15  Consultation Rate Analysis

This chapter connects consultation data with the department’s full case database to compute consultation rates — both department-wide and per pathologist.

[1] TRUE

15.1 Monthly Case Volume

The department’s total monthly case volume provides context for understanding consultation patterns. Cases are counted as unique biopsy numbers per month from the full departmental database (329,997 unique cases across 65 months).

Monthly Unique Cases Processed by the Department. Shaded Region Indicates Months with Suspected Data Quality Issues (< 500 Cases).

15.1.1 Yearly Summary

Annual Case Volume Summary
Year Months Total Cases Mean/Month Median/Month Min Max Flagged Months
2,020 3 1,273 424 1 1 1,271 2
2,021 12 53,605 4,467 4,532 3,292 6,221 0
2,022 12 65,196 5,433 5,506 4,634 6,056 0
2,023 12 67,973 5,664 5,788 4,849 6,284 0
2,024 12 69,902 5,825 5,868 4,939 6,676 0
2,025 12 71,134 5,928 5,894 5,281 7,116 0
2,026 2 914 457 457 8 906 1

15.2 Consultation Rate as Percentage of Total Cases

By comparing monthly consultation counts with total cases processed, we can assess how frequently consultations are triggered relative to overall workload.

The consultation rate is computed as:

\[\text{Consultation Rate (\%)} = \frac{\text{Unique Cases with Consultation per Month}}{\text{Total Unique Cases per Month}} \times 100\]

Over the 41 months with normal data quality (overlap period: Aug 2022 – Dec 2025), the consultation rate ranges from 0.31% to 3.54% (median 1.87%).

Monthly Case Volume (Bars) and Consultation Rate (Line). Red-Shaded Months Have Data Quality Flags.

15.2.1 Seasonal Pattern

Seasonal Consultation Rate Distribution by Calendar Month (Excluding Flagged Months)

15.2.2 Monthly Consultation Rate Table

Monthly Consultation Rate
Month Total Cases Cases with Consultation Rate (%) Data Quality
2022-08 5,795 18 0.31 Normal
2022-09 5,045 49 0.97 Normal
2022-10 5,281 84 1.59 Normal
2022-11 5,732 107 1.87 Normal
2022-12 5,999 117 1.95 Normal
2023-01 5,889 128 2.17 Normal
2023-02 5,030 73 1.45 Normal
2023-03 5,987 84 1.40 Normal
2023-04 4,849 58 1.20 Normal
2023-05 5,719 75 1.31 Normal
2023-06 5,372 65 1.21 Normal
2023-07 5,412 65 1.20 Normal
2023-08 5,857 89 1.52 Normal
2023-09 5,653 69 1.22 Normal
2023-10 5,936 59 0.99 Normal
2023-11 5,985 68 1.14 Normal
2023-12 6,284 75 1.19 Normal
2024-01 6,327 83 1.31 Normal
2024-02 5,969 101 1.69 Normal
2024-03 6,187 104 1.68 Normal
2024-04 4,974 69 1.39 Normal
2024-05 6,676 134 2.01 Normal
2024-06 4,939 101 2.04 Normal
2024-07 5,900 121 2.05 Normal
2024-08 5,742 110 1.92 Normal
2024-09 5,322 105 1.97 Normal
2024-10 5,826 146 2.51 Normal
2024-11 5,836 97 1.66 Normal
2024-12 6,204 140 2.26 Normal
2025-01 5,928 210 3.54 Normal
2025-02 5,347 170 3.18 Normal
2025-03 5,780 180 3.11 Normal
2025-04 5,859 144 2.46 Normal
2025-05 6,166 177 2.87 Normal
2025-06 5,281 141 2.67 Normal
2025-07 6,078 188 3.09 Normal
2025-08 5,414 151 2.79 Normal
2025-09 5,723 166 2.90 Normal
2025-10 6,160 175 2.84 Normal
2025-11 6,282 217 3.45 Normal
2025-12 7,116 30 0.42 Normal

15.3 Per-Pathologist Monthly Consultation Rates

For each pathologist, we compute monthly rates of consultations asked and answered, restricted to months when they were actively employed.

15.3.1 Consultation Heatmap

Monthly Unique Cases with Consultation (Asked + Answered) per Pathologist. Only Months During Active Employment Are Shown.

