Added value of point-of-care testing for Group A Streptococcus in community pharmacy sore throat pathways: Analysis of the Wales Sore Throat Test and Treat service

This study evaluating the Wales Sore Throat Test and Treat service found that relying solely on the FeverPAIN clinical scoring system for Group A Streptococcus diagnosis results in significant overtreatment and undertreatment, particularly among children, thereby supporting the integration of point-of-care testing into community pharmacy pathways to enable more targeted antibiotic prescribing.

The Gist

Imagine you have a sore throat. You walk into a pharmacy, and the pharmacist needs to decide: "Is this a virus (which needs rest and water) or is it a bacteria called Group A Strep (which needs antibiotics)?"

For years, the rulebook (called FeverPAIN) has been like a weather forecast. It asks questions like: "Do you have a fever? Is your throat red? Do you have a cough?" Based on the answers, it gives you a score. If the score is high enough, the rulebook says, "It's probably Strep, give them antibiotics."

But here's the problem: The weather forecast isn't always right. Sometimes it predicts a storm when it's just cloudy, and sometimes it says "sunny" when a hurricane is coming.

This study looked at a massive experiment in Wales where pharmacists didn't just rely on the weather forecast. Instead, they used a rapid test kit (like a pregnancy test, but for your throat) to get a "yes or no" answer right there in the shop. They compared the old rulebook against the new test kit for over 73,000 people.

Here is what they found, broken down simply:

1. The Rulebook is a Bit Clumsy

When the researchers looked at the results, they found that the FeverPAIN rulebook was only right about half the time when it came to deciding who actually had the bacteria.

• The "False Alarm" (Over-treatment): About 23% of people who got antibiotics based only on the rulebook didn't actually have the bacteria. They were given medicine they didn't need. It's like buying an umbrella because the forecast said "rain," but it turns out to be a beautiful sunny day.
• The "Missed Storm" (Under-treatment): About 44% of people who did have the bacteria were told by the rulebook that they were fine and didn't get antibiotics. It's like the forecast saying "sunny," but you're actually getting soaked in a downpour.

2. Kids Are the Tricky Ones

The study found that the rulebook was especially bad at guessing what was happening with children aged 6 to 10.

• Kids in this age group had the highest rates of the bacteria.
• However, the rulebook was the least accurate for them. It was like trying to use a map designed for adults to navigate a playground; the landmarks just didn't match up. The "weather" for kids was much more chaotic than the rulebook expected.

3. The "Test Kit" is the Better Compass

The rapid test (POCT) acted like a flashlight in the dark.

• When pharmacists used the test, they could see exactly who had the bacteria and who didn't.
• This meant they could stop giving antibiotics to people who didn't need them (saving the "umbrellas" for real storms) and make sure the kids who did need medicine actually got it.

4. Why Does This Matter?

You might think, "So what if we give a few extra antibiotics?" But there are two big reasons this matters:

• Superbugs: Antibiotics are powerful tools. If we use them when we don't need to, the bacteria learn how to fight back, creating "superbugs" that we can't kill later. It's like using a sledgehammer to crack a nut; eventually, the sledgehammer breaks, and you can't crack anything.
• Health Costs: Unnecessary medicine costs money and can cause side effects for patients who are already feeling sick.

The Bottom Line

The study concludes that relying on the old rulebook (FeverPAIN) alone in pharmacies is like trying to drive a car with a foggy windshield. You might get to the destination, but you're likely to take a wrong turn or hit a bump.

By adding the rapid test kit to the process, pharmacists can clear the fog. They can see the road clearly, treat the right people, and stop wasting medicine on the wrong ones. The authors are asking health officials to update the rules so that more pharmacies can use these "flashlights" to keep everyone safer and healthier.

Technical Summary

1. Problem Statement

Acute sore throat is a frequent cause of primary care consultations in the UK, yet the majority of cases are viral. Despite this, antibiotics are prescribed in approximately 60% of consultations, driving unnecessary antimicrobial use and resistance.

• Diagnostic Challenge: Differentiating Group A Streptococcus (GAS) from viral infections based on clinical symptoms alone is difficult.
• Current Guidelines: The National Institute for Health and Care Excellence (NICE) recommends using clinical prediction rules (FeverPAIN and Centor) to guide antibiotic prescribing.
  • FeverPAIN: Scores fever, purulence, rapid onset, inflamed tonsils, and absence of cough (Max 5). Threshold for antibiotics: $\ge$4.
  • Centor: Scores exudate, lymphadenopathy, fever, and absence of cough (Max 4). Threshold for antibiotics: $\ge$3.
• The Gap: Previous studies suggest these rules have poor predictive value at the individual level (e.g., <25% of patients meeting thresholds are microbiologically confirmed).
• Policy Divergence:
  • Wales & Northern Ireland: Utilize a "Test and Treat" model in community pharmacies where Point-of-Care Testing (POCT) is used to confirm GAS before prescribing.
  • England: The 2019 NICE review concluded POCTs were not cost-effective compared to clinical rules alone; consequently, England's 2024 Pharmacy First service does not include POCT for sore throats.
• Objective: This study evaluates the diagnostic performance of FeverPAIN and Centor scores against POCT results in a real-world community pharmacy setting to determine if clinical rules alone are sufficient for decision-making.

