Percuma & Serta-merta — Tiada Pendaftaran Diperlukan

Likelihood Ratio Calculator

The Likelihood Ratio Calculator computes positive likelihood ratio (LR+) and negative likelihood ratio (LR-) from a diagnostic test's sensitivity and specificity. Evaluate how much a test result shifts the probability of disease for evidence-based clinical decisions.

Likelihood Ratio Calculator — Pratonton Nisbah Langsung
Likelihood Ratios
A
B
⚖️
Proportion Solver
A : B = C : D — Enter any 3 values
:
=
:
📊
Results
Visual ratio breakdown
Solved Proportion
Simplified
Percentages
Decimal
Fraction
Visual Ratio
A
B
Part A: —
Part B: —
    Ratio Simplifier
    Reduce any ratio to its simplest form
    :
    📊
    Simplified Result
    Reduced to lowest terms
    Simplified Ratio
    GCD Used
    Percentages
    Decimal Ratio
    Fraction
    Visual Ratio
    A
    B
    Part A: —
    Part B: —
      📐
      Ratio Scaler
      Multiply a ratio by a scale factor
      :
      ×
      📊
      Scaled Result
      Ratio after scaling
      Scaled Ratio
      Original
      Factor
      Percentages
      Simplified
      Visual Ratio
      A
      B

        🕐 Recent Calculations

        📭
        No calculations yet. Start computing above!

        What is a Likelihood Ratio?

        A likelihood ratio quantifies how much a test result changes the probability of having a condition. The positive likelihood ratio (LR+) measures how much a positive test increases disease probability. The negative likelihood ratio (LR-) measures how much a negative test decreases it. LR+ above 10 or LR- below 0.1 are considered strong evidence.

        Likelihood ratios combine sensitivity and specificity into a single metric that is independent of disease prevalence. Clinicians use LRs with pre-test probability (Bayesian reasoning) to calculate post-test probability, making them more useful than sensitivity/specificity alone for clinical decision-making.

        Formula & Persamaan Yang Digunakan

        Likelihood Ratio Calculator ini menggunakan 5 persamaan utama:

        1 Positive Likelihood Ratio
        LR+ = Sensitivity / (1 - Specificity)

        Sensitivity 95%, Specificity 90%: LR+ = 0.95 / (1 - 0.90) = 0.95 / 0.10 = 9.5.

        2 Negative Likelihood Ratio
        LR- = (1 - Sensitivity) / Specificity

        Sensitivity 95%, Specificity 90%: LR- = (1-0.95) / 0.90 = 0.05 / 0.90 = 0.056.

        3 Post-Test Odds (Fagan Nomogram)
        Post-Test Odds = Pre-Test Odds × Likelihood Ratio

        Pre-test probability 20% → odds = 0.25. LR+ = 9.5 → Post-test odds = 0.25 × 9.5 = 2.375 → probability = 70.4%.

        Cara Menggunakan Kalkulator Ini

        Untuk menggunakan Kalkulator Nisbah ini, ikuti 3 langkah berikut:

        1

        Masukkan Nilai

        Taip nilai nisbah yang diketahui ke dalam ruang input. Biarkan satu ruang kosong — itu ialah nilai tidak diketahui yang diselesaikan oleh Kalkulator Nisbah.

        2

        Pilih Mod

        Pilih mod nisbah — Selesaikan, Permudahkan, atau Skalakan. Setiap mod menggunakan persamaan berbeza untuk nilai input anda.

        3

        Dapatkan Hasil

        Klik Kira. Skrin hasil memaparkan jawapan dengan bar nisbah visual, carta pai, dan pecahan jalan pengiraan langkah-demi-langkah.

        Contoh Masalah & Penyelesaian Langkah-demi-Langkah

        Berikut ialah 3 contoh masalah dengan penyelesaian langkah-demi-langkah menggunakan Kalkulator Nisbah ini:

        Input 1 Test with 90% sensitivity, 85% specificity
        1 LR+ = 0.90 / (1-0.85) = 0.90/0.15 = 6.0
        2 LR- = (1-0.90) / 0.85 = 0.10/0.85 = 0.118
        3 LR+ of 6.0: moderate evidence for disease when positive
        4 LR- of 0.12: moderate evidence against disease when negative
        LR+ = 6.0, LR- = 0.12 — moderately useful test
        Input 2 Calculate post-test probability
        1 Pre-test probability: 30% → odds = 0.30/0.70 = 0.429
        2 Test positive, LR+ = 10
        3 Post-test odds = 0.429 × 10 = 4.29
        4 Post-test probability = 4.29/(1+4.29) = 81.1%
        Positive test raises probability from 30% to 81%
        Input 3 Highly sensitive test: 99% sensitivity, 50% specificity
        1 LR+ = 0.99 / 0.50 = 1.98 (weak positive evidence)
        2 LR- = 0.01 / 0.50 = 0.02 (very strong negative evidence)
        3 A negative result almost rules out disease
        4 A positive result barely changes probability
        LR+ = 2.0 (weak), LR- = 0.02 (excellent rule-out)

        Soalan Lazim

        What is a good likelihood ratio?

