Governing Knowledge, Not Gadgets: What Keeps Chinese Maternal Health Physicians Satisfied at Work

A physician’s daily work in a maternal and child health hospital runs on knowledge: updated clinical evidence, training that keeps skills current, and an organizational culture that either values or ignores all of it. A new study from Shenzhen, China, now suggests that it is precisely these human and institutional arrangements—rather than the technology that supposedly underpins them—that are most closely tied to how satisfied doctors feel about their jobs. The research, published in Frontiers in Public Health, examined how physicians’ perceptions of their hospitals’ knowledge infrastructure capability relate to their job satisfaction in a health system under intensifying pressure.

The study, led by Shuyan Jin of Shenzhen Maternity and Child Healthcare Hospital, Alexandra Fernandes of ISCTE-IUL in Lisbon, and Dong Wang of Southern Medical University, is unusual in that it unfolds in two phases. The first phase built and validated a measurement scale for what the authors call perceived knowledge infrastructure capability, or KIC—the foundational organizational capacity that enables effective knowledge management through technological, structural, and cultural conditions. The second phase applied that scale in a cross-sectional survey of physicians in Shenzhen’s maternal and child health institutions, using structural equation modeling to probe its associations with three distinct facets of job satisfaction: intrinsic, extrinsic, and general satisfaction.

The context matters. China’s maternal and child health institutions are government-established, non-profit specialized providers with a public-health mandate, delivering both preventive services and basic medical care for women and children. Declining fertility rates have reduced birth volumes, while a rising share of advanced-age mothers and high-risk pregnancies has raised the technical complexity and risk profile of the work that remains. Shenzhen, a megacity of more than 17 million permanent residents—about 7.89 million of them women, with migrants accounting for more than 65 percent of the population—exemplifies these shifting demands. Against this backdrop, physician job satisfaction is not a soft metric: nationwide surveys indicate that clinical physicians’ satisfaction in China remains at a low level, and low satisfaction among healthcare workers has been linked to turnover intention, with direct implications for patient safety and service quality.

In the first phase, the researchers began with a preliminary framework of three dimensions—Technology, Structure, and Culture—comprising 24 items drawn from existing knowledge management literature. They then ran two rounds of a modified Delphi consultation with 18 experts from Beijing, Shanghai, Shenzhen, and other regions, including hospital directors, management department heads, and researchers in hospital operations, clinical practice, health policy, and medical research. Expert authority coefficients ranged from 0.70 to 1.00, with a mean of 0.86, and all 18 experts returned their assessments. Following predefined screening criteria—items with mean total scores below 4.00 and coefficients of variation below 20 percent were deleted, for instance—the first round removed eight items, merged two, and optimized four.

What happened next reveals something important about measuring organizational life across cultures. Expert interviews, a 20-person pre-survey, and focus group discussions with five experts converged on the same observation: in the Chinese hospital context, respondents found it difficult to distinguish items about organizational structure from those about culture. As the authors explain, formal institutions and informal relationships and cultural norms are often highly embedded in Chinese management practice, jointly shaping how knowledge is created, shared, and retained. Drawing on the concept of knowledge governance mechanisms, the team consolidated the Structure and Culture dimensions into a single integrated construct, KGMs, and removed four further items for conceptual overlap. The final scale retained just two dimensions and 11 items: Technology, covering knowledge management systems and the hardware, software, networks, and databases supporting collaboration inside and beyond the hospital, and KGMs, covering training and continuing education systems, academic exchange, innovation incentives, colleagues’ recognition of knowledge’s value, expertise-based rewards, and senior management’s explicit emphasis on knowledge.

To validate this refined scale, the team surveyed 1,037 valid respondents from 49 healthcare institutions across 10 provinces and autonomous regions and 15 cities spanning eastern, central, and western China. Confirmatory factor analysis showed acceptable fit, with a comparative fit index of 0.980, a Tucker–Lewis index of 0.973, an RMSEA of 0.052, and an SRMR of 0.050. Cronbach’s alpha coefficients were 0.849 for Technology and 0.975 for KGMs, composite reliability exceeded 0.70 for both dimensions, average variance extracted exceeded 0.50, and the heterotrait–monotrait ratio of 0.706 fell below the 0.85 threshold for adequate discriminant validity. A single-factor comparison model fitted markedly worse, suggesting that common method bias was unlikely to be driving the observed covariation among items.