15.3.3 Per-Pathologist Cases Reported and Consultation Rate

Monthly Cases Reported and Consultation Rate per 100 Cases for Top Pathologists

15.3.4 Summary Statistics

Per-Pathologist Consultation Summary (During Active Employment)
Pathologist Active Months Cases Asked Cases Answered Total Cases Mean/Month Median/Month Max/Month Cases Reported Rate/100 Cases
P8 41 618 348 966 23.6 24.0 42 8,005 12.07
P2 41 230 661 891 21.7 21.0 41 13,975 6.38
P5 41 145 731 876 21.4 19.0 45 17,800 4.92
P9 41 119 685 804 19.6 18.0 43 8,557 9.40
P11 41 105 522 627 15.3 10.0 48 6,759 9.28
P13 38 533 66 599 15.8 12.5 45 32,321 1.85
P10 41 375 215 590 14.4 10.0 40 14,948 3.95
P21 27 185 399 584 21.6 21.0 47 9,651 6.05
P17 41 180 357 537 13.1 12.0 25 12,579 4.27
P23 24 98 403 501 20.9 19.5 41 4,401 11.38
P6 18 108 246 354 19.7 17.0 41 7,208 4.91
P19 41 99 227 326 8.0 7.0 16 10,324 3.16
P28 24 188 122 310 12.9 2.0 41 1,829 16.95
P4 41 223 79 302 7.4 6.0 21 19,255 1.57
P18 41 174 82 256 6.2 5.0 17 12,616 2.03
P24 22 131 120 251 11.4 12.0 29 4,259 5.89
P1 41 132 28 160 3.9 3.0 18 16,194 0.99
P27 14 59 94 153 10.9 10.5 24 886 17.27
P3 11 138 14 152 13.8 13.0 26 3,721 4.08
P16 41 79 72 151 3.7 3.0 11 7,548 2.00
P22 17 148 1 149 8.8 8.0 27 3,046 4.89
P14 17 111 8 119 7.0 7.0 20 6,100 1.95
P7 23 87 3 90 3.9 3.0 17 7,962 1.13
P29 7 87 1 88 12.6 10.0 31 1,689 5.21
P25 21 51 17 68 3.2 3.0 8 7,628 0.89
P26 18 43 6 49 2.7 1.0 10 9,542 0.51
P30 5 9 18 27 5.4 7.0 10 583 4.63
P15 19 24 1 25 1.3 0.0 5 3,100 0.81
P31 7 16 0 16 2.3 0.0 14 508 3.15
P32 3 6 8 14 4.7 6.0 8 306 4.58
NoteInterpretation

This table shows each pathologist’s consultation activity during their period of active employment, counted as unique cases. “Cases Asked” is the number of distinct cases for which they requested a colleague’s opinion, and “Cases Answered” is the number of distinct cases for which they provided one. The “Rate/100 Cases” column normalizes consultation activity against each pathologist’s total reported caseload, allowing fair comparison between high-volume and low-volume pathologists.

15.4 Department Cases per Pathologist from External Data

Monthly Cases Reported per Pathologist from the Department Case Database

15.5 Per-Pathologist Consultation Rate (% of Cases)

For each pathologist, what fraction of their total reported cases resulted in a consultation — either requested (asked) or provided (answered)? This reveals individual consultation behaviour and how it evolves over time.

Per-Pathologist Consultation Rate as Percentage of Total Cases Reported (Unique Cases)
Pathologist Active Months Total Cases Cases Asked Cases Answered Total Consultation Cases Ask Rate (%) Answer Rate (%) Overall Rate (%)
P28 12 1,829 188 122 310 10.28 6.67 16.95
P8 41 8,005 618 348 966 7.72 4.35 12.07
P23 23 4,401 98 403 501 2.23 9.16 11.38
P9 41 8,557 119 685 804 1.39 8.01 9.40
P11 41 6,759 105 522 627 1.55 7.72 9.28
P27 5 886 31 36 67 3.50 4.06 7.56
P2 41 13,975 230 661 891 1.65 4.73 6.38
P21 27 9,651 185 399 584 1.92 4.13 6.05
P24 21 4,259 131 120 251 3.08 2.82 5.89
P29 6 1,689 87 1 88 5.15 0.06 5.21
P5 41 17,800 145 731 876 0.81 4.11 4.92
P6 18 7,208 108 246 354 1.50 3.41 4.91
P22 17 3,046 148 1 149 4.86 0.03 4.89
P32 2 306 6 8 14 1.96 2.61 4.58
P17 41 12,579 180 357 537 1.43 2.84 4.27
P3 11 3,721 138 14 152 3.71 0.38 4.08
P10 41 14,948 375 215 590 2.51 1.44 3.95
P30 3 583 6 13 19 1.03 2.23 3.26
P19 41 10,324 99 227 326 0.96 2.20 3.16
P31 3 508 16 0 16 3.15 0.00 3.15
P18 41 12,616 174 82 256 1.38 0.65 2.03
P16 41 7,548 79 72 151 1.05 0.95 2.00
P14 16 6,100 111 8 119 1.82 0.13 1.95
P13 38 32,321 533 66 599 1.65 0.20 1.85
P4 41 19,255 223 79 302 1.16 0.41 1.57
P7 22 7,962 87 3 90 1.09 0.04 1.13
P1 36 16,194 132 28 160 0.82 0.17 0.99
P25 21 7,628 51 17 68 0.67 0.22 0.89
P15 10 3,100 24 1 25 0.77 0.03 0.81
P26 18 9,542 43 6 49 0.45 0.06 0.51
NoteInterpretation

