Assessing heterogeneity in menstrual cycles by means of a multilevel latent class approach

ABSTRACT

In this paper, we study the problem of heterogeneity in cervical mucus hydration at different times relative to the mucus peak both between cycles and women, specifying and estimating appropriate multilevel latent class models for longitudinal data. We estimate multilevel and growth latent class models which classify women on the basis of the evolution of cervical mucus characteristics observed over the fertile period of each menstrual cycle taking into account that we observe a different number of cycles per woman and correlation over time between consecutive observations. The effect of potential covariates on mucus evolution patterns is as well evaluated. Results confirm the existence of heterogeneity in mucus evolution between cycles and women. Moreover, an important significant effect of a woman’s age is found.

Notes

¹ Model parameters are estimated by means of maximum likelihood. The estimation procedure was carried out with different sets of starting values, in order to avoid local maxima. Responses to items were treated as measured on an ordinal scale (Goodman 1979). For the best-fitting model, the value of the log-likelihood is equal to −14,105, BIC = 28,868, the parameters are equal to 85. Convergence is reached after 250 iterations of the EM algorithm, followed by 13 iterations of Newton Raphson. All models were estimated with Latent Gold 5.0 (Vermunt and Magidson 2013).

² Model parameters are estimated by means of maximum likelihood. For the best-fitting model, the value of the log-likelihood is equal to −14,605, BIC = 29,269, the parameters are equal to 29. Convergence is reached after 30 iterations of Newton Raphson. All models were estimated with Latent Gold 5.0 (Vermunt and Magidson 2013).