Assessing quality of maternity care in Hungary: expert validation and testing of the mother-centered prenatal care (MCPC) survey instrument

Women in Hungary and in the CEE region report negative experiences with pregnancy care. It is unknown how common these experiences are. High numbers of women pay their obstetricians with informal cash payments, sometimes called “tips”, in order to obtain higher quality care. We don’t know if, when women pay informally, they actually get higher quality care.

In order to quantitatively explore the experience of quality maternity care in Hungary, we assembled a multi-disciplinary expert panel to adapt English-language maternity care surveys to the Hungarian context. We instructed the experts to think broadly about all aspects of care that may be important to women.

Generally, the experts found that English-language surveys could be easily adapted, and they helped us narrow the number of survey questions from 155 to 117. Because the informal payment, or “tip” system, is specific to Hungary, the experts developed new questions using Hungarian words to represent this practice. We then tested all questions on two groups of post-partum Hungarian women who use the internet: a random, representative sample and another group recruited from online maternity care forums.

We found that the new questions about informal cash payments made sense to women and that women reported more positive experiences with care when they paid informally. Women’s responses to an open-ended question revealed that we addressed the majority of care dimensions that mattered to them.

In conclusion, we developed a survey to comprehensively explore the maternity care experience in Hungary. Our survey process and questions may be useful to explore the maternity systems of surrounding countries.

Person-centered care has been associated with engendering the most optimal relationship between patient and provider in all medical specialties [1]. Balanced sharing of information, individualized care plans, and continuous emotional support are elements that have been shown to improve birth care outcomes and increase satisfaction with the birth experience [2, 3]. The World Health Organization’s (WHO) vision for quality of care for pregnant women and newborns mandates both the provision of evidence-based medical services and curation of the maternal experience [4, 5]. However, in a systematic review of 65 studies across 34 countries Bohren and colleagues confirmed that few tools exist to measure the experience of respect or mistreatment in maternity care [6].

While the CEE region demonstrates generally favorable maternal health indicators [7], Miteniece et al.’s systematic review of 20 investigations into the region’s quality of maternity care confirmed the need to also assess the professional, technical, and informational aspects of maternity care [8]. Birth care providers are often trained with outdated curricula [9], resulting in overapplication of non-evidence-based techniques. Doctor-patient communication is often poor, with providers lacking the skills to interpret what mothers need during pregnancy care [10, 11]. Women themselves play an important role in the doctor-patient interaction during pregnancy, including uptake of information, optimizing health behaviors, and adherence to care. However, there is scarce information about whether women in the CEE region have autonomy over their childbearing experiences. Mitenice et al. conclude that the evidence on these and other aspects of quality maternity care in the CEE region are derived in large part from qualitative designs, and few studies provide nationally representative data [8].

To date, there have been conflicting quantitative investigations of the quality of maternity care in Hungary. In a 1998 survey of academic obstetric departments, Hagymasy found high levels of “family-centered obstetrics”, defined as involvement of fathers, an upright birthing position, and skin-to-skin to contact [12]. In contrast, a 2004 “birth guide”, compiled by surveying a convenience sample of women and hospital staff, revealed significant variations in the quality of information provided to pregnant women and in respectful treatment from staff [13].

Neither of these previous surveys explored the effects of informal cash payments, even though in Hungary more than 60% of pregnant women pay informally for birth care [14]. Informal and formal cash payments may affect quality of maternity care in terms of affordability and accessibility of services. As with fee-for-service official payment schemes, informal payments may generate unnecessary use of services with “doctors recommending procedures in order to increase their income rather than for therapeutic benefit [15, 16].” Even though quantitative data is lacking, qualitative studies from Serbia and the Ukraine found that women pay informally to have a “chosen” obstetrician attend their births, and that they perceive many benefits to this continuity relationship—mainly, receiving more respectful care [10, 17]. However, given the lack of representative data to measure and monitor the quality of maternity care, the extent to which women actually benefit from these payments is unknown.

Investigators in the United States and in Canada have used cross-sectional surveys to assess women’s experiences of quality maternity care, addressing issues of evidence-based care, doctor-patient communication, and the process of decision-making in birth. The American survey Listening to Mothers 3 (LTM3) has been administered three times to a representative cohort of U.S. women [18]. Changing Childbirth in British Columbia (CCinBC) incorporated items from LTM3 but also used community-based participatory research methods to develop new items pertaining to women’s maternity care preferences, their decision-making, and perceptions of autonomy and respect [19, 20]. Given the existence of high-quality, English-language survey items, we decided to adapt and content validate these items for use in the Hungarian context with the primary aim of creating a comprehensive questionnaire to explore quality maternity care. Specifically, we examined quality care according to the rates of obstetric procedures, several measures of the experience of care, as well as the prevalence of informal cash payments.

Eleven of the 31 invited multi-disciplinary experts completed the entire validation process. The final panel consisted of: research and a clinical psychologists [2]; obstetrician-gynecologists [2]; a lawyer expert in birth issues [1]; directors of non-governmental organizations [2]; a midwife [1]; a doula [1]; an epidemiologist [1]; a mother [1]. One of the psychologists runs a support group for new mothers, and the NGO directors lead initiatives on expanding pregnancy and birth options. The doula herself is a mother and has supported birthing mothers in Hungary. Thus, 6 of the 11 experts had personal experiences as, or close relationships with active users of Hungarian maternity care.