This table expresses consultation activity as a percentage of each pathologist’s total caseload, using unique case counts. The “Ask Rate” shows what fraction of their cases prompted them to seek a second opinion, while the “Answer Rate” shows how often they served as the consulted expert relative to their caseload. These percentage-based rates control for differences in overall workload, revealing individual consultation behavior independent of case volume.

Per-Pathologist Consultation Rate as Percentage of Total Cases. Blue = Consultations Requested (Asked), Red = Consultations Provided (Answered).

15.5.2 Pathologist Consultation Behaviour Heatmap

Monthly Consultation Rate (Asked per 100 Cases) by Pathologist. Darker Colors Indicate Higher Consultation Request Rates.

15.5.3 Yearly Pathologist Consultation Rate

Yearly Ask Rate (% of Cases) by Pathologist — Columns Are Years, Last Two Columns Are Overall Totals
Pathologist 2022 2023 2024 2025 Total_Cases Overall_Rate_Pct
P28 NA NA NA 10.28 1,829 16.95
P8 6.41 7.21 8.14 8.45 8,005 12.07
P23 NA NA 2.24 2.22 4,401 11.38
P9 1.01 0.60 1.56 2.34 8,557 9.40
P11 2.05 1.34 1.22 2.53 6,759 9.28
P27 NA NA NA 3.50 886 7.56
P2 0.91 1.71 1.91 1.84 13,975 6.38
P21 NA 0.35 1.33 3.49 9,651 6.05
P24 NA NA 2.65 3.37 4,259 5.89
P29 NA NA NA 5.15 1,689 5.21
P5 0.25 0.57 0.61 2.00 17,800 4.92
P6 2.20 1.30 1.82 NA 7,208 4.91
P22 NA NA 2.87 6.47 3,046 4.89
P32 NA NA NA 1.96 306 4.58
P17 0.55 0.73 2.14 1.93 12,579 4.27
P3 3.89 3.53 NA NA 3,721 4.08
P10 0.61 0.66 2.79 6.28 14,948 3.95
P30 NA NA NA 1.03 583 3.26
P19 0.13 0.77 0.98 2.23 10,324 3.16
P31 NA NA NA 3.15 508 3.15
P18 0.63 1.05 0.99 2.28 12,616 2.03
P16 0.52 0.35 1.34 2.41 7,548 2.00
P14 0.58 2.45 NA NA 6,100 1.95
P13 0.65 0.88 1.47 2.48 32,321 1.85
P4 2.60 0.98 0.72 1.27 19,255 1.57
P7 0.97 0.82 1.73 NA 7,962 1.13
P1 0.20 0.40 2.53 2.73 16,194 0.99
P25 NA NA 0.80 0.60 7,628 0.89
P15 0.67 0.84 NA NA 3,100 0.81
P26 NA NA 0.68 0.33 9,542 0.51

15.6 Consultation Rate by Biopsy Section

By joining the section (biopsy workstation) assignment data with consultation records, we can determine which biopsy sections generate consultations most frequently.

Of the 4532 unique consultation Case IDs, 96.2% match the section database. Auxiliary sections (Immun, Histo, Frozen, Sitoloji, Rezeksiyona Ek, Konsültasyon, Genetik, DIF) are excluded when a case also belongs to a primary section — this removed 69,003 auxiliary-only rows, ensuring cases are counted under their primary workstation.