The second phase turned to Shenzhen’s 11 maternal and child health institutions, of which three were selected to reflect differences in institutional size, stage of development, and service volume. Of 366 valid physician questionnaires, 231, 101, and 34 came from the three institutions respectively. The sample was 75.7 percent female; 45.6 percent of participants were aged 35 to 44, 58.5 percent held postgraduate degrees, 41.3 percent held intermediate professional titles, and 79.0 percent worked more than 40 hours per week, with 51.9 percent performing four or more night shifts per month. Job satisfaction was measured with the widely used 20-item short form of the Minnesota Satisfaction Questionnaire, which yields intrinsic satisfaction scores from 12 items tapping the work itself—ability use, accomplishment, creativity, responsibility, and variety—extrinsic satisfaction scores from six items covering compensation, policies, advancement, supervision, and working conditions, and a general satisfaction score across all 20 items. Internal consistency was high, with alpha coefficients of 0.949, 0.937, and 0.971 respectively.

Before modeling, the researchers observed statistically significant differences in satisfaction scores across age, professional title, years of service, monthly income, and weekly working hours. Physicians under 35 reported higher scores than those aged 35 to 44; those with fewer than 10 years of service outscored the 10-to-19-year cohort; and those earning 15,000 CNY or less per month scored higher on extrinsic and general satisfaction than the highest earners above 20,000 CNY. These characteristics were then included as covariates in three separate structural equation models, each with acceptable fit—RMSEA values of 0.035, 0.038, and 0.036 for the intrinsic, extrinsic, and general satisfaction models respectively.

The central result was strikingly asymmetric. KGMs showed significant positive associations with all three satisfaction outcomes: a standardized beta of 0.727 (95 percent confidence interval 0.553 to 0.901) for intrinsic satisfaction, 0.679 (0.492 to 0.866) for extrinsic satisfaction, and 0.724 (0.549 to 0.899) for general satisfaction. Technology, by contrast, showed no statistically significant direct association with any of the three outcomes. An exploratory mediation analysis then tested whether Technology might work on satisfaction indirectly, through KGMs. The direct associations remained non-significant, but the indirect associations through KGMs were statistically significant for all three outcomes—standardized betas of 0.600 for intrinsic, 0.561 for extrinsic, and 0.598 for general satisfaction. The authors interpret this cautiously: technological capability itself may not automatically translate into higher job satisfaction; its value may need to be realized through corresponding organizational governance mechanisms, consistent with prior research noting that technological infrastructure functions as an enabling condition whose effectiveness depends on alignment with organizational structure, management practices, and cultural environment.

The authors are careful about what this cross-sectional study can and cannot show. The observed associations do not establish temporal or causal relationships, and the application data came from only three institutions in a single city, limiting generalizability to other regions or healthcare settings. The study focused exclusively on physicians, so extrapolation to nurses or medical technicians warrants caution, and because all variables were self-reported by the same respondents, common method variance cannot be fully excluded despite the bias checks. The first Delphi round also showed relatively low consensus—Kendall’s coefficient of 0.084 for importance ratings—which rose to 0.278 in the second round after feedback. Future work, the authors suggest, should span multiple regions, include other professional groups, and adopt longitudinal designs. Still, for hospital administrators in Shenzhen and similar settings, the practical message is concrete: strengthening training and continuing education systems, academic exchange platforms, innovation incentives, expertise-based recognition, and visible managerial commitment to knowledge may offer a practical approach to supporting physician job satisfaction—while new technology, to pay off, must arrive wrapped in the governance that makes it useful.

Subject of Research: The association between perceived knowledge infrastructure capability and physician job satisfaction in Chinese maternal and child health hospitals

Article Title: Perceived knowledge infrastructure capability and physicians’ job satisfaction in Chinese maternal and child health hospitals: evidence from Shenzhen

Article References: Jin, S., Fernandes, A., & Wang, D. (2026). Perceived knowledge infrastructure capability and physicians’ job satisfaction in Chinese maternal and child health hospitals: evidence from Shenzhen. Frontiers in Public Health, 14, Article 1964250. https://doi.org/10.3389/fpubh.2026.1964250

Image Credits: AI Generated

DOI: 10.3389/fpubh.2026.1964250

Keywords: knowledge infrastructure capability, knowledge governance mechanisms, job satisfaction, physicians, maternal and child health, Shenzhen, China, structural equation modeling, cross-sectional survey, healthcare management, Delphi method, occupational health