Strengths and limitations
This study has several limitations that need to be noted. First, given the characteristics and temporal nature of our cross-sectional design, it was impossible to make a causal inference between potential associated factors and subfertility or infertility (presented as longer TTP in our study). Second, information on TTP and other measures was self-reported, which might introduce recall bias. To resolve this problem, we designed a set of questions to collect accurate data on duration among different populations, and logical errors were further excluded from the final analysis. Third, we only enrolled couples who conceived or were attempting to conceive at the time of the investigation instead of couples without intent to become pregnant, which might not be representative of women at risk for pregnancy. However, not all couples who engaged in unprotected sex without intention to conveive were whom truly desired to become pregnant, and the quality of information on TTP in this population is probably poorer. Finally, our data were collected from 2010 to 2011, when only couples who were both “only children” were encouraged to have their second child in China. Interpretation of the results for such group under complicated political restrictions is challenging.
Nevertheless, our study has obvious strengths. First, the integration of both retrospective and cross-sectional designs was applied. Thus, we not only analyzed TTP in pregnant women but also utilized current approaches to examine TTP in women attempting to conceive. Second, the couples in our study were sampled from the general population with adequate representation, making it possible to inquire the TTP in cases of unsuccessful attempts and infertility. Third, dissimilar to other several population-based studies that used traditional binary classification to estimate infertility with “yes” or “no,” our study acquired the total TTP distribution in couples at risk for pregnancy, which provided a more sensitive indicator of fecundity and its associated risk factors. Finally, this approach could facilitate comparisons across different types of population-based and clinical studies, which might help improve the public health guidelines and clinical recommendations.