Consultation Rate by Biopsy Section
Section Total Cases Cases with Consultation Consultation Rate (%)
ENDOKRIN 2,542 211 8.30
NEFRO 508 34 6.69
BASBOYUN 3,692 233 6.31
İMMUN 46,025 2,591 5.63
GIS-REZ 2,828 159 5.62
KONSÜLTASYON 2,460 130 5.28
AKC 3,351 169 5.04
DİF 562 27 4.80
HISTO 21 1 4.76
KM-YUM D 9,421 328 3.48
PANKREAS 480 16 3.33
KC 4,094 124 3.03
REZEKSİYONA EK 23,476 661 2.82
HEMATO 9,764 249 2.55
SİTOLOJİ 11,573 269 2.32
GYN 29,129 609 2.09
MEME 9,085 172 1.89
GIS 86,997 1,520 1.75
FROZEN 72 1 1.39
DERİ 18,266 253 1.39
ÜRO 8,048 101 1.25
PRENATAL 84 1 1.19
NÖRO 4,124 43 1.04
GENETIK 675 1 0.15
SMEAR 116,620 55 0.05
KARD.VASK 106 0 0.00
NoteInterpretation

This table shows what percentage of cases from each biopsy section (workstation) led to a consultation. Sections with higher rates handle case types that more frequently require a second opinion – this typically reflects greater diagnostic complexity or subspecialty overlap rather than individual pathologist uncertainty.

Note: The KONSULTASYON section represents cases that were explicitly routed through the consultation workflow — its high rate is expected by definition.

Consultation Rate by Biopsy Section (Sections with >= 100 Cases). Higher Rates Indicate Sections Where Consultations Are More Frequently Requested.

15.6.1 Section Volume vs Consultation Rate

Relationship Between Section Case Volume and Consultation Rate. Bubble Size Represents Absolute Number of Consultations. Labels Show Sections with >= 500 Cases.

15.7 Consultation Rate by Section Over Time

15.7.2 Section x Month Heatmap

Heatmap of Consultation Rates by Section and Month. Darker Colors Indicate Higher Consultation Rates.

15.7.3 Yearly Section Summary

Yearly Consultation Rate (%) by Section
Section 2022 2023 2024 2025 Total_Cases
SMEAR 0.04 0.02 0.06 0.13 116,620
GIS 1.74 1.52 2.31 3.38 86,997
İMMUN 4.42 5.75 7.04 10.35 46,025
GYN 1.50 1.46 2.62 5.38 29,129
REZEKSİYONA EK 2.88 2.76 3.93 5.07 23,476
DERİ 0.95 1.49 2.51 1.87 18,266
SİTOLOJİ 2.47 2.82 3.38 4.27 11,573
HEMATO 1.08 1.52 2.56 8.76 9,764
KM-YUM D 3.92 4.98 4.99 6.40 9,421
MEME 2.02 1.59 2.12 4.22 9,085
ÜRO 3.38 1.63 1.31 1.85 8,048
NÖRO 0.53 1.87 1.29 1.35 4,124
KC 5.03 4.62 3.37 4.03 4,094
BASBOYUN 6.39 9.72 10.80 8.63 3,692
AKC 2.48 5.47 6.29 10.63 3,351
GIS-REZ 4.71 6.38 7.98 10.20 2,828
ENDOKRIN 11.31 8.00 14.96 13.01 2,542
KONSÜLTASYON 4.44 5.82 6.68 12.81 2,460
DİF 0.00 0.89 7.75 8.16 562
NEFRO 0.00 0.00 14.39 9.43 508
PANKREAS 4.65 6.00 2.50 3.96 480
KARD.VASK 0.00 0.00 0.00 0.00 106

15.8 Statistical Analysis of Section-Level Consultation Rates

[1] TRUE

15.8.1 Do Consultation Rates Differ Across Sections?

A chi-square test of independence was performed to determine whether consultation rates vary significantly across biopsy sections. To satisfy independence, inferential tests are restricted to cases with a single non-auxiliary section assignment.

Inference subset: 250,357 single-section cases across 18 sections (excluded 65,112 multi-section cases).
χ²(17) = 2195.2, p < 2.2e-16, Cramér's V = 0.094 (95% CI: 0.089–0.097)

Consultation rates differ **highly significantly** across biopsy sections. The effect size (Cramér's V) indicates a **small-to-medium** association.

15.8.2 Which Sections Deviate from the Department Average?

Each section was compared against a leave-one-out department reference rate (computed from all other sections) using exact binomial tests with Benjamini-Hochberg correction.

