This household survey found there were few barriers to seeking health care outside the household in the Iraq. The large majority of Iraqi children and adults sick in the past two weeks had received care outside the household (86.0-91.9%). The high utilization is consistent with Articles 30 and 31 of the Iraqi constitution guaranteeing the right of all citizens to health care .
Respondents were generally satisfied with primary health care services available from public and private sources, and reported trust in health workers is high. The majority of respondents felt patients were treated with courtesy. Around a third of persons expressed some concern about the security situation when seeking health care services. When analyzed separately from all services combined, the pattern of satisfactions with public sector PHCC and other sources of primary care were generally similar. However, high satisfaction may in reality reflect low expectations from health services. In clinic exit interviews (IMC, unpublished) patients stated they did not expect doctors to communicate instructions for taking medicines, potential adverse effects, or information about their illness as "doctors have no time for this."
Almost all of those interviewed were aware of the services of the public sector PHCCs. These clinics were the site of treatment for the last illness in about 40% of children but in only 21% of adults. The remainder used private clinics and to a lesser extent public hospitals. Women delivering in the past year preferred private clinics for antenatal services, but most delivered in public hospitals. Only about 5% of the surveyed households had deliveries which occurred outside the formal health sector. This is far less than earlier national data have reported.  Many Iraqis feel that delivery in a private hospital frequently results in delivery by cesarean section. Few PHCCs have delivery services, which accounts for the three percent of pregnancies that were delivered here. Of children born in the past year, 55%, received their initial care in PHCC clinics.
Immunization coverage was found to be complete for age for measles in 69% of children 23 months and under and 65% for DPT3 by immunization card. While this is lower than reported values for Syria, Iran and Jordan, it does suggest that immunization has been relatively effective, despite on-going conflict .
Respondents were asked what type of services they would recommend to others for a variety of childhood and adult conditions. Although many would recommend the PHCCs for most of conditions suggested, for no conditions would more than half of respondents recommend PHCCs. Private doctors' offices would be recommended for 26-48% of outpatient conditions.
Of respondents who reported that they used or knew their local PHCC well, 63% reported that medicines were frequently not available. At their local PHCC, 64% felt health workers were too rushed, perhaps reflecting the general loss of Iraqi doctors from death and immigration. A shortage of female health workers at the local facilities was noted by 60%. In general, residents of PHCC service areas were satisfied with services, and felt they were treated courteously by competent health workers. The data suggest that PHCC facilities were well respected in the community. Traditionally, the health services in Iraq have been very hospital focused, with limited investment in primary health care. However, recent efforts have been made by the Ministry of Health and USAID to improve the quality of primary health care in Iraq .
Using household asset indicators it becomes evident that poorer households utilize PHCC services more than richer households. This emphasizes the role of PHCCs in providing the guaranteed right to health for all Iraqis, as set out in the Iraqi constitution. The PHCC services were in fact free, with few costs being reported for direct services. Poor households were no more likely to feel that hospitalization would produce financial difficulties than rich households. This also suggests that informal payments were not being extracted for hospital care. When free medications were not available from public facilities, as appears to be common in PHCCs, 57.8% of poorer households felt that they would have difficulty affording medicines, compared with 38.1% of richer households. The survey found expenditure on health in the past month was equivalent to $US77.20 which compares with $US50 found four years earlier in the large Iraq Family Health Service (IFHS) . While our data would include some inflationary cost, it is likely that access and wealth of our catchment area population differed significantly for national values recorded four years earlier, perhaps being more urban than the IFHS sample. A slightly smaller percent of the 2010 study population could meet outpatient costs out of pocket than in 2006/7 (75.7% vs 86.1% in 2010), the significance of this is not clearly, though the 2010 study population being almost entirely urban or peri-urban, with the previous study having a greater rural population.
Overall, 15% of Iraqi population was thought to have been displaced within Iraq by 2008, using the International Organization for Migration estimates . These estimates are very similar to the 14.2% of the population in this study (2010) who reported moving in the past five years. Not surprisingly, the majority of these moves were because of insecurity. Our 2010 migration findings could be an underestimate, as Erbil, a common destination for displaced Iraqi minorities, was not included in this survey for logistical reasons. Displaced populations are often marginalized from health care services, so it is reassuring to find no differences in utilization among the displaced living in the service area. Neither were there any differences in household assets from people who had moved into the area within five years.
Measuring the community perceptions of health services as reported here has an advantage over the commonly used health facilities exit interviews, which can suffer from a "gratitude bias" where patient satisfaction perceived health worker performance may be overstated [24, 25]. While this household survey approach may allow a more settled reflection on the most recent consultation and incorporation of individual outcome indicators, there is a risk that recall bias may blur some details of the most recent visit. However, an advantage of the survey approach can be the capture of overall impressions, potentially based on multiple visits.
A survey of this nature has a number of limitations. Only households less than 2 km from the PHCC were included. Persons living further away from the health facilities may have different perceptions and utilization patterns than those living closer. Further, we included only those PHCCs staffed by medical doctors, excluding the large numbers managed by nurses and medical auxiliaries. Participants may also over-rate the value of services for fear of losing access to services, either individually or through health policy changes. In the analysis we included recent users with those having an acquaintance with conditions at the health facilities, though not recent users. This could dilute the results, but when analyzed separately there seemed to be little difference in the perceptions between these two groups. This survey has the potential biases of cluster surveys, collecting information from similar households and with insufficient cluster size to make comparisons among clinics and within governorates. However, population data were not available for a simple random sample. In the absence of objective PHCC performance data and interviews with professional staff, the user perceptions we recorded provides only part of the picture of primary care in Iraq. Nevertheless, user and community perceptions are key drivers of utilization practices, and are important for the planning of health services. Selecting from the catchment areas of PHCCs may have meant that low and middle income households are over-represented in the study, as public sector facilities are typically not sited in high income neighborhoods. Further studies to examine quality of Primary Health Care services from among a sample of various care sources could validate some of the observations reported here against objective criteria.