2. Methodology

• Study Design: Cross-sectional analysis of routine clinical data.
• Setting & Population:
  • Service: Wales Sore Throat Test and Treat (STTT) service.
  • Period: November 2018 to September 2024.
  • Participants: 73,617 patients aged $\ge$6 years presenting with acute sore throat to 577 community pharmacies.
  • Inclusion Criteria: Patients assessed with FeverPAIN or Centor, eligible for POCT (FeverPAIN $\ge$2 or Centor $\ge$3), and actually underwent the test.
• Intervention/Testing:
  • Pharmacists administered the OSOM® Strep A Test (reported sensitivity 96%, specificity 98% vs. culture).
  • Antibiotics were supplied only if the POCT was positive, following patient discussion.
• Data Analysis:
  • Primary Outcomes: GAS positivity rates and diagnostic performance metrics (Sensitivity, Specificity, PPV, NPV, AUROC) of FeverPAIN against the POCT result (used as the reference standard).
  • Thresholds: FeverPAIN was analyzed against the NICE antibiotic threshold ($\ge$4). Centor could not be analyzed against its threshold ($\ge$3) because the POCT eligibility criteria used the same cutoff, leaving no comparator group.
  • Stratification: Analyses were stratified by age groups (6-10, 11-15, 16-40, 41+).
  • Sensitivity Analyses: Included imputation for eligible patients who were not tested and analysis of ineligible patients who were tested.

3. Key Results

• GAS Positivity:
  • Overall POCT positivity was 37.0% (27,220/73,617).
  • Positivity was highest in children aged 6–10 years (47.0%) and lowest in those aged $\ge$41 years (29.5%).
  • No clear temporal trend in positivity was observed over the study period, despite a surge in consultations during winter 2022/23.
• Diagnostic Performance of FeverPAIN (at threshold $\ge$4):
  • Sensitivity: 55.0% (95% CI: 54.4–55.6%) — Indicates nearly half of true GAS cases would be missed.
  • Specificity: 77.0% (95% CI: 76.6–77.4%).
  • Positive Predictive Value (PPV): 57.6% — Meaning 42.4% of patients scoring $\ge$4 were POCT-negative (false positives).
  • Negative Predictive Value (NPV): 75.1% — Meaning 24.9% of patients scoring <4 were POCT-positive (false negatives).
  • AUROC: 0.70 overall, dropping to 0.69 for children aged 6–10 years (lowest performance).
• Clinical Implications (Counterfactual Analysis):
  • If antibiotics were prescribed based solely on FeverPAIN $\ge$4 (ignoring POCT):
    • Undertreatment: 44% of POCT-positive patients (true GAS cases) would not receive antibiotics.
    • Overtreatment: 23% of POCT-negative patients (viral/non-GAS) would receive unnecessary antibiotics.
  • Age Disparity: Undertreatment was most severe in children (6–10 years), where 46% of true GAS cases would be missed. Overtreatment was highest in adults (16–40 years).

4. Key Contributions

• Real-World Evidence in Pharmacies: This is one of the largest studies (n=73k) evaluating diagnostic rules specifically within a community pharmacy setting, addressing a gap identified in previous NICE reviews which lacked pharmacy-specific data.
• Quantification of Discordance: The study provides precise metrics showing that clinical prediction rules (FeverPAIN) have limited accuracy for individual decision-making in this setting, with significant discordance between scores and microbiological reality.
• Age-Stratified Insights: It highlights that the diagnostic limitations are most pronounced in children (6–10 years), a demographic with the highest GAS prevalence but the lowest AUROC for FeverPAIN.
• Policy Relevance: The findings directly challenge the 2019 NICE conclusion that POCTs offer no additional benefit, suggesting that the absence of POCT in England's Pharmacy First service may lead to higher rates of both overtreatment and undertreatment compared to the Welsh model.

5. Significance and Conclusion

• Limitations of Clinical Rules: The study concludes that FeverPAIN alone is insufficient for identifying GAS in community pharmacies. Relying solely on these rules leads to a "double error": missing nearly half of bacterial infections (risking complications) and prescribing antibiotics for nearly a quarter of viral cases (driving resistance).
• Value of POCT: Integrating POCT into pharmacy pathways significantly improves diagnostic accuracy, allowing for targeted antibiotic prescribing. The data suggests that POCT availability correlates with lower antibiotic supply rates (29.9% in Wales vs. 65.5% in England's Pharmacy First).
• Recommendations:
  • Re-evaluation of Guidelines: The authors call for a re-evaluation of the role of POCTs in acute sore throat management, particularly for pediatric populations.
  • Future Research: Economic evaluations are needed to assess the cost-effectiveness of POCT integration, considering downstream outcomes like reduced complications, lower resistance, and avoided repeat consultations.
• Final Verdict: The Wales STTT service model demonstrates that combining clinical assessment with rapid POCT offers a superior strategy for managing sore throats in community pharmacies compared to clinical rules alone.