Binomial tests: section consultation rate vs leave-one-out department reference. BH-corrected p-values.
Section Total Cases Rate (%) Direction Odds Ratio Reference Rate (%) p (raw) p (adj.) Significant
SMEAR 116,591 0.05 Lower 0.04 1.10 0.0000000 8.89e-323 Yes
GIS 68,090 1.02 Higher 2.25 0.46 0.0000000 5.47e-78 Yes
BASBOYUN 2,607 4.49 Higher 8.21 0.57 0.0000000 8.55e-63 Yes
İMMUN 988 4.96 Higher 8.76 0.59 0.0000000 3.72e-28 Yes
KM-YUM D 6,227 1.61 Higher 2.78 0.58 0.0000000 8.31e-18 Yes
ENDOKRIN 507 4.54 Higher 7.85 0.60 0.0000000 5.91e-13 Yes
GYN 21,762 0.98 Higher 1.71 0.57 0.0000000 1.80e-12 Yes
SİTOLOJİ 5,486 1.44 Higher 2.46 0.59 0.0000000 7.06e-12 Yes
REZEKSİYONA EK 436 3.21 Higher 5.45 0.61 0.0000007 1.36e-06 Yes
AKC 421 1.90 Higher 3.17 0.61 0.0046967 0.0071 Yes
GIS-REZ 514 1.75 Higher 2.92 0.61 0.0047425 0.0071 Yes
NÖRO 414 1.93 Higher 3.22 0.61 0.0042549 0.0071 Yes
KONSÜLTASYON 163 2.45 Higher 4.11 0.61 0.0181108 0.0251 Yes
HEMATO 4,712 0.38 Lower 0.62 0.61 0.0397096 0.0511 No
MEME 3,234 0.46 Lower 0.76 0.61 0.3647431 0.4377 No
DERİ 14,215 0.65 Higher 1.07 0.61 0.5171258 0.5818 No
ÜRO 3,393 0.68 Higher 1.11 0.61 0.5798803 0.614 No
KC 418 0.72 Higher 1.18 0.61 0.7454676 0.7455 No

15.8.3 Correlation: Section Volume and Consultation Rate

Spearman's ρ = -0.587, S = 1538, p = 0.0104

There is a significant negative correlation between section case volume and consultation rate. Sections with higher caseloads tend to have  lower  consultation rates.

15.8.4 Temporal Trend in Department-Wide Consultation Rate

TFPW Mann-Kendall (monthly deseasonalized): τ = 0.533, p = 0.0000, lag-1 rho = 0.333

There is a statistically significant increasing trend in the department-wide monthly consultation rate over time.

15.8.5 Seasonal Pattern in Consultation Rates

Kruskal-Wallis test: H(11) = 3.42, p = 0.9839, ε² = 0.000

No significant seasonal pattern was detected in monthly consultation rates. The consultation rate appears stable across calendar months.

15.9 Section → Responder Expertise

Which biopsy sections generate consultations directed to which subspecialty groups? For example, do GIS section cases predominantly get referred to GI pathologists, or do they also flow to hematopathologists and other experts?

[1] TRUE
Consultation Distribution by Section and Responder Expertise (% of Section Consultations)
Section BST Breast GIS General Gyn Hemato Lung Pancreas Skin Uro Total
İMMUN 14.5 7.5 26.8 1.2 12.7 18.6 3.4 6.4 7.2 1.7 3,241
GIS 2.8 2.0 54.3 2.0 2.8 12.2 2.4 20.8 0.4 0.3 2,055
REZEKSİYONA EK 9.2 4.2 40.9 1.5 14.1 13.6 5.5 7.2 3.0 0.8 867
GYN 5.0 0.7 7.1 2.4 55.1 15.3 2.2 1.0 11.0 0.4 721
KM-YUM D 37.6 4.1 20.1 0.0 10.2 16.0 2.3 2.0 5.1 2.8 394
SİTOLOJİ 39.7 4.3 15.7 0.6 15.1 13.2 4.0 3.7 1.8 1.8 325
HEMATO 12.2 4.7 12.9 0.7 5.0 54.7 4.0 1.4 2.2 2.2 278
BASBOYUN 41.1 0.4 41.8 0.8 0.0 11.0 0.4 0.4 4.2 0.0 263
DERİ 27.2 5.4 20.3 0.4 3.4 19.2 1.1 0.0 22.6 0.4 261
ENDOKRIN 12.0 1.3 17.6 0.0 0.4 3.0 0.0 0.0 64.8 0.9 233
GIS-REZ 9.7 3.5 48.9 1.8 11.5 9.3 2.2 12.3 0.0 0.9 227
MEME 7.0 62.1 0.5 1.9 8.4 9.8 0.0 0.0 4.2 6.1 214
AKC 25.4 3.9 8.3 1.5 15.1 14.6 24.9 2.0 1.0 3.4 205
KC 15.6 3.4 36.3 0.6 5.6 16.8 3.4 13.4 0.6 4.5 179
KONSÜLTASYON 21.3 4.7 25.4 1.2 14.8 19.5 3.0 5.3 2.4 2.4 169
ÜRO 23.1 36.1 14.8 0.9 3.7 12.0 0.0 2.8 1.9 4.6 108
SMEAR 24.1 5.2 12.1 1.7 48.3 3.4 1.7 1.7 1.7 0.0 58
NÖRO 38.0 6.0 14.0 0.0 4.0 30.0 2.0 2.0 0.0 4.0 50
NEFRO 5.9 0.0 91.2 0.0 0.0 2.9 0.0 0.0 0.0 0.0 34
DİF 3.7 0.0 96.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 27
PANKREAS 16.7 0.0 22.2 5.6 5.6 27.8 0.0 16.7 0.0 5.6 18
GENETIK 0.0 0.0 33.3 0.0 33.3 0.0 0.0 33.3 0.0 0.0 3

Heatmap Showing Which Expertise Groups Respond to Consultations from Each Biopsy Section. Color Intensity Represents the Percentage of That Section’s Consultations Directed to Each Expertise Group.

