensional grid to evaluate patterns of association.

Types of Variable Correlations:

1. Positive Correlation:

   • As x increases, y tends to increase.
   • Trend line slopes upward from left to right.
   • Clinical Example: Patient body mass index (BMI) vs. risk of developing Type 2 diabetes.

2. Negative Correlation:

   • As x increases, y tends to decrease (inverse relationship).
   • Trend line slopes downward from left to right.
   • Clinical Example: Weekly aerobic exercise hours vs. resting heart rate.

3. No Correlation (Zero Correlation):

   • Points are randomly scattered with no discernible linear pattern.
   • Clinical Example: Patient height vs. serum sodium concentration.

  POSITIVE CORRELATION       NEGATIVE CORRELATION       NO CORRELATION
    ^   .  *                   ^ *                        ^    .   *
    |  . *                     |   *                      |  *   .   *
    | . *                      |     *                    |    *   .
    +------------->            +------------->            +------------->

[!IMPORTANT] Correlation Does NOT Equal Causation: A strong statistical correlation between two variables does not prove that variable x causes variable y. Both variables may be influenced by an unmeasured third variable (confounder).

Line of Best Fit (Trend Line)

A line of best fit is a straight line drawn through scatter plot points that minimizes the overall distance to all data points. It is used to make predictions:

• Interpolation: Estimating y for an x-value within the range of plotted data.
• Extrapolation: Estimating y for an x-value outside the range of plotted data.

---

Basic Single-Event Probability

Probability quantifies the numerical likelihood that a specific event will occur. The probability of an event A, denoted P(A), is defined as:

$P(A) = \frac{\text{Number of Favorable Outcomes}}{\text{Total Number of Possible Outcomes}} = \frac{n(A)}{n(S)}$

Core Properties of Probability:

1. Bounded Range: Probability is always between 0 (impossible) and 1 (certain): $0 \le P(A) \le 1$
2. Representation Formats: Can be written as a fraction, decimal, or percentage (e.g., 1/4 = 0.25 = 25%).
3. Complement Rule: The probability that event A will not occur is: $P(\text{not } A) = 1 - P(A)$

Worked Example 2: Single-Event Patient Selection Probability

A clinic blood bank donor registry contains records for 100 patients categorized by blood type:

• Type A: 35 patients
• Type B: 25 patients
• Type AB: 10 patients
• Type O: 30 patients

Question: If a patient chart is selected at random, what is the probability that the patient has Type A blood? What is the complement probability of selecting a patient who does NOT have Type A blood?

• Step 1: Calculate P(Type A). $P(\text{Type A}) = \frac{35}{100} = 0.35 = 35\%$
• Step 2: Apply Complement Rule for P(not Type A). $P(\text{not Type A}) = 1 - P(\text{Type A}) = 1 - 0.35 = 0.65 = 65\%$

---

Compound Probability of Independent Events

Two events A and B are independent if the occurrence of event A has no effect on the probability of event B.

Multiplication Rule for Independent Events:

The probability that both independent events A and B occur together is the product of their individual probabilities:

$P(A \text{ and } B) = P(A) \times P(B)$

Worked Example 3: Clinical Diagnostic Monitor Reliability

An intensive care unit uses two independent electronic cardiac monitoring sensors (Sensor A and Sensor B) to detect arrhythmia. Each sensor operates independently with an accuracy rate of 95% (0.95).

Question: What is the probability that both sensors function accurately during a patient monitoring cycle?

• Step 1: Identify Individual Probabilities. $P(\text{Sensor A}) = 0.95, \quad P(\text{Sensor B}) = 0.95$
• Step 2: Multiply Probabilities on 4-Function Calculator. $P(\text{Both Accurate}) = 0.95 \times 0.95 = 0.9025$
• Step 3: Convert to Percentage. $0.9025 \times 100 = 90.25\%$

---

4-Function Calculator Strategies & Common Student Traps

[!TIP] Probability Multiplication Strategy: When multiplying fractions on a 4-function calculator (e.g., 3/4 * 1/2), convert each fraction to a decimal first (0.75 * 0.50 = 0.375) to avoid fraction multiplication errors.

Common Pitfalls to Avoid:

1. Confusing Negative Correlation with No Relationship: A negative correlation represents a strong, highly predictable inverse relationship—do not mistake it for lack of correlation.
2. Mixing Axis Assignment: Always ensure the independent variable (cause/input) is placed on the x-axis and the dependent variable (outcome/effect) on the y-axis.