Figure 1 summarizes the survey development and validation process. Only 3 items scored below the commonly used I-CVI cut off at or below 0.8. The LTM3 question “Is your baby during this time period living?” received an I-CVI of 0.8. Experts felt this question used harsh language and might turn women away. Another question from LTM3, “Did you get your first prenatal visit as early in your pregnancy as you wanted?” scored 0.76; experts felt this question was not relevant to a socialized health system. As a group the informal payment questions received scores (I-CVI 0.93) above the cut off for inclusion. However, experts consistently commented on the lack of relevance to the intrapartum context of questions developed for general inpatients. Table 1 lists the nine new informal payment questions that we developed with expert input.

After the informal payment questions the next most challenging group of items referred to the overlap between payments and the model of care (doctor, midwife). Experts decided that the process of paying a provider informally hinged on the model that the majority of women select prior to delivery. In accordance with previous research and their own system knowledge, experts then decided to use the word “chosen” (választott) to refer to the continuity prenatal relationship that women pay for informally. The word “chosen” was then applied to the private and public models of care to yield three models of chosen doctor care and one model of chosen (hospital) midwifery care. Two models of “not chosen” care represented the default model provided by the state insurance system. Independent home birth midwifery was its own category. These linguistic results are shown in Table 2.

Ipsos field tested the survey and confirmed that the duration of participant engagement required approximately 30 min. Altogether, Ipsos sent 892 e-mail invitations to their panel with a response rate of 67%. Reasons for drop out were: 14 (1.6%) quota full, 115 (12.9%) screened out, 163 (18.3%) terminated the survey. In addition, 657 completed surveys were obtained through convenience internet sampling. Table 3 shows demographic indicators for the sample with the corresponding most recent census data listed below the table. Overall, the representative sample compared well to recent census data. Pearson Chi² statistics show that the convenience sample was statistically significantly more highly educated women (Chi² = 341.8, p < 0.0001), lived in the capital (Chi² = 128.2, p < 0.0001), and had higher average net incomes (t = −16.02, p < 0.0001).

Table 4 reveals the informal payment practices in the representative sample according to the model of care categories. Excluding the categories with fewer than five respondents, we see that the response percentages to the informal payment question ranged between 75 and 86%. Pearson Chi² statistics showed that the share of women who paid informally was significantly different across the groups (Chi² = 183.6; p < 0.0001). ANOVA test shows that the amount of informal payment is also significantly different across groups (F = 6.73, p < 0.0001).

Regarding informed consent practices, Table 5 shows the responses from the representative sample as to whether a woman’s permission was obtained prior to undergoing a cesarean (N = 244) or an episiotomy (N = 257). Pearson Chi² statistics showed that the permission practices were significantly different across provider types for cesarean section (Chi² = 39.2, p = 0.003) but were not significantly different for episiotomy (Chi² = 18.6, p = 0.414). MADM scores were significantly different across permission categories (ANOVA results for caesarean: F = 14.50, p < 0.0001, for episiotomy: F = 10.34 p < 0.0001).

Table 6 shows the coded results from an open-ended question from LTM3. Thematic analysis of the open-ended responses (N = 1015) from the entire sample identified 13 priority areas of the maternity care experience, 9 of which were addressed by the questionnaire.

We used a standardized and rigorous methodology to develop and validate a survey instrument that comprehensively examined maternity care experiences in Hungary, thus filling an important gap where no government-sponsored data exists. The process included collating validated items from the international literature, adapting them to the Hungarian context by expert panel, designing region-specific new items, and validating the content. To the best of our knowledge, no other group has undertaken this task. We found that existing English-language survey items concerning the experience of maternity care were clear and relevant to the Hungarian context. This is likely due to the fact that many of the issues related to excessive obstetric procedures, poor communication, and the lack of maternal autonomy that we found in Hungary are also common in the United States and Canada [8, 18, 19].

Our expert process proved effective at identifying survey domains that required additional adaptation. For example, our maternity care experts identified that informal payment questions developed for general inpatients required adaptation. Our expert panel integrated linguistic, system, and user expertise to develop new survey items specific to the CEE region. We believe this was a result of the collaboration across our diverse panel. Some argue that content experts should have significant research or clinical experience. However, inclusion of “lay” experts has been found to be appropriate in many situations [27] and is consistent with the principles of patient-centered research [32].

To test reliability, we administered the survey to two samples of service users: a randomly selected representative sample and a parallel convenience sample. The instrument performed well in both groups: it was user friendly, feasible to distribute in an online format, and captured information on several domains relevant to maternal experience of care during pregnancy and childbirth.

In our 30-min survey 67% of the items addressed issues of person-centered care. We found that the extent of informed consent and autonomy (MADM scores) varied significantly across the model of care categories. We also found lower MADM scores in the women who had cesareans and episiotomies performed without their consent. Lack of consent for procedures was a common theme in the responses to the open-ended question. These findings are discussed in detail in a separate paper [33] and are supported by qualitative studies that show that women pay informally to receive care that they perceive to be more respectful [10, 17]. Analysis of our model of care categories showed extensive overlap between informal payments and the use of the word “chosen”. Women who went to their local clinic without choosing a doctor paid informally 17% of the time—the lowest frequency of all the models of care. We believe the statistically significant different distribution of informal payments across the care categories validates these categories for future research in Hungary and the CEE region. Because informal payments may distort health care services in ways that require policy intervention [21], reliable survey items are necessary to evaluate their effects [23].

We developed a reliable and relevant survey instrument to evaluate evidence-based care and maternal experiences in Hungary. This survey instrument can be easily adapted for use in other Central and Eastern European countries, where informal payments, the variable application of evidence, and concerns with respectful provider-patient relationships are similar. We plan to utilize the data resulting from this survey to inform interprofessional education and elucidate determinants of high quality maternity care in Hungary. A survey similar to ours could be used to regularly monitor trends in Hungarian maternity care as well as for cross-country comparisons in the CEE region, where representative data on quality maternity care is lacking.