Alluvial (Sankey) Diagram Showing the Flow of Consultations from Biopsy Sections to Responder Expertise Groups. Only Flows with >= 5 Consultations Are Shown for Clarity.

15.10 Section → Consultation Topic

What types of questions arise from each biopsy section? This reveals section-specific consultation patterns — for instance, whether GYN cases generate more dysplasia questions than GIS cases, or whether hematopathology queries dominate certain sections.

Consultation Topic Distribution by Section (% of Section Consultations)
Section Cytology/FNA Diagnosis/Tumor Type Dysplasia/Grade Hematopathology Inflammatory/Non-neoplastic Metastasis/Origin Neuroendocrine Other Sarcoma/Mesenchymal Second Opinion/Review Staging/TNM Margin/Resection IHC/Biomarkers Total
İMMUN 8.5 6.8 19.3 18.9 8.0 11.5 4.4 6.9 6.1 1.2 6.3 1.4 0.7 3,241
GIS 1.8 2.2 40.9 28.8 9.9 2.9 3.5 5.4 1.3 0.3 1.9 0.2 0.8 2,055
REZEKSİYONA EK 7.7 2.5 25.5 20.6 11.9 13.6 4.7 3.2 3.7 0.3 5.2 0.9 0.0 867
GYN 8.9 12.6 25.2 1.4 9.3 4.7 1.2 23.7 3.5 4.0 3.6 1.5 0.3 721
KM-YUM D 4.1 10.2 6.1 15.5 8.9 14.0 4.1 9.1 19.5 1.0 6.3 0.8 0.5 394
SİTOLOJİ 79.1 3.4 2.2 3.7 1.8 4.3 3.4 0.0 0.9 0.0 1.2 0.0 0.0 325
HEMATO 38.8 4.0 1.8 28.4 5.4 12.6 1.4 2.9 2.2 0.4 1.8 0.0 0.4 278
BASBOYUN 0.0 12.2 20.9 13.7 22.8 2.3 1.5 13.7 4.9 1.1 4.6 2.3 0.0 263
DERİ 7.3 11.9 10.3 15.3 9.2 3.4 0.8 8.8 11.5 1.9 7.7 10.0 1.9 261
ENDOKRIN 1.3 3.9 3.4 2.6 14.6 4.3 1.3 18.9 0.4 0.4 46.8 2.1 0.0 233
GIS-REZ 5.7 1.3 5.3 14.5 6.2 37.0 8.4 3.1 11.0 0.0 3.5 4.0 0.0 227
MEME 1.4 4.2 20.6 12.6 2.8 29.4 2.3 15.9 4.2 2.3 2.8 1.4 0.0 214
AKC 8.8 17.1 3.9 16.6 7.3 11.7 7.8 4.4 11.2 2.9 8.3 0.0 0.0 205
KC 2.2 8.4 11.2 12.3 6.7 26.8 18.4 3.4 6.1 0.0 4.5 0.0 0.0 179
KONSÜLTASYON 5.9 10.7 18.9 17.8 6.5 17.8 5.3 2.4 8.3 2.4 3.0 0.0 1.2 169
ÜRO 5.6 6.5 18.5 12.0 9.3 6.5 0.9 10.2 24.1 0.0 5.6 0.9 0.0 108
SMEAR 98.3 0.0 1.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 58
NÖRO 18.0 0.0 20.0 20.0 8.0 12.0 2.0 8.0 2.0 0.0 6.0 0.0 4.0 50
NEFRO 0.0 0.0 11.8 8.8 38.2 0.0 5.9 0.0 0.0 0.0 32.4 0.0 2.9 34
DİF 0.0 0.0 7.4 7.4 37.0 0.0 3.7 3.7 0.0 0.0 37.0 0.0 3.7 27
PANKREAS 5.6 5.6 0.0 38.9 0.0 27.8 0.0 0.0 16.7 0.0 5.6 0.0 0.0 18
GENETIK 0.0 0.0 0.0 0.0 0.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3

Heatmap of Consultation Topics by Biopsy Section. Color Intensity Represents the Percentage of That Section’s Consultations Related to Each Topic.

