FSCI

Kaplan–Meier curves indicated delayed conception during the post-FMD period across the study conditions (Figure 3). Heifers that did not conceive within the 180-day follow-up period were treated as right-censored observations. Cox proportional hazards analysis (Table 5) showed that, in Condition 1, the hazard of conception was significantly lower after FMD infection than before FMD infection (HR = 0.56; 95% CI: 0.46–0.67; p < 0.001). No significant baseline differences in conception hazard were detected between Condition 1 and Condition 2 (HR = 0.73; 95% CI: 0.49–1.10; p = 0.14) or Condition 3 (HR = 0.89; 95% CI: 0.72–1.09; p = 0.25). A significant interaction was observed between period and Condition 3 (HR = 1.38; 95% CI: 1.02–1.88; p = 0.04), indicating that the decline in the hazard of conception after FMD exposure was less pronounced in Condition 3 than in Condition 1. The interaction for Condition 2 was not significant (HR = 1.55; 95% CI: 0.88–2.72; p = 0.13). The proportional hazards assumption was met (global test, p = 0.49).

Figure 3: Unadjusted Kaplan–Meier survival curves for the interval from first AI to conception up to 180 days in HF heifers across three study conditions. Only heifers that remained in the herd and were eligible for reproductive follow-up were included in the analysis. Heifers that did not conceive by the end of the observation period were right-censored. The x-axis represents the number of days required to achieve conception, whereas the y-axis represents the cumulative probability of animals remaining non-pregnant during the observation period. (A) Condition 1; (B) Condition 2; (C) Condition 3.

Reproductive record exit

Figure 4 shows the unadjusted Kaplan–Meier survival curves for time to reproductive record exit during the post-FMD period in HF heifers across the three study conditions. Although the curves partly overlapped, the pattern of herd retention changed over time, indicating non-constant differences in reproductive record exit hazard among conditions. Because the proportional hazards assumption was violated in the standard Cox model, reproductive record exit was further evaluated using a piecewise Cox proportional hazards model with follow-up divided into 0–90-day and 91–180-day intervals (Table 6).

Using Condition 1 as the reference group, no significant difference in reproductive record exit hazard was observed for Condition 2 during the first 90 days (HR = 0.73; 95% CI: 0.22–2.37; p = 0.60), whereas Condition 3 showed a significantly higher hazard than Condition 1 (HR = 1.97; 95% CI: 1.23–3.17; p < 0.01). During the later follow-up period (91–180 days), the time-varying effect remained non-significant for Condition 2 (HR = 0.64; 95% CI: 0.15–2.69; p = 0.54), whereas Condition 3 showed a significant reduction in hazard relative to its earlier period effect (HR = 0.32; 95% CI: 0.16–0.63; p < 0.01). Overall, the piecewise Cox model was significant (likelihood ratio test, p = 0.004), indicating that the association between exposure condition and reproductive record exit hazard varied over time.

Figure 4: Unadjusted Kaplan–Meier survival curves showing time to reproductive record exit after foot-and-mouth disease (FMD) infection or vaccination in Holstein–Friesian heifers. Condition 1 = Heifers clinically affected by FMD during the outbreak; Condition 2 = Heifers developing clinical FMD after vaccination; Condition 3 = Clinically healthy heifers after vaccination.

Variable | Coefficient | HR | 95% CI | p-value
FMD after | −0.59 | 0.55 | 0.46–0.66 | <0.01
Condition 2 | −0.31 | 0.73 | 0.49–1.10 | 0.13
Condition 3 | −0.12 | 0.88 | 0.71–1.09 | 0.23
FMD after × Condition 2 | 0.44 | 1.56 | 0.89–2.73 | 0.12
FMD after × Condition 3 | 0.32 | 1.40 | 1.03–1.91 | 0.03

Condition | Time interval (days) | n/N | Coefficient | HR | 95% CI | p-valueᵃ
2 | <90 | 3/60 | −0.32 | 0.73 | 0.22–2.37 | 0.60
 | >90 | 6/60 | −0.45 | 0.64 | 0.15–2.69 | 0.54
3 | <90 | 33/248 | 0.68 | 1.97 | 1.23–3.17 | <0.01
 | >90 | 21/248 | −1.14 | 0.32 | 0.16–0.63 | <0.01