        LR+ > 10 is strong evidence for disease. LR+ 5-10 is moderate. LR+ 2-5 is weak. LR- < 0.1 strongly rules out disease. LR- 0.1-0.2 is moderate. LR values near 1.0 provide no useful information.

        Why are likelihood ratios better than sensitivity/specificity?

        LRs combine both metrics into a single number and are independent of disease prevalence. They can be applied directly to individual patients using pre-test probability. Sensitivity and specificity are prevalence-independent but don't directly give post-test probability.

        How do I use the Fagan nomogram?

        Draw a line from your pre-test probability (left axis) through the likelihood ratio (middle axis) to find the post-test probability (right axis). This visual tool quickly converts LRs into clinically useful probability changes.

        Can likelihood ratios be used for multi-level test results?

        Yes. Interval likelihood ratios can be calculated for different result ranges (e.g., low, moderate, high values) rather than just positive/negative. Each interval has its own LR for more nuanced interpretation.

        What is pre-test probability?

        Pre-test probability is your estimated chance of disease before conducting the test, based on prevalence, symptoms, and clinical judgment. It's the starting point for Bayesian reasoning with likelihood ratios.

        Ketahui Mengenai Nisbah

        Apakah itu nisbah?

        Nisbah ialah perbandingan antara dua atau lebih kuantiti yang menunjukkan saiz relatif satu kuantiti berbanding yang lain. Ditulis sebagai A : B, ia bermaksud 'bagi setiap A unit kuantiti pertama, terdapat B unit bagi kuantiti kedua.' Sebagai contoh, nisbah 3 : 4 bermaksud bagi setiap 3 bahagian A, ada 4 bahagian B. Nisbah digunakan dalam masakan, pembinaan, kewangan, sains, dan kehidupan seharian.

        Bagaimanakah saya menyelesaikan perkadaran?

        Perkadaran ialah persamaan yang menyatakan bahawa dua nisbah adalah sama: A : B = C : D. Untuk menyelesaikan nilai yang hilang, gunakan pendaraban silang. Jika D tidak diketahui: D = (B × C) / A. Ini berfungsi kerana dalam nisbah yang sama, hasil darab silang sentiasa sama: A × D = B × C. Penyelesai Perkadaran kami melakukan ini secara automatik — masukkan mana-mana 3 nilai sahaja dan ia akan mencari yang ke-4.

        Bagaimanakah saya memudahkan nisbah?

        Untuk memudahkan nisbah, cari Faktor Sepunya Terbesar (FSTB) bagi kedua-dua nombor dan bahagikan setiap satu dengannya. Sebagai contoh, 24 : 36 — FSTB bagi 24 dan 36 ialah 12. Jadi 24 ÷ 12 = 2 dan 36 ÷ 12 = 3, memberikan nisbah dipermudahkan 2 : 3. Penyelesai kami secara automatik mencari FSTB dan mengecilkan nisbah anda ke bentuk paling mudah.

        Apakah itu penskalaan nisbah dan bilakah ia berguna?

        Menskalakan nisbah bermakna mendarabkan kedua-dua bahagian dengan faktor yang sama untuk mencipta nisbah setara, lebih besar (atau lebih kecil). Contohnya, menskalakan 2 : 5 dengan faktor 3 menghasilkan 6 : 15. Ini sangat berguna untuk resipi (cth., menggandakan resipi), pembinaan (skalakan pelan), membancuh larutan, atau sebarang situasi di mana anda perlu mengekalkan kadar nisbah yang sama pada saiz berbeza.

        Apakah perbezaan antara nisbah dan pecahan?

        Nisbah A : B membandingkan dua kuantiti antara satu sama lain (bahagian-ke-bahagian), manakala pecahan A/B biasanya mewakili hubungan bahagian-ke-keseluruhan. Walau bagaimanapun, sebarang nisbah boleh dinyatakan sebagai pecahan: 3 : 4 adalah setara dengan 3/4 = 0.75. Perbezaan utama ialah konteks — nisbah membandingkan kuantiti secara bersebelahan, manakala pecahan mewakili sebahagian daripada keseluruhan.