15.10.1 Comparing Dysplasia and Hematopathology Questions Across Sections

Comparison of Selected Consultation Topics Across Biopsy Sections. Bar Length Shows the Percentage of Each Section’s Consultations Devoted to Each Topic.

Alluvial (Sankey) Diagram Showing the Flow of Consultations from Biopsy Sections to Consultation Topics. Only Flows with >= 5 Consultations Are Shown.

15.11 The “Low Threshold” Hypothesis

NoteWhat is the “Low Threshold” Hypothesis?

In traditional (glass slide) pathology, requesting a colleague’s opinion requires physically carrying slides to another workstation. This effort acts as a friction filter – pathologists may rely on their own judgment for borderline cases rather than walk down the hall. In a fully digital environment, consultation is as easy as clicking a button. The “Low Threshold” hypothesis predicts that removing this friction increases consultation volume (more consultations happen) while decreasing intensity per consultation (each one is resolved faster, because many of the additional consultations are quick, low-complexity “micro-consults” that would not have occurred in the analog world).

If the data supports this hypothesis, we would expect to see: (1) a negative correlation between monthly consultation volume and median turnaround time, (2) an increasing proportion of rapid consultations over time, and (3) formal trend tests confirming rising volume and falling TAT.

[1] TRUE

15.11.2 Volume vs Intensity: Dual-Axis Visualization

Monthly Consultation Volume (Bars) and Median Turnaround Time (Red Line). A Negative Relationship Supports the Low Threshold Hypothesis.

15.11.3 Correlation: Volume and Turnaround Time

Spearman's rho = -0.586, S = 18212, p = 0.0001 (n = 41 months)

A **significant negative correlation** was found between monthly consultation volume and median TAT. Months with more consultations tend to have shorter turnaround times — consistent with the Low Threshold Hypothesis.

Scatter Plot of Monthly Volume vs Median TAT with LOESS Trend. Each Point Is One Month.

15.11.4 “Micro-Consult” Detection

If the Low Threshold Hypothesis holds, we expect the proportion of rapid consultations – what we term “micro-consults” – to increase over time. These are consultations resolved in under 1 hour or under 4 hours, suggesting quick confirmatory checks rather than deep diagnostic deliberation.

Proportion of Rapid Consultations Over Time. An Upward Trend in Sub-1h and Sub-4h Consultations Supports the Micro-Consult Pattern.

15.11.5 Formal Trend Tests (Autocorrelation/Seasonality-Adjusted Mann-Kendall)

We apply trend-free prewhitened Mann-Kendall tests (with monthly deseasonalization) to three monthly time series. The hypothesis predicts: (1) increasing volume (positive tau), (2) decreasing median TAT (negative tau), and (3) increasing proportion of sub-4h consultations (positive tau).

Adjusted Mann-Kendall Trend Tests (TFPW + monthly deseasonalization) for the Low Threshold Hypothesis
Time Series τ (Tau) p-value Lag-1 ρ Expected Direction Observed Direction Significant Supports Hypothesis
Monthly Volume 0.536 1.19e-06 0.191 Positive (increasing) Increasing Yes Yes
Monthly Median TAT -0.562 3.55e-07 0.251 Negative (decreasing) Decreasing Yes Yes
Monthly % Under 4h 0.736 0.00e+00 0.164 Positive (increasing) Increasing Yes Yes

15.11.6 Interpretation

Summary of Evidence for the Low Threshold Hypothesis:

The data provide substantial support for the Low Threshold Hypothesis. Of the three predicted trends, 3 out of 3 were confirmed by statistically significant adjusted Mann-Kendall tests.

The Spearman correlation between monthly volume and median TAT (rho = -0.586, p = 1e-04) confirms a significant negative relationship: months with higher consultation volume tend to have shorter turnaround times.

The friction-filter concept. In the analog world, physical effort acts as a friction filter — pathologists might rely on their own judgment for borderline cases rather than walk down the hall. In the digital world, clicking a button to share a case takes seconds, effectively removing this friction. The consequence, if the hypothesis holds, is a shift in the composition of consultations: the overall volume rises not because difficult cases become more common, but because easier cases that previously would have been handled alone now enter the consultation workflow as quick, confirmatory ‘micro-consults.’ This pattern has important implications for quality assurance — even if individual consultations are less intense on average, the aggregate effect on diagnostic accuracy could be substantial, as borderline cases that previously went unreviewed now receive a second pair of eyes.

15.12 Discussion: Literature Context

15.12.1 Consultation Rates in Context

The consultation rates observed in this study can be contextualized against published benchmarks from multi-institutional studies. The CAP Q-Probes data from 180 laboratories and 2,746 intradepartmental consultations established a baseline consultation rate of approximately 0.5% of surgical pathology cases (Renshaw et al. 2002). In these studies, the consulted pathologist confirmed the original diagnosis in 70.5% of cases, added significant diagnostic information in 15.9%, and identified a total discordance rate of 4.7% (1.0% major) (Farooq et al. 2021).

Renshaw et al. (Renshaw et al. 2002) established an early benchmark, finding that 4.3% of in-house consultations resulted in a diagnostic change, with approximately 1% being clinically significant. In a more recent study of mandatory intradepartmental peer review at a tertiary hospital, Solivas and Diwa (Solivas and Diwa 2024) reported a 3.1% discordance rate among 5,377 surgical pathology cases, with gastrointestinal, gynecologic, and head & neck pathology showing the most discordant cases.

The section-level variation in consultation rates observed here aligns with the well-documented subspecialty differences in diagnostic difficulty. Elmore et al. (Elmore et al. 2015) demonstrated that breast biopsy concordance among pathologists was only 75.3%, with atypia diagnoses achieving just 48% agreement. Peck et al. (Peck et al. 2018) reported that inaccurate diagnoses (major discordance) ranged from 3% to 9% among different specimen groups, with gynecology, dermatopathology, and gastrointestinal specimens presenting the greatest diagnostic challenge. These findings support the hypothesis that section-level consultation rates reflect inherent diagnostic complexity.

15.12.2 Digital Pathology and Consultation Workflows

This study was conducted in a fully digital pathology environment, which has important implications for consultation patterns. Hanna et al. (Hanna et al. 2019) demonstrated that digital pathology implementation reduced glass slide archive requests by 93%, off-site requests by 97%, and improved surgical case turnaround time by one day. Digital telepathology workflows at Mount Sinai Hospital reduced consultation turnaround from 86 hours to 35 minutes — a 98% reduction (Haghighi et al. 2021).

A systematic review and meta-analysis by Azam et al. (Azam et al. 2021) covering 25 studies and 10,410 histology samples showed 98.3% overall concordance between digital and light microscopy. Major discordances were primarily related to nuclear atypia assessment, dysplasia grading, and malignancy evaluation (57%), challenging diagnoses (26%), and identification of small objects (16%). These are precisely the types of cases most likely to prompt intradepartmental consultation.

During the COVID-19 pandemic, Hanna et al. (Hanna et al. 2020) validated remote digital pathology diagnosis, achieving 100% major diagnostic equivalency between digital and glass slide diagnoses across 108 cases (254 parts, 1,196 slides). The framework for integrating digital pathology into clinical practice (Hanna et al. 2022) and quality management systems for digital pathology operations (Ardon et al. 2023) provide the operational context in which the consultation patterns analyzed here occur.

15.12.3 Workload Implications

The per-pathologist consultation rate trends documented in this study connect to the broader literature on pathologist workload and workforce challenges. Bonert et al. (Bonert et al. 2021) found that workload units rose 23% over nine years while case volume decreased 6%, indicating increasing specimen complexity. The same group (Bonert et al. 2022) documented substantial workload inequality (Gini coefficients 0.05–0.23) within hospital groups, with high individual workload and maldistribution associated with significant absences and departures.

The declining pathologist workforce — a 17.5% decrease in active US pathologists from 2007 to 2017 (Metter et al. 2019) — increases the importance of efficient consultation workflows. No standardized workload measurement tool currently exists for pathologists (Hanna et al. 2024), and wRVU productivity varies 4- to 7-fold between subspecialties (Parkash et al. 2018), making consultation workload assessment an important complement to case-volume-based metrics.

15.12.4 Quality Assurance Framework

The CAP/ADASP guideline (Nakhleh et al. 2016) establishes intradepartmental consultation as a primary mechanism for interpretive diagnostic error reduction, recommending case review conferences and focused second review for high-risk cases. Goebel et al. (Goebel, Ettler, and Walsh 2018) found that all surveyed centers had formalized processes for documenting consultations, yet 92% of academic and 90% of community pathologists also requested undocumented informal consultations — suggesting that formal consultation tracking (as in this study) captures only a fraction of actual consultation activity.

Electronic LIS-based tracking systems, such as the module developed by Dunbar et al. (Dunbar et al. 2022) that processed 3,049 consultations in its first year, enable the type of systematic analysis presented here. The UCLA telepathology network (Chong et al. 2019) further demonstrated that digital infrastructure facilitates subspecialty consultation across geographically distributed sites with 97